The Oxford Companion to Philosophy Part 44 pdf

basis of our dubitable, if for the most part correct, beliefs
about the empirical world.
But Husserl disagreed with Descartes in one crucial
respect. Descartes moved swiftly from the proposition
that ‘I think’ to the conclusion that I am a ‘thinking thing’.
The belief that I am a thinking thing is itself, Husserl
claims, to be bracketed. I, who am conscious of objects,
am neither a thinking substance, nor an embodied person,
nor even the stream of my experiences—for I am con-
scious of, and in that sense distinct from, my experiences;
I am the pure or transcendental ego, what Kant called the
‘I think’ which ‘must be able to accompany all my repre-
sentations’. The transcendental ego or ‘transcendental
subjectivity’ cannot itself be bracketed, any more than
Cartesian doubt can extend to the existence of the
doubter. Thus only transcendental subjectivity is ‘non-
relative . . . while the real world indeed exists, but in respect
of essence is relative to transcendental subjectivity, in
such a way that it can have its meaning as existing reality
only as the intentional meaning-product of transcendental
subjectivity’. Husserl here infers an idealist conclusion,
namely that objects are constituted by consciousness and
could not exist without it, from the the true premiss that
nothing can be conceived without being an object of con-
sciousness. The error depends on either or both of two
confusions: (1) between an intentional and a real object—
in conceiving an object, I make it an object of my con-
sciousness, but I do not thereby make it a real object, e.g. a
tree; (2) between making something my intentional
object by conceiving it and conceiving it as my intentional
object—I cannot think of a possible lifeless universe with-

out making it the object of my thought, but I do not
thereby think of it as an object of my thought and thus sup-
pose myself to be one of its inhabitants. (It is a mistake to
suppose that Husserl’s *idealism can only be avoided if we
reject the methodological use of epoche¯.) In his Cartesian
Meditations (1931; tr. The Hague, 1960) Husserl tried to
relieve phenomenology of the charge that it entails solip-
sism by explaining how one transcendental ego can experi-
ence another transcendental ego on a par with itself.
From the Ideas to the Cartesian Meditations, Husserl’s
enterprise is avowedly akin to Descartes’s Meditations and,
unavowedly, to Fichte’s *Wissenschaftslehre. But his last
great (unfinished) work, The Crisis of European Sciences and
Transcendental Phenomenology (1936; tr. 1954) is closer to
Hegel’s Phenomenology of Spirit. For it purports to show,
‘by way of a teleological–historical reflection on the ori-
gins of our critical scientific and philosophical situation,
the inescapable necessity of a transcendental–phenom-
enological reorientation of philosophy’. This is at odds with
his earlier approach in at least two respects: (1) a historical
or causal account of the genesis of our consciousness was
excluded or *‘bracketed’ in his earlier works; (2) in so far as
Husserl is now concerned not so much with particular
past events, as with the eidos of history, with the essential
historicity of consciousness, its burden of preconceptions
derived from the traditions of its social milieu, this casts
doubt on his own attempt to found a rigorous science, free
of all preconceptions, that goes directly ‘to the things
themselves’. In part 3 of the Crisis, and in other papers
intended for incorporation in it (such as ‘The Origin of

Geometry’) he develops the concept of the ‘life-world’
(Lebenswelt), the intersubjective world of our natural, pre-
theoretical experience and activity, which, he believes,
was neglected by philosophers such as Kant in favour of
the world of theoretical science. But the ‘theoretical atti-
tude’ (exemplified, for Husserl, by Galileo) arose historic-
ally, in ancient Greece, against the background of the
life-world, and the life-world essentially persists even after
the development of the theoretical ‘spirit’. Even the physi-
cist thinks of the sun as rising and setting, and as marking
the phases of his practical life. Husserl’s account of the life-
world, of its essential priority to theory, and of the emer-
gence of theory from it, owes something to the eidetic
method and to epoche¯: to describe the essential structures of
the life-world involves suspending our scientific presup-
positions and our practical engagement with the life-
world. Nevertheless, some philosophers, notably Merleau
Ponty, see the Crisis as a distinct departure from Husserl’s
earlier work.
Husserl has had an immense influence in continental
Europe. Phenomenological analysis has been applied to
psychology (Pfander), law (Reinach), values, aesthetics,
and religion (Scheler). Even philosophers who reject
Husserl’s theoretical doctrines have benefited from his
meticulous analyses of particular phenomena. But
thinkers such as Heidegger, Sartre, and Merleau-Ponty
have used phenomenology in the service of philosophical
positions quite different from Husserl’s own, and his hope
that his rigorous science would put an end to radical philo-
sophical disagreements has remained unfulfilled. m.j.i.

D. Bell, Husserl (London, 1991).
J. Derrida, Edmund Husserl’s ‘Origin of Geometry’ (New York, 1978).
—— Speech and Phenomena (Evanston, Ill., 1973).
H. Dreyfus (ed.), Husserl, Intentionality and Cognitive Science (Cam-
bridge, Mass., 1982).
F. A. Elliston and P. McCormick (eds.), Husserl: Expositions and
Appraisals (Notre Dame, Ind., 1977).
E. Levinas, The Theory of Intuition in Husserl’s Phenomenology
(Evanston, Ill., 1973).
Hutcheson, Francis (1694–1746/7). An academic philoso-
pher of Irish origin who taught (and was criticized by)
Adam Smith at Glasgow University and strongly influ-
enced Hume, he was the main representative of the
*‘moral sense’ doctrine in ethics, which he inherited from
Shaftesbury. The main thrust of his philosophy was to
emphasize feeling rather than reason or intuition as the
source of what we think of as moral knowledge, though it
is unclear whether this feeling detects special moral qual-
ities in actions or situations, as we feel the warmth of fire,
or whether we simply have feelings of approval or disap-
proval towards their non-moral properties. This latter
interpretation would place Hutcheson as an ancestor of
the twentieth-century *emotive theory of ethics, and
similar theories, but the eighteenth century was less
410 Husserl, Edmund
hypothetico-deductive method 411
sensitive than the twentieth to precise semantic analyses
of the meanings of words and phrases. a.r.l.
M. P. Strasser, Francis Hutcheson’s Moral Theory (Wakefield, NH,
1990).

hylomorphism. The doctrine that sensible things are
composites of matter (Greek hule¯) and form (morphe¯).
Against atomists, who explained big things in terms of
varying arrangements of small things, Aristotle found his
model in sculptures ‘formed’ from matter by the artist.
Once reified into metaphysical constituents, forms are
treated as primitive explanatory entities accounting for
both static and dynamic structure of things (e.g. the sub-
stantial form of bovinity explains both the organic differ-
entiation of cow bodies and their distinctive digestive
processes). m.m.a.
*atomism.
Marilyn McCord Adams, William Ockham (Notre Dame, Ind.,
1987), chs. 15–17.
Montgomery Furth, Substance, Form and Psyche: An Aristotelian
Metaphysics (Cambridge, 1988).
hylozoism: see panpsychism.
hypothesis. A hunch, speculation, or conjecture pro-
posed as a possible solution to a problem, and requiring
further investigation of its acceptability by argument or
observation and *experiment. Hypotheses are indispens-
able to human thinking, and used by everyone from detect-
ives (Sherlock Holmes) to metaphysicians. They form
the basis of an influential account of scientific method
(*hypothetico-deductive method), which is closely
related to the claim, associated with Popper, that scientific
*theories are empirical hypotheses and remain so, how-
ever successful they are at withstanding repeated attempts
to falsify them. a.bel.
*evidence.

Larry Laudan, Science and Hypothesis (Dordrecht, 1981).
hypothetical imperative: see categorical imperative.
hypothetico-deductive method. A theory in science is a
general statement (or *hypothesis) from which particular
inferences may be deduced. Thus, from the theory ‘All
planets have elliptical orbits’, given the information that
Mars is a planet, we can deduce that Mars has an elliptical
orbit. Observations (here, of Mars’s orbit) can then be
seen as confirming or falsifying the hypothesis.
In the twentieth century Karl Popper and many other
philosophers of science before him saw the core of science
as the deployment of what is called the hypothetico-
deductive method. An unfortunate consequence of this
view has been a concentration on the formal relationships
between theories and the statements which follow from
them. There has been consequent lack of interest in the
relationship of theories to the actual practices, evidential
and experimental, from which they emerge, despite
the fact that even testing a theory is in practice more com-
plex than the hypothetico-deductive model suggests.
a.o’h.
*Bacon.
E. Nagel, The Structure of Science (London, 1961).
I and thou. The relation of subject to subject celebrated in
Buber’s ethical and religious philosophy: I may contem-
plate a tree ‘as a picture: a rigid pillar in a flood of light . . .
overcome its uniqueness and form so rigorously that I
recognize it only as an expression of the law . . . dissolve it
into a number . . . But it can also happen, if will and grace
are joined . . . that I am drawn into a relation, and the tree

ceases to be an It. The power of exclusiveness has seized
me.’ Placing relationality (and so, intersubjectivity) ahead
of both subject and object, Buber’s approach springs from
Kant’s foregrounding of the subject morally and in consti-
tuting experience and from Hegel and Feuerbach’s insist-
ence that dialectic constitutes the self. Similar thoughts
unfold in Mead, Peirce, and Dewey; they take flight as
fully fledged metaphysical themes in Cohen, Rosenzweig,
and Buber. l.e.g.
Martin Buber, I and Thou (1922), tr. Walter Kaufmann (New York,
1970).
Ibn Gabirol, Solomon (c.1022–c.1058). Philosopher and
poet, author of the Fons Vitae, an intricate, highly abstract
dialogue on Neoplatonic metaphysics. Since the Arabic
original is lost, it was not known until Salomon Munk
(1845) recognized Shem Tov ibn Falaquera’s Hebrew
quotations from it that the Avecebrol of the Latin manu-
scripts was the well-known Hebrew poet Ibn Gabirol.
Born in Malaga and raised in Saragossa, Ibn Gabirol relied
on the idea of intellectual or ‘universal’ matter to explain
the emergence of multiplicity from God’s unity. Matter
here is the passive or receptive aspect of every being but
God. Ibn Gabirol’s On the Improvement of the Moral Qualities
(tr. Steven Wise (1902)) offers a physiological treatment of
ethics based on the theory of the four humours. ‘The
Kingly Crown’, his Neoplatonic poem on the descent and
destiny of the soul, is included in the Sephardic liturgy of
the Day of Atonement. l.e.g.
*Neoplatonism.
Solomon Ibn Gabirol, Fons Vitae ex Arabico in Latinum translatus ab

Johanne Hispano et Dominico Gundissalino, ed. C. Baümker
(Munich, 1892–95).
Jacques Schlanger, La Philosophie de Salomon Ibn Gabirol (Leiden,
1986).
Ibn Khaldu¯n (1332–1406). Born in Tunis, he was one of the
most creative of Muslim statesmen and political thinkers,
widely acclaimed by modern historians as the greatest
philosopher-historian. In his major theoretical work, The
Prolegomena, he introduced the notion of natural causality
in history, in contrast to Islamic theology, and called for
the definition and study of sociological and political
processes (considered to be the principles of historical
methodology) with the express investigative intention of
recovering historical accuracy. He defined and claimed to
be the originator of a ‘science of culture’ (‘umra¯n) that
would study cultures in multiple stages in their natural
human, social, and political development. His methodol-
ogy emphasizes the study of environmental impact on
social organizations and economic processes that define
value, prosperity, and culture. h.z.
Muhsin Mahdi, Ibn Khaldu¯n’s Philosophy of History: A Study of the
Philosophical Foundation of the Science of Culture (Chicago, 1964).
Ibn Rushd: see Averroës.
Ibn Sı¯na¯: see Avicenna.
idealism, British. Movement in nineteenth- and twenti-
eth-century British philosophy according to which ultim-
ate reality is mental or spiritual, or at least not physical.
Bradley, Green, and Bosanquet think matter is not real.
Physical objects and the subjective points of view of con-
scious individuals stand in a system of *internal relations

called ‘the absolute’ or ‘absolute mind’, so British idealism
is mostly a kind of quasi-Hegelian *absolute idealism, and
*truth is ultimately construed as *coherence. McTaggart,
although denying the existence of God and the reality of
time and matter, holds that we are essentially immortal
souls. His influential argument that time is not real puta-
tively shows that tense is incoherent. Change depends on
tense, time depends on change, so time depends on tense,
so there is no time. Tense is incoherent because it entails a
contradiction to say, for example, that the same (token)
event happened last week and will happen the day after
tomorrow.
Newton might be an early if unwitting exponent of
absolute idealism, depending on the meaning of his claim
that absolute space and time are ‘the sensoria of God’.
Physical objects and events logically depend upon space
I
and time. Space and time logically depend upon the pres-
ence of God, so physical objects and events logically
depend upon the presence of God.
More recently in Britain, *Timothy Sprigge has argued
for a kind of absolute idealism. Non-absolute or individu-
alistic versions of idealism have been defended by John
Foster and Howard Robinson. They argue that it is inco-
herent to maintain that ultimate reality is physical. Foster
argues for a quasi-phenomenalist idealism on which
physical facts are wholly constituted by experiential facts,
even though physical concepts are not analytically
reducible to experiential concepts. Foster therefore rejects
physical realism and any naïve realist view on which we

are directly acquainted with physical objects in sense
perception.
Although late twentieth-century British philosophy
was dominated by the materialist and positivist ideas com-
mon to English language philosophy, it is not impossible
that idealism will have a new role once non-reductivist
explanations of consciousness come to be taken more ser-
iously. It is now clear, if it was not clear then, that material-
ism, naturalism, and positivistic science have wholly failed
to provide any kind of explanation of the existence and
reality of one’s own subjective mind. s.p.
F. H. Bradley, Ethical Studies (Oxford, 1876).
—— Appearance and Reality (Oxford, 1893).
C. D. Broad, An Examination of McTaggart’s Philosophy, 2 vols.
(Cambridge, 1933, 1938).
John Foster, The Case for Idealism (London, 1985).
T. H. Green, Prolegomena to Ethics (Oxford, 1883).
J. McT. E. McTaggart, The Nature of Existence, 2 vols. (Cambridge,
1921 and 1927, repr. 1968).
Stephen Priest, Theories of the Mind (London, 1991).
Howard Robinson (ed.), Objections to Physicalism (Oxford,
1993).
Timothy Sprigge, The Vindication of Absolute Idealism (Edinburgh,
1983).
idealism, German. Movement in late eighteenth- and
early nineteenth-century German philosophy according
to which ultimate reality is mental or spiritual, or at least
not physical. Kant, Hegel, Fichte, and Schelling are usually
classed as German idealists, and Leibniz and Schopen-
hauer are sometimes included.

Arguably, Kant is not any kind of idealist. He called his
critical philosophy ‘transcendental idealism’, but realized
his mistake when readers understood this as a kind of ideal-
ism. In the second (B) edition of the Critique of Pure Rea-
son he expressly included a new chapter called ‘The
Refutation of Idealism’ (B 274–9). ‘Idealism’ in ‘transcen-
dental idealism’ is in fact correctly understood as ‘anti-
realism’. Kant thinks there are no metaphysical
propositions, so transcendental idealism is metaphysical
anti-realism. Kant makes many claims consistent with
idealism, but they are misunderstood as entailing ideal-
ism. Famously, he insists that we have knowledge only of
appearances, not of things-in-themselves. This means we
can only know things as they appear to us, not as they
could not appear to us. There is nothing to suggest that
appearances are mental (‘A motor car appeared around
the street corner’). ‘We can only know possible objects of
experience’ means: We can only know things as they
would appear to us if we were experiencing them. This
does not entail that those things are mind-dependent, only
that our knowledge has a perspectival nature. Kant makes
it clear that the notion of a noumenon is a ‘limiting’ con-
cept, and ‘existence’ is a category, so whatever interpret-
ation we give to ‘noumena’, it had better not imply that
there are any. Kant thinks we can know objects only if
they are spatio-temporal, Euclidian, Newtonian, count-
able, could interact causally, and are substances. If this
epistemological conservatism carries any ontological
implication, it is materialism, not idealism.
There is a huge secondary literature on Kant, most of it,

from Hegel to contemporary British, German, and Ameri-
can commentators, mistakenly construing Kant as some
sort of idealist. In the hands of Fichte, transcendental
idealism does become a kind of idealism. Fichte rejects
things-in-themselves, which he misunderstands as in hidden
ontological duplication of appearances. Although Kant
insists that there is no substantial subject or ego (for
example in the *paralogisms), and depicts the transcenden-
tal unity of apperception as the purely formal possibility that
any thought of mine may be prefaced by ‘I think [that] ’ (in
the Transcendental Deduction), Fichte nevertheless postu-
lates an inner ego as owner of one’s thoughts and experi-
ences, a self-consciousness that ‘posits’ the external world
and, with dubious coherence, itself. Self-consciousness
presupposes an ‘I–not I’ distinction, it arguably making
little sense to postulate a portion of what is that I am
without positing a distinct portion that I am not.
In the System of Transcendental Idealism (1800) Schelling
construes consciousness and the unconscious as recipro-
cally constituting. In the Exposition of my System of
Philosophy (1801) Schelling advocates an ‘identity’ or
‘indifference’ between nature and spirit, object and sub-
ject; but ‘identity’ should not be understood here as
‘numerical identity’ or ‘qualitative identity’. In a partial
anticipation of Hegelian dialectic, a and b are identical if
and only if, if not a then not b, and if not b then not a, so it
is legitimate to regard a and b as parts or ‘moments’ of a
larger whole. What a is a part of is nothing other than what
b is a part of. Ultimately, nature and mind are aspects of the
Absolute. Because nature and spirit are mutually depend-

ent in Schelling’s philosophy, it is better construed as
*neutral monism than idealism. In Bruno (1802) the
Absolute is identified with God. In Of Human Freedom
(1809) God is identified with the source of the universe as
pure, infinite free will.
Hegel’s philosophy may be understood as the applica-
tion of Kant’s critical philosophy to the totality of what is.
The triadic arrangement of Kant’s table of categories and
the synthetic resolution of the Third Antinomy in which
thesis and antithesis are both true, albeit at different levels
of explanation, provide Hegel with the immediate model
for his dialectic: the exhibiting of philosophical problems
as prima-facie contradictions which, on closer inspection,
idealism, German 413
turn out to be mutually consistent depictions of a larger
whole. By ‘logic’ Hegel does not mean logic but, roughly,
the exhibiting of semantic, psychological, and ontological
relations between universals.
Hegel made two attempts at writing his system: the
Phenomenology of Spirit (1807) and the Science of Logic
(1812–14), and, secondly, the so-called Encyclopaedia of the
Philosophical Sciences (1815–31), including the Lesser Logic
and Philosophy of Mind. In writing two fundamentally dif-
ferent kinds of philosophy book, phenomenology books
and dialectic books, Hegel, the greatest synthesizing
philosopher, failed to ‘overcome’ the greatest antithesis of
Western philosophy: between empiricism and rational-
ism. Kant, in his own view, had done this. s.p.
Henry Allison, Kant’s Transcendental Idealism (New Haven,
Conn., 1983).

Graham Bird, Kant’s Theory of Knowledge (London, 1961).
Stephen Priest, Theories of the Mind (Harmondsworth, 1991).
—— (ed.), Hegel’s Critique of Kant (Oxford, 1987).
idealism, philosophical. Philosophical idealism is not the
same as idealism considered as an attitude to be observed
in life; it is not the pursuit of an ideal. It is, rather, a meta-
physical theory about the nature of reality, and thus pre-
supposes a distinction between *appearance and reality,
drawn in an other than common-sense way. It maintains
in general that what is real is in some way confined to or at
least related to the contents of our own minds. Plato’s the-
ory of Forms is sometimes said to be a species of idealism
on the grounds that his Forms are also called Ideas. But
those so-called Ideas were not merely contents of our
minds; indeed Plato explicitly rejects that supposition in
his Parmenides. It has been argued by Myles Burnyeat that
idealism proper could not appear before Descartes had
argued for the epistemological priority of access to our
own minds. Although this has been disputed, there is
much to be said for the thesis. At all events, whether or not
there are to be found any indications of belief in philo-
sophical idealism before Descartes’s time, it certainly
needed his argument to provide it with any basis. Yet
Descartes was not himself an idealist.
What are the reasons, therefore, for thinking that real-
ity is confined to the contents of our minds—*ideas, as
they were called by Descartes and others at his time?
Berkeley, who was the first idealist proper, generalized
Locke’s arguments to the effect that where the perception
of qualities of things, such as colour, taste, and warmth, is

circumstance-dependent (i.e. relative to the context in
which perception takes place, e.g. the illumination, the
condition of our tongue, or the temperature of our hands)
those qualities cannot be real properties of things. Berke-
ley argued that this applied to all *perception. Since per-
ception is, he thought, a matter of having sensations or
ideas, and since to be is to be perceived (Berkeley’s car-
dinal thesis), only sensations or ideas can properly be said
to be or to be real. He summed this up towards the end of
his Three Dialogues by a twofold thesis maintained, he said,
both by philosophers and by the ‘vulgar’: those things they
immediately perceive are the real things, and the things
immediately perceived are ideas which exist only in the
mind. There are many points to question about this thesis,
including the whole idea of immediate perception and
the claim that, whatever immediate perception is, it is
confined to sensations or ideas.
The theory of perception involved, the so-called repre-
sentative theory of perception, according to which what
we perceive is at best mere representations of things,
remained part of the apparatus of empiricist thought, and
is implicit in Hume’s doctrine that what we are given is
impressions, of which ideas are in some way copies. In the
eighteenth century only Reid challenged the theory,
because he thought it led Hume to absurdity. But the the-
ory is still there in the thought of Kant, who held that
perception provides us only with representations (Vorstel-
lungen), however mediated by concepts. Kant held, how-
ever, that a mere subjective, Berkeleian idealism would
not do in that it did not make it possible to distinguish

properly what is objective from what is subjective in the
sense in which flights of fancy are subjective. Kant, fol-
lowed to this extent by Schopenhauer, thought that ideal-
ism must be transcendental, which he tried to define by
saying that ‘appearances are to be regarded as being, one
and all, representations only, not things-in-themselves,
and that time and space are therefore only sensible forms
of our intuition, not determinations given as existing by
themselves, nor conditions of objects viewed as things-in
themselves’ (Critique of Pure Reason, A 369). The main
point is that Kant thought that he could distinguish
between appearances or representations of perception
and *things-in-themselves, but that the conditions for a
further distinction within appearances between what is
objective and what is merely subjective could also be set
out. The spatial–temporal features of objects as given in
experience are thus, he said, empirically real but transcen-
dentally ideal. Kant thought that he could, in these terms,
show the unacceptability of Berkeleian idealism.
We have, so far, a distinction between two forms of
idealism. Post-Kantian philosophy supplied a third, which
became known as absolute idealism. Fichte began it
(although he called his form of idealism ‘critical idealism’)
by rejecting what Kant had had to say about things-in-
themselves, seeking to make the distinction between self
and non-self purely within what is due to the activities of
the self. Hegel took it much further, thinking that he could
demonstrate, first, the identity of consciousness with its
object and, second, the identity of consciousness with self-
consciousness. This led, to simplify his argument grossly,

to the idea of a universal self-consciousness, a universal
‘notion’ (Begriff ), which is what reality is. This is the
*Absolute (a term introduced by Fichte), the one uncondi-
tional entity. While both Berkeleian, subjective idealism
and Kantian, transcendental idealism, in effect construe
reality in terms of the contents of an individual mind,
absolute idealism tends to construe it in terms of some
interpersonal consciousness. Indeed, in it the distinction
between one self and another tends to lapse, leading, as is
414 idealism, German
explicit in F. H. Bradley’s Appearance and Reality, to a form
of *monism, according to which there is only one thing,
distinctions within which are simply appearance. This is
clearly heady stuff.
All these forms of idealism have in common the view
that there is no access to reality apart from what the mind
provides us with, and further that the mind can provide
and reveal to us only its own contents. This second con-
sideration is supposed to follow from the first, but it does
not do so unless one invokes additional considerations,
such as those adduced by Berkeley when he claims that the
circumstance-dependence of judgements of perception
show that the objects of perception are mind-dependent.
The effect of the latter is to assimilate perception to
*sensation. Something like pain is taken as the paradigm of
sensation, and it is then argued that the feeling of warmth
can be assimilated to pain, and that other forms of percep-
tion can be assimilated to that, in that, perhaps, they are all
subject to bodily and other, contextual conditions. Reid
reasonably defined sensation in terms of the idea that it

has no object other than itself. He did not think that this
was true of perception, despite what Hume and other
empiricists had said; he thought that perception involves
concepts and beliefs, but that these are of objects distinct
from what takes place in the mind. But, in fact, however
correct this last point is, one can show that it is so only by
meeting the arguments which try to assimilate perception
to sensation. G. E. Moore thought that he could refute
idealism by drawing attention to the distinction within
experience between the experience itself and its object.
But one needs to show in addition that that object can be
extra-mental.
Recent arguments for what has been called *anti-
realism raise difficulties over the idea that there can be
forms of understanding reality which are verification-
transcendent, so that there are problems about our attach-
ing content to something if there is no way of verifying
whether it does or does not hold good. Transcendental
idealism can be construed similarly as a form of anti-
realism in that Kant argues for limits on what can be
understood if it cannot be brought under the conditions of
objective judgement. However, anti-realism does not
quite entail idealism. Kant depends also, for his transcen-
dental idealism, on a representationalist view of percep-
tion, holding that sensible intuitions (i.e. what is given in
perception) which are brought under concepts in judge-
ment about the experienced world take the form of
representations. Kant inherited that view from his prede-
cessors, and accepted it because it seemed obvious. But it
is not obvious (though something approaching it has

become the vogue today, particularly among cognitive
scientists, who hold that the mind’s workings have to do
with mental representations). It is not obvious because it
is assumed that the stimulation of our sense-organs pro-
duces not merely sensations in the ordinary sense, but
something that performs the role of representing what-
ever produces the stimulation, so that it is this which
we are directly aware of (or which the mind is directly
concerned with), rather than the object itself. At the same
time, without that misconceived view of perception (or
something like it) idealism cannot get off the ground.
Idealism has been very pervasive in the history of phil-
osophy since at any rate the eighteenth century. It has
been less prevalent in recent times, but tendencies
towards representationalism are liable to push adherents
of that view in its direction. Consequently, defenders of
idealism in some form are still to be found. Indeed, many
beginners in philosophy seem to think that it is the most
obvious philosophical theory, although nobody before
Descartes would have thought just that. It is also note-
worthy that Berkeley thought that his idealism amounted
to a defence of common sense and that, as indicated
earlier, it was what philosophers and the ‘vulgar’ had in
common. Later forms of idealism have been less ‘obvious’
because they are more sophisticated and more complex.
Schopenhauer characterized his form of transcendental
idealism by saying that it amounted to the doctrine of ‘no
object without a subject’, and he defended that, partly by
appealing to Berkeley and partly by arguing that if we try
to imagine a world without a knowing subject, we are

bound to realize that we are involved in a contradiction.
For what we shall be imagining is something that is indeed
dependent on a knowing consciousness—our own. But
while it is clear that it is impossible for us to imagine any-
thing without an imaginer existing, namely ourselves, it
does not follow that we cannot imagine a scene in which
no conscious beings exist. It does not follow, that is, that
there cannot be objects the existence of which does not
depend on their being for a subject. As Hegel might have
put it, and despite his arguments to the contrary, what is
‘in-itself’ need not be ‘for another’. In the end, the only
positive argument for idealism of any form is to be found
in the representative theory of perception, and that theory
is false. d.w.h.
*for-itself and in-itself; idealism, British; idealism,
German.
G. Berkeley, Three Dialogues between Hylas and Philonous (1713).
D. W. Hamlyn, Metaphysics (Cambridge, 1984), ch. 2.
G. Hegel, The Phenomenology of Spirit, tr. A. V. Miller (Oxford,
1967).
I. Kant, Critique of Pure Reason, tr. N. Kemp Smith (London, 1929).
Godfrey Vesey (ed.), Idealism, Past and Present (Cambridge, 1982).
ideal observer theory. A theory of justification in ethics.
The theory is that moral judgements can be justified by
appeal to what an ideal observer or ‘impartial spectator’
would do or say in a given situation. The theory has been
developed from its origins in the British moralists of the
eighteenth century, but it still has a problem in providing
a non-circular account of an ideal observer. r.s.d.
Roderick Firth, ‘Ethical Absolutism and the Ideal Observer’,

Philosophy and Phenomenological Research (1952).
ideals, moral. Two levels of moral standards have been
distinguished in ethical theory: ordinary moral standards
and extraordinary moral standards. The first level is
ideals, moral 415
confined to standards in the common morality that apply
to everyone—the moral minimum. The second level is a
morality of aspiration; here individuals adopt moral stand-
ards that do not hold for everyone. These ideals transcend
what we appropriately expect of others and thus are
aspirational ideals of individual excellence. In so far as a
person aspires to moral goals that surpass the conventional
moral point of view, the person accepts moral ideals.
Those who fulfil these ideals can be praised and admired,
whereas those who fall short of ideals cannot be rightly
blamed or condemned by others.
*Supererogation is a category of moral ideals pertaining
principally to actions, rather than to virtues or motives. The
etymological root of supererogation means paying or per-
forming in addition to what is owed. It has several defining
conditions. Supererogatory ideals and acts are those
which exceed what is expected or demanded by the com-
mon morality; they are intentionally undertaken for the
welfare of others (although the actor need not intend to
act from an ideal). A supererogatory ideal is optional—
neither required nor forbidden by the common morality.
Omission of a supererogatory act is not morally wrong
and not condemnable by common-morality standards.
Nevertheless, individuals who act from ideals often do
not consider their actions to be morally optional. Many

heroes and saints describe their actions in the language of
obligation and even necessity: ‘I had to do it’, ‘It was my
duty.’ The point of this language is to express a personal
sense of obligation. Some philosophical accounts deny the
literal appropriateness of this language, interpreting it as a
form of moral modesty designed to deflect merit or
praise that might be showered on the person. But a
broader and more sympathetic interpretation is that a
personal norm is accepted by the person as establishing
what ought to be done from a pledge or assignment of per-
sonal responsibility.
Supererogatory acts done from moral ideals typically
would be required were it not for an abnormal depriv-
ation or risk present in the particular circumstances, but
the individual elects not to invoke an exemption from
acting based on the abnormal situation. The individual
therefore does not make a mistake in regarding the action
as personally required and can view failure as grounds for
guilt, even though no one else is free to view the act as
obligatory.
Not all ideals are exceptionally arduous, costly, or risky.
Examples of less demanding ideals include generous gift-
giving, volunteering for public service, forgiving
another’s costly error, devoted and extended kindness,
and complying with requests made by other persons
when these exceed the requirements of the common
morality. Many everyday actions exceed obligation with-
out being at the highest level of ideals.
Aristotelians have held that the ideal of an admirable
life of moral achievement is central to the very nature of

ethics, not merely a second level beyond ordinary moral-
ity. Each individual should aspire to a level as elevated as
his or her abilities permit. Some persons are more able
than others, and for this reason they merit more praise,
acknowledgement, and admiration. The Aristotelian
model does not expect perfection, but rather that one
strives toward perfection. Ideals are thus central in this
model, not merely ornaments to an already commend-
able life. t.l.b.
*absolutism, moral.
R. Crisp and M. Slote (eds.), Virtue Ethics (Oxford, 1997).
Joel Feinberg, ‘Supererogation and Rules’, Ethics, 71 (1961); repr.
in Judith J. Thomson and Gerald Dworkin (eds.), Ethics (New
York, 1968).
David Heyd, Supererogation: Its Status in Ethical Theory (Cam-
bridge, 1982).
J. O. Urmson, ‘Saints and Heroes’, in A. I. Melden (ed.), Essays in
Moral Philosophy (Seattle, 1958).
ideas. These are entities that exist only as contents of
some mind. Ideas in this sense should be distinguished
from Plato’s Ideas or *Forms, which are non-physical but
exist apart from any conscious beings. The image of a Pla-
tonic Form that occurs in a person’s mind would be an
idea in our sense. Beginning in the seventeenth century all
objects of consciousness were held to be ideas. For
example, we are conscious of ideas when we imagine,
remember, dream, or think about some concept or
proposition. Ideas are subjective in that individuals can be
aware only of their own ideas. If two individuals are imag-
ining Pegasus or thinking about the Pythagorean theorem

each is directly aware of a distinct idea, although these
ideas may share many features. This is analogous to the
sense in which two reproductions of the Mona Lisa are
distinct objects even though most of their properties are
identical, but it is impossible for one individual to inspect
another’s ideas.
Reflection on the nature of our perceptual experience
led Descartes and Locke, among others, to argue that
even when we are perceiving we are directly aware of
ideas, not physical objects. For example, touching a hot
object or walking into a wall may cause someone to feel
pain. The pain is caused by a physical interaction between
the object and the perceiver’s body, but the pain exists
only as long as the individual is conscious of it. Moreover,
pain is subjective: if two people walk into the same wall
each experiences a distinct pain that exists only in that indi-
vidual’s experience. Thus to experience pain is to experi-
ence an idea. But all perceptual experience arises in the
same way as does pain. We feel warmth or solidity, see
colour or shape, and so forth, because an object acts
causally on our sense-organs. The item we become
directly aware of as a result of this interaction is an idea.
This thesis receives further support when we consider the
way in which the apparent size, shape, and colour of a
physical object, such as a distant tower, changes as our dis-
tance from the tower and our angle of observation
changes. Since the tower presumably remains unchanged
it is concluded that we do not directly perceive the tower,
but rather ideas that are caused to exist in our minds by the
interaction between the tower and our sense-organs.

416 ideals, moral
Once the doctrine that we experience only ideas was
accepted, three major philosophical problems immedi-
ately arose. First, are the ideas we experience adequate
copies of items that exist apart from our experience? In the
case of perception it is generally agreed that pain does not
exist apart from experience and philosophers argued that
other ideas which physical objects cause us to experience
may not actually characterize those objects. Thus philoso-
phers sought criteria for assessing which of our perceptual
ideas characterize items in the physical world. Second,
ideas are mental entities which, according to Descartes’s
analysis, have nothing in common with physical objects.
Thus it is unclear how interactions between a physical
object and a human body (also a physical object) can gen-
erate ideas. This question is one aspect of the *mind–body
problem. Third, if we are directly aware only of our own
ideas, it becomes problematic how we know that any-
thing exists other than these ideas. This question also
arises for ideas of non-physical objects such as God. After
Descartes the doctrine that we are aware only of our own
ideas became widely accepted and the three problems just
noted became central problems of epistemology and
metaphysics. h.i.b.
*phenomenalism; concepts; content; innate ideas.
G. Berkeley, The Principles of Human Knowledge, in Philosophical
Works (London, 1975).
R. Descartes, Meditations on First Philosophy, in The Philosophical
Works of Descartes, ii, tr. J. Cottingham, R. Stoothoff, and D.
Murdoch (Cambridge, 1984).

D. W. Hamlyn, Sensation and Perception (London, 1961).
ideas, innate: see innate ideas.
ideas of reason. This is Kant’s expression for the products
of *reason. In the Critique of Pure Reason Kant argued that
there are three such ideas corresponding to the self, the
world, and God; and that human reason is subject to an
unavoidable ‘transcendental illusion’ through which it
assumes the existence of non-empirical objects corres-
ponding to these ideas, but that they nevertheless have an
important regulative function in the systematic unifica-
tion of experience. h.e.a.
Norman Kemp Smith, A Commentary to Kant’s Critique of Pure
Reason, 2nd edn. rev. and enlarged (New York, 1962).
identity. The word ‘same’ is used sometimes to indicate
similarity (qualitative sameness), as in ‘Rachel is the same
age as Tony, and the same height as last year’, sometimes
to indicate that what is named twice should be counted
once (numerical sameness), as in ‘The morning star and the
evening star are the same planet’. The word ‘identical’ can
also have the former sense (identical twins, identical
dresses) as well as the latter; hence philosophers are
liable to discuss both kinds of sameness under the label
‘identity’.
Similarity comes in degrees and ways. Jane may be
exactly the same in looks as she was, or as her sister, but
only roughly the same in weight. Leibniz’s thesis of the
identity (i.e. numerical identity) of indiscernibles (i.e.
qualitative identicals) states that no two things can be
exactly the same in every way, sharing all their qualities.
This is disputable, but becomes a tautology if numerical-

identity-with-a is allowed to count among the qualities of
a. The converse thesis (often called Leibniz’s law), that
things differing in quality must be two, is harder to doubt.
But it must not be interpreted in such a way as to banish
change, since a can have some quality that b used to lack,
and still be numerically the same as b: many things persist
through change. Hume thought that in the ‘proper’ sense
identity over time requires changelessness. That would be
true if the proper sense of identity were exact qualitative
identity; but in fact the numerical sense is no less proper,
merely different.
Different kinds of thing have different criteria of numer-
ical identity. For example, mathematical classes are the
same if and only if they have the same members; contrast
regiments, clubs, etc., which can survive the addition or
withdrawal of members. The criterion of identity of many
things is vague, especially over time. For example, even
though the plural ‘clouds’ shows that we sometimes
count clouds, as we do not at a single time count fog,
nevertheless the question ‘How many clouds are there in
the sky?’ will rarely have even an approximate answer,
unless it is ‘none’. Sometimes the criterion is purely con-
ventional (one road runs all the way from Edinburgh to
London) or stipulated for a particular purpose (you are to
count books by titles, not volumes or copies of titles or
copies of volumes). Conflicting criteria are even allowed
to coexist. For example, someone might count St Mark’s
and the Palazzo Ducale as different buildings and also as
parts of the same building.
Numerical identity is an *equivalence relation, i.e. tran-

sitive, symmetric, and strongly reflexive. But philoso-
phers have sometimes proposed criteria, e.g. of personal
identity, that fail to respect these properties. For example,
it is logically (if not physically) possible that two different
persons should both be linked by memory chains with a
given later person. But it is not logically possible that two
different persons should be numerically the same person
as a given person. For the same reason ‘is a clone of’ lacks
the formal properties of numerical identity; and although
‘lies on the same, non-branching line of clone-descent
with’ puts things right formally, it would be odd if, for
example, the identity of last year’s bulb with its current
clone depended on there happening to be no current
rivals. Some philosophers have concluded that a criterion
of identity need not be logically equivalent to identity, i.e.
equivalent in all logically possible situations, whether or
not within our experience. Others infer that the search for
a precise and helpful criterion cannot always succeed.
Geach and others have used Locke’s myth of the prince
whose soul comes to ‘inform the Body of a Cobler’ to
argue that numerical identity is relative to sorts, so that it
is logically possible, for example, that the prince of today
should be the same person but not the same man as the
prince of yesterday, even though both are persons and
identity 417
both are men. (Locke’s own conclusion was different,
though just as odd.) If this is right, the number you get
depends on what sort of things you are counting things as.
In any case it depends on what sort of things you are count-
ing, as Frege saw: e.g. one atlas is many maps. c.a.k.

*identity of indiscernibles; of identity criterion.
D. Hume, A Treatise of Human Nature (1739), ed. L. A. Selby-Bigge,
3rd edn. (Oxford, 1978), i. iv. 6.
J. Locke, An Essay Concerning Human Understanding, 2nd edn.
(1694), ed. P. H. Nidditch (Oxford, 1975), ii. xxvii.
M. K. Munitz (ed.), Identity and Individuation (New York, 1971).
D. R. P. Wiggins, Sameness and Substance (Oxford, 1980).
identity, criterion of. A criterion of *identity is a principle
specifying, in a non-circular way, the identity-conditions
of objects of a given kind. Objects of different kinds can
have different identity-conditions. Thus, a criterion of
identity for rivers might specify that if x is a river and y is a
river, then x and y are the same river if and only if x and y
have the same source and the same mouth.
So, one common form for a criterion of identity to take
is this: if x is a K and y is a K, then x and y are the same K if
and only if x and y stand in relation R. But there is another
form of identity criterion, commonly associated with
Frege, and exemplified by his criterion of identity for the
directions of lines: the direction of line x is the same as the
direction of line y if and only if line x is parallel with line y.
e.j.l.
E. J. Lowe, ‘What is a Criterion of Identity?’, Philosophical
Quarterly (1989).
identity, the paradox of. Wittgenstein says (Tractatus
Logico-Philosophicus, 5.5303): ‘Roughly speaking, to say of
two things that they are identical is nonsense, and to say of
one thing that it is identical with itself is to say nothing at
all.’ If identity is a relation it must hold either between two
distinct things or between a thing and itself. To say that A

is the same as B, when A and B are distinct, is bound to be
false; but to say that A is the same as A is to utter a tautol-
ogy. Different solutions have been found by different
philosophers for this paradox, which is discussed by Plato,
Hume, and Frege, amongst others. Frege dealt with the
paradox by making a distinction between the *sense and
reference of an expression. Wittgenstein’s solution is to
deny that identity is a relation. Anything useful that is said
by means of ‘is the same as’ can be said by a sentence con-
taining a repeated expression. Thus instead of saying ‘The
author of the Iliad was the same as the author of the
Odyssey’ we can say, repeating the ‘x’, ‘For some person, x,
both x wrote the Iliad and x wrote the Odyssey’, and for
‘Florence is the same as Firenze’ ‘For some city, x, both x
is called Florence and x is called Firenze’. c.j.f.w.
C. J. F. Williams, What is Identity? (Oxford, 1989).
identity of indiscernibles. The doctrine of the *identity
of indiscernibles has various formulations, ranging from a
trivially true version to the metaphysically weighty ver-
sion employed by Leibniz. Here is a trivially true version:
for any individuals x and y, if, for any property f, x has f if
and only if y has f, then x is identical with y. Let the prop-
erty f be the property of being identical with y. Surely y has
it. But, then, if x has every property yhas, then x has it also.
Hence, x is identical with y. Here is Leibniz’s version: for
any individuals x and y, if for any intrinsic, non-relational
property f, x has f if and only if y has f, then x is identical
with y. Thus, according to Leibniz’s version, if x and y are
distinct individuals, they can not differ simply with respect
to extrinsic, relational properties; they must differ with

respect to some intrinsic, non-relational property as well.
Clearly, the exact content of Leibniz’s version of the
identity of indiscernibles turns on how we understand the
notion of an intrinsic, non-relational property. Subse-
quent to Leibniz, philosophers have formulated versions
of the identity of indiscernibles intermediate in strength
between his strong version and the trivial version first
mentioned. Others have offered alleged counter-
examples to various of the intermediate versions, many
having their origin in Kant’s examples of *incongruent
counterparts. Consider an exactly matching pair of gloves,
suppose the entire universe consists in the left glove facing
the right glove. There are two distinct gloves. But what is
the difference between the two? Consideration of such
alleged counter-examples has yielded insights concerning
the notion of an intrinsic, non-relational property, as well
as the nature of space. r.c.sle.
*identity, criterion of.
I. Kant, Prolegomena to Any Future Metaphysics, tr. Peter G. Lucas
(Manchester, 1953), sect. 13.
G. W. Leibniz, ‘On the Principle of Indiscernibles’, in Leibniz:
Philosophical Writings, ed. and tr. G. H. R. Parkinson and M.
Morris (London, 1973).
identity theory of mind, the. The contemporary
mind–body identity theory, developed in the late 1950s, is
that mental events are (that is, are identical with) physical-
biological processes in the brain. Pain, for example, is
nothing over and above a neural state in the central ner-
vous system, presumably the excitation of certain neurons
(‘nociceptive neurons’) in the brain. Although minds as

substantival entities (e.g. Cartesian mental substances)
have largely disappeared from philosophy, we can formu-
late an identity theory for minds as well: minds are brains
(of appropriate complexity)—or to have a mind is to have
a brain.
The identity theory in this form identifies psychological
types (properties, kinds) with physical types (properties,
kinds). This is why the theory is sometimes called ‘type
*physicalism’. When pain is identified with, say, the exci-
tation of c-fibres, it is pain as a type of event that is being
claimed to be a neural event. The identity can also be put
in terms of properties of events, as follows: the property of
being a pain event is identical with the property of being a
c-fibre stimulation event. Of course, if pain is identical
with a neural event type, individual occurrences of pain
will also be identical with individual events falling under
that neural type.
418 identity
Proponents of the identity theory often invoke consid-
erations of *simplicity (*Ockham’s razor) in its support.
We observe a regular correlation between pain and a cer-
tain neural state, N. Such correlations cry out for an explan-
ation: Why is it that pain is experienced just when N
occurs? Why don’t we experience, say, an itch when N
occurs? But there seems no way to give a more basic
explanation of why the pain–N correlation holds, and we
seem forced to accept it as a brute, unexplainable relation-
ship whereby a mental state ‘dangles’ from a physical
process. However, by identifying pain with N, and other
mental states with their neural substrates, we can, it is

argued, be rid of these ‘nomological danglers’, and
simplify both our ontology and our theory.
Mind–body identity is often likened by its advocates to
certain identities discovered by empirical science such as
‘The temperature of a gas is the mean kinetic energy of its
molecules’, ‘Light is electromagnetic radiation’, and ‘The
gene is the DNA molecule’. Just as scientific research has
uncovered these ‘theoretical identities’, research in neuro-
physiology has shown, goes the argument, that pain is the
excitation of certain neurons, and similarly for other men-
tal states.
Another major argument for the identity theory centres
on considerations of mental causation. That mental
events are sometimes causes and effects of bodily events is
part of deeply entrenched common sense, and it is also a
widely shared assumption of philosophers and working
psycho-logists. It has been notoriously difficult, however,
to explain how *mental causation is possible, as long as
mental phenomena are thought to lie outside the physical
domain. On the identity theory, however, the problem
simply vanishes: there no longer is a special problem
about how one’s desire for a drink of water can cause one’s
limbs to move, since to have a desire for water is for the
brain to be in a certain neural state. On this approach,
then, mental causation turns out to be merely a special
case of physical causation.
An important objection against the identity theory
exploits the *variable realizability (or multiple realizabil-
ity) of mental states. Consider pain: the neural substrate of
pain in humans may be the excitation of c-fibres, but there

is ample reason to doubt that the same neural state sub-
serves pain in all pain-capable organisms (think of octo-
puses). Moreover, there seems no a priori reason to
exclude non-biological systems as psychological systems.
It appears then that pain as a type cannot be identified with
any single physical kind. The best we can do, the objection
goes, is the token-identity theory (or ‘token physicalism’),
which only identifies each instance (or ‘token’) of pain
with an instance of some physical kind, without identify-
ing kinds with kinds. Another major objection to the
identity theory is based on the observation that the
phenomenological features of the mental (e.g. the hurtful-
ness of pain, the visual qualities of an after-image), with
their characteristic subjectivity and privacy, could not be
identical with the neural properties of the brain which are
entirely objective and publicly accessible. j.k.
*materialism; anomalous monism; union theory;
dualism.
D. Armstrong, A Materialist Theory of Mind (London, 1968).
T. Honderich, Mind and Brain (Oxford, 1988).
C. Macdonald, Mind–Body Identity Theories (London, 1989).
U. T. Place, Identifying the Mind: Selected Papers (New York, 2003).
ideology. In its original use, ideology was to be a general
‘science of ideas’, of their elements and relations (Destutt
de Tracy, 1754–1836). Although interest in ideology in this
broad sense has persisted—sometimes with a more a
priori character, sometimes more sociological—perhaps
the most important usage in contemporary philosophy
and politics is narrower and more normative, standing for
a collection of beliefs and values held by an individual or

group for other than purely epistemic reasons, e.g. bour-
geois ideology, nationalist ideology, or gender ideology.
The normative use of the term typically involves two
elements.
First, a particular style of explanation in which the preva-
lence of certain beliefs and values is attributed (to some
significant degree) to a non-epistemic role that they serve
for the individuals who hold them or for society at large.
This role can be specified in terms of the satisfaction of the
non-epistemic interests of certain groups, or in terms of
social–symbolic functions such as stabilization or legit-
imization of the status quo.
Second, a particular style of criticism in which beliefs
and values are called into question precisely by giving this
sort of interest-based or social–symbolic explanation of
their prevalence—an explanation characteristically not
known by the believers themselves.
Thus, Karl Marx and Friedrich Engels argued that the
dominant ideas in any epoch not only reflect the experi-
ence of the dominant class, but also serve its interests.
Dominant ideas do this in part by ‘inverting’ various fea-
tures of social reality—reifying the historically contingent
and class-bound as necessary and universal, or reversing
the role of cause and effect in thinking about economic
activity or ‘human nature’—in ways that make the social
order seem natural, inevitable, or just. More recently,
members of the *Frankfurt School have developed a con-
ception of ideology as a communicative structure system-
atically distorted by power relations; Jürgen Habermas in
particular developed a notion of ‘ideological critique’ that

stresses the failure of certain beliefs and values to with-
stand open, uncoerced, but none the less interest-
involving, intersubjective discussion. (In more orthodox
sociology, Karl Mannheim and others have emphasized
the social function of ideologies in opposing change or
lessening apparent value conflicts.)
A number of important questions arise. What is the
critical force of calling a belief ideological? Perhaps it
is an epistemic defect to hold a belief in part for
reasons ‘hidden’ from oneself and that involve interest
rather than evidence, but perhaps all belief-forming
processes involve unacknowledged causes and (at
least some) non-epistemic interests. Moreover, such
ideology 419