The Oxford Companion to Philosophy Part 55 pdf

they are, why should I not adopt them anyway, without
bothering whether they are God’s?
Another version of the question focuses not on our indi-
vidual lives but on the whole scheme of things: what is the
point of it all? An implication of this, in the spirit if not in
the letter, seems to be that without some overall purpose
in things all our own projects are somehow worthless or
doomed to frustration. But why should that be so? Often
the underlying thought seems to be that real values can
only exist if they are permanent. But why should some-
thing in itself valueless acquire value by being permanent,
or belonging to a set of things which is permanent? The
value of my having just passed my exam and the disvalue
of having painfully stubbed my toe are surely not affected
if the sun will explode in eight billion years and I myself
face annihilation somewhat sooner? Perhaps the thought
is that our projects will fail unless ultimately ‘God is on our
side’. But our short-term projects often succeed. Some-
times events may later make us wish they had not done so,
but this is relatively rare, and often success is definite and
there are no hidden snags.
But now perhaps the question broadens into something
else: what are the conditions for our lives to reach ultimate
success? Many philosophers have held, with Sidgwick,
that ultimately nothing can be of value but certain con-
scious states, for how could values exist without conscious
beings to appreciate them? But recently this inference has
been attacked. No doubt a lifeless desert would lack value
(pace G. E. Moore, who thought that if it was beautiful it
would not), but perhaps the value of at least many con-
scious states presupposes that their owners value other

things; how, for instance, could one see any value in the
state of mind consequent on fulfilling one’s ambition to
climb Everest if one saw no value in having climbed Ever-
est (which is not itself a state of mind)? The question then
becomes: how should we assess these further values? Can
any rational grounds be given for pursuing some of them
rather than others; or one life plan rather than another?
A further, and age-old, question which arises out of this
concerns the value to us of things that happen after our
*deaths, so that we cannot know about them. ‘Call no
man happy until he is dead’ said the Greek sage Solon; but
how can he be happy then? Suppose someone dies after an
apparently happy and successful life, but his achievements
are then shown to be nugatory, for reasons he could not
have anticipated, and his children all come to grief: would
we still call him a happy man, who lived a happy life? If
not, happiness cannot be a state of mind, and even if the
meaning of life is to acquire happiness, it cannot be simply
to acquire a state of mind. a.r.l.
John Cottingham, On the Meaning of Life (London, 2002).
E. D. Klemke (ed.), The Meaning of Life, 2nd edn. (Oxford, 2000).
R. Nozick, The Examined Life (New York, 1989), ch. 10. Like
Wiggins (cited below), criticizes Sidgwick’s outlook, though
without mentioning him.
H. Sidgwick, The Methods of Ethics, 7th edn. (London, 1907), iv. xiv.
D. Wiggins, Needs, Values, Truth (London, 1987), essay iii, esp.
sects. 1–6, 10–15.
life, philosophy of: see abandonment; absurd; Arendt;
Aristippus; Buddhist philosophy; Chinese philosophy; exis-
tentialism; freedom and determinism; Hindu philosophy;

Indian philosophy; Kierkegaard; life, meaning of; Marcel;
Marxist philosophy; moral philosophy, history of; moral
philosophy, problems of; nirvana; pessimism and opti-
mism; Plato; religion, history of philosophy of; religion,
problems of the philosophy of; Schopenhauer; Spinoza.
life, quality of: see quality of life.
life and death. Biological life is best understood as a *fam-
ily resemblance concept, because there is no one property
common to all and only animate things. To be alive is to
have some of: an organic chemistry, a digestive and excre-
tive system, a reproductive capacity, a genetic make-up.
Being alive is a property distinct from being sentient, con-
scious, or intelligent. For example, grass is alive but lacks
those three properties. If there is artificial intelligence,
then some computer is intelligent but not alive. Death is at
least the cessation of life, but it is philosophically contro-
versial what death consists in. Biologically and medically,
death is brain death, or brain-stem death.
Should we fear death? Parfit has suggested that it is not
personal identity that should matter to us, and one’s own
death is not the destruction of a unique Cartesian self but
the ending of a certain series of connected experiences and
actions. Nagel has argued that there is a fundamental
problem about being someone: the unique fact of some-
one’s being oneself resists *reductionism in psychological
theories of personal identity. Should we fear death less if
there is life after death? Bernard Williams argued that
immortality is not desirable, because everlasting life
would become tedious. Perhaps this is not right. Perhaps
the afterlife would be infinitely interesting and fulfilling.

Immortality requires that the person who lives after
death be numerically identical with the person who lives
the earthly life. For example, any resurrected person has
to be oneself rather than some being qualitatively similar
but numerically distinct from oneself. Arguably, this
requires that each of us is a soul that could remain
unchanged by such profound physical and psychological
changes. Arguably, Nagel’s problem of being someone
can be solved only if we are souls.
Why are the living the living? Why are the dead the
dead? These are hard philosophical questions to answer
once it is realized that any life has some duration ending in
death. To be one of the living is to be living now, or to be
actual. To be one of the dead is not to be actual but to have
been actual. It is an unsolved philosophical problem what
actuality is. We do not have an answer to ‘Why is it now
now?’, which means, ‘Why is this particular time the
present?’ s.p.
Thomas Nagel, Mortal Questions (Cambridge, 1979).
—— The View from Nowhere (Oxford, 1986), esp. 223–31.
Derek Parfit, Reasons and Persons (Oxford, 1984).
Sogyal Rinpoche, The Tibetan Book of Living and Dying (San Fran-
cisco, 1992).
520 life, the meaning of
Bernard Williams, Problems of the Self (Cambridge, 1973) esp. ch. 6:
‘The Makropulos Case’.
Palle Yourgrau, ‘The Dead’, Journal of Philosophy, 84 (1987).
life and science. Science describes and explains the world
in terms of causal regularities. Effects simply follow causes
because that is the way the world is ordered, not because it

is better that one thing happens rather than another. The
theories of science are mathematically based abstractions,
prescinding, as far as possible, from much that is important
in the world as lived and experienced by human beings (the
Lebenswelt of the phenomenologists). In the Lebenswelt
descriptions and explanations are irreducibly normative,
coloured by values, feelings, and emotions. To describe a
person as temperate or handsome, or a landscape as beauti-
ful, is to praise them, implying that they are better than if
they had turned out some other way. Reconciling the
value-free theories of science with what we say and think in
the Lebenswelt has troubled philosophers since the time of
Kant. The best hope seems to be to regard neither scientific
nor everyday accounts as exhaustive of the whole of reality,
but both as valid within their own spheres. a.o’h.
*phenomenology; science, history of the philosophy of;
science, problems of the philosophy of.
A. O’Hear, The Element of Fire: Science, Art and the Human World
(London, 1988).
life-world. The universally structured realm of beliefs,
assumptions, feelings, values, and cultural practices that
constitute meaning in everyday life. In criticism of the
classical theory of knowledge (Descartes to Kant), the
concept of the life-world is first introduced as the insur-
mountable basis for scientific experience. Scientific theo-
ries are seen as ‘idealized constructions’ (Husserl),
dependent on immediate sense-perception which itself,
however, is part of the human everyday world that is
taken for granted. Accordingly, the life-world as such is
understood as the unproblematic and pre-scientific pre-

supposition of any understanding and meaning, providing
an implicit *background of once explicitly held or
intended and now ‘sedimented’ beliefs, assumptions, and
practices. Whereas the life-world has first been conceptu-
alized as the world of the subject (Husserl, Schütz), more
recently its genuinely social character has been empha-
sized (Gadamer, Habermas). h h.k.
*Frankfurt School.
J. Habermas, ‘The Concept of the Lifeworld and the Hermeneu-
tic Idealism of Interpretive Sociology’, in Theory of Communica-
tive Action, ii (Boston, 1987).
light of nature. In Cartesian philosophy, the faculty of the
soul by which knowledge, especially a priori knowledge,
is discerned. Descartes’s ‘principles of natural light’ are
put forward as fundamental metaphysical truths; for
example; an (efficient) cause must contain as much reality
as an effect; what can exist from its own power always
exists; if I doubt, then I exist; deception is caused by some
defect; if something happens, it cannot then not have hap-
pened; nothing can cause nothing. If p is a principle of nat-
ural light, then p is self-evident, p is indubitable, and p is a
necessary truth. According to Descartes, there is no fac-
ulty more authoritative than the light of nature which
could be deployed to call its findings into question. s.p.
Philosophical Works of Descartes, tr. E. S. Haldane and G. R. T.
Ross, 2 vols. (Cambridge, 1967).
Anthony Kenny, Descartes: A Study of his Philosophy (New York,
1968).
linguistic acts. Things done with words, an account of
which may cast light on human language and its use.

J. L. Austin believed that the study of *language had
been too much focused on words, and the study of action
too much focused on ‘ordinary physical actions’. His
stated overall project in How to Do Things with Words was
to characterize performatives—utterances on the occa-
sion of which something is done rather than stated. He
wanted to draw a line between performatives and another
sort of utterance which he thought had received all the
attention and at whose expense performatives had previ-
ously been ignored. (*Constatives.) But Austin’s attempt
to draw the line undermined the assumption that there
was a line to be drawn in the first place, and this made way
for the idea that all utterances have a performative dimen-
sion. Thus Austin’s work led to ‘speech-act theory’, a
branch of language studies premissed originally in the
thought that speech is a species of action.
In any use of language—any occasion of someone’s
speaking, that is—there are many things the speaker
does—many linguistic acts she performs. (For example, an
action might be someone’s doing at least these four things:
uttering the words ‘It’s 10 o’clock’, saying what time it is,
reminding Jane that it’s time to go to the lecture, alarming Ted.)
Each linguistic act corresponds to a type of action; and a
principled way of organizing linguistic acts provides a
framework into which the particularities of occasions on
which one or another is done can potentially be fitted so as
to provide for illuminating accounts of speech-actions.
The classification of linguistic acts which Austin got
started may be thought of as a means of imposing system
on to the actual data of linguistic communication.

Austin’s own primary classification was into locutionary
(which incorporates phonetic, phatic, and rhetic), illocu-
tionary, and perlocutionary. Each of these categories sub-
sumes some range of acts; and an action of speaking is
typically a speaker’s performing some act within each
range. Locutionary acts are of saying something; illocu-
tionary acts are acts done in saying something; and per-
locutionary acts are acts done by saying something. (In the
example, uttering the words . . . is a phonetic act, saying that
. . . a locutionary one, reminding . . . (arguably) an illocution-
ary one, and alarming . . . (arguably) a perlocutionary one
(‘arguably’, because Austin in fact had difficulty in making
the illocutionary–perlocutionary distinction clearly).
The idea that speaking a language is engaging in behav-
iour of a rule-governed kind was developed by John Searle
linguistic acts 521
in Speech Acts. He attempted to account for a variety of
phenomena in the setting of an institutional theory of
communication, and to clarify particular speech-acts, e.g.
referring (sometimes called a subsentential speech-act,
because it is done using a word or two rather than a whole
sentence) and promising.
Subsequent work in speech-act theory has been con-
fined to the area that comes under Austin’s illocutionary
head, so that what is usually meant by ‘a speech-act’ is in
the category that Austin called illocutionary. Speech-act
theory may then be thought of as a branch of pragmatics.
It can be divided into two types, depending on the attitude
taken to that which determines a speech-action to be of
the illocutionary act it is of. In the work of such linguists as

John Ross and Jerrold Katz, illocutionary force is absorbed
into a more or less formalized account of locution. In the
work of Searle and others, illocutionary force is a function
of unformalized circumstances. The latter kind of speech-
act theory is more in keeping with its Austinian
beginnings. j.horn.
J. L. Austin, How to Do Things with Words (Oxford, 1961).
John Searle, Speech Acts (Cambridge, 1969).
S. L. Tsohatzidis (ed.), Foundations of Speech Act Theory: Philosoph-
ical and Linguistic Perspectives (London, 1994).
linguistic philosophy. Linguistic philosophy may be
regarded either as a variant form, or as a competitor, of
*analytic philosophy. The latter arose from the early col-
laboration of Wittgenstein and Russell. Linguistic philos-
ophy was more particularly Wittgenstein’s creation,
although it had some elements of affinity with the philo-
sophical practice of G. E. Moore, and, in its later develop-
ment at Oxford with Ryle and Austin, showed some
dependence on the thought of their Oxford predecessors
Cook Wilson and H. A. Prichard.
The central principle of linguistic philosophy is that the
traditional problems of philosophy (or metaphysics) are
not genuine problems at all but confusions generated by
misunderstandings about language or by the misuse of it.
The apparent problems cannot be solved; but they can be
dissolved, confusion can be dispelled. A philosophical puz-
zle is created by an inclination to assert something
absurdly at variance with common sense for what seem
convincing reasons (that we have no knowledge of, or that
there are, no material things, people other than ourselves,

past events, laws of nature).
Moore’s defence of *common sense was direct and prim-
itive. It rested on Thomas Reid’s assumption that the beliefs
of common sense ‘are older and of more authority than the
arguments of philosophy’. Holding out his hand, he said
that he knew for certain that this was a hand and, since a
hand is unquestionably a material thing, it followed that he
knew for certain that there was at least one material thing.
This was more a rhetorical device for showing that philoso-
phers commonly do not mean what they say than a way of
getting to grips with what it is that concerns them.
Wittgenstein’s technique was much more elaborate.
He himself compared it to psychotherapy, in which a kind
of intellectual neurosis is relieved by a long-drawn-out
process of reminding the puzzled philosopher of the way
in which the crucial terms in the expression of his puzzle-
ment are ordinarily used. Wittgenstein’s treatment
resembled psychotherapy not only in its apparently inter-
minable duration but also in its failure to bring about last-
ing cures. But some successes should be acknowledged.
No one will ever now suppose that understanding is a
matter of inward illumination. We tell whether someone
has understood a lesson in long division or French pro-
nunciation by his capacity to do some sums or make the
correct sounds, and so does he, whatever flashes of inward
illumination he may, or may not, have had.
Devotional commentators on Wittgenstein have
argued that there are not two Wittgensteins, but one,
developing a single line of thought. There are, indeed,
common elements in the thinking of the earlier and the

later Wittgenstein. Both are centrally concerned with lan-
guage, both insist that philosophy is not only quite distinct
from science, but that it is an activity rather than a theory
of any kind whatever. But what was formerly seen as ‘the
logical clarification of thoughts’, the revelation by analysis
of the formal structure which is hidden by ordinary
language, is explicitly rejected by the later Wittgenstein
and replaced by an absolutely opposed conception of the
matter.
*Language, on this new view, has no logical essence. It
is an accumulation of a great number of different *‘lan-
guage-games’, of which the reporting or description of
facts is just one. Each of these has its own way of working
and they are no more identical in essential form than ordi-
nary games, being related to one another, as ordinary
games are, only by ‘family resemblance’, an idea on which
Wittgenstein laid much stress. Just as it is not the universal
function of sentences to describe, so it is not the universal
task of the words making up those sentences to name or
refer to objects, concrete or abstract, or to ideas or images
in the minds of their users. The meaning of a word or sen-
tence lies in the rules for its actual use in real life, not philo-
sophical reflection; these rules are best discerned in the
activity of learning how to use the expressions involved;
they are the result of decisions which can be altered; but
these conventions must be public and shared, a *private
language is impossible. That last point is argued for with
something very like a traditional philosophical argument,
and has not been found universally convincing. In the
same spirit, Wittgenstein argues that the elemental truths

Moore thought he could prove in his blunt way are really
background assumptions without whose acceptance
nothing we could recognize as doubt or its settlement
could take place.
It is impossible not to see Wittgenstein’s loosely
affirmed principle that ‘inner processes stand in need of
outward criteria’, in which he claimed some kind of nec-
essary connection between mental states and their behav-
ioural manifestations, as lying behind the less cautious
view of Ryle on the same topic. Ryle proposes something
like a generally applicable pattern of analysis of categorical
522 linguistic acts
linguistics, philosophical relevance of 523
statements about mental events and processes into collec-
tions of hypothetical statements about what those
referred to would do if certain conditions were satisfied.
Ryle held that the familiar dualist conception of *mind and
body as distinct worlds with proprietary kinds of event
going on in them was a large-scale ‘category mistake’, in
which the matters under discussion were treated as
belonging to the wrong logical class, as happens more
obviously in wondering what colour a number is or what
is the weight of a shadow.
Ryle’s preoccupation with thinking paralleled Wittgen-
stein’s with meaning. To think what one is doing is not to
carry out some sequence of bodily movements while con-
sciously rehearsing some appropriate sequence of inner
thoughts. It is to make the bodily movements in an intelli-
gent way, reacting quickly and adequately to obstructions
and difficulties.

The most exquisite of linguistic philosophers was J. L.
Austin, who from 1945 until his death in 1960 exercised a
powerful influence in Oxford, which rapidly faded away
after that. Austin’s acute sensitivity to nuances of meaning
led him to stress that the language we actually use is the
evolutionary by-product of its long and various applica-
tion. Philosophers, he held, persistently over-simplify,
running together words which, although similar, are by
no means identical in meaning: ‘look’ with ‘appear’ and
‘seem’, ‘inadvertently’ with ‘accidentally’ and ‘uninten-
tionally’. Admiration for the refinement and, indeed, cor-
rectness of these distinctions is compatible with doubt
about whether they cut any philosophical ice.
Large claims were made for his identification and nam-
ing of *‘performative’ utterances, such as ‘I promise to pay
you back’ and ‘I name this ship Gladys’, which rather con-
stitute than describe the performance of promising or
naming. His suggestion that ‘I know that so-and-so’ is also
performative as a kind of guarantee of the speaker’s claim
did not survive inspection. Something of the flavour of this
detection of performativeness is present in the account of
truth given by Sir Peter Strawson, ironically enough in a
powerful criticism of Austin’s own attempt to attempt to
rehabilitate the *correspondence theory of truth. To say ‘p
is true’, Strawson held, is not to say something about ‘p’,
such as that it corresponds to the facts, but is at once to
assert it and to confirm its assertion or suggestion by some-
one else. In his later work, from Individuals (1959), he
moved on from linguistic philosophy to a sophisticated
kind of Kantianism, reinforced by the analytic philosophy

of the twentieth century, which aimed to set out the gen-
eral presuppositions of the possibility of articulate dis-
course about our experience. In his earlier phase he had
produced a powerful criticism of the account of reference
embodied in Russell’s theory of *descriptions and he went
on to point out the lavishly Procrustean distortions of the
logical rules of ordinary, natural language made by mod-
ern, mathematically inspired formal logic. a.q.
K. T. Fann (ed.), A Symposium on J. L. Austin (London, 1959).
Anthony Kenny, Wittgenstein (Harmondsworth, 1975).
C. W. K. Mundle, Critique of Linguistic Philosophy (Oxford, 1970).
J. A. Passmore, A Hundred Years of Philosophy (London, 1957),
ch. 18.
O. P. Wood and G. Pitcher (eds.), Ryle: A Collection of Critical
Essays (London, 1970).
linguistics, formal. An empirical discipline which pro-
vides a mathematical framework for characterizing prop-
erties of possible human languages, with different
branches characterizing phenomena at different though
related levels in the speech chain from sound to meaning.
Each branch of linguistics provides a theory that isolates a
unit of linguistic significance, such as property of sound,
form, or meaning, which it analyses and relates to notions
analysed at the levels above and below. The minimal units
of analysis are abstract notions used to segment the con-
tinuous sound signal of human speech into phonemes,
then syllables, then words and morphemes, constituents
and phrases, sentences and discourse structures. Corre-
sponding to these levels we have the theories of phonetics,
phonology, morphology, syntax, semantics, and dis-

course representation theory. The general theory of lan-
guage comprising these subtheories will also contain a
formal treatment of the learnability of languages. b.c.s.
M. Atkinson, D. Kilby, and I. Roca, Foundations of General Linguis-
tics, 2nd edn. (London, 1988).
linguistics, philosophical relevance of. Linguistics bears
on certain issues in *epistemology and in *philosophy of
mind and, of course, is directly relevant to philosophy of
language. For example, Chomsky has forcefully argued
that the exigencies of language learning strongly favour
*rationalism over *empiricism in the traditional debate on
*innate ideas. The facility with which children learn their
native languages, despite the severely limited quantity and
variety of data available to them, indicates that language
acquisition is hardly a matter of stimulus generalization.
Rather, Chomsky proposes, we possess a language faculty
specially equipped for acquiring languages with just those
features that distinguish natural human languages. Char-
acterizing these features is the task of what he calls univer-
sal grammar. Also, he draws a distinction between
competence and performance. Though applied to know-
ledge of language, it is relevant to philosophy of psychol-
ogy in general, for it points to the distinction, crucial to
*cognitive science, between explaining abilities and
explaining behaviour. Chomsky’s account of knowledge
of language suggests that Ryle’s distinction between
knowing-that and knowing-how is not exhaustive, and it
provides an antidote to Wittgenstein’s and Kripke’s scepti-
cism about *rules and rule-following.
Underlying these issues is the question of how gram-

matical information is represented and utilized in lan-
guage production and comprehension. The findings of
psycholinguistics strongly support the claim that gram-
matical information is not an artefact of theory but is psy-
chologically real. Although the categories and principles
of modern linguistic theory generally do not correspond
to those of school grammar, and are not otherwise
intuitively accessible to language-users, people’s linguistic
behaviour and grammatical judgements seem sensitive to
such principles. Accordingly, it is plausible to suppose that
knowledge of language is not conscious but tacit, and that
this knowledge includes representations of sentential
structure corresponding to the categories of linguistic the-
ory (not that there is any consensus on the best such the-
ory). Otherwise, there would seem to be no explanation
for a multitude of linguistic regularities or for the robust-
ness of people’s linguistic intuitions.
Various branches of linguistic theory are relevant to the
philosophy of language. Syntactic theory sheds light on
such concepts as argument structure, binding, scope, and
*logical form. Also, because some semantic information is
encoded structurally rather than lexically, the theory of
meaning (*semantics) in philosophy cannot ignore syntac-
tic theory in linguistics. Linguistic semantics illuminates
such concepts as *ambiguity, *vagueness, and *synonymy,
and offers a framework for explicating the distinction
between *analytic and synthetic statements. Linguistic
*pragmatics overlaps with the philosophical theory of
*speech acts, and, as Grice showed, the distinction
between pragmatic and semantic questions has important

consequences for a variety of philosophical issues.
From a linguistic standpoint, the logical form of a sen-
tence is not the form of a logical formula used to represent
the proposition expressed by the sentence, but a level of
syntactic structure. It is often construed as the input to
semantic interpretation, the stage of a grammar at which
information about the semantic contents of lexical items is
applied to yield, as a function of this level of syntactic struc-
ture, the semantic content of the entire sentence. Logical
form helps explain scope relations among quantified noun
phrases, connectives, and other operators, and binding and
anaphoric relations between noun phrases and pronouns.
Also, there is often reason to impute hidden variables to
the logical forms of sentences containing expressions of
certain sorts, such as relational terms like ‘local’ and ‘alien’,
and temporal adverbs and connectives, like ‘usually’ and
‘whenever’. These latter terms, as they occur in such sen-
tences as ‘Usually Abe drinks wine with dinner’ and
‘Whenever Bob is late for work, he skips breakfast’, func-
tion as quantifiers, and, plausibly, these sentences contain
event variables bound by those quantifiers.
Lexical semanticists distinguish polysemy from lexical
ambiguity. The ambiguity of nouns such as ‘club’, ‘joint’,
and ‘trunk’, each with several unrelated meanings, is a lin-
guistic coincidence, evident from their translation into
other languages. However, polysemous words, such as the
verbs ‘call’, ‘go’, and ‘play’, have closely related meanings
that need to be explained systematically. From a philo-
sophical standpoint, the phenomenon of polysemy chal-
lenges the simplistic view that for a word to have several

literal uses is just a matter of its having several senses. This
is clear from the case of adjectival modification, as illus-
trated by the different relations between adjective and
noun in ‘fast runner’, ‘fast track’, and ‘fast race’, or between
the nouns in ‘child abuse’ and in ‘drug abuse’. k.b.
L. Antony and N. Hornstein (eds.), Chomsky and his Critics
(Oxford, 2003).
N. Chomsky, Knowledge of Language (New York, 1986).
A. George (ed.), Reflections on Chomsky (Oxford, 1989).
T. Parsons, Events in the Semantics of English (Cambridge, Mass.,
1990).
G. Preyer and G. Peter (eds.), Logical Form and Language (Oxford,
2002).
Y. Ravin and C. Leacock, Polysemy: Theoretical and Computational
Approaches (Oxford, 2000).
linguistic turn. Collective designation for a range of oth-
erwise quite disparate trends in twentieth-century
thought. What they all have in common is an appeal to
language, to *discourse, or forms of linguistic representa-
tion as the furthest point that philosophy can reach in its
quest for knowledge and truth. There are no ‘facts’ out-
side language, and no ‘reality’ other than that which pre-
sents itself under some linguistic description. Thus
philosophers can only be deluded if they seek to render
language more accurate or perspicuous by removing its
various natural imperfections—ambiguity, metaphor,
opaque reference, etc.—and achieving a crystalline trans-
parency of logical form. Rather they should follow
Wittgenstein’s example and acknowledge the open multi-
plicity of *‘language-games’ (or cultural *‘forms of life’),

each with its own criteria for what counts as a valid or
meaningful utterance. In short, the proper business of phil-
osophy in this therapeutic mode is to cure language of its
abstract cravings and (in the words of Stanley Cavell) to
‘lead it back, via the community, home’.
The project thus described was pursued most zealously
by J. L. Austin and the proponents of so-called ‘ordinary
language’ philosophy. ‘Our common stock of words’,
Austin wrote, ‘embodies all the distinctions men have
found worth drawing . . . in the lifetimes of many genera-
tions: these are surely likely to be more numerous, more
sound, and more subtle . . . than any that you or I are likely
to think up in our arm-chairs of an afternoon—the most
favoured alternative method’. But the trouble with this
approach, as many have felt, is its tendency to consecrate
the nuances of received (‘common-sense’) wisdom while
failing to address more substantive philosophical issues.
Thus it can easily give rise to an outlook of laissez-faire rel-
ativism or an inert consensus based recommendation that
philosophy should cease asking awkward questions and
be content—in Wittgenstein’s phrase—to ‘leave every-
thing as it is’. c.n.
J. L. Austin, ‘A Plea for Excuses’, in Philosophical Papers (London,
1961).
Richard Rorty (ed.), The Linguistic Turn: Recent Essays in Philosoph-
ical Method (Chicago, 1967).
—— Philosophy and the Mirror of Nature (Oxford, 1980).
Ludwig Wittgenstein, Philosophical Investigations, tr. G. E. M.
Anscombe (Oxford, 1953).
literature and philosophy. Some philosophical writing,

though not very much, it has to be said, displays literary
merit. Plato was a great writer and the British Empiricists
524 linguistics, philosophical relevance of
are admired for the vivacity of their prose. However,
imaginative literature—poetry, drama, and the novel—
has presented problems to philosophers. Plato’s hostility
to art in general is well known. For him, art was a rival in
the pursuit of truth, and liable to corrupt. Plato’s antago-
nism, though shared to an extent by Tolstoy, finds few
modern supporters. But, as usual, Plato raises, albeit
obliquely, profound questions about literature.
There are two major issues. Firstly, there is what is
sometimes described as the problem of belief. If a work of
literature asserts or assumes propositions which I know or
believe to be false, what difference does that make? If I do
not share Milton’s metaphysics, am I debarred to that
extent from an appreciation of Paradise Lost? If I do not
share his beliefs, can I nevertheless empathize with the
poet by ‘suspending my disbelief’ or ‘making believe’ that
these beliefs are true? Certainly I can be moved by *poetry
or prose which proceeds on assumptions which I do not
share. An atheist may find the poetry of George Herbert
moving. Yet there are limits. I may bridle at the anti-
Semitism in Pound’s cantos. I do not suspend my disbelief
in order to enter sympathetically into the world of a racist.
And we certainly will resist literature which tries to sub-
orn us. We cannot take seriously and may even resent fic-
tion which, as we say, verges on propaganda. So although
we can learn from literature we certainly do not com-
monly learn by absorbing maxims; it exemplifies and dis-

plays truths rather than argues for them. Indeed it is not
part of a proper reaction to literature qua literature to
assess the validity of the arguments it contains. Its ‘truth-
fulness’ is, pari passu, not a matter of the truth of the claims
it makes. The famous generalizations about marriage and
the family which begin Pride and Prejudice and Anna Karen-
ina are not exceptionless; it would be easy to find counter-
examples. The truth of literature is generally a matter of
the convincingness of the characters it portrays. Seen in
this context, Plato’s reservations about the arts seem less
strange. If we believe that philosophy can increase our
knowledge by its criticism of superstition and specula-
tion, then philosophy teaches in a way which literature
does not.
What literature does do is to offer us imaginary scenes,
concentrated and complex settings, in which imaginary
beings act. It is sometimes described as inductive. From lit-
erature I may learn about individual human propensities
and peculiarities. In order that it can do this in its own
peculiar fashion, it is necessary that literature move and
involve us, and this raises the second of the two problems
to which I alluded at the beginning. Not only am I moved
by Hamlet’s conversation with the ghost of his father,
even though I do not believe in ghosts, I can be deeply
moved by the death of Anna Karenina, even though I
know that she has no existence outside these pages. But
how can I be moved by the fate of somebody who does not
exist? Is it that she is ‘really dead’ but only in the ‘possible
world’ of Tolstoy’s novel? Am I being irrational? It cannot
be that I am moved by the general truth that there are real

women who escape from a boring life into an ultimately
unsatisfactory affair. I know there are such women but it is
the fate of Anna which moves me and not theirs.
*Imagination is crucial here for I can moved by what I can
imagine and the fact that I can be so moved is an important
factor in planning the course of my life. Literature, we
could say, is important because it nurtures the imagin-
ation in ways which moral maxims or philosophical dis-
course cannot. r.a.s.
*fiction.
David Caute, The Illusion (London, 1971).
John Hospers, ‘Implied Truth in Literature’, Journal of Aesthetics
and Art Criticism (1960).
Peter Lamarque and Stein Haugom Olsen, Truth, Fiction and Lit-
erature (Oxford, 1994).
Colin Radford, ‘How can we be Moved by the Fate of Anna Karen-
ina?’, Proceedings of the Aristotelian Society, supp. vol. (1975).
Morris Weitz, ‘Truth in Literature’, Revue Internationale de
Philosophie (1955).
Locke, John (1632–1704). The foremost English philoso-
pher of the early period of modern post-Cartesian philoso-
phy was educated at Westminster School and Christ
Church, Oxford. Besides studying, and then teaching, sub-
jects such as logic, moral philosophy, rhetoric, and Greek,
he had a deep and abiding interest in medicine. Through
the Earl of Shaftesbury, he became involved in Protestant
politics; this resulted in exile in Holland from 1683 to 1689,
when, after the ‘Glorious Revolution’ which put William
of Orange on the throne, he returned to England and a life
of private study and public service.

He wrote widely—on various branches of philosophy,
on education, economics, religion, and medicine. He is
best known for his Treatises of Government (1690), Letter
Concerning Toleration (1689), and his Essay Concerning
Human Understanding (1690).
The Treatises, which contain Locke’s political philoso-
phy, were composed in the years of the Exclusion Crisis,
during which Locke’s patron, Shaftesbury, and others,
sought to exclude James, then Duke of York, from the suc-
cession to the throne, and argued for government by con-
sent and for the right to religious dissent.
The First Treatise, contains criticism of Robert Filmer’s
theory (Patriarcha (1680) ) of absolute monarchy and the
divine right of kings. Locke found this account of political
*authority, according to which God granted Adam
absolute and total political authority, unworkable. It
could not be used to justify any actual political authority;
we cannot show of any particular ruler that he is one of
Adam’s heirs. In an alternative account the Second Treatise
argues that though subjects do have a duty to God to obey
their ruler, their ruler’s power is not God given or
absolute, and it goes along with duties to his subjects.
Locke’s account begins with the idea of a *state of
nature, in which people live, free from external authority,
in families and loose groups. In this state people have a
duty to God not to ‘harm another in his life . . . liberty, or
. . . goods’ (sect. 6), and so have a corresponding right to
defend against such attack. But these rights and duties
Locke, John 525
may not actually be respected and obeyed; we may lack

the power to defend our rights, or may go too far in our
own defence. For such reasons people agree to ‘enter into
society to make . . . one body politic, under one supreme
government’ (sect. 89). Leaving the state of nature, they
‘set up a judge . . . with authority to determine all . . . con-
troversies and redress . . . injuries’ (sect. 89). But this
authority is not absolute; they are answerable to ‘the will
and determination of the majority’ (sect. 96). Popular con-
sent not only creates, but also produces, the continued
existence of a Lockean political society.
A distinction between tacit and explicit *consent pro-
vides Locke’s answer to the objection that there is no his-
torical evidence for his account of the creation of political
authority, and that people are simply born into civil soci-
eties and come under their laws and authority without
choice. By remaining in society, one gives one’s tacit con-
sent to it. Locke’s suggestion that a person is always ‘at lib-
erty to . . . incorporate himself into any other community,
or . . . to begin a new one’ (sect. 121) is even less plausible
now than then. But his account can be seen as a pic-
turesque way of analysing the structure of legitimate
political authority, and of revealing it to be essentially
based on the consent of the governed. The notion of tacit
consent is given further substance by Locke’s allowing the
possibility of legitimate resistance or *revolution. ‘The
community perpetually retains a supreme power of sav-
ing themselves from . . . their legislators, whenever they
shall be so foolish or so wicked, as to . . . carry on designs
against the[ir] liberties and properties’ (sect. 149).
Locke’s Letter Concerning Toleration, written during

exile in Holland, considers how far the state can legiti-
mately concern itself with religious practices. Convinced
that Christianity as such requires toleration, Locke is
scathing about states which persecute religious dissidents
‘with a pretence of care of the public weal’, and of national
churches which persecute them ‘under pretence of reli-
gion’ (17). He then argues further against religious intoler-
ance by the state by differentiating ‘the business of civil
government from that of religion’ (17). States are consti-
tuted ‘only for the procuring, preserving, and advancing
their [subjects] civil interests’: ‘life, liberty, health …
money, land, houses’ (17); so a ruler’s duty is never ‘to be
extended to the salvation of souls’ (18). This conclusion is
based on three considerations. First, the ‘civil magistrate’
has no more duty than anyone else has to concern himself
with ‘the care of souls’: God has given him no authority ‘to
compel any one to his religion’ (18); nor have his subjects
left it to him to ‘prescribe … what faith or worship [they]
shall embrace’ (18). Second, the means used by a ruler can-
not bring about convictions in his subjects’ minds. Civil
power consists in outward force and earthly penalties, and
these are unable to bring about ‘inward persuasion’ (18):
‘the understanding … cannot be compelled to the belief of
any thing by outward force … [which has no] efficacy as to
make men change the inward judgement that they have
framed of things’ (18). We cannot simply believe what
we are told to believe. Thirdly, even if it were possible,
imposition of a religion by the state would in many cases
not help to save souls. While there is, Locke believed, ‘but
one truth, one way to heaven’, there nevertheless is a

‘variety and contradiction of opinions in religion, wherein
the princes of the world are … divided’ (19); so state inter-
ference in religious matters would mean that ‘men would
owe their eternal happiness or misery to the places of their
nativity’ (19).
But though the state’s sole concern is to protect the civil
interests of its subjects, laws might still properly be
enacted which have consequences for some sects or
churches. Things prejudical to the state and its members
‘ought not to be permitted to the churches in their sacred
rites’ (37); nor are ‘opinions contrary to human society, or
to those moral rules which are necessary to the preserva-
tion of civil society’ (45) to be tolerated. Equally, a church
should not be tolerated by the state if membership
involves allegiance to another earthly power. It is Roman
Catholics whom Locke has in mind in this case, but they
are not the only group to whom toleration is not to be
extended: atheists are similarly untrustworthy. ‘Those are
not at all to be tolerated who deny the being of God.
Promises, covenants, and oaths, which are the bonds of
human society, can have no hold upon an atheist. The tak-
ing away of God … dissolves all’ (47).
Locke resists the suggestion that toleration should be
denied to certain dissenting, non-conformist sects, even
though these had sometimes been ‘nurseries of factions
and seditions’ (47). Their civil unruliness was not rooted in
their very nature, he thought, but had developed precisely
through lack of toleration. ‘If men enter into seditious
conspiracies, it is not religion inspires them to it … but
their suffering and oppressions’ (48–9).

Locke also concerned himself with interdenomina-
tional toleration. What he says turns on his account of the
nature of a church as a society of people who have volun-
tarily come together ‘in order to the public worshipping of
God, in such manner as they judge acceptable to him, and
effectual to the salvation of their souls’ (20). Societies need
rules and regulations, and in the case of a church the
authority to make and apply them resides in members.
Locke explicitly denied that a church requires ‘a bishop, or
presbyter with ruling authority derived from the very
apostles’ (20–1). Christ did not say that churches should
have governments of these kinds; he promised to be pre-
sent simply ‘wheresoever two or three are gathered
together in his name’ (21).
In short, then, ‘the care … of every man’s soul … is to be
left unto himself’ (28). But Locke is not allowing that
someone might enter into private and solitary commu-
nion with God: God should be publicly worshipped, and
people should ‘meet together … to own to the world that
they worship God’ (32). He is not allowing, either, that
there are many, equally good ways to eternal happiness:
there is ‘one only narrow way which leads to heaven’ (30).
But it is not for the state or some national church to pre-
scribe what it is. ‘Those things … every man ought sin-
cerely to inquire into himself’ (29). Even though there is
526 Locke, John
only one narrow way to heaven I cannot be on it unless I
am thoroughly persuaded in my own mind: ‘I cannot be
saved by a religion that I distrust, and by a worship that I
abhor’ (32).

Locke is best known for his Essay Concerning Human
Understanding, which contains his theory of knowledge.
Believing we have been put here by God with some expec-
tation of an afterlife, Locke’s aim is to discover what kind
of things God has fitted us to know, and so how we should
direct and use our intellect and understanding. His pur-
pose is ‘to enquire into the original, certainty, and extent
of human knowledge’ (ii. i. 2).
He maintains from the outset that none of our ideas or
knowledge (whether theoretical or ethical) is innate: the
mind at birth is like ‘white paper’ (ii. i. 2), and all our ideas
are derived from experience. But such experience-based
ideas are only ‘the materials’ of reason and knowledge (ii.
i. 2). Knowledge itself is not ‘made out to us by our senses’
(Draft A, 157). It is a product of reason working out the
connections between those ideas. Locke’s *empiricism
about ideas is combined with a *rationalism about knowl-
edge: without reason all we have is belief, not knowledge.
His claim that all our *ideas, the materials of knowl-
edge, come from experience is facilitated by a distinction
between simple and complex ideas—the former being
unanalysable and indefinable, the latter being mentally
constructible out of simples. Complex ideas are of various
sorts: substances (e.g. gold, lead, horses), which represent
things in the material world; modes (e.g. triangle, grati-
tude), which are ‘dependences on, or affections of sub-
stances’ (ii. xii. 4); and relations (e.g. parent, whiter). He
defends his view that all our ideas derive from experience
by consideration of such cases as ‘space, time, and infinity,
and some few others’ (ii. xii. 8) such as perception, solidity,

memory, number, volition, pure substance in general,
cause and effect, identity. Besides offering these as difficult
test cases, Locke obviously finds them philosophically
interesting too.
Locke’s discussion of ‘pure *substance in general’
(ii. xxiii. 2) became notorious, and there are different
accounts of what he means. Often he is taken to be reject-
ing the kind of view which was later held by Bertrand Rus-
sell, according to which a material thing is no more than ‘a
bundle of properties’. He is often, that is, supposed to hold
that, in addition to properties, things have a ‘substratum’
which ‘supports’ their properties. According to another
interpretation, Locke’s ‘substratum’ should not be seen in
the context of abstract logical questions about the differ-
ence between ‘things’ and ‘properties’. It should be identi-
fied simply with matter as understood by the
‘corpuscularians’ of his century, who revived classical
Greek atomism, or, more specifically, with particular
arrangements of corpuscles of that matter, arrangements
which Locke calls the real essences of material things.
His discussion of identity has been of lasting interest.
There is, he points out, a relativity about identity. ‘Is this
what was here before?’ It depends what kind of thing this is
meant to be. If a mass of matter, it is the same if it consists
of the same particles; if a living body, this need not be so:
‘a colt grown up to a horse . . . is all the while the same . . .
though there may be a manifest change of the parts’
(ii. xxvii. 3). Identity consists here in matter’s being con-
tinuously arranged in a similar way so that it ‘partakes of
the same life’ (ii. xxvii. 4). The point is important for his

distinction, made in connection with *personal identity,
between the idea of ‘man’ and that of ‘person’. A man’s
identity is basically no different from that of any other ani-
mal. But a person is not simply a living body. Identity here
is that of ‘a thinking intelligent being, that has reason, and
reflection, and can consider itself as itself, the same think-
ing thing in different times and places’ (ii. xxvii. 9). Locke’s
description of a person as ‘a thinking intelligent being’
does not mean that the continuity of self-consciousness
which constitutes personal identity is the continuity of
some immaterial substance, which is self-conscious, for he
is clearly unhappy with this view of Descartes’s.
In fact the Essay contains a fair amount of criticism of
Descartes: the identification of extension as the whole
essence of material substance, and the claim that the mind
is always thinking, are further things to which he objects.
Nevertheless, it was that ‘justly-admired gentleman’
Descartes who rescued Locke from the obscurantism (as it
seemed) of the then-prevailing Aristotelian scholasticism
to which he had been exposed as a student (Works, iv. 48).
From Descartes too Locke takes his central and hard-
worked notion of an ‘idea’ (‘whatsoever is the object of
the understanding when a man thinks’ (Essay, i. i. 8) ) as an
essentially mind-dependent thing, rather than a Platonic
entity with a reality of its own quite independent of any
relation it might have to our minds.
Locke refers to the ‘vague and insignificant forms of
speech’ of the scholastics in his ‘Epistle to the Reader’.
They come in for criticism in book iii, ‘On Words’, where
he rejects the idea that classificatory words stand for ‘real

essences’, understood, not as Locke prefers (as corpuscu-
lar constitutions), but as so-called substantial forms
which, by being embodied in things, make them to be of
one sort or another. Instead, he argues that classification is
a matter of human interests and convenience, and that
general words stand for ‘nominal essences’, abstract ideas
which we ourselves construct. Generality and universal-
ity, he says, ‘belong not to the real existence of things; but
are the inventions and creatures of the understanding,
made by it for its own use’ (iii. iii. 11).
In book iv, *knowledge is defined as ‘the perception of
the connection and agreement, or disagreement . . . of any
of our ideas’ (iv. i. 2). Some propositions are true because
the relevant ideas are connected and related in such a way
as to make them true. Any number is even or odd by virtue
of there being a connection between the idea of ‘number’
and those of ‘evenness’ and ‘oddness’. It is by ‘perceiving’
these relations by the light of our reason that we come to
have knowledge. Where either intellectual incapacity or
lack of any actual connection, means we can perceive no
connection, then, ‘though we may fancy, guess, or believe,
yet we always come short of knowledge’ (iv. i. 2).
Locke, John 527
The definition of knowledge as the perception of con-
nections between ideas is ill suited to a third degree of
knowledge, our ‘sensitive knowledge’ of the existence of
things ‘without us’ (iv. ii. 14) which correspond to our ideas.
Moreover, though the certainty of sensitive knowledge is
not so great as that of the other two degrees, it still, Locke
says, deserves the name of knowledge; and he is dismis-

sive of those who might be sceptical about the existence of
an external world. Because of his talk of a correspondence
between external things and our ideas, Locke has usually
been taken to be a representational realist about percep-
tion; but in recent years, and despite his saying that the
mind ‘perceives nothing but its own ideas’ (iv. iv. 3), some
have interpreted him as a direct realist.
Locke’s definition of knowledge seems perfect for our a
priori knowledge in a subject such as geometry, which
deals with modes such as ‘triangle’. But what of our
knowledge in the area of what was known as ‘natural phil-
osophy’, for example, the knowledge that the substance
gold is malleable and graphite not? This is surely based on
observation and experience and not on intellectual per-
ception of any connection between ideas. Locke recog-
nizes such cases where, because there is ‘a want of a
discoverable connection between those ideas which we
have . . . we are . . . left only to observation and experi-
ment’ (iv. iii. 28) and explicitly says that they do not con-
stitute ‘knowledge’, but what he calls belief or opinion.
The contrast between knowledge proper and belief or
opinion is inherited from the scholastics. But, unlike
them, Locke does not think that ‘opinion’ or ‘belief ’ about
the properties and behaviour of substances in the material
world is not worth having. He clearly supports the idea of
a systematic observationally and experimentally based
study of nature, a study of the kind being pursued by his
colleague and friend, the chemist Robert Boyle, whom he
refers to as one of the ‘master builders’ of the ‘common-
wealth of learning’.

The reason why ‘natural philosophy is not capable of
being made a science’ (iv. xii. 10), i.e. into a systematic
body of knowledge as Locke defines it, is that we do not
know the real essences, the corpuscular constitutions, of
the substances with which it is concerned. The fact that we
can perceive no connection between being gold and being
malleable does not mean that there is not one. The prop-
erties of gold depend on or result from its corpuscular con-
stitution, and if we knew just how its corpuscules are
structured and arranged, we would be able to see just why
it has those properties. If, that is to say, our idea of gold
were an idea of its real essence, we might see a connection
between being gold and being malleable.
But this limitation on our knowledge is no cause for
pessimistic concern. The ‘belief’ and ‘opinion’ we have
about the properties of substances in the world are suffi-
cient for daily practicalities. ‘Men have reason to be well
satisfied with what God hath thought fit for them, since he
has given them . . . whatsoever is necessary for the conve-
niences of life’ (i. i. 5).
Unlike ‘natural philosophy’, geometry is a science and
falls on this side of the horizon of our knowledge. This is
because it deals not with substances (e.g. gold, lead) but
with modes (e.g. the triangle), whose real essences we
know. As with a substance such as gold, it is because a
mode such as a triangle is what it is that it has the proper-
ties it has; but whereas in the first case we do not know the
real essence, in the second, Locke says, we do. It is because
it is a figure of three lines enclosing a space that a triangle’s
external angle equals its internal opposites; and because

our idea of a triangle is an idea of that real essence we can
see a connection between being a triangle and having
angles like that.
Our knowledge is bounded by our ideas, and, in gen-
eral, extends only so far as they are ideas of real essences.
But geometrical figures are not the only things whose real
essences we might know. The ideas of morality are modes
too, and Locke thinks that, with proper application, a sys-
tematic science of ethics similar to that of geometry could
be developed.
Yet though moral principles are neither innate nor easy
to acquire by reason, no one need remain ignorant of his
duties and obligations; for the Bible teaches us them too.
This need not mean taking things on authority and aban-
doning all thought of moral knowledge. We can in hind-
sight find rationally justifying arguments for what the
Bible first suggests. Nevertheless, some people, ‘per-
plexed in the necessary affairs of life’ (i. iii. 25) may have no
time for this, and their morality must be a matter of ‘faith’
or ‘belief’.
Locke’s general conclusion concerning the extent of
our knowledge is, then, that not only has God ‘put within
the reach of [our] discovery [beliefs sufficient for] the com-
fortable provision for [this] life’ (i. i. 5), he has also put
within our grasp knowledge of ‘the way that leads to a bet-
ter’. He has given us the means to acquire knowledge of
‘whatsoever is necessary for . . . the information of virtue’
(i. i. 5).
Many of the early reactions to the Essay were critical.
Locke was sometimes supposed to be a sceptic; but

though he does put limits on our ability to know and
understand, he is hardly pessimistic about the human situ-
ation. He explicitly aimed to defeat the despairing idea
‘that either there is no such thing as truth at all; or that
mankind hath no sufficient means to attain a certain
knowledge of it’ (i. i. 2). Nevertheless, his polemic against
innate ideas was taken to imply an impersonal deism, and
his suggestion that matter might think (despite stressing
that ‘all the great ends of morality, and religion, are well
enough secured, without philosophical proofs of the
soul’s immateriality’ (iv. iii. 6) ) was pointed to with hor-
ror. Berkeley, the first great British philosopher after
Locke, reacted against what he saw as the sceptical and
atheistical consequences of Locke’s philosophy.
The framework of Locke’s approach to the human
mind influenced psychology and epistemology for a
long time. David Hartley (1705–57), Joseph Priestley
(1733–1804), Francis Hutcheson (1694–1747), James
Mill (1733–1836), and Étienne Condillac (1715–80) all
528 Locke, John
approached this problem by analysing experience, after
the manner of Locke, into elements and their combina-
tions and associations.
Many of the ideas in the Essay (the stress on observation
and the corpuscular theory of matter, the attack on the
scholastics, the place of reason in religion) can be found in
Locke’s lesser contemporaries too. But he was a powerful
and vigorous spokesman for them. Along with Isaac New-
ton, he became one of the figureheads of the Age of
*Enlightenment. Both then, and in our own time, he is val-

ued for a judicious, sober reasonableness, and an individu-
alistic insistence that opinions are to be weighed carefully
on their merits by each of us, independently of what
others, particularly those in majority or authority, say.
‘Trial and examination must give [truth] price.’ r.s.w.
Richard Ashcraft, Locke’s Two Treatises of Government (Lon-
don, 1987).
Michael Ayers, Locke, 2 vols. (London, 1992).
John Dunn, John Locke (Oxford, 1984).
Nicholas Jolley, Locke: His Philosophical Thought (Oxford, 1999).
John Locke, Draft A of Locke’s Essay Concerning Human Under-
standing (1671), in Drafts for the Essay Concerning Human
Understanding, and other Philosophical Writings, vol. 1, ed. P. H.
Nidditch and G. A. J. Rogers (Oxford, 1990).
—— Two Treatises of Government (London, 1690). The partially
modernized quotations are from Peter Laslett’s critical edn.
(Cambridge, 1960).
—— Essay Concerning Human Understanding (London, 1690). The
partially modernized quotations are from P. H. Nidditch’s
authoritative edn. (Oxford, 1975).
—— Works (London, 1823).
—— A Letter Concerning Toleration (London, 1689). References
are to the edn. by John Horton and Susan Mendus (London,
1991).
R. S. Woolhouse, Locke (Brighton, 1983).
locutions: see linguistic acts.
logic, deontic: see deontic logic.
logic, formal or symbolic: see modern logic; traditional
logic; history of logic; logical theory; calculus; proposi-
tional calculus; propositional logic; sentential calculus;

predicate calculus; modal logic; deontic logic; many-
valued logics; relvance logic; tense logic; higher-
order logic; intuitionist logic; logicism; logical harmony;
metalogic; logical paradoxes; logical notations; Appendix
on logical symbols.
logic, higher-order: see higher-order logic.
logic, history of. Aristotle was the first thinker to devise a
logical system. He drew upon the emphasis on universal
definition found in Socrates, the use of *reductio ad
absurdum in Zeno of Elea, claims about propositional
structure and *negation in Parmenides and Plato, and the
body of argumentative techniques found in legal reason-
ing and geometrical proof. Yet the theory presented in
Aristotle’s five treatises known as the Organon—the
Categories, the De interpretatione, the Prior Analytics, the
Posterior Analytics, and the De sophisticis elenchis—goes far
beyond any of these.
Aristotle holds that a proposition is a complex involving
two terms, a *subject and a predicate, each of which is rep-
resented grammatically with a noun. The logical *form of
a proposition is determined by its quantity (universal or
particular) and by its quality (affirmative or negative).
Aristotle investigates the relations between two propos-
itions containing the same terms in his theories of oppos-
ition (*square of opposition) and conversion. The former
describes relations of *contradictoriness and *contrariety,
the latter *equipollences and *entailments.
The analysis of logical form, opposition, and conver-
sion are combined in syllogistic, Aristotle’s greatest inven-
tion in logic. A *syllogism consists of three propositions.

The first two, the premisses, share exactly one term, and
they logically entail the third proposition, the conclusion,
which contains the two non-shared terms of the pre-
misses. The term common to the two premisses may
occur as subject in one and predicate in the other (called
the ‘first figure’), predicate in both (‘second figure’), or
subject in both (‘third figure’). A given configuration of
premisses and conclusions is called a ‘mood’.
In the scholastic period, mnemonic names for the valid
moods canvassed in the Prior Analytics were devised. Two
first-figure valid moods were considered perfect and not in
need of any further validation: *Barbara (consisting
entirely in universal affirmatives) and *Celarent (consist-
ing in a universal negative and a universal affirmative,
concluding in a universal negative). For the validation of
the rest, Aristotle used three techniques: reduction, where
a given mood is transformed through conversions into
Barbara or Celarent; reductio ad absurdum; and ekthesis,
which proceeds by selection of an arbitrary individual. He
regularly describes moods by using variables in place of
terms. To reject a proposed inference he typically gives a
list of terms that, when substituted as values of the term-
variables, produce true premisses and false conclusion.
This is similar to the modern technique of constructing
‘counter-arguments’ to establish invalidity.
Aristotle may also be credited with the formulation of
several metalogical theses, most notably the law of *non-
contradiction, the principle of *excluded middle, and the
law of *bivalence. These are important in his discussions
of *modal logic and tense logic. Aristotle referred to cer-

tain principles of propositional logic and to reasoning
involving hypothetical propositions. He also created two
non-formal logical theories: techniques and strategies for
devising arguments (in the Topics), and a theory of fallacies
(in the De sophisticis elenchis). Aristotle’s pupils Eudemus
and Theophrastus modified and developed Aristotelian
logic in several ways.
The next major innovations in logic are due to the
Megarian–Stoic School. They developed an alternative
account of the syllogism, and, in the course of so doing,
elaborated a full *propositional logic which complements
Aristotelian term logic. There are fragmentary records
of debates over the *truth-conditions for various
logic, history of 529