Inquiry, 45, 59–80

The Happy Truth: J. L. Austin’s How To

Do Things With Words
Alice Crary
The New School

This article aims to disrupt received views about the significanc e of J. L. Austin’s
contributio n to philosophy of language. Its focus is Austin’s 1955 lectures How To
Do Things With Words. Commentators on the lectures in both philosophica l and
literary-theoretica l circles, despite conspicuou s differences , tend to agree in
attributing to Austin an assumption about the relation between literal meaning and
truth, which is in fact his central critical target. The goal of the article is to correct
this misunderstandin g and to show that Austin is deeply critical of a picture of
correspondenc e between language and the world which nearly half a century after he
delivered his lectures continue s to structure philosophica l discussions of language.

And forget, for once and for a while, that ... curious question
‘Is it true?’ May we?1

In real life, as opposed to the simple situation s envisaged in

logical theory, one cannot always answer in a simple
manner whether [a statement] is true or false.2

I. Introduction
Over the last couple of decades, J. L. Austin’s work has slid into a state of
respectable semi-obscurity. While some bits of philosophical terminology he
introduced remain in circulation (e.g. his talk of ‘constative’ and
‘performative’ utterances and his system of classifying utterances according
to their ‘illocutionary forces’), even these bits tend to be regarded, not as
expressing groundbreaking perspectives, but rather as accommodating
themselves comfortably within dominant research programs. My aim in this
article is to raise a question about this image of Austin as a philosophically
relatively innocuous – if also somewhat eccentric and engaging – Ž gure. I will
be concerned above all with his 1955 lectures How To Do Things With Words.
What I hope to show is (i) that one of Austin’s larger goals here is to criticize
an ideal of the ‘statement’ as always in some way reporting on the world or
imparting information about the facts either truly or falsely3 and also (ii) that,
in criticizing this ideal, Austin is developing a powerful attack on a picture of
correspondence between language and the world that continues, nearly half a
century after he delivered his lectures, to structure philosophical discussions
of language.
The picture of correspondence between language and the facts that Austin

# 2002 Taylor & Francis

60 Alice Crary

takes to be implicit in a traditional ideal of the ‘statement’ is one on which the

business of corresponding to the facts is the prerogative of what might be
called bi-polar ‘statements’ or propositions, i.e. ‘statements’ or propositions
that always describe states of affairs either truly or falsely.4 In his lectures,
Austin addresses himself to a philosophical audience whose members for the
most part regard this picture as containing an uncontroversial (or even
unquestionable) representation of what it is for language to correspond to the
facts.5 His aim is to discredit the idea that to talk about bits of language
having a bearing on the world just is to talk about ‘statements’ or propositions
that are bi-polar in this sense. He proceeds by arguing that this idea is
nourished by a view of meaning on which sentences possess what are
sometimes called literal meanings (i.e. meanings they carry with them into
every context of use) and by arguing that this view fails to withstand critical
scrutiny.
In reading Austin as attacking a familiar idea of literal sentence-meaning, I
part company with received philosophical interpretations of his lectures.
Philosophical commentators typically assume that Austin takes the bi-polar
‘statement’ – or some other conception of the ‘statement’ that likewise
presupposes the idea of literal meaning – to contain an accurate representation
of what it is for language to have a bearing on the world. Thus, e.g., in his
earliest writings on Austin, John Searle claims both that Austin’s study of
speech acts constitutes a critique of traditional philosophica l views of the
‘statement’ and also that it is nevertheless possible to assimilate his critique
within the framework of the bi-polar ‘statement’.6 In slightly more recent
writings, Searle no longer insists that Austin’s doctrine of speech acts is best
understood as consistent with bi-polarity. But he continues to maintain that it
should be understood as taking for granted a notion of literal meaning.7
The assumption that Austin takes an ideal of the literal ‘statement’ to
capture the notion of correspondence between language and the world also
enters into discussions of his work within literary-theoretical circles, albeit in
a very different manner. There has been a marked growth of interest in
Austin’s lectures within such circles since Jacques Derrida claimed to Ž nd
signiŽ cant afŽ nities between Austin’s view of language and his own
deconstructivist view.8 Theorists who sympathize with Derrida in this matter
typically take Austin to be agreeing with representatives of deconstruction in
expressing skepticism about the very idea of a fully objective relationship
between language and the world. Further, they typically take him to be
deriving his alleged skepticism from an attack on the idea of literal sentence-
meaning. The upshot is a picture of Austin as moving from the denial that
sentences have literal meanings to the conclusion that no bit of language can
describe how things are in a thoroughly objective manner.9
Although literary-theoretical commentators thus tend to disagree with
philosophers like Searle, who portray Austin as preserving the idea of literal
 The Happy Truth 61

sentence-meanings, their disagreement takes place against the background of

a philosophically signiŽ cant form of agreement. These commentators by and
large read Austin as claiming that there can be no such thing as literal
sentence-meaning and that therefore we must acknowledge that no mode of
discourse can have a whole-heartedly objective bearing on the world. They
arrive at the conclusion that Austin rejects the possibility of objective
correspondence because they take him to be maintaining that there is an
important sense in which an ideal of the literal ‘statement’ does justice to
what such correspondence amounts to. They thus in effect agree with Searle
and others in thinking that Austin is properly read as assuming that the literal
‘statement’ contains a faithful representation of what it would be for language
to have an objective bearing on the world.
My goal in what follows is to show that this assumption, far from informing
Austin’s work, is one of his main targets, and that we need to discard it in
order to locate the central, philosophically most provocative line of thought in
his lectures. Austin, as I read him, is arguing that a traditional philosophical
ideal of the (literal, bi-polar) ‘statement’ drives the development of an
inaccurate picture of the relation between language and the world. Since the
picture he is criticizing still dominates research in philosophy of language,
there is an important respect in which his argument retains its original
interest.

II. Constatives and Performatives

Austin opens his lectures with a discussion of a traditional philosophica l ideal
of the ‘statement’ on which, in his words, ‘the business of a “statement” can
only be to “describe” some state of affairs, or to “state some fact”, which it
must do either truly or falsely’ (p. 1). His initial discussion of this ideal takes
the form of a sketch of ways in which he thinks it controls philosophical
investigations of language. He mentions the writings of philosophers who
treat the ‘descriptive statement’ as monopolizing the serious business of
language. Their work has, he claims, been frustrated by the continual
discovery of combinations of words which have the grammatical form of
statements but which cannot plausibly be regarded as describing states of
affairs or stating facts. He points out that, while some philosophers who are
committed to upholding this ideal of the ‘statement’ have insisted upon
regarding certain allegedly aberrant linguistic constructions as nonsensical
‘pseudo-statements’ (here he evidently has in mind various logical
positivists), yet others have felt uncomfortable about an image of us as
speaking a lot of nonsense. This lack of comfort is, he claims, partly
responsible both for the resurgence of philosophica l interest in bits of
language which have the grammatical form of statements but do not ‘record
62 Alice Crary

or impart straightforward information about the facts’ and also for the fact
that some philosophers have become concerned with asking ‘whether many
apparent pseudo-statements really set out to be statements at all’ (p. 2). In this
connection, Austin touches on the efforts of philosophers who take an interest
in utterances which are not evidently nonsensical, but which do not state facts
or describe states of affairs. He mentions the work of those emotivists who
maintain that ‘ethical propositions’ are perhaps intended’, not to report on
features of the world, but rather ‘to evince emotion or to prescribe conduct or
to in uence it in special ways’ (pp. 2–3).
Austin moves from surveying ways in which philosophica l research into
language has been structured by an ideal of the ‘descriptive statement’ to
outlining a new system for classifying utterances – a system which, like those
proposed within earlier research, leaves the ideal in place. He draws attention
to utterances which look like ‘statements’ but which do not meet the
speciŽ cations of the traditional ideal. He proposes the term ‘performative’ for
utterances which are not properly characterized as describing a state of affairs
and in which the uttering of a sentence (e.g. ‘I promise’) is the doing of an
action (e.g. making a promise) which would not normally be described as
saying something (p. 5). And he proposes the term ‘constative’ for utterances
which, in so far as they describe a state of affairs or state a fact (e.g. ‘That dog
is dangerous’), appear to attain to the traditional ideal of the ‘statement’.
Although the classical ‘statement’ – or, as he now prefers to put it, the
constative utterance – can be judged true or false depending on whether it
accurately or inaccurately describes a state of affairs or states a fact, the
performative utterance is not appropriately assessed in these terms.
It merits emphasis that this initial classiŽ cation of utterances into
constatives and performatives takes for granted the ideal of the ‘statement’
that Austin highlights in the opening moments of his lectures. This is
important because, later in his lectures, Austin will declare that he Ž nds this
ideal deeply problematic. Indeed, even at this early stage, he says that he
wants to question ‘an age-old assumption in philosophy – the assumption that
to say something, at least in all cases worth considering, i.e. all cases
considered, is always and simply to state something’ (p. 12).10 Now, it may
appear that Austin, in isolating a class of utterances that do not meet the
speciŽ cations of the traditional ‘statement’, has already washed his hands of
the assumption that ‘to say something is always to state something’.
Similarly, it may appear that the emotivist, in describing a set of utterances
which are best understood as involving the expression of (non-cognitive)
feeling, has broken with this assumption. Austin, however, believes that
genuine liberation from the assumption requires more than the creation of
new categories of utterances to coexist alongside the traditional ‘statement’.
At various points in his Ž rst several lectures, he suggests that philosophers’
preoccupation with the category of the ‘statement’ has its roots in a confused
 The Happy Truth 63

view of meaning – a view which makes it appear possible to classify linguistic

formulae into those for use in making ‘statements’ and those for use in
producing other kinds of utterances. Further, he attempts to distance himself,
not only from this view of meaning, but also from the idea, suggested by the
view, that there are at least some linguistic formulae that are perfectly suited
for making ‘statements’. What emerges, as we will see, is that Austin believes
there is something philosophicall y suspicious about classiŽ catory systems
(such as his own talk of constative and performative utterances and the
emotivist’s talk of factually signiŽ cant statements and ‘non-cognitive’
utterances) which depend for their integrity on the Ž xity of the category of
the ‘statement’. Hence, when, at an early juncture in his lectures, he declares
his intention to dislodge an ‘age-old assumption’ to the effect that ‘to say
something is always and simply to state something’, he is hinting at
something that he will make explicit later in his lectures, viz., that he thinks
his original distinction between constative and performative utterances
cannot, in the Ž nal analysis, be preserved.
In his second lecture, Austin studies the utterances he has provisionally
identiŽ ed as performatives. He considers various dimensions in which we
assess these utterances – dimensions which he refers to as those of
‘happiness’ and ‘unhappiness’. His inventory of types of unhappiness – or
‘infelicities’ – is a list of ways in which the circumstances of performative
utterances can fail to be appropriate. Very generally, Austin tells us that in
order for a performative to be ‘happy’, the circumstances in which a person
utters a set of words must satisfy the conditions of a conventional procedure
having a certain conventional effect – a procedure which involves the uttering
of those words by a person of a certain standing in a certain situation. (Thus,
e.g., in order for my utterance of the sentence ‘I name this yacht The Midnight
Special’ to be happy, there must be an established procedure for naming
yachts, and I must be an appropriate person in appropriate circumstances for
performing that procedure.) A performative will be ‘unhappy’ in some way if
one or more of these conditions fails to be met.
Austin divides his list of infelicities into two categories: (i) ‘misŽ res’,
infelicities which result in the relevant act – i.e. the act for the performing of
which the verbal formula in question is issued – not being performed (e.g. I
utter the formula used for marrying two persons but I say it to two
dachshunds, so that the act in question, marrying, does not come off at all);
and (ii) ‘abuses’, infelicities which result, not in the non-performance of the
act in question, but rather in its being performed in something less than an
exemplary fashion (e.g. when I say to someone ‘I promise to ...’ without
having any intention of fulŽ lling my promise, I do, in spite of my bad
intentions, none the less make a promise).
Austin claims that his discussion of infelicities shows that the performative,
although it is not per se ever true or false, ‘still [is] subject to criticism’ (p.
64 Alice Crary

25). He takes himself to have demonstrated that we can critically assess

performative utterances by appealing to the circumstances that must obtain
for them to be happy.
Austin also claims to have shown that it is not always possible neatly to sort
a particular utterance either into the class of constatives or into the class of
performatives. He mentions several utterances which, although they qualify
as perfomatives, nonetheless resemble constatives in having a bearing on the
facts. He considers, e.g., a case in which a person comes out with the sentence
‘I warn you that that bull is about to charge’ in a situation in which the
relevant bull is not on the verge of charging. Austin argues that while this
utterance is a performative (since the person who comes out with it is
performing the act of warning) it fails to do justice to the facts in the same
way a statement might. What this person says is ‘false or (better) mistaken, as
with a statement’ (p. 55).
Austin casts further doubt on the prospects for a neat classiŽ cation of
performatives and constatives when, after having considered the infelicities
of the performative, he directs attention towards ways in which even
utterances that meet the speciŽ cations of constatives can be unhappy –
towards ‘ways we can do wrong, speak outrageously, in uttering conjunctions
of “factual” statements’ (p. 47). Here Austin is not merely concerned with the
fact that, for a given factual statement (or set of factual statements), there are
situations in which uttering it would be inappropriate or bizarre (e.g. at a
party, I say to a group of my friends, ‘so-and-so does not beat his children’,
when none of us has any grounds for even suspecting that so-and-so might
beat his children); he is not merely concerned with the fact that putative
constative utterances can suffer from the sorts of infelicities he calls ‘abuses’.
He hopes to show that even a ‘factual statement’ (e.g. ‘John’s children are all
bald’) can ‘misŽ re’ – can turn out to be ‘void’ – if it is issued in sufŽ ciently
untoward circumstances (perhaps John has no children) (pp. 50–51). His
thought is that the conditions of a conventional procedure need to be satisŽ ed
for the constative utterance to be constituted as the utterance it is – and hence
that the failure of certain conditions to be satisŽ ed results, not merely in the
utterance’s not being fully appropriate but, further, in it not being fully
determinate which utterance it is.11
At this point, Austin takes himself to have shown both that ‘considerations
of the happiness and unhappiness type may infect statements (or some
statements)’ and also that ‘considerations of the type of truth and falsity may
infect performatives (or some performatives)’ (p. 55). This leads him to
speculate that ‘[p]erhaps indeed there is no great distinction between
statements and performative utterances’ (p. 52). But he does not give up hope
of ‘some precise way in which we can deŽ nitely distinguish the performative
from the constative utterance’ (p. 55). Instead he searches for a grammatical
criterion which will clearly isolate the performative.
 The Happy Truth 65

Austin notes that the performatives he has thus far considered (e.g. ‘I name
this ship ...’, ‘I give and bequeath this watch ...’, ‘I bet you sixpence that ...’,
etc.) all have verbs in the Ž rst person singular present indicative active. He
accordingly asks whether this mood and tense are distinctive of the
performative. What he discovers is that there are performatives that do not
observe this rule. There are verbal formulae in the Ž rst person plural (e.g. ‘We
promise ...’), in the second person (e.g. ‘You are hereby authorized to pay ...’),
and in the third person (e.g. ‘Passengers are warned ...’) which are sometimes
issued with performative force (p. 57). He concludes that ‘mood and tense ...
break down as absolute criteria’ for the isolation of the performative (p. 58).
Austin now considers another suggestion for isolating the performative.
The performative might be distinguished by a special vocabulary. This
suggestion, however, is undermined by the observation that for any candidate
pair of performative and ‘performative words’ it is possible to produce the
performative without the words, or to use the words without producing the
performative (p. 59).
These failures lead Austin to ask whether a more complex criterion – one
‘involving both grammar and vocabulary’ (p. 59) – is required, and to propose
such a criterion for ‘explicit performatives’. The mark of an explicit
performative is that it makes explicit the agent who is performing the action in
question and also that its verb or verb phrase makes explicit what action is
being performed on the occasion of utterance (e.g. ‘I promise to pay you
tomorrow’). Austin considers the suggestion that we can demarcate
performative language by counting as performatives utterances which are
‘reducible, expandable, or analysable’ into this explicit form (pp. 60–61). But
the suggestion encounters intractable difŽ culties. Some verbal formulae (e.g.
‘I approve’) seem sometimes to be explicit performatives (e.g. ‘I approve’
when it has the performative force of giving approval) and sometimes to be
descriptives (e.g. ‘I approve’ when it means ‘I favor this’). Austin concludes
that ‘it is often not easy to be sure that, even when it is apparently in explicit
form, an utterance is performative or that it is not’ (p. 91), and he abandons his
attempt to Ž nd a precise method of distinguishing constatives and
performatives.
It seems clear that Austin’s narrative about his failure to Ž nd a grammatical
criterion for distinguishing constative and performative utterances is intended
to establish that there is something confused in principle about the idea of
such a criterion. Recall his discussion of how constative utterances can be
infelicitous and can even ‘misŽ re’. Here Austin is attacking the idea that we
can isolate constative combinations of words in advance of a consideration of
ways in which those combinations of words are used on particular occasions.
He challenges our tendency to regard the contexts in which strings of words
are used as essentially irrelevant – or ‘merely pragmatic’. This relatively
dismissive treatment of context appears justiŽ ed where the ‘statement’ is
66 Alice Crary

understood as perfectly suited (i.e. without regard to the circumstances in

which it is made, the purpose for which it is made, or the audience to which it
is made) for describing a state of affairs either truly or falsely. Implicit in this
understanding of the ‘statement’ is a picture of the sense of a sentence as Ž xed
independently of its being used to say something to someone on a particular
occasion. On this picture, the assignment of meaning to a combination of
words is accomplished in a manner which is insulated from mishaps that
would render it ‘void’. The picture makes room for the idea of linguistic
constructions which are ‘purely descriptive (or constative)’ in that they
describe a bit of the world or make a factual claim about it in any context of
use. Austin is accordingly putting critical pressure on the picture when he
mentions cases in which putative constative utterances ‘misŽ re’ so that we
cannot isolate the ‘proposition’ which (according to the picture) is invariably
expressed by a proper ‘statement’.12
This illuminates the signiŽ cance of Austin’s practice of placing ‘statement’
in single quotes when he talks about a traditional philosophical ideal of the
descriptive ‘statement’ (a practice I have adopted in this article).13 Since
Austin rejects the idea that it is possible to determine the meaning of verbal
formulae apart from a consideration of the circumstances of their use, he
thinks it is important clearly to distinguish merely grammatical entities from
the use of such entities to say things in particular contexts. He proposes to use
the term sentence strictly for a class of merely grammatical entities and to use
the term statement (without single quotes) for utterances of sentences in
particular situations.14 In putting single quotes around ‘statement’ when he is
discussing a familiar philosophical ‘ideal of the statement’, Austin indicates
that he thinks a particular linguistic ideal, in so far as it depicts meanings as
belonging to isolated combinations of words, is properly understood as
containing, not (as it tends to be portrayed) an account of the workings of
statements, but rather an account of the workings of sentences.15
Here Austin repudiates the idea that we can somehow identify sentences as
such (i.e. considered apart from the circumstances of their use) as either
‘constative’ or ‘performative’ and distances himself from the possibility of a
criterion for distinguishing performative and constative language that could
be spelled out in terms of grammar. The search for a grammatical criterion for
distinguishing constatives and performatives, as he sees it, is thus doomed to
failure from the start.
It would, however, be a mistake to conclude that Austin wants to prevent us
from drawing distinctions between different ways in which language
functions. Austin is discouraging philosophers from studying the workings
of language by looking at isolated sentences. But he thinks that we can
productively study language if we take as our object what he regards as its
minimal units: statements (without single quotes) or complete acts of speech.
Austin represents his own investigations of language as exclusively
 The Happy Truth 67

concerned with language in this sense, with speech acts. He writes that ‘[t]he
total speech act in the total speech situation is the only actual phenomenon
which, in the last resort, we are engaged in elucidating’ (p. 147). And he
signals that he thinks we can classify bits of language with regard to the kind
of speech act that they are – with regard, as he will put it, to their
‘illocutionary forces’.16

III. The ‘Doctrine of Illocutionary Forces’

Austin begins his discussion of the illocutionary act by offering the following
gloss. The illocutionary act is, he writes, the ‘performance of an act in saying
something as opposed to [the] performance of an act of saying something’ (p.
99). He declares that one of his main aims in his lectures is to distinguish the
illocutionary act both from what he calls the ‘locutionary act’, the act of
saying something ‘with a certain sense and a certain reference’, and also from
what he calls the ‘perlocutionary act’, the act of producing certain
consequential effects by saying something (p. 103).
The distinction between locutionary and illocutionary acts is among the
more controversial points of Austin’s lectures.17 Austin himself seems to
invite debate on this topic by expressing reservations at several points about
the adequacy of his own conception of the locutionary act. He says, e.g., that
he is inclined to think that ‘a great many further reŽ nements would be
possible and necessary if we were to discuss [the locutionary act] for its own
sake’ (p. 95); and at one point he even says that a distinction between
locutionary acts and illocutionary acts ‘is only adumbrated’ in his lectures and
that he himself is not sure ‘if this distinction is sound’ (p. 148, emphasis in the
original). It would, however, be wrong to take the moments at which Austin
voices a lack of conŽ dence in his conception of the locutionary act as a sign
that he suspects the things he says about the locutionary act, taken as a whole,
are  awed. This becomes clear once we locate his worries about his account
of the locutionary act within the architectonic of the lectures.
The main point of Austin’s treatment of the locutionary act is – not to
isolate the locutionary act itself, but rather – to avert what he regards as a
natural misunderstanding of the illocutionary act. He wants to make it clear
that when he talks about the illocutionary forces of utterances he is not talking
about ways in which grammatically correct and meaningful sentences – or
‘locutions’ – are used. He says he believes that the ‘descriptive fallacy’ (i.e.
the fallacy of thinking that the proper business of all ‘statements’ is to
describe a state of affairs truly or falsely) typically arises when questions
about all different sorts of features of language, including the illocutionary
forces of utterances, are conceived as questions of ‘locutionary usage’ (pp.
99–100).
68 Alice Crary

When Austin speaks of ‘locutions’ (or ‘locutionary usage’) in thus

anticipating a misunderstanding of his notion of the illocutionary act, he is not
employing the term ‘locution’ in a sense that he himself ultimately wants to
endorse. His thought is that if we think of questions about illocutionary force
as questions about the manner in which particular ‘locutions’ are used –
where ‘locutions’ are understood as grammatically correct and meaningful
sentences – then, in speaking about the illocutionary force of a particular
utterance, we will be at least tacitly assuming that we can make out what
‘locution’ is at issue without Ž rst grasping how, in the situation at hand, the
speaker is using a set of words to say something to someone. He worries that
if we conceive of the meaning of a combination of words in this way as
somehow Ž xed independently of its being used to say something to someone
on a particular occasion – so that it appears possible to tell what ‘locution’ is
at issue even if we cannot as it were specify the ‘illocutionary force’ with
which it is intended – then it will seem to make sense to think that some
sentences are (on account of the ‘locutionary acts’ we imagine are invariably
performed when they are uttered) ideally suited for making factual claims.
We will be inclined to embrace a version of the view of meaning, underlying a
traditional philosophical ideal of the ‘statement’, that he attacks in his
discussion of constative and performative utterances.
To the extent that Austin endorses a conception of the locutionary act at all,
he endorses a conception of it according to which ‘[t]o perform a locutionary
act is in general ... also and eo ipso to perform an illocutionary act’ (p. 98,
emphasis in the original). Austin stresses that he thinks that whenever I say
anything (except things like ‘ouch’ and ‘damn’) I perform both a locutionary
act (i.e. an act of saying something with a certain meaning) and an
illocutionary act (i.e. an act of saying something with a certain force). (See,
e.g., pp. 113 and 132.) He is drawn towards this view by the thought that there
is no such thing as identifying the meaning of a combination of words (or: no
such thing as identifying the ‘locutionary act’ performed when a combination
of words is uttered) independently of an appreciation of how those words are
being used to say something to someone on a particular occasion (or:
independently of an appreciation of their ‘illocutionary force’). Our ability to
say what a bit of language means, as he sees it, is parasitic on our
understanding of it as a complete act of speech or illocutionary act. His
refusal to regard locutionary and illocutionary acts as separate kinds of acts
thus stems from the view that it is only in so far as the circumstances in which
a combination of words is uttered satisfy the conditions of a certain
conventional procedure – the conditions for performing a certain illocu-
tionary act – that it is constituted as the (meaningful) utterance that it is.18
Let me offer one illustration of the kinds of considerations that make
Austin suspicious of the idea that sentences have ‘literal’ or ‘conventional’
meanings that they carry with them into every context of their use.19 Consider
 The Happy Truth 69

the sentence ‘The car is red.’ If any sentence has a clear ‘literal meaning’, this
sentence seems like a good candidate. It appears to have a constant and
unvarying descriptive meaning – one which its indexical elements (along with
contextual clues) will pin down to the particular contexts in which it is used.
Imagine the following situation:
Two people, A and B are sitting at a table that has on it a checker set and
a small green toy car. A picks up the green car and says: ‘The car is red.’
B is confused. Nonetheless, because she takes herself to understand
(what she thinks of as) the ‘literal meaning’ of A’s sentence, she thinks
she can determine (by appealing to this ‘literal meaning’ along with a
few contextual clues) what factual statement A is making. Suppose that B
now responds: ‘Of course it’s not red. It’s green.’ A laughs and rejoins:
‘That’s not what I meant. This checker set is missing one of its red
pieces. I was suggesting that we use this little car here as one of the red
pieces!’
The correct lesson to draw from this example is not that the sentence ‘The car
is red’ as A uses it does not have the particular ‘literal meaning’ that B takes it
to have. The example leaves open the question of whether the sentence as A
uses it is in the end best understood as having what B takes to be its ‘literal
meaning’. Indeed it would be in keeping with the point of the example to
append to it a scenario in which B, after she has understood what A is using
the sentence ‘The car is red’ to say, tries to Ž gure out whether or not the
sentence as A uses it has (what B thinks of as) its ‘literal meaning’. Which
decision B ultimately reaches is of no importance. What the example teaches
is that it is only once B has understood how A’s sentence is to be taken that she
is in a position to make such a decision at all. Since it is only once B has
understood how A’s sentence is to be taken that she is in a position to say what
the sentence as A uses it means, it is only at this point that she is in a position
to consider what similarities and differences obtain between the meaning it
has in this context and (what she takes to be) its ‘literal meaning’. So it is not
in virtue of her grasp of (what she thinks of as) the ‘literal meaning’ of A’s
sentence that B understands what A is saying. This means that B’s appeal to
the ‘literal meaning’ of A’s sentence does not here play the role that it is often
called on to play in philosophy.
Austin’s account of locutionary and illocutionary acts, taken as a whole,
brings into question the idea that we might develop a theory that could be
used to identify the locutionary acts performed whenever particular sentences
are used. But his strategy for combating this idea involves allowing himself at
Ž rst to entertain it. After initially elaborating his (suspect) notion of the
locutionary act, he criticizes it by rejecting as  awed an idea that it
presupposes, viz., that it is possible to isolate the locutionary act that is
performed when a particular sentence is employed in the absence of a grasp of
70 Alice Crary

the illocutionary force with which it is being used. Austin’s misgivings about
the viability of his own notion of the locutionary act are best understood as
directed towards his initial description of it and should not be interpreted as a
sign that he lacks conŽ dence in his overall treatment.
After contrasting the illocutionary act with the locutionary act, Austin goes
on to contrast it with the perlocutionary act. He distinguishes the sorts of
conventional effects that belong to the illocutionary act (e.g. my saying
something which has the illocutionary force of a promise has the effect of my
being committed by my promise [p. 102]) from the sorts of consequential
effects that uttering a string of words may have (e.g. your saying ‘Look what
you’re doing!’ may have the effect of my stopping my work and looking to
see whether I have perhaps spilled my coffee or put my elbow in the jam). He
calls the production of the latter kind of effects a perlocutionary act (p. 101)
and says he wants to keep us from confusing illocutionary acts with
perlocutionary ones (p. 103). In urging us not to understand his remarks about
illocutionary force as remarks about perlocutions, Austin is, once again,
guarding against the picture of meaning at play in a traditional philosophical
ideal of the ‘statement’. What interests him is the fact that it is possible to
identify the perlocutionary (or consequential) effects of an utterance without
understanding what someone is saying in uttering the relevant string of words.
It follows that a system of classifying utterances according to their
perlocutionary forces would be one that did not presuppose an understanding
of the utterances in question. Our employment of such a system would be
consistent with the view that it is possible to grasp the meaning of a sentence
without considering the force with which it is issued in a particular context;
we could study perlocutions without distancing ourselves from the notion of
literal sentence-meaning. The result is that if we read Austin as concerned not
with illocutionary effects but with perlocutionary ones, we will be inclined
not to recognize that one of his central concerns is criticizing this notion.
One further comment is in order here. After contrasting illocutionary acts
with locutionary and perlocutionary acts, Austin goes on to classify utterances
according to their illocutionary forces. At this juncture, he starts to refer to
those utterances that have the illocutionary forces of promises, bets, warnings,
etc., as ‘promises’, ‘bets’, ‘warnings’, and he starts to use the term ‘statement’
(without single quotes) – not, as he did previously, to refer to all historical
utterances of sentences, but rather – to refer exclusively to those utterances
that have the illocutionary force of statements.20

IV. The ‘True/False Fetish’

Austin’s criticism of the idea that sentences as such may be constative is the
centerpiece of his argument against bi-polarity – against what he sometimes
 The Happy Truth 71

calls the ‘true/false fetish’. He realizes that, to the extent that it seems possible
to identify sentences as such as constative, it will seem reasonable to think
that we can characterize the relation of correspondence between language and
the facts without investigating how what we say bears on the facts in
particular circumstances. Now it will seem as though we are justiŽ ed in
conceiving ‘true’ and ‘false’ as terms of assessment which (because, by our
lights, they pertain to a relationship between sentences and world which is not
caught up with our conventional procedures for performing speech acts) are
essentially different from the terms of assessment (e.g. ‘bizarre’, ‘appro-
priate’, ‘rough’, ‘misleading’, etc.) employed in assessing speech acts. Austin
accordingly believes that if we abandon the idea that we can grasp the
meaning of sentences independently of a grasp of the (illocutionary) force
with which they are used in particular contexts, we will want to say that the
terms we use for appraising statements ‘overlap with those we use in the
appraisal of performatives’ (pp. 141–2).21 We will take ‘true’ and ‘false’ to
stand, not for a ‘simple’ relation (i.e. one ‘uncontaminated’ by conventions of
illocutionary force), but rather for ‘a general dimension of being a right or
proper thing to say as opposed to a wrong thing, in these circumstances, to this
audience, for these purposes and with these intentions’ (p. 144). Austin
writes:
Is the constative, then, always true or false? When a constative is confronted with
the facts, we in fact appraise it in ways involving the employment of a vast array of
terms which overlap with those that we use in the appraisal of performatives. In real
life, as opposed to the simple situations envisaged in logical theory, one cannot
always answer in a simple manner whether it is true or false (pp. 141–2).22

By arguing along these lines, Austin wishes to get us to jettison the view
that ‘true’ and ‘false’ are essentially different from evaluative terms we use in
the assessment of speech acts and to leave room for the possibility that some
bits of language that report on the facts are most accurately characterized, not
either as true or as false, but rather as, say, rough or deceptive. But it would be
misleading to portray him as simply claiming that sometimes statements are
neither true nor false – that sometimes they are ‘approximately true’ or ‘not
strictly false’. This claim is consistent with the view of literal sentence-
meaning Austin rejects. Someone who endorsed the rejected view might
maintain that there are cases in which we know the literal meaning of a
sentence that is being used to make a ‘statement’ – and in which, moreover,
we also know how the indexical elements of the sentence pin it to the
particular context of use – but in which we nevertheless cannot determine
whether the relevant ‘statement’ is true or false. The idea might be that our
inability to make judgments of truth or falsity in these cases is a function of
(what might be thought of as) the indeterminacy of literal meaning. However,
although a conception of the indeterminacy of literal meaning might thus
72 Alice Crary

plausibly be taken to motivate various things Austin says in criticizing an

ideal of the ‘statement’ as bi-polar, the resultant critique of bi-polarity would
not be Austin’s.
Austin’s critique of the bi-polar ‘statement’ is grounded in his critique of
the idea of literal sentence-meaning. It follows, not only that there is no such
thing as assimilating his critique of bi-polarity while retaining this idea, but
also that he should be read as leaving undisturbed forms of bi-polarity that do
not presuppose it. Austin leaves us free to think of statements (as opposed to
‘statements’) as bits of language which state facts either truly or falsely as
long as we acknowledge: (i) that statements are a class of utterances, (ii) that
there may be utterances which are like statements in every respect except that
they are not either strictly true or strictly false, and (iii) that statements, in
addition to being aptly described as either true or false, may aptly be
described as, e.g., appropriate, bizarre, misleading, etc.

V. A Note on the Reception of Austin’s Work

There is a striking mismatch between the interpretation of Austin’s lectures
presented in the preceding sections and established philosophical interpreta-
tions. Philosophical commentators rarely represent Austin’s critique of a
traditional ideal of the ‘statement’ as driven by a critique of the idea of literal
sentence-meaning. Indeed, it is rare for philosophical commentators even to
suggest that criticizing such an idea of meaning is one of his central
philosophical concerns. Searle’s in uential interpretation is, as I mentioned at
the outset, representative in this respect. One of my goals in closing is to show
that this feature of Searle’s interpretation is an expression of a tendency to
allow certain philosophical prejudices – prejudices foreign to Austin’s
thought – to shape his interpretative strategy.
Although Searle says he agrees with basic tenets of Austin’s doctrine of
speech acts, he also self-consciously distances himself from Austin’s claim
that some speech acts may have truth-values.23 He claims that ‘statement’ is
ambiguous in that it can sometimes mean the act of stating (e.g. ‘The
statement of our position took all of the morning session’)24 and can also
sometimes mean what is stated (e.g. ‘The statement that all men are mortal is
true’).25 Searle refers to the former as ‘statement acts’ and to the latter as
‘statement objects’, and he tries to show that Austin winds up defending the
view that some speech acts are capable of being true or false because he
unwittingly slides between these two meanings of ‘statement’. According to
Searle, when Austin says that statements are speech acts, he is talking about
‘statement acts’, and when he says that statements are capable of being true or
false, he is talking about ‘statement objects’. So, as Searle sees it, it does not
 The Happy Truth 73

follow from Austin’s discovery that statements are speech acts that some
speech acts are capable of being true or false.26
Searle is not unaware that Austin’s talk of statements as speech acts is
intended to draw attention to the role the circumstances in which a statement
is made play in making up its content. Searle repeatedly observes that one of
Austin’s larger philosophical ambitions is to show that features of the
circumstances in which a sentence is uttered play a role in constituting its
meaning. And he tries to accommodate this moment in Austin’s thought by
claiming that features of the way in which a sentence is used (such as its
‘stress and intonation contour’) affect (what Searle calls) its ‘intended
meaning’.27 Searle’s thought seems to be that, once he has made this initial
claim about the signiŽ cance of the circumstances in which linguistic
expressions are used, he can then go on – without betraying the spirit of
Austin’s thought – to advance a further claim about how sentences have (in
addition to their ‘intended meanings’) ‘literal meanings’ they carry with them
into every context of use. Searle thus believes he is licensed to present himself
as faithful to Austin’s thought in arguing that the way in which a sentence
corresponds to the world is a function of elements of its ‘literal meaning’.28
And he believes he is licensed to claim that Austin would have supported an
ideal of the ‘statement’ that presupposes some idea of literal sentence-
meaning if not for (what Searle regards as) an error in Austin’s analysis of
statements.
Searle does not argue that it is possible to defend the attribution to Austin of
an idea of literal sentence-meaning by mustering internal, textual evidence.
What he suggests instead is that it is possible to defend this attribution by
appealing to evidence of an external kind. In relatively recent writings, Searle
defends an assumption about how it is necessary to retain the idea of literal
sentence-meaning in order to preserve the notion of objective truth.29 And he
signals that he thinks this assumption speaks in favor of searching for an
interpretation of Austin’s doctrine of speech acts which is consistent with the
idea of such meaning.
It is worth noting in this connection that, although Derrida is conspicuously
critical of Searle’s interpretation of Austin’s lectures,30 his interpretative
strategy nevertheless resembles Searle’s in an important respect. Derrida
(together with members of the group of commentators on Austin he has
in uenced) agrees with Searle in approaching Austin’s lectures equipped with
the assumption that the notion of objective truth stands or falls with the idea
of literal sentence-meaning.
The point here is not that the assumption operates uniformly across the
different interpretations. Commentators who follow in Derrida’s footsteps
differ from Searle in taking the assumption to support an interpretation of
Austin on which he appears to be – not receptive to the idea of literal
sentence-meaning, but rather – uneasy about the very notion of objective
74 Alice Crary

truth. What accounts for this difference from Searle, given the shared
interpretative framework, is the fact that, unlike Searle, the members of this
group of commentators start with the recognition that a central aim of
Austin’s lectures is to discredit the idea of literal sentence-meaning.31
One characteristic expression of this form of recognition, within the
writings of Derrida and other literary-theoretical commentators, concerns
their understanding of Austin’s view of truth. It is typical of Derrida – and, for
instance, Stanley Fish – to claim that Austin rejects a familiar conception of
truth and falsity that takes for granted that we can understand what someone is
saying independently of a grasp of how her words satisfy the demands of a
conventional procedure for saying something on a particular occasion. Thus
Derrida writes that Austin hopes to free us ‘from the true/false opposition ... at
least in its classical form’.32 And Fish tells us that Austin wishes to wean us
from the view that answers to ‘the question “Is it true or false?”... [are]
absolute judgments made independently of any particular set of circum-
stances’.33
In so far as the account of Austin’s view of truth that Derrida and Fish
champion portrays Austin as repudiating a traditional philosophica l regime of
truth (and, in particular, one that presupposes the idea of literal sentence-
meaning), it resembles the account presented in this article. But this
resemblance coexists with a philosophically fundamental difference. Within
the Derridean account – in sharp contrast to the account laid out above – an
interpretatively compelling story about how Austin abandons a familiar
regime of truth serves as a preface to an interpretatively quite dubious story
about how he is skeptical about the very notion of objective truth. Austin is
read as maintaining that there is no such thing as a statement that possesses
the kind of universal authority that is distinctive of objective truth. He is taken
to be denying the existence of statements that are objectively true in the sense
that anyone who is thinking about the pertinent matter correctly ought to
recognize their correctness. The larger suggestion of the Derridean account of
Austin’s view of truth is that Austin should be read as proposing to exchange
an understanding of truth and falsity as terms of assessment with universal
authority for an understanding of them as terms of assessment whose
authority is merely relative to the linguistic conventions that govern particular
discursive practices. The idea, in Derrida’s parlance, is that Austin wants to
‘substitute [the authority of the truth value] for ... the value of force’.34 Fish in
turn claims that Austin jettisons a conception of truth as concerned with ‘a
special relationship [language] bears to the world’, in favor of a conception of
it as concerned with ‘a special relationship it bears to its users’.35
It is at this juncture that we see the in uence, in the writings of Derrida and
Fish, of the very basic exegetical assumption shared with Searle. Like Searle,
Derrida and Fish assume that we cannot rid ourselves of the idea of literal
sentence-meaning without placing the very notion of objective truth at risk.
 The Happy Truth 75

Where they differ most dramatically from Searle is simply in maintaining that
Austin is willing to run this risk.
This brings me to my main criticism of prevailing interpretations of
Austin’s lectures. An assumption about a direct conceptual dependence of
objective truth on literal sentence-meaning is foreign to Austin’s thought.
Moreover, it is just as foreign when made by philosophers like Searle as it is
when made by literary-theoretical commentators like Derrida. The difŽ culty
isn’t merely that Austin himself never explicitly makes such an assumption –
that he launches an assault on the idea of literal sentence-meaning without
any indication that he takes his assault to threaten his right to talk about
objective truth.36 The more serious difŽ culty is that the assumption depends
for its apparent plausibility on an internally inconsistent understanding of the
implications of abandoning the idea of literal sentence-meaning.
This more serious difŽ culty might be expressed as follows. To give up the
idea of literal sentence-meaning is to give up an understanding of ourselves as
capable of identifying and assessing bits of language from outside
‘conventions of illocutionary usage’. If we present ourselves as breaking
cleanly with this idea, we commit ourselves to the view that there is no such
thing as a linguistically formulable conception of the world that is accessible
independently of our dealings with (irredeemably conventional) illocutionary
forces. Now we are in effect claiming to have repudiated any thought of an
extra-conventional standpoint from which we might Ž nd a priori metaphy-
sical grounds for representing our preferred conception of the world as merely
relative to a set of illocutionary conventions; or, in other words, from which
we might Ž nd a priori metaphysical grounds for representing our preferred
conception as falling short of objective authority. Indeed, given the
constraints of our envisioned philosophical outlook,34 there cannot even be
a meaningful question about whether our preferred conception of the world
lacks (universal, objective) authority in relation to some hypothetical, extra-
conventionally available alternative. For it is part of this outlook that the idea
of such an alternative is intrinsically confused. The end-result is that we
cannot consistently treat our rejection of the idea of literal sentence-meaning
as undermining the notion of objective truth. If we try, we wind up presenting
ourselves both as repudiating any thought of an extra-conventional standpoint
on language and, at the same time, also as endorsing a view of language which
(in so far as it represents us as hopelessly cut off from objective truth as a
result of our failure to adopt such a standpoint) remains structured by such a
thought. In order to distance ourselves from the idea of literal sentence-
meaning in a rigorously consistent manner, we therefore need to resist the
temptation to represent any distance we achieve as simultaneously a departure
from the notion of objective truth.
The line of reasoning traced out in the last paragraph doesn’t amount to an
attack on the thinker who both protests that talk of objective truth presupposes
76 Alice Crary

a standpoint on language outside conventions of illocutionary usage and, in

addition, also tries to defend her right to protest by explicitly endorsing the
idea of literal sentence-meaning. What the above line of reasoning shows is
simply that a protest of the pertinent sort depends for its legitimacy on the
idea of such meaning. The lesson we should take from the line of reasoning is
– to put it in terms of the concerns of this article – that if Austin succeeds in
demonstrating there is a  aw inherent in the idea of literal sentence-meaning,
he will by the same token have succeeded in demonstrating the emptiness of
the protest.
One way of summarizing the issues raised by this brief survey of
commentary on Austin’s lectures is to say there are no antecedent
philosophical obstacles, of the sort Searle envisions, to depicting Austin as
launching a direct attack on the idea of literal sentence-meaning. If we
approach Austin’s lectures with an unprejudiced eye, we see that he wishes to
get us to take seriously the possibility that the idea of such meaning is
philosophically bankrupt. When Austin’s readers turn to seriously investigat-
ing this possibility, it will be a sign that the reception of this region of his
philosophy has at last begun.37

NOTES

1 Austin (1979b, 185 n.2).

2 Austin (1962, p. 142). All unattribute d page reference s in the body and notes of this article
are to this text.
3 I am following Austin in placing single quotes around ‘statement’. Austin uses this notation
to indicate his reservation s about referring to a traditional linguistic ideal – one often
characterize d in philosoph y as an ‘ideal of the statement’ – as an ideal of the statement.
Throughou t this article, I speak of ‘statements’ (with single quotes) when talking about the
philosophica l ideal that Austin wants to challenge , and I speak of statements (without single
quotes) when talking about things that Austin himself says about statements. I discuss the
philosophica l rationale behind this notation below, in section II.
4 I am borrowing this talk of ‘bi-polar propositions ’ from the early Wittgenstei n (1979).
Wittgenstei n speaks of ‘the bi-polarit y of the proposition ’ in connectio n with the thought
that it is essential to a proposition that it be either true or false. He writes: ‘Every
proposition is true-false. To understan d it we must know both what must be the case if it is
true and what must be the case if it is false. Thus the proposition has two poles
correspondin g to the case of its truth and the case of its falsehood ’ (pp. 98–99).
5 Austin gave his lectures decades before Dummett (1981) argued that a three-value d logic
(i.e. one making room for the value ‘neither true nor false’ alongsid e the values ‘true’ and
‘false’) might furnish a better account of the sense of assertions than a traditional two-
valued logic. It is, however, not obvious that Dummett’s view of language can be excused
from membership in the class of views Austin is attacking simply on account of its
depiction of assertion s as ‘tri-polar’ and not ‘bi-polar’. Dummett’s view resembles the
views Austin is attacking in that, like them, it represent s the sense of a sentence as Ž xed by
the condition s under which it is true and all other conditions. Where Dummett’s view
differs is in its construal of such ‘other’ conditions as including the conditions under which
a sentence is false and the condition s under which it is neither true nor false – as opposed to
only the conditions under which the sentence is false.
6 At this stage in his career, Searle, although he acknowledge s that some things Austin says
about the truth and falsity of statements are at odds with a conceptio n of the statement as
 The Happy Truth 77

‘bi-polar’, claims that Austin’s hostility to this conception is a function of an error in his
analysis of statements and that, when allowances are made for this error, the doctrine of
speech acts becomes fully compatible with bi-polarity. See Searle (1968, pp. 422–4) and
(1969b, chs 1–3).
7 Searle (1978).
8 Derrida (1988c).
9 Stanley Fish’s writings contain one of the clearest and most in uential readings of Austin
along these lines. See especiall y Fish (1980). Fish (1989) points out that his interpretatio n
of Austin resembles Derrida’s in important respects. I touch brie y on Derrida’s and Fish’s
work on Austin below in section V.
10 This passage continues: ‘[t]his assumption is no doubt unconscious , no doubt is wrong, but
it is wholly natural in philosoph y apparently’.
11 A number of commentators have argued that Austin overestimates the difŽ culty of
preserving a sharp distinctio n between constative s and performatives . According to Max
Black, Austin would be in a position to retain his original distinctio n if he simply stipulate d
that no utterance in which a speaker makes a truth claim counts as a performativ e – not
even if the speaker makes the claim in ‘a special way indicated by the utterance ’ (e.g. ‘I
state that it is time to do it’). Black claims that if we impose this constrain t on what counts
as a performativ e ‘many if not all of the difŽ culties that Austin encountere d [in attempting
to distinguis h constatives and performatives ] will be overcome’ (1969, p. 406). Here Black
is considering passages in which Austin suggests that his initial distinction between
constatives and performative s is threatene d by the discover y that some performatives , while
not true or false per se, have some bearing on the facts. Black also consider s passages in
which Austin suggests that statements, like performatives , are the performance s of
conventiona l procedures . Black says he agrees with Austin that there is a sense in which
even statements are conventional . He adds that if we are to safeguard the constative /
performativ e dichotomy we need to distinguis h between the sense in which the action
performed when a speaker utters a performativ e formula is conventiona l and the sense in
which any act of speech is conventional , and he suggests that this task will be a quite
straightforwar d one (1969, pp. 406–8). But Austin gives us reason to think that the task of
making a general distinctio n between the conventiona l characte r of performative s and that
of constatives will be more difŽ cult than Black imagines. Indeed Austin suggests that the
task is a hopeless one. When he points out that constative s are in general subject to the same
range of infelicitie s to which performative s are subject, he is effectivel y pointing out that
(at least some) constativ e utterances are the performances of conventiona l procedure s in
just the same way in which performative s are the performance s of conventiona l procedures .
12 Austin writes that ‘in order to explain what can go wrong with statements we cannot just
concentrat e on the proposition involved (whatever that is) as has been done traditionally ’
(p. 52).
13 See above, note 3.
14 Austin (1979c, esp. pp. 119–20) explicitly discusse s this distinctio n between sentences and
statements. He sometimes restricts the applicatio n of the term ‘statement’ to speciŽ c sorts
of utterance s (1979, p. 131). I touch on this further restriction below, at the end of section
III.
15 At the opening of his lectures, Austin claims that philosopher s endorse a traditiona l ideal of
the ‘statement’ in part because they fail to distinguish sentences from statements. They tend
to obscure the fact that ‘not all “sentences ” are (used in making) statements’ through ‘some
loose use of “sentence ” for “statement”’ (p. 1).
16 Indeed Austin believes that we can classify bits of language as either constative s or
performatives – as long as we are clear that what we are doing is classifying acts of speech
accordin g to whether or not, in performing them, the relevant speaker constate s something
and not classifyin g sentences accordin g to whether or not they are suited for constatin g
things. The fact that Austin is willing to use a vocabular y of ‘constatives ’ and
‘performatives ’ later in his lectures should, for this reason, not be taken as a sign that he
equivocate s about whether or not it is possible to sort sentence s as such into constatives and
performatives.
17 In this connection , see Searle (1968).
78 Alice Crary

18 This part of Austin’s lectures sheds helpful light on the bits of his work outside How To Do
Things With Words that have attracted the most critical attention – bits in which he
investigate s ways in which we ordinaril y use a phrase or a sentence and conclude s that it is
‘meaningless’ to use that sentence or phrase in some set of circumstance s in which it is not
ordinaril y used. In the light of Austin’s discussio n of locutionary and illocutionar y acts, it
seems reasonabl e to read Austin as claiming that when expressions are used in
circumstance s which are so different from the circumstance s of their ordinary use that
we have no clue what someone might be saying in using them, it is best to regard those
expression s as mere nonsense . But most commentators take Austin to be making – not a
logical objection to our coming out with combination s of words which, in the particula r
circumstance s of utterance , make no sense, but rather – a pragmatic objection to our
coming out with combinations of words which, while meaningful , are strange or bizarre in
the situations in which they are uttered. Thus, e.g., Searle (1969a) takes Austin to be
concerne d with showing that we don’t ordinaril y advance factual claims unless there is
some reason to doubt their truth. He reads Austin as maintaining that if there is no reason to
doubt the truth of a particula r claim, making it will be ‘bizarre’ or ‘odd’. (For a favorabl e
assessment of Searle’s interpretation , see Grice [1989, pp. 3–21].) Alan White (1969)
suggests a helpful criticism of relevant portions of Searle’s interpretation . White argues that
Searle’s Austin differs from the true Austin in the following respect. Where the former
might subscrib e to the motto ‘if you have nothing worth saying, then keep it to yourself’,
the latter would more likely champion the motto with which Wittgenstei n closes the
Tractatus: ‘Whereof one cannot speak, thereof one must be silent.’ White’s use of this
Tractarian motto draws attention to a striking parallel between Austin’s thought and a
central moment in Wittgenstein’s philosophy , viz., the moment at which Wittgenstei n
employs ‘nonsense ’ as a term of philosophica l assessment . When Wittgenstei n says that a
form of words that we are tempted to utter in philosophy is nonsense , he is making – not a
pragmatic objection to a form of words that attempts to say some impermissible (or bizarre
or strange) thing that cannot properly be formulated , but rather – a logical objection to a
form of words that fails to say anything , that hasn’t yet been given any sense. For
discussio n of the signiŽ cance of this aspect of Wittgenstein’s philosophy , see, e.g., the
essays in Crary and Read (2000).
19 For a straightforwar d version of this sort of view of ‘literal meaning’, see Grice (1989).
20 See above, note 14.
21 One way in which Austin attempts to demonstrat e overlap between the terms we use for
appraisin g statements and the terms we use for appraisin g speech acts (or performatives ) is
by listing a series of sentences and asking us to determine whether they are true or false.
(See the discussio n of ‘France is hexagonal’, ‘Lord Raglan won the Battle of Alma’, and
‘All snow geese migrate to Labrador’ [pp. 142–3].) The point of this exercise is to show
that we can only make judgments of truth or falsity if we imagine a sentence being used, in
some situation, to make a statement.
22 For the signiŽ cance of Austin’s willingness to continue to talk about constativ e and
performativ e utterances even late in his lectures, see above, note 16.
23 This is a central project of Searle (1968).
24 Searle (1968, p. 422).
25 Searle (1968, p. 422). Notice that Searle here uses the term ‘statement’ to refer to what
Austin would call a sentence . If we wanted a statement in Austin’s sense, we might take:
‘The statement the minister made to the family of the deceased to the effect that all men are
mortal is true.’
26 Searle (1968, pp. 421–3).
27 Searle (1968, pp. 415–16). See also (1969b, pp. 44–49).
28 Searle takes the ‘literal meaning’ of a sentence to include elements that have the function of
correspondin g to the world (Searle describe s these elements as making up the
‘propositiona l content’ of the sentence ) and also elements that place constraint s on the
illocutionar y force with which the sentence can be uttered (1968, p. 412). His thought is
that, for any given illocutionar y act, it is possible to come up with a linguistic formula
which (on account of constraints its ‘literal meaning’ places on illocutionar y force) can only
be used to perform that illocutionar y act. (See also 1969b, pp. 20–21.) Thus, e.g., he thinks
 The Happy Truth 79

that there are linguistic formulae which (on account of their ‘literal meanings’) can only be
used to make promises and also linguistic formulae that can only be used to make
statements. In presentin g his view of the relation between ‘literal meaning’ and
illocutionar y force, Searle resurrect s in a slightly new guise an idea Austin plainly rejects
– namely that some combination s of words are ideally suited (without regard to the
circumstance s in which they are used) for making statements.
29 Searle (1992).
30 See the exchang e between Derrida and Searle that includes Derrida’s ‘signature Event
Context’ (Derrida [1988c], Ž rst published in vol. 1 of Glyph in 1977), Searle’s response in
‘Reiterating the Differences : A Reply to Derrida’ (Searle (1977), publishe d in vol. 2 of
Glyph) and Derrida’s counter to Searle in ‘Limited Inc a b c ...’ (Derrida [1988b], which
also Ž rst appeared in vol. 2 of Glyph).
31 Derrida (1988c) tells us that Austin’s efforts to discredit this idea – which Derrida refers to
as Austin’s ‘critique of linguisticism ’ – represen t the moment in Austin’s thought ‘that most
intereste d and convince d [him] in Austin’s undertaking ’ (p. 19). It is, however, important to
acknowledg e that, although Derrida thus presents himself as a fan of Austin’s attack on a
classical ideal of the ‘statement’, one of Derrida’s main concerns in his discussion s of How
To Do Things With Words is criticizin g what he sees as Austin’s tendency to allow this
ideal to continue to shape his own thinking. (See esp. Derrida [1988c, pp. 13–15].) Derrida
is committed to defendin g the somewhat peculiar exegetica l position that, when Austin
draws what by Derrida’s own lights counts as the main conclusion of his lectures (viz., that
there is no such thing as identifying the locutionar y act someone performs in using certain
words apart from an appreciatio n of the illocutionar y force with which it is uttered in
particula r circumstances) , he is assenting to a view that he himself in various ways actively
resists. (See Fish [1989] for a favorabl e assessment of this Derridean position. ) I am
inclined to think that Derrida’s claim that Austin betrays his own central insight rests on a
misinterpretation , but I cannot take up this matter in this article. (For a detailed and very
helpful discussion, see Cavell [1994].) My primary concerns here are: (i) the fact that
Derrida, unlike Searle, takes Austin to be rejecting the idea of literal sentence-meanin g and
also (ii) the further fact that the conclusion s Derrida draws from this Austinian gesture of
rejection reveal that, like Searle, he is drawing on an assumption about a conceptual
dependenc e of objective truth on literal sentence-meanin g – an assumption which (I am
arguing) is not at home in Austin’s thought.
32 Derrida (1988c, p. 13).
33 Fish (1980, p. 198), emphasis in the original.
34 Derrida (1988c, p. 13), emphasis added.
35 Fish (1980, p. 241).
36 See sections III and IV, above.
37 This article grew out of a series of conversation s I had in the mid-1990s with three of my
teachers – Nuel Belnap (with whom I read several of Austin’s papers in the Spring of 1993
while serving as the University of Pittsburgh Philosophy Department’s Alan Ross Anderson
Fellow), Stanley Cavell (with whom I discusse d the exchang e between Austin and Derrida
on several occasion s during the academic year 1993–94), and Annette Baier (with whom I
read How To Do Things With Words in the Fall of 1994). I am grateful to all three for their
guidance , and to Cavell and Baier for their help with an early draft. I owe thanks to Jim
Conant and Nancy Bauer for stimulating discussions about Austin over the years, as well as
for many constructiv e criticisms of this article; to Cora Diamond and Elijah Millgram for
their encouragemen t and for their helpful comments on a relatively recent draft; and to an
anonymous referee for useful suggestions about preparin g the article for publication .

R E FE R E NC E S

Austin, J. L. 1962. How To Do Things With Words. Cambridge, MA: Harvard University Press.
Austin, J. L. 1979a. Philosophica l Papers, 3rd ed., in J. O. Urmson and G . J. Warnock (eds).
Oxford: Oxford University Press.
Austin, J. L. 1979b. ‘A Plea for Excuses’, Philosophica l Papers, 175–204.
80 Alice Crary

Austin, J. L. 1979c. ‘Truth’, Philosophical Papers, 117–33.

Black, Max 1969. ‘Austin on Performatives’, in K. T. Fann (ed.), Symposium on J. L. Austin,
pp. 401–11.
Cavell, Stanley 1994. ‘Counter-Philosoph y and the Pawn of Voice’, A Pitch of Philosophy :
Autobiographica l Exercises. Cambridge, MA: Harvard University Press, pp. 53–128.
Crary, Alice and Read, Rupert (eds) 2000. The New Wittgenstei n. London: Routledge.
Derrida, Jacques 1988a. Limited Inc. Evanston, IL.: Northwestern University Press, pp. 1–23.
Derrida, Jacques 1988b. ‘Limited Inc a b c ...’, Limited Inc. pp. 29–110.
Derrida, Jacques 1988c. ‘Signature Event Context’, reprinted in Limited Inc. pp. 1–23.
Dummett, Michael 1981. Frege: Philosophy of Language, 2nd ed. Cambridge, MA: Harvard
University Press.
Fann, K. T. (ed.) 1969. Symposium on J. L. Austin. London: Routledge & Kegan Paul.
Fish, Stanley 1980. ‘How to Do Things With Austin and Searle: Speech-Act Theory and
Literary Criticism’, Is There a Text in This Class? The Authority of Interpretiv e
Communities. Cambridge , MA: Harvard University Press, pp. 197–245.
Fish, Stanley 1989. ‘With Compliments of the Author: Re ections on Austin and Derrida’,
Doing What Comes Naturally: Change, Rhetoric and the Practice of Theory in Literary and
Legal Studies. Durham, NC: Duke University Press, pp. 37–67.
Grice, Paul 1989. Studies in the Ways of Words. Cambridge, MA: Harvard University Press.
Searle, John 1968. ‘Austin on Locutionar y and Illocutionar y Acts’, Philosophical Review 77:
405–24.
Searle, John 1969a. ‘Assertions and Aberrations ’, in K. T. Fann (ed.), Symposium on J. L.
Austin, pp. 205–18.
Searle, John 1969b. Speech Acts: An Essay in the Philosophy of Language. Cambridge:
Cambridge University Press.
Searle, John 1977. ‘Reiterating the Difference : A Reply to Derrida’, in H. Sussman and S.
Webster (eds), Glyph I. Baltimore: Johns Hopkins University Press.
Searle, John 1978. ‘Literal Meaning’, Erkenntnis 13, 207–24.
Searle, John 1992. ‘Rationality and Reason, What is at Stake?’, Daedelus 122: 55–84.
White, Alan 1969. ‘Mentioning the Unmentionable’, in K. T. Fann (ed.), Symposium on J. L.
Austin, pp. 219–25.
Wittgenstein , Ludwig 1979. ‘Notes on Logic’, Notebooks 1914–1916, 2nd ed., in G. H. von
Wright and G. E. M. Anscombe (eds), trans. G. E. M. Anscombe. Oxford: Basil Blackwell,
pp. 93–107.

Received 7 November 2001

Alice Crary, Department of Philosophy, Graduate Faculty of Political and Social Science,
The New School, 65 Fifth Avenue, New York, NY 10003, USA. E-mail: CraryA@
newschool.edu