Vanderveken, Daniel, and Susumu Kubo, eds (2001) Essays in
Speech Act Theory. John Benjamins Publishing Company, vi+324pp,
paperback ISBN 90-272-5094-4 (EUR) / 1-55619-836-1 (US & Canada),
Pragmatics and Beyond New Series 77.
Anne Reboul, Institute for Cognitive Sciences, CNRS, France.
INTRODUCTION
Vanderveken and Kubo's collection of papers is mainly
oriented toward illocutionary logic (Searle and Vanderveken
1985, Vanderveken 1990, 1991) rather than to "classical"
speech act theory (Austin 1962, Searle 1969). Though this
may mean that more technical knowledge may be needed to
read it, it also offers a good panorama of what is being
done in contemporary speech act theory. However, it should
be clear that this is no a beginner book: a good grounding
in both "classical" speech act theory and illocutionary
logic (not to say in logic in general) is needed and some
philosophical knowledge would be helpful.
I will begin by a general presentation of the book overall
organization, then go to a summary of each paper in the
book, including the introductory chapter. I will pick out
here and there a chapter for more extended discussion when
either its excellency or its deficiency seems to mark it
out for discussion and I will end with a general appraisal
of the book.
ORGANIZATION OF THE BOOK
The book is divided in three parts of four chapters each.
However, it seems that some chapters are in the wrong part
of the book and that the aesthetically pleasing symmetry of
three equal parts might be more artificial than anything
else. The first part, "General Theory", collects a (very
big) paper by Vanderveken which sums up illocutionary logic
and extend it beyond illocutionary acts to conversation; an
historical paper by Leclerc on the precursory aspects of
classical French grammar and its successors, Searle's paper
on performatives advocating an approach where they are to
be considered as declaratives and, finally, a paper by de
Sousa Melo on directions of fit between mind and the world.
Part II, "Discourse and interlocution", begins with a
chapter by Trognon on the application of illocutionary
logic to conversational interaction. It is followed by two
chapters, both excellent, which deal, respectively, with
the difference between locutionary and illocutionary acts
(by Davis) and with a situation semantics formalization of
illocutionary acts based on an ascriptional approach (by
Yamada). Both of these papers could presumably have found a
legitimate niche in the first part as they both deal with
general points in speech acts theory and do not seem to be
especially oriented toward discourse or interlocution. The
last paper in this part, by Moulin and Rousseau, whatever
its merit as a modelization of conversation has nothing to
do with speech acts theory as it endorses Searle's
pessimism (1992) about the application of speech act
theory to conversation.
Part III, "Speech Acts in Linguistics", combines two very
interesting papers on, respectively, illocutionary
morphological markers in Japanese (by Kubo), and reporting
speech acts in indirect discourse (by Yamanashi), with a
paper by Moeschler, casting strong doubts on the
application of speech act theory to conversation and a
paper by Dominicy and Franken comparing Relevance Theory
approach to speech acts (Sperber & Wilson 1995) with
illocutionary logic. One can wonder why Moeschler's paper
does not appear in the previous part of the book given that
it is as centrally concerned as Trognon's by interlocution
and conversation and unless one wants to put Sperber and
Wilson's theory in Linguistics (contrary to their theory,
by the way) with illocutionary logic does not seem to have
much to do with the main theme of this part.
SUMMARY OF THE CHAPTERS
The "Introduction" is devoted to a general presentation of
speech acts theory, tracing its historical origins to both
French classical Grammar and Frege's notion of force, as
well as to Wittgenstein's notion of a language game. Though
none of this comes as a surprise, it is well worth
remembering. It then presents Austinian and Searlian speech
acts theory before turning to illocutionary logic and its
possible extension to conversation. It also includes a
short presentation of each paper in the book.
The first chapter, "Universal Grammar and Speech Act
Theory" by Vanderveken, is devoted to a general
presentation of illocutionary logic and its extension to
conversation. It also aims to ask and answer the question
of the transcendent features that any natural language must
have to fill its expressive and communicative functions.
Not surprisingly these features are to be discovered
through illocutionary logic and are (in the last few lines
of the paper) linked to Chomsky's hypothesis of a Universal
Grammar. The paper makes explicit the link between
illocutionary logic and semantic logic of a Montagovian
type, success and satisfaction being generalizations of
truth, though they are not reducible to it. This is where
the notion of direction of fit comes in, dictating how the
correspondence between words and world should go. There are
four directions of fit (words-to-world; world-to-words;
both; none), corresponding to five illocutionary points
(assertive, commissive, directive, declaratory and
expressive). Universals of speech act theory are
ontological, logical, semantic, pragmatic, and cognitive.
Illocutionary logic can be extended to conversation because
one can consider an extended unity of meaning, the
intervention, which has a discursive goal, a mode of
achievement of discursive goal, thematic conditions,
background conditions and sincerity conditions.
Interventions comprise a master speech act which must be
satisfied and non-defective if the intervention is to be
satisfied. The whole chapter is quite useful as a general
presentation of illocutionary logic and of its extension to
conversation though Vanderveken does not seem to have
answered Searle's qualms about it (see Searle 1992).
Leclerc's paper, "Verbal Moods and Sentence Moods in the
Tradition of Universal Grammar", has a mainly historical
import. It traces the ancestry of speech act theory to the
history of grammar and to a tradition initiated in the
Universal Grammar of Lancelot, Nicole and Arnaud in
seventeenth century France. As Leclerc points out, verbal
moods and sentence moods were not as clearly distinguished
then as they are now and both were considered as "merely
different but equivalent conventional devices expressing
our most important acts of thought or operations of the
mind" (64). There were two mains approaches to the subject,
a reductionist approach (whereby non assertive moods would
be reduced to assertion) and a conception in which moods
are considered as markers for different acts of thought.
The paper shows that though Port-Royal Grammar was
non-reductionist in its approach, as were some of its
successors (notably Du Marsais, the first grammarian for
Diderot and d'Alembert's Encyclopedia and the British
grammarians Harris and Gregory), others were reductionist
(notably Beauz�e, Du Marsais' replacement grammarian for
the Encyclopedia, and Beattie). This scholarly paper shows
how the same problems which gave rise to speech acts theory
were approached and understood in previous works.
The next chapter, "How performatives work" by Searle, aims
(and succeeds I think) at giving a satisfactory theory of
performatives. Under the current view, performatives are
statements (or assertions) whose peculiarity is that in
making them the speaker indirectly performs another speech
act. Searle's intention is to give a characterization of
performatives; indicate the condition of adequacy which
should be met by any solution; show the failures of some
analyses; introduce the necessary elements for a solution
and suggest a solution. According to him, performatives
enter speech acts theory as "some illocutionary acts
[which] can be performed by uttering a sentence containing
an expression that names the type of speech act" (86), i.e.
only so-called "explicit" performatives are performatives.
Under this informal definition, performatives rise the
question of how saying and performing are linked. This is
not irrelevant to the fact that one cannot lie or say a
falsehood in saying a performative. And finally are
performative verbs ambiguous between a performative and a
non-performative meaning? Searle begins by rejecting former
accounts, based on indirection, or on the idea that
performatives are assertives and comes to a declarative
account of performatives. As declaratives do, performatives
have a double direction of fit, which cannot be derived
from assertion. There are both extra-linguistic
(institutional) declarations and linguistic declarations
and performatives are linguistic declarations (the fact
created by linguistic declarations are themselves
linguistic). This explains why literal meaning is enough
for a performative. This is because:
a) performative verbs describe a class of actions where the
manifestation of an intention to perform them suffices
for their performance;
b) the notion of intention is part of the meaning of
performative verbs;
c) performative utterances are both self-referential and
executive.
This analysis, as Searle points out, has two consequences:
the performative (declaration) is primary; there is no
specific property which define performative verbs. The
class is delimitated by the way the world is.
"Possible Directions of Fit between Mind, Language and the
World", by de Sousa Melo, tries to show that, contra Searle
(1983), there are not three but four directions of fit
between mind and the world, i.e. the same number as those
of illocutionary acts. Restricting herself to conceptual
thoughts (representations of fact), she distinguishes
between mental acts (which have their public counterparts
in speech acts), which are judgments, attempts,
commitments, definitions, expression of emotions, etc. and
mental states, i.e. propositional attitudes. The two
classes are of course linked. According to de Sousa Melo,
Searle (1983) restricts directions of fit for the mind to
three because he mainly considers mental states
(propositional attitudes) and ignores mental acts.
Reintroducing mental acts should lead to the recognition of
four directions of fit for the mental. Searle recognizes
the three following directions of fit: mind-to-world
(beliefs, judgments); world-to-mind (intentions and
desires); none (feelings and expressions). De Sousa Melo
wants to introduce the reciprocal direction of fit under
the following argument: illocutionary acts correspond to
conceptual thoughts and declaratives (which have the double
direction of fit) are illocutionary acts. As such they are
accompanied by conceptual thoughts, which, given the double
direction of fit for declaratives, must themselves the
double direction of fit. This is of course rather a poor
defense of four directions of fit for the mental given that
declarations, whether or not they are accompanied by a
double-directed conceptual thought, are not in themselves
mental but, at best, a public expression of a conceptual
thought. De Sousa Melo sees this objection and attempts to
answer it through two arguments: there can be mental
declarations through which we classify, name, possibly in
mentalese, a claim so vague that it is impossible to assess
it; as the mind is part of the world, each mental act
modifies the world. This is basically her argument and I
must own that I find it deeply baffling. The only way of
understanding the very notion of a direction of fit between
mind (or language) and the world is through a principled
distinction between, respectively, mind and the world on
the one hand, and language and the world on the other hand.
If we fail to do that, the very notion of direction of fit
becomes trivial: indeed, by parity of reasoning, any
locutionary act should have the world-to-words direction of
fit, whatever the meaning of its content, if any, because
language is part of the world and its very existence is a
fact about the world. This means that every act, whether
mental, linguistic or anything else has the direction
world-to-mind/words/action. This does not prove, in the
Searlian sense, that there exist four mental directions of
fit. In fact, it does not prove anything that I can see.
We now come to the second part of the book, "Discourse and
Interlocution", which opens with Trognon's paper, "Speech
Acts and the logic of mutual understanding". This is a
difficult paper to assess because no notion is defined,
there is a constant equivocation in terminology and the
English is really very poor. What is more, the paper does
not seem to have been reread either at the manuscript stage
or in proofs before publication. Trognon begins by
introducing a few concepts of ethnomethodology before
turning to the extension of speech acts theory to "mutual
understanding" (which seems to be another term here for
"conversation", but, as no definition is given, it is hard
to be sure). He begins by quoting Searle's (1992) on the
limits of such an extension, ending this section of his
paper by the following rather mysterious sentences: "How
can the sequential character of conversation be used to
define the illocutionary force of a speech act if the act
does not determine the speech acts that follow it? The
solution lies in taking into consideration the semantic
properties of speech acts" (126). The first sentence seems
utterly mysterious: shouldn't it be the illocutionary force
of a speech act that should be used to define the
sequential character of a conversation, rather than the
reverse? And wasn't Searle just taking into consideration
the semantic properties of speech acts when he raised
doubts about the determination of the following speech acts
through the illocutionary force of an initial speech act?
The content of the rest of the paper rests on what was
called by the Geneva school of conversational analysis (see
Roulet et al. 1985) the "principle of dialogical
interpretation", according to which an initial speech act
is interpreted via the following reaction of the addressee.
Here, Trognon misuses Sperber & Wilson's notion of "mutual
manifestness" (1995), saying that the addressee's reaction
makes his interpretation of the initial speaker's act
"mutually obvious". This seems to be a mistake in
terminology. The pair of moves constituted by the original
act and its addressee's reaction is an "interpretation
relationship". Section 3.3.2., which follows, is so full of
typos that it is hard to understand it (e.g. "condtions",
"an illocution interpretation", "applying this low to all
speech acts", "sucesssful", etc.). The main import of the
paper seems to be in the notion of an interpretation
relationship, itself based on the principle of dialogical
interpretation, which has been rejected as non-predictive
(see Moeschler's paper in the present book). As such it is
not exactly convincing and it is hard to see what
illocutionary logic can add to that rather overworked and
tired approach. The paper ends with a quotation from Livet,
mistranslated from French and which, in the absence of a
context, remains more or less impossible to understand.
By contrast, Davis' paper may be the best in the book and
is indeed excellent. It deals with the distinction between
utterance acts and speech acts and aims (successfully I
think) to show that they are not merely different
descriptions of a single act, but, indeed, different acts.
This goes against Davidson's extensional treatment of
events. This, as Davis remarks, is because the problem is
intentional (in Brentano's sense) and intensional (in the
sense familiar from opaque contexts). To show that
utterance acts are not identical with the corresponding
illocutionary acts, Davis adapts Burge's argument for the
non-identicality between brain states and belief states,
itself an adaptation from Putnam's notorious Twin Earth
argument. The argument is too sophisticated for me to go
into the details but basically the idea is the following:
there are two planets, Earth and Twin Earth which are
physically identical but for one difference, the fact that
on Twin Earth, the stuff that passes for water is XYZ.
Ruth, an Earthling, says "Water is good to drink" and so do
Twuth, her Twin Earth counterpart. It is easy to account
for Ruth's discourse: she believes that water is good to
drink. It's not as easy for Twuth's discourse, because
neither Twuth, nor the people in her speech community, have
any idea of water. Thus, though Ruth and Twuth have uttered
the same sentence type (their utterance acts are indeed
identical), the illocutionary acts that they performed are
different. Hence, utterance acts and illocutionary acts are
not identical. This is very roughly the basic content of
Davis' very elegant paper, though it needs to be read
carefully to see that the consequences of the argument
extend far beyond the confines of speech acts theory.
The next paper, "An Ascription-Based Theory of
Illocutionary Acts" by Yamada, is another original and
rewarding paper. Basically, it is a convincing attempt to
establish a theory of speech acts, based on ascription of
actions to agents, avoiding propositions as contents for
speech acts and treating illocutionary acts as acts. Though
the second point is clearly a departure from illocutionary
logic, Yamada replaces propositions as content for
illocutionary acts by treating such contents through an
extension of Austin's theory of truth and, notably, his
notions of demonstrative and descriptive conventions.
Yamada uses a version of Situation Theory as the tool
through he builds his theory of illocutionary acts. He
begins by giving an outline of Situation Theory and of some
of its extensions which he deems necessary for a theory of
illocutionary acts, defines illocutionary commitment,
conventional effects, his Austinian theory of content, etc.
using the elements previously introduced in a particularly
clear way. Though the paper does need some concentration,
it is crystal clear and not really hard to read. I think
that it should be an obligatory reading for anyone
interested in both illocutionary acts and situation theory.
The last chapter of this part of the book, "An approach for
modelling and simulating conversations", though its theme
is squarely in accord with the theme of "interlocution",
does not really fit in this book, given that it accepts
Searle's verdict on the application of speech acts theory
to conversation, namely, that it is not useful. What is
more, it does not attempt to show this but takes it for
granted. This means that it has nothing to say about speech
act theory. It is a pity because the paper (mainly its
second part where the system is actually presented) is
interesting, but it is doubtful whether it will find its
public in a collection devoted to speech act theory. It is
a mystery why it was included in this book.
The third part, "Speech Acts in Linguistics", opens by a
chapter on Japanese illocutionary markers ("Illocutionary
Morphology and Speech Acts") by Vanderveken's co-editor,
Susumu Kubo. Kubo begins by noting that Japanese is
notorious for being a language in which sentence-final
affixes indicate the speaker's intentions and, more
specifically, its illocutionary intentions. Kubo makes some
observations about the combinations of Japanese verbs
(corresponding to English illocutionary verbs), which may
or may not be used performatively, with illocutionary
affixes and also with diverse forms of honorific or
derogatory Japanese formulas. This leads him to propose a
compositional treatment for illocutionary affixes in an
illocutionary morphology of his own. Though it is hard for
me as neither a speaker of Japanese nor a specialist in
Japanese linguistics to assess his proposition, regarding
its adequacy or originality, the paper is very clear and
well-constructed and makes quite fascinating reading.
The next paper by Yamanashi bears the rather misleading
title "Speech-Act Constructions, Illocutionary Forces, and
Conventionality". In fact, it is entirely devoted to an
exploration of the use of performative verbs in the quoting
part of reported speech, i.e. "he
advised/affirmed/directed/predicted/etc." As he notes, such
usages are not free: though one can use performative verbs
to report indirect speech acts, this is not possible when
the same performative verb was used in the original
discourse, i.e. the sentence "'I affirm that I am King of
the universe' he affirmed" is not acceptable. In the same
way, hedged performative (i.e. "I must tell/order/etc.
you ...") cannot appear in the quoting part of the reported
speech. One does not have "'you are a shrew', John ought to
tell her". The paper is devoted to making explicit these
constraints and is both original and quite interesting.
Though the author does not say so, data such as those he
reports could presumably be used to show the difference in
nature between types of reported speech. For instance, the
restrictions he notes for direct and indirect reported
speech do not seem to be valid for reported speech and
thought. One could have, for instance, "She was a shrew,
John was sorry/had/ought/etc. to say". Thus this paper
opens a whole new domain of investigation.
Moeschler's paper, "Speech act theory and the analysis of
conversations" (subtitled "Sequencing and interpretation in
pragmatic theory") should, as was said before, have been
put in the previous part of the book, as it is an obvious
rejoinder to some of Trognon's claims. The paper compares
the discourse analysis and the pragmatic theory approaches
to conversation. According to Moeschler, "the main purpose
of discourse analysis is the definition of necessary and
sufficient conditions for sequencing and interpretating
utterances in discourse" (239). However, sequencing and
interpretation cannot be accounted for independently and,
what is more, speech act theory cannot shed any light on
them because it is neither a theory of interpretation nor a
global theory of action. Moeschler begins by reiterating
Searle's criticisms of the application of speech act theory
to conversation based on the example of so-called question-
answer pairs, showing that illocutionary force has no
predictive import relative to conversational sequencing.
Moeschler then turns to discourse analysis and its emphasis
on functional constraints on sequencing. After describing a
version of the Geneva school model of conversation,
Moeschler points out that sequencing is generally regarded
as one version of the more general question of the
coherence of discourse which has come to be regarded as
more a problem of interpretation than of structure.
Interpretation has been regarded in conversational and
discourse analysis as susceptible to a solution through the
principle of dialogical interpretation (also used, albeit
implicitly, by Trognon in his paper). This principle meets
with destructive objections. Thus conversational or
discourse analysis should be abandoned and a radical
pragmatic interpretation, in the theoretical framework of
Relevance Theory should be adopted. The paper is clear.
Whether it is right or not is presumably up for discussion
and it is a pity that the present book does not offer one.
The final paper, "Speech Acts and Relevance Theory" by
Dominicy and Franken, is devoted to a comparison of the
treatment of a few examples in both Speech Act Theory and
Relevance Theory. The paper is not always very clear. It
deals with one of Sperber and Wilson's contentions about
the typology of speech acts, that is, that though it may be
a legitimate theoretical aim, it is not used by speakers in
as much as not all so-called illocutionary acts need to be
recognised as belonging to such and such a type to be
interpreted. Indeed, Sperber and Wilson (1995) distinguish
three classes of speech acts: basic (linguistic) speech
acts of three types, saying that, telling to and asking wh,
which should be recognised as such; institutional acts
(baptism, declaration of wars, etc.) which must be
recognized as such but are non linguistic; and, finally, a
third class of acts which may but need not be recognized as
such to be understood.
It is the third class that Dominicy and Franken
are interested in and they choose a few examples
over which they compare Relevance-based accounts
and Speech Acts-based accounts. Those examples are
imperatives used in advice, permission, good wishes,
audienceless cases, etc. To take the first case, the
example runs: "PETER: Excuse me, I want to get to the
station. MARY: Take a number 3 bus". According to Sperber
and Wilson, Mary communicates that taking a 3 bus is
desirable from Peter's point of view but not especially
from her own. Dominicy and Franken deny that this is so.
They may be right, but as their discussion relies on Speech
Act Theory, which Sperber and Wilson reject in this
instance, it is hard to see whether they are or not. They
then move on to permission, where much the same comment
applies. The mixture of speech acts theory and relevance
makes their discussion difficult to follow. They then
discuss several notions of desirability, aiming to
distinguish "desirable for", "desirable to" and "desirable
from the point of view of". Apparently both Relevance and
Speech act theory are wanting regarding that distinction.
They conclude this section by claiming that the speaker of
a permission expressed through an imperative presents the
state of affairs as desirable both to her hearer and to
herself. This is clearly false. If I say to my fifteen
years old son who is pestering me to buy him a motorcycle
"OK, go ahead, buy a motor cycle and kill yourself but just
leave me alone", I do give him permission to bring about a
state of affairs which he deems to be desirable to and for
himself (being the owner of a motorcycle) but which I deem
to be undesirable for me (I will be anxious about his
security). By the way, the very wording of the permission
in their example ("PETER: Can I open the window? MARY: Oh,
open it, then", italics theirs), seems to sustain such a
view (it certainly does not present Mary as overjoyed about
the opening of the window, which is weird if it is a
desirable state of affairs for her). They then go on to
discuss, more convincingly I think, other examples where
echoic uses may or may not be involved and propose an
analysis of some such imperatives as conditional
illocutionary acts, which seems to me quite interesting.
Their general conclusion is that speech act theory is
better regarding illocutionary acts though relevance is
superior regarding echoic and ironic use. There are two
errors regarding Relevance which it may be useful to point
out: on page 263, it is said that in Relevance,
"cooperation only takes place when it is needed to produce
relevance", which seems highly doubtful; on page 264, it is
said that in Relevance, the three basic types of speech
acts are assumed "by default". Given that one of the major
tenets of Relevance Theory is a strong rejection of default
assumptions, this, again, seems doubtful.
GENERAL CRITICISM
There are two qualities which one should look for in a
collection of papers: that the contributions should be
original (i.e. that they should not have been published
before or that they should not be easily accessible and
that they should bring new ideas), and that the papers
should be in accordance with the general theme of the book.
There is one exception to the second rule, which is Moulin
and Rousseau's paper which does not seem to have anything
to do with speech act theory. Though a previous version of
Vanderveken's paper has been published before in French, it
was not as easily accessible as it is in the present book.
Given the widespread distribution of Linguistics and
Philosophy, this may not be the case for Searle's paper,
but its quality more than warrants a re-edition. Basically
all other papers in the book, with the exceptions of de
Sousa Melo's (which seems utterly mistaken), and Trognon's
(which is a first draft rather than a well-argumented
paper) are well worth reading with special stars to Davis'
and Yamada's papers.
BIBLIOGRAPHY
Austin, John L. (1962) How to Do Things with Words, Oxford,
Clarendon Press.
Roulet, Eddy et al (1985) L'Articulation du discours en
fran�ais contemporain, Berne, Peter Lang.
Searle, John R. & Vanderveken, Daniel (1985) Foundations of
Illocutionary Logic, Cambridge, Cambridge University Press.
Searle, John R. (1969) Speech Acts: an Essay in the
Philosophy of Language, Cambridge, Cambridge University
Press.
Searle, John R. (1983) Intentionality, Cambridge, Cambridge
University Press.
Searle, John R. (1992) "Conversation", in H. Parret &
Verschueren, J. (eds), On Searle on Conversation,
Amsterdam/Philadelphia, John Benjamins.
Sperber, Dan & Wilson, Deirdre (1995) Relevance:
Communication and Cognition, Oxford, Basil Blackwell. 2nd
edition.
Vanderveken, Daniel (1990/1991) Meaning and Speech Acts:
vol. I & vol II, Cambridge, Cambridge University Press.
BIOGRAPHICAL SKETCH
Anne Reboul is a First Class Research Fellow at the French
Center for Scientific Research (CNRS) in France. She has a
Ph.D. in Linguistics (EHESS, Paris) and a PhD. In
Philosophy (University of Geneva, Switzerland). She has
written some books, among which an Encyclopedic Dictionary
of Pragmatics and quite a few papers in French and English,
on pragmatic and/or philosophical subjects.
|
