Download PDF
Porto Alegre
Livros Grátis
Milhares de livros grátis para download.
Dissertation presented as a prerequisite for
obtaining the Master degree from the Post-
graduation Program of Faculdade de Letras
of Pontifícia Universidade Católica do Rio
Grande do Sul.
Advisor: Ph.D. Jorge Campos da Costa
Porto Alegre
Ph.D. Jorge Campos da Costa - for his classes, guiding sessions and especially for
showing me there is much more about logic than meets the eye.
M.A. Maria Luíza Baethgen Oliveira and Ph.D. Ubiratan P. de Oliveira (my film
mentors) for granting me access to their library and filmotheque and for always being
solicitous to exchange ideas and points of view about Woody Allen and his work.
Ph.D. José Marcelino Poersch - for stressing out the transitory character of truth and
the need to keep an open mind about one’s assumptions and system of beliefs.
Ph.D. Ana Maria T. Ibaños - for being a brilliant linguist and for her continual support
and overrated trust in my potential.
Ph.D. Cristina Becker Lopes Perna - for one never forgets their first truly great
CAPES – for subsidizing education and granting scholarships.
M.A. Aline Aver Vanin – for her generosity and willingness in helping me format the
current study.
The ten subjects who partook in this project – for their patience and commitment.
Organic life, we are told, has developed gradually
from the protozoon to the philosopher, and this
development, we are assured, is indubitably an
advance. Unfortunately it is the philosopher, not
the protozoon, who gives us this assurance.
Bertrand Russell – Mysticism and Logic
A semantics/pragmatics interface may resolve problems of indeterminacy of meaning.
The current study attempts to suggest models of inferential architecture based on Grice’s
Amplified Model, as proposed by Costa (1984, 2004), and on the Relevance Theoretic
Framework, as envisaged by Sperber and Wilson (1986, 1995), for analyzing and interpreting
Woody Allen’s Match Point (2005). Concomitantly, an empirical investigation takes place.
Findings seem to suggest that meaning is inferentially processed on the basis of non-trivial
logical deduction, and that the more viewers are able to predict and relate, the higher their
understanding of a cinematic text will be.
Key words: semantics/pragmatics interface, inference, implicature, Woody Allen’s Match
Uma interface entre a semântica e a pragmática pode resolver problemas de
indeterminação de significado. O presente estudo almeja sugerir modelos de arquiteturas
inferenciais com base no Modelo Ampliado de Grice, proposto por Costa (1984, 2004) e no
Arcabouço Teórico da Relevância, como vislumbrado por Sperber e Wilson (1986, 1995),
para analisar e interpretar Match Point (2005) de Woody Allen. Concomitantemente uma
investigação empírica se desenvolve. Os resultados parecem sugerir que o significado é
processado inferencialmente, com base na lógica dedutiva não-trivial e que quanto mais um
expectador for capaz de fazer relações e previsões, maior será o seu entendimento sobre o
texto cinematográfico.
Palavras-chave: interface semântica/pragmática, inferência, implicatura, Match Point de
Woody Allen
1 INTRODUCTION ………………………………………………………………………….
2 THEORETICAL FRAMEWORK ………………………………………………………..
2.1 THE ORIGINS OF PRAGMATICS ………………………………………………….
2.2 GRICE’S THEORY OF IMPLICATURES ………………………………………….
2.2.1 Implicatures and Their Properties ………………………………………………….
3 METHODOLOGICAL APPROACH …………………………………………………..
4.1 MODELING THE CONTEXT …………………………………………………….
4.1.1 The Author …………………………………………………………………………
4.1.2 The Film ……………………………………………………………………………
4.1.3 Theme ……………………………………………………………………………….
4.1.4 The Characters ………………………………………………………………………
4.1.5 Intertextuality ………………………………………………………………………..
4.2 THE ANALYSIS OF MATCH POINT ……………………………………………….
5 FINAL CONSIDERATIONS ………………………………………………………………
REFERENCES ………………………………………………………………………………..
First and foremost, the current study characterizes Linguistic Theory as being
essentially interdisciplinary. Consequently, the present research will foment the idea that
interfaces, both internal and external, are necessary to attain its main objective; that is
resolving problems of indeterminacies of meaning.
As a matter of fact, the quest for the nature of meaning in natural language has been an
inexhaustible source of studies since the time of the ancient Greek philosophers to present
days. More recently, a little over a century ago, logicians, philosophers and linguists have
promoted significant contributions to the aforementioned search by investigating grounds that
had never been fronted before. Such grounds, initially seen as unstable and slippery, gradually
started to become more accurately demarcated and solidified, inciting a few bold ones to
venture into them and explore natural language through a more powerful magnifying glass,
which enabled the semantic sub-theory to get a complementary ally in its search for
signification – the pragmatic sub-theory.
The present theoretical study aims at analyzing the way in which implicit meaning
underlying Woody Allen’s Match Point
is obtained. For carrying out this intent, it is divided
into three chapters: the first one begins to delineate a panorama on the evolution of the
pragmatic theory from its classical period, when pragmatics appeared sort of uncommitted
with any scientific methodology until the 1950’s, when it began to undergo some degree of
systematization by being characterized as the study of language in use. From then on, the
modern period took its course, allowing pragmatics to evolve considerably. Contributions
promoted by Austin (1962), who envisaged language in use as a form of action and not a mere
way to describe reality, Searle (1969), who prompted the reformulation of Austin’s Speech
Acts theory, the theory of implicatures conceived by the philosopher Paul Grice (1975), and
the Relevance theoretic framework as proposed by Sperber and Wilson (1986), cast new light
onto the matter of signification advertising that an interface between semantics and
pragmatics was the means by which the quest for meaning should be achieved. From then on,
newer contributions have appeared such as the ones fostered by Bach (1994), who devised the
concepts of impliciture in juxtaposition to explicature and the ones offered by Levinson
Match Point. Written and directed by Woody Allen. PlayArte Home Video, Brasil, 2006.
(2000) with his notion of presumptive meanings. Although these latest contributions have
fostered additional advances to pragmatics, they will not take part in this study, which is
solely based upon the fundamental cornerstones of the sub-theory; i.e., inferences and
The chapter advances and Grice’s theory comes into the foreground, establishing an
external interface between semantics/pragmatics (the former accounts for logical entailments
and the latter is in charge of pragmatic implicatures) and formal sciences. Grice’s essential
notion of the implicits of natural language such as speakers’ meaning and implicatures are
developed along with communicative conventions outlined by his cooperative principle and
the maxims of conversation. A sub-section to explore the properties of implicatures
comprising calculability, cancelability, non-detachability, among others is elaborated as well.
The chapter gets to its closure with the introduction of Sperber and Wilson’s RT
(Relevance Theory), which enables the semantic/pragmatic sub-theories to undergo interface
with natural sciences. RT accounts for a cognitive approach to pragmatics, where meaning is
inferentially processed. The underlying concepts of its theoretic framework, including the
notion of relevance, ostensive stimulus, cognitive environment, contextual effects and
ostensive inferential communication are outlined, nurturing the qualitative idea that relevance
is a low cost/higher benefit property and that comprehension is non-demonstratively
determined. Both Grice’s and S&W’s RT will enable a further connection, namely an
interface between semantics/pragmatics and social sciences (encompassing communication).
The second chapter presents and contrasts architectural designs for modeling
inferential calculation that will serve as basic tools for analyzing Woody Allen’s film.
Initially, Grice’s Classical Model for inferential calculation is introduced and exemplified,
followed by its reformulation, as envisaged by Costa (1984, 2004), being entitled Grice’s
Amplified Model. Finally, the RT model is also proposed and illustrated. After such models
are presented, an additional interface between cinema and literature is incited, disclosing an
account of the quintessential factors that promote inference generation in both types of
communicative mediums.
The third chapter handles Woody Allen’s Match Point. The instigation for
approaching the inferential architecture underlying meaning in this cinematic text lies behind
the phenomenon that its author, nominated ‘the most European of the American filmmakers’,
is not usually appreciated by audiences who are neither in tune with his modus operandi of
presenting his views on the role and pursues of man in the world, nor totally acquainted with
his fancy contextual models that, in this particular film, may presuppose the viewers to
establish connections with a diversified range of specific background knowledge such as:
tennis, Fiodor Dostoevsky’s Crime and Punishment, opera arias and extracts, London
locations, not to mention an array of polemic themes ranging from luck versus determinism,
to ethics and morality. As a consequence, his movies are commonly referred to as ‘box office
poison’, influencing the American and Latin American moviegoing audience to have a
negative appraisal on them, or even avoid watching them at all, possibly because of the higher
effort demanded for processing the interpretation of many of the utterances that corroborate to
yield positive cognitive effects.
Before the analysis of Match Point takes its due course, extra linguistic components
are gathered into a database that will include facts about Woody Allen’s life and his films,
information about this particular cinematic text such as the setting, characters’ insights,
themes and the sort of intertextuality that this movie entails. Such a database will be partially
accountable for modeling the film context and guide the possible readers of the current
research to draw further considerations as to the communicative intentness comprised within
Concomitantly with the film analysis, an empirical investigation involving ten subjects
(five male and five female) who had never been exposed to Match Point before is going to be
described, aiming at evincing the sort of calculations these subjects were more likely to
undertake for attaining comprehension.
The present study apparently seems to contribute to the line of research that it is
inserted into (Logic and Natural Language) for introducing the incipient notion of
‘convergent implicatures’ attributable to correlated objects (animate or inanimate) that are
prone to lead someone to make use of exactly the same set of implicated premises for deriving
distinct implicated conclusions.
Finally, the findings of the current research are reported and considerations are drawn
about the hypotheses it intended to validate, namely that meaning is inferentially processed,
that indeterminacies of meaning can be resolved by means of an interface between semantics
and pragmatics and that the more viewers are likely to predict and relate, the more recovery of
the director/writer’s intentions there is, and the more positive cognitive effects.
According to Yule (2003), pragmatics is a branch of the linguistic theory that
is concerned with the study of meaning by a speaker/writer and its interpretation by a
Since its beginnings, pragmatics has appeared with an interdisciplinary character, for
linguists, logicians, mathematicians and philosophers showed a tendency to push to the edges
any phenomena that had to do with everyday language use, thus generating a sort of
wastebasket that comprised all sorts of evidence that could not be dealt within the formal
systems of language analysis.
Therefore, the initial mise-en-scène for the appearance of the pragmatic sub theory
was characterized by the presence of unwanted material that could neither be investigated by
means of purely structural (syntactical) processes nor merely determined by the truth-
conditions (semantics) of its propositional content. Let´s exemplify this situation a little
further. Consider the sentence underneath.
[1] The piano played Chopin and silence reigned supreme.
From a syntactic standpoint, this sentence is correct since it abides by all the rules that
establish its proper structure and rules out any inappropriate ordering such as ‘Played the
Chopin piano’.
On the other hand, as far as semantics is concerned, there are a few setbacks. First,
both the verbs ‘played’ and ‘reigned’ require an animate entity as their subjects; a condition
that can not be satisfied by either ‘the piano’ or by ‘silence’, not to mention that the verb
‘play’, in the sense utilized, does also require an inanimate entity as its internal argument,
condition which is not contended by the animate noun Chopin’. Second, the verb ‘play’ in
the sense of ‘playing a musical instrument’ entails a non-silent action that goes against the
second proposition ‘silence reigned’, even though, while considering the truth-conditions of
the propositions (p & q) expressed by the sentences ‘The piano played Chopin’ and ‘silence
reigned supreme’, it could be asserted by all means whatsoever that if p is true and q is also
true, so will p & q.
As one may see, sentence [1] exemplifies a case of more being communicated than
actually is said. This simple case, as well as many others dealing with the everyday world of
language use, can illustrate the sort of left over material that was unlikely to be investigated
just by formal approaches to language analysis.
In accordance with Costa (1984, 2004), Gottlob Frege (1892) was apparently one of
the first scholars to detect that presupposition should be connected to context use. According
to Penco (2006), Frege posited that logic, as a science that deals with the rules and tests of
sound thinking as well as proof by reasoning, should bear the purpose of serving as an
instrument for analyzing both scientific and natural language. From this onset, Frege
contemplated that every sort of linguistic expression comprised in logical language should
have a sense and a reference. Thus, every linguistic expression (singular term or ‘proper
name’ as used by Frege) must refer to a single object, expressing that the use of a ‘proper
name’ (singular term) must necessarily presuppose the existence of the individual (object)
denoted by its name and that every linguistic predicate must refer as well to the conceptual
relations and properties specified by such a name. Such a notion of context provided grounds
for a semantic distinction between the truth-conditions expressed by the terms in logical
language and by the terms (names) in natural language. While truth in the former is warranted
by the signification of the variables, in the latter is assumed based upon a certain degree of
extra linguistic intuition.
These Fregean postulates for logical theory are based on the principle of
compositionality, which states that both the sense and reference of the parts determine the
sense and reference of the whole statement. From these premises, Frege asserted that if a
sentence contains a term with no reference, the whole sentence is devoid of reference as well,
since it has no truth-value, being neither true nor false.
Taking into account the abovementioned consideration, Penco (2006) also states that
Bertrand Russell (1905) rebelled against such an idea, insisting that Frege’s mistake lay in
holding that names have both a sense and a reference.
Penco (2006) also states that Russell conceived a theory where the meaning of a
proper name should be exclusively restricted to its reference to an object. Such a point of view
advocated that proper names, in natural language, are abbreviations of defined descriptions
lacking any kind of property. In his 1905 article entitled On Denoting, Russell came up with
his ‘Theory of Descriptions’, which attempted to uncover the real logical forms that lie behind
the accidental and irrelevant linguistic forms present in natural language. In accordance with
Monk (1997), Russell expressed that denotation’ is a notion quite different from reference’.
By referring Russell meant the sort of linguistic relation between a word or a string of words
and one object, for example between a name and a person. On the other hand, according to
Russell’s Theory of Descriptions, proper names do not possess a denoting capability as such
ability can only be attributed to concepts. Therefore, denoting in natural language is never
obtained by means of a name, but only by a conceptual description. His theory asserts that
every and each denotative sentence is in its nature devoid of meaning. In other words,
denotative sentences are ‘incomplete symbols’ which may just acquire meaning within a
propositional context.
Goldstein et alii (2007) assert that Russell’s reasoning appeared to be more
epistemological than semantic, since to truly nominate a thing seems to suggest a very close
epistemological relationship to such a thing, an intimate acquaintance, as said by Russell, with
this thing. From such assertions, Russell derived that things which one is not especially
acquainted with can never be nominated. At the same time, Russell also affirmed that only
‘sense data’ (sprung from perception or sensation) may indeed provide such a familiar
relationship with objects. Thereupon, Russell concluded that since most things labeled by a
name are not sense data in their essence, they are not really names but a plethora of defined
descriptions in disguise.
Levinson (1983: 172) professes that “Russell’s analysis remained largely unchallenged
until Strawson (1950) proposed a quite different approach”. But for chronological purposes,
Strawson´s approach will be dealt with later on when the 1950’s come into the foreground.
Costa (1984, 2004) as well as Marcondes (2005) state that Morris was actually the first
philosopher to officially come up with the term pragmatics. Most likely influenced by the
American philosopher Charles Sanders Peirce´s work on semiotics (1897), Morris (1938)
devised a semiotic triangle to explain the function of signs in language analysis. In accordance
with Morris´ model, the three vertices of such a semiotic triangle would be accountable for
differing processes in relation to the usec40511(o )-143.612(r)3.21279(i)1.40381(1-.0204(e)-2.80762()-64(of)-7.42551( )-1e)-2.80761.40511(y )-143.617(i)1.400.42551( )-9(r)3.21279(i)--2.81ne ey orrt prionmc4051183068(t)1.40511( )-1279(i)1.40381(1-.0204)6.0217(e)-2.Td[(e)7.88ia clc ngnce309.96 -19.h.80762( 1(a)1.06464.89)1.40381(,)--441.489(t) 0 Td[(l)1.40511hrons r eeei, he ec
semiotics that deals with the relation between signs or linguistic expressions and their users,
as well as the way such signs are interpreted and made use of by their users.
Marcondes (2005) also adds that traditionally, both syntax and semantics were paid
higher attention to. On one hand, syntax studies the relations among the several signs and how
these basic linguistic unities are connected to each other to form complete propositions.
Therefore, syntactic theory is characterized as a formal science that defines the rules for
forming propositions, as abstract entities, from the set of possible combinations among the
different signs. Closely related to syntax is semantics that is concerned with the relation
between the signs and the objects they refer to, aiming at attaining the meaning of those
linguistic signs. As such, semantic theory is concerned with the meaningful content of the
signs and the truth-conditions of the propositions in which these signs are inserted into. On
the other hand, pragmatics had appeared in this context sort of uncommitted with a specific
scientific methodology to study language in use, being utterly dependable upon differing
As attested by Marcondes (2005), The German philosopher Rudolf Carnap (1891-
1970) resumed Morris´ studies and started developing a little further the distinction between
syntax, semantics and pragmatics as fields of study for language analysis. In 1942 he brought
into the light the consideration that a pragmatic study of language would seem misleading
since the heterogeneity and multiplicity of possible uses of linguistic expressions would
plausibly compromise a more systematic and theoretical approach to language. For Carnap,
every pragmatic attempt to analyze language would presuppose an abstraction and
generalization of such an array of multiple uses, striving to obtain common elements to
address language more methodically. According to Carnap, syntax and semantics should
really deserve higher homage as scientific theories for being able to convert concrete language
in use into more abstract levels of generalization. Thus, Carnap stated that semantics could de
facto have autonomy for abstracting specific variations of usage and drawing considerations
about the meaning of the terms, apart from their uses. And as for syntax, Carnap believed that
its task consisted in abstracting meaning so that linguistic expressions could be handled as
mere classes of signs for deriving the formal rules by which such categories interrelate.
It was not until the early 1950´s that such a pessimistic view about a pragmatic
approach to language analysis began to change.
As previously mentioned, Peter Strawson (1950) came up with a different
interpretation for analyzing language with no disregard or sign of contempt for its possible
uses. Levinson (1983) sheds light into his proposal by highlighting the most significant
conclusions that Strawson got to while scrutinizing Russell’s aphorisms.
In 1905, Russell tried to provide an explanation for the fact that sentences that lacked
proper referents, like for example [1], could be meaningful.
[1] The high priestess of Delphi is sick
According to Russell, sentence [1] can indeed be meaningful if just taken to be a false
assertion, since the ‘the high priestess of Delphi’ professes the existence of such an individual
(despite the fact there are no such high priestesses in Delphi in present-day terms).
Succeeding Russell’s view, Strawson (1950) rejected it holding that Russell failed in
distinguishing sentences from the uses of sentences to make statements. Following this view,
he asserted that what can be taken as true or false are not sentences but utterances. Therefore,
the statement uttered in [1] could very well be true in 2000 B.C. and false by A.D. 1640. What
actually mattered, Strawson argued, is that there is a noteworthy relationship between [1] and
[1] The high priestess of Delphi is sick
[2] There is a present-day high priestess of Delphi
That is to say, “that [2] is a precondition for [1] being judged as either true or false”
(LEVINSON, 1983:172). Such a particular relationship, as advocated by Strawson, is
different in nature from logical implications or entailments, deriving solely from complex
conventions about the utilization of referring expressions in day-by-day language. Hence,
Strawson (1950) started denominating such sort of presuppositional relationship between [1]
and [2] as pragmatic inferences, expressing that linguistic expressions in ordinary language do
not abide by the same Russellian rules, applicable to logical language.
From then on, in conformity with Marcondes (2005), the pragmatic theory started
unfolding into two lines of development within philosophy of language parameters: one
dealing specifically with deixis or indexical expressions and the other considering semantic
meaning as determined by language in use.
Following the first line, Yehoshua Bar-Hillel (1954) asserted that indexical expressions
(comprising both personal and demonstrative pronouns as well as time and place adverbs) are
dependable upon the context so that their meaningful content may have their reference
figured. For instance:
[1] They are coming here tomorrow
The understanding of such a sentence may not be entirely attainable if the reference of the
indexical expressions ‘they’, ‘here’ and ‘tomorrow’ are not taken into account. Therefore, one
may comprehend how vital it is to specify the context in which the sentence was uttered.
Pragmatics, according to this view, should contemplate both the contribution of such
indexical expressions as well as the need for interpreting them within the context of use, so
that the semantic meaning of the sentences in which they are applied may be thoroughly
attended to.
The second line also considers the notion of context and conceives the idea that
language is a form of action and not a way of describing reality.
In tune with this line, Marcondes (2005) declares that the Austrian philosopher
Ludwig Wittgenstein (1899-1951) developed predominantly in his ‘Philosophical
Investigations’ (1953) what may be fancied as a pragmatic conception. He contended the
meaning of a word as an indication of its use within a certain context and introduced the
notion of ‘language game’. From this standpoint, Wittgenstein envisaged that meaning should
not be construed as something stable and settled, inherently derive957(s)6.020417(m)1.4.6383(212.4459(a)-2.8-)10.6383(e.4485(r)10892( )-239.362(83(f)-7.423p0417(m(t)-9.23449(i)1.40381(oo9 )-26.5957(s)6.0204(t10.6383(e50]TJ-318 -1(l19.2l)1.40511(e)-2.80762-154.258(i)1.s)-4.6166( )-303.191(l)12.0421(isc)-13.4472(h )-110762(e)-2.807un279( )-5.31915(a)-2.80899.44 Td[(not32.976.809(w)9.23h6166(t)1.40381(i)1.40381(g)10.(l)12.0421l15(a)-2.80899.44 Td1279(l)12.0492(ge)-2.80891(g)10.6383(e)-13.1(n )-164.894t)-9.23384( )-122.349.44 Td[80892(o)10.6383(m)15.31915(d)10.4(a)-2.80763068(a)-13.44.894(i)12.0434 )-122.349.440 Td4459( )-143.617(c)7.8306d)10.6383(t32.976.up79(l)12.0[(not2.0421(isc)-1312.04342.0421p0762(c)-2.80762(l)1.40251(c)7.83068(a)-13.17(c)-13.4u91(de)-2.80762(c)-2.80762(l)1.404342.042162(i)-9.410.6(n )-143.617(c)]TJ306.24 0 Td[(ont)12.0434( )-1107204(e)-2.80762( )-154.2(c)-13.44504342.04212005))3.212250]TJ-352.23(,)-5.31383(f)-9.2t3( )-26.59511(e)-13.4459(32.976.279(l)12.0o9.362(t)1.4l(l19.2l)1381(ng)103319(i)1.40510511(e)-2.410.6)10.6383( )29.56 TL( )72 -19.44 Tk421(e)-2.80892(t)1.403813os)630381(h )-303.191(t)1.40381os)63049(or)3.21279(d )-15466(he)-2.80892( )-23.617(c)7.8306d)1020.6383(a)-13.4472(t)1.400381osedorrac(c)]TJ4(t10.6383(e50(a)-2.80763(t)1.40511(h )-90.4255(t)-9.23311os)6300511(r)3.21279( )-271.277(m)12.0434(1(r)3.20892(t)1.403813os)63084(i)1.40381(ont)1.40381(une)-)-2.8-e0762(r)3.21279(e)-203834(e)-13.4459(d )-164.894(m)-9.2331324.312.v83(e)-2.80892( )-2.80768(c)-2.803813osin ortng rt 92.553(oo9 )-26.59xtoo92(e)-13.4485(r)3.21(l19.2l)1.40511(e)-24ey
Just like in any sort of game such as chess, tennis or poker, the use of the term ‘game’
may imply different rules or even different characterizations, still it is possible to hold that all
of these games may possess some trivial features that enable one to get closer (to a higher or
smaller degree) to the other. Similarly, in language games words or expressions must be used
in accordance with the rules that define the dos and do nots as to the proposed objectives.
Hence, such usage rules (as a matter of fact pragmatic rules) make possible to ascribe the sort
of actions carried out by the players (speakers, writers). Analyzing the meaning of words
consists in situating them within the context of the game in which they appear and drawing
considerations about what the players intend to do with them.
The language games proposed by Wittgenstein have decisively contributed to the
development of philosophy of language by casting broader light into the idea that what is
apposite to language analysis are the functions that words may exert upon the language
games. Pursuing this view, rules of usage may define the purpose and the way in which words
can be utilized, at the same time they may attribute the circumstances under which such
usages may be employed.
Costa (1984, 2004) argues that Wittgenstein’s ‘language games’ furnished one of the
epistemological grounds for the appearance of the contemporary theories about the meaning
of words. In fact, Costa also asserts that up to Wittgenstein’s ‘language games’ the pragmatic
theory was in its classical period, characterized by incipient signs of its existence but without
being properly approached. From the 1960’s on, the modern period takes its course and
pragmatic phenomena starts to be more fully developed, hurrying theoreticians to define the
scope and specific object of study appropriately ascribed to a pragmatic theory.
The first modern theory that attempted to explain pragmatic phenomena in a
systematic manner was ‘The Speech Acts Theory’, proposed by the British philosopher John
Langshaw Austin (1911-1960), and published posthumously in 1962.
According to Marcondes (2005), Austin tried to demonstrate that language in use
could truly be regarded as the object of study for a systematic approach to pragmatics as long
as suitably developed conceptual gears were applied.
Following this line, Austin (1962) held that speech acts are the basic constituents for
using and interpreting natural language. He professed that speech acts are not mere sentences
endowed with truth-conditions as conceived by the traditional semantic theories of meaning.
Contrarily, he assumes that sentences are either used to describe facts and events as well as to
perform something. The examples underneath provide an illustration of that.
[1] Jordan was smooching Debbie in the living room.
[2] I promise I will pay you as soon as I get the money.
While sentence [1] describes a fact and event that may be true or false in relation with the
reality that it represents, sentence [2] performs the act of promising something without taking
into account whether it is true or false, for not describing a mere fact. Depending on the
circumstances in which it was uttered and the consequences of the realization of the act, one
may regard sentence [2] as successful or unsuccessful.
Later on, Austin realized that such a dichotomy between representative and
performative acts was rather inappropriate. He perceived that representative acts also possess
a performative dimension, i.e., describing is also a speech act that may be successfully or
unsuccessfully carried out. Simultaneously he noticed that performative acts also share a
representative function, since they keep a closer connection with the fact that is likely to take
place. From such evidences, Austin conceived a performative function of language that
should be applicable to all natural language analysis. From this general assumption, Austin
proposed that speech acts should be considered as the essential unities of signification,
consisting of three articulated dimensions: the locutionary act, the illocutionary act and the
perlocutionary act.
A. The Locutionary Act
All utterances comprise some sort of sense and reference. Such a feature is constituted
by three distinct acts: the phonetic act, evincing what sounds are pronounced; the
phatic act, comprising lexical and syntactical parameters which will determine the
order in which words are pronounced, and the rhetic act, expressing that the words
uttered will convey a sense and refer to something else.
B. The Illocutionary Act
An utterance is formed with some sort of function in mind such as offering, ordering
or promising. The illocutionary act expresses the force of an utterance, so that the
intentness comprised in it may be revealed.
C. The Perlocutionary Act
It is responsible for generating the effect intended by the illocutionary act, based on
the assumption that the addressee will recognize the effect intended by the speaker.
One of the main objectives of Austin’s theory is to define, in the sense of making
explicit, the illocutionary force that is comprised within the intended speech act that was
carried out, so that the identification of its typology may be attained.
Such a process relies on an array of presupposed conditions and circumstances for the
successful realization of speech acts, consisting predominantly of a combination of speaker’s
intentions and social conventions. Speaker’s intentions are thought to be subjective as well as
socially determined, whereas social conventions come up with diversified degrees of
formality, depending on the social event in which they are applied to. Such social conventions
make up the explicit and/or implicit rules that determine the customary pattern of behavior
that characterizes the speaker’s successful way of socially interacting. Austin has technically
denominated felicity conditions’ such expected or appropriate circumstances for the
performance of a speech act as intended by the speakers.
One of Austin’s latest contributions to his ‘Speech Acts Theory’ was proposing the
classification of the illocutionary forces comprised in utterances into five general types:
declarations, representatives, expressives, directives, and commissives.
A. Declarations – speech acts that change the world via their utterance.
Example: a. Priest: By the power invested in me I now pronounce you husband
and wife.
B. Representatives speech acts that reflect the speaker’s beliefs about the case or
event. Representatives consist of statements, assertions, conclusions and
descriptions, as illustrated below.
Example: a. The sun rises in the east and sets in the west.
b. Nazis hate Jews.
c. That was the hottest summer in my recollection.
C. Expressives – speech acts expressing psychological states, determining the way the
speakers feel.
Example: a. I’m really sorry to have disturbed you.
b. Congratulations, honey!
D. Directives speech acts designed to get someone else to do something, expressing
what the speakers want.
Example: a. Would you please close the damn door?
b. Keep your mouth shut, little rascal.
E. Commissives speech acts that express what the speakers intend to do in terms of
future action.
Example: a. I’ll crush your face just like a rotten potato.
As one may see, Austin was primarily concerned with devising a methodology that
enabled implicit elements to be brought into the surface, so that philosophical problems could
be dealt with by means of language in use. For him, language was a form of action that
provided the means for the realization of acts through the use of words.
Some years later, the American philosopher John R. Searle resumed Austin’s ideas,
and as stated by Costa (1984,2004), predicated that meaning and illocutionary act should
never be dealt with as synonyms. Bearing these assumptions in mind, Searle reformulated
Austin’s speech acts typology, coming up with:
A. Utterance Acts – articulation of phonemes.
B. Propositional Acts – speakers refers to an object, predicating something about it.
C. Illocutionary Acts – assertions, promises, requests, inter alia.
D. Perlocutionary Acts – the effect intended by an illocutionary act.
The reformulation of such components, as mentioned by Marcondes (2005), derives
from the development of an initial idea proposed by Searle in Speech Acts (1969). In
accordance with this idea, Searle attests that a speech act results from the combination of a
proposition p, endowed with a specific semantic content, establishing its relation with world
facts, thus likely to be either true or false, and an illocutionary force f that is added to the
proposition, so that the performance of a speech act may be carried out. Consider the ensuing
[1] The company is shut down.
[2] Shut down the company.
[3] If the company were shut down….
Statement [1] contains exactly the same propositional content of statement [2]
(imperative) and statement [3] (conditional). The difference among such utterances lies in the
sort of illocutionary force that is incremented to the same propositional content.
For analyzing the dissimilarities found among the different types of speech acts, one
must take into consideration not only the utterance itself and the semantic meaning of the
terms and expressions employed. It is also essential to proceed the identification of contextual
elements such as the role of the speaker, the set of situational rules and procedures that are
demanded under a specific circumstance, whether the participants of the social interaction are
abiding by such rules, and the intentions and objectives of both the speakers and the listeners.
Following this vein, it is possible to assert that the Speech Acts Theory encompasses
extra linguistic components that are fundamental for the realization of such acts. Thus, from
this pragmatic outlook on the theory that envisages language as a communicative act, one may
assert the need for allying the context in use with linguistic components in a manner that the
analysis of meaning may be more thoroughly grasped. In other words, The Speech Acts
Theory promotes the understanding that a pragmatic approach to the analysis of language
must be inserted along the semantic and syntactic pathway in order to enable that implicit
meaning be inferred conjoined with the explicit meaning of words, their syntactic alignment
and the truth-conditions of the propositions in which they appear.
Along with the Speech Acts Theory, Grice’s Theory of Implicatures and the Relevance
Theoretical Framework, proposed by Sperber and Wilson (both specified in the ensuing
sections), have constituted the essential cornerstones for the modern developments of a
pragmatic sub theory, attempting to attribute to it the status of an interdisciplinary science
with a well-defined object of study, and aiming at helping uncovering indeterminacies of
Levinson (1983) observes that semantic content seems likely to be the solid basis over
which other displays of meaning strike. From this assertion as well as from verified linguistic
phenomena that indicate the need of conjoined efforts between semantics and pragmatics,
Costa (1984, 2004) also endorses the view that the pragmatic theory is meant to complement
the semantic pathway. Such an interface between a semantic theory that considers the logical
truth-conditions of propositions, and a pragmatic theory anchored on Grice’s Theory of
Implicatures as well as Sperber and Wilson’s inferential/cognitive model delimitates the scope
in which the current study is going to be based on. Thereupon, the present theoretical study
aims at demonstrating as well that while entailments should be handled by semantic means,
implicatures should be dealt with via pragmatics.
The British philosopher Henry Paul Grice (1913-1988) came up in the late 1950’s with
a semantic theory that established the distinction between speaker’s meaning and the literal
one. Bearing such considerations in mind, in 1967 he launched an article entitled On Logic
and Conversation (published in 1975) that has impacted admirably the development of the
pragmatic theory, for introducing a highly efficient conceptual system to handle signification
in natural language and distinguishing when to make use of strictly logical procedures for
language analysis.
In that article, Grice introduces a new approach to deal with implicit meanings that are
commonly found in the utterances of people engaged in conversation. He points out that
implicit meaning should be understood as the amount of extra linguistic elements that are
beyond the literal meaning expressed by the words and sentences uttered by the speakers.
Still, it is the sort of meaning that in most cases the addressee is able to uncover naturally, but
within the realm of linguistic studies, only a pragmatic sub theory would be able to properly
deal with, since it is the type of meaning that is dependable upon the conversational context in
which words and sentences are attached to.
From this line of reasoning, one might consider that ‘literal meaning’ is what is said
explicitly by the chain of words articulated by the speakers, while ‘speaker’s meaning or
‘meaning-nn’, as proposed by Grice, could possibly be defined as what is ‘not said’, being
prone to be inferred by the addressee on the basis of contextual evidences. From this onset,
Grice comes up with a technical nomenclature, instituting the terms ‘implicate’, ‘implicatum’
and ‘implicature’ in order to compile around them an explanatory system of signification for
handling the meaning of what is ‘not said’.
In accordance with Grice’s postulates, such an additional meaning does not alter in
conformity with the listener’s interpretation of what is said; unlikely, meaning-nn can undergo
calculation by taking into account the set of sentences that were actually said as the dialogue
unfolded. Paraphrasing Faria (1999), it is the context promoted by the conversational piece
that provides sufficient and necessary grounds for the addressee’s understanding that the
speaker intended to say p + q, when actually they said just p.
For Grice, there are two fundamental kinds of implicatures: conventional and
conversational. The ensuing examples illustrate them.
[1] Angus is Scotch. Therefore, he is closefisted.
[2] He is rich but not stuffy.
‘Conventional Implicatures’ are those that are not necessarily dependent on the
conversational context, and their signification is imparted on the basis of explicit
presuppositions derived from the use of certain lexical items.
Thence, statement [1] explicitly says that Angus is Scotch and also that he is
closefisted. The connective ‘therefore’ generates such a conventional implicature because it
imparts the presupposition or stereotyped notion that Scottish people are closefisted by nature,
something that was evidently not literally said but conventionally implied.
Example [2] also implies that the rich are stuffy, although not said literally but inferred
from the use of the connective ‘but’.
As Costa (1984,2004) asserts, conventional implicatures result from the illocutionary
force imparted on certain lexical items, being therefore easily inferred by the participants of a
speech act.
[3] (Student) What do you really think about my master’s paper?
(Professor) You presented your ideas so well that I think you should seriously consider a
career outside academic life.
The above example illustrates a conversational implicature, which is dependant on the
conversational context and determined by general principles of the communicative exchange,
as proposed by Grice. Before defining it in a more accurate manner, let’s just say for the
moment that in example [3] the professor’s reply to the student’s question is ironic. In
accordance with the context, it implies that the student’s written production was so poor that
he or she should entirely reconsider the options at hand and direct his/her efforts to another
area of expertise; one which preferably does not involve academic research. Such
signification is by no means literally comprised in the professor’s utterance, but easily
inferred by the addressee on the grounds of logical reasoning and contextual evidences.
Before contemplating the conceptual idea of conversational implicature, it is
mandatory to address what Grice denominated the ‘Cooperative Principle’ and the ‘maxims of
Taking into account that data exchange is the utmost purpose of conversation, Grice
(1975) proposes a theoretical rational scheme that accounts for cooperative ends in
conversation. Such a theoretical model states that in conversation, people do not merely utter
sentences in a haphazard fashion; rather, they abide by certain conventional as well as
conversational principles. These principles seem to be subconsciously accepted and followed
by speakers/writers. The definition of Grice’s cooperative principle may be stated as follows:
contribute what is required by the accepted purpose of the verbal exchange.
Grice (1989) claims that the cooperative principle stands for the set of specific logical
assumptions about the cooperative nature of ordinary verbal interaction, and such a
cooperative nature intrinsically implies that speakers should use language sincerely (maxim of
quality), perspicuously (maxim of manner) and relevantly (maxim of relevance), while
providing sufficient information (maxim of quantity) in order to converse in an efficient,
rational, cooperative way.
The scheme that follows underneath reports the maxims of conversation in a more
suitable form.
A. Maxim of Quantity refers to the quantity of information that must be provided in a
message, being divisible into two sub maxims: (I) Make your contribution as
informative as it is required for the purposes of the exchange. (II) Do not make your
contribution more informative than it is required.
B. Maxim of Quality refers specifically to making a contribution that is true, being
related with two sub maxims: (I) Do not say what you believe to be false. (II) Do not
say that for which you lack adequate evidence.
C. Maxim of Relation – be relevant.
D. Maxim of Manner refers to being clear enough so that the message may be
effectively conveyed. It is connected with the following sub maxims: (I) Avoid
obscurity of expression. (II) Avoid ambiguity. (III) Be brief (avoid unnecessary
prolixity). (IV) Be orderly.
Although Grice attests that there may be other maxims such as the maxim of
politeness, he professes that in communication the four above-quoted maxims are enough to
guarantee and deal with the phenomenon designated as conversational implicature.
Levinson (1983) revisits Grice’s theory asserting that in communicating, people utilize
a set of specific logical assumptions originated from basic rational considerations that guide
the conduct of conversation. Such rational assumptions, or conversation maxims, can be
treated as sub principles of the cooperative principle, serving as general guidelines for an
effective use of language in conversation to improve cooperative exchanges.
In accordance with Grice (1975), while communicating people may handle the
cooperative principle in at least two different ways, depending on how speakers relate
themselves to the conversation maxims: (1) If speakers are observing the maxims in a direct
way, the may rely on the addressee to amplify what they say by means of unequivocal
inferences denominated ‘generalized conversational implicatures’ and ´conventional
implicatures’, which are not necessarily dependent on the conversational context, and while
the former are thought to take into account the logical standard meaning presented by
speakers´ utterances, the latter are assumed to be explicit presuppositions derived from the use
of certain lexical items. (2) If speakers deliberately and ostensibly flout some of the maxims
in order to explore them for communicative purposes, they make use of inductive inferences
within the context, being thus entitled ‘particularized conversational implicatures’ or simply
‘implicatures’. GRICE (1989:30) predicates that “when people flout the maxims, we have
implicatures”. Such floutings of the maxims may generate many figures of speech such as
irony and metaphor, which in turn will lead listeners on to a broader range of distinctive
deductive inferences.
In Studies in the Way of Words (1989), Grice claims that for figuring out the additional
conveyed meaning of a certain implicature, the listener will rely on the following evidences:
(1) The conventional meaning of the words uttered;
(2) The cooperative principle and the maxims of conversation;
(3) The linguistic context of the utterance;
(4) One’s background knowledge (culturally pre-existing knowledge structures that
are used to interpret new experiences);
(5) The fact that all these previous relevant items are part of both the speaker and
addressee’s mutual knowledge.
Following Costa’s trail (1984, 2004), there appears to be three distinct situations that
may yield implicatures, depending on how the speakers relate themselves to the cooperation
I. No maxim is flouted
[1] (A) – I surely could use a drink.
(B) – There’s a pub two blocks away.
[2] (A) – I can’t find one of my shoes.
(B) – Come on Cinderella, let’s go!
In both examples (B) seems liable to be respecting the cooperation principle, despite
the fact the maxim of relevance appears to be breached. As a matter of fact, (A) is able to
figure out the conversational implicature just because he/she unconsciously believes that (B)
is abiding by the rules of the conversation.
In [1], (A) figures out that (B) is saying that there is a pub close by because (A)
believes that (B) is implying the pub must be open and selling beverages that may provide a
good solution for the satisfaction of the desire stated by (A). Example [2] is a little more
refined in the sense that (A) perceives that (B) is being cooperative, and by providing a
cultural context mutually known by (A) and (B) that comprises the knowledge that Cinderella,
as a fairy tale character who loses one of her crystal shoes, is being comparable to (A) for the
reason of not being able to find one of his/her shoes. At the same time, (A) can infer that (B)
is being ironic and conveying the explicit message that there is no more time to spare,
simultaneously with the implicit one that (A) must find a fast solution for the shoe matter.
II. A maxim is flouted so that another one may be not
[1] (A) How can I say ‘thank you’ in Czech?
(B) I’m not a walking encyclopedia.
[2] (A) Don’t you think my girlfriend is hot?
(B) For me, my buddies’ girlfriends are just like one of the guys.
Both examples illustrate that the replies provided by (B) flout the maxim of quantity,
since they are not as informative as demanded by the conversation piece. What seems to be
most likely is that speakers (B) are replying in this manner to imply that they are unable or
unwilling to provide a more satisfactory answer to (A)’s requests. In case [1], instead of
providing an accurate answer, (B) decided to respect the maxim of quality by offering a vague
answer that conveys an implicit yet clear message that he/she is unable to come up with the
desired answer, for not possessing the necessary information required. In case [2], a simple
yes/no reply to (A)’s question would suffice. Instead, (B) opts for flouting the maxim of
quantity as well as the maxim of relevance in an attempt to keep himself uncommitted with
the possible unfoldings of (A)’s request. It seems that, by providing a neutral answer to (A),
speaker (B) is unwilling to express what he really thinks about (A)’s girlfriend. His reply may
not be regarded as an insult; one as if it asserted that his buddy’s girlfriend looks like a man or
lacks the degree of femininity, commonly attributable to women. On the contrary, speaker (A)
may infer among several other possibilities that his friend (B) is refusing to provide the
requested answer just because he does not wish to offend him by saying something
inappropriate such as ‘your girlfriend is a dog; really ugly’ or ‘she’s so delicious I could eat
her with a spoon’.
III. A maxim is flouted in order to generate a conversational implicature
Figures of speech as attested by Costa (1984, 2004) generally fall under this case.
A. Maxim of quantity is flouted due to lack of information
[1] (A) In your opinion, who is better George Bush or Hugo Chaves?
(B) A nazi is a nazi.
The tautology expressed by (B)’s statement violates the maxim of quantity by being
utterly redundant. Yet, (B) is being cooperative and saying more than actually he/she did. As
a matter of fact, (B) is implying that both Bush and Chaves are as catastrophic to their
countries as nazis were for Germany along the period of World War II. By making use of
such tautology, speaker (A) may infer that speaker (B) is expressing that both presidents are
fascist dictators who aim the autocratic centralization of government for attaining personal
objectives of their own. Such a tautological reply may also imply that speaker (B) thinks that
Bush is just as bad as Chaves.
B. Maxim of quantity is flouted due to excess of information
[2] (A) How are you doing?
(B) I lost my job, my husband left me, I was evicted and I’ve just found out I got
Such an excessive amount of information comprised in (B)’s reply is possibly meant
to imply that her life is going really badly. By emphasizing all the misfortunes that have
befallen upon (B), speaker (A) may also gauge the level of stress, anxiety and desperation that
(B) is undergoing.
C. Maxim of quality is flouted due to false assumptions
[3] (A) Does she really love you?
(B) Yeah. As much as an ulcer.
This example characterizes what is known as irony. (B)’s reply expresses something
that he does not believe to be true; namely that the woman mentioned in (A)’s question really
loves him. Nobody loves an ulcer, apart from masochists. Therefore, by stating something that
is held to be contrary to the truth, speaker (B) is actually implying the opposite, i.e. she does
not love him.
[4] (A) Are you cold?
(B) My feet are like blocks of ice.
Example [4] characterizes a metaphor. Both (A) and (B) bear in mind that a foot can
not be a block of ice, except for sculpting purposes in very stern climates. Therefore, (B)’s
reply implies the assumption that he/she is feeling his/her feet extremely cold.
[5] (A) You look awful! Have you been crying?
(B) I cried a river of tears.
This example illustrates what is known as hyperbole; an extravagant exaggeration
used as a figure of speech. Speaker (A) clearly understands that speaker (B) has implied that
he/she cried a great deal, in spite of holding that it is not humanly possible to cry as much as
required by the water volume of a river.
D. Maxim of quality is flouted due to lack of appropriate evidence
[6] (A) What is Angelina Jolie doing these days?
(B) She must be after some hot chick for fun.
Despite the fact there is lack of evidence to corroborate what (B) is saying, speaker
(A) understands that (B) means to imply that the famous celebrity is a lesbian with a grand
appetite for other beautiful women. Such implicatures may be derived from assumptions that
come about by means of rumors, overgeneralizations based upon observation as well as other
E. Maxim of relation is flouted
[7] (A) How much do you earn in your current position?
(B) I must see the dentist.
(A) Hey, How much?
(B) I think I’ve got a cavity in one my molars.
It may be assumed in the abovementioned context that speaker (B) is deliberately
flouting the maxim of relevance to imply that he is unwilling to answer (A)’s question. As the
dialogue unfolds, (B) keeps on beating round the bushes as if (A) were interested about (B)’s
dental problems so that (A) may figure that wild horses will not drag the answer out of
F. Maxim of manner is flouted due to ambiguity
[8] (A) Did you enjoy the food at the new Russian restaurant?
(B) The restaurant is lavishly decorated.
This case illustrates that (B)’s reply to (A)’s question is ambiguous. By stating that
the decoration at the new Russian restaurant is lavish, (A) may infer that is the only good
feature worth mentioning about the place. In other words, by saying nothing about the food,
that is clearly the main topic of conversation, (B) may be implying that he did not enjoy the
food he/she ate there.
G. Maxim of manner is flouted due to vagueness
[9] (Lloyd) Let’s do that thing?
(Debbie) What thing are you talking about?
(Lloyd) That thing we like so much to do for relaxing.
(Debbie) Oh, that thing! Are you in the mood, naughty boy?
(Lloyd) I surely am.
Such a dialogue illustrates an intimate exchange between Debbie and Lloyd. Lloyd
purposefully avoids saying what exactly ‘that thing’ is for a diversified number of possible
reasons. Perhaps there may be some other people listening to what they are saying, or maybe
‘that thing’ is not supposed to be uttered because Lloyd does not want to sound blunt, or still
Lloyd believes that by being subtle Debbie will acquiesce more easily to his desire. Despite
the vagueness involved in ‘that thing’, Debbie infers what he is talking about and by hinting
clues such as ‘are you in the mood’ and ‘naughty boy’, one may infer that perchance they are
talking about sexual intercourse.
H. Maxim of manner is flouted due to lack of conciseness
[10] (A) What is a platypus?
(B) Platypus, plural form platypuses or platypi, is a small aquatic egg-laying
mammal of Australia, also known as the ‘land down under’, with webbed feet and
a fleshy bill like a duck’s. The etymology of the word is New Latin, from Greek
platypous meaning “flat-footed”, from platys “broad, flat” plus pous “foot”.
By answering in such a prolix manner, (B) may be implying that he /she is a true
savant and parade around his/her level of specialization. At the same time, (A) might possibly
infer that speaker (B) is conceited and loves showing off, among other possibilities.
I. Maxim of manner is flouted due to lack of order
[11] (A) Do you speak English very well?
(B) Me speak well very English.
Speaker (B) deliberately breaks the order to imply that he/she is not a fluent English
speaker, or perhaps to imply that he/she speaks well but is teasing speaker (A).
2.2.1 Implicatures and Their Properties
Before systematizing the properties of the implicatures, it seems advisable to
distinguish conventional implicatures from conversational implicatures. While the former are
attached to the conventional force derived from the meaning of the linguistic expressions, and
recognized by the listeners on the basis of their linguistic intuition, the latter are dependant on
deductible calculation from the listener, so that their meaning may be attained. Grice (1989)
attributes five properties to conversational implicatures: cancelability, calculability, non-
detachability, non-determinability and non-conventionality, not to mention the fact that they
may be carried by virtue of the manner of expression.
1. Cancelability
A conversational implicature is always susceptible to being cancelled since the due
respect to the cooperative principle may be breached under a particular context or by the
addition of an extra phrase. Examples [1] and [2] illustrate the respective cases.
[1] (A) John, can I borrow some aspirin?
(B) Don’t do that!
Speaker (A) might have understood that (B) refused her request of borrowing aspirin.
But (B) may be violating the cooperative principle just because (A) has messed up his work
papers and, consequently, he utters a protest of indignation.
[2] (A) How many pets have you got?
(B) Two… two dogs and an iguana.
Initially (A) infers that (B) has got two pets and no more. But by adding the phrase
‘two dogs and an iguana’, (B) implies something else; namely that he/she has got three pets
altogether. Thus, (A)’s initial inference is cancelled.
2. Calculability
Grice (1975) claims that every conversational implicature must be derived from
logical calculation; i.e. from assumptions based on what is said by the speaker, the listener
will get to deductible conclusions. For effectively attaining the desired conclusions, Grice
provides a classical model of inferential calculation (dealt with in the next chapter) that
supposedly follows a series of steps common to logical reasoning. For the moment, let’s
consider the following example, somewhat plagiarized from Costa (1984, 2004).
[1] (A) My head is throbbing with pain.
(B) There’s a drugstore round the corner.
Believing that (B) is abiding by the cooperative principle, (A) reckons that if (B) said
that there is a drugstore round the corner, it is because (B) intends to imply that: (A) had
better go to the drugstore, the drugstore must be open, the drugstore must have headache
medicine, (A) should acquire it, and that all these steps will help solving (A)’s problem.
3. Non-detachability
As mentioned before, conversational implicatures are not dependant upon the meaning
of the words uttered. Non-detachability is a property that states the implicit content of a given
utterance may be preserved if the expression(s) triggering it is (are) replaced by synonym
expressions. Though largely disputable by Grice’s critics, who affirm the implicatures derived
from floutings of the maxim of manner are immune to this property, let’s consider the
examples underneath.
[1] (A) I’m so hungry I could eat a horse.
(B) Dinner will be ready soon.
[2] (A) I’m starving.
(B) The food is almost on the table.
Both exchanges warrant the same implied meaning; that is, the solution for (A)’s
problem lies on waiting a little while until the food is ready to be served. Therefore,
independently of the mode of expression utilized in [1] or [2], the same conversational
implicature is attained.
4. Non-determinability
Metaphors epitomize the undeterminable character of conversational implicatures.
Example: (A) Jealousy - a green-eyed monster!
Such an utterance violates the maxim of manner, allowing the listeners to infer an
open set of possible implied meanings. (A) could be implying that green as a color symbolizes
envy, and that envy comes from the Latin expression invidére, closely connected with vidére
meaning to see, thus alluding to the green eyes. Still monster could be a symbol for an
extremely wicked or cruel person. Therefore, one could claim that jealousy, in the sense of
envy, turns a person into a wicked creature. Independently of the intended meaning, one may
notice that the sort of calculation involved in this inference may lead listeners on to
contemplate (A)’s utterance in a myriad of alternative ways, characterizing its indeterminacy.
Only the context in which utterance (A) appeared may limit the choices of which inferences
are more likely to take their course.
5. Non-conventionality
Such a property establishes the distinction between what Grice denominates
conventional and conversational implicatures. As a matter of fact, Grice (1989) asserts that a
conversational implicature is liable