3. Literal meaning vs.

Speaker’s meaning
Let us take stock. The view discussed in section 1 was based on the following
assumptions:
(i) semantics and pragmatics are two complementary, non-overlapping disciplines.
12
(ii) pragmatics deals with the use of language.
(iii) semantics deals with content and truth conditions.
Since words-world relations in natural language (hence content and truth
conditions)
cannot be studied in abstraction from use, those assumptions form an inconsistent
triad —
or so it seems. Semantics cannot be legitimately contrasted with pragmatics,
defined as the
theory of use, if semantics itself is defined as the study of words-world
relations.
In section 2 we entertained the possibility of giving up (iii). Following Katz, we
can retreat to the view that semantics deals with the conventional meaning of
expressiontypes
(rather than with content and truth conditions). But we have just seen that the
meaning of at least some expressions is best construed as a convention governing
their
use. It follows that the theory of meaning and the theory of use are inextricably
intertwined, in a manner that seems hardly compatible with (i). Be that as it may,
most
semanticists are reluctant to give up (iii). For both philosophical and technical
reasons,
they think the denotation relation must be the cornerstone of a theory of meaning.
As
David Lewis wrote in a famous passage, “semantics with no treatment of truth
conditions
is not semantics” (Lewis 1972: 169). I will return to that point below (§4).
An attempt can be made to save the triad, by focusing on the distinction between
LITERAL MEANING and SPEAKER’S MEANING. What a sentence literally means is
determined by the rules of the language — those rules that the semanticist attempts
to
capture. But what the speaker means by his utterance is not determined by rules. As
Grice
emphasized, speaker’s meaning is a matter of intentions: what someone means is what
he
or she overtly intends (or, as Grice says, “M-intends”) to get across through his
or her
utterance. Communication succeeds when the M-intentions of the speaker are
recognized
by the hearer.
This suggests that two distinct and radically different processes are jointly
involved
in the interpretation of linguistic utterances. The process of SEMANTIC
INTERPRETATION
is specifically linguistic. It consists in applying the tacit theory that speaker-
hearers are
said to possess, and that formal semantics tries to make explicit, to the sentence
13
undergoing interpretation. By applying the theory, one can deductively establish
the truth
conditions of any sentence of the language. To do so, it is argued, one does not
need to
take the speaker’s beliefs and intentions into account: one has simply to apply the
rules.
In contrast, the type of competence that underlies the process of PRAGMATIC
INTERPRETATION is not specifically linguistic. Pragmatic interpretation is involved
in the
understanding of human action in general. When someone acts, whether linguistically
or
otherwise, there is a reason why he does what he does. To provide an interpretation
for the
action is to find that reason, that is, to ascribe to the agent a particular
intention in terms of
which we can make sense of the action.
Pragmatic interpretation thus construed is characterized by the following three
properties:
• CHARITY. Pragmatic interpretation is possible only if we presuppose that the
agent is
rational. To interpret an action, we have to make hypotheses concerning the agent’s
beliefs
and desires, hypotheses in virtue of which it can be deemed rational for the agent
to behave
as she does.
• NONMONOTONICITY. Pragmatic interpretation is defeasible. The best explanation we
can offer for an action given the available evidence can always be overridden if
enough
new evidence is adduced to account for the subject’s behaviour.
• HOLISM. Because of its defeasibility, there is no limit to the amount of
contextual
information that can in principle affect pragmatic interpretation. Any piece of
information
can turn out to be relevant and influence the outcome of pragmatic interpretation.
The three features go together. Jointly they constitute what we might call the
HERMENEUTIC character of pragmatic interpretation. It strikingly contrasts with the
algorithmic, mechanical character of semantic interpretation.
It is important to realize that, on this view (which I will shortly criticize),
semantic
competence involves more than the ability to determine the context-independent
meaning
of any well-formed expression in the language. It also involves the ability to
assign values
to indexical expressions in context. Those assignments are themselves determined by
14
linguistic rules, which linguistic rules constitute the context-independent meaning
of
indexical expressions. In virtue of its linguistic meaning, an indexical expression
like I
tells you three things: (i) that it needs to be contextually assigned a value; (ii)
which aspect
of the situation of utterance is relevant to determining that value; and (iii) how
the value of
the indexical can be calculated once the relevant feature of the context has been
identified.
If one adds to one’s knowledge of the language a minimal knowledge of the situation
of
utterance — the sort of knowledge which is available to speech participants qua
speech
participants — one is in a position to assign contextual values to indexicals,
hence to
determine the truth conditions of the utterance.
From what has been said, it follows that the context of use plays a role both in
semantic and in pragmatic interpretation. But it plays very different roles in
each, and it
can even be denied that there is a single notion of “context” corresponding to the
two
roles. According to Kent Bach, there are two notions of context: a narrow and a
broad one,
corresponding to semantic and pragmatic interpretation respectively.
Wide context concerns any contextual information relevant to determining the
speaker’s intention and to the successful and felicitous performance of the speech
act...
Narrow context concerns information specifically relevant to determining the
semantic
values of [indexicals]... Narrow context is semantic, wide context pragmatic.6
In contrast to the wide context, which is virtually limitless in the sense that any
piece of
information can affect pragmatic interpretation, the narrow context is a small
package of
factors involving only very limited aspects of the actual situation of utterance:
who speaks,
when, where, to whom, and so forth. It comes into play only to help determine the
reference of those few expressions whose reference is not fixed directly by the
rules of the
language but is fixed by them only “relative to context”. And it does so in the
algorithmic
and non-hermeneutical manner which is characteristic of semantic interpretation as
opposed to pragmatic interpretation. The narrow context determines, say, that I
refers to
John when John says I quite irrespective of John’s beliefs and intentions. As
Barwise and
Perry write, “even if I am fully convinced that I am Napoleon, my use of ‘I’
designates
15
me, not him. Similarly, I may be fully convinced that it is 1789, but it does not
make my
use of ‘now’ about a time in 1789” (Barwise and Perry 1983: 148).
The view I have just described is very widespread and deserves to be called the
Standard Picture (SP). It enables the theorist to maintain the three assumptions
listed at
the beginning of this section. Semantics and pragmatics each has its own field of
study.
Semantics deals with literal meaning and truth conditions; pragmatics deals with
speech
acts and speaker’s meaning. To be sure, the “context” plays a role in semantic
interpretation, because of the context-dependence of truth-conditional content. But
that is
not sufficient to threaten assumption (i). Because of context-dependence, semantics
cannot
deal merely with sentence-types: it must deal with OCCURRENCES or sentences-in-
context.
But this, as Kaplan writes, “is not the same as the notion, from the theory of
speech acts,
of an utterance of an expression by the agent of a context” (Kaplan 1989b: 584):
An occurrence requires no utterance. Utterances take time, and are produced one at
a
time; this will not do for the analysis of validity. By the time an agent finished
uttering
a very, very long true premise and began uttering the conclusion, the premise may
have
gone false... Also, there are sentences which express a truth in certain contexts,
but not
if uttered. For example, ‘I say nothing’. Logic and semantics are concerned not
with
the vagaries of actions, but with the verities of meanings. (Kaplan 1989b: 584-5)
Moreover, as we have seen, the context which is appealed to in semantic
interpretation
differs from the “wide” context which features in pragmatic interpretation.
Semantics
deals with occurrences, narrow contexts, and literal meaning; pragmatics deals with
utterances, wide contexts, and speaker’s meaning. Appearances notwithstanding, the
two
types of study do not overlap.