iv. If f is an expression in CS
a
and f cannot be determined from
SYNTAX
a
by application of Rules R
1
...R
n
, then “ f amounts
to the presupposition of the antecedent, constructed by sub-
stituting variables for the [necessary elements] in the CS of
the antecedent” (Culicover and Jackendoff 2005:276)
c. Phonological interpretation rules:
should
p
/S2d/, etc.
Finally, one could imagine that the speaker who utters the sounds corre-
sponding to Jill should, too has produced a defective syntactic structure and
something with a non-type- meaning (perhaps something of type ),
and the hearer arrives at the assertion made and the proposition meant by
means of a pragmatic process. On such an approach, only a sentence
pragmatic
is produced.
The picture that emerges can be conveniently viewed in the following di-
agram adapted from Stainton 2006:37; the various types of ellipses can be
seen as operating at various levels or mappings.
(10)
a. Sound pattern (P) ⇔
phon
Syntax (S) ⇔
sem
Encoded content (M)
⇔
prag
Speech act content (A)
b. P ⇐ellipsis
syntactic
⇒ S ⇐ellipsis
semantic
⇒ M ⇐ellipsis
pragmatic
⇒
A
Stainton’s project in his 2006 book and in a great many articles leading up
to it is to defend the following premises:
(11)
Premise 1: Speakers genuinely can utter ordinary words and phrases
in isolation, and thereby perform full-fledged speech acts.
Premise 2: If speakers genuinely can utter ordinary words and phrases
in isolation, and thereby perform full-fledged speech acts, then such-
and-such implications obtain. (Stainton 2006:3)
Premise 1 (P1) can also be stated in the terms introduced above:
(12)
“ordinary words and phrases, with the syntax of words and phrases,
are not [always, JM] sentences
syntactic
or sentences
semantic
, but they are
[sometimes, JM] nevertheless sentences
pragmatic
.” (Stainton 2006:32)
Stainton has set the bar very high for anyone who wishes to dispute him in
his carefully argued and excellent (in many senses—methodological, exeget-
ical, analytical, and empirical) book. Here I wish mostly to concentrate on
the data he adduces to establish P1 and his discussion of it and of the various
responses to it. The reason he is so persuasive in his claiming of P1 is exactly
because of the wide range of data he considers, the kinds of sophisticated
views he has of it, and of the ecumenical nature of the data sources he brings
to bear on the question.
He is primarily interested in three sets of cases (all examples except the
ones from Mason’s novel are from Stainton 2006 in various places, occasion-
ally slightly modified; a convenient listing of them can be found on p. 83),
depending on what kind of phrase gets pronounced.
The first group consists of the pronunciation of sounds corresponding to
predicates which denote properties. (Stainton is admirably careful about his
phrasing of these things, which I’ll try to emulate, though I may fall into
sloppy ways at times when the details shouldn’t matter.)
(13)
Properties applied to a manifest object
a. Sanjay and Silvia are loading up a van. Silvia is looking for a
missing table leg. Sanjay says, ‘On the stoop.’
b. Anita and Sheryl are at the cottage, looking out over the lake.
Watching a boat go by, Anita says, ‘Moving pretty fast!’
c. Jack holds up a letter and says, ‘From Spain!’
d. A car dealer points at a car and says, ‘Driven exactly 10,000km.’
e. On a bottle of cold medicine: ‘Recommended for ages 6 and
older.’
f. She looked up at Nok Lek, who watched the forest nervously.
“I told you, one of Anthony Carroll’s best men.” (Daniel Mason,
The piano tuner
, Vintage: New York, 2002, p. 159)
In this first case, consider (13a). In it, Stainton claims (and I concur) that
Sanjay can be taken as having asserted of the table leg a de re proposition,
namely that it, the table leg, is on the stoop. The criterion used for judg-
ing whether an assertion of a proposition has occurred is whether or not we
have an intuition that Sanjay can be right or wrong, whether the proposi-
tion expressed can be true or false, and hence whether or not Sanjay can lie.
Here, Stainton rightly claims that we have the strong intuition that if Sanjay
knew that the table leg was not on the stoop, his utterance of (13a) would
count as a lie. Stainton claims that Sanjay makes his assertion by virtue of
the meaning of the phrase uttered, which has the syntax merely of a preposi-
tional phrase PP, not embedded in further, unpronounced, syntactic structure,
and has the meaning of a property of type
. This property is semanti-
cally unsaturated, but needs an argument to be the content of an assertion.
This argument is provided by the actual table leg, here manifest in Sperber
and Wilson’s 1986 sense (though not necessarily the object of direct percep-
tion), combined with the content of the uttered phrase. This ‘combining’ is
by function-argument application not to items of particular types in the type-
theoretic sense of (7b) above, but of mental representations (in Mentalese).
These mental representations come about, in this case, through two differ-
ent mechanisms: the representation of the property comes about through the
decoding of the linguistic signal and is the output of the language faculty;
the representation of the object comes about through other faculties of the
mind, be they memory, vision, systems that regulate planning, goal-setting,
understanding intentions of agents, etc.
The second subcase is that of the pronunciation of sounds corresponding
to noun phrases which denote individuals, such as names and definite descrip-
tions as in the following.
(14)
Individuals as arguments of a manifest property
a. A woman is coming through a door, and a linguist turns to her
friend and identifies the new arrival by saying, ‘Barbara Partee.’
b. A girl is doling out jam and says, ‘Chunks of strawberries.’ Her
mother nods and says, ‘Rob’s mom.’
c. After some weeks one summer of unusually cold weather in Man-
itoba (a part of Canada where the summers are usually warm),
Alice, looking at the sky, says to Bruce (who has just returned
from a trip to Spain), ‘Nova Scotia.’
d. Edgar didn’t have time to ask what this was, for at that instant,
from behind the stage rose a plaintive wail. He caught his breath.
It was the same tune he had heard that night when the steamer had
stopped on the river. He had forgotten it until now. “The ngo-gyin,
the song of mourning,” said Nash-Burnham at his side. (Daniel
Mason, The piano tuner, Vintage: New York, 2002, p. 140)
In these cases, the relevant property may be something like ‘(is) the iden-
tity of the person coming through the door’, ‘(is) the person responsible
for there being chunks of strawberries in the jam’, or ‘(is) the song being
heard’. These properties, in their Mentalese representations, are combined by