Three types of ellipsis
∗
Jason Merchant
Abstract
The term ‘ellipsis’ can be used to refer to a variety of phenomena:
syntactic, semantic, and pragmatic. In this article, I discuss the recent
comprehensive survey by Stainton 2006 of these kinds of ellipsis with
respect to the analysis of nonsententials and try to show that despite
his trenchant criticisms and insightful proposal, some of the criticisms
can be evaded and the insights incorporated into a semantic ellipsis
analysis, making a ‘divide-and-conquer’ strategy to the properties of
nonsententials feasible after all.
1.
Introduction
A character in Carlos Fuentes’s 2002 novel The Eagle’s Throne (trans. Kristina
Cordero; Random House: New York, 2006) says with self-disdain and melan-
choly (p. 93):
Did you know I’ve learned to speak like an Anglo-Saxon, without articles or
context?
“Exactly.”
“Done.”
“Nothing.”
“Careful.”
“Perfect.”
∗
This paper owes an enormous debt to a large number of people over the years since I
first began working on it, but special mention must be made of Rob Stainton, whose work
was the original impetus for it and whose comments at Rutgers and in Paris in 2007 led to
numerous improvements. Thanks also to Ernie Lepore, François Recanati, Jason Stanley, and
the participants in the Leverhulme Foundation workshop organized by Laurence Goldstein at
Canterbury in 2008.
“Warned.”
“Face consequences.”
I say these things, nothing else.
As it turns out, such seemingly telegraphic speech is by no means limited
to the Anglo-Saxon world. The question is just what such utterances could
and do mean ‘without context’ and with, and what exactly a speaker who
utters such phrases says, means, and conveys.
Speakers convey information by a variety of means: the one studied most
by those of interested in language and meaning is the content conveyed with
linguistic means, a content whose nature is determined by the context of an
utterance and the meaning of the elements used in the utterance, by virtue
of their form and other factors. One of the most interesting current ques-
tions regarding this fact is where and how to draw the boundary line between
pragmatics and semantics. A standard approach is to distinguish between
speaker’s meaning and sentence meaning, but the latter term—sentence—has
a number of uses (and the former isn’t simple either) that must be distin-
guished.
A very salutary typology of things we call ‘sentences’ is provided by
Stainton 2006, as in (1):
1
(1)
Three senses of ‘sentence’ (Stainton 2006:31)
a. sentence
syntactic
: an expression with a certain kind of structure/form
b. sentence
semantic
: an expression with a certain kind of content/meaning
c. sentence
pragmatic
: an expression with a certain kind of use
We standardly conceive of an utterance of for example (2a) as consisting
of a 4-tuple of the form in (2b), which follows the general pattern given in
(2c), where the first member of the 4-tuple is the phonological representation
P
, the second the syntactic S, the third the semantic M, and the fourth the
‘speech act content’ C
SA
(the particular representations used here for illustra-
tive purposes are of course in their details immaterial).
(2)
a. Abby left.
b. < /æbi lEft/, [
S
[
NP
Abby ] [
V P
left ] ], left(abby), left(abby)
M
,g,w,i
=
1 >
1
See also Bloomfield 1914 for a tracing of the notion ‘sentence’ in ancient and 19th
c. grammarians and for critical discussion.
c. < P, S, M,C
SA
>
In the standard case, the three final members of the 4-tuple correspond
to sentences in the syntactic, semantic, and pragmatic sense intended, and
have characteristic types: for sentences
syntactic
, this type is S (or its modern
descendants in some theories, TP, IP, or CP), for sentences
semantic
, the type is
(or ), and for speech acts, the type has no standard name known to
me (nor representation, for that matter—that in (2b) is just roughly sketched
as the kind of thing that could be the argument of an ‘assertion’ operator), but
it ranges over things like assertion, command, question—call it type SA.
The main task of natural language theorists is to give a general account of
how the four members of such tuples are related to one another. One widely
adopted view takes it that there are mappings between the representations as
follows (other views have also been proposed, that allow direct interactions
between the phonology and semantics, for instance):
(3)
a. P ⇔
phon
S
b. S ⇔
sem
M
c. M ⇔
prag
A
On this view, the equivalence between both the propositional content and
illocutionary force of (4a) and those of (4b) gives rise to various analytical op-
tions, all of which have the common goal of capturing the fact that a speaker
can use (4a) to assert that Jill should collect butterflies, just as much as she
could use (4b), and that this is a contingent fact about English.
(4)
VP ellipsis in English
a. Bill should collect butterflies. Jill should, too.
b. Bill should collect butterflies. Jill should collect butterflies, too.
The first possibility is what Stainton perspicuously calls ellipsis
syntactic
,
which involves positing an unusual mapping between the syntax and phonol-
ogy, but claims that otherwise (4a) and (4b) are identical. In particular, the
phrase structure and lexical insertion rules of English work in both cases as
usual (illustrated with the structures in (5a,b), the semantic combinatorics
work as usual (say, via functional application, as in (6)), but there is some-
thing special about the pronunciation of the unheard VP. One (lexicalist) way
of cashing this out is the following: posit a special feature, E (for Ellipsis),