banner
SIL International Home

SIL Electronic Book Reviews 2009-012

URL: http://www.sil.org/silebr/2009/silebr2009-012

A Theory of Ellipsis

By Marjorie J. McShane

Oxford: Oxford University Press, 2005. Pp. 272. hardback $109.99. ISBN 978-0195176926 (hardback), 0-19-517692-8 (hardback).

Reviewed by Stephen H. Levinsohn

SIL International

McShane begins by highlighting the existence, not only of syntactic ellipsis, but also of semantic ellipsis. “Syntactic ellipsis is the nonexpression of a word or phrase that is, nevertheless, expec¬ted to occupy a place in the syntactic structure of a sentence. For example, in Mary got an A on the math test and Louise O a B, the verb ‘got’ in the second conjunct is elided” (p. 3). In contrast, semantic ellipsis is “the nonexpression of elements that, while crucial for a full semantic interpre¬tation, are not signaled by a syntactic gap”. For instance, the meaning of He is reading Tolstoy is actually “He is reading a book written by Tolstoy”, and the elision of “a book written by” is not permitted in Chinese (ibid.).

“Ellipsis is a universal property of natural language, but its scope and means of realization differ substantially from language to language … Working on ellipsis … requires reference to syntax, lexical semantics, discourse, prosody, semantics and stylistics” (ibid.). McShane considers that “current treatments have one thing in common: they address those subtypes of ellipsis that are most salient for a given established framework” (p. 4). Her interest, in contrast, is in “painting the whole canvas of ellipsis at a single go, considering all the factors from all the realms that deter¬mine when a speaker of some language should, or even virtually must elide categories”. Also, “the results must be generalizable, with parametric variation across languages” (ibid.).

As I read the book, I wondered whether it really represented a theory of ellipsis. McShane insists that it does, in the light of the definition offered by Nirenburg and Raskin’s (2004). She states, “a theory can be defined as a combination of its function and its components. Its function must be to promote (1) selection of the best description methodology and (2) subsequent evaluation of the quality of the description, leading to iterative improvements in both methodology and descrip¬tion” (p. 5). If understood that way, then McShane’s book certainly presents a theory of ellipsis.

The following “snapshot” of the “top-level classification of elliptical phenomena” and a “repre¬sentative sample of subtypes” discussed in the book will give you a feel of its comprehensive nature (pp. 6–7):

Syntactic ellipsis with coreference (i.e., with an accessible syntactic antecedent)

Syntactic ellipsis without coreference (i.e., with no syntactically accessible antecedent)

Semantic ellipsis (ellipsis of meaningful elements, but with no syntactic gap)

Unexpressed morphemes

Language strategies

McShane points out (p. 7) that the many references to object ellipsis in the above “snapshot” are likely to surprise native speakers of English, since standard English does not employ object ellip¬sis. In many languages, though, object ellipsis is very frequent and study of the phenomenon “offers particularly rich insights into the difficult problems” associated with ellipsis (ibid.).

On p. 137, McShane cites Hindi in connection with “so-called backward Gapping … in which the antecedent clause follows the Gapped clause”. It would have been helpful to point out that back¬ward Gapping of verbs correlates with OV constituent order (see Whaley 1997:274 for discussion of this point).1

Students of Russian will benefit most from the detailed discussion of the topics listed above, as each one is illustrated with examples from natural text in that language.2 Comparisons are also drawn with ellipsis strategies in Polish and Czech (the nine comparisons on pp. 70-75 are parti¬cularly interesting). Analysts of other languages will benefit most from the tables of parameters and values that need to be taken into account to describe the different elliptical phenomena (see also the algorithms on pp. 77–79). The following table (p. 202) will give you a feel of what McShane offers.

TABLE 13.1 A short inventory of parameters and values to describe subject ellipsis
Parameters Values
Is subject ellipsis employed? Yes
No
What can the referents be? Speaker
Hearer
Third person (person or thing
Does verbal morphology provide clues
for recovering the elided subject? 		Yes
No
(If yes above … ) Which subject features
are reflected by verbal morphology?
(possibly different for different tense/mood,
etc., combinations)		 Gender
Person
Number
Etc.
What can an elided subject refer to? Concrete person or thing
Generalized person or thing
What can the status of the referent be? Syntactically accessible
Understood extralinguistically
Understood by world knowledge
How “necessary” is subject ellipsis when
the subject can be recovered by morphological
or contextual means 		Virtually obligatory
Very common
Optional

Although McShane both acknowledges that working on ellipsis requires reference to discourse and has a short section on “discourse-oriented languages” (pp. 17–19), I was disappointed by the lack of consideration of discourse factors in ellipsis. For example, her section on the ellipsis of conjunctions (pp. 178–181) does not mention the role of conjunctions in packaging information or in constraining particular interpretations (see Levinsohn 2006). When translating from one language to another, I fear that such a lack would result in decisions about elision being taken only on impressionistic grounds.

This is reflected in a discussion of syntactic ellipsis in which she asks not only Can the category be inserted?, but also Would the category be inserted? (p. 23). It is true that “ellipsis judgments are not a yes-no option; they represent a scale of acceptability” (p. 24). However, more than acceptability has to be taken into account. When either option is acceptable, the question must also be posed: what is the discourse implication of choosing to elide or not to elide?

When not talking about ellipsis per se, McShane is sometimes careless in her use of linguistic terminology. While discussing actives and passives, for instance, she writes that the active “emphasizes” the agency of the actor, whereas the passive “emphasizes” the “patiency” of the patient. “Thus”, she continues, “the contrast between active and passive can be a contrast in focus or point of view” (p. 27). Such inexact use of linguistic terms is particularly annoying because, on three occasions, she cites Lambrecht’s work on Information Structure.

The terminology employed in the section on “The Assertion and Elaboration Strategy” is unhelp¬ful, too. McShane is discussing “clause complexes (i.e., juxtaposed clauses), in which the first clause asserts something and the second clause explains, embellishes, or otherwise comments upon it” (pp. 55–56). The Elaboration in the Russian example that follows appears to be a relative clause, but this is never stated.

Typos on pp. 50–52 have the potential to confuse the reader. In table 3.5 on p. 50, the Key for the third and fourth entries should be ‘c.’ and ‘d.’, not ‘a.’ and ‘b.’. Then, on p. 51, the final ØACC in example (14) should read itACC. Finally, example (18) on p. 52 is of Type VI, not Type VII.

The book is generally easy to understand, but there are some occasional lapses. For example, how many times did you have to read the following sentence to make sense of the apostrophe: “The structures’ being joined by a colon or dash is a strong indicator that the relationship between them is Assertion and Elaboration…” (p. 77)?

One editorial feature that I found unhelpful was the policy of placing tables at the end of a page, instead of immediately after a paragraph that ended, “as shown in table X”. This unnecessary separation of the text and table is particularly annoying on pp. 95, 107, 133, 150, 206 and 231.

This book was an eye-opener to me as to the range of factors to be taken into account when ana¬lysing ellipsis in a language. I therefore conclude that it should be essential reading for anyone planning to include a section on ellipsis in a grammatical description. And I would certainly want a personal copy if I was learning Russian and was concerned about when to use ellipsis correctly!

Notes

1

Although the index has an entry for Hindi on p. 254, it does not include a reference to p. 137.

2

Russian examples are presented in both Cyrillic and transliteration, followed by a word-by-word and free transla¬tion. It is unfortunate that, when an example comes at the bottom of a page, the first two lines often appear on one page, with the other lines on the next. See, for example, pp. 37-39.

References

Levinsohn, Stephen H. 2006 The Relevance of Greek Discourse Studies to Exegesis. Journal of Translation 2.2.

Nirenburg, Sergei, and Victor Raskin. 2004. Ontological Semantics. MIT Press.

Whaley, Lindsay J. 1997. Introduction to Typology: The Unity and Diversity of Language. London: SAGE Publications.