[This reposting of Thomas Payne's review of Corbett (2000)
includes a correction to the animacy hierarchy chart (the
error was made by the editor, not the reviewer), and a
biographical statement by the reviewer. --Ed.]
Corbett, Greville G. (2000) Number. Cambridge Textbooks in
Linguistics. Cambridge University Press. Hardback edition,
xx + 358 pages, including author, language and subject
indices.
Review by Thomas E. Payne, University of Oregon and SIL
International.
Number is an area of morphosyntactic typology that has
often been taken for granted by linguistic researchers.
In this comprehensive typological survey, Corbett
convincingly shows that the complexity and diversity of
number systems in the languages of the world far surpasses
what is commonly assumed.
This book would probably serve as the main text only for
graduate seminars in number systems. However, it will
certainly be a useful resource for undergraduate and
graduate students who are doing major research projects or
term papers. I suspect that the book will be most useful to
individuals engaged in grammatical description of particular
languages or families of languages. It provides a thorough
account of how the meaning categories in number systems
are organized (chapters 1-4), how they are expressed
grammatically (chapters 5-6), their extended uses (chapters
7-8), and how they rise and evolve diachronically. Though
this is not a statistical study, data from some 250
languages, selected to illustrate the full range of variety
of number systems, are represented. Many of these languages
have not previously been cited in the typological literature.
The book begins by calling attention to a series of common
myths concerning number. These are:
1. Number is just an opposition of singular vs. plural.
2. All relevant items (nouns, for instance) will mark number.
3. Items which do mark number will behave the same.
4. Number must be expressed.
5. Number is a nominal category.
The remainder of the book consists largely of detailed
evidence dispelling these myths. For example, concerning
number categories, Corbett clearly describes and exemplifies
unfamiliar categories such as "general number," "paucal,"
"greater paucal," "greater plural" (the opposite of paucal),
and "determinate plural." In addition to singular, plural,
dual and trial categories, Corbett also addresses the
question of whether a category of "quadral" exists in any
language. The conclusion is that, if such categories exist,
they are very rare. Corbett argues that those that have been
proposed in the literature (all from the Austronesian
family), are best treated as paucals. The most complex
systems express up to five number categories For example,
Lahir (Oceanic) has singular, dual, trial, paucal and plural
categories. Susurunga, another Oceanic language, has
singular, dual, paucal, "greater paucal," and plural
categories.
Concerning which items will take number marking, the
evidence supports the common observation that number is
more likely to be expressed on nouns that are higher on an
"animacy" hierarchy, than on those that are lower on the
hierarchy. The hierarchy (better termed "empathy" than
"animacy") that is relevant for number is the following (p.
57):
1st person > 2nd per > 3rd per > kin > human > animate > inanimate
| ---------------Pronouns-----||
The universal that is supported by the data is that if a
particular number distinction (e.g., dual vs. trial,
singular vs. plural) occurs at any point on this hierarchy,
it will occur at all points to the left. In some languages
this reaches the level of obligatoriness, whereas for other
languages number marking remains optional, but is more
common at the higher positions.
One entire chapter, and parts of several others, is (are?)
dedicated to agreement phenomena relating to number. (The
previous sentence illustrates one typological parameter of
number agreement that varies from language to language --
namely whether unequal conjuncts in a conjoined noun
phrase trigger singular or plural agreement). A
distinction is made between semantic agreement and
grammatical agreement. For example, a noun like _committee_
is grammatically singular (a standard plural form also
exists), but semantically can be construed as singular or
plural. Agreement in both American and British English can
depend on the grammatical form ("The committee is . . .),
or the semantics ("The committee are . . ."), though
British speakers are more likely to choose the semantic
basis for agreement. When it comes to agreement in
demonstratives, however, both British and American English
are sensitive to the form ("This committee . . .," rather
than "*these committee . . ."). A universal is proposed
whereby agreement is more semantically based the farther to
the right the agreeing element is on the following "Agreement
Hierarchy":
Attributive (adj., demons., ...) < predicate < rel. pron. < pers. pron.
Many fascinating examples from around the world are cited
in support of this hierarchy.
Another chapter (chapter 7) is devoted to extended uses of
number marking. A common extended use of plural
morphology is to express honorific status. Others include
associative, anti-associative, approximative, evasive,
exaggeration and intensification. The associative use is
when a plural marker on a proper name refers to the person
designated by the proper name plus his or her companions or
family. An "anti-associative" is a quite distinct plural
marker that occurs on a common noun to form an expression
that refers to the item plus its users. The Eskimo
languages all have such a special use of the plural for
vehicles plus their passengers/riders, e.g., siikilit
(plural) 'bicycle plus rider.' The evasive, exaggeration
and intensification uses can all be illustrated marginally
in English. Such expressions as "She's got all kinds of
houses and cars and airplanes," to mean simply "she is
well-off" illustrates exaggeration/intensification. The
evasive is illustrated in the tendency to use a plural
pronoun to refer to a singular referent in order to avoid
designating a particular sex, as in "Every student should
see their advisor." Apparently this kind of usage occurs in
other languages as well. For example, in Alamblak (a Papuan
language) the plural can be used to refer to a singular
unborn child in order to avoid commitment to the as-yet
unknown sex.
Chapter 8 deals with "verbal number." This is a type of
verbal aspect in which a verb is marked as referring to
more than one event. The reasons given by Corbett for
treating this under the heading of number are, 1) the
conceptual parallel between nominal number referring to
number of things, and verbal number referring to number
of events, 2) verbal number often exhibits some of the same
restrictions as nominal number (e.g., verbal number is more
likely to be expressed if the actor is high on the animacy
hierarchy), and 3) there is a tradition in some language
families (e.g., Chadic languages) of referring to these
aspects as "plural verbs." Considering this rather marginal
relation to prototypical number systems, it seems to me a
bit unbalanced to devote an entire chapter, out of nine, to
verbal number. A good portion of this chapter is devoted to
distinguishing verbal number from agreement with nominal
arguments, for example Huichol (Uto-Aztecan) has singular
and plural verb stems depending on the number of the
"participant most directly affected" by the situation
described by the verb. The stem alternation is quite
distinct from the system of verb agreement, which is
expressed by prefixes and is strictly based on a
nominative/accusative system. For this reason, Corbett
considers the Huichol system to be an instance of verbal
number. Several similar examples are given from various
parts of the world.
Finally, chapter 9 deals with various topics not touched on
in the other eight, including the diachronic rise and fall
of number systems. In comparison to the thoroughness with
which the other topics in this book are addressed, the
section on diachrony is a bit disappointing. Few concrete
examples are given, at least in comparison to the richness
of exemplification in the rest of the book. Other topics
treated in chapter nine are the interaction between number
and other nominal categories (e.g., gender, obviation,
definiteness, case and person), the acquisition of number,
and the psycholinguistics of number, and how number marking
is deployed in various languages.
While this reviewer may have preferred a more in-depth
treatment and exemplification of diachrony, and
comparatively reduced treatment of "verbal number," there
is no question that this book will be considered a "classic"
of typological literature. It is a thorough and engaging
treatment of a quite under-appreciated yet pervasive
functional category. The book should serve as a model for
future typological studies of other areas of linguistic
structure.
Thomas Payne is an international linguistics consultant with
SIL International and a research associate at the University
of Oregon. He specializes in linguistic typology and
grammatical description of endangered and underdescribed
languages.
|
