LINGUIST List 4.306

Sun 25 Apr 1993

Disc: Markedness

Editor for this issue: <>

Directory

1. , Re: 4.295 Marking
2. Robert Beard, Marking--Russian Number

Message 1: Re: 4.295 Marking

A correction on Old French number: those masculine nouns that had an
-s suffix in the nominative singular also had -s in the accusative
plural, though not in the nominative plural. In other words, in Old
French there was no *mere plural* morpheme for masculine nouns at all, just
as there was none in Latin, is none in Russian, etc. etc. Neither was
there a *mere singular* affix of the sort being sought.

--Leo Connolly

 In his response to Croft, Manaster-Ramer confuses 'grammatical cate-
gory' with 'morpholgical expression' (of a grammatical category). There
is now a significant body of literature on the problem written by
supporters of Word-and-Paradigm and Lexeme-Morpheme Base Morphology. The
correlation between category and expression (e.g. affixation) may be
many-one, one-many, one/many-zero, zero-one/many. The fact that the
grammatical properties [-Singular, +Plural] (of the grammatical category
'Number') are expressed by several morphological markings comes as no
surprise to morphologists.
 Manaster-Ramer asks: "If we found a language in which a special form
was used only with the numeral for '2', would that mean that this language
has no dual?" That depends. If there are dual agreement categories
reflected in the adjective and verbs, yes, there is a dual. If such
agreement is not present, no, there is no basis to claim a dual. In other
words, affixation is no proof of a category or its properties. For this
reason the phrasal evidence for dual suggested by Manaster-Ramer, e.g.
dva/tri/Cetyre krasnyx/*krasnogo karandaSa, proves just the opposite of
what he argues. The use of plural agreement in the adjective proves
conclusively that the noun is plural, i.e. [-Singular, +Plural], and not
dual. The agreement with verbs also reveals no special agreement pattern
for dual (or paucal). Verbs agreeing with quantified nouns in Russian
usually fall in the neuter singular or plural. If there were a dual or
paucal in Russian, one would expect one agreement pattern for this pro-
perty and the other for plural. But the difference is one of style and
both may be used for any quantification beyond 'one'.
 Russian also has a set of indelinable nouns, nouns with no case
markers at all. All these nouns express case, number, and gender,
however, in agreement, e.g. dva krasnyx kenguru 'two (male) kangaroos'
dve krasnyx kenguru 'two (female) kangaroos'. Agreement is always
the normal plural (or neuter singular). Here the categories are
clearly present but there is no affixation on the noun.
 Croft's assessment of the Russian data was therefore correct.
There are ways to demonstrate grammatical categories and affixation
is one of them. However, it is wholly unreliable under the false
assumption of any sort of direct relation between morphological form
and function. --RBeard

Robert Beard, rbeardbucknell.edu