LINGUIST List 4.295

Fri 23 Apr 1993

Disc: Markedness

Editor for this issue: <>

Directory

1. , Subject: Number markedness
2. Paul T Kershaw, Kiowa number

I am not wholly convinced by some of what was said in recent
postings on this subject:

 Date: Tue, 20 Apr 93 3:10:50 EDT
 From: Paul T Kershaw <kershawpstudent.msu.edu>

and
 Date: Wed, 21 Apr 93 22:31:22 PDT
 From: Bill Croft <wcroftCSLI.Stanford.EDU>

In Bill Croft's, I am surprised by the following statement in
response to Paul Kershaw:

<< The Russian data does not have to do directly with the grammatical
<< category of number. They have to do with the form of nouns after
<< numerals, in which (in standard Russian) a noun takes the nominative
<< singular after 1, the genitive singular after 2-4, and the genitive plural
<< after 5-10. This pattern is repeated after numerals that end in 1, e.g. 21,
<< 31, 101, etc. and likewise for numerals ending in 2-9. However, it is
<< not found in other syntactic environments. While this pattern is
<< interesting in its own right (see for example the Greenberg articles on
<< number in his recent anthology "On Language", ed. K. Denning and S.
<< Kemmer, Stanford), it isn't a part of grammatical number. (In fact, in
<< many languages, the noun form after the numeral has no number
<< marking at all.) Russian has suffixes indicating both number and case
<< which are obligatory in all syntactic contexts. There is in fact an
<< anomalous pattern there, namely that for certain declension classes the
<< genitive plural is zero-marked (see below).

I do not see why these facts have nothing to do with number.
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?

Also, the facts are more complicated. First, a few nouns
 have a different form when used with numerals ending
in 2-4 than they do in genitive singular, notably, chas 'hour'.

Second, we must consider the syntax of the whole phrase, not
just the form of the noun, I think, and it is significant that
adjectives cooccuring with numerals ending in 2-4 do NOT take
singular-looking forms.
For ex., krasnogo karandasha is the gen. sg. of 'red pencil', while
'2 red pencils' is 'dva krasnyx/*krasnogo karandasha'.

Thus, it is appropriate to say, I think, that Russian has
special number categories that appear in the presence of
these classes of numerals.

And further:

<< So much for the motivated exceptions. Now for the real exceptions.
<< The Russian zero-marked genitive plural is one; another is Old French
<< and Old Provencal, in which the nominative singular ends in -s and the
<< plural in zero. However, this anomalous situation corrected itself by
<< the modern versions of the latter languages, and is in the process of
<< doing so for some Slavic languages other than Russian (see Greenberg,
<< "Some methods of dynamic comparson", in the aforementioned
<<
<< Bill Croft

The Russian zero-marked genitive plural is a property
of some classes of nouns only. Other classes take the
suffixes -ov or -ej. And I think that something like
this must have been true of Old French and Old Provencal
nominative singulars (surely the feminines in -e did not
have a nominative in -s, did they?).

It would be interesting to ask whether any language that
has counterexamples like these to the usual assumptions
about number marking is any more consistent than these
languages. That is, is there any language where EVERY
plural form is zero and EVERY singular is marked (or
where every plural is longer than the corresponding
singular)? If not and if there are languages where
plurals are always longer than singulars (e.g., Turkish),
then I think we do have a fairly solid universal here,
or do we?

And in Paul Kershaw's summary, I was not quite happy with the
statement about Sinhalese for a similar reason:

>> Sinhala (Paolillo) inanimate nouns have
>> no
>> overt affix in the singular, while the singular affix is -a and the
>> indefinite
>> affix (available only in sg.) is -k. This pattern does not carry through to
>> the animates, where all three noun types (plur, sg.def., sg.indef.) have
>> endings.

Again, while more or less accurate, this account leaves out
some (to my mind crucial) facts. First, a few inanimates have
overt endings in the plural and in the singular (the reverse of
the normal situation), e.g., kaTA (T = alveolar as opposed to
dental, A = schwa) 'mouth', raTA 'country', paarA 'road', gee 'house',
dee 'thing', all of which take the plural suffixVal. Second, when
we look beyond the direct case, we find that the plural is more
marked than the singular. Thus, consider the NORMAL inanimate
pattern

 Sing. Plural

Direct potA pot
Dative potATA potvAlATA
Genitive potee potvAlA
Instru. poten potvAlin

[The final n is actually a velar nasal.]

TheVAl- we find in all the oblique plural forms is
the same morpheme as theVal we found in the anomalous
plurals like kaTAval from kaTA, the a/A alternation being automatic.

Thus, again, on the whole, the Sinhalese plural is not really
zero-marked or less marked than the singular.

Having been confused by the two accounts of Kiowan grammar (one by Bill Croft,
which you all have seen), I sought out the horse's mouth, to wit, Laurel J.
Watkins 1984 A Grammar of Kiowa. Lincoln: U of Neb Press. The number
business is described on pp. 78-92 (sec. 3.12).

There are three numbers and four classes of nouns. The numbers are singular,
dual, and plural. The classes are: animate (I), inanimate tangible/count
(II), inanimate intangible/mass (IV), and other (III). The Roman numerals are
Watkins', the category labels mine, and there is of course a great deal of
mismatch between class membership and the real world -- "star", for example, is
class I, as are the loanword "car" and "knee", while "foot" is class II.
Watkins doesn't provide enough data, though to support or refute Croft's claim
that there is "no rhyme or reason" to membership.

In all cases, there is no overt marking on the noun for dual. The other
classes can be distinguished by the four possibilities of overt marking on
singular and plural, namely:
 SG PL
 I -- ga
 II ga --
 III ga ga
 IV -- --
Ambiguities are resolved in some cases bacause the verb carries an agreement
prefix that also indicates number.

I hope this description, for those of you interested, has been cleare than the
previous ones.

Paul Kershaw, Michigan State University