Count-mass distinctions

[Maria Koptjevskaja Tamm]

Dear colleagues,
I am interested in the mass-count distinctinction across languages,
primarily in how it can be manifested and in whether there are
languages which can do without it.
	According to traditional wisdom, the English word cow is a
typical count noun: it refers to discrete entities with a
well-defined shape and precise limits. A single cow is easily
distinguishable from several ones (apart from abnormal cases), and
this is linguistically reflected in the number opposition (the
singular -ø vs the plural -s) marked on the corresponding noun. As
discrete entities, cows can be counted ­ correspondingly, the noun
cow easily combines with numerals, e.g. one cow, three cows etc. and
with the denumerating measure word many.
	Water, on the other hand, is a typical mass (uncountable)
noun: it refers to homogeneous undifferentiated stuff without any
certain shape or precise limits. The mass term itself lacks a
built-in mode of dividing reference, with the effect that it lacks a
number opposition (i.e., has a defective paradigm as compared to
count nouns), cannot combine with numerals and with the indefinite
article a and takes the measure word much instead of the denumerator
many. Water, as most stuff, can be measured, but this presupposes a
procedure of dividing the stuff up in portions which are further
counted ­ e.g. three glasses of water, one bottle of water etc.
	The traditional wisdom has been reconsidered in various
works. Thus, it is known that many nouns can occur in both countable
and uncountable contexts ­ philosophic and linguistic literature
abounds not only in examples like "Hetty likes to gorge herself on
cake" and "Whenever Hetty gobbles down a cake, her diet ¹starts
tomorrow¹" (Allan 1980:546) or "I'd like to have one water and three
coffees", but also in such pearls as "There was dog all over the
street" and "I don¹t like shelf ­ I¹d rather eat table" (uttered by
one intelligent termite to another). However, there are clear lexical
preferences for which nouns tend to occur as count nouns and which as
mass nouns.
	In this connection I have two queries:
1. What are the linguistic correlates of the count-mass distinction
in various languages?

a. The most obvious ones are mentioned above: mass nouns lack number
oppositions (apart from special contexts), do not combine with
numerals, with the indefinite article and with certain denumerators,
such as "many".
b. In some Spanish dialects certain nouns have a mass number in
addition to the normal singular-plural opposition, e.g. Œhair¹ ­ pílu
(sg) vs. pélos (pl) vs. pélo (mass). Also adjectives have a separate
mass agreement form, which can be used even with those nouns that,
for purely morphological reasons, lack a special mass form, e.g. la
maéra tába sék-o Œthe wood (mass) was dry¹ vs. la maéra tába sék-a
Œthe (piece of) wood was dry¹ (ex. from Lena dialect quoted in
Corbett 1999).
c. The use of the partitive article in French and of the genitive
case for marking objects in Russian in certain contexts are strongly
favoured by mass nouns.

Any more?

2. Do all languages manifest the count-mass distinction?

Two types of languages have been quoted for which the count-mass
distinction seems to be irrelevant or at least of much less
importance than, say, for English, since these languages combine all
nouns with numerals in more or less the same fashion, without
distinguishing between "three cows" and "three glasses of water".
First, numeral-classifier languages like Vietnamese where all nouns
in numeral constructions must be accompanied by a classifier. Second,
in some languages, like Tagalog, all nouns, even the most "mass-like"
ones, seem to be able to combine with numerals and plural markers
directly - the most "mass-like" ones are then interpreted as
referring to portions of the mass.

The question is, thus, whether the count-mass distinction is
completely irrelevant for such languages or whether nouns can still
be categorized on the basis of their lexical preferences for being
used in count or mass contexts.

I would be most grateful for any comments, examples and advice!
--
Maria Koptjevskaja Tamm
Office: Dept. of linguistics, Stockholm university
106 91 Stockholm, Sweden
Tel.: +46-8-16 26 20
http://www.ling.su.se/staff/tamm

Home: Vaesterled 166
167 72, Bromma, Sweden
Tel.: +46-8-26 90 91