The count/mass distinction in: The Noun Phrase

Jan Rijkhoff linjr at HUM.AU.DK
Fri Aug 23 09:18:05 UTC 2002

(an earlier version of this message was distributed in a distorted form)

Various typological aspects of the count-mass distinction are discussed in
my book 'The Noun Phrase' (Oxford Studies in Typology and Linguistic
Theory, OUP 2002).

For example, Chapter 2 contains a cross-linguistic investigation of nouns
that are employed to refer to a singular discrete spatial entity (or
individual) such as 'dog' or 'knife' (or, rather, to refer to an entity
whose ontological counterpart or SEIN-correlate is a singular discrete
spatial object; see note ** below on Referents). I argue on the basis of
morpho-syntactic evidence that four noun types (or nominal subcategories)
are normally used for this purpose: SINGULAR OBJECT NOUNS ('count nouns'),
SET NOUNS, SORT NOUNS, and GENERAL NOUNS (the last three noun types are all
transnumeral; SORT NOUNS and GENERAL NOUNS require a classifier when
modified by a numeral).

Ultimately this leads to a more comprehensive semantic classification of
six major nominal subcategories (also including MASS NOUNS and COLLECTIVE
NOUNS) which are defined in terms of two features: Shape and Homogeneity.
Section 2.5 is devoted to the semantics of incorporated and predicate nouns
and in the last section I contend that essentially each nominal subcategory
defines a different SEINSART of a spatial property, i.e. an alternative way
in which a nominal property is specified for the features Shape and
Homogeneity (cf. AKTIONSART in verb semantics, which concerns the way a
verbal property or relation is represented in the temporal dimension).
   Since this is a linguistic and not an ontological classification, there
is no direct relationship between noun type and (real world) entity type.
This is, of course, precisely the reason why it is possible for different
noun types to be used across the world's languages to refer to entities
whose SEIN-correlate in the external world is a single, discrete spatial
object: singular object nouns, set nouns, sort nouns and general nouns. In
other words, speakers of a different language do not necessarily use a
member of the same nominal subcategory when they talk about, for instance,
a knife.

Now, to return to one of Masja's queries: what I have called GENERAL NOUNS
are used to refer to masses and discrete objects (see also e.g. John Lucy
(1992: 74) on Yucatec Maya in his book 'Grammatical categories and
cognition: a case study of the linguistic relativity hypothesis', CUP).

As to number marking, which is generally considered to be a characteristic
of count nouns, I would like to point out that I distinguish between number
marking (used obligatorily on SINGULAR OBJECT NOUNS and COLLECTIVE NOUNS)
and the new grammatical category of NOMINAL ASPECT MARKING (often optional
and restricted to the transnumeral set nouns, which are attested as a minor
or major noun type in most of the world's languages). Nominal aspect
marking differs from number marking in that it specifies a qualitative
(rather than a quantitative) distinction: it signals whether the speaker
refers a singleton set or to a collective set.

The Noun Phrase appeared a couple of weeks ago and is based on data from a
representative sample of 50 languages (plus facts from many other
languages). Chapters 2-6 aim to give a cross-linguistic overview of the
semantic and morpho-syntactic properties of the constituents of the NP. In
chapter 7 I propose a new, typologically adequate, semantic model of the
underlying structure of the NP (suggesting that NPs and clauses can be
analysed in a similar fashion). In Chapters 8-10 I demonstrate that
NP-internal ordering patterns can be accounted for in terms of a few
general ordering principles. The final chapter is an epilogue.

** (see above): The immediate referent of an NP is not object in the real
world, but rather a mental construct that is created, stored, and retrieved
in the minds of the speech participants. This allows for possible
discrepancies between (linguistic) properties of discourse referents and
(ontological) properties of their real-world counterparts or
SEIN-correlates (if they exist).

Jan Rijkhoff

>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.
>	In this connection I have two queries:
>1. What are the linguistic correlates of the count-mass distinction
>in various languages?
>2. Do all languages manifest the count-mass distinction?

Jan Rijkhoff, Institut for Lingvistik, Aarhus Universitet
Jens Chr. Skous Vej 7 (467-517), DK-8000  Aarhus C, DENMARK
Phone (+45) 8942 6550 * Fax (+45) 8942 6570 * E-mail linjr at
-------------- next part --------------
A non-text attachment was scrubbed...
Name: not available
Type: text/enriched
Size: 4820 bytes
Desc: not available
URL: <>

More information about the Lingtyp mailing list