The (neuro)-psychology of mass and count nouns

Abstract

A basic distinction in our knowledge, already present in pre-linguistic infants, is that between compact, enduring things (objects) and the stuff (substance) of which they are constituted. The former are counted, the latter are measured. Most languages reflect this distinction in the categories of count and mass nouns. Count nouns, like cat, chair or fork, apply (Macnamara & Reyes, 1994) to perceptual entities that in combination do not yield another entity of the same kind. Mass nouns, like water, salt or oil denote, instead, monadic entities whose boundaries are perceptually inaccessible: the samples to which they are applied are taken as constituting in combination another sample. The structure of substances, designated by mass nouns, is arbitrary, while the structure of objects, designated by count nouns, is not arbitrary. In languages like English, French or Italian, syntactic properties also distinguish the two lexical categories of mass and count nouns (see Gillon, 1992). Cardinal numerals and quasi-cardinal numerals (e.g., several) modify count nouns, never mass nouns. Moreover, quantifiers like ‘‘little’’ or ‘‘much’’ modify mass nouns, never count nouns, whereas ‘‘few’’ and ‘‘many’’ modify count nouns, never mass nouns. Count nouns admit a morphological contrast between singular and plural; mass nouns do not, being almost always singular. The pronoun ‘‘one’’ may have as its antecedent a count noun, not a mass noun. Mass nouns with singular morphology do not tolerate the indefinite article, whereas singular count nouns do. Finally, mass nouns occur only with plural form of those quantifiers whose singular and plural forms differ. A categorical organization of noun processing has been shown to be supported in the brain in many instances (Semenza, 1999; Pulvermueller, 1999): the main question addressed here is whether count and mass nouns are differentially processed in the brain. Neuropsychological investigations conducted so far are recent and relatively few. Significant findings concern morpho-syntactic, conceptual, semantic, and lexical aspects, while anatomo-physiological studies are even more limited. The most convincing findings distinguishing mass and count nouns in neuropsychological literature concern morpho-syntactic aspects. Indeed, Grossman, Mickanin, Onishi, and Hughes (1995) had shown that early dementing patients are particularly sensitive to subtle syntactic distinctions such as those mentioned for mass and count nouns. In the first single case extensive report addressing this issue, a patient, whose grammar was otherwise perfect, was described, who, as a consequence of focal brain damage, showed an isolated deficit in the use, across a series of tasks, of the grammatical properties of mass nouns (Semenza, Mondini, & Cappelletti, 1997). Mondini, Jarema, and Liguori (2004) have recently reported the reverse pattern: their patient exhibited a general syntactic deficit while his performance was flawless in mass/count syntactic tasks. An interesting finding has been reported by Vigliocco, Vison, Martin, and Garret (1999): anomic patients may be able to apply proper mass/count lexical-syntactic rules to words they cannot retrieve. At a general, conceptual level, mass kinds were found (Borgo & Shallice, 2003) to associate with living entities rather than with artifacts within a herpes encephalitis patient’s memory disorder. A dissociation between mass and count nouns at the semantic and lexical level has been difficult to find. Indeed, in lexical retrieval past studies suggest that mass and count nouns may be supported by largely overlapping regions. Repeated investigations, in fact, involving a considerable number of aphasic patients, found only one subject, who had a huge left hemisphere lesion, featuring an over time stable, reliable dissociation (count worse than mass) in naming the two categories (Semenza, Mondini, & Marinelli, 2000).