Search Results

This book has presented a cross-linguistic overview of the constituents that together make up the integral noun phrase (NP). Most NP constituents were discussed in some detail, including articles, demonstratives, numerals, adjectives, nouns, adnominal possessor constructions, and relative clauses. The discussion is mainly restricted to lexical NPs which are headed by an underived noun and which are used to refer to a single, discrete, first order (that is, spatial) entity. The immediate referent of an NP is not an entity in the real world, but rather a mental construct, whose properties need not coincide with the properties of its ontological correlate in the external world (if it exists). The various chapters touched on topics such as Seinsarten, nominal classification, qualifying modifiers, quantifying modifiers, localising modifiers, the underlying structure of noun phrases, and NP internal syntax.

It is tempting to suppose that all languages represent objects in comparable ways. Typological research has shown, however, that while there are many cross-linguistic similarities in this semantic realm, there are also numerous differences. This chapter describes some of these findings and explores their implications for cognitive neuroscience. The first section discusses plant, animal, and artifact concepts jointly, but in a manner that still respects their different treatments by typologists and neuroscientists. Then the subsequent section focuses on a fourth domain, namely body parts. Next, the chapter considers some of the ways in which objects are represented by the following kinds of closed-class items and constructions: grammatical-semantic splits involving possession, and nominal classification systems. Although both of these forms of object representation have been intensively investigated in typology, they have been almost completely neglected in neuroscience; hence, they are especially relevant to the latter field of study.

The entrenched nature of the gender/classifier dichotomy stands in the way of better typologies of nominal classification. How can we move beyond it to a more integrated view of nominal classification? Looking at a range of kinds of data from the Australian language Mawng, it is clear that our understanding of many less well-known nominal classification systems reflects a lack of data on how the system is used. Mawng has what seems like a well-behaved system of five genders, including gender agreement in the verb. However, the genders, like classifiers, play a crucial role in constructing meaning in discourse, often in the absence of nouns. Nominal classification systems must be contextualized in terms of their roles in constructing meaning in discourse, in order to do them justice in typologies. Greater emphasis on the flexibility of nominal classification systems and less on the role of nouns will also move efforts forward.

This chapter establishes that not all languages have a grammaticized mass/count distinction and consequently we have to distinguish between ontological properties associated with nouns and categorical properties associated with a functional category dominating these nouns. It is argued that the categorical properties associated with the mass/count distinction are tied to a functional category identified as nominal inner aspect. This category can host the feature responsible for the mass/count distinction (i.e, [± bounded]). It is further shown that languages lacking a categorical mass/count distinction come in at least two varieties. They can lack the functional category which may host the [±bounded] feature (Halkomelem). Alternatively, they can associated a different feature with inner aspect. In particular, it is shown that in Blackfoot [±animate] associates with inner aspect. Consequently, in this language, it is animacy, rather than mass/count which serves as the nominal classification device.