Dies ist das Newsblog des Sprachwissenschaftlichen Instituts an der Ruhr-Universität Bochum.





Ruhr-Universität Bochum
Sprachwissenschaftliches Institut



Powered by PivotX - 2.3.11 
XML-Feed (RSS 1.0) 
XML: Atom Feed 

« Vortrag von Hana Fili… | home | Vortrag von Guillaume… »

Vortrag von Paolo Acquaviva am Dienstag, 23.01.2018 (16:00 Uhr - GB 3/159)

Montag, 08. Januar 2018. Aus der Kategorie 'Vortragsreihe'.

Das Sprachwissenschaftliche Institut lädt herzlich ein zum Vortrag von

Paolo Acquaviva (University College Dublin)

Internalist semantics and the grammatical construction of individuals.


Short abstract:

Our use of language presupposes a domain of entities, but this domain is at least in part a result of a conceptualization encoded in language. How to analyze linguistic conceptualization without falling into a simplistic Sapir-Whorf relativism? I address this challenge by distinguishing a basic domain of abstract entities, each named by a noun, from the domain of discourse referents, denoted by DPs. In between, grammar provides a template organizing part-structural information in different ways across languages. This explains a cluster of phenomena relative to kind-interpretation, number, and countability, unifies the analysis of nouns with that of names, and makes possible a predictive theory of possible nouns in natural language. In this way, lexical semantics can be integrated with a “grounded” approach to cognition, as the form for representing the substance provided by the mental recreation of experience.

Full abstract:

Most people would agree that language expresses a conceptualization of reality. But going beyond this vague statement is made difficult by a tendency to frame it in terms of unsatisfactory alternatives: that the entities and their properties are just “out there”, or that they are “created” by language. My goal is to outline an approach to the role of language in the conceptualization of entities that rests on a falsifiable empirical basis and does not fall into a simplistic and unpredictive Sapir-Whorf relativism.

At the most basic level an entity is an abstract object without internal structure, semantically modelled as a kind. Water, river, event, time, but also Socrates or Hamlet all name, at this level, mind-internal abstract objects, whether or not we may associate properties to them. All nouns are names in this basic ontology, and different labels identify different abstract objects. On the other hand, the discourse referents that speakers talk about are denoted by complete DPs, and do not necessarily require a noun to be spoken about (look at that). We make identifying reference acts to such referents by means of DPs, and we identify entities and their types by means of nominalized roots (or ‘nouns’): the two are sharply distinct. In between, grammar provides a structural template which encodes information about part structure, organized in various ways across languages. This explains in a novel and unified fashion a cluster of phenomena relative to the interplay of morphology and lexical semantics: the correspondence between a particular interpretation and a particular morphology in furniture­-type mass nouns, and restrictions on the kind reading for plurals including mass plurals like waters and beginnings.

Another important consequence of this approach is that it unifies the analysis of nouns and names: at the basic level, all entity labels are names, and the difference arises with the syntactic construction of a DP. This explains the parallel structure of this is called John and this is called water, but also a range of non-canonical uses of names, including the mass reading of too much Falstaff. Finally, the hypothesized interpretive template makes it possible to contemplate a predictive theory of possible nouns in any natural language: any entity can be thought, but language constrains what can be encapsulated as a noun; for instance not world-particular objects, or properties like roundsquare.

Finally, any theory of linguistic conceptualization should be compatible with what is known about the psychological representation of concepts. I will argue that viewing linguistic conceptualization as an interpretive template makes it easier to integrate lexical semantics with a “grounded” approach to cognition as the mental recreation of experience. In this way, linguistic theorizing on the content of nouns does not set up psychologically unsupported models of lexical concepts, but identifies the framework for representing through words what we know about entities.