Cognitive Linguistics Croft & Cruse 10

Download Report

Transcript Cognitive Linguistics Croft & Cruse 10

Cognitive Linguistics
Croft & Cruse 10
An overview of construction
grammars
(part 2, 10.2.2 through end)
10.2.2 Lakoff (1987) & Goldberg
(1995)
• Q&A:
– 1) What is the status of the categories of the
syntactic elements in construction grammar?
• Participant roles are non-reductionist – the
complex event is the primitive, but syntactic roles
are reductionist, assuming primitive roles such as
subject, object
– 2) What sorts of syntactic relations are
posited?
• NA and see above
10.2.2 Lakoff (1987) & Goldberg
(1995)
• Q&A:
– 3) What sorts of relations are found between
constructions?
• Radial categories, with central, prototypical members and
other members related via metaphorical extension
– 4) How is information stored in the construction
taxonomy?
• It is stored redundantly at various levels, in conformity with
the Usage-Based model and evidence that information is
stored redundantly in the mind
10.2.3 Cognitive Grammar
(Langacker)
• Cognitive Grammar’s model of syntactic
representation is a construction grammar
model, emphasizing symbolic and
semantic definitions of what is traditionally
analyzed as syntax
• A symbolic unit links a semantic pole to a
phonological pole
10.2.3 Cognitive Grammar
(Langacker), cont’d.
• Q&A:
– 1) What is the status of the categories of the
syntactic elements in construction grammar?
• Syntactic categories such as noun, verb, subject,
object are abstract semantic construals of the
content of their denotations. So they are semantic
and based on construal of experience.
– 2) What sorts of syntactic relations are
posited?
• Valence is handled as elaboration sites for
substructures, and they represent a gradient
10.2.3 Cognitive Grammar
(Langacker), cont’d.
• Q&A:
– 3) What sorts of relations are found between
constructions?
• Again, we have a radial category with prototype
and (metaphorical) extensions
– 4) How is information stored in the
construction taxonomy?
• Like Lakoff & Goldberg, Langacker’s is a UsageBased model
10.2.4 Radical Construction
Grammar (Croft)
• Adopts the radial category structure and
Usage-Based model we see with Lakoff,
Goldberg, and Langacker
• Thoroughly non-reductionist
• Rejects autonomous syntactic relations
• Brings in models of conceptual space and
semantic map as organizing principles
10.2.4 Radical Construction
Grammar (Croft), cont’d.
• Q&A:
– 1) What is the status of the categories of the
syntactic elements in construction grammar?
• Constructions are the basic units, and syntactic
categories are defined in relation to constructions.
There are no atomic schematic units defined
independently of constructions. Constructions are
organized into radial categories with prototypes
and extensions
10.2.4 Radical Construction
Grammar (Croft), cont’d.
• Q&A:
– 2) What sorts of syntactic relations are
posited?
• Relations between parts of a construction are
defined in purely semantic terms – there are no
syntactic relations in Radical Construction
Grammar. When you look cross-linguistically, you
find that syntactic relations are not uniform, and
their semantic content is not the same.
10.2.4 Radical Construction
Grammar (Croft), cont’d.
• Q&A:
– 3) What sorts of relations are found between
constructions?
• Constructions define all the relations. So the
Transitive construction defines both the Transitive
Verb and the Transitive Object. And the
Morphological Verb construction defines the Verb
as the stem and inflections, uniting Transitive and
Intransitive Verbs. It’s constructions all the way up.
10.2.4 Radical Construction
Grammar (Croft), cont’d.
• Q&A:
– 4) How is information stored in the
construction taxonomy?
• Information is represented redundantly in
conformity with the Usage-Based model
– Conceptual space – functional properties, based on
(universal?) experience
– Semantic map – how items map onto the conceptual
space
• Constructions are language-specific