Top Banner
Cognitive-Functional Linguistics – Some Basic Tenets III Rolf Theil Bergen, June 19, 2006
40

Cognitive-Functional Linguistics

Jan 09, 2016

Download

Documents

Conor

Cognitive-Functional Linguistics. – Some Basic Tenets III Rolf Theil Bergen, June 19, 2006. The Emergent Grammar 1. Grammatical structure “grows upwards” from instances. There is no sharp boundary between lexicon and grammar. - PowerPoint PPT Presentation
Welcome message from author
This document is posted to help you gain knowledge. Please leave a comment to let me know what you think about it! Share it to your friends and learn new things together.
Transcript
Page 1: Cognitive-Functional Linguistics

Cognitive-Functional Linguistics

– Some Basic Tenets III

Rolf TheilBergen, June 19, 2006

Page 2: Cognitive-Functional Linguistics

June 19, 2006 RT/CFL 2

The Emergent Grammar 1

Grammatical structure “grows upwards” from instances.

There is no sharp boundary between lexicon and grammar.

They are different aspects of the same network of extension and instantiation relations.

Page 3: Cognitive-Functional Linguistics

June 19, 2006 RT/CFL 3

The Emergent Grammar 2

Grammar and lexicon constitute construc-tions with varying degrees of schematicity.“Grammatical constructions” are more schematic

than “lexical constructions”.Grammar cannot be neatly divided into

syntax and morphology.Syntax and morphology constitute construc-

tions of varying size and complexity.“Syntactic constructions” are bigger and more

complex than “morphological constructions”.

Page 4: Cognitive-Functional Linguistics

June 19, 2006 RT/CFL 4

The Emergent Grammar 3

Common structural principles hold across phonology, semantics, pragmatics, morpho-logy, syntax, and other aspects of language.

The morphological network illustrated earlier is governed by exactly the same principles as those governing the syntactic aspects of the lexicon-morphology-syntax continuum.

Page 5: Cognitive-Functional Linguistics

June 19, 2006 RT/CFL 5

The Emergent Grammar 4

In our presentation of The Emergent Grammar, we left out some aspects that we shall take a look at now:

Entrenchment Composition Categorization

Page 6: Cognitive-Functional Linguistics

June 19, 2006 RT/CFL 6

Entrenchment 1

Usage affects grammatical representation in the mind.

Frequency of use correlates with entrenchment.

Constructions that are more frequently processed become more entrenched in the language system.

Entrenchment may be interpreted as resting activity.

Page 7: Cognitive-Functional Linguistics

June 19, 2006 RT/CFL 7

Entrenchment 2

There are two main types of frequency:Token frequencyType frequency

Token frequency gives rise to the entrenchment of instances.

Type frequency gives rise to the entrenchment of more abstract schemas.

Page 8: Cognitive-Functional Linguistics

June 19, 2006 RT/CFL 8

Entrenchment 3

Token frequency gives rise to the entrenchment of instances.

Example Each time the constructions (or words) [[dansa]

/ [DANCE, PAST]] and [[so:g] / [SEE, PAST]] are used, their mental representations are strengthened.

[[so:g] / [SEE, PAST]] is much more frequent than [[dansa] / [DANCE, PAST]].

A subsequent reduction in use weakens the entrenchment.

Page 9: Cognitive-Functional Linguistics

June 19, 2006 RT/CFL 9

Entrenchment 4

Assuming that [[so:g] / [SEE, PAST]] as well as [[dansa] / [DANCE, PAST]] have unit status, we can represent the differences in entrench-ment by letting the degree of entrenchment correlate with the thickness of the lines used to draw the “unit boxes”:

so:g / SEE, PAST dansa / DANCE, PAST

Page 10: Cognitive-Functional Linguistics

June 19, 2006 RT/CFL 10

Entrenchment 5

Type frequency gives rise to the entrenchment of more abstract schemas.

Example [[so:g] / [SEE, PAST]] and [[lo:g] / [LIE, PAST]] are the

only instances of the schema [[Co:g] / [VERB, PAST]]. [[dansa] / [DANCE, PAST]] and thousands of other

verbs instantiate the schema [[…a] / [VERB, PAST]]. The entrenchment of a schema is governed by its

number of instances. This is illustrated on the next slide.

Page 11: Cognitive-Functional Linguistics

June 19, 2006 RT/CFL 11

Entrenchment 6 A weakly entrenched schema with a few strongly

entrenched instances. A strongly entrenched schema with many weakly

entrenched instances.

so:g / SEE, PAST lo:g / LIE, PAST

Co:g / VERB, PAST

kasta / THROW, PAST

spe:la / PLAY, PAST

dansa / DANCE, PAST

hopa / JUMP, PAST joba / WORK, PAST

σ…a / VERB, PAST

Page 12: Cognitive-Functional Linguistics

June 19, 2006 RT/CFL 12

Entrenchment 7

Productivity is a matter of degree. Productivity is a matter of how available a

pattern is for the sanction of novel expressions. Sanction: the motivation afforded a novel structure by

the conventional units of the language.

Productivity amounts to likelihood of being selected as the active structure used to categorize a novel expression.

Strongly entrenched schemas are more easily activated (they have a higher ‘resting activity’) than weakly entrenched schemas.

Page 13: Cognitive-Functional Linguistics

June 19, 2006 RT/CFL 13

Composition 1

Composition: The relation between component structures and the composite structure that derives from them.

A composite structure is a structure that results when two or more component structures combine.

[[hopa] / [JUMP, PAST]] is a composite structure. The component structures are:

[[hopa…] / [JUMP, TNS]][[σ…a] / VERB, PAST]]

The relationship between the structures is shown on the next slide.

Page 14: Cognitive-Functional Linguistics

June 19, 2006 RT/CFL 14

Composition 2

hopa / JUMP, PAST

hopa…… / JUMP, TNSTNS σ…σ…a / VERBVERB, PAST

Component structure Component structure

Composite structure

Page 15: Cognitive-Functional Linguistics

June 19, 2006 RT/CFL 15

Composition 3

Composition between two structures is only possible if at least one of them is partly schematic, like the red, outlined red, outlined partsparts of the structures to the right.

σ…σ…a / VERBVERB, PAST

hopa…… / JUMP, TNSTNS

Page 16: Cognitive-Functional Linguistics

June 19, 2006 RT/CFL 16

Composition 4

A schematic part of a structure is called an elaboration site.To elaborate: to instantiate a schema.An elaboration site: those facets of one component structure that another component structure serves to elaborate.

σ…σ…a / VERB VERB , PAST

hopa … … / JUMP , TNSTNS

Page 17: Cognitive-Functional Linguistics

June 19, 2006 RT/CFL 17

Composition 5

The component structures of a composite structure (= a complex construction) are not like classical morphemes that are stacked together to form more complex edifices, where form and meaning are parts of the individual “building blocks”, e.g. as in:

{/dans/‘dance’}+{/a/‘past’}

Page 18: Cognitive-Functional Linguistics

June 19, 2006 RT/CFL 18

Composition 6

The component structures are more or less schematic structures that are integrated with each other.

The structure below, which is the CFL counterpart of the classical morpheme {/a/‘past’}, is a schematic word:

σ…σ…a / VERBVERB, PAST

Page 19: Cognitive-Functional Linguistics

June 19, 2006 RT/CFL 19

Composition 7

Classical morphemes are assumed to predict the form and meaning of a complex word.

CFL component structures motivate the form and meaning of a composite structure. Example on the next slide.

Page 20: Cognitive-Functional Linguistics

June 19, 2006 RT/CFL 20

Composition 8

The morphemic analysis of fireman is:{/faɪəɹ/‘fire’} + {/mæn/‘man’}

Fireman has two related meanings:‘person whose job is putting out fires’‘person who looks after the fire in a steam

engine or furnace’Neither meaning is predictable from the

morphemic representation.

Page 21: Cognitive-Functional Linguistics

June 19, 2006 RT/CFL 21

Composition 9

CFL analysis of fireman

faɪəɹmən / PERSON WHOSE JOB IS PUTTING OUT FIRES faɪəɹmən / PERSON WHO LOOKS AFTER THE FIRE IN A STEAM ENGINE

faɪəɹ / FIRE σ…mən / PERSON ASSOCIATED WITH ‘THING’ mæn / MAN

…mVn / PERSON

faɪəɹmən / PERSON ASSOCIATED WITH FIRE

Page 22: Cognitive-Functional Linguistics

June 19, 2006 RT/CFL 22

Composition 10

A grammar is not a “generative” description, providing a formal enumeration of all and only the well-formed sentences of a language.

Nor is the grammar a device that carries out a series of operations and gives well-formed sentences as its output.

Page 23: Cognitive-Functional Linguistics

June 19, 2006 RT/CFL 23

Composition 11

Putting together novel expressions is some-thing that speakers, not grammars, do.

It is a problem-solving activity that demands a constructive effort and occurs when linguistic convention is put to use in specific circumstances.

Creating a novel expression is not necessari-ly different in fundamental character from problem-solving activity in general.

Page 24: Cognitive-Functional Linguistics

June 19, 2006 RT/CFL 24

Categorization 1

Linguists are gradually coming to appreciate the critical significance of categorization to linguistic structure.

The role of categorization is especially prominent in cognitive grammar, which invokes it for several basic functions.

P. 369 in R. W. Langacker: Foundations of Cognitive Grammar. Vol. 1.

Page 25: Cognitive-Functional Linguistics

June 19, 2006 RT/CFL 25

Categorization 2

There are different models of categorization.

The strict Criterial-Attribute Model (the “Aristotelian model”), despite its dominance in the Western intellectual tradition, cannot be accepted unquestioningly as the basis for language structure and behavior.

Page 26: Cognitive-Functional Linguistics

June 19, 2006 RT/CFL 26

Categorization 3

The Criterial-Attribute Model A class is characterized by means of a list

of defining features. All members of the class fully possess

every property on the list; no nonmembers possess all of the listed properties.

Class membership is an all-or-nothing affair; a sharp distinction is drawn between those entities that are in the class and those that are not.

Page 27: Cognitive-Functional Linguistics

June 19, 2006 RT/CFL 27

Categorization 4

Well-known problems with the Criterial-Attribute Model – A It often happens that certain class

members lack a property so fundamental (on intuitive grounds) that it can hardly be denied criterial status: Flightless birds and egg-laying mammals

are familiar illustrations.

Page 28: Cognitive-Functional Linguistics

June 19, 2006 RT/CFL 28

Categorization 5

Well-known problems with the Criterial-Attribute Model – B A set of properties sufficient to pick out all

and only the members of a class might still be incomplete and inadequate as a characterization of that class.

If the semantic specifications [FEATHERLESS] and [BIPED] were in fact adequate as criterial features for the class of humans, we would nevertheless hesitate to accept these features as a comprehensive or revealing description of our species.

Page 29: Cognitive-Functional Linguistics

June 19, 2006 RT/CFL 29

Categorization 6

Well-known problems with the Criterial-Attribute Model – C Yet another problem is that speakers do

not adhere rigidly to criterial attributes in judging class membership:

I’ve never seen an orange baseball before! Look at that giant baseball! This tennis ball is a good baseball. Who tore the cover off my baseball? My baseball just exploded!

More on the next slide

Page 30: Cognitive-Functional Linguistics

June 19, 2006 RT/CFL 30

Categorization 7

Well-known problems with the Criterial-Attribute Model – C Speakers do not adhere rigidly to criterial attributes

in judging class membership: A speaker will not hesitate to call something a

baseball even if it happens — to be the wrong color (yellow) to be the wrong size (giant) to be wrong in virtually all criterial properties (tennis

ball) to be drastically deformed (without cover) to have ceased to exist (exploded)

Page 31: Cognitive-Functional Linguistics

June 19, 2006 RT/CFL 31

Categorization 8

Two other models, more directly grounded in cognitive concerns, appear to offer more revelatory and empirical-ly adequate accounts of linguistic cate-gorization:

The Prototype Model The Schema Model

Page 32: Cognitive-Functional Linguistics

June 19, 2006 RT/CFL 32

Categorization 9

The Prototype Model A prototype is a typical instance of a

category, and other elements are assimi-lated to the category on the basis of their perceived resemblance to the prototype; there are degrees of membership based on degrees of similarity.

Page 33: Cognitive-Functional Linguistics

June 19, 2006 RT/CFL 33

Categorization 10

The Prototype Model

The category plus the extensions constitute the category.

PROTOTYPE

EXTENSIONEXTENSION

EXTENSION / PROTOTYPE

EXTENSION

EXTENSION

Page 34: Cognitive-Functional Linguistics

June 19, 2006 RT/CFL 34

Categorization 11

The Schema Model A schema is an abstract characterization

that is fully compatible with all the members of the category it defines (so membership is not a matter of degree).

It is an integrated structure that embodies the commonality of its members, which are conceptions of great specificity and detail that elaborate the schema in contrasting ways.

Page 35: Cognitive-Functional Linguistics

June 19, 2006 RT/CFL 35

Categorization 12

The Schema Model

The schema plus the instances constitute the category

SCHEMA

INSTANCE/SCHEMA

INSTANCE INSTANCEINSTANCE

Page 36: Cognitive-Functional Linguistics

June 19, 2006 RT/CFL 36

Categorization 13

The Prototype Model and the Schema Model are intimately associated and are describable as aspects of a unified phenomenon. Categorization by extension typically

presupposes and incorporates schematic relationships.

A schema expresses the commonalities between a prototype and an extension.

Page 37: Cognitive-Functional Linguistics

June 19, 2006 RT/CFL 37

Categorization 14

Categories are networks containing prototypes and extensions, schemas and instances:

SCHEMA

INSTANCE/SCHEMA

INSTANCE / PROTOTYPE INSTANCE / EXTENSIONINSTANCE / EXTENSION

Page 38: Cognitive-Functional Linguistics

June 19, 2006 RT/CFL 38

Categorization 15: Baby

BABY1. She had a baby2. Baby carrots3. Hey, baby!4. Baby / babe5. He’s such a baby.6. Mr Platt is the baby

in his family

P. 169 in Goldberg (2006):

Constructions at work.

1. A human, B infant, C small, D cute, E emotionally immature, F youngest in a family

2. C small (and cute?)3. A human, D cute4. A human, D cute, G

sexy, H female, I adult5. A human, E emotionally

immature6. A human, F youngest in

the family

Page 39: Cognitive-Functional Linguistics

June 19, 2006 RT/CFL 39

Categorization 16: BabyA human, B infant, C small, D cute, E emotionally immature, F

youngest in a family, G sexy, H female, I adult

1. baby/ABCDEF

2. baby/C (D?)

3. baby/AD 4. baby/ADGHI

5. baby/ AE

6. baby/AFI

baby/AF

baby/A

Page 40: Cognitive-Functional Linguistics

June 19, 2006 RT/CFL 40

Categorization 17: Norw. /p/

[pʰ] [p] [pʷ][pʷʰ][pˑ][pˑʷ]

unvcd, bilab, plos

unvcd, unasp, bilab, plosunvcd, asp, bilab, plos

unvcd, unasp, bilab, plos, long unvcd, bilab, plos, short

unvcd, bilab, plos, rndunvcd, bilab, plos, unrnd