Conversationalization in public discourse

Download Report

Transcript Conversationalization in public discourse

Conversationalization
in public discourse
Tryntje Pasma
Kirsten Vis
1
Overview
•
Introduction
•
•
•
•
Conversationalization
VU-Ster project
Metaphor
Subjectivity
2
Conversationalization
•
Hypothesis by Norman Fairclough (CDA)
•
“the modelling of public discourse upon the
discursive practices of ordinary life, ‘conversational’
practices in a broad sense” (Fairclough, 1994: 253)
•
movement towards norms of ‘casual’ conversation
in university brochures, news reports etc.
•
•
Example:
Di’s butler bows out . . . in sneakers.
(headline Daily Mirror)
•
•
Conversational vocabulary
Graphic devices
3
Di’s butler bows out . . . in sneakers.
•
description of certain stylistic markers as
‘conversational’ is problematic (Pearce (2005))
•
Which word is conversational?
• ‘bows out’ (vs. ‘resigns’)?
‘sneakers’ (vs. ‘trainers’)?
• ‘Di’ (vs. ‘Diana’ / ‘Princess Diana’)?
Lexical density
Tense
•
•
•

Intuitively plausible but intuitive approach
4
VU-Ster project
•
Goal
• empirically test Fairclough’s conversationalization
hypothesis for Dutch public discourse
•
Corpus analysis
• Dutch news from 1950 <-> 2002
• Dutch news from 2002 <-> Dutch conversations from 2002
News
• 1950: 30,000w
• 2002: 50,000w
• 5 national newspapers; different sections
Conversations
• 50,000w from Corpus of Spoken Dutch
• 30 complete spontaneous conversations
•
•
5
To metaphor…

Conversationalization includes:


Biber’s features of involved vs informational production


colloquial vocabulary; phonic, prosodic and paralinguistic
features of colloquial language; direct address (you and we);
repetition; lack of subject-verb agreement
involved: causative subordination; wh-questions/clauses; etc.
Can same be observed for metaphor?

seen as conventional, stylistic property, rhetorical effect
6
Metaphor in conversation

Studies of metaphor in conversations




Focus on certain forms and functions (Cameron 2003, 2008;
Drew & Holt 1995) in certain settings
Idiomatic expressions
Delexicalised verbs (lexical density)
Position in sentence
7
Idiomatic expressions

Their role in English conversations



Cheshire (2005): fixed expressions function as a means to
help speakers keep up with the demand of online speech
production
Drew and Holt (1995): idiomatic expressions in naturally
occurring conversations seem to be used predominantly for
topic summarizing and topic termination purposes
Their role in Dutch conversations

Termination and summary function; topic transition and
start of new topic
8
Example
78: ja je gaat vanavond maar weer flink te s aan de zuip.
79: ik heb een kater vandaag gewoon.
80: ongelooflijk.
81: ik heb uh helse pijnen doorstaan.
82: ik ben nog maar net uit b uit bed.
83: net nou eigenlijk net.
84: ja dat kasteelbier van jou dat uh dat ga dat hakt erin als een
kasteel de volgende dag.
85: ja die zijn inderdaad genadeloos ja.
86: ik heb trouwens uh...
87: ben net even naar de videotheek geweest.
88: en daar lag gewoon Lars Von Trier The Idiots bij de videotheek.
89: bij in de vijfhoek.
(fn000496)
9
Idioms in news similar?

Occur more often in news than in conversation


Functions are similar


roughly 1 per 500 words
a lot of the examples fit within termination, summary,
transition function
Example:
Bijbelimporteur drijft wig tussen China en de VS
(Vbu2)
10
Position of metaphor

Do metaphor-related words occur at
beginning, middle or end of a sentence

What is expected? Are expectations different
for conversation and news?

What are the results? Are results different for
conversation and news?
11
Results conversation
12
Results news
13
Form, function, position

Comparison between conversations and
news


patterns on different levels
diachronic element for conversationalization


with respect to form and position
Conceptual analysis of metaphor

patterns in registers
14
To subjectivity
•
Definition:
the degree to which the presence of the
speaker (/writer) is felt
e.g. when speaker gives opinion or shows (un-)certainty
•
Why subjectivity?
•
•
Examples:
•
•
•
•
Presence of speaker in conversations
It is a beautiful city.
Maybe your friend will come to the party.
John must be ill.
SPEAKER subjectivity
15
Analysis
•
Two levels
1.
2.
Text: coherence relations
Sentence / word: lexico-grammatical
16
1. Text level: coherence relations
•
•
•
Relations between text parts like Cause-Consequence,
Contrast, Evidence etc.
Capture part of what makes a text a text (rather than a random
set of sentences)
Starting point:
Rhetorical Structure Theory (RST; Mann & Thompson, 1988)
•
fairly exhaustive list of 24 well-defined relations
17
Subjective relations
10 subjective relations
•
1.
2.
3.
4.
5.
6.
7.
8.
9.
10.
Antithesis
Concession
Concessive opposition
Enablement
Enumeration
Evidence
Evaluation
Interpretation
Justify
Motivation
Conversationalization hypothesis:
•
•
The relative amount of subjective relations has increased
over time.
18
Subjective relations in news
1950
2002
per 10,000 w
per 10,000 w
Antithesis
7.6
6.8
0.138149
Concession
28.5
29.3
0.033829
Concessive opposition
12.8
8.2
3.431172
Enablement
0.0
1.6
***
Enumeration
4.3
5.4
0.39456
Evaluation
10.0
6.4
2.558607
Evidence
26.1
37.8
5.966287
increase
Interpretation
16.2
6.4
15.36982
decrease
Justify
1.9
0.2
***
Motivation
0.0
0.0
***
107.5
102.2
Chi2
0.396887
no
change
20
Conclusion coherence relations
•
Overall number of subjective relations has not
changed significantly, but the nature of the textual
subjectivity has:
‘Old’ newspapers interpret more,
‘new’ newspapers prove / conclude more
•
Explanation:
•
back to texts
21
2. Sentence/word level
•
Following Bekker (2006), Scheibman (2002), Wiebe (2005)
•
Intensifiers
•
•
Modal verbs
•
•
think, say
First and second person pronouns
•
•
maybe, presumably, certainly
Verbs of cognition
•
•
must
Modal adverbs
•
•
very, enormously
I, you
Direct questions
•
uncertainty; listener is addressed
22
Subjective elements
Chi2
1950
2002
conversations
10,000 w
10,000 w
1p sg pron
8.6
32.0
32.6702
348.4
1p pl pron
21.9
22.7
0.049368
84.3
2p sg pron
4.8
15.2
13.30681
248.6
2p pl pron
0.0
0.0
***
10.2
modal verb
36.7
42.6
1.29107
57.3
modal adverb
33.3
32.3
0.048473
57.7
particle
76.1
90.3
3.530204
196.0
verb of
cognition
53.2
35.0
12.38341
66.0
intensifier
249.1
250.8
0.017301
792.4
question
13.8
8.0
5.241389
207.4
deictic
27.6
23.7
0.901491
45.5
10,000 w
23
Conclusion lexico-grammatical analysis
•
•
Only marginal support for conversationalization
hypothesis
More research is needed
•
Verbs of cognition
•
•
only 1st person
Direct speech
•
•
exclude character speech
not straight-forward: e.g. Semi-Direct Speech
24
Conclusion lexico-grammatical analysis
•
•
Only marginal support for conversationalization
hypothesis
More research is needed
•
Verbs of cognition
•
•
only 1st person
Direct speech
•
•
•
•
exclude character speech
not straight-forward: e.g. Semi-Direct Speech
De enige werkelijke oplossing is de sluiting van het terrein,
stelt Molenman. (Nbi1)
The only real solution is closing the area, says Molenman.
De Ned. marine had een zeer gunstige indruk op hem
gemaakt, zo zei hij. (TRObu2)
The Dutch navy had made a very favourable impression on
him, he said.
25
What’s next?
•
•
•
Refine lexico-grammatical analysis
Automatic lexico-grammatical analysis of larger
corpus
Qualitative studies
•
•
•
•
•
conversation 1950
analysis of perspective
etc.
Reception experiment
Possibly automatic analysis of adjectives and
nouns with help from Computational Lexicology
& Terminology Lab
26
Thank you!
27