Hans-Jrg Schmid, GCLC 6, Erlangen 1 October 2014 1. Introduction: - - PowerPoint PPT Presentation

hans j rg schmid gclc 6 erlangen 1 october 2014
SMART_READER_LITE
LIVE PREVIEW

Hans-Jrg Schmid, GCLC 6, Erlangen 1 October 2014 1. Introduction: - - PowerPoint PPT Presentation

Hans-Jrg Schmid, GCLC 6, Erlangen 1 October 2014 1. Introduction: my plans for today a model of language: the Entrenchment and Conventionalization Model a glimpse of how key aspects of the EC-model can be tested Entrenchment and


slide-1
SLIDE 1

Hans-Jörg Schmid, GCLC 6, Erlangen 1 October 2014 1

Entrenchment and conventionalization. Testing some predictions of the model against historical data

Hans-Jörg Schmid, Ludwig-Maximilians-Universität München

1

  • 1. Introduction: my plans for today
  • a model of language: the Entrenchment and Conventionalization Model
  • a glimpse of how key aspects of the EC-model can be tested

Inspired by the work of Peter Auer Alice Blumenthal-Dramé Joan Bybee Bill Croft Nick Ellis Chuck Fillmore Dirk Geeraerts Adele Goldberg Franziska Günther Thomas Herbst Martin Hilpert Paul Hopper Daphné Kerremans Ron Langacker Annette Mantlik John Searle John Sinclair Anatol Stefanowitsch Michael Tomasello Elizabeth Traugott Alison Wray

2

  • 2. Some (more or less) indisputable facts about language
  • Linguistic knowledge is hosted in language users’ minds
  • Linguistic knowledge is represented in language users’ brains
  • Linguistic knowledge is shared by the members of a speech community
  • No two members of a speech community have identical linguistic knowledge
  • Individual and shared linguistic knowledge are both stable and dynamic
  • Language has, among others, a communicative function
  • The “structure of language” emerges from use in the service of the

communicative function under the influence of social, cognitive and pragmatic factors

  • Stability and change of language structure are determined by social, cognitive

and pragmatic factors

3

  • 3. Towards a model which tries to do justice to these facts
  • ... hosted in language users’ minds
  • ... represented in language users’ brains
  • ... shared by members of speech community
  • No two members ... have identical knowledge
  • ... knowledge ... are both stable and dynamic
  • ... a communicative function
  • ... emerges from use
  • ... determined by ... factors
  • cognitive
  • neurocognitive
  • sociocognitive
  • variational
  • adaptive
  • pragmatic
  • emergentist
  • usage-based
  • parsimonious, elegant, testable ...

4

slide-2
SLIDE 2

Hans-Jörg Schmid, GCLC 6, Erlangen 1 October 2014 2

  • 4. A blueprint of the Entrenchment and Conventionalization Model (EC-Model)

sociopragmatic processes CONVENTIONALIZATION cognitive processes ENTRENCHMENT cognitive forces emotive- affective forces communication usage pragmatic forces social forces

5

form meaning function form meaning function form meaning function form meaning function form meaning function form meaning function form meaning function form meaning function form meaning function form meaning function

  • linguistic knowledge is available to individuals in the form of networks of

associations showing different and constantly changing degrees and types

  • f entrenchment. (For descriptive purposes, these associative networks can

be understood in terms of constructions on different levels of schematicity.) symbolic association pragmatic association paradigmatic association syntagmatic association

situation

Entrenchment in associative networks (Langacker 2008: 226, Traugott and Trousdale 2013: 50ff, Schmid

2014 and Ms.)

  • 4. A blueprint of the Entrenchment and Conventionalization Model (EC-Model)

6

Example: What can I do for you?

  • 4. A blueprint of the Entrenchment and Conventionalization Model (EC-Model)

F: What Vaux/mod NP Vfull M/F: ‘asking about possible event’ wh-question construction

7

Example: What can I do for you?

symbolic association pragmatic association paradigmatic association syntagmatic association

  • 4. A blueprint of the Entrenchment and Conventionalization Model (EC-Model)

F: What Vaux/mod NP Vfull M/F: ‘asking about possible event’ F: Why Vaux/mod NP Vfull M/F: ‘asking for reason’ F: Who Vaux/mod NP Vfull M/F: ‘asking for person’

8

slide-3
SLIDE 3

Hans-Jörg Schmid, GCLC 6, Erlangen 1 October 2014 3

symbolic association pragmatic association paradigmatic association syntagmatic association

  • 4. A blueprint of the Entrenchment and Conventionalization Model (EC-Model)

F: What F: can F: I F: do F: for F: you M/F: ‘wh-question’ M/F: ‘permission’M/F: ‘S’ M/F: ‘action’ M/F: ‘benef.’ M/F: ‘H.’ F: What Vaux/mod NP Vfull M/F: ‘asking about possible event’ F: Why Vaux/mod NP Vfull M/F: ‘asking for reason’ F: Who Vaux/mod NP Vfull M/F: ‘asking for person’ Example: What can I do for you?

9

symbolic association pragmatic association paradigmatic association syntagmatic association

  • 4. A blueprint of the Entrenchment and Conventionalization Model (EC-Model)

F: What F: can F: I F: do F: for F: you M/F: ‘wh-question’ M/F: ‘permission’M/F: ‘S’ M/F: ‘action’ M/F: ‘benef.’ M/F: ‘H.’ F: What Vaux/mod NP Vfull M/F: ‘asking about possible event’ F: What can I do M/F: ‘offer’ S F: Why Vaux/mod NP Vfull M/F: ‘asking for reason’ F: Who Vaux/mod NP Vfull M/F: ‘asking for person’ Example: What can I do for you?

10

symbolic association pragmatic association paradigmatic association syntagmatic association

  • 4. A blueprint of the Entrenchment and Conventionalization Model (EC-Model)

F: What F: can F: I F: do F: for F: you M/F: ‘wh-question’ M/F: ‘permission’M/F: ‘S’ M/F: ‘action’ M/F: ‘benef.’ M/F: ‘H.’ F: What Vaux/mod NP Vfull M/F: ‘asking about possible event’ F: What can I do for you? F: What can I do M/F: ‘offer’ M/F: ‘please tell me your request’ S S F: Why Vaux/mod NP Vfull M/F: ‘asking for reason’ F: Who Vaux/mod NP Vfull M/F: ‘asking for person’ Example: What can I do for you?

11

F M/F F M/F F M/F F M/F F M/F F M/F F M/F F M/F F M/F

  • linguistic knowledge comes to be shared by the individuals in a speech

community by means of conventionalization but differs interindividually in terms of degrees and types of entrenchment (Booij 2010; 93, Boye and Harder 2012:

8)

Speaker 1 Speaker 2 Speaker 3 Conventionalization

  • 4. A blueprint of the Entrenchment and Conventionalization Model (EC-Model)
  • e.g. What can I do for you today?

F: wh-interrogative M/F:‘wh-question’ F: what can I V M/F: ‘offer’ F: what can I do for you M/F: ‘please tell me your request’ S S

12

slide-4
SLIDE 4

Hans-Jörg Schmid, GCLC 6, Erlangen 1 October 2014 4

  • 4. A blueprint of the Entrenchment and Conventionalization Model (EC-Model)

Entrenchment processes

  • take place in the mind
  • perate over associations
  • affect the strength of associations
  • affect the organisation of the cogni-

tive network Conventionalization processes

  • take place in social situations
  • perate over utterances
  • affect the behaviour of speakers
  • affect the organisation of “language”

in society meet via pragmatic associations

13

  • 4. A blueprint of the Entrenchment and Conventionalization Model (EC-Model)

Entrenchment processes

  • association
  • routinization and automization
  • schematization

Conventionalization processes

  • co-adaptation
  • diffusion
  • normation

14

  • 4. A blueprint of the Entrenchment and Conventionalization Model (EC-Model)

Definition entrenchment continuous routinization and re-

  • rganization of all types of associations,

depending on exposure to and frequency

  • f identical or similar processing events,

subject to the exigencies of the social environment Definition conventionalization continuous mutual coordination and matching of communicative knowledge and practices, subject to the exigencies

  • f the entrenchment processes taking

place in individual minds

15

increases likelihood of

  • 5. How does entrenchment work?

percept/ exposure motor control system sensory system processing cognitive system memory trace strengthening

  • f association

production ease of activation usage event usage event society mind

provides causes leaves brings about provides causes

reconfiguration

  • f network

increases

repetition causes routinization

entrenchment

16

slide-5
SLIDE 5

Hans-Jörg Schmid, GCLC 6, Erlangen 1 October 2014 5

  • routinization of symbolic associations
  • speedy and effortless activa-

tion of ‘constructions’

COTEXT COTEXT-

  • FREE

FREE ENTRENCHMENT ENTRENCHMENT

  • routinization of syntagmatic associations
  • chunking
  • weakening of paradigmatic

associations of parts

COTEXUAL COTEXUAL ENTRENCHMENT ENTRENCHMENT

  • routinization of pragmatic associations
  • context-dependent speed and

effort of activation of ‘constructions’

CONTEXTUAL CONTEXTUAL ENTRENCHMENT ENTRENCHMENT

  • routinization of paradigmatic associations
  • schemas, paradigms, analogy

SCHEMA SCHEMA-

  • FORMATION

FORMATION

  • routinization of schemas
  • productivity, ‘generative’

competence Effects of entrenchment in associative networks (Schmid Ms.)

  • 5. How does entrenchment work?

17

  • 6. How does conventionalization work?

society mind speaker 1 speaker 2

utterance 1

priming effect memory trace usage event 1 usage event 2 speaker 2 speaker 1 or n

utterance 1’

situation 1 speaker 2 speaker 3

utterance 1’’

usage event 3 situation 2 incipient diffusion entrenchment co-adaptation

18

  • 7. Forces acting on entrenchment and conventionalization

sociopragmatic processes CONVENTIONALIZATION cognitive processes ENTRENCHMENT cognitive forces

similarity contiguity salience categorization gestalt processing

emotive- affective forces

egocentrism emotion need for admiration fun aesthetic sensibility

communication usage pragmatic forces

meaning comprehending acting accomplishing telling arguing

social forces

empathy identity solidarity prestige

frequency

19

  • 8. A second example introducing the case study

“29th. [October 1664) All the talk is that De Ruyter is come over-land home with six or eight of his captaines to command here at home, ...” (The Diary of Samuel Pepys, kept from January 1660 to May 1669, first published 1825)

Portrait of Samuel Pepys by J. Hayls. Oil on canvas, 1666, 756 mm × 629 mm National Portrait Gallery, London. Source: Wikipedia, s.v. Samuel Pepys 20

slide-6
SLIDE 6

Hans-Jörg Schmid, GCLC 6, Erlangen 1 October 2014 6

All the talk is that De Ruyter is come

  • ver-land

F: DET N be that M/F: ‘’THING concept encapsulates proposition’ F: all the talk is that M/F: ‘there is a lot of talk about X’ F: DET N(linguistic) be that M/F: ‘speech-reporting noun encapsulates message’ substantive, lexically filled construction schema(tic construction) sub-schema

  • 8. A second example introducing the case study

21

All the talk is that De Ruyter is come

  • ver-land

F: all the talk is that F/M: ‘there is a lot of talk about X’ F: DET N(linguistic) be that F/M: ‘speech-reporting noun encapsulates message’

  • 8. A second example introducing the case study

22

DET N(linguistic) be that ‘speech-reporting noun encapsulates message’ All the talk is that De Ruyter is come

  • ver-land

F: DET N be that F/M: ‘’thing’ concept encapsulates proposition’

  • 8. A second example introducing the case study

23

  • 9. The case study (with Annette Mantlik, Heidelberg)

Corpus

  • approx. number of words

text types time number of tokens Gutenberg 18,500,000 fiction, essays, chronicles 1250-1871 1024 OED3 20,000,000 (est.) mixed 1250- 1871 194 Old Bailey 17,000,000 court proceedings 1300-1871 153 Helsinki 1,500,000 mixed 730-1710 13 PCEEC 2,200,000 letters

  • c. 1410-1695

140 Paston 200,000 letters 1422-1509 9 Oxford 140,400 letters 1796-1817 23 Shakespeare 900,000 drama, poetry 1590-1612 17 1573 Target: N + BE + that construction: all the talk is that, the problem is that

24

slide-7
SLIDE 7

Hans-Jörg Schmid, GCLC 6, Erlangen 1 October 2014 7

  • 10. Predictions of the model regarding entrenchment

Definition entrenchment continuous routinization and re-

  • rganization of all types of associations,

depending on exposure to and frequency

  • f identical or similar processing events,

subject to the exigencies of the social environment

25

repeated processing of entrenchment effects measurable in usage in terms of identical constructions

a....a.....a.....a....a....a ab...ab...ab...ab...ab

cotext-free entrenchment: routinization of symbolic association

  • relative frequency of

construction as produced by an individual speaker schematic constructions with variable slot fillers

axb ...ayb... azb...

schema-formation and routinization of paradigmatic associations

  • range of nouns
  • innovations and hapaxes as signs
  • f productivity
  • paradigmatic competitors

sequences of identical or similar constructions

a.b.c....a.b.c....a.x.c.... ab.cd.ef....ab.cd.xf...

cotextual entrenchment: routinization of syntagmatic association

  • preferences for certain nouns
  • emergence of patterns and

chunks identical or similar constructions in similar situations and/or with similar functions s

axb...ayb ...azb...

contextual entrenchment: routinization of pragmatic associations

  • relative frequency under specific

pragmatic conditions

(Schmid 2010, 2014, Ms., Schmid and Küchenhoff 2013)

  • 10. Predictions of the model regarding entrenchment

26

Frequency per million words: selected essay authors

0,00 50,00 100,00 150,00 200,00 250,00 300,00

1643 1796

  • 11. Cotext-free entrenchment

27

0,00 20,00 40,00 60,00 80,00 100,00 120,00

Frequency per million words: selected fiction authors

1667 1871

  • 11. Cotext-free entrenchment

28

slide-8
SLIDE 8

Hans-Jörg Schmid, GCLC 6, Erlangen 1 October 2014 8

1

David Hume Range of nouns/preference for nouns: entrenched schema?

  • 12. Schema-formation

29

1

David Hume Range of nouns/preference for nouns: entrenched schema? hapax innovation

  • 12. Schema-formation

30

1 2 3 4 5

Geoffrey Chaucer Range of nouns, preference for nouns, innovations and hapaxes as signs of productivity hapax preference innovation

  • 12. Schema-formation

31

2 4 6 8

Thomas Paine

5 10 15 20 25

Edmund Burke

2 4 6 8 10 12 14 16

Samuel Johnson

32

slide-9
SLIDE 9

Hans-Jörg Schmid, GCLC 6, Erlangen 1 October 2014 9

2 4 6 8 10 12

Jane Austen

5 10 15

Charles Dickens

1 2 3 4

George Eliot

33

1 2 3 4 5 6 7

Samuel Pepys

1 2

consequence reason cause

George Berkeley Paradigmatic competitors

  • 13. Paradigmatic associations

34

2 4 6 8

them therein. Here all the discourse is, that now the King is of opinion to have the Parliament uld depart the town. All the discourse now-a-day is, that the King will come again; and for all I see, it i the reason of all this. The great discourse now is, that the Parliament shall be dissolved and another cal died the last week. All the news now is, that Sir Jeremy Smith is at Cales with her part mighty well. All the news now is that Mr. Trevor is for certain to be Secretary in Morr which was made him. All the news from London is that things go on further towards a King. That the Ski they must be purified. The worst news he tells me, is that Mr. Chetwind is dead, my old and most ingenious a to the office, where all the news this morning is that the Dutch are come with a fleet of eighty sail to monk's soldiers. 4th. All the news to-day is, that the Parliament this morning voted the House to be ine gentleman. 27th. All the news this day is, that the Dutch are, with twenty- two sail of ships of ing leave of him. But the best piece of newes is, that instead of a great many troublesome Lords, the wh as very noble. 29th. All the talk is that De Ruyter is come over-land home with six or eigh , and hath good rest. All the talk is that my Lord Sandwich hath perfected the peace with Sp ent, which vexed me. The great talk is, that the Spaniards and the Hollanders do intend to set m good night. 16th. The talk upon the 'Change is, that De Ruyter is dead, with fifty men of his own ship

Preference for nouns and emergence of chunks and patterns Samuel Pepys

  • 14. Emergence of patterns and chunks

35

1 2 3 4

and give concern to your honest friends: and all my pleasure is, that I can and will make you amends for the disturbanc man, and deserved this tribute to her memory. All my pleasure now is, that she knew not half my wicked pranks, and that I e heart to such protection as I could find. All my comfort is, that your advice repeatedly given me to the same purpo think so. All the comfort I know of in children, is, that when young they do with us what they will, and al iscretions to do as you think fit in it. All my concern is, that this daring and foolish project, if carried on, w

  • n her refusal to receive her. All her consolation is, that her unhappy situation is not owing to her own ina

, was no more? Her principal consolation, however, was, that she should not long survive her. She hoped, she s r may be refused to my corpse; in this case my desire is, that it may be interred in the churchyard belonging to know, has always been my principal care. All my fear is, that, when she comes to the point, the over-niceness o ust think so, let him say of me what he will. All my fear is, that, as he knows I am in disgrace with a family whose

  • me tidings of his lady.' And now, Belford, all my hope is, that this fellow (who attended us in our airing to Ham
  • f such as think your going away a fault. The hope is, that things will still end happily, and that some peop

Samuel Richardson Preference for nouns and emergence of chunks and patterns

  • 14. Emergence of patterns and chunks

36

slide-10
SLIDE 10

Hans-Jörg Schmid, GCLC 6, Erlangen 1 October 2014 10

directed vnto youe , whiche myn advise and counsail is that youe shall in any wise ensue , and that your chauncelour shall , whiche myn advise and counsail is that youe shall in any wise ensue , and that your chauncelour shall do the semblable Sir Robart Wyngfeld within the Marches of Calays , the kinges pleasure is that ye and thother Commyssioners shall I pray you shew him on my behalf that my poure and frendelie aduise is that his grace shall liberally wryte to And the pleasure of his highnes is that the due dettes of the said howses well prouyde without Covyn it canne not be enhabited , his graces pleasure is that your lordship shall cause some oone or two of the consail there The kinges pleasure is that you shal sende a spedy aunswer herof , and therwith signifie suche occurrantes as For remedy whereof the kinges graciouse pleasur is that ye shal cause the said persones suspected to be thoroughly Wherfor his gracious pleasure and commandment is that forasmoch as we hope chiefly and specially the Frensh King Wherfor his gracious pleasure and commandment is that forasmoch as we hope chiefly and specially the Frensh Kin ye can not so sone com the kinges maiestes will is that ye shal make the shortiest abode there ye can knowe herafter The kinges Maiestes pleasure is that you shal vieu his graces howse there called Thexchequie the Kinges Maieste whose graciouse pleasure is that with all celerite ther $shall $be {TEXT:shalbe} that the worst may be provided for , his graces pleasure is that it $shall $be {TEXT:shalbe} diligently forseen ther

2 4 6 8 10

Letters by Thomas Cromwell Preference for nouns and emergence of chunks and patterns

  • 14. Emergence of patterns and chunks

37

Effect of communicative function on emergence of patterns (Schmid 2014)

and give concern to your honest friends: and all my pleasure is, that I can and will make you amends man, and deserved this tribute to her memory. All my pleasure now is, that she knew not half my e heart to such protection as I could find. All my comfort is, that your advice repeatedly given

  • u know, has always been my principal care. All my fear is, that, when she comes to the point,

ust think so, let him say of me what he will. All my fear is, that, as he knows I am in disgrace with

  • me tidings of his lady.' And now, Belford, all my hope is, that this fellow (who attended us in
  • f such as think your going away a fault. The hope is, that things will still end happily, and that
  • rk.] which was made him. All the news from London is that things go on further towards a King. That the

they must be purified. The worst news he tells me, is that Mr. Chetwind is dead, my old and most ingenious and to the office, where all the news this morning is that the Dutch are come with a fleet of eighty sail to ull of Monk's soldiers. 4th. All the news to-day is, that the Parliament this morning voted the House to be ys a fine gentleman. 27th. All the news this day is, that the Dutch are, with twenty- two sail of ships of t taking leave of him. But the best piece of newes is, that instead of a great many troublesome Lords, the w s word, which was very noble. 29th. All the talk is that De Ruyter is come over-land home with six or eigh e is much better, and hath good rest. All the talk is that my Lord Sandwich hath perfected the peace with Sp ye can not so sone com the kinges maiestes will is that ye shal make the shortiest abode there ye can knowe herafter The kinges Maiestes pleasure is that you shal vieu his graces howse there called Thexchequie the Kinges Maieste whose graciouse pleasure is that with all celerite ther shall be that the worst may be provided for , his graces pleasure is that it shall be diligently forseen ther

Richardson: expressive Pepys: reporting Cromwell: directive

  • 15. Contextual entrenchment

38

The usage profiles of comparable authors differ in terms of findings indicating David Hume Samuel Pepys Edmund Burke relative frequency of construction cotext-free entrenchment low high high range of nouns schema- formation and routinization of paradigmatic associations very limited limited wide innovations and hapaxes as signs of productivity

  • nly one

yes yes paradigmatic semantic competitors no yes yes preferences for certain nouns cotextual entrenchment no strong strong emergence of patterns and chunks no yes yes relative frequency under specific pragmatic conditions contextual entrenchment no yes, one function yes, several functions

  • 16. Discussion

39

David Hume Samuel Pepys Edmund Burke

  • 16. Discussion

...

40

slide-11
SLIDE 11

Hans-Jörg Schmid, GCLC 6, Erlangen 1 October 2014 11

  • 17. General discussion
  • so it’s associations all over the place, rather than constructions all the way up

and down

  • different types of associations
  • though associations form a huge, messy, endless and constantly changing

network, the processes and forces shaping the organization of the network are limited

41

  • 17. General discussion
  • try not to reify constructions without sound evidence
  • consider on which level of schematicity a potential construction could be

available

  • consider on which level of schematicity a potential construction could be

available to whom

42

  • 17. General discussion
  • include the social side of language, i.e. conventionalization processes and their

determinants

  • keep apart cognitive and social processes, but study how they interact
  • don’t forget that language has a communicative function: study the

fundamental role of pragmatic associations

  • don’t forget that frequency is not just out there to be counted in the corpus but

has cognitive, social and pragmatic sources

  • understand that frequency is both a source and an effect of entrenchment and

conventionalization

43

Selected references Blumenthal-Dramé, Alice . 2012. Entrenchment in Usage-based Theories: What Corpus Data Do and Do not Reveal about the Mind. Berlin etc.: Walter de Gruyter. Bybee, Joan. 2010. Language, usage and cognition. Cambridge/New York: Cambridge University Press. Croft, William. 2000. Explaining Language Change: An Evolutionary Approach. Harlow/New York: Longman. Ellis, Nick C. and Diane Larsen-Freeman, eds. 2009. Language as a Complex Adaptive System. Chichester: Wiley-

  • Blackwell. [= Language Learning 59: Suppl. 1]

Hilpert, Martin. 2014. Construction grammar and its application to English: Edinburgh: Edinburgh University Press. Langacker, Ronald W. 2008. Cognitive grammar: a basic introduction. Oxford/New York: Oxford University Press. Schmid, Hans-Jörg. 2010. Does frequency in text really instantiate entrenchment in the cognitive system? In Quantitative methods in cognitive semantics: Corpus-driven approaches, Dylan Glynn and Kerstin Fischer (eds.), 101–

  • 133. Berlin etc.: de Gruyter Mouton.

Schmid, Hans-Jörg. 2014. “Lexico-grammatical patterns, pragmatic associations and discourse frequency.” In Thomas Herbst, Hans-Jörg Schmid & Susen Faulhaber (eds.), Constructions – collocations – patterns. Berlin etc.: Mouton de Gruyter. Schmid, Hans-Jörg. Ms. A framework for understanding linguistic entrenchment and its psychological foundations in memory and automatization. (at present available at: http://www.anglistik.uni-muenchen.de/personen /professoren/schmid/schmid_publ/ entrenchment-intro.pdf; July 2014); to appear in: Entrenchment, memory and

  • automaticity. The psychology of linguistic knowledge and language learning, Hans-Jörg Schmid (ed.). Boston: APA and

Walter de Gruyter. Schmid, Hans-Jörg and Helmut Küchenhoff. 2013. “Collostructional analysis and other ways of measuring lexicogrammatical attraction: theoretical premises, practical problems and cognitive underpinnings”. Cognitive Linguistics 24.3, 531-577. Tomasello, Michael. 2003. Constructing a Language. A Usage-Based Theory of Language Acquisition. Cambridge, MA/London: Harvard University Press. Traugott, Elizabeth Closs and Graeme Trousdale. 2013. Constructionalization and constructional change. Oxford: Oxford University Press.

44