Priming and language evolution

1
Priming and language evolution

• Gerhard Jäger
• (University of Bielefeld)
• Anette Rosenbach
• (University of Düsseldorf/ University of
  Paderborn)

Language Change and Evolution Integrating
Formal and Functional Perspectives King's
College, London, 15-16 December 2006
2
Integrating formal and functional perspectives on
language change and evolution

• 1. Psycholinguistic account of language change
  (based on priming)
• ? functionalist approach to language change
• 2. Priming and language evolution
• ? formalization of functionally-driven language
  change in an evolutionary model of language change

3
Outline

• I. Priming and language change
• Grammaticalization and the hypothesis of
  unidirectionality
• Priming
• Unidirectional change and priming 2 case studies
• How priming leads to unidirectionality ...
• II. Priming and language evolution
• Priming and exemplar dynamics
• Some computer simulations
• George Prices General Theory of Selection
• Modeling unidirectionality in Prices framework

4
1. Grammaticalization and the hypothesis of
unidirectionality
5
1.1 Grammaticalization

• lexical elements gt grammatical elements
• example going to in English
• Im going (to London) to visit my aunt.
• Im going to read tomorrow.
• Its going to rain.
• Its gonna rain.

lexical (main) verb locomotion going (to)
auxiliary verb time (future) gonna
6
1.2 Hypothesis of unidirectionality

• controversial issue (see e.g. special issue of
  Language Sciences 23 Newmeyer 1998 Lass 2000
  Haspelmath 2004)
• consensus most grammaticalization processes
  cannot be reversed
• Open question Why should that be so ?

7
1.3 Traditional explanations for unidirectionality

• economy vs. clarity (e.g. von der Gabelentz
  1901)
• maxim of extravagance (Haspelmath 1999)

8
1.4 Present account

• psycholinguistic motivation of unidirectionality
• falsifiable predictions
• couched within an evolutionary approach to
  language change

9
2. Priming
10
2. Priming

• cognitive mechanism
• priming preactivation
• stimulus ( prime) target

11
Priming example (1)

• Repetition priming (direct priming)
• At what time does your shop close?
• ? at six
• (b) What time does your shop close?
• ? six oclock
• (cf. Levelt Kelter, 1982)

12
Priming example (2)

• associative priming (indirect priming)
• (cf. Flores dArcais Schreuder, 1987)

doesnt
primes
prime
13
Priming psycholinguistic evidence

• form priming
• phonological, morphological, lexical, and
  syntactic priming
• meaning priming
• semantic priming
• (cf. e.g Bock 1986, 1987 Flores dArcais
  Schreuder 1987 Bock Loebell 1990 Zwitslerlood
  1996 Pickering Branigan 1999 Levelt et al.
  1999, Branigan et al. 2000 Pickering Garrod
  2004, inter alia)

14
3. Unidirectional change and primingempirical
evidence

• 3.1 Space gt time
• 3.2 Phonological reduction

15
3. 1 Space to time

• space-time correspondences in language

• unidirectional development space gt time (e.g.
  Heine et al. 1991
• Haspelmath 1997, Heine Kuteva 2002, Hopper
  Traugott 200385)

16
Case study Ispace gt time

• Boroditsky (2000)
• evidence from experimental priming studies

17
Temporal metaphor

• 2 dominant spatial metaphors to sequence events
  in time (cf. e.g. Clark, 1973)

ego-moving metaphor
The revolution is before us.
time-moving metaphor
The revolution was over before breakfast.
(from Boroditsky, 20005)
18
Spatial metaphor
ego-moving metaphor
object-moving metaphor
(from Boroditsky, 2000 6)
19
Boroditsky (2000) experiment 1

• primes
• ego-moving spatial e.g. The dark can is in front
  of me.
• object-moving spatial e.g. The light widget is
  in front of the dark widget.
• targets ambiguous temporal sentences, e.g.Next
  Wednesdays meeting has been moved forward two
  days.

20
Boroditsky (2000) results

• ego-moving spatial prime ego-moving
  temporal interpretation
• (The dark can is in front of me.)
  (meeting on Friday)
• object-moving spatial time-moving
  temporal prime interpretation
  
• (The dark widget is in front
  (meeting on Monday)
• of the light widget.)
• ? space can prime time !

21
Boroditsky (2000) experiment 2

• 4 primes
• spatial
• ego-moving e.g. The flower is in front of me.
• object-moving e.g. The hat-box is in front of
  the Kleenex.
• temporal
• ego-moving e.g. On Thursday, Saturday is before
  us.
• time-moving e.g. Thursday comes before Saturday.
• 2 targets
• ambiguous time questions e.g. Next Wednesdays
  meeting has been moved forward two days.
• ambiguous space questions e.g. Which one of
  these widgets is ahead ?

22
Boroditsky (2000) results from experiment 2
(from Boroditsky 2000 14)
23
3.2 Case study II phonological reduction

• phonological reduction in grammaticalization
• e.g. going to gt gonna let us gt lets
• unidirectional development (e.g. Hopper
  Traugott 19932003)

24
Shields Balota (1991)

• experimental study on effects of repetition on
• word length
• amplitude

25
Shields and Balota (1991)

• 3 conditions
• Identical prime and targetHer cat chases our cat
  under the table.
• (2) Related prime and targetHer dog chases our
  cat under the table.
• (3) Unrelated prime and targetHer son chases our
  cat under the table.

26
Shields and Balota (1991) Results

• duration
• (cat) cat 329 msec
• (dog) cat 340 msec
• (son) cat 350 msec

27
Shields and Balota (1991) Results

• amplitude
• (in comparison to reference vowel)
• (cat) cat -1.62 dB
• (dog) cat -0.11 dB
• (son) cat 0.23 dB

28
Shields Balota (1991)

• Summary of results
• repetition and associative priming ? shortening
• repetition priming ? reduced amplitude
• ? evidence for asymmetric priming

29
4. How priming leads to unidirectionality...
30
4.1. From usage to grammar

• priming short-lived phenomenon in language usage
• open question preferences in grammar ?
• language usage
• ? problem of linkage (Kirby 1999)

31
4.2 Implicit learning

• structural priming evidence for persistence of
  priming effects
• persistence of structural priming effects beyond
  adjacent sentence, over 10 intervening sentences
  in experimental studies (Bock Griffin 2000)
• temporal persistence in spontaneous speech
  (Szmrecsanyi 2006 Recchia Bresnan 2006)
• evidence for cumulative priming effects (e.g.
  Luka Barsalou 2005)
• ? implicit learning, modelled within a
  connectionist neural network (Bock Griffin
  2000 Chang et al. 2000 Chang et al. 2006)

32
4.3 Implicit learning(cf. Chang et al. 2006)

• mechanism modelling learning of syntactic
  representations
• error-based learning
• ? priming effects (in language usage) have an
  immediate impact on grammar !

33
4.4 Synthesis

• asymmetric priming
• 1. concept or form A concept/ form B
• presupposes sufficient similarity between A and
  B
• repeated usage implicit learning
• 2. asymmetric priming grammar
• (language usage) (competence)
• 3. grammaticalization (unidirectional pathways)
  

34
Diachronic unidirectional pathways

• decomposable into atomic steps of
• asymmetric priming
• in language use

35
Outlook for further research

• Hypothesis of asymmetric priming
• falsifiable predictions
• can be tested with present-day speakers
• Prediction unidirectional pathway A gt B
• ? A primes B, but B does not prime A

36
Using the present to understand the past

• Uniformitarian assumption
• The same cognitive (psycholinguistic) mechanisms
  operating today, have also been operating in past
  speakers.
• ? psycholinguistic mechanisms of language change
  (cf. also Tabor 1994 Rosenbach 2002 Kempson
  Cann 2005, this workshop)

37
Conclusion I

• psycholinguistic mechanisms and methods
• language change