Transcript Document

Competition, Copying and Cues: The
Acquisition of Wh-questions in English
and Norwegian
Marit Westergaard
Department of Language &
Linguistics/CASTL
17.07.2015
University of Tromsø
1. Introduction
(1) Dad:
Child:
Dad:
Child:
Where’s Mommy?
She goed to the store.
Mommy goed to the store?
NO! (annoyed) Daddy, I say it that
way, not you!
(Pinker 1999: 199)
17.07.2015
University of Tromsø
2
•
Optionality in the child grammar
Wh-questions in English:
(2) Why he can’t hit? (Adam 3;4.01)
(3) What am I saying? (Adam 3;4.01)
•
Optionality in the input
Wh-questions in Norwegian (Tromsø):
(4) Ka sir du? / Ka du sir?
what say you / what you say
‘What are you saying?’
17.07.2015
University of Tromsø
3
2. The Structure of the Target Languages
English: Subject-auxiliary inversion
(5)
Peter will eat the olives.
What will
Peter
eat?
• No inversion in embedded questions/declaratives:
(6) I don’t know [*what will Peter eat]
(7) *Then will Peter eat.
•
No inversion w/lexical verbs, except be:
(8) *What ate Peter? / (9) Where was Peter?
Residual Verb Second (V2) (Rizzi 1996)
17.07.2015
University of Tromsø
4
Norwegian: ‘Classical’ V2
Dialects: No strict V2 in wh-questions (e.g. Vangsnes 2006).
(10)
kor er mitt fly? (INV, file Ole.17)
where is my plane
‘Where is my plane?
(11)
kor
vi lande henne? (INV, file Ole.17)Non-V2: [pron+V]
where we land LOC
‘Where do we land?’
V2:
Non-V2:
17.07.2015
subject is discourse new
subject is discourse given
University of Tromsø
V2: [be+DP]
(Westergaard 2003)
5
• Long (phrasal) wh-elements require V2:
(12)
Korfor kommer du? /*Korfor du kommer?
why come you
‘Why are you coming?’
• Embedded questions require non-V2 (like English)
• Subject questions require non-V2:
(13)
17.07.2015
Kem som kommer? /*Kem kommer?
who SOM come
‘Who is coming?’
University of Tromsø
6
•
•
•
•
Word order variation in wh-questions in adult grammars
dependent on:
clause type
(question vs. decl., main vs. embedded)
wh-element
(short vs. long, subject vs. non-subject)
verb
(lexical verbs vs. aux and/or be)
subject
(given vs. new)
=> quite a bit of detail must be learned from input.
17.07.2015
University of Tromsø
7
3. Competition, Copying or Cues
Generative grammar (Competition)
• Children endowed with a Universal Grammar (UG)
containing major word order parameters, e.g. +/-V2.
• Children only need to be exposed to a few examples to set
the parameter and generalize to all cases (e.g. Wexler 1999).
• Children’s mistakes due to competing parameter settings.
17.07.2015
University of Tromsø
8
Constructivist accounts (Copying)
• No UG - children learn from input only.
• Early grammar has no categories (N or V) or rules (e.g. Saux inversion/V2).
• Children sensitive to frequent word combinations in the
input, e.g. wh-word+aux.
(2’)
(3’)
Why he can’t hit? (Adam 3;4.01)
What am I saying? (Adam 3;4.01)
[why+can’t]
[what+am]
Rowland&Pine 2000, 2003, Rowland et al 2003, Ambridge et al 2006.
17.07.2015
University of Tromsø
9
Cues
• Cue is piece of (hierarchical) structure, produced in a
child’s I-language on exposure to triggers in the input.
(14)
Cue for V2 syntax: CP[XP CV...]
(Lightfoot 2006: 86)
 BUT: Given the variation in adult languages, cues must be
much more fine-grained - i.e. micro-cues.
(Westergaard 2007, forthcoming, Lightfoot&Westergaard 2007)
Micro-cues
(15)
17.07.2015
Cue for V2 in wh-questions (English): IntP[(wh) IntI...
University of Tromsø
10
Predictions
• Setting major word order parameter:
Massive overgeneralization
• Copying frequent word combinations:
Some overgeneralization (frequent => infrequent, i.e.
embedded questions, questions with long wh-elements)
• Micro-cues:
Generally target-consistent production
17.07.2015
University of Tromsø
11
4. Acquisition data - Norwegian
Corpus of Norwegian child language (Tromsø), Anderssen (2006).
Long wh-phrases
Target-consistent V2 (96%, 97/101)
(Westergaard 2003, 2005).
Embedded wh
Target-consistent non-V2 (99.1%,107/108)
Monosyllabic wh-words
Target-consistent V2 and non-V2
17.07.2015
University of Tromsø
12
Word order dependent on information structure:
(16)
kor
e babyen? (Ina 2;1.0)
where is baby.DEF
‘Where is the baby?’
(17)
ka løva
like å spise mamma? (Ann 2;6.21)
what lion.DEF likes to eat mommie
‘Mommie, what does the lion like to eat?’
17.07.2015
University of Tromsø
13
Subject questions
• Functional element som missing at early stage.
• Initial word combination ka som/kem som - not copied:
(18)ADULT: nei og nei ka som skjer der
no and no what SOM happens there
‘Oh no, what is happening there?’
CHILD: nei og nei ka skjer der. (Ole 2;1.5)
17.07.2015
University of Tromsø
14
5. Acquisition data - English
• As soon as aux appear, target-consistently inverted.
• No overgeneralization of S-aux inversion to other clause types
or verb types. (Radford 1992, Roeper 1999, 2007)
Adam?
Questions with be
(Westergaard 2008)
Target-consistent inversion 96.4% (455/472)
(19)
17.07.2015
where is a box? (Adam 3;0.11)
University of Tromsø
15
Embedded wh-clauses
Target-consistent non-inversion 94.2% (97/103)
(20)
So we can know [where the mailman is]. (Adam 3;2.21)
[where + is]
frequent combination in main clauses
Long wh-elements
Target-consistent inversion 91.7% (35/39)
(21)
17.07.2015
What kind of butterfly is this?
University of Tromsø
(Adam 3;3.18)
16
What’s Adam’s problem?
•
Distinction aux/be
aux - inverted 34.2% (25/73), age 3;2-3;5
•
Distinction between wh-elements
what - inverted 96.6% (689/713)
why - inverted 11.9% (7/59)
(2’) Why he can’t hit? (Adam 3;4.01)
17.07.2015
University of Tromsø
17
6. Acquisition Data - Summary
• Children zoom in on target word order in different contexts
early - making distinctions between linguistically relevant
(sub-) categories - micro-cues.
• No copying of frequent word combinations - no
competition between major parameter settings.
• Given the complexity of word order in wh-questions, a cue
may be delayed.
• Adam’s grammar is conservative/makes finer distinctions
than the target language (even smaller micro-cues) - e.g.
does not generalize be to aux, what to why (cf. Tromsø).
17.07.2015
University of Tromsø
18
7. Conclusion
(1’) Mommy goed to the store.
(2’) Why he can’t hit?
• Overgeneralization of past tense morphology -ed vs.
‘underapplication’ of S-aux inversion.
• No principled reason why goed is ungrammatical - -ed rule
applies blindly to any verb.
• Strictly speaking, no principled reason why (2) is
ungrammatical either (cf. Indian Vernacular English).
BUT: There ARE principled reasons for when the S-aux/V2
rule applies - micro-cues, preventing children from
overgeneralizing.
17.07.2015
University of Tromsø
19
• Learning the target form has no effect on
meaning/communication.
goed = went
Why he can’t hit? = Why can’t he hit?
• Yet, well before the age of 3, children distinguish between
main and embedded wh-questions, subjects and nonsubjects, phrases and heads, auxiliaries and lexical verbs...
Why should they care?
17.07.2015
University of Tromsø
20
Language is certainly a powerful tool for communication,
but children could not acquire its details by figuring out
which ones help in communication; they learn the whole
language, with all its strengths and weaknesses, because
they just can’t help it.
(Pinker 1999: 194)
17.07.2015
University of Tromsø
21
References
Ambridge, B., Rowland, C. F., Theakston A. L. & Tomasello, M. 2006.. Comparing
different accounts of inversion errors in children’s non-subject wh-questions:
‘What experimental data can tell us?’ Journal of Child Language 33, 519-557.
Anderssen, M. 2006. The Acquisition of Compositional Definiteness in Norwegian.
Doctoral Dissertation, University of Tromsø.
Bhatt, R. M. 2004. Indian English: syntax. In B. Kortmann, K. Burridge, R. Mesthrie,
E. W. Schneider & C. Upton (eds.), Handbook of Varieties of English 2:
Morphology and Syntax. Berlin/New York: Mouton de Gruyter. 1016-1030.
Brown, R. 1973. A First Language: The Early Stages. Cambridge, MA: Harvard
University Press. Diessel, H. & Tomasello, M. (2001). The acquisition of finite
complement clauses in English: a usage-based approach to the development of
grammatical constructions. Cognitive Linguistics 12, 97-141.
Lightfoot, D. 2006. How New Languages Emerge. Cambridge, UK: Cambridge
University Press.
Lightfoot, D. & Westergaard, M. 2007. Language Acquisition and Language Change:
Inter-relationships. Language and Linguistics Compass.
MacWhinney, B. 2000. The CHILDES Project: Tools for analyzing talk. 3rd Edition.
Vol. 2: The Database. Mahwah, NJ: Lawrence Erlbaum Associates.
17.07.2015
University of Tromsø
22
Pinker, S. 1999. Word and Rules: The Ingredients of Language. London:
Weidenfeld & Nicolson.
Radford, A. 1992. The acquisition of the morphosyntax of finite verbs in English.
In J. M. Meisel (ed.), The Acquisition of Verb Placement: Func–tional
Categories and V2 Phenomena in Language Acquisition, 23-62. Dordrecht:
Kluwer.
Rizzi, L. 1996. Residual verb second and the wh-criterion. In A. Belletti & L.
Rizzi (eds.), Parameters and Functional Heads. 63-90. Oxford: Oxford
University Press.
Rowland, C. F. & Pine, J. M. 2000. Subject-auxiliary inversion errors and whquestion acquisition: ‘What children do know?’ Journal of Child Language 27,
157-181.
Rowland, C. F. & M. Pine, J. M. 2003. The development of inversion in whquestions: a reply to Van Valin. Journal of Child Language 30, 197-212.
Rowland, C. F., Pine, J. M., Lieven, E. M. V. & Theakston, A. L. 2003.
Determinants of acquisition order in wh-questions: re-evaluating the role of
caregiver speech. Journal of Child Language 30, 609-635.
Vangsnes, Øystein A. 2006. ‘Microparameters for Norwegian wh-grammars.’
Linguistic Variation Yearbook 5, pp. 187-226. Amsterdam and Philadelphia:
John Benjamins.
17.07.2015
University of Tromsø
23
Westergaard, Marit R. 2003. ‘Word Order in Wh-Questions in a North Norwegian
Dialect: Some Evidence from an Acquisition Study.’ Nordic Journal of Linguistics
26.1, 81-109.
Westergaard, Marit and Kristine Bentzen. 2007. ‘The (Non-) Effect of Input
Frequency on the Acquisition of Word Order in Norwegian Embedded Clauses.’ In
Insa Gülzow and Natalia Gagarina (eds.), Frequency Effects in Language
Acquisition: Defining the Limits of Frequency as an Explanatory Concept, [Studies
on Language Acquisition], 271-306. Berlin/New York: Mouton de Gruyter.
Westergaard, Marit. Forthcoming. ‘Microvariation as Diachrony: A View from
Acquisition.’ Accepted for publication in Journal of Comparative Germanic
Linguistics.
Westergaard, Marit. Forthcoming. ‘Acquisition and Change: On the Robustness of the
Triggering Experience for Word Order Cues.’ Accepted for publication in Lingua.
Westergaard, Marit. 2008. ‘Item-based vs. Rule-based Learning: The Acquisition of
Word Order in Wh-Questions in English and Norwegian.’ Ms., University of
Tromsø.
Wexler, Kenneth. 1999. ‘Very early parameter setting and the unique checking
constraint: a new explanation of the optional infinitive stage.’ In Antonella Sorace,
Caroline Heycock and Richard Shillock (eds.), Language Acquisition: Knowledge
Representation and Processing, special issue of Lingua. 23-79. Amsterdam:
Elsevier.
17.07.2015
University of Tromsø
24