Reading

1
Reading

• For the lecture on Thursday March 3rd (wk 4)
• Meisel, J.M. (to appear) The bilingual child.
  In. T.K. Bhatia W.C. Ritchie (eds.) The
  Handbook of Bilingualism. Oxford Blackwell.
• Lakshmanan, U. (1995) Child second language
  acquisition of syntax. Studies in Second Language
  Acquisition 17 301-329
• Both are available in my pigeon hole please
  return immediately after copying

2
Negation and Verb Movement

• Linguistic Research on Second Language
  Acquisition Development Lecture 3

3
Todays lecture

• Brief look at L2 negation data
• Syntactic analysis of negation (and verb
  movement)
• How properties of negation can tell us something
  about the acquisition of verb movement
• Verb movement studies

4
Types of negation

• Sentential negation
• Constituent negation
• Anaphoric negation

5
Which is which?

• Does Bob eat cakes? No, he eats pies.
• Bob doesnt eat cakes.
• Bob doesnt eat cakes. He eats pies.

6
Which is which?

• Does Bob eat cakes? No, he eats pies.Anaphoric
  negation
• Bob doesnt eat cakes.Sentential negation
• Bob doesnt eat cakes. He eats pies.
• Constituent negation

7
Early work on L2 negation

• Cancino, Rosansky Schumann (1978)
• L1 Spanish, L2 English
• 6 subjects 2 children, 2 teenagers, 2 adults
• Longitudinal study
• Spontaneous data

8
Early work on L2 negation
9
A change in approach

• 1970/80s descriptive generalisations
• Late 1980s / 1990s
• Analysis of L2 data using linguistic theory
• How can the tools of linguistic theory help us to
  understand the L2 acquisition process?
• (How can L2 acquisition help us to understand
  language (i.e. contribute to linguistic theory)?)

10
Recent work on L2 negation

• Negation as it relates to other properties of the
  interlanguage (IL) grammar
• Verb movement
• Whats that?

11
Verb movement

• French
• Cécile boiti souvent ti du café
• Cécile souvent boit du café
• English
• Melinda often drinks coffee
• Melinda drinksi often ti coffee

12
Analysis of negation

• What is the position of elements such as not in
  English and (ne) pas in French in the syntactic
  structure of a sentence?
• Why does negation appear to the right of thematic
  verbs in French and to the left of these verbs in
  English?

13
NegP

• Single category in each language, Neg, which
  appears in same place in structure
• Neg projects to Neg, which projects to NegP
• NegP is (optionally) selected by I(nflection)
• not/nt and pas are in Spec, NP
• ne is in Neg (i.e. the head of NegP)

14
French vs. English

• French
• Cécile I (ne) boiti NegP pas VP ti du café
• Cécile NegP (ne) pas VP boit du café
• English
• Melinda I does NegP not VP drink coffee
• Melinda I drinksi NegP not VP ti coffee

15
French vs. English

• Strength of inflection
• French strong
• Strong I(nfl) attracts verb verb-raising
• English weak
• Weak I(nfl) allows verb to remain in VP

16
L2A of verb movement
17
L2A of verb movement
18
L2A of verb movement

• What does the L2er have to acquire?
• ? L1 English, L2 French

19
L2A of verb movement
20
Hawkins, Towell Bazergui (1993)

• Subjects
• Intermediate L2ers (n75)
• Advanced L2ers (n29)
• Task
• Grammaticality judgement
• Sentences with/without verb raising over
  negation, adverbs, floating subject quantifier

21
Hawkins et al (1993)

• Results
• Both groups correctly accepted sentences with
  thematic verbs raised over negation and rejected
  thematic verbs in situ
• 40 of intermediate subjects incorrectly accepted
  both raised and non-raised orders with
  manner/frequency adverbs

22
Hawkins et al (1993)

• Conclusion Difference between verb raising over
  negation (acquired earlier) and verb raising over
  adverbs.

23
L2A of verb movement

• What does the L2er have to acquire?
• ? L1 English, L2 French
• ? L1 French, L2 English

24
L2A of verb movement
25
? L1 French, L2 English

• Series of studies by Lydia White and colleagues
• Additional factor role of explicit and negative
  evidence
• Whats that?

26
Explicit/Negative evidence

• Explicit evidence/instruction
• explanation of grammatical rules
• Negative evidence
• information about what is ungrammatical

27
White (1991) Background

• L1 acquisition
• interaction of Universal Grammar with positive
  evidence (utterances in the input)
• incorrect hypotheses are disconfirmed by positive
  evidence
• negative evidence plays minor role

28
White (1991) Background

• L2 acquisition
• interaction of UG with positive evidence
• incorrect hypotheses (resulting from L1 transfer)
  cannot always be disconfirmed by positive
  evidence
• negative evidence may be necessary

29
White (1991) Learnability

• French L2er of English must discover that
• the L2 allows SAV order
• the L2 does not allow SVAO order

30
White (1991) Learnability

• Whats the evidence in the input?
• SAV utterances show possibility of this order
• no positive evidence that SVAO is impossible
• non-occurring
• other input shows considerable freedom wrt adverb
  placement
• ? negative evidence might be necessary

31
White (1991) Hypotheses

• Transfer of L1 parameter settings ? French
  learners of English will assume that SVAdvO is
  possible English word order
• Specific teaching (incl. negative evidence) will
  allow L2ers to master possibility of SAdvV and
  impossibility of SVAdvO

32
White (1991) Subjects

• Francophone learners of English as second
  language in Québec, Canada
• 11/12 year olds
• intensive ESL programme, emphasis on
  communicative teaching
• very little previous knowledge of English
• very little contact with English outside classroom

33
White (1991) Subjects

• Adverb group
• given explicit instruction, including negative
  evidence, about adverb placement
• n82
• Question group
• given explicit instruction, including negative
  evidence, about question formation
• n56

34
White (1991) Treatment

• Started after c. 3 months in ESL programme
• Pre-test determine knowledge of adverb placement
  using three different tasks
• Treatment intensive teaching on adverbs or
  questions for 2 weeks
• Post-test 1 immediately after treatment
• Post-test 2 c. 5 weeks after treatment
• Follow up 1 year after treatment

35
White (1991) Treatment

• Adverb group Week 1
• 5 hours intensive work on adverb placement
• adverbs of frequency and manner
• emphasis on meaning (in context) and form
• illustration of different positions for adverbs
• sentences with main verbs
• teachers pointed out and corrected errors

36
White (1991) Treatment

• Adverb group Week 2
• 2 hours of follow-up activities
• Question group
• similar activities but with focus on question
  formation

37
White (1991) Tests

• Same three tests used on all four testing
  occasions
• Used manner/frequency adverbs often, always,
  sometimes, usually, quickly, slowly, quietly,
  carefully
• Tested different adverb positions with transitive
  verbs AdvSVO, SAdvVO, SVAdvO, SVOAdv
• Also tested intransitive verbs with PPs SVAdvPP,
  SAdvVPP

38
White (1991) Tests

• ? Grammaticality judgement task
• sentences presented in continuous story
• subjects asked to correct any sentences they
  thought were incorrect
• 16 sentences with adverbs (grammatical and
  ungrammatical) plus 17 fillers

39
White (1991) Tests

• ? Grammaticality judgement task

Billy eats always his peas
40
White (1991) Tests

• ? Preference task
• written
• subjects read pair of sentences and circle option
• 28 pairs with adverb positions, 4 fillers

• Susan often plays the piano
• Susan plays often the piano
• only a is right / only b is right / both right /
  both wrong / dont know

41
White (1991) Tests

• ? Manipulation task
• subject asked to form as many sentences as
  possible with sets of word cards
• 4 sets of cards, 3 with SVO and 1 with SVPP
• subjects tested individually

42
White (1991) Tests

• ? Manipulation task

Susan
piano
plays
the
often
43
White (1991) Tests
Susan
often
plays
the
piano
44
White (1991) Tests
often
Susan
plays
the
piano
45
White (1991) Tests
Susan
plays
often
the
piano
46
White (1991) Results

• ? Grammaticality judgement task
• SVAO error score no. of SVAO sentences
  incorrectly left unchanged (max. 4) plus any
  other sentences incorrectly changed to SVAO
• Max score 16, but most are between 0-10
• Failure to alter less than 3/7 incorrect
  distractor sentences ? exclusion
• Adverb group 37 subjects (out of 82)
• Question group 38 subjects (out of 56)

47
White (1991) Results

• Figure 1. Judgement task SVAO error scores

48
White (1991) Results

• ? Grammaticality judgement task
• SAV score no. of SAV sentences correctly left
  unchanged) plus any other sentences correctly
  changed to SAV
• Max score 16

49
White (1991) Results

• Figure 2. Judgement task SAV scores

50
White (1991) Results

• ? Grammaticality judgement task
• clear differences between adverb and question
  groups
• differences for both ungrammatical (SVAO) and
  grammatical (SAVO) orders
• explicit evidence is more effective than
  naturalistic positive evidence alone

51
White (1991) Results

• ? Preference task
• SVAO error score no. of responses showing
  incorrect preference for SVAO order or allowing
  both SVAO and ASVO/SAVO/AVOA (i.e. a grammatical
  order in English)

52
White (1991) Results

• Figure 3. Preference task SVAO error scores

53
White (1991) Results

• ? Preference task
• clear differences between adverb and question
  groups
• again, explicit evidence appears to more
  effective than naturalistic positive evidence
  alone

54
White (1991) Results

• ? Manipulation task
• SVAX score no. of sentences produced with this
  order (max 4)
• SAV score no. of sentences produced with this
  order (max 4)

55
White (1991) Results

• Figure 4. Manipulation task SVAX scores

56
White (1991) Results

• Figure 5. Manipulation task SAV scores

57
White (1991) Results

• ? Manipulation task
• similar trends to other tasks
• lasting decrease in no. of errors for adverb
  group
• plus increase in production of targetlike SAV
  orders

58
White (1991) Summary

• Support for transfer of L1 parameter settings
  (hypothesis 1) production and preference for
  SVAO in pre-test
• BUT note that targetlike SAV is never totally
  rejected

59
White (1991) Summary

• Support for effectiveness of negative evidence
  adverb group consistently perform better than
  question group
• positive evidence alone didnt help the question
  group to acquire the properties associated with
  adverb placement

60
White (1991) Summary

• Adverb group did not forget what they had been
  taught
• Evidence for restructuring of underlying grammar?
• Knowledge could be conscious rather than
  unconscious

61
Conscious vs. unconscious

• Why does this matter?

62
Conscious vs. unconscious

• Competence vs. performance
• Acquisition vs. Learning (Krashen 1982)
• Learned Linguistic Knowledge vs. Learned
  Linguistic Behaviour (Schwartz 1993)
• Interface vs. non-interface position
• Crucial if we want to test for the availability
  of UG in L2A, we need to access competence

63
Adverb placement Obj vs. PP

• ? Preference task SVO vs. SVPP
• SVO John quickly drinks his beer
• vs. John drinks quickly his beer
• SVPP (manner adverbs)
• John quickly walks to school
• vs. John walks quickly to school
• SVPP (frequency adverbs)
• John often walks to school
• vs. ?John walks often to school

64
Adverb placement Obj vs. PP

• ? Preference task SVO vs. SVPP
• Are L2ers able to generalise from SVO to SVPP?
• Are they sensitive to structural differences
  between these two types of sentences without
  being taught this?

65
Back to Hawkins (2001)

• Exercise 7 (p. 120-1)

66
Schwartz Gubala-Ryzack (1992)

• critique of White (1991)
• negative evidence may have had an effect on
  L2ers behaviour but not on underlying grammar,
  i.e. interlanguage grammar was not restructured

67
Schwartz Gubala-Ryzack (1992)

• Why negative evidence didnt have an effect
• SVAdvPP is also incorrectly ruled out
• evidence of overgeneralisation of surface rule
  (Nothing can come between verb and object) rather
  than restructuring of IL grammar, i.e. change in
  metalinguistic/conscious/Learned Linguistics
  knowledge

68
Schwartz Gubala-Ryzack (1992)

• Why negative evidence didnt have an effect
• L2ers in adverb group IL grammar is not a
  natural language grammar
• only wild grammar could rule out SVAdvPP and
  still allow SVAdvO
• changes in behaviour were short-term only

69
White (1991) Follow-up

• Same three tests carried out with adverb group
  one year later
• Comparable uninstructed group (but not original
  question group)
• Did subjects retain what they had learned?

70
White (1991) Follow-up

• Figure 6. Follow-up study. Judgement task SVAO
  error scores

71
White (1991) Follow-up

• Results children reverted to same level as
  pre-test, i.e. prior to treatment
• Similar results for other tasks
• Conclusion instruction was very effective in
  short-term, but didnt have lasting effect
• White (1992) reply to Schwartz Gubala Ryzak
  (1992)

72
More studies on L2 verb movement and negation

• Beck (1998) Local impairment hypothesis (verb
  movement in L2 German)
• Eubank (1996) Valueless features (negation in L2
  English)
• Prévost White (2000) Missing surface
  inflection hypothesis

73
And finally

• Hawkins (2001) Exercise 4
• Evaluating research design