CLL lecture: Theoretical issues in SLA research

1
CLL lecture Theoretical issues in SLA research

• 12 October 2004
• Florencia Franceschina

2
Learnability

• Q1 How is it possible for humans (but not
  animals or machines) to learn to understand and
  produce sentences of the language(s) they are
  exposed to?

3
Approaches to Q1

• Interactionist/sociocultural modelse.g.,
  Schumanns (1978) Acculturation/Pidginization
  Hypothesis
• Cognitive modelse.g., Bates and MacWhinneys
  (1985, 1989) Competition Model
• UG-based modelse.g., White (1989, 2003), Flynn
  et al. (1998), Schwartz (1998), Archibald (2000),
  Herschensson (2000), Balcom (2001), Hawkins (2001)

4
Evidence for UG in FLA

• FLA is
• Quick
• Effortless
• Uniform across stages of acquisition
• Robust in terms of noisy/variable input
• It shows equipotentiality across learners

5
Evidence for UG in FLA

• Poverty of the Stimulus (PoS) our linguistic
  knowledge is underdetermined by the input
• Example 1 Structure Dependence Principle
• Example 2 OPC

6
Structure Dependence Principle

• This is my cat Is this my cat?
• Paws has drunk his milk Has Paws drunk his
  milk?
• My cat is the best Cat my is the best?

7
Overt Pronoun Constraint

• (1) John believes that he is intelligent
  ?English ? Japanese
• (2) John believes that _ is intelligent ?
  English ?Japanese
• Montalbetti (1984)
• Kanno (1997)

8
Overt Pronoun Constraint

• Japanese
• (1) Johni believes that hei/j is intelligent
• (2) Johni believes that _i/j is intelligent
• (3) Everyonei believes that theyj are
  intelligent ? OPC
• (4) Everyonei believes that _i/j are intelligent

9
UG, principles and parameters

• The aspects of I-language which are common to
  all of us are known as UG, and the theory of UG
  will state the commonalities that hold across all
  possible languages (often called principles of
  UG) and in what ways individual I-languages may
  diverge from these commonalities (known as
  parameters of variation of UG or just
  parameters).
• Adger (2003 16)

10
Evidence for UG in SLA?

• Target-like outcomes are the norm in FLA
• Vs.
• Non-target-like outcomes are typical in SLA

11
Accounts of divergent outcomes

• Account 1 No access to UG
• Clahsen and Muysken (1986)
• Schachter (1988)
• Fundamental Difference Hypothesis (Bley-Vroman,
  1990)

12
Accounts of divergent outcomes

• Account 2 Full access to UG
• Flynn (1987)
• Epstein, Flynn and Martohardjono (1996)
• Scharwtz and Sprouse (1996)

13
Accounts of divergent outcomes

• Account 3 Partial access to UG
• Smith and Tsimpli (1995)

14
Accounts of divergent outcomes

• More recent proposals
• Account 2 Missing Surface Inflection
  Hypothesis (MSIH, Prevost and White, 2001)
• Account 3 Failed Functional Features
  Hypothesis (FFFH, Hawkins and Chan, 1997)

15
A case study Patty

• Consider the case of Patty (Lardiere, 1998a, b)
• Which of the two recent accounts is supported by
  the data?
• What other information about Pattys L2 knowledge
  would be useful to help us reach a definite
  conclusion?

16
The initial state

• Account 1 Full Transfer/Full Access (FT/FA,
  Schwartz and Sprouse, 1994, 1996)
• Account 2 Minimal Trees (Vainikka and
  Young-Scholten, 1994, 1996)

17
Reading

• Mitchell, R. and F. Myles 1998 Second language
  learning theories. London Arnold.
• Hawkins, R. 2001 Second language syntax. A
  generative introduction. Oxford Blackwell.
  (Chapter 8)
• White, L. 2003 Second language acquisition and
  Universal Grammar. (2nd edition) Cambridge CUP.
  (Chapter 2)

18
From theory to data

• Exercise