Teacher- and learner-led discourse as tools for L2 grammatical development in task-based Spanish instruction

1
Teacher- and learner-led discourse as tools for
L2 grammatical development in task-based Spanish
instruction

• Paul D. Toth
• University of Wisconsin-Madison
• ptoth_at_wisc.edu
• 2007 TLBT Conference, University of Hawaii

2
Instruction L2 grammatical development

• Provision of comprehensible L2 input via
• Modifications to instructional speech or
  materials
• Opportunities for learner negotiation
• Attention directed to L2 form-meaning
  relationships via
• Salience in instructional speech or materials
• Explicit, metalinguistic information about the L2
• Feedback on learner performance
• Opportunities for L2 output (Swain, 1985, 1995,
  2000)
• Learners pushed to encode meaning in
  morphosyntax
• Test hypotheses about L2 form-meaning
  relationships
• Notice gaps in L2 grammar
• Conceptualize L2 grammar through metatalk

3
Task-Based Instruction

• Requires learners to use language, with
  emphasis on meaning, to attain an objective
  (Bygate, Skehan, Swain, 2001, p. 11)
• Focused tasks target the purposeful use of
  specific L2 structures to express meaning (Ellis,
  2003, p. 16)
• Descriptions adjective agreement
• Narration past tense and aspect marking
• Requests of others subjunctive mood
• Explaining procedures impersonal passive
• Narrating spontaneous events inchoative verbs

4
Learner-Led Discourse

• Strengths
• More like real world communication (Nunan, 1987)
• Participatory structure more suitable for
  negotiation, especially during information gap
  tasks (Pica, 1987 Pica et al., 1993)
• More discourse turns per learner more
  opportunities for negotiation (Lee, 2000 Long
  Porter, 1985)
• Greater linguistic autonomy and self-regulation
  (van Lier, 1996)
• Learners assist each other during task
  performance (Donato, 1994 Swain, 1998, 2000
  Swain Lapkin, 1995)

5
Learner-Led Discourse

• Weaknesses
• Learners often produce minimal utterances
  (Seedhouse, 1999)
• Learners are poor L2 models for each other
  (Prabhu, 1987)
• Learners prefer to focus on lexical rather than
  morphosyntactic L2 issues when negotiating
  (Buckwalter, 2001 Morris, 2002 Williams, 1999)
• Suggested Remedies
• Make target forms useful or essential to task
  performance (Loschky Bley-Vroman, 1993 Fotos,
  2002)
• Precede tasks with pre-task warm-up to orient
  learners to necessary language follow tasks with
  post-task activity to lend accountability to
  learner performance (Skehan, 1996, 1998)

6
Teacher-Led Discourse

• Strengths
• Teacher input and support provides expert
  scaffolding for task performance (Adair-Hauck
  Donato, 1994 Antón, 1999 McCormick Donato,
  2000).
• Teacher feedback has been shown to benefit
  non-turn-taking listeners as well as active
  discourse participants (Ohta, 2000, 2001).
• Weaknesses
• Far fewer speaking turns per learner (Lee, 2000)
• IRF sequences (Initiate, Response, Feedback)
  often limit learner utterances and prevent
  development of broader interactional competence
  (Brooks, 1993 Hall, 1995, 2004 Leemann-Guthrie,
  1984 Mehan, 1979 Nunan, 1990)

7
Teacher-Led Discourse

• Suggested Remedies
• Design whole-class activities as collaborative
  communication tasks, rather than mechanical
  grammar drills (DeKeyser, 1998 Wong VanPatten,
  2003)
• Teachers should build their turns upon topical
  content of learner utterances, as follow up
  moves (Johnson, 1995 Toth, 2004 Wells, 1998)
• Solicit multiple learner responses to teacher
  questions before moving onto another question
  (Toth)

8
Motivation for comparing TLD LLD

• Importance of interlocutors and interaction in L2
  acquisition
• Little previous research
• Pica (1987), Doughty Pica (1986) More
  negotiation for LLD in information exchange
  tasks similar amounts of negotiation in more
  open-ended collaborative discussion
• Fotos (1993, 1994) TLD and LLD classes perform
  nearly equally, with TLD group noticing one of
  target structures more frequently
• Calls for further research
• Pica (1994) Benefits of negotiated interaction
  in learner dyads need to be supported by
  quantitative assessments of learning outcomes
• DeKeyser (2003), Doughty (2003), Pica (2005)
  Quantitative studies of learning outcomes through
  LLD negotiation need to be conducted in
  ecologically-valid classroom contexts, rather
  than only in laboratory settings.

9
Spanish se
Se used to derive intransitive syntax from a
transitive verb (Dobrobie-Sorin, 1998 Montrul,
2004 Raposo Uriagereka, 1996)
X
a. Ellos prepararon la comida. AGENT
PATIENT They prepared the food.

• ? Ellos se prepararon.
• AGENT
• They prepared themselves / each other.

X
b. Ellos prepararon la comida. AGENT
PATIENT They prepared the food.

• ? Se preparó la comida.
• PATIENT
• The food was prepared / One prepared
  food.

anticausative se
X
c. Ellos cocinaron la comida. AGENT
PATIENT They cooked the food.
? Se cocinó la comida.
PATIENT The food Ø cooked / was
cooked / One cooked food.
10
Research Questions

• Question 1 Will LLD provide an advantage in
  grammaticality judgments for Spanish
  anticausative se when compared to TLD?
• Question 2 Will LLD provide an advantage over
  TLD in performance with anticausative se on
  sentence-level picture descriptions?
• Question 3 Will excerpts of classroom
  interactions reveal differences in the way
  learners in each group attend to the
  form-meaning relationships associated with
  anticausative se and use the target form for
  output?

11
Method Participants

• 6 intact classes of 2nd semester beginning L2
  Spanish in two large, public American
  universities with identical Spanish curriculums.
  Each group comprised of two classes.
• Teacher-Led Discourse (TLD) n 28
• Learner-Led Discourse (LLD) n 25
• Control Group (C) n 25
• Native Speaker comparison group n 30

12
Method Instruction
Sequence of lesson topics for treatment groups
anticausative se
13
Method Instruction

• Standard 50-minute daily lesson
• Whole-class warm-up activity, reminiscent of
  previous days tasks (5 mins.)
• Explicit grammar explanation for current days
  topic (5 mins.)
• LLD 2 passes through pre-task, task, post-task
  sequence, with most tasks designed as two-way
  information gaps (40 mins.)
• TLD 4-6 tasks mirroring those of the LLD group,
  implemented as whole-class, collaborative
  interaction. (40 mins.)

14
Method Instruction

• Spotting differences activity
• LLD implemented as a two-way information gap in
  small groups
• TLD implemented as whole-class collaborative
  discourse

15
Method Assessment

• Experimental Design
• Pre-test,
• Immediate posttest
• Delayed posttest (24 days after instruction)
• Two test versions, piloted on two native
  speakers, and randomly assigned to learners. Then
  rotated over the three test administrations
• Grammaticality judgment (GJ) task
• Picture description task
• Lesson on se of unplanned occurrences recorded
  and transcribed in each group

16
Method GJ Task
Sample items from the grammaticality judgment task
17
Method Picture Description Task
Sample item from the picture description task
18
Results Picture Description Task
increase 0.02
increase 0.31
increase 0.46
19
Results Picture Description Task

• NS mean 0.48

20
Results GJ Task
increase 0.09
increase 0.36
increase 1.07
21
Results GJ Task

• NS mean 2.24

22
Results Transcripts
1. LLD Information gap activity
?
?
?
23
Results Transcripts
2. LLD information gap activity
?
?
?
24
Results Transcripts
3. TLD whole-class collaborative discourse
IRF
?
25
Results Transcripts
3. TLD whole-class collaborative discourse
(cont.)
?
?
26
Results Transcripts
4. TLD whole-class collaborative discourse
?
?
?
27
Results Transcripts
5. TLD whole-class collaborative discourse
?
?
?
?
?
?
28
Results Transcripts
5. TLD whole-class collaborative discourse
(cont.)
?
?
29
Results Transcripts
6. LLD information gap activity
?
?
?
?
30
Results Transcripts
7. LLD Information gap activity
?
?
?
31
Results Transcripts
7. LLD Information gap activity (cont.)
?
?
?
?
32
Results Transcripts
7. LLD Information gap activity (cont.)
?
?
?
33
Discussion

• Under the best circumstances, learners attention
  to target forms may be limited in LLD
• Developmental needs that focus attention other
  areas of L2 morphosyntax
• Widely-observed tendency to focus on lexis rather
  than morphosyntax, and to prioritize getting
  meaning across over formal accuracy
• Preference for self-correction rather than
  other-correction (Buckwalter, 2001 Seedhouse,
  2004)
• Participatory roles that, while increasing
  turn-taking, do not authorize individuals to
  assist in procedures for making output

34
Discussion

• In TLD, attention to target forms may be more
  consistent
• Provision of accurate input models and cues for
  using target form
• Feedback centers on target form
• Cumulative benefit of feedback to others, if
  relevance is maintained across discourse turns
• Participatory roles allow teacher-expert to
  directly assist learners in formulating
  utterances
• Following Ohta (2001), potential for
  collaborative listeners to indirectly realize
  output benefits if they are cognitively engaged.

35
Discussion

• Teachers as providers of procedural assistance in
  output processing
• Assistance with linguistic task of utterance
  formulation and morphosyntactic assembly, rather
  than conceptual or analytical scaffolding
  (Wood, Bruner, Ross, 1976).
• Proactive, simultaneous assistance to learner
  rather than reactive and subsequent feedback, as
  in clarification requests, confirmation checks,
  or recasts. (Long, 1981, 1996).
• Assistance utilizing L2 morphosyntax that is more
  complex than the learners extant interlanguage,
  OR
• Useable metalinguistic information that can guide
  learners toward incorporating new forms into
  their L2 speech.

36
Discussion

• Hypothesized benefits of procedural assistance
• Some current models of language processing hold
  that hierarchical morphosyntactic relationships
  are computed on-line, during comprehension or
  production (Harrington, 2001 Juffs, 2004
  Pritchett, 1992)
• Parsing, or processing, L2 form-meaning
  relationships may be key to a transition theory
  that explains how the L2 linguistic properties
  become incorporated into interlanguage grammars.
  (Carroll, 2001 Gregg, 2001 Pienemann, 1999)
• Procedural assistance may allow learners to
  implement, or proceduralize, the declarative L2
  metalinguistic knowledge they have, increasing
  the complexity of L2 utterances that they can
  process
• If learners are able to assemble more complex
  utterances with the assistance of an expert, this
  may facilitate incorporation of these structures
  into the implicit L2 grammatical system.

37
References
38
References
39
References