SLA Processability Seminar - Program

Friday 1st of October 9.00 am to 5.30 pm
Room BA-02.LR28 which means : Lecture Room 28, Building 2, Bankstown Campus, Bullecourt Avenue, MILPERRA, NSW .

Proposed Program

Morning session. Chair: Howard Nicholas, La Trobe University
Arrival at 9.00 am  for 9.15 am start  
9.00-9.15 Introductions and Welcome
9.15 Manfred Pienemann , University of Paderborn :     Extending  Processability  Theory
9.50 Bruno Di Biase, UWS:     Developing the Topic, or: what happens after SVO?
10.15 Satomi Kawaguchi UWS:    Topic assignment  in Japanese L2 
10.45 Junko Iwasaki; Edith Cowan University : Child acquisition of Japanese L2 - A processability account.   11.15   Coffee Break
11.30  Yuki Itani-Adams UWS:    A comparison of the development of Japanese and English in a bilingual child.
12.00 Lesley Cioccarelli, Wanda Gunawardana, Yuka Sakamoto (Yanyin Zhang) University of Canberra; A processing-based analysis of the acquisition of English L2 morphology
12.25 Aya Harada, UWS: The Acquisition of English L2 Modals by Japanese L1 learners. 
12.50 Rebecca Taylor, UWS: Spanish L2: A processability view"

1.15    Lunch Break

Afternoon session. Chair: Manfred Pienemann, University of Paderborn
2.15    Michael Harrington, University of Queensland: Speed of lexical access as an index of L2 proficiency
2.45    Howard Nicholas, La Trobe University: An issue for processability theory
3.15   Wen  Xiong,  La Trobe University Acquisition criteria in SLA (with reference to learners of Chinese L2)
3.45    Ruying Qi, UWS:  The transition from nominal to pronominal person reference in the early language of a Mandarin-English Bilingual child
4.15   Loan Dao, Australian National University: English verbal and nominal morphology in Vietnamese learners of EFL in Ho Chi Min City.
4.45 Louise  Jansen, Australian National University: On the fitness of the fit of implicational scales: A discussion paper on the use of implicational scaling in the context of Processability Theory
5.15-5.30  Manfred Pienemann: Conclusions & Close ABSTRACTS

Abstracts

Abstract 1. Manfred Pienemann, Paderborn University
Extending Processability Theory: A-structure mapping as a principle of PT

Conceptually, PT is based on the notion of information transfer in sentence generation. This notion is formally modelled within LFG. This is possible because of the complementary relationship between LFG (as theory of grammatical representation) and IPG (or its variants as summarised in Levelt 1989 - as a theory of grammatical processing). The key instrument for formally modelling levels of processability has been the analysis of feature unification.

In this paper I will specify a second set of principles that can contribute to the formal modelling of levels of processability, namely the mapping of a(rgument)-structure onto f(unction)-mapping. The basic rationale of this proposal is quite straight-forward and has first been utilised for a range of structures in Italian and Japanese by Di Biase and Kawaguchi (2002). First of all, one needs to bear in mind that the mapping of a-structure onto f-structure is the driving force behind LFG and that it coincides with IPG at the level of processing. The relationship between a-structure and f-stucture is a many-to-many (i.e. a non-linear) relationship. These circumstances of language generation give rise to the "linearisation problem" (Levelt 1981, 1989). In configurational languages, second language learners initially solve the linarisation problem by a direct (i.e. linear) mapping of a-structure onto f-structure. This results in canonical word order as the initial organising principle in syntax. According to the new proposal, the remainder of the acquisition process can be seen as the cummulative adaptation of the IL to the mapping principles of the L2. For instance, Adv-fronting (previously explained with reference to saliency) can be described as the introduction of a degree of non-linearity in the mapping process where a thematic role that is low on the thematic hierarchy is mapped on the non-core grammatical function FOCUS.

Applying these principles of function mapping to PT, yields the inclusion of a whole range of phenomena in the processability hierarchy. The list of structures that can be integrated utilising this architecture include the following phenomena (also treated in Bresnan 2001): passives, raising, causative, topicalisation, and predicative complements. Some of these phenomena occur very much later in SLA than Cancel Inversion (i.e. past level 6). It is therefore likely that the processability hierarchy can be extended by further levels. The formal implementation of all of these phenomena into the overall framework is the object of future research.

One additional benefit of including function mapping principles in PT is that this adds to the plausibility of the processability hierarchy. The inclusion of function mapping specifies in detail which "simplified procedures" take the place of the S-procedure before the latter develops properly (i.e. at levels 1 to 4).

Abstract 2 by Howard Nicholas; School of Educational Studies, La Trobe University

"An issue for processability theory"

For most work on second language acquisition within the processability framework, a contrast has been assumed between on the one hand routines and formulae and on the other hand, what I have previously referred to as 'combinatorial speech'. It has generally been assumed that issues of second language processing are directed at combinatorial speech while routines and formulae constitute exceptions outside the ambit of the theory so that 'exceptional' items of second language performance can be excluded from analysis. In this paper I reflect on whether it may be possible to re-address some of these issues. In doing this, I draw on the doctoral work of Priscilla Clarke who explored the use of routines and formulae by four Vietnamese background pre-school children acquiring English as a second language. I outline her framework for differentiating formulae and routines of different kinds and speculate about the implications of her framework for models of second language acquisition.

Abstract 3 By Wen Xiong , School of Educational Studies, La Trobe University

Title: The acquisition criteria of Second Language Acquisition

How to define the items or rules become available to the learner of a second language, or how to track that the learner has acquired something is one of central issues with divergent views in decades. Most of the studies in language acquisition, including both L1 and L2 studies, have adopted accuracy or mastery criteria to define it. In this paper, emergence as an acquisition criterion has been discussed in terms of definition and relevance area. Further exploration in applying emergence as a criterion has been made in the area of word learning and development, because in previous studies this criterion has been applied only to the development in syntax and morphology.  Two kinds of emergence, single emergence and systematic emergence are used to label the different time points when the learner's interlanguage develops recognizably more complex features. Based on the perspective, the uses of Chinese auxiliary verb group (NVG) words are examined and their different emergences in distributional patterns will be illustrated.

Abstract 4 by Louise Jansen , School of Language Studies, Australian National University

On the fitness of the fit of implicational scales: A discussion paper on the use of implicational scaling in the context of Processability Theory.

This is a discussion paper in which I would like to focus on two seemingly disparate issues in the relationship between theory (PT) and data.

One is the observation that language-specific evidence for some stages appears to be solid while for others we find exceptions. I would like to propose that there is a qualitative, rather than only quantitative difference between these two types of evidence. I note that the Multidimensional Model recognized only the former, not the latter as evidence for developmental phenomena in SLA . Somewhere along the line this strict criterion has been weakened through the application of a scalability measure, whereby a less than 100% fit is considered equally acceptable as empirical support for the developmental dimension. I propose to challenge the "fitness" of such a fit in a PT context .

The other issue is my observation that many students and researchers appear to fail to appreciate the significance of applying an emergence criterion in a PT context. I think in part this is due to the fact that people intuitively (and understandably) equate the meaning of the word "acquisition" with "mastery". This, when coupled with an insufficient grasp of PT's explanatory goal leads time and again to publications in which the predictions of PT are challenged in the context of a mastery, rather than an emergence criterion. In this paper I would like to provide what I hope is a clearer explanation of the significance of applying an emergence criterion in a PT context.

I will draw the two issues together through an analogy and develop each for discussion.

Abstract 5 Lesley Cioccarelli, Wanda Gunawardana, Yuka Sakamoto (Yanyin Zhang), University of Canberra

A processing-based analysis of the acquisition of English L2 morphology

This study seeks to determine the acquisition stages of four ESL learners at beginner level in a migrant English class in the ACT. The learners come from Laos , Indonesia , Vietnam and Russia . Their L2 speech data was collected using communicative tasks detailed in Pienemann (1998, p. xx). The analysis will focus on both syntactic and morphological structures in the data corpus. Since the study is still in progress, only findings of the L2 developmental state of morphology will be covered in our presentation.

Abstract 6 Junko Iwasaki; Edith Cowen University , WA

Child Acquisition of Japanese L2 - A Processability Account

This is a study of the acquisition of Japanese morpho-syntactic structures through a case study of a seven-year-old Australian boy acquiring Japanese as a naturalistic second language learner. Data were collected through audiotaping approximately 90 minute conversations between the child and other Japanese speakers fortnightly over a period of one year and nine months. Hence the data constitutes a total of 26 oral samples. Specifically the points of emergence for three verbal morpho-syntactic structures, namely verbal inflection, the V -te V structure and the passive/causative structure, were investigated within a framework of Processability Theory (PT) (Pienemann, 1998).

The results of the current study clearly indicate that a developmental sequence of acquisition of verbal morho-syntax does exist in the interlanguage of the naturalistic child learner of JSL, just as have been found with adult learners of JSL. The child acquired the three structures in the order of verbal inflection > the V -te V structure > the passive/causative structure as hypothesised by Di Biase and Kawaguchi (2002), thus following the acquisition order of the L2 processes predicted in PT, i.e., lexical > phrasal > interphrasal. Therefore, the findings of this study provide further support for the applicability of PT to the acquisition of JSL, not only by adult learners, but also by a child learner. It is therefore possible that within a framework of PT, maturational constraints do not impact on the acquisition order of these linguistic features.

The results of the current study have relevance not only to SLA theory, but also with respect to pedagogical development. Firstly, the results of the current study indicate that both the instructed adult learners and the uninstructed child learner of JSL acquired the three verbal morpho-syntactic structures in the same order, confirming that the availability of instruction does not affect the developmental sequence of these structures. This gives an endorsement to the claim by Pienemann (1998) that "teachability is constrained by processability (p. 250)". Secondly, there was some discrepancy in the internal order of the acquisition of verbal affixes found between the results of the current study and those of studies by Di Biase and Kawaguchi (2002). Therefore, this may be in fact an area where JSL teachers may be able to differentiate the points of emergence for verbal affixes through instruction according to the age or needs of learners.

Abstract 7 : Loan Dao, ANU

English verbal and nominal morphology ‘s’ in Vietnamese learners of EFL in Ho Chi Minh City: A variational or developmental feature?

In his Dissertation, Malcolm Johnston (1997) presents his study of naturalistic data from 24 Polish and Vietnamese adults of ESL for English grammatical rules which include verbal morphology such as regular/irregular past marking, third person singular –s marking, and nominal morphology such as the plural –s, the possessive –s. While the results of his study support Processability Theory (PT) in relation to the acquisitional hierarchy applied to English (Pienemann 1998), Johnston states that “It is possible that of the three -s morphemes, the third person singular is in fact a developmental feature, while the plural and genitive are not … In regard to nominal inflectional morphology, … there exist substantial doubts about the developmental status of these markers. There is some evidence that regular plural marking may be a variational feature. Due to lack of data the status of the genitive is unclear, but, equally, it may not be developmental in nature.” (p 269).

This paper, which forms part of my PhD, attempts to shed more light onto the above question. It studies oral production records of 36 formal Vietnamese learners of English as a Foreign Language, from year 7 to year 12, at 3 secondary and 3 high schools in Ho Chi Minh City. Each subject was given a series of 12 elicitation tasks (3 for lexical plural -s, 3 for phrasal plural -s, 3 for possessive -‘s, and 3 for third person singular -s), of 30 minutes duration altogether.

Implicational scaling and quantitative analysis are used in this study. The analyses are based on types (lexical variation), not tokens, and are performed on three emergence-based acquisition criteria, each requiring at least five obligatory contexts: (1) following Pienemann (1998), two type applications (p 147); (2) three type applications; and (3) four type applications. Implicational scaling using Guttman procedure is then applied to determine the Coefficient of Scalability (Cscal), “the figure that indicates whether a given set of features are truly scalable” (Hatch & Lazaraton, 1991, p 212).

Preliminary results lend support towards Johnston’s doubt “that of the three -s morphemes, the third person singular is in fact a developmental feature, while the plural and genitive are not.”