|
|
 E-mail this message to a friend
|
|
Title:
|
Inflection As a Marker of Language Impairment in Second Language Learners: A cross-linguistic study
|
|
Author:
|
Madhavi Raman
|
|
Email:
|
click here to access email
|
|
Degree Awarded:
|
English and Foreign Languages University
, Centre for English Studies
|
|
Degree Date:
|
2006
|
|
Linguistic Subfield(s):
|
Applied Linguistics
|
|
Director(s):
|
R. Amritavalli
Michael Ullman
|
|
|
Abstract:
|
|
This thesis investigates the possibility of identifying language impairment in child learners of English as a second language (ESL) through a series of tests in inflectional morphology in English and in their L1 (here Malayalam). Our study assumes an approach to reading problems within the paradigm of Specific Language Impairment (SLI), seeing oral language problems as predating reading difficulties that arise in reading for meaning rather than in the mere decoding of text.
The investigation of SLI in a second language is particularly problematic as performance in the second language could be confounded by the variable of acquisition itself. In Chapter 2, we raise the issue of testing second language populations using diagnostic tests that are norm-referenced for monolingual English-speaking populations and caution against instances of “missed identity” and “mistaken identity”. Very recently, Paradis (2005) has suggested that language impairment in L2 learners should be identified by norm-referencing their performance in the L2 with their own L1 performance. This has been an exploratory study in that direction. We investigate difficulties with inflectional morphology using parallel past tense and plural production and judgment tasks in English and Malayalam. These tasks serve as a probe for the dual-mechanism model of computation and representation of inflections (regular and irregular past tense and plural) in an ESL context.
Our main conclusion from our sample of 17 Malayalam-English bilingual children (6 girls, 11 boys, mean age 8 years, 5 months), is that persistent problems with inflectional morphology in English imply corresponding problems in Malayalam: i.e., problems in the L2 manifest in the L1 as well. This is evident in the performance of the “outliers” (whose score is below the “lower fence” of the group, i.e., 1.5 inter quartile ranges below the 25th percentile). Disturbingly, 7 subjects (2 girls, 5 boys) out of our sample population of 17 are identified as outliers in English and Malayalam in a minimum of one pair of tasks and a maximum of 10 tasks (out of 32). Interestingly, a child from the upper end of this continuum was independently clinically identified (albeit on tests norm-referenced for monolingual English populations) while our work was in progress.
As hypothesized, performance on real regular verbs is superior to that on irregular verbs, thereby confirming the dual-mechanism hypothesis in the context of L2 acquisition. Indeed it seems that real irregulars are encountered so sparsely as to be perceived as novel forms.
The hypothesis that a lack of overregularization of irregular verbs will serve to differentiate the at-risk group from the others does not receive support from our data. Rather, it is the error pattern (such as the use of unmarked forms, the incorrect acceptance of stem forms, the rejection of past and plural marked forms) that serves to distinguish these two populations. If normality and impairment are two ends of a cline, then children prone to errors, outliers, and those at risk for SLI, lie at progressive points along that cline.
Our hypothesis that nominal inflections are easier than verbal inflections is confirmed for English; there is no significant difference between verbs and nouns in Malayalam. This reflects the acquisitional status of the two languages. Correspondingly, performance on Malayalam verbs is better than on English verbs; but there is no L1 advantage in the case of nouns.
|
|
|
|
|
|
Page Updated: 29-Nov-2009
Please report any bad links or misclassified data
LINGUIST Homepage | Read
LINGUIST | Contact us
While the LINGUIST List makes every effort to ensure the linguistic relevance of sites listed
on its pages, it cannot vouch for their contents.
|
|
