Review Jurnal

On the Role of L1 Markedness and L2 Input Robustness in Determining Potentially Fossilizable Language Forms in Iranian EFL Learners' Writing

Musa Nushi a *

a Shahid Beheshti University,Iran

 Nushi, M. (2016).

Musa Nushi is an assistant professor of TEFL at Shahid Behehsti University. His research interests lie mainly in the interface of second language acquisition and second language instruction, with particular emphasis on the role of corrective feedback in the L2 development.

. Journal of Language and Linguistic Studies, 12(2), 66-86.

 Abstract

Han’s (2009, 2013) selective fossilization hypothesis (SFH) claims that L1 markedness and L2 input robustness determine the fossilizability (and learnability) of an L2 feature. To test the validity of the model, a pseudo-longitudinal study was designed in which the errors in the argumentative essays of 52 Iranian EFL learners were identified and categorized based on a researcher-developed error taxonomy. Next, the learners were provided with implicit and explicit corrective feedback on those errors to see if there existed any errors that would persist despite learners’ motivation and the pedagogical intervention to eliminate them from their writing. ANOVA results revealed that the errors in the pronoun, word order, passive voice and possessive categories persisted in the written output of the participants. A sub-classification of errors in the pronoun and possessive categories showed that deletion and redundant addition of subject pronouns, lack of agreement between pronouns and their antecedents, vague or ambiguous pronoun references in the pronoun category, and a wrong use of the apostrophe (‘) or apostrophe + s (‘s) with regular plurals in the possessive category proved most resistant to correction, pointing to their tendency towards becoming fossilized. That tendency, however, could not be accounted for by the SFH.

© 2016 JLLS and the Authors - Published by JLLS.

Keywords: Fossilization; errors; corrective feedback; markedness; input; writing.

1.   Introduction

Fossilization, introduced by Selinker (1972), is said to be a distinctive characteristic of secondlanguage (L2) learning (Han, 2004; Selinker, 1996; Tarone, 1994). Most pervasive among adult L2 learners (Han & Odlin, 2006; Kellerman, 1995; E. Lee, 2009; Schachter, 1996), fossilization has been characterized as premature cessation of learning, even though the learner possesses a strong motivation to learn, is exposed to frequent and rich input, and has numerous opportunities for practice (Selinker, 1972). Selinker introduced the concept based on his observation that it was rare for learners of an L2 to succeed in obtaining full native-speaker competence. He put the success rate at “a mere 5%” (p. 212), although later estimates put the figure higher (e.g., Birdsong, 2004; Montrul & Slabakova, 2003; White & Genesee, 1996). Ever since Selinker’s introduction of the term, a considerable amount of second language acquisition (SLA) research has been devoted to fossilization of linguistic forms both within and across learners in an attempt to figure out why the outcome of L2 learning for many is incomplete and fragmentary when compared with that of first language acquisition (see Han, 2004; E. Lee, 2009; Long, 2003 for a review).

Despite extensive literature on fossilization, spanning over four decades, researchers in the field of SLA (e.g., Acton, 1984; Birdsong, 1992, 2004, 2006; Han, 2004, 2009, 2011, 2013; Han & Odlin, 2006; Kellerman, 1995; Long, 2003; Selinker, 1972, 1996) have repeatedly raised two problems regarding the construct.

 First, fossilization lacks a unified definition. In the absence of a comprehensive conceptual definition of fossilization, the term has been (mis)used simply as “a protean, catch-all” term (Birdsong, 2004, p. 87) to describe "any lack of progress in L2 learning regardless of its nature" (Shin, 2009, p. 60). It has also meant that the choice of the linguistic targets in many of the studies on fossilization has been quite arbitrary, that is, they were chosen because they were thought to be fossilized or fossilizable, often without proper explanation as to how fossilization or their fossilizability was determined.

The second problem relates to the fact that fossilization has not been adequately investigated empirically. Han and Odlin (2006, p. 5) point out that “not only has there been a continuous paucity of [empirical] evidence, but the existing [empirical] evidence is also suspect, due to various conceptual and methodological shortcomings.” Moreover, there has been little investigation by SLA theorists of how to prevent or counter fossilization, and little explanation as to why some adult L2 learners manage to overcome certain areas of stability in interlanguage (IL) and reach a high level of proficiency.

  

The SFH presents a model of fossilization with intriguing possibilities for instructed SLA theory and practice. From a theoretical perspective, the SFH offers a scientific means to move the issue of fossilization in L2 learning “beyond its hitherto primarily argumentative basis” and towards “a more tangible and precise understanding” (Han, 2009, pp. 157-158).

Han (2009), however, states that the SFH has not been described empirically and calls for an inclusive research into the observed persistent non-target-like behaviors, both within and across learners, in order to develop the SFH into a comprehensive analytic model of fossilization. One possible avenue to validate and refine the SFH is to see whether it can account for the typical and residual errors in the Persian-speaking EFL learners’ written output.

2.    Method

The main purpose of this study, which is a follow-up to an initial study was to find out whether the SFH could account for the persistent erroneous linguistic features in Iranian EFL learners’ written output. To that end, it utilized corrective feedback approach (Kellerman, 1989) to the study of fossilization. The corrective feedback approach examines L2 learners’ reaction to feedback and assumes that errors immune to pedagogic intervention should be made the linguistic focus of investigations of fossilization (Thep-Ackrapong, 1990, as cited in Han, 2004). The study also adopted a pseudo-longitudinal method of data collection (Kellerman, 1989). In this method, learners of different proficiency levels are used as informants to construct a diachronic view of the IL structures under scrutiny.

Sample / Participants

Fifty two male and female EFL learners, selected out of a pool of 95 through purposive sampling (Macaro & Masterman, 2006), took part in the study. The participants were all undergraduate and graduate non-English majors in several universities in Tehran and ranged from 18 to 38 in age (mean age 27.5). They were enrolled in EFL classes in five English language institutes. Their classes were held twice a week with each session lasting approximately 105 minutes.

Instrument

 English language proficiency test A teacher-made English proficiency test was developed and then validated to determine the subjects’ level of proficiency. It comprised of 100 multiple-choice items which measured the knowledge of English grammar and structures, vocabulary and reading.

Writing tasks

As part of their coursework, the learners were assigned three topics representing the argumentative rhetorical mode and were asked to write a 300-word essay on each topic.

Rating scale

The learners’ essays were evaluated using a 6-point holistic scoring rubric patented after the Test of Written English (TWE) scoring guide, the written component of TOEFL. The scores were the basis for classifying the learners into three levels of writing proficiency.

Motivation questionnaire

To measure the learners’ motivational intensity towards language learning, Takahashi’s (2005) questionnaire was used with one minor modification. The scores obtained from the questionnaire were checked for reliability.

Data collection procedures

The study comprised two stages. The first stage consisted of the identification and treatment of errors in the argumentative essays written by 52 Iranian EFL learners. The purpose of the first stage was to obtain a picture of common errors in the learners’ argumentative writing and to provide pedagogical focus on those errors (in the form of implicit and explicit corrective feedback) in an attempt to prevent their reappearance on future writing tasks.

Error identification and treatment

The data for this stage were collected from 52 adult EFL learners over a period of three months. The learners were taught by the researcher himself. He started off by administering the 100-item test to categorize the learners into homogenous groups based on their English proficiency. Based on their scores, the students were divided into the pre-intermediate, intermediate and advanced proficiency levels. To see whether there were any significant differences across the levels, a one-way ANOVA was conducted. The results revealed a significant difference across the three levels (F(2, 49) = 251.211, p < .05). Post-hoc comparisons showed that each level was significantly different from the other two groups. 

Next, the learners were assigned one topic in the argumentative mode and were asked to write a 300-word essay about it. Prior to writing, however, the teacher provided the students with a sample argumentative essay and taught them the key principles of writing in that mode. After collecting the first draft of their writings, the teacher and another EFL colleague with 12 years of teaching experience independently assessed the quality of each composition using the 6-point holistic scoring rubric and grouped them into pre-intermediate, intermediate and advanced writing proficiency levels. To identify and classify the errors in the learners' essays, the researcher reviewed several error taxonomies in the literature (e.g., Chandler, 2003; N. Lee, 1990; Richards & Sampson, 1974) but could not come up with a model that accommodated all the errors. Each taxonomy seemed to have been designed for a specific purpose and population and was either too broad or too narrow in its classification of linguistic errors.

3.   Results

To answer the first research question, that is, if there existed some errors that proved impervious to either implicit or explicit corrective feedback in both Stage 1 and 2 (a harbinger of fossilization), the researcher sought to find those errors that received corrective feedback in Stage 1 but kept reappearing in Stage 2. The assumption was that if there were some errors which, despite being treated with corrective feedback in the previous stage, continued to show up in the next stage and required explicit corrective feedback to be removed, then there could have been cases where fossilization might have been setting in. A repeated measures ANOVA with one within group factor, corrective feedback type, with three levels (1 = no feedback, 2 = implicit feedback and 3 = explicit feedback), and one between group factor, writing proficiency, with three levels was conducted (since there were few learners making the negation, the verb to be and the comparative errors, ANOVA analyses could not be run for these categories and hence they were not counted as data). Table 1 revealed there was a reduction in the means of all error categories when the L2 learners were provided with either implicit or explicit feedback as opposed to when there was no feedback. The main effect of feedback was significant for all error types.

4.   Discussion

There is now ample research showing that the L1 preprograms L2 learners and that L1 knowledge interacts with the available L2 input to influence adult L2 learning (e.g., Gass & Selinker, 1992; Kellerman, 1984; Kellerman & Sharwood-Smith, 1986; Sorace, 1993). However, attributing the selectivity of the (non-)learning of L2 features solely to L1 markedness and L2 input robustness seems a little too simplistic. First, such a contention overlooks the complexity surrounding the concept of markedness (see Battistella, 1990; Eckman, 1977; Hume, 2008; Hyltenstam, 1987 for a discussion). Han (2009) conceptualizes markedness as a property of languages determined by the frequency of a linguistic feature in the learners’ L1 and its FMF variability. However, as Hume (2008) has rightly observed, the markedness value of a given form is dependent upon the expectations that a language user has about the linguistic form in question and those expectations are guided by the experiences of the learner with that form. It means that the markedness value of a structure for a lawyer is different from the one for a musician; thus, it would be misleading to consider the markedness of a form in a language as a quantitative static attribute, the way Han (2009) does. Similarly, Han’s operational definitions of the frequency and variability of a language feature are vague and hard to objectify as these variables seem highly dependent on the individual experiences of language learners. In addition, she does not make it clear how much frequency or variability makes a feature frequent or variable. In other words,

5.   Conclusions

The SFH ignores ample individual differences such as the learners’ reasons for learning (or motivation) and satisfaction of communication needs. These factors, along with a host of others, many of which may not yet be known, can interact to codetermine the tendency of a given linguistic feature to become fossilized. Motivation in language learning is so important that Gardner (1985), one of the most prominent researchers in the area of motivation in L2 learning, identified it as the single most influential factor in learning a new language. Motivated L2 learners generally seek out more exposure to and practice opportunities in the TL to further their language learning career (Cheng & Dörnyei, 2007; Scarcella & Oxford, 1992). Besser (2002, as cited in Han, 2009) has also said that learners’ satisfaction of communication needs can affect the fossilization process. Skehan (1998, p. 61) stated that “if communicative effectiveness is achieved, the erroneous exemplar may survive and stabilize, and becomes a syntactic fossil.” Similarly, Ellis (2002) believes that “successful use of communication strategies will prevent acquisition” (p. 212).

The model does not make it clear either how certain underlying individual differences, such as cognitive resources and abilities, affective and personality-related differences (Dörnyei, 2005) can affect the fossilization process. De Graaff (1997), for instance, showed that L2 learners’ grammatical sensitivity and ability to infer the meanings of words from a text were positively related to their ultimate attainment. Finally, Han’s model fails to address the issue of modality of performance (Skehan, 1996, 2002, 2009) and how and why some language learners use a linguistic form in an almost native-like fashion in one mode (e.g., written) but not in another (e.g., spoken). Equally, it might be said that this study was flawed and that is the reason for the contradictory findings. It may be argued, for example, that the input these learners were provided with was not robust enough, which, if true, could turn the whole equation around and the model would correctly have predicted the fossilization tendency of at least some of the specific errors that persisted in the learners’ written performance. However, the feedback provided to the learners was indeed robust, based on Han’s (2009) definition of L2 input robustness, because for 6 consecutive months the teacher-researcher provided the learners with individualized corrective feedback which specifically pointed out to them the erroneous use of a linguistic feature in terms of its form, meaning or function. In addition, he created a wide variety of oral and written comprehension and production activities in the classroom that specifically focused on the commonly occurring deviant forms in their writings. Moreover, the linguistic scope of learners’ errors was limited, so it was possible for the teacher to focus on these common errors several times during the course of the instructional treatment.

It could also be argued that these errors have not fossilized yet and the stoppage was just a learning plateau. This criticism cannot be justified in view of the pseudo-longitudinal nature of this research, as the errors repeated themselves not only among learners of a particular proficiency group but also across learners of different proficiency levels and instructional stages.

There are points that call for caution, however, and the model should not be written off immediately. First, only two error categories (pronouns and possessives) out of the four resistant and persistent error categories were looked into for validation purposes. Second, the SFH is as much about learnability as it is about fossilizability, so another way of testing the validity of the model is to see if it can account for the errors that disappeared from the performance

6.    References

Acton, W. (1984). Changing fossilized pronunciation. TESOL Quarterly, 18(1), 71-85.

Battistella, E. L. (1990). Markedness: The evaluative superstructure of language. Albany, NY: State University of New York Press.

Birdsong, D. (1992). Ultimate attainment in second language acquisition, Language, 68(4), 706-755.

Birdsong, D. (2004). Second language acquisition and ultimate attainment. In A. Davies & C. Elder (Eds.), Handbook of applied linguistics (pp. 82-105). London: Blackwell.

Birdsong, D. (2006). Why not fossilization. In Z. H. Han & T. Odlin (Eds.), Studies of fossilization in second language acquisition (pp. 21-34). Clevedon: Multilingual Matters.

Carduner, J. (2007). Teaching proofreading skills as a means of reducing composition errors. Language Learning Journal, 35(2), 283-295.

Chandler, J. (2003). The efficacy of various kinds of error feedback for improvement in the accuracy and fluency of L2 student writing. Journal of Second Language Writing, 12, 267-296