The Routledge Handbook of Second Language Acquisition

Susan M. Gass, Alison Mackey, Susan M. Gass, Alison Mackey

• Publisher: Routledge
• Year: 2013
• Print ISBN: 9780415479936
• eBook ISBN: 9781136666889
• Topic: Languages & Linguistics
• Subtopic: Linguistics
• Index: Languages & Linguistics

The Routledge Handbook of Second Language Acquisition brings together fifty leading international figures in the field to produce a state-of-the-art overview of Second Language Acquisition.

The Handbook covers a wide range of topics related to Second Language Acquisition: language in context, linguistic, psycholinguistic, and neurolinguistic theories and perspectives, skill learning, individual differences, L2 learning settings, and language assessment. All chapters introduce the reader to the topic, outline the core issues, then explore the pedagogical application of research in the area and possible future development.

The Routledge Handbook of Second Language Acquisition is an essential resource for all those studying and researching Second Language Acquisition.

Part I

Language in context

1

Interactionist approach

Alison Mackey, Rebekha Abbuhl, and Susan M. Gass

Introduction

In the 30 years since the initial formulations of the Interaction Hypothesis (Long, 1980, 1981), there has been an explosion of studies investigating the ways in which interaction can benefit second language acquisition (SLA), with the most recent work documenting its evolution from hypothesis to approach (Gass and Mackey, 2007a). This review begins with an overview of the historical background of the interactionist approach and then discusses the core issues surrounding it, examines some of the ways in which data are collected in this area of SLA, and explores the practical applications of the approach. Directions for future research will be addressed in the final section.

Historical discussion

The roots of the interactionist approach can be traced to several lines of research that began in the 1970s. At this time, researchers became increasingly interested in the types of discourse patterns found in native speaker and learner conversations. In particular, they examined “foreigner talk” and the ways in which native speakers modified their speech so as to make it more comprehensible for learners (e.g., Ferguson, 1971). Paralleling a similar line of work in first language acquisition (which examined caretaker talk and the ways in which parents modified their speech for young children), researchers in this area argued that modifications such as repetition and syntactic simplifications served to make the input more comprehensible to second language (L2) learners, and in this way promoted the acquisition of the target language.

This focus on input and comprehensibility could also be found in another strand of research that influenced the development of the interactionist approach: Krashen's Input Hypothesis (Krashen, 1977, 1980). According to this hypothesis, input that was comprehensible (but slightly above the learner's current level of proficiency) was the driving force behind language acquisition. In Krashen's view, if a learner was exposed to this type of input, and, at the same time, had a “low affective filter” (i.e., low levels of anxiety and negative feelings associated with learning the L2), acquisition of the non-native language would automatically (i.e., subconsciously) take place. For Krashen, any mechanism that served to make the input comprehensible (e.g., simplifying the grammar of a written or oral text) was of value; however, interaction between native speakers and learners did not hold any special place in his theories.

However, from the late 1970s researchers began to accord more importance to interaction itself. Wagner-Gough and Hatch (1975), for example, maintained that researchers needed to examine “the relationship between language and communication if we are looking for explanations of the learning process” (p. 307), while Hatch (1978a, 1978b) argued that interaction might be an actual site for L2 learning and not just a means of observing what had already been learned. In Hatch's (1978b) words, “one learns how to do conversation, one learns how to interact verbally, and out of the interaction syntactic structures are developed” (p. 404).

Drawing upon the work of these researchers, Long formulated his initial version of what became known as the Interaction Hypothesis (Long, 1980, 1981). In Long's view, both comprehensible input and L2 development stemmed from the conversational modifications that occurred when native speakers and non-native speakers (NNS) worked to resolve a communication difficulty. Although these modifications (commonly referred to as interactional adjustments during negotiation for meaning) were not, of course, the only means of achieving message comprehensibility, Long suggested these modifications were positioned to promote comprehensible input, and, ultimately, L2 acquisition.

Early research using the interactionist approach framework sought to describe the frequency and types of interactional modifications used in native speaker-learner and learner-learner pairings, examine the relationship between negotiation of meaning and learners’ comprehension, and compare the effects of premodified and interactionally modified input (e.g., Doughty and Pica, 1986; Gass and Varonis, 1985, 1986; Long, 1983a, b; Loschky, 1994; Pica and Doughty, 1985; Pica et al., 1987; Porter, 1986; Varonis and Gass, 1985). As several researchers, including Mackey (1999), Ellis (1999) and Spada and Lightbown (2009), pointed out, this early research tended to be descriptive and did not seek to provide direct evidence that interaction was causally linked to L2 acquisition.

Early versions of the Interaction Hypothesis incorporated Krashen's claims about comprehensible input being necessary and sufficient for development in the L2: “Access to comprehensible input is a characteristic of all cases of successful acquisition, first and second ... greater quantities of comprehensible input seem to result in better (or at least faster) acquisition ... and crucially, lack of access to comprehensible input ... results in little or no acquisition” (Long, 1983b, p. 210). However, a number of researchers took issue with Krashen's claims about comprehensible input and SLA. Swain (1985, 1995), for example, argued that while comprehensible input was necessary for L2 acquisition to occur, it was far from sufficient. Based on her work with French immersion students in Canada, she argued that if learners do not have regular opportunities to speak or write the language (that is, to produce output), their production skills (speaking and writing) would lag considerably behind their comprehension skills (listening and reading). This observation served as the starting point for Swain's Output Hypothesis (1985), which posits that producing output plays a crucial role in the development of the L2, as it (a) gives learners the opportunity to practice and thus to automatize the production of the language; (b) allows learners to test hypotheses concerning the L2; (c) forces learners to focus on structure of the language; and (d) draws learners’ attention to gaps in their interlanguage (1995; see also Swain, 2005). Swain suggested that second language learners need to be pushed to produce output, arguing that “being pushed in output, it seems to me, is a concept parallel to that of the i + 1 of comprehensible input. Indeed, one might call this the comprehensible output” (1985, p. 249). As we discuss later in this chapter, Swain's Output Hypothesis is subsumed in later versions of the Interaction Hypothesis.

The idea of comprehensible input being insufficient for second language acquisition was also discussed by White (1991, 2003). Approaching the role of input from a different theoretical perspective, White claimed that it is unlikely that French learners of English acquire the rule that delimits adverb placement in English (the strict adjacency principle in case assignment) with positive evidence alone because the positive evidence they receive does not contain the information that what is acceptable in French adverb placement may not be acceptable in English. To elaborate, in English, adverbs cannot be placed between the verb and the direct object (thus “Mary watches often television” is ungrammatical). However, the same word ordering —V Adv O—is possible in French (“Mary regarde souvent la television”). Thus, drawing upon his/her L1, the native speaker of French learning English might assume that a sentence such as “Mary watches often television” is acceptable because positive evidence (i.e., English input) provides no information about non-acceptable utterances. White points out that if a learner were solely dependent on comprehensible input for making progress in the L2, they would have to notice the absence (referred to as indirect negative evidence) of a particular structure (for example, the V Adv O structure in English). While this is theoretically possible, the language acquisition process may be facilitated if the learner receives assistance in the form of correction or instruction. In essence, what White argued was that comprehensible input is insufficient for certain aspects of L2 development. However, as we will see below and in Chapter 2, the interactionist approach takes into account the important construct of feedback that helps to account for how learners receive information about incorrect utterances.

Other extensions of the Interaction Hypothesis grew out of points made about the role of attention in second language learning. Work in the early 1990s proposed that only consciously noticed features of the input became intake (e.g., Schmidt, 1990, 1993). According to Schmidt's Noticing Hypothesis, “subliminal language learning is impossible, and ... noticing is the necessary and sufficient condition for converting input to intake” (1990, p. 129) (see also Chapter 15). Researchers were quick to point out, however, that L2 learners’ input processing strategies might make certain portions of the L2 difficult to notice. As VanPatten (1989) noted, L2 learners typically practice a form of selective attention, focusing on a limited and thus less overwhelming portion of the input. Those aspects of the input that are salient and meaningful are typically those that draw the learners’ attention; features that lack saliency or communicative value (such as determiners in English, gender agreement in Spanish, or postpositions in Japanese) may pass under the learner's radar, so to speak. For this reason, researchers argued, L2 learners might benefit from having their attention drawn to formal features of the target language. This claim, too, made its way into a later version of the Interaction Hypothesis.

Drawing upon the work of these researchers, Long presented a reformulated version of the Interaction Hypothesis in 1996, which stated that “Negotiation for meaning, and especially negotiation that triggers interactional adjustments by the native speaker or more competent interlocutor, facilitates acquisition because it connects input, internal learner capacities, particularly selective attention, and output in productive ways” (p. 451). The following section deals with the major tenets of the most recent updates of this approach (Gass and Mackey, 2007a), along with current research.

Core issues

In the most recent version of the interactionist approach, Gass and Mackey (2007a) note that, “it is now commonly accepted within the SLA literature that there is a robust connection between interaction and learning” (p. 176). The interactionist approach posits that the interactional “work” that occurs when a learner and his/her interlocutor (whether a native speaker or more proficient learner) encounter some kind of communication breakdown is beneficial for L2 development. For example, when a learner experiences difficulty understanding his/her interlocutor or making himself/herself understood, discourse strategies such as clarification requests, confirmation checks, repetitions, and recasts may be employed to help resolve the difficulty. In this manner, the learner may have received input that has been modified often in an effort to make it more comprehensible. The process of interacting with another individual may also serve to draw the learner's attention to some sort of “gap” (in Schmidt's terms, 1990) between his/her interlanguage and the target language. For example, the learner may become aware of a difficulty using a particular linguistic feature. Aware of the gap, the learner may pay more attention to the subsequent input, something that is believed to be essential for L2 acquisition. The learner may also have his/her attention drawn to this gap through either explicit feedback (such as metalinguistic corrections) or potentially more indirect forms of feedback (such as recasts). Both of these terms will be explained in what follows, together with examples (also see Gass and Mackey, 2007a; Mackey and Abbuhl, 2006; Mackey, 2007 for recent overviews).

This constellation of features—interactionally modified input, having the learner's attention drawn to his/her interlanguage and to the formal features of the L2, opportunities to produce output, and opportunities to receive feedback—are the core components of the interactionist approach and have been investigated in nearly a hundred empirical studies since the mid-1990s. Researchers have found that interaction and its concomitant features are beneficial for a range of morphosyntactic features, including articles (Muranoi, 2000; Sheen, 2007), questions (Mackey and Philp, 1998; Mackey, 1999; Philp, 2003), past-tense formation (Doughty and Varela, 1998; Ellis, 2007; Ellis, Loewen, and Erlam, 2006; McDonough, 2007), and plurals (Mackey, 2006a). These results appear to hold true for children as well as adults (Mackey and Oliver, 2002; Mackey and Silver, 2005; Van den Branden, 1997), classroom as well as laboratory settings (Gass et al., 2005; see Mackey and Goo, 2007 for a meta-analysis in this regard), and for a range of languages, including French (Ayoun, 2001; Swain and Lapkin, 1998, 2002), Japanese (Ishida, 2004; Iwashita, 2003), Korean (Jeon, 2007), and Spanish (de la Fuente, 2002; Gass and Alvarez Torres, 2005, reprinted 2011; Leeman, 2003).

The current research agenda has moved away from investigating whether interaction impacts L2 outcomes to determining: (a) which aspects of the L2 benefit the most from interaction; (b) how individual difference variables mediate the relationship between interaction and L2 development; and (c) what forms of interaction (and in particular, what types of feedback) are the most beneficial for L2 learners (how various types of interactional feedback differentially impact various L2 forms).

The question of whether interaction differentially affects L2 development, in other words, whether interaction (more specifically, interactional feedback) works for all L2 forms or only for some forms, but not others, has been raised by a number of recent researchers (e.g., Jeon, 2007; Long, 2007; Long et al., 1998; Mackey et al., 2000). In particular, it has been proposed that aspects of the L2 that possess both transparency and high communicative value (such as lexis)—in comparison with less salient and more complex features such as morphosyntax—may receive the greatest benefit from interaction. For example, in one recent study, Jeon (2007) compared the effects of interaction on a range of morphosyntactic and lexical targets in Korean. Using a controlled pre-test-post-test design, Jeon found that her Korean as a foreign language learners experienced more gains with the lexical targets (concrete nouns and action verbs) and one of the morphosyntactic targets (object relative clause constructions) than with the highly complex morphosyntactic target of honorific subject-verb agreement. Jeon suggested that the low saliency and communicative value of the latter target might have mitigated the effect of interaction. In light of these and similar findings, as well as the fact that few studies investigating the effects of interaction have addressed p...

Table of contents

• Frontcover
• The Routledge Handbook of Second Language Acquisition
• Routledge Handbooks in Applied Linguistics
• Title Page
• Copyright
• Contents
• Illustrations
• Abbreviations
• Contributors
• Introduction
• PART I
• PART II
• PART III
• PART IV
• PART V
• PART VI
• PART VII
• Glossary
• Index