Understanding Language Learning Through Second Language

FiGuRe 1.2 Visual Representation of Vocabulary to Facilitate Acquisition

Source: From Hershberger/Navey-Davis/Borras, Plazas, 3e. © 2008 Cengage Learning.
(2) concepts such as comprehensible input and the learning-acquisition distinction are not clearly defined or testable; and (3) his model presents far too simplistic a view of the acquisition process (Brown, 2007; Lightbown, 2004; McLaughlin, 1987). Furthermore, use of the acquisition-rich environment diminishes the role of the learner in the foreign language classroom by highlighting the role of the teacher as the source of comprehensible input and by failing to recognize the function of learner-to-learner talk (Platt & Brooks, 1994). See
Latifi, Ketabi, and Mohammadi (2013) for the results of a recent interview with Krashen.⁴
Few would deny that Krashen’s model sparked a great deal of thought and discussion in the profession regarding the role of input in language learning and prompted many language teachers to provide more comprehensible TL input in their classrooms. Nevertheless, many of his claims paint an unclear picture of the role of classroom instruction in language learning and remain to be empirically tested. Despite this criticism on theoretical and empirical grounds, however, Krashen’s work prompted teachers to re-assess their use of the TL in their classes.
Language processing and Variability in performance. Krashen’s claim that only acquisition, and not learning, leads to spontaneous communication has been criticized because it fails to account for the fact that what learners can do with language often varies within a single learner, over time, within contexts, and across different learners. Historically, SLA researchers have attempted to explain this variability in performance by acknowledging the different ways in which learners access and process their knowledge so they can use it in communication (Bialystok, 1981, 1982; Ellis, 1997; Gass & Selinker, 1994; McLaughlin, Rossman, & McLeod, 1983; Ortega, 2009; Tarone, 1983). There is general agreement that learners’ mental representation of L2 is both implicit and explicit, that “consciousness” is a key characteristic that distinguishes them,and that these two types of knowledge are accessed through two distinct types of processing (Ellis, 2008). These types of processing deal with the notion of automaticity, the ability to access linguistic knowledge in a routinized manner using implicit (that is, unconscious, procedural) knowledge. For example, automaticity (that is, automatic processing) can be seen in the following brief exchange between two people who pass in the hallway at work:
Speaker1: Hi.
Speaker2: Hi, how are you?
Speaker1: Fine, and you?
Speaker 2: Fine. (Gass, 2013, p. 256)
The elements of language in this conversation have become so automatized, that is, used automatically, that we may answer “Fine” even before the question is asked. Automatic processing has been described as typically quick, unconscious, and effortless (Gass, 2013). The second type of processing is called controlled processing, in which the learner is thought to access explicit (that is, declarative, conscious) metalinguistic knowledge pertaining to language structures and linguistic concepts, which may slow down response time. As an example, if we are asked the same question as above, “How are you?”, in an unfamiliar language, we would need more time to access an appropriate response in a controlled manner using our explicit knowledge of the lexicon of that language (Gass, 2013). The current view of the differences in these two types of processing involves establishing associations or links in long-term memory, which leads to routinization and greater automaticity.
There is much less consensus in the field concerning the relationship between implicit and explicit knowledge—that is, whether that which is explicit can actually become implicit, what type of interface (if any) there is between these two types of knowledge—and there continues to be debate on this issue (Ellis, 2008; Gass, 2013; Lantolf & Poehner, 2014). Recent research in the area of cognitive neuroscience has contributed to this discussion by positing the existence of two different memory systems in the brain (Paradis, 2009; Ullman, 2005). Procedural memory underlies implicit linguistic competence and enables language acquisition to occur. Paradis (2009) describes this type of memory as “a functional system capable of generating sentences . . . acquired incidentally, by focusing one’s attention on something other than what is internalized” (pp. 4–5). According to Paradis, this competence cannot be described since it is unconscious, and it is accessed through automatic (unconscious) processing. The example above of the two speakers greeting each other is an example of procedural memory. It should be noted that the primary characteristic of this type of competence is “systematicity (the virtual lack of variance in performance in all contexts and in all components of linguistic structure, from prosody to phonology, to morphology, to syntax and lexical retrieval),” not speed or accuracy alone (Paradis, 2009, p. 35).
From a neurolinguistic perspective, declarative memory underlies explicit declarative (metalinguistic) knowledge, that is, knowledge about the language system that is typically learned in a formal L2 classroom situation and is accessed through controlled (conscious) processing. This type of knowledge, however, is thought to go beyond textbook rules of grammar, usage, and exceptions, that is, “rules of thumb” (Lantolf & Poehner, 2014, p. 124; Negueruela, 2003) and verb conjugations, by encompassing “high quality conceptual knowledge” (Lantolf & Poehner, 2014, p. 64). This type of conceptual knowledge deals less with correct vs. incorrect forms and more with the types of language choices one makes to express one’s intended meaning. As an example, textbooks often present rules of thumb for using the preterite and imperfect tenses in Spanish. According to Lantolf and Poehner (2014), these types of rules “are at best descriptively appropriate for specific contexts of use and at worse they are often misleading” (p. 69). Learners might benefit more from building a conceptual understanding of how verbal aspect, rather than verb tense, can affect how readers make sense out of a story or how writers can manipulate aspect to depict how actions are carried out in the past (e.g., simultaneously, sequentially, focusing on the beginning, middle, or end of an action, as background events or foregrounded actions in a narrative), as illustrated in a study described in Lantolf and Poehner (2014). The field has yet to develop concrete ideas for how conceptual understanding might be developed in the classroom to replace an exclusive discrete rules-ofthumb approach that is currently characteristic of the vast majority of language textbooks and pedagogical approaches. It is the viewpoint of Teacher’s Handbook that an approach designed around dialogic thinking and interactions could build these types of conceptual understandings, as in the PACE Approach. See Chapter 7 for a full discussion of this approach.
A key concept in this declarative/procedural framework is that language users compensate for their gaps in implicit competence by resorting to declarative knowledge, as in the earlier example in the situation of having to consciously access language rules, concepts, or words, for example, when encountering a familiar situation in an unfamiliar language. For both Paradis and Ullman, processing is either automatic or controlled, and there is no continuum between the two to the extent that declarative knowledge changes into procedural competence over time. On the contrary, in this view, as one’s implicit competence becomes increasingly available, one relies on declarative knowledge less. In sum, while declarative knowledge/memory plays a role in language acquisition, exactly what that role is, how it contributes to acquisition and performance, what should be done with it and to it during instruction, and what it becomes in the process is currently a topic of wide debate in the field (R. Donato, personal communication, June 11, 2014). Nonetheless, what seems to be evident is that, to the extent that L2 teaching is communicative, “it may involve some procedural memory (and result in some implicit linguistic competence)...” (Paradis, 2009, p. x).
Several caveats about automaticity offer important implications for teachers. First, automaticity does not consist simply of accelerated behavior but rather should be understood as performance over which one has greater control (Gass, 2013). Secondly, learners do not just keep on adding grammatical rules until they can use them automatically. Rather, consistent and repeated practice changes the knowledge representation by making the stored knowledge become more specific or more analyzed through a process called restructuring (McLaughlin, 1990). According to Lightbown and Spada (2013), “restructuring may account for what appear to be bursts of progress, when learners suddenly seem to ‘put it all together,’ even though they have not had any new instruction or apparently relevant exposure to the language” (p. 110). Further, it could explain backsliding, when learners overgeneralize new rules into contexts to which they do not apply, such as applying the English -ed past tense ending to an irregular verb, as in the case of “*I eated.”
Restructuring, mistakenly appearing to be error-ridden performance or forgetting, often results in U-shaped patterns of learning,which illustrate three stages of language use, as depicted in Figure 1.3. In Stage 1, the learner produces a form that is error-free, perhaps as a result of attention on that form in classroom practice. In Stage 2, the learner produces the form but with errors, which may arise for a variety of reasons: the learner is tired, the communicative situation is too demanding, or the new learning leads to restructuring of existing linguistic knowledge through overgeneralization of a language concept, as in the example of “*I eated” previously cited. In the face of these circumstances, the learner “makes the very error that he or she had so recently appeared to have learned to overcome” (Segalowitz, 2003, p. 397). Finally, in Stage 3, the learner once again uses the correct form as in Stage 1, having presumably restructured his or her understanding of the original structure and perhaps of new material.