30 Sociocultural theory and the zone of proximal development

Transcription

1 30 Sociocultural theory and the zone of proximal development Amy Snyder Ohta 30.1 Introduction There has been increasing interest in perspectives on second language acquisition that are grounded in discourse, culture and ecologies, interpenetrating mind and learning with social interaction, culture and society, and considering L2 developmental processes as they occur in multiple settings of language use; such approaches have grown so much in recent years that Swain and Deters (2007) recognize a new mainstream emerging in SLA theory (2007: 820). Sociocultural theory, a neo-vygotskian theory of human cognitive, psychological, social and historical development, born of Russian cultural historical psychology, is one approach impacting research and thinking in the field of L2 development. The zone of proximal development (ZPD) is arguably the most widely applied construct from sociocultural theory, not only in educational research, but also in inquiry related to second and foreign language acquisition. Vygotsky (1978 [ ]) defined the ZPD as: the distance between the actual developmental level as determined by independent problem solving and the level of potential development as determined through problem solving under adult guidance or in collaboration with more capable peers. (Vygotsky 1978: 86) The purpose of this chapter is, after theoretically situating the ZPD in sociocultural theory and considering how the ZPD is introduced in Vygotsky s writings in contrast to how it is used today (Kinginger 2002; Chaiklin 2003), to examine selected socioculturally oriented SLA research relevant to the ZPD. Along with considering how the ZPD is useful in understanding SLA processes, the chapter also considers expanded notions of the ZPD that contribute to understanding L2 development. 648

2 Sociocultural theory and the zone of proximal development Background Sociocultural theory Sociocultural theory (SCT) is an integrative approach to human development and cognition built upon the work of Vygotsky, his students and colleagues, and contemporary scholars (see also Chapter 11, this volume). Lantolf and Thorne (2007) provide an overview of SCT, tailored to those interested in SLA; sources that are helpful in understanding Vygotskian thought include, of course, works such as Vygotsky (1978 [ ], 1987 [1934]) and Van der Veer and Valsiner (1994), as well as secondary sources, such as Wertsch (1985), Moll (1990), Newman and Holzman (1993) and Van der Veer and Valsiner (1991). Application of SCT to L2 learning focuses on Vygotsky s genetic, or developmental, method, and the mediated origin of the individual s higher mental processes in social processes. These are two of the three major themes of Vygotsky s work according to Werstch (1985). The third, which focuses on understanding the tools/signs that mediate human mental processes, has received less attention. In this section, I will focus on the first two of these three major themes in Vygotsky s work, considering Vygotsky s genetic method and the role of mediation in human higher mental processes, followed by a brief explanation of activity theory. Following this background, I will then move to the main topic of this chapter, the ZPD. Vygotsky s genetic method In SCT, the word genetic means developmental. Vygotsky s genetic method, an approach which considers development-in-process, differentiates sociocultural theory from traditional western approaches to science. SCT integrates theory and method into a tool-and-result (Newman and Holzman 1993) approach: namely, factors that have traditionally been considered to cause development are dynamically integrated such that human cognition is interactively formed through social interactional and cultural historical processes that cannot be separated from either the cognitive processing or cognitive mechanisms being formed. The genetic method is concerned with analysis of wholes and avoidance of reductionisms, because when processes are decomposed, they may not be properly understood. Vygotsky recommended using units of analysis, with a unit being a part of the whole, comprising all of the key characteristics of the whole (Vygotsky 1987: 46). Vygotsky died before fully fleshing out this approach, but scholars working in sociocultural theory work to preserve the integrity of the processes they are studying. Mediation Mediation is the most important notion in sociocultural theory, referring to the process through which human activity, including mental activity, incorporates a range of tools, and how these tools function to transform

3 650 AMY SNYDER OHTA activity and mind. Mediation means that thinking incorporates culturally constructed artifacts, concepts and activities, including language (Lantolf and Thorne 2006: 79). The origin of human mental processes is the mediation of mind via historically and culturally embedded social interactive processes. Human psychological and cognitive processes are mediated throughout the lifespan and continually formed through these processes. The idea of mediation encompasses physical tools, which not only work to change the physical environment but also become integrated in human culture and cognition, as well as psychological, semiotic tools, the most prominent of which is language. The incorporation of semiotic tools into human mental functioning occurs from the earliest human development, when language is used by caregivers to interact with and regulate the child s functioning via social mediation. The idea of regulation also relates to mediation, whether the individual is object regulated, other-regulated, or self-related, three types of regulation that are understood to be on a continuum, moving toward higher mental functions. What Vygotsky termed the general genetic law of cultural development relates to the role of socially mediated activity in human development, with cognitive development proceeding through a process of other-regulation: Every function in the child s cultural development appears twice: first, on the social level, and later, on the individual level; first, between people (interpsychological), and then inside the child (intrapsychological). This applies equally to voluntary attention, to logical memory, and to the formation of concepts. All the higher functions originate as actual relations between human individuals. (Vygotsky 1978: 57) Human development s socially mediated nature is also seen in the role of language activity in human development. Language is actively internalized through interactive processes such that it becomes a meditational means through which people regulate mental functions, including consciousness and thinking. Sociocultural scholars are interested in private speech, that is, speech for the self, and how it develops through internalization processes into inner speech and thought. L2 development is conceptualized as a process of re-mediating human mental processing with a new set of semiotic tools the language being learned. Private speech provides a window on these developmental processes (Ohta 2001c; Guerrero 2005; Woodall 2002; Negueruela, Lantolf, Jordan and Gelabert 2004; J. S. Lee 2006). Activity theory Activity theory (Engeström 1987, 2001) is a sociocultural-theoretic approach that focuses on the role of human activity in cognitive development to a greater degree than did Vygotsky, who was more interested in semiotic mediation of human higher mental processes (Lantolf and Thorne 2006). Some L2 researchers identify most strongly with activity theory and use

4 Sociocultural theory and the zone of proximal development 651 this term, rather than SCT, to describe their theoretical framework. However, it is important to keep in mind that activity theory and SCT cannot be distinctly separated. Activity theory is part of a sociocultural approach to development whose influence is most strongly seen in understandings of how social practice underlies and permeates human psychological and cognitive development The zone of proximal development The ZPD brings together the concepts of mediation, its role in the regulation and development of higher mental functioning, and the transformative role of imitating in internalization and development. The ZPD can be best understood as a kind of activity, as mediational processes whereby a person is enabled to do more than s/he could have accomplished individually through some sort of assistance. Incorporating assistance means that the individual s activity and thinking are facilitated, shaped, or guided by another s, words, tone of voice, gesture, eye gaze, facial expressions and deployment of other artifacts (visual aids, writing, etc.). The ZPD strongly relates to the general genetic law of cultural development, showing the earlier developmental processes where new knowledge and skills are realized collaboratively between people. Vygotsky s ZPD In Vygotsky s time, the ZPD was originally applied (and intended to be applied) to understanding mental age levels of children for the purposes of better projecting a child s educative potential (Van der Veer and Valsiner 1991). Vygotsky s definition, cited in the introduction (Vygotsky 1978), talks about an actual and a potential developmental level of a child, and how that potential level can be determined by seeing how a child is able to solve problems in collaboration with an adult. Van der Veer and Valsiner (1991) discuss Vygotsky s concern that IQ testing was being used to establish a child s mental age by investigating what a child could do without assistance. Children of widely varying abilities received the same IQ score, resulting in inaccurate educational placements. Vygotsky suggested that if, instead, the mental age/iq were to be compared with what a child could do with some assistance, one could more effectively determine a child s educability, and make better educational placements. The ZPD today Most researchers today use the ZPD differently than Vygotsky intended (Chaiklin 2003; Kinginger 2002). It is most often applied to better understand learning processes, classroom interaction and human development, including SLA, while still also being applied as a tool integrating assessment and instruction, as in Dynamic Assessment (DA) (Poehner 2007). The ZPD today is still applied to child development and now it is also common to apply the

5 652 AMY SNYDER OHTA ZPD to adult learning in formal and informal settings (McCafferty in press). The original formulation s mention of a teacher or more expert peer has been replaced today with an understanding that ZPDs also form in interactions between true peers who may or may not have greater expertise (Ohta 2001c), and between individuals as well as in groups (Guk and Kellogg 2007). In addition, mediation in the ZPD is understood to reveal potential development at the time of these collaborative experiences while also impacting the whole development of the person and mind. In this sense, today s ZPD is understood both as an assessment and as a developmentally significant activity related to the forming of mind. The dialogicality and social character of cognition developed via the ZPD is retained through adulthood in the development of help-seeking and help-incorporating processes that include a variety of media (books, electronic resources, social media) in addition to face-to-face interaction (Ohta 2006, 2010). It is also the case that researchers today may talk about processes that could be called the ZPD or which are related to the ZPD, but do not name them as such, and thus may not refer to the ZPD when it would be relevant to do so. For example, Knouzi, Swain, Lapkin and Brooks (2010) describe development in terms of the internalization of social speech leading to self-regulation for particular languagerelated tasks, processes that are often understood as part of the ZPD. There are some concerns regarding certain modern applications of the ZPD. Chaiklin (2003) and Kinginger (2002) point out that contrary to Vygotsky s own emphasis on wholistic thinking and integration of tool-and-result, it has become rather common in educational circles to apply the ZPD in a cause-and-effect manner as a pedagogical technique. This tends to occur when the ZPD is taken out of context and not integrated into the broader theory which gives the ZPD its meaning. Chaiklin s (2003) particular concern is that Vygotsky developed the ZPD as an assessment tool, not as a metaphor to guide our understanding of collaboration in human development. In terms of collaboration, Chaiklin insightfully notes that the effectiveness of collaboration was accepted by Vygotsky, who said that it is a well known fact that with collaboration, direction, or some kind of help the child is always able to do more and solve more difficult tasks tha[n] he can independently (Vygotsky 1987: 209). Vygotsky s innovation was suggesting that this well known fact be applied in determining a child s mental age or educability: The child s potential for moving from what he can do to what he can do only in collaboration is the most sensitive index of the dynamics of development and the degree of success that will come to characterize the child s mental activity (Vygotsky 1987: 210). Chaiklin s (2003) careful readings of original source materials and explication of the original meanings of the ZPD in Vygotsky s work and application are helpful for understanding the context and original meanings of the term. However, through the applications of the ZPD to diverse teaching and learning contexts, instructional content and learning across the lifespan, including to adult SLA, the ZPD has come to mean something somewhat

6 Sociocultural theory and the zone of proximal development 653 different than it did in Vygotsky s time. While we need not accept all applications of the ZPD as legitimate, L2 researchers Lantolf and Thorne (2006) also caution against taking a doctrinal approach to the ZPD; SCT and the ZPD are living concepts. It is legitimate for L2 researchers to apply the ZPD differently than Vygotsky did. Contextualizing the ZPD in Vygotsky s work and SCT goes far in avoiding mechanistic applications. A broadening of the ZPD s applicability has been useful and meaningful in L2 research L2 research and the ZPD SLA, including L2 teaching and learning, presents a wide range of problems for L2 researchers to consider. Sociocultural theory, in its integration of research and practice, embodies Vygotsky s concern about solving the problems of his day and is readily applied to modern contexts. Lantolf (2008), for example, echoes Vygotsky s commitment to praxis, a commitment to interconnections between practice, research and theory. Socioculturally oriented L2 researchers are most interested in development in the context of SLA that is, as L2 development, not general cognitive development. The words development, learning and acquisition are often used interchangably, as in this chapter. Depending on the study, L2 development may mean that learners showed progress in listening comprehension through teacher-mediated dynamic assessment (Ableeva 2002), more expertly used L2 syntax via peer collaboration (as in Ohta 2000a), or developed a higher level of linguistic and metalinguistic awareness via collaboration with peers or tutors (as in Swain and Lapkin 1998, 2002; Lapkin, Swain and Smith 2002; Guerrero and Villamil 2000; Aljaafreh and Lantolf 1994). All of these are examples of L2 developmental processes. Because language activity is what is being examined, and language use varies according to different settings, the question of whether particular forms are learned or acquired may or may not be considered. Some researchers conduct followup work, such as post-tests to investigate what learners have retained from classroom interventions (such as Lapkin and Swain, 2000), and or conduct longitudinal studies which show evidence of development over time (Ohta 2001c). Sociocultural L2 research is deeply process-oriented, valuing the understanding of how development unfolds in situ. Moment-by-moment analyses of discourse are often used to reveal developmental processes, called microgenesis (Wertsch 1985). In terms of the ZPD, while microgenetic linguistic development may or may not mean that the learner will be able to perform similarly later, or in a different context later the ZPD shows potential development (Negueruela 2008). And, in embracing a view of the whole learner, in and out of the classroom, SCT-oriented L2 researchers also investigate SLA in a wide variety of contexts: classroom instruction including pedagogies from communicative-approach college foreign language classrooms that incorporate a variety of collaborative tasks (Anton 1999; Ohta 2000a), to elementary

7 654 AMY SNYDER OHTA school classrooms (Poehner 2009), to teacher-fronted EFL classrooms abroad (Sullivan 2000); language learning in and out of classrooms abroad (Kinginger 2008); and advanced Asian language learning in classroom, self-study and study/soujourn abroad (Ohta 2006, 2010). In the next section, several L2 studies related to the ZPD will be presented using a taxonomy developed by expanding that of Lantolf and Thorne (2006). L2 studies that are relevant to understanding and using the ZPD will be discussed in the following categories: (i) L2 skills development; (ii) dynamic Assessment and tutor/teacher feedback; (iii) peer collaboration; and (iv) expanding the ZPD. Any taxonomy is difficult to apply because these are overlapping, not discrete categories. Considering the contributions of particular studies in certain areas outlined here does not, of course, limit the broader range of contributions made by each study L2 studies and the ZPD The ZPD and L2 skills development L2 skill building has been considered to be linguistic or metalinguistic (Kinginger 2002), though the two can be difficult to tease apart. Linguistic skill building may be conscious or unconscious, and has a reflective/metalinguistic component, though these components may or may not be visible in the words transcribed for the discourse data analyzed in various studies. For example, beginning learners, guided in teacher-fronted or peer learning settings to manipulate the nuts and bolts of a new grammatical system, are building a foundational level of familiarity and expertise with new forms and how to make and use them. A basic understanding and ability to produce and recall forms is needed for the language to be manipulable in the learner s speech and verbal thought, which form the basic tools of oral and mental manipulation and rehearsal, including metalinguistic reflection. Two studies which have implications for skill building in the ZPD will be considered here: Swain and Lapkin (1998) and Ohta (2001c). Swain and Lapkin (1998) investigated the interactions of French immersion middle-schoolers. Their detailed analysis of two middle-schoolers working on a jigsaw speaking and writing task allows insight into how learners assist each other with form meaning connections in both French and their L1, English. This study provides examples of linguistic and metalinguistic skill building in the ZPD. Swain and Lapkin address a particular problem in French immersion programs: grammar instruction that focuses on isolated rules, paradigms and the manipulation of form, rather than on relating form to function (1998: 325). Swain and Lapkin s analysis focuses on language-related episodes (LREs), which they define as any part of a dialogue where the students talk about the language they are producing, question their language use, or correct themselves or others (1998: 326). This includes supportive interactions in

8 Sociocultural theory and the zone of proximal development 655 the ZPD. Their data show how the pair gave and received assistance, mostly settling on correct forms, but not always. The forms that the learners collaboratively agreed upon, whether correct or incorrect, were selected by students as correct answers in individually tailored post-tests. This evidences the power of the ZPD in L2 skill building (students learned in the ZPD, at least in the short term), but also leads to the question of whether collaboration might cause students to build up misunderstandings or learn incorrect forms. However, it is important to note that this was a research study and not an ordinary classroom setting; teachers were not permitted to intervene. When students disagreed on a form, teachers were not available to answer questions or help to resolve differences in opinion. Had the task been implemented in a regular classroom context, the results might have been quite different because a teacher would have been available to help resolve learner questions. For example, Ohta (2001c) found that in ordinary college foreign language classrooms, students in pairs/groups who disagreed on forms asked the circulating teacher for assistance, or realized their mistakes during follow-up whole-class review led by the teacher. Swain and Lapkin (1998: 333), following their analysis, recommend that classroom tasks involving collaboration must be followed by opportunities for teacher feedback to minimize the potential for learning incorrect forms. Ohta (2001c) followed college-level learners of Japanese over an academic year of instruction. In terms of linguistic skill building, the study documents growth in learner use of listener responses over the period of the study, skills which were modeled by teachers, with assistance provided in teacher-fronted contexts tapping into a whole-class ZPD. Data from early in the academic year show that learners did not utter listener responses, but produced consecutive initiation/response (question/answer) pairs, or, sometimes, laughing (Ohta 2008) or pausing between initiation/response sequences. The following example from Ohta s corpus shows this lack of follow-up turns. Ungrammatical Japanese production is reflected in ungrammatical English translations. However, the nature of the grammatical errors is different. The purpose of these translations is to give the flavor of the Japanese, not to precisely pinpoint the nature of the errors. (1) 1 S: Uh (.) biiru o nomimasu? Uh (.) you drink beer? 2 C: Iie (.) uh (.) nomimasen No(.)uh(.)Idon t 3 S: Nihongo no benkyoo o shimasu ka? Do you do Japanese s study? 4 C: Hai (.) hai uh (.) benkyoo shimasu. Yes (.) yes uh (.) I do. 5 S: Eiga (.) ni (.) ikimasu ka? Do (.) you (.) go to movies?

9 656 AMY SNYDER OHTA 6 C: Iie (.) ikimasu (.) ikimasen. Eiga:: eiga ni (.) ikimasu ka. No (.) I go (.) I don t go. Do you (.) go to movies? 5 S: Hai ikimasu Yes, I do.(11/27) Learners at this early stage did not evidence errors in the follow-up turn; rather, they simply failed to produce any language in the follow-up turn. These turns, however, are very important in Japanese, where verbal listener responses, called aizuchi, are much more common than in English (Maynard 1989; Ohta 1995, 1997, 1999). These turns provide an opportunity for aligning responses which create a related, friendly tone in conversation. The classroom data show how teachers provided instruction on how to do these responses. This, however, was not sufficient to result in learner production of these turns, as students still did t produce them. In addition to instruction, teachers also, then, provided support during learner production tasks, in both teacher-fronted and pair-work settings, allowing students an opportunity to immediately produce such turns. In the following excerpt, a teacher, observing various pairs during a pair interview task, created a whole-class ZPD by calling out a reminder to the entire class, asking them to include the listener response Aa soo desu ka Oh really? : (2) 1 R: Ah shimbun:: (.) o yomimasu? Do you read the newspaper? 2 D: Sorry? (.) Hai (.) o shimasu Sorry? (.) Yeah (.) do. 3 R: Dono: (.) yomimasu ka? Which (.) you read 4 D: (3) Um (3) USA Today hehe (.) o yomimasu. USA Today o: (.) o yomimasu (3) Um (3) I read hehe (.) USA Today. I read (.) USA Today. 5 T: ((from a distance)) Minasan itte kudasai yo:: aa soo desu ka::. Everyone please say Oh really? 6 R: Ah soo desu ka::. Oh really? 7 D: Koohii de- koohii o nomimasu? Coffee with- do you drink coffee? (11/27) (adapted from Ohta 2001c: 197) Note in the above example that Rob and his partner did not produce listener responses after their first two initiation/response (question/answer) sequences, but Rob uttered a listener response in line 6 after the teacher s line 5 exhortation. By the end of the academic year, all of the students showed greater skill in producing appropriate follow-up turns, including the affective marking with ne that is critical for assessments (content-related emotional or evaluative comments) to be appropriate in Japanese. The following example from Ohta s

10 Sociocultural theory and the zone of proximal development 657 corpus illustrates a student s appropriate use of such a response, which was common by the year s end: (3) 1 C: Ka- kase o hiku: (.) um (.) kaze o hiku (.) toki doo shimasu ka? When you ha- have a culd (.) um have a cold (.) what do you do? 2 S: Uh soo desu ne::. Ha- hot mizu o uh hot mizu o nondari nemasu. Uh netari shimasu. Hmmm. I drink ha- hot water uh, I drink hot water sleep. Uh and sleep. 3 C: Unn ii desu ne. Oh that s good. (5/22) Skill development relates to both linguistic and paralinguistic skills. Haught and McCafferty s (2008) study of two Russian ESL learners relates the ZPD and skills development to development of language, prosodic features and gestures in weekly drama workshops. This study addresses the problem of language learning as development of the whole person, and how language development is a more holistic process than has been considered in much SLA research. Haught and McCafferty are particularly interested in the relationship between thinking and gesture in language development (for a review, see Stam in press and Gullburg in press). They consider the mediational role of gesture, prosody and body movements in ZPDs formed in the interactions between the students and their drama instructor. In these drama workshops, activities revolved around theater games, improvisations, tongue twisters and the rehearsal of written dramatic scripts (Haught and McCafferty 2008: 151). The students were uncomfortable with improvisation, but enjoyed working with written scripts they reported learning words and expressions through these opportunities. Activities included acting out scenes and discussing characters, situations and points of view. Discussion revolved around how to deliver these lines effectively, including talk about the meaning of the lines and their cultural context. Haught and McCafferty found that the students learned more than their lines they also mirrored the actor-instructor s prosody and gestures, incorporating these into their growing L2 repertoire, and increasing both linguistic and paralinguistic awareness. Repeatedly practicing delivery while imitating the instructor s prosody and gesture of emotionally laden lines in context provided learners with the opportunity to use language with high emotional and cultural content and to connect form, meaning, prosody, posture, facial expression and gesture in unique ways as the actor-instructor coached their performances. Haught and McCafferty found that the drama workshops comprised a context for learners to be supported in developing a range of L2 skills in the ZPD. It is also a good example of learner agency in the ZPD: learners followed their own interests in a drama workshop where they focused on their own areas of focus in working to develop their L2 proficiency.

11 658 AMY SNYDER OHTA Dynamic Assessment and tutor/teacher feedback in the ZPD Research on teacher feedback, or corrective feedback, in SLA has tended to focus on what type of feedback, for example, explicit or implicit, might be most effective for SLA (R. Ellis, Loewen and Erlam 2006). From a sociocultural approach, it is understood that feedback must be individually tailored and provided with sensitivity to the learner s developmental level; in this view, the best type of corrective feedback for a particular situation will vary depending on learner needs. SCT researchers are most interested in how various types of feedback, flexibly, responsively and collaboratively provided, are involved in moment-by-moment developmental processes. SCToriented studies of teacher feedback focus on interaction between teacher and student(s), and how feedback relates to learners unfolding needs as revealed, moment-by-moment in the discourse. There are a number of interesting sociocultural studies considering the role of tutor or teacher feedback on language development in the ZPD, in both individual tutoring sessions (Aljaafreh and Lantolf 1994; Nassaji and Swain 2000) and teacher-fronted classroom instruction (Anton 1999; Ohta 1995, 1999, 2000a, 2000b, 2001a, 2001b, 2001c; Yoshida 2009, 2010; Poehner 2009; Lantolf and Poehner 2011). In this section, three classroom studies will be considered: Ohta s longitudinal classroom study (Ohta 2000b, 2001c), Poehner s case study of an elementary classroom language learning game, from a DA perspective (Poehner 2009; Lantolf and Poehner 2011), and Yoshida s classroom study that incorporated stimulated recall to better understand the characteristics of corrective feedback noticed by learners (Yoshida 2009, 2010). Ohta (2000b, 2001c) expanded the notion of classroom corrective feedback by teachers to include considering how a teacher s feedback to an individual learner impacts others in the class. This was accomplished via individual recording of learners private speech, utterances for the self. Private speech data reveals how learners who were not directly addressed, uttered replies sotto voce, which Ohta termed vicarious response, and benefited from what Ohta termed incidental recasts; the latter emerge when a vicarious response containing an error contrasts with the teacher s utterance. Data captured private speech episodes where, as learners struggled to produce a vicarious response in private speech, a ZPD emerged as a contrasting, correct form was provided by the teacher and incorporated into the learner s utterance. In this way, Ohta s study shows how ZPDs form in teacher-fronted, whole-class contexts as students benefit from feedback the classroom teacher provides to nominated individuals. Yoshida s (2009, 2010) socioculturally oriented study of classroom interaction builds upon Ohta s work by incorporating stimulated recall data. Yoshida analyzed teacher feedback in her ten-week study of six Australian learners enrolled in the second year of university Japanese classes. She also found teacher feedback to impact learners beyond the particular student being addressed. Analyses of stimulated recall data show that the usefulness of teacher feedback to non-addressed students depends on the students level

12 Sociocultural theory and the zone of proximal development 659 of verbal engagement. The learners who were engaged in vicarious responses while the teacher called on and interacted with others were most likely to report noticing and attending to feedback addressed to their classmates. The results suggest that the level of learner engagement may relate to the extent to which a learner partakes in a whole-class ZPD. Poehner and Lantolf (Poehner 2009; Lantolf and Poehner 2011) further investigate the idea of the role of teacher feedback in a whole-class ZPD in their study of an elementary school Spanish teacher s implementation of a language learning game in a Spanish lesson. They present the teacher s methodology as an application of Dynamic Assessment (DA) (Poehner and Lantolf 2005; Poehner 2007, 2008, 2009; Lantolf and Poehner 2011), an interactive, instructional approach to assessment and instruction in the ZPD. Previous L2 literature in this area to date has applied DA to individual teacher/learner interaction during tutoring sessions. In their study, the teacher created her own rubric to guide her provision of feedback to students who made language errors, as follows: 1. Pause 2. Repeat the erroneous phrase with question intonation 3. Repeat just the error 4. Ask the student what is wrong with the utterance 5. Point out the incorrect word 6. Ask an either/or question 7. Provide the correct language 8. Provide an explanation (adapted from Poehner 2009: 481) In applying this rubric to implementation of a game involving the description of animals (the grammatical structure being taught was adjective noun agreement), the teacher moved through these prompts in sequential order, from less to more explicit feedback, depending upon whether or not a child was able to produce the correct form. Her implementation of this rubric resulted in strongly form-focused interaction during the classroom game. Analysis shows how three children varied in the number of prompts the teacher implemented before they produced the correct form. In some ways, feedback caused misunderstanding: for example, prompt 2, repeating the error with question intonation, means that the teacher uttered incorrect language. In the data, a student misinterpreted such a prompt, an NTRI (next-turn-repair initiation), as a content-related correction (See Foster and Ohta 2005 for a discussion of the ambiguity of NTRIs); the student did not understand the teacher was speaking incorrect Spanish, something that might not be expected teacher behavior. This misunderstanding, however, also revealed the student s lack of awareness of the error and lack of facility with Spanish grammar. The teacher then provided more focused feedback, using five more prompts until finally, using prompt 6 (ask an either/or question) resulted in the child s provision of the correct form. All in all, the

13 660 AMY SNYDER OHTA class seemed to benefit from these corrective procedures as measured by how two subsequent learners played the game, requiring less mediation. Lantolf and Poehner (2011) argue that, while they cannot be sure why subsequent learners required successively less intervention by the teacher, the learners most likely benefitted from participation in a collective ZPD (2011: 24); in other words, the teacher s feedback, though tailored to an individual student, tapped into the ZPD of the entire class, allowing others to benefit from the dynamic assessment process. The studies of teacher feedback reviewed above show developmental processes as learners interact with the teacher s corrective feedback. Poehner and Lantolf show how teacher feedback, moving from implicit to more explicit feedback, functioned in an L2 classroom game, as children performed more accurately as the game proceeded. Via analysis of vicarious responses, Ohta s and Yoshida s research shows learners making use of corrective feedback addressed to classmates. Yoshida takes this a step further in her finding that learners who witness teacher feedback to classmates are much more likely to notice that feedback when they, themselves, are verbally engaging in the classroom activity. Through these processes, teacher feedback allows whole-class ZPDs to emerge as learners are assisted to better understand and produce utterances beyond their current developmental level, showing their developmental potential. Yoshida s findings also underscore the importance of learner characteristics in this case, level of engagement, understood here as verbal participation in classroom activities as it relates to the impact of assistance in the ZPD Peer collaboration and the ZPD Vygotsky s definition of the ZPD requires an adult or more capable peer. Modern researchers applying Vygotsky s ideas have considered how ZPDs emerge in interactions between true peers. Analyses of pair and group work of L2 learners have produced repeated findings that ZPDs form, even among true peers, thus expanding our understanding of the ZPD in L2 learning settings. Donato (1994) is one such early study showing formation of ZPDs in small group learning settings, as learner collaboration allowed individuals to function at a higher level. Many researchers cited already in this chapter have built on this early work. More recently, Guk and Kellogg (2007) studied how ZPDs form among elementary school students doing group work in their foreign language classes. Along with collaboration allowing students to share their strengths, which Donato termed collective scaffolding, Ohta (2001c) suggests other mechanisms at work that promote the formation of ZPDs, even when greater expertise may be lacking. Pair and small group collaboration allows learners to take on different roles, as speaker, addressee, or auditor (conversational participant who is not currently addressed.

14 Sociocultural theory and the zone of proximal development 661 These terms are borrowed from Bell (1984), a sociolinguist interested dialectical variation and how speakers adapt their language to their interlocutors, depending upon whether the interlocutor is an addressee, auditor, overhearer (present in the area and known to participants, but not part of the conversational group) or eavesdropper (present in the area but unknown to participants). Ohta (2001c) applies Bell s terms in understanding the impact of various participant roles on cognitive processes in language learning. While the speaker is encumbered by the speaking role and has little working memory available to solve language problems emerging during online production, addressees and auditors are actively involved in the pragmatic notion of projection (Levinson 1983), mentally moving beyond the speaker s words to what might be coming next. Projection involves anticipatory thought, as the listener tracks the speaker s utterances, including potentially formulating what might come next in inner speech. For learners engaged in pair or group work, projection allows the listener to chime in to help the speaker with problems that emerge, even in cases where the listener could not have produced the entire utterance him/herself. Simply being placed into groups and assigned language tasks, however, is not sufficient for ZPDs to form. Peers vary in how collaborative and mutually supportive their interactions are (Storch 2002). In addition, the classroom context is critical in supporting peer interaction. DiNitto (2000) shows the importance of context for the emergence of ZPDs in classroom group work. She studied group work in an intact Japanese language class that was ordinarily conducted using a teacher-fronted instructional approach where learners performed and adapted memorized dialogues in class while the teacher promptly corrected all errors (Noda 1998). DiNitto, who taught the class, was trained in this teaching methodology. However, one quarter she added occasional group work tasks into her curriculum without modifying her pedagogy in other ways. These tasks were different from the dialogue recitations that dominated her pedagogy; they involved situations that were different from those presented in memorized material, such as using new vocabulary to create maps, giving and receiving directions, role-playing restaurant ordering scenes and talking about the content of short reading passages. During one of these tasks, DiNitto video- and audiorecorded group work conducted by the students in two groups of four. While one of the groups collaborated and showed interactions of the type DiNitto expected, another reflected the norms of the teacher-fronted classroom one student, Walt, took on the role of teacher and conducted the group work in a teacher-fronted fashion. DiNitto notes that the language used in Walt s group reproduced the norms of her methodology, including: the explicit recitation of grammar rules, the IRF [initiation/response/follow-up sequence]-style response, the roles of expert and novice, and the privileging of form over meaning and accuracy over communication (2000: 201). In terms of the ZPD, clearly the broader context in which group work is embedded appears to be critical for its success while DiNitto was at first

15 662 AMY SNYDER OHTA surprised by the non-collaborative group s performance, she cites Coughlan and Duff s (1994: 190) assertion that second language data cannot be neatly removed from the sociocultural context in which [they were] created or collected. She states, Students were experiencing group work as a new activity, one for which they were ill-prepared and which was by nature at odds with the classroom setting they had acclimated themselves to over the previous twenty weeks of instruction. This last point is key, for as Barnes and Todd (1995: 16) argue, for success [in small group work] it is essential for the teacher to help students grasp what kind of learning is required. This did not happen in the classroom under review. However, I believe that help[ing] students grasp what kind of learning is required implies much more than teaching the students how to collaborate in the L2. It requires an underlying philosophy of language learning that was not present in this classroom. (DiNitto 2000: 201) She concludes by questioning: We could ask why Walt s group did not succeed, but a more revealing question is, Why should they have succeeded? Why should they have opted for a model of learning other than the dominant learning style of the classroom? (DiNitto 2000: 201) DiNitto (2000) does point out that the classroom environment is not the only factor determining whether or not learner collaboration can be productive one of the two groups in her class did construct interactions where learners supported each other s language use in the ZPD. Her main goal in presenting analysis of the classroom and two group task performances is to highlight the power of the dominant classroom methodology and underlying ideology and its influence on learners and learning. Ohta (2000b, 2001a, 2001b, 2001c) investigated the interactions of learners in their first or second year of college Japanese language classes. Ohta s longitudinal study provides insight into the process of learning a new language that is typologically distant from the learners L1s. She collected audio and video data in intact classes as implemented by the lecturers and teaching assistants in charge of instruction, without any intervention by the researcher. Close miking of students enabled collection of student talk during group work and private speech, whispered talk to the self, that occurred during whole-class, teacher-fronted instruction. In terms of the peer collaboration in the ZPD, Ohta s analyses of interaction in learner learner pairs show numerous examples of peers providing one another with finely tuned, developmentally appropriate support. Aljaafreh and Lantolf (1994) point out how assistance in the ZPD, in order to be effective, must give the speaker increasing responsibility for his/her own performance, and that assistance must be negotiated, with the goal of providing just enough assistance (1994: 469). Ohta s (2001c) Japanese

16 Sociocultural theory and the zone of proximal development 663 language learners naturally seemed to follow a principle of providing just enough assistance, moving from more implicit to more explicit forms of assistance as needed. For example, a common form of peer assistance was to wait, allowing the partner time to finish an utterance or to self-correct. In the following excerpt Kuo-ming s pauses evidence his peer interlocutor s strategy of waiting. Pause length is shown in seconds (i.e. (1) meaning 1 second) or, where less than one second, the notation (.) is used. (4) 1 Km: um (1) um (1) suteki no:: (.) um ske-suteki na sofa desu ka? (2) Desu ne or desu k- (.) haha desu ne:, (.) um (1) um (1) is it the (.) nice ((wrong particle)) um ni-nice ((correct particle)) sofa ((wrong final particle))? (.) ((re-states copula with corrected final particle and wrong final particle)) (.) ((chooses correct final particle)) (Ohta 2001c: 90) Another common form of assistance was the prompt, where a peer assisted by providing a partial repetition, stopping before the trouble-source to allow the partner to do his/her own reformulation. (5) 1 Pf: Anoo sumimasen. Um excuse me. 2 Sr: Hai Yes 3 Pf: Ima nanji desu ka? What time is it? 4 Sr: Um (2) kuji? Um (.) Nine o clock? 5 Pf: Kuji Nine o clock 6 Sr: I guess that s ten after (.) 7 Pf: Kuji::: Nine oclock::: 8 Sr: juppun desu. Ten minutes (( kuji juppun, literally nine o clock ten minutes, is the correct way to tell the time)). (Ohta 2001c: 91) Besides prompting, peers also assisted via co-construction, chiming in what the partner needed to produce an utterance. In these and other ways, peers tailored their support, allowing their interlocutor to do more than s/he could have done individually. Ohta (2001c) also found that negotiating assistance was about more than just providing nudges of support to classmates in fact, students were sensitive to their own needs and limitations. Students who expressed doubts

17 664 AMY SNYDER OHTA about a form received help from peers, and when a peer wasn t sure, inquiring of the circulating teacher, or if the teacher wasn t accessible, making a note to check the form later. These findings underscore the importance of the agency of the learner in working together with the teacher and classmates to create contexts effective for language learning not only by providing developmentally appropriate assistance to one another but also by judging for themselves when they needed such assistance, and by making notes of what required further inquiry. Ohta (2001c) found that learners were able to make use of a variety of learning activities, from the traditional to the innovative, mechanical to creative, in their language learning. Interestingly, findings suggest that while some task types may be more effective than others in terms of providing a context for linguistic skill building in the L2, this also depends on the learners own orientations toward tasks, again highlighting the role of learner agency in SLA. As mentioned in the previous section, another important finding related to the ZPD is that learners also received assistance during whole-class, teacher-fronted interactions that they vicariously participated in. Kitade (2008) investigated peer collaboration in advanced content-based Japanese language classes in the context of asynchronous computermediated communication (ACMC) with Japanese key-pals (see Chapter 14, this volume). ACMC refers to exchanges in this case. Kitade investigated both how Japanese key-pals, native-speaking partners assigned to pairs of American learners of Japanese, adjusted their messages in order to better suit the learners developmental levels, as well as analyzing the collaboration between the paired American learners that occurred offline, during the process of reading messages by their Japanese key-pals and co-authoring responses. Kitade s goal was to more fully understand the impact of ACMC in providing language learning opportunities. Like earlier studies of collaborative dialogue (i.e. Swain and Lapkin, 1998), Kitade also included individually tailored post-tests to investigate whether learners retained the knowledge co-constructed in collaboration with their peers. Kitade found that collaboratively working in class on reading and replying to Japanese key-pal s was quite productive for the students, as students were able to provide and receive assistance in the ZPD from each other and from instructors, incorporating this assistance into their replies. Kitade explains how the in-class collaboration provided important developmental opportunities: had they done these assignments at home, working independently, they would have missed the language learning that occurred via metalinguistic talk during peer collaboration (2008: 77). Consistent with results of previous research on collaborative writing (see, for example, Storch and Wigglesworth 2007; Wigglesworth and Storch 2009), Kitade found that peers working together produced higher-quality texts than did students who worked alone. And, dyad-specific post-tests demonstrated the effectiveness of peer collaboration: 73 percent of post-test items were correct, consistent

18 Sociocultural theory and the zone of proximal development 665 with previous research findings (Swain and Lapkin 1998). Collaborating during production (writing messages) appeared to have a different effect than collaboration during receptive tasks (reading messages). Collaborative writing resulted in greater retention of new vocabulary than did collaborative reading and discussion of messages received. All in all, collaboration during writing tasks enhanced learning by providing a context for mutual assistance in the ZPD: Although ACMC [asynchronous computer mediated communication] activities are frequently conducted as outside-the-classroom assignments, the findings in this study indicate the significance of the in-class ACMC activity, since this entails the beneficial aspects of offline talk. Although reference to online dictionaries is useful, the learners retrospective interviews suggest that there are limitations in the scope of these dictionaries. Unlike the receptive mode (reading), which requires only comprehension, the productive mode requires the selection of the correct linguistic knowledge and awareness of how to apply that knowledge in a particular context. [The] collaborative peer context is able to meet such complicated demands that cannot be solved using dictionaries. (Kitade 2008: 78) Kitade found variability among dyads in how helpful collaboration was, and in how successful they were in managing the interaction in productive ways. For example, some pairs made good use of the instructor s availability to solve intractable problems, while others did not. Overall, however, she found that the opportunity for collaboration was beneficial to students who chose to collaborate Expanding the ZPD For adult learners, benefitting from giving and providing assistance in the ZPD is not only dependent on having opportunities to collaborate, as Kitade (2008) pointed out, but also on individual differences in how learners manage challenges that arise during language learning. The ZPD is the locus of semiotic mediation in learning and development; in the ZPD, learners interactively incorporate assistance in their activity in a way that shows learning as a process of the present (actuality) on the basis of anticipation of immediate future possibilities and through construction of reality out of these anticipated possibilities (Valsiner 2001: 86). And, studies such as those reviewed above have shown how opportunities for interaction and collaborative assistance not only show learners developmental potential, but also promote development of linguistic skills that serve as resources to learners beyond the present task. Recently, some studies have begun to shed light on how, for adults, the ZPD is not only a place of mediation by other persons, but that ZPDs also form when adults actively engage with a variety of resources in language