LINGUIST List 28.716
Tue Feb 07 2017
Review: English; Japanese; Applied Ling; Lang Acquisition: Shintani (2016)
Editor for this issue: Clare Harshey <clarelinguistlist.org>
Marie Jouannaud <marie-pierre.jouannaud
Input-based Tasks in Foreign Language Instruction for Young Learners E-mail this message to a friend Discuss this message
Book announced at http://linguistlist.org/issues/27/27-1922.html
AUTHOR: Natsuko Shintani
TITLE: Input-based Tasks in Foreign Language Instruction for Young Learners
SERIES TITLE: Task-Based Language Teaching 9
PUBLISHER: John Benjamins
REVIEWER: Marie Jouannaud, Université de Grenoble - Alpes
Reviews Editor: Helen Aristar-Dry
Natsuko Shintani’s _Input-based tasks in Foreign Language Instruction for Young Learners details the results of a method comparison study with 6-year-old beginners in English in Japan. The methods compared are traditional Presentation-Practice-Production (PPP), supported by skill building theory, and Task-based Language Teaching (TBLT), congruent with newer interactionist approaches to FL acquisition. In so doing it also provides the reader with a very clear description of the Japanese context (which translates well to other large countries whose citizens might feel “forced” to learn English), a brief but comprehensive summary of the literature on intentional vs. incidental learning, error correction, and task repetition, ideas for tasks in the classroom (mainly for beginning learners), samples of classroom interaction which clearly exemplify the differences between the 2 methods presented, and last but not least, a blueprint for how to conduct a classroom experiment.
Chapter 1 “Getting started with task-based teaching” presents an engaging narrative detailing how the author came to Task-based Language teaching (TBLT) after she grew dissatisfied with the results of the students enrolled at her private language school and started experimenting with information-gap activities. She follows Ellis (2003) in characterizing a language task as an activity focusing on meaning, including a “gap” that triggers communication, letting learners use their own resources (linguistic or otherwise), and having a clearly defined outcome other than the use of language. With 6-year-old beginners, she decided to use “listen-and-do” tasks.
The title of Chapter 2, “Task-based language teaching in “difficult” contexts: Pedagogical issues”, refers to the difficulty of implementing authentic communication tasks in a country such as Japan, where English is both embraced as a necessary communication tool in a modern world, and seen as a potential threat to traditional Japanese culture. The author’s thesis is that some problems can be circumvented if a task-based approach is chosen: the fact that the learners share an L1, for example, is less of an issue if the task itself requires the use of English for its successful completion (and the use of L1 is natural in any case, especially in collaborative contexts as a mediating tool). TBLT might help with lack of motivation, if tasks are well-designed and students are made to feel successful, even when they are beginners or their level is very low, as the students in this study. Other common obstacles are less important in a young learner context: the English syllabus for elementary schools in Japan is less specified, and thus less clearly structural, than for older learners, and the competitive entrance examinations to senior high school and university (with no oral or interaction component) are far into the future. Other factors are less easy to deal with: many teachers in Japan, especially in elementary schools, are not confident that their level is high enough to teach English, let alone with TBLT; even when they are dissatisfied with the status quo, they clearly need support, and the training they receive at the moment is inadequate.
Chapter 3 “Theoretical Foundations of Task-based Language Teaching” justifies the decisions made in the study: in order to compare output-based PPP and input-based TBLT with 6-year-old beginners, the target items chosen were vocabulary items (taught explicitly in PPP, and picked up incidentally in TBLT), and 2 grammatical structures, plural –s and copular “be” (incidental in both conditions).
It also summarizes the relevant research on incidental learning, comprehensible input, the role of interaction (including negotiation of meaning), error correction, the role of output, and task repetition.
It ends with a description of the steps taken to mitigate the problems often associated with method comparison studies, i.e. the fact that one never knows what really happens in the classrooms and whether and how the methods compared are really different. A simplified version of Conversation Analysis was chosen to analyze the in-class interactions.
Chapter 4 “Introducing the comparative method study of PPP and TBLT” details the operationalization of both methods in the study: 45 6-year-olds were recruited and divided into 3 groups (PPP, TBLT and control), and each group was divided into 2 sub-groups so that no class had more than 9 students. The treatment spanned 9 lessons and each lesson contained the following activities:
- Production-based PPP: Listen and repeat/ Guess hidden items/ Throw dice and name item/ Production bingo/ Kim’s game (really a matching game), with 3 sequences of 2 lessons spent on 14 vocabulary items each, and 3 review lessons at the end;
- Input-based TBLT: the 3 tasks covering all 36 vocabulary items (24 nouns and 12 adjectives) were repeated for each of the 9 lessons: Help the zoo and the supermarket / Help the animals/ Comprehension bingo. There was no pre-task or post-task phase (the vocabulary was not pre-taught).
- Control group: unrelated activities (songs, etc.)
A battery of tests was carefully constructed, comprising production tests (assumed to favor PPP students) and comprehension tests (TBLT students), both discrete (e.g. name individual items) and contextualized (e.g. play a game with the teacher). The design adopted was pretest- treatment- immediate posttest- delayed posttest (after 4 weeks).
Chapter 5 “Comparing the process features of the two types of instruction” analyzes the types of exchanges using Conversation Analysis and shows that both conditions were clearly different. As expected, the questions in PPP are mostly display questions initiated by the teacher, whereas in TBLT all are referential questions (i.e. ‘real’ questions asked because one does not know the answer), initiated more often by students than by the teacher (so that the students did produce language even though they were not required to). In both methods, students mostly produced one-word utterances, but the teacher’s turns were twice as long in TBLT compared to PPP, and the students in TBLT produced half as much as the PPP students. The author concludes with the analysis of repair moves and shows that in PPP, these were mostly initiated and executed by the teacher (e.g. as recasts), whereas in TBLT they usually involved collaboration between the students and the teacher.
Chapter 6 “Learning Vocabulary through PPP and TBLT” presents the results for all 3 groups. All students were complete beginners at pretest. The control group did not improve significantly and was outperformed by both experimental groups with large effect sizes. For nouns, although the PPP group scored higher than TBLT on 3 tests out of 4, the difference was not significant; the TBLT group scored higher than PPP on the last test (contextualized listening) with a small but significant effect size. This showed that although both groups tended to outperform the other in the modality they had been taught with (output for PPP, input for TBLT), they also developed almost as good proficiency in the other modality, and their results were essentially indistinguishable.
For adjectives, the TBLT group significantly outperformed the PPP group on all measures. Shintani explains that this might be because the use of adjectives was always contextualized with the TBLT group, i.e. the adjectives were used by the teacher only to help the students identify the necessary referent (e.g. if they did not understand “squirrel”, she would tell them it was very small –perhaps with an accompanying gesture, or brown –and would touch a brown object in the classroom). They were used by students in the same way when they did not understand a noun (in which case they would ask, “Big? Blue?”. When the question was in the L1, the teacher would answer in English). In the PPP group, adjectives were introduced and practiced in the same way as nouns, i.e. with flashcards. The absence of difference for nouns is explained by the fact that both groups used flashcards for nouns.
Shintani also looks at the difference in quantity and quality of input, and concludes that the quantity of input was similar (although it was provided almost exclusively by the teacher in TBLT, and was roughly evenly divided between teacher and students in PPP). Qualitatively, the target items were mostly produced in isolation in PPP (including by the teacher), and were usually embedded in meaningful utterances in TBLT (especially for adjectives).
Chapter 7 “Incidental acquisition of grammatical features in PPP and TBLT” shows that four fifths of the TBLT students vs. only one fifth in PPP, were able to pick up one of the features (plural), but only in comprehension. Almost no-one showed any productive use of copula “be”. These findings are explained by the fact that TBLT task completion required comprehension of plural –s, but production of neither feature was necessary (“2 crocodile” and “peacock orange” are both comprehensible and do not impede communication). Comprehension of the copula (which is redundant anyway) was never required. Although the teacher did produce the plural in the PPP classes and recast students’ utterances when they did not repeat the plural ending, these recasts were mostly not picked up.
Chapter 8 “Theoretical implications of the study” summarizes the results and proposes avenues for further research, in particular to avoid the confound between input and TBLT on the one hand and output and PPP on the other.
The last chapter “Pedagogical implications of the study” deals with the counterarguments that may have occurred to the reader while reading the preceding chapters. The author argues for focused tasks (with predetermined target items) as more realistic in elementary school, and underlines how important task repetition is for the success of tasks (at least for this age group, repeating the same tasks 9 times in a row is not a problem). She believes that listen-and-do tasks such as used in this study with less than 10 children could be transferred to large traditional classrooms, and that teachers can develop their confidence in their own FL communication skills when using TBLT, as it forces them to communicate with their students.
The book ends with a conclusion, a list of references and appendices detailing some of the statistical analyses used in the study (too complex for this reviewer).
This book is trying to provide an answer to the perennial question, “What is the best language teaching method?”, with a comparison between a “traditional” method, PPP, and one method currently widely recommended, task-based language teaching (TBLT). The results are clearly in favor of TBLT, but it must be said that the way both methods were operationalized is not straightforward, especially as regards the roles of comprehension and production. Shintani says (p.31) that “PPP is by definition production-based”, but only the third P stands for “production”, and the Presentation and Practice sections should probably include a lot more listening (and presumably often do in real classrooms), as it has been known for quite some time that comprehensible input is an essential ingredient of L2 acquisition (Krashen 1981). Moreover, the 3rd P is supposed to stand for (free) Production, whereas all the activities presented really were “practice” activities: there was no progression toward freer production (which a “task” would have provided cf. Ellis (2003)).
TBLT itself is usually associated with production, so much so that one of its perceived weaknesses is insufficient input (Swan 2005). One of the strengths of the book is to remind the reader that tasks are compatible with input-based instruction, but that is certainly a very marginal kind of task, especially when all the input is provided by the teacher (and indeed the tasks for higher level students presented at the end of the book involve very little input by the teacher). TBLT is also associated with group work, but the children here worked individually, although they were free to interact, and the size of each class (7 on average) was not much bigger than a group in a regular classroom. Lastly, the second and third tasks may not be seen as true communicative tasks in other interpretations of the method, such as that of the Common European Framework of Reference for Languages (Goullier 2007). The first task (Help the zoo) was the only one where there was clear contextualization. It is much harder to think of a context for the second activity, where the children heard sentences like “the tiger needs the batteries” and had to choose the corresponding noun flashcards (“tiger” and “batteries”). The 3rd one (comprehension bingo) is also the 4th activity in the PPP lesson plan, except that it is comprehension-based in one case and production-based in the other. It is not clear why it is called a task in one case and not the other. In the Conversation Analysis results presented in chapter 5, Task 1 is indeed the one which triggered the most interaction (p. 93 and p.108).
Because of this, it seems that what was really compared was contextualized comprehension-based vs. decontextualized production-based instruction, with the former coming out ahead. This is acknowledged by Shintani in Chapter 8, but only in passing and not referred to in the rest of the book. This does not detract from the quality of the study, however. On the contrary, it is because it was so well designed and reported on that it is possible for practitioners working within different traditions to draw conclusions relevant to their contexts. I agree with Long (2009) that it is more useful to focus on methodological principles and lesson design features than to cling to method labels.
Another goal of the study was to study the difference between intentional learning (PPP) and incidental (TBLT), but I am not sure that it was helpful to use the term “incidental” for all of the target features in TBLT. Admittedly, none of them were pre-taught and the outward goal for the students was only performance of the tasks, but the teaching process was very different for nouns, all represented by flashcards and often used out of context, adjectives, used only in context, plural –s, which gave rise to several focus-on-form episodes, and copular “be”, which never did. It would have been interesting to ask each group what they had learned and observe the difference in their reports. It is harder for the teacher-researcher to decide what was intentional from the students’ point of view.
In the end, if my calculations are correct and if we assume that each lesson lasted one hour (this was not clearly stated in the text, although the introduction mentions 40-, 60- and 90-minute lessons), the PPP students learned to comprehend and produce on average 2.6 and 1.3 words per hour respectively, and the TBLT students 2.9 and 1.6. This is similar to what was reported in Alcaraz (2011) with 6- to 9-year old beginners in Spain (between 3 and 4 words per hour) using a communicative approach. The slightly lower results in this study might be explained by the fact that care was taken not to include cognates if at all possible, leading to the inclusion of infrequent words (‘ladle’) and adjectives which were probably too close in meaning for young beginners, e.g. big/small, heavy/light and long/short.will
The remarks above notwithstanding, this is an excellent book, well written, thought-provoking and with just the right balance between theory and practice. It fulfills its main goal by showing that task-based instruction is feasible with young beginners, at least as effective as other methods, and conducive to natural FL classroom interactions, provided it is input-based and teacher-led. It is highly recommended to SLA researchers and practitioners, including classroom teachers and teacher trainees.
Alcaraz, G. 2011. Quantity and Rate of Vocabulary Acquisition in the Context of Nonsystematic input: Elementary Education Students of English as a Foreign Language. PhD Dissertation. University of Murcia, Spain.
Ellis, R. 2003. Task-based Language Learning and Teaching. Oxford University Press.
Goullier, F. 2007. Council of Europe Tools for Language Teaching: Common European Framework and Portfolios. Didier/Council of Europe 2007.
Krashen, S. 1981. Second Language Acquisition and Second Language Learning. Oxford: Pergamon Press.
Long, M. H. 2009. Methodological principles for language teaching. In M. H. Long & C. J. Doughty (Eds.), The handbook of language teaching (pp. 373–394). Oxford: Blackwell. Will will will
Swan, M. 2005. Legislation by hypothesis: The case of task-based instruction. Applied Linguistics 26 (3), 376-401.
ABOUT THE REVIEWER
Marie-Pierre Jouannaud is a lecturer and teacher trainer in the Foreign Languages Department of Université Grenoble Alpes, France. She has an MA in linguistics. Her research interests within SLA include classroom research, blended learning and language assessment.
Page Updated: 07-Feb-2017