Errors and intelligence in computer-assisted language learning: Parsers and pedagogues (review)

Form-focused instruction - that is, 'any planned or incidental instructional activity that is intended to induce language learners to pay attention to linguistic form' (Ellis, 2001, pp. 1-2) - is a feature of proficiency-oriented communicative L2 classrooms. Research has shown that students desire feedback (Schulz, 2001) and that feedback can be very effective (Lyster, 2004). In computer-assisted language learning (CALL), difficulty arises when the goal is meaningful communication balanced with appropriate feedback on Students' errors.

This book promises to be a comprehensive overview of research in the areas of CALL, applied linguistics, and computational linguistics. It is touted as appropriate for upper-level undergraduates and graduate students and researchers interested in the above-mentioned areas as well as for those who work in second language acquisition, language pedagogy, or artificial intelligence. The central focus of the book is parser-based CALL, which is rooted in 'a formal linguistic representation of natural language input' and which identifies 'the grammatical functions of the parts of a sentence' (p. 2). The authors note that this is often identified by the 'misleading term' 'Intelligent CALL' (ICALL).

The book is divided into six chapters. Chapter 1 ('Introduction') provides a solid overview of the field. As is the case throughout the book, Heift and Schulze provide concise definitions of terms. Chapter 2 ('NLP in CALL') focuses mainly on the role of parsers. The authors discuss natural language processing (NLP), a subfield of artificial intelligence; the chapter ends with a thorough discussion of the research into and projects developed relying upon NLP resources.

The main focus of chapter 3 ('Error Analysis and Description') is the classification of errors in NLP. The authors point out that if error analysis in CALL is to be successful, it must begin with the appropriate classification of errors. Current research in this area has shown the difficulty of considering errors in semantics and pragmatics and of distinguishing errors from non-errors.

In chapter 4 ('Feedback'), Heift and Schulze begin with a real-world definition of the term 'feedback.' They thoroughly discuss related notions, including reinforcement, repair, and correction, and provide concrete examples of feedback options within CALL software. They also point to the importance of learner control and provide researchers and developers with theory from the areas of SLA and CALL, as well as the results of studies on learner uptake that will assist them in creating effective software.

The focus of chapter 5 ('Student Modeling') is individualized student tracking, which has not traditionally received much attention in CALL. The authors note here, as they do throughout the book, that the task at hand - in this case, initialization of the student model - is 'time-consuming and laborious' (p. 188); however, they cite research showing that learners benefit from the tracking of their errors.

Chapter 6 ('The Past and the Future') provides an overview of the research conducted in the last 25 years, and the authors note a number of obstacles that have been overcome in that time. Heift and Schulze point to the fact that pedagogy should lead future developments. They discuss the importance of training future users of ICALL technology in both the practicalities and the limitations of the software; they call for greater collaboration among researchers, especially in the areas of computational linguistics and CALL, and posit that the fields of ICALL and L2 pedagogy will continue to inform one another.

Common themes throughout the book include the rapidity of technological change, the tedious nature of a great deal of the work at hand, and, most importantly, the interaction among the related fields. The long list of abbreviations provided at the beginning of the volume is greatly appreciated. The authors tend to focus on their own research, which is in the area of German L2 learning; however, given the importance of this research in the field, this focus is justified. Figures provided in the book are often blurry, and at...