635 Studies in Second Language Learning and Teaching Department of English Studies, Faculty of Pedagogy and Fine Arts, Adam Mickiewicz University, Kalisz SSLLT 11 (4). 2021. 635-639 http://dx.doi.org/10.14746/ssllt.2021.11.4.8 http://pressto.amu.edu.pl/index.php/ssllt Book Review Learning words from reading: A cognitive model of word-meaning inference Author: Megumi Hamada Publisher: Bloomsbury Academic, 2021 ISBN: 978-1-3501-5368-4 Pages: 168 Acquiring words is an essential, yet daunting, part of learning any language. In English, current estimates indicate that learners need to be able to understand around 8,000-9,000 word families, that is, headwords with their inflections and derivations (Bauer & Nation, 1993), to attain satisfactory levels of reading com- prehension (Nation, 2006). To put this into perspective, 8,000 word families could amount to over 34,000 individual words (Schmitt, 2010). Academic texts appear to be even more demanding, typically requiring knowledge of around 14,000 families (Webb & Paribakht, 2015). Second language (L2) learners cannot achieve such knowledge only by intentionally attempting to commit each word to memory. Instead, most words must be acquired incidentally, that is, as a by- product of meaning-focused tasks such as reading. The problem is that inci- dental learning through reading is slow and necessitates vast amounts of com- prehensible input. Thus, seeking to improve incidental vocabulary learning and teaching, the book Learning Words From Reading: A Cognitive Model of Word- meaning Inference by Megumi Hamada proposes an eponymous model of reading 636 that draws on second language acquisition (SLA) and psycholinguistic research to illustrate “how the meanings of words encountered during reading are learned” (p. viii). The model is hence the brainchild of a timely collaboration between two related and yet distinct research traditions. And it is perhaps this much-needed interdisciplinary dialogue that confers the most value on the book. The book is divided into two parts. The first part, with three chapters, summa- rizes research on lexical learning and reading, especially from the fields of applied lin- guistics and SLA. Chapter 1 provides a historical and theoretical overview of incidental learning through reading. Chapter 2 explains what it means to know a word, fore- grounds research on the mechanisms underpinning lexical learning through reading, and outlines findings on input processing. Then, the chapter discusses the effective- ness of incidental learning through reading and compares it to learning yielded by meaning-given (intentional-learning) methods (e.g., provision of L1 translations or L2 definitions). The author concludes that “learning from reading may not be as effective as expected in the long term” (p. 25) and that meaning-given methods are “more effective than incidental learning” (p. 26). The third and last chapter in Part 1 delves more deeply into some linguistic and contextual factors that affect incidental vocab- ulary learning; then, it underscores the importance of strategy use in successful text comprehension and word-inferencing; finally, it reviews studies that investigate inten- tional-learning tasks conducted concurrently with or following reading tasks. Chapter 3 yields at least two important conclusions: (1) reading-strategy instruction improves text comprehension but may not have “any reliable effects on the success of word- meaning inference during reading” (p. 34); and (2), echoing a conclusion from Chap- ter 2, reading provides opportunities for lexical learning, intentional-learning tasks, especially output-production tasks performed during or after reading, significantly enhance vocabulary learning. The second part of the book, Chapters 4 to 7, draws on studies from psycholinguistic research on reading. Chapter 4 outlines several theories and models from L1 and L2 research. It explores the cognitive processes that learn- ers go through when inferring the meaning of a word in a text and introduces the cognitive model of word-meaning inference. Essentially, the model incor- porates the reading models discussed in the chapter to summarize the higher- and lower-level processes that readers go through to obtain the word-form and contextual information necessary for successful word inferencing (and thus lexical learning). The following two chapters then detail how such word- form (Chapter 5) and contextual information (Chapter 6) is utilized by readers. Finally, Chapter 7 focuses on practical ways to improve L2 reading comprehen- sion and vocabulary learning. The chapter suggests tasks that may improve word-recognition, sentence-processing, and reading-comprehension skills, all 637 of which may facilitate text comprehension and, as the author claims, also word inferencing and learning. Hamada’s book is clearly written, and each chapter is introduced with a short overview and concluded with a concise but highly useful, informative sum- mary. The book is indubitably a welcome addition to the SLA and psycholinguis- tic literature and may be an important asset for anyone seeking to better under- stand the reading process and its potential to foster incidental lexical learning. The conciseness and overall clarity of the text make it accessible to specialists and graduate students, but also to more adventurous teachers and undergrad- uate students. Additionally, in her quest to develop the cognitive model of word- meaning inference, the author covers vast theoretical ground. This provides the reader, especially the uninitiated one, with extensive knowledge of the topics covered, even though these topics are seldom discussed in depth. It is also this remarkable display of breadth of knowledge that lays the foundations for the cognitive model proposed in the book. To me, the model and its theoretical un- derpinnings are the most praiseworthy aspect of Hamada’s work. It is refresh- ing, even reassuring, to understand the need to intertwine two similar research traditions that rarely talk to each other in one, rather simple, model. Yet, such simplicity may constitute the most significant omission of the book, as the extensive theoretical background covered does not appear to come together nicely in the proposed model. I feel that the topics discussed may be too broad while lacking depth. This seems true especially in the case of the first part, which highlights research on vocabulary and reading in SLA. For instance, Hamada spends considerable time reviewing intentional-learning tasks, such as meaning-given output-production tasks, and concludes that these task types yield more lexical learning than reading alone (see above). Nevertheless, and with no clear rationale, the author neither accounts for these tasks in the model, nor does she discuss how these tasks may supplement word-learning through infer- encing. This begs the question as to why studies investigating such tasks were reviewed in the first place. The second part, focused on the psycholinguistics of reading, is significantly more relevant to the model. However, I believe that also in this case too many models and theories are introduced only to be glossed over, and it remains far from clear how they all contribute to the model. In short, I often felt that there was too much information being provided too quickly, and I looked forward to a section (maybe a chapter) that would eventually build a coherent whole, which, however, never materialized. In the end, the book ap- pears to cover too much superficial ground, which may be taxing and often mis- leading, especially to less experienced readers. I am also not convinced that the main premise of the model can withstand scrutiny. Its main assumption seems to be that by focusing on the development of 638 word-recognition, sentence-processing, and contextual-processing skills, teachers may help learners comprehend texts better, which facilitates word inferencing and thus enhances lexical learning. Although intuitive, this assumption is not sup- ported by the research findings reported in the book. On the contrary, on page 34 the author reviews a study by Kern (1989) which shows that strategy instruc- tion enhances reading comprehension but not lexical learning (see also above). A final point of contention concerns Hamada’s recommendations regard- ing proficiency levels and inferencing. As the author mentions, readers need to understand around 96-98% of the words in a text for unassisted comprehension and successful word inferencing to occur (e.g., Hu & Nation, 2000). Thus, Hamada asserts that “word-meaning inference is not recommended for begin- ning-level students” (p. 123) and claims that learners may need to know 3,000 word families before inferencing skills can be taught. Put differently, the author appears to imply that learners who know fewer than 3,000 word families are beginners who cannot benefit from word inferencing. First, these are not begin- ners, but rather learners of intermediate (B1/B2) proficiency in English (Milton & Alexiou, 2009; see also below). Second, the author seems to ignore the exist- ence of simplified texts, which occur in at least two forms. There are the short, simplified texts that are pervasive in textbooks across proficiency levels; and there are graded readers (simplified versions of unabridged books), typically used in extensive reading programs, which are discussed by the author. For in- stance, the Penguin Readers series (www.penguinreaders.co.uk) contains seven levels. Level 1 books (A1 level, as per the website) utilize the most common 500 word families; Level 7 books (B2 level) make use of about 2,500 families. In other words, comprehensible input (i.e., where 96-98% of the words are known to readers) is available at almost any level. Thus, it is unclear why the author makes such recommendations. The book also does not clarify why teaching reading skills to facilitate comprehension and lexical learning (i.e., the main premise of the model) may be more beneficial to vocabulary acquisition than, for example, the reading of simplified texts supplemented by intentional-learning tasks. Despite these critical comments, Megumi Hamada’s book is a valuable contribution to the literature. For one thing, the book provides an extensive overview of two main fields, that is, SLA and psycholinguistics, thus creating a dialogue between two complementary, yet distinct, research traditions. This di- alogue allows graduate students and researchers to broaden the scope of their knowledge and is useful to teachers, who may be willing to better understand the intricacies underpinning their day-to-day practice. Secondly, and being per- haps Hamada’s most novel addition to the literature, the book highlights the need to consider word-recognition, sentence-processing, and textual-compre- hension skills when seeking to investigate lexical learning through reading. Such 639 discussion may alert researchers to unexplored confound variables and may, hopefully, help improve the reliability of future research findings. Reviewed by Breno Silva University of Warsaw, Poland brenotesol@gmail.com References Bauer, L., & Nation, P. (1993). Word families. International Journal of Lexicogra- phy, 6, 253-279. Hu, M., & Nation, I. S. P. (2000). Vocabulary density and reading comprehension. Reading in a Foreign Language, 23(1), 403-430. Milton, J., & Alexiou, T. (2009). Vocabulary size and the Common European Frame- work of Reference for Languages. In B. Richards et al. (Eds.), Vocabulary studies in first and second language acquisition (pp. 194-211). Palgrave. Nation, I. S. P. (2006). How large a vocabulary is needed for reading and listen- ing? The Canadian Modern Language Review, 63, 59-81. Schmitt, N. (2010). Researching vocabulary: A vocabulary research manual. Pal- grave Macmillan. Webb, S., & Paribakht, T. S. (2015). What is the relationship between the lexical profile of test items and performance on a standardized English profi- ciency test. English for Specific Purposes, 38, 34-43.