|AUTHOR: María del Pilar Agustín Llach
TITLE: Lexical Errors and Accuracy in Foreign Language Writing
SERIES TITLE: Second Language Acquisition
PUBLISHER: Multilingual Matters
Trevor Jenkins, Freelance Communication Support Worker/Community (Sign Language) Interpreter; Computing Scientist
The study reported here concerns lexical errors (rather than grammatical or semantic errors) appearing in the writing of second language users. In this particular study the language pairing is Spanish and English. Errors are limited to misspellings, borrowings, calques, and other lexical issues. Grammatical and semantic errors were excluded from analysis. These errors are used as a metric for assessing written work of a group of language learners. Two surveys of the same group were taken 2 years apart, conducted after approximately 400 hours of tuition and then after a further 200 hours.
The book is presented in two separate sections. The first is a review of previous research into foreign language students’ lexical errors. While there appears to be a substantial body of work on the general area, the author has identified that little investigation has been conducted into the errors made by younger learners (below grade 6). The second is a practical study conducted by the author, addressing this gap in research.
The first chapter surveys the current theories of vocabulary acquisition by second language (L2) learners. A brief summary of the organisation of cognitive lexicons is given with concentration upon how this might be achieved by children. What few studies there have been of L2 acquisition are split between bilingual/home acquisition and formal tuition. However more studies have been conducted of the former than the latter despite the majority of L2 acquisition being gained in a purely taught situation.
The second chapter reviews the published literature for variables in L2 learning that have been proposed by others. A variety of situations are surveyed, not only those focusing on purely lexical issues but also on speaking or listening ability, and fluency improvement. One factor from those previous studies that the author considers at length is learner age. Adult learners are said to make fewer mistakes in their L2 usage. The author suggests that this is because their first language (L1) cognitive lexicons are firmer and more comprehensive, whereas younger learners have not established a solid lexicon for their L1. Further she identifies that little research has been conducted on L2 acquisition by children under 11 years of age even when, as is clearly the case in Spain, long-term formal tuition is given to such children. Two propositions are made concerning learner proficiency: first that errors decrease with proficiency and second that the type of errors made change with proficiency. These two notions are the foundation for the specific exemplar study reported in part two of the book.
The third chapter looks at vocabulary acquisition and its consequent use in a written productive setting. Once more the author reviews the existing literature on both L1 and L2 writing skills.
Chapter 4 concludes the first part of the text by examining types of lexical errors reported in the literature. Agustín provides a taxonomy of those reported errors.
The second part of the text covers the author’s own work, and constitutes half the text (by page count). Some of the description here provides an overview of the statistical analysis undertaken although the various results are presented in summary form.
The study is presented in sufficient detail that it could be repeated with a different language pair from Agustín’s own of Spanish and English.
Chapter 6 is the meat of the book, at over 30 pages, covering the proposition that Agustín raises in the first part that learners' lexical proficiency improves while at the same time the errors made are of a more sophisticated form. This is a longitudinal experiment with the performance of the same group of learners being studied over a two-year period. The underlying source of the errors present in the subjects’ samples are analysed and comparisons between the various kinds of errors presented and analysed.
The following chapter (7) considers whether there is a correlation between lexical errors and the quality of the written samples. As might be expected the fewer overall errors in a subject’s text then the better the quality of it. However, there are surprises with some errors (especially calques) where their occurrence is an indicator of well composed ideas and structure.
The final chapter of analysis considers the impact that receptive proficiency has upon productive proficiency. Two aspects of vocabulary development are covered, how large are subjects’ native language vocabulary and similarly how large is their corresponding second language lexicon. Agustín raises and substantiates what would appear to be obvious that the subjects cannot be expected to use a wide variety of L2 vocabulary if they do not already have a well developed L1 lexicon. She mentions the subsidiary issue that her young subjects have not yet solidified their L1 lexicon.
Agustín’s text provides a useful overview of the SLA literature that non-specialists pronouncing on the subject would do well to acquaint themselves with; for example, shortly after this book was published the British Secretary for Education (Michael Gove) suggested that children as young as 5 years old should be taught a foreign language a suggestion that goes against research reported by Agustín that adults make better progress than such young children [Wintour and Watt (2011)]. So the publication of this book is timely. It ought to have an effect on the implementation of that suggestion.
However, the book is not perfect. There are holes in the argumentation and in the coverage of alternative approaches to evaluating learners acquisition of second language skills. The author asserts, without support, that the process of writing is difficult and fails to distinguish between the cognitive formation of ideas in the writer's own mind and the physical or mechanical transcription of that on to some shareable medium as the late Russian author and Nobel Laureate for Literature Solzhenitsyn describes so eloquently in his book “The Oak and the Calf” (1975).
The author presumes that the dialect of English being evaluated is British, and does not mention the lexical differences between English and American English. The selection of this variant of English as the L2 is presumably because the teaching syllabus and examination materials are mentioned as from the University of Cambridge. In addition there is no consideration of register. One example ''My class is huge'', which she asserts is wrong could be correct for colloquial English as a statement of cohort size, which is the meaning that Agustín believes her subject intended. Similar no consideration of regional usage is mentioned. There is a parallel here with the use of a lexical sign for “business” in British Sign Language; where different regions lexical signs are treated as either WORK or BUSINESS (Dowe and Squelch, 2010).
The author does not consider the effect that mobile phones may have upon children and especially their use of “text speech”. It is unclear when the survey work was conducted, though the references would suggest it occurred sometime in the mid-/late-2000s but the increasing use of such phonetic transcriptions in written (typed/keyed) discourse may not have influenced the subjects too greatly. However, very young children are being given mobile phones and the use of these contracted orthographies may affect both their L1 and their acquired L2(s).
A glaring failing of the book is that there is little by way of comparison to L1 learners’ lexical errors other than to the learners’ (involved in the study itself) not yet having stabilised their own L1 lexicon. It remains to be seen therefore whether any conclusions drawn from this exemplar project can be applied to other pairs of L1 and L2. This would need to be addressed directly in further research.
Neither is there any discussion of the errors made by proficient L1 users. The British newspaper the Guardian is a cause célèbre and had a reputation for its spelling mistakes, such that it was and still is referred to as the Grauniad by its own readers. The problem is not limited to native English speakers in Britain as the Swedish-based online newspaper The Local publishes an English language edition, with many of its reports containing similar lexical errors to those described in this book, for example http://www.thelocal.se/36734/20111013/ .
The two languages involved in the study (L1 Spanish and L2 English) share common linguistic ancestry and therefore have relatively similar tense systems. It is not obvious whether these results would hold for pairs of languages where one has no tense system (such as Chinese or (British) Sign Language).
Although the author describes the statistical procedures used, it would have been helpful to have the source included from whichever package was used.
In any other context it would be inappropriate to consider an author's own lexical errors. However, when the subject itself is under examination, as in this text, then it becomes appropriate. One section where they obscure her own meaning is in the review of lexical errors taxonomies. (She has not been well served by her copy editor or publisher in this regard.)
In various places we see the author imposing her own synforms (a word that appears in the text and is defined as “similar lexical forms”). This may kindly be seen as the notion attributed to computing scientist Brian Randell who observed that “in English every noun can be verbed” (Jackson 1983). These synforms become an issue because they raise the reader's processing cost (Gutt 1992) or as Agustín herself says (p98) “lexical errors may greatly irritate or disturb the interlocutor.'' One such that occurs through this chapter is ''delimitation'', which appears to be a synform for ''delineation'' or the possibly more natural ''delimited''. The Oxford English Dictionary has its last quotation for this word as 1884 (OED). This word appears to be used metaphorically but the domain from which it is drawn (typically trans-national border disputes) does not truly accommodate such metaphoricalisation (Lakoff and Turner, 1989; Lakoff and Johnson, 2003; Pinker, 2008).
There are several places in the book where markup in the original typescript was not correct. Just as inappropriate lexical choices prove irritating for the reader so are these, which as we have already observed is an issue that Agustín herself raises.
The sources examined by Agustín would provide a useful introduction to the field. However, there is a preponderance of Spanish language studies amongst them. This is most obvious in the reproduction of the learner survey questions in the native Spanish. We mentioned at the beginning that the references cited in this text are primarily in Spanish. Although not a problem per se citations to multilingual documents such as those from the Council of Europe are given as if in English but then the bibliographic entry itself is presented only in Spanish. This detracts from the usefulness of the book because the reader needs to scan through the entire bibliography hunting for the relevant entry.
There is no mention of the use of corpora and their associated techniques either providing exemplars of L1 usage or of analysis of the written samples. Although the use of corpora in L2 acquisition is relatively new there is already a substantial body of literature for it to have either influenced the original study or the literature survey within part 1. A suitable introduction to the use of corpora in L2 acquisition work can be found in Tono (2009).
The author's subject group is drawn from different schools but no analysis is provided on the variability of the teaching methods employed. Teachers and teaching methods are considered to be homogeneous.
Many of the studies considered have related orthographies, that is, they are Latinate writing systems. The author has identified that the similarity of Spanish to English causes students problems when writing and specifically the phonetic nature of Spanish writing and the non-phonetic nature of English. There appears to be no mention of work with dissimilar scripts for example, English/Arabic, English/Chinese or between an L2 with an orthography and an L1 without. Examples of the latter would be the indigenous signed languages around the world which is a vital issue for the Deaf who are forced to learn the language of their usually hearing L2 teachers to transcribe their L1.
There is no index. With word processors now pervasive there is no excuse for an academic study not to have an index; even if it originated as a PhD thesis prepared for publication.
Even with the issues that this book raises and the omissions and oversights, it still provides a useful resource for students/researchers commencing work on L2 acquisition. The literature survey in the general part 1 and the taxonomical analysis of that literature in the specific part 2 make it a useful resource.
Dowe, S. and L. Squelch. (2010). The BSL interpreter: help or hindrance; benefit or barrier? Supporting Deaf People Conference (SDCP). Online conference. http://www.online-conference.net/sdp2010/programme.htm#benefit
Gutt, E. A. (1992). Relevance Theory: A Guide to Successful Communication in Translation. SIL.
Jackson, M. (1983). System Development. Prentice-Hall International.
Lakoff, G. and M. Johnson. (2003). Metaphors We Live By. Second edition. University of Chicago Press.
Lakoff, G. and M. Turner. (1989). More Than Cool Reason. University of Chicago Press.
OED. Oxford English Dictionary. Second edition, 1989; online version September 2011. http://www.oed.com/view/Entry/49404
Pinker, S. (2008). The Stuff of Thought: Language as a Window into Human Nature. Penguin Press.
Solzhenitsyn, A. (1975). ''The Oak and the Calf”. Translated from Russian by Harry Willets, Harper Colophon Books, 1981.
Tono, Y. (2009). Integrating Learner Corpus Analysis into a Probabilistic Model of Second Language Acquisition. In Paul Baker (ed) (2009) Contemporary Corpus Linguistics, London: Continuum, pp184-203.
Wintour, P. and N. Watt (2011). Michael Gove proposes teaching foreign languages from age five. Guardian. Online. Available at http://www.guardian.co.uk/politics/2011/sep/30/michael-gove-teaching-languages-conference Last accessed 30 September 2011.
ABOUT THE REVIEWER