The Linguist

The Linguist 61,3 - June/July 2022

The Linguist is a languages magazine for professional linguists, translators, interpreters, language professionals, language teachers, trainers, students and academics with articles on translation, interpreting, business, government, technology

Issue link: https://thelinguist.uberflip.com/i/1469166

Contents of this Issue

Navigation

Page 19 of 35

20 The Linguist Vol/61 No/3 2022 thelinguist.uberflip.com FEATURES What can research tell us about vocabulary size, language proficiency and the GCSE curriculum, asks James Milton O fqual, in its recent review of research into language teaching in schools, 1 advocates for teaching a relatively small number of almost exclusively frequent words at GCSE – words from within the most frequent 2,000 lemmatised words. In accordance with this, the GCSE MFL (modern foreign languages) curriculum, which is set at A2/B1 level, now prescribes a vocabulary content that is a fraction of the size required for equivalent qualifications overseas, and for previous age-equivalent exams in the UK. What, then, does research tell us about the vocabulary needed for proficiency in a foreign language? Learners need thousands of words to become communicative. CEFR levels can be linked to normative vocabulary sizes, and vocabulary size below about 1,500 words will almost always signal that a learner is at an elementary (CEFR A1) level and effectively uncommunicative in the foreign language. 2 At the other end of the proficiency scale, Norbert Schmitt summarises research which shows that foreign language proficiency is defined, not entirely but very largely, by the learner's vocabulary size, and that 8,000- 9,000 words are needed for high levels of language performance (CEFR C2). 3 Lars Stenius Stæhr, for example, finds that vocabulary size can explain more than 50% of variance in performance in reading and writing exam scores. 4 He concludes that vocabulary size is the most important factor in determining learner proficiency. This is supported by J Charles Alderson, who asserts that language ability can be largely described as a function of vocabulary size. 5 Learning a large vocabulary is probably an essential step to learning other aspects of language, such as grammar. Learners do not learn all the words they are taught. Research suggests that learners retain, perhaps, about half of the words they are taught, when learning in good and supportive learning environments. 6 If the DfE proposes, as it does, 7 teaching only 1,200- 1,700 words in the foreign language, probably only about 600-800 words will be retained by the average learner. According to my own research, uptake of vocabulary at GCSE can fall well below that 50% figure. 8 Less frequent words are required for communication. Language is always about something. It has content, and a lexicon of frequent structure words rarely contains the specific vocabulary needed to provide this content and to talk about something, or anything. If you want to order breakfast, for example, you cannot do this with a lexicon comprising only highly frequent structural words such as 'and', 'the' and 'but'. You will also need less frequent, content vocabulary, such as 'tea', 'egg', 'toast' and 'sausage'. In well-developed frequency lists, such as those used for the Routledge Frequency Dictionaries, there is very little of this content vocabulary on any specific topic in the most frequent 2,000 words. For a learner to become expert, these less frequent content words are not optional, they are essential. High levels of coverage do not mean high levels of comprehension. A comparatively small number of highly frequent words comprise a large percentage of the words that are used in connected speech and writing. Words like articles in English ('the', 'a', 'an') are used repeatedly and might comprise 10% of any well-constructed corpus. A learner knowing 2,000 words might have 80% 'coverage' therefore, but would probably take little or no understanding from anything but the most artificially constructed text, and will struggle to communicate on any topic. Something like 100% coverage is needed for good comprehension; that is to say, in any text that a learner reads or hears they will need to know all – or almost all – the words to fully understand it. 9 Marcella Hu and Paul Nation's study of unknown vocabulary density and reading comprehension shows that where 80% of the words in a reading passage were known by testees, they did not have sufficient understanding to answer straightforward comprehension questions on the text. Even when 90% or 95% of the words were known, most participants could not gain adequate comprehension, and 98% coverage was needed for unassisted comprehension. For that to occur across a wide range of topics, about 8,000-9,000 words are needed. 10 What vocabulary should be taught? Ofsted's Curriculum Research Review (OCRR) is clear that able learners taking a GCSE language qualification should become 'expert' language users – or C2 level – but this is only possible with a vocabulary of at least 8,000 words. The review, further, gives a strong steer away from teaching the less frequent, theme- or topic-related vocabulary required for effective communication. Its claims that, by giving 80% coverage, high-frequency words are sufficient for comprehension are misleading. In a recent tweet, probably intended to endorse the When size matters

Articles in this issue

Links on this page

Archives of this issue

view archives of The Linguist - The Linguist 61,3 - June/July 2022