English corpora under Japanese eyes (review)

This collection was published to commemorate the tenth anniversary of the Japan Association for English Corpus Studies(JAECS) and contains eleven papers by members of this association, as well as an overview article by Stig Johansson (3–24), who sketches the evolution of corpus linguistics, beginning with corpora before the computer era and ending with recent multilingual developments. The papers are preceded by a foreword by the president of the JAECS and a preface by the editors. The topics discussed in this collection are heterogeneous and cover contemporary English as well as earlier varieties, English literature, and issues concerning English language teaching.

There are two contrastive studies in this volume, both based on parallel corpora: Mitsumi Uchida and Tomohiro Yanagi (27–48) examine the English passive be to construction, which is not translated with its straight forward French être à counterpart due to the latter’s structural ambiguity with respect to voice. Makoto Shimizu and Masaki Murata (71–91) confirm their earlier findings that English reflexives generally do not correspond to Japanese reflexives.

Mayumi Nishibu (49–69) investigates definite notional subjects in existential there constructions. She distinguishes various uses of the definite article (with cataphoric use being the most frequent), examines the discourse functions of these constructions, and also lists those nouns that tend to occur in the definite NPs.

Makimi Kimura (93–113) compares two seemingly synonymous words (magnate and the Japanese loanword tycoon) and discovers a number of differences that she suggests should be used to update the entries in OED2. Satoko Takami (115–35) uses statistical methods to identify adjectives that occur with significantly different frequencies in tabloid and broadsheet newspapers and describes the morphological and semantic characteristics of these two groups of words.

Yoshiyuki Nakao, Akiyuki Jimura, and Masatsugu Matsuo’s report (139–50) deals with computer-assisted philology rather than corpus linguistics proper: they are in the process of building a collation of two machine-readable Chaucer manuscripts, which they use here to discover dissimilarities in spelling. Ohkado Masayuki (151–68) takes a generative approach and discards the assumption that there is an independent leftward verb movement operation in Old English subordinate clauses.

Satoru Tsukamoto (169–84) describes a method for dating texts based on the frequencies of a range of text-internal syntactic features that are characteristic of certain periods of time. Shin’ichiro Ishikawa (187–212) analyzes basic color terms in the novels of D. H. Lawrence and finds out that black, white, blue, and grey play a prominent role in his works. A quantitative and qualitative study of the cooccurring words reveals some of the implications these preferences may have.

The last two studies deal with English as a foreign language. Tomoko Kaneko (215–30) investigates the (in)correct use of past-tense forms in an error-tagged corpus containing spoken material by Japanese learners of English. Kiyomi Chujo (231–49) assesses the vocabulary levels of various English textbooks and tests by comparing them with a lemmatized frequency list from the British National Corpus. She finds out that there is a considerable gap between the vocabulary level of high-school textbooks and the vocabulary required for passing university entrance exams or certain proficiency tests in Japan.

This volume offers insight into the varied ways English corpora are employed in Japan, but it also shows that ‘Japanese eyes’ are not much different from other people’s.