SURVEY OF ENGLISH USAGE. Short form SEU. A survey founded in 1959 by Randolph QUIRK at the U. of Durham, England, and which he took with him when he moved to U. College London (UCL) in 1960. Quirk remained its director until 1981, after which his successor to the Quain professorship of English, Sidney GREENBAUM, was its director from 1983 until his death in 1996. The aim of the SEU has been to provide resources for the accurate description of the GRAMMAR available to, and used by, adult educated native speakers of BrE. It is assumed that grammarians cannot rely simply on their own knowledge to provide the grammatical data at the disposal of speakers of English in a range of stylistic contexts. The major activity of the Survey has therefore been the collection and analysis of representative samples of spoken and written/printed BrE. A subsidiary activity has been the development and administration of experiments eliciting use and judgement among native speakers. The elicitation tests have supplemented corpus data on features in divided use or found to be infrequent in the corpus, such as: the choice of the subjunctive; the should-construction; the indicative after verbs such as demand and recommend; the normal positions of different types of adverbials.

Content and analysis

The CORPUS consists of 200 samples (‘texts’), each of 5,000 words, for a total of 1m words. The TEXTS cover a wide range of subject matter, situations, and degrees of formality. The 100 spoken texts comprise: face-to-face conversations and telephone coversations; discussions; interviews; broadcast commentaries; lectures; demonstrations; sermons; committee meetings; and dictations. The 100 written/printed texts comprise: scripted material that was read aloud, such as drama and news broadcasts; manuscript material, such as social and business letters; printed publications, such as learned and popular books on various topics, fiction, news reports, and legal and administrative documents. The original Corpus is in photocopied booklets and slips, each slip (6 × 4 inches) containing 17 lines, including four lines of overlap with preceding and following slips to provide context. For each collected feature there is one slip that is marked for that item. The Survey has collected: 65 grammatical features (such as adverbs, names, negation, direct speech); over 400 specified words or phrases that were felt to have grammatical significance (such as a, is could, not, and, in spite of); for spoken texts, about 100 prosodic features (such as rising and falling tones) and paralinguistic features (such as laughing, sobbing); for written texts, all punctuation marks. Scholars from many parts of the world have visited UCL to consult the files.

Survey publications

Numerous books and articles have drawn on data from the Corpus (including its computerized spoken texts in the LONDON-LUND CORPUS) and from the results of experiments conducted at the Survey or by its associates. Prominent among them are two major reference works: A Grammar of Contemporary English (Longman, 1972) and A Comprehensive Grammar of the English Language (Longman, 1985), both by Randolph Quirk, Sidney Greenbaum, Geoffrey Leech, and Jan Svartvik. Two advanced-level textbooks based on the 1972 reference grammar are used in universities and colleges worldwide: A University Grammar of English, by Quirk and Greenbaum (Longman, 1973; US title A Concise Grammar of Contemporary English, Harcourt Brace Jovanovich, 1973) and A Communicative Grammar of English, by Leech and Svartvik (Longman, 1975, 1994). An advanced-level textbook based on the 1985 reference grammar has also been published: A Student's Grammar of the English Language, by Greenbaum and Quirk (Longman, 1990). See INTERNATIONAL CORPUS OF ENGLISH.