Linguistic Circle
Charlotte Brewer (University of Oxford), 'OED3: how a Victorian dictionary has morphed into a 21st century one'
Abstract This paper reviews the various stages of compilation, publication and revision of the OED (from 1884-2012) in order to assess its current ‘incarnation’ as OED Online (www.oed.com). Re-launched in December 2010, the website for OED has transformed the ways in which dictionary users can access the information originally lodged, somewhat inaccessibly, in 22,000 printed pages. OED Online facilitates many different types of search and analysis of this vast quantity of linguistic data and meta-data—e.g. of the number of new words entering the language (or at any rate recorded in the dictionary) in any year and how that has changed over time, of the chief quotation sources for the dictionary, of grammatical features of lexis, of vocabulary used in definitions and etymologies and the various linguistic and social assumptions such terms encode, and so on.
Some of this transformation has been bought at a high cost, however. The quarterly updating of the website with new entries, quotations, and information of all types, spanning the whole alphabet range, makes it hard for scholars to be sure they are consulting (and then citing) a stable, authoritative resource: an entry may change in major or minor ways within a few months, with previous revisions disappearing without trace (some entries, e.g. for nigger, have been revised many times). And the splicing of newly revised entries with original, unrevised ones – the latter in some cases over a hundred years old – is confusing for both scholars and the public. As of February 2012, for example, slang is still defined as ‘The special vocabulary used by any set of persons of a low or disreputable character; language of a low and vulgar type’; dialect as ‘One of the subordinate forms or varieties of a language arising from local peculiarities of vocabulary, pronunciation, and idiom… a provincial method of speech’. These demonstratively misleading definitions, in entries that look entirely the same as the most recently revised ones, date from 1911 and 1895 respectively; it is prohibitively difficult for users to work this out.
Acknowledging the outstanding achievement of the technological and lexicographical achievements to date, the paper ends with some suggestions for future directions of change.