January 1999 newsletter

Antedatings and the OED

One of the main functions of the OED, as a historical dictionary, is to record the earliest known use of a word (or sense of a word) in English. A very satisfying aspect of the current revision programme is our ability to push back the history of individual words and senses by adding to entries earlier examples than those found in OED2. We can do so because we have access to a vast number of primary and secondary sources which were not available to previous OED editors.

New primary material includes (as well as re-edited and re-dated editions of works regarded as part of the literary canon) a wealth of non-literary material from throughout the English-speaking world, such as letters, diaries, wills, inventories, and local history records. New secondary sources include major historical and regional dictionaries like the Middle English Dictionary, the Dictionary of the Older Scottish Tongue, and the various national dictionaries of the English-speaking world.

The last ten years has seen a revolution in the amount of research that editors can do at their desks by searching online databases of text. As well as corpora compiled specifically for the project, such as our own Historical Corpus and the British National Corpus, we have access to numerous databases compiled by other bodies and institutions. These include the Brown Women Writers Corpus, the Chadwyck-Healey Literature Online databases, and the Making of America. One such database, indeed one of the largest, is the American journal archive JSTOR, which Fred Shapiro of Yale University Law Library has been searching against the OED, with the impressive results described in his article below.

The proliferation of new resources means that we are now managing to antedate more than one in four senses in the OED - that is, we are discovering that over a quarter of the language has a longer history than could be demonstrated a century ago.

Some of these antedatings are dramatic. For example, a sense of open familiar to us all ('of a shop, office, etc.: ..available for business') has been taken back from a first citation of 1824 to documents from the fifteenth century, with a precursor in Old English. Many other antedatings, while less dramatic, are very significant (for example, many words first attributed to Shakespeare and other Early Modern writers can now be shown to go back to Middle English and even Old English).

Antedatings enable us not only to push back the history of words, but also to fill out the social context within which they originated. The word 'parlour', used adjectivally to designate a person who professes belief in a cause but does not actively support it (as 'parlour Bolshevik, socialist', etc.), is attributed in OED2 to radical thinkers of the early 20th century, but can now be antedated, perhaps unsurprisingly, to the 1790s and the work of Samuel Taylor Coleridge ('parlour and tavern patriots').

As well as telling us something about the social context in which words appeared, antedatings can also tell us about the habits of mind of the individuals who first used them. The adjective 'Byronian', recorded in the OED from 1822, seems in fact to have been coined (somewhat tongue-in-cheek) ten years earlier by that arch-egoist Byron himself. It appears, in his recently published letters and journals, in a letter of 1812: 'Of Common Sense, [I have seen] so little, that I mean to leave a Byronian prize at each of our "Alma Matres" for the first Discovery'.

Many other antedatings with political and social significance are mentioned by Fred Shapiro and Alan Hartley below. Perhaps one of our most satisfying recent discoveries is that 'pastrami', first recorded in the OED from 1940 in the letters of Groucho Marx, can now be antedated to 1920. For this we have to thank one of our regular American contributors, Stuart Silverstein.

This discovery serves to remind us yet again that there are perhaps no better finders of antedatings than interested and informed general readers - those non-professional lexicographers with a feel for historical language and an eye for an antedating. So do, please, send in any quotations for words or senses you come across which are earlier than OED's first usage. We are always grateful for antedatings from any verifiable source.