A Scholarly Everest Gets Bigger

By Paul Gray

There will be no greater publishing event this century than the appearance of the new OED.

-- Anthony Burgess

This eye-popping blurb -- about a dictionary, no less -- may seem a bit of a stretcher. But the Oxford English Dictionary is not just another reference book, an arcane preserve of scholars and authors, like Burgess, who use language to make their livings. Since its completion in 1928, exactly 71 years after it was proposed at a meeting of the Philological Society in London, the OED has stood as the ultimate authority on the tongue of Shakespeare and the King James Bible, not to mention the language of tradespeople and the slang of the streets. Relatively few speakers of English consulted it, to be sure; but many were reassured by the knowledge that it, an Everest of scholarship, was there.

And it is here again, updated and expanded, a mammoth historical progress report on a language with a vocabulary, the world's largest, that grows by an estimated 450 words a year. The second edition, in 20 volumes, dedicated to Queen Elizabeth II, will be officially unveiled at celebrations in London next week. It is a statistician's dream. It contains 21,728 pages and defines 616,500 words and terms, using nearly 60 million words (34% more than the first OED) to do so; it also costs $2,500.

What that money will buy is the most pertinent fact about the OED2, at least to prospective customers. In essence, the new edition collates into alphabetical order three distinct elements: 1) the first OED, largely unchanged, although some errors and lapses have been corrected; 2) the contents of the four supplements to the first edition, which appeared between 1972 and 1986; and 3) roughly 5,000 words or expressions that have gained currency since the early '70s.

These last entries are likely to attract most of the preliminary attention. The OED2 co-editors, John Simpson and Edmund Weiner, note that the generating ferment in English has shifted from the literary world toward those of science, business, medicine and North American slang. In fact, a partial listing of what the language has been up to lately is enough to inspire depression: brain-dead, nose job, right-to-die, acid rain, crack, heat-seeker, asset stripping, greenmail, petro-currency, barf, drunk tank. There is not much here that would inspire Keats to write an ode.

On the other hand, every new term, however inelegant, is given the treatment that long ago distinguished the OED from all competitors. This dictionary does not merely give etymologies, pronunciations and definitions; it also provides a word's earliest known appearance in print and uses quotations to illustrate the context in which the word has been used and all shifts of meaning to which it has been subjected. Hence AIDS (another lamentable addition to the lexicon) is defined and then traced back to its presumptive print debut, in the Morbidity & Mortality Weekly Report of Sept. 24, 1982.

Invisible behind the plethora of pages in the OED2 are hidden commands heralding an advance that will revolutionize the way it -- and perhaps all reference books -- is used in the future. Amazingly, this entire venture was conceived and completed within a span of seven years. A. Walton Litz, a professor of English at Princeton and a member of the Oxford University Press advisory council, says, "I've never been associated with a project, I've never even heard of a project, that was so incredibly complicated and that met every deadline." Some of this speed and success can be attributed to the efficient cooperation among firms in Britain, Canada and the U.S., all of whom contributed essential parts to the larger whole. But the principal reason why this edition was prepared so rapidly can be cited in a word that did not appear in the first OED: computerization.

In short, the new dictionary, all 350 million characters of it, now exists as a data base, an electronic version stored in a massive computer memory. At first glance, this may seem unremarkable; the difference between a lot of ; words on the page and on some terminal screen appears to be chiefly one of weight. But that is not the whole story. Electronic information can be made available to interested readers in a manner not possible through print. The task of devising software that would ferret out new uses for the OED2 was assigned to the University of Waterloo in Ontario. There a team of computer scientists led by Gaston Gonnet and Frank Tompa, both 40, responded with a vengeance.

They devised programs that can search through the OED2 and come up with information that would have taken weeks, years or lifetimes to assemble before. Word of this advance in data retrieval has been spreading among computer and dictionary buffs for months. Tompa has letters on his desk asking how many words entered English directly from German and how many references to the Malay language appear in the dictionary. Child's play, apparently. He is more interested in the broader possibilities. "It would be relatively straightforward," he says, "to compile dictionaries for distinct historical periods, to produce something, say, that would present only the vocabulary available to Shakespeare. The same thing could be done with reference to important legal documents, pointing out what the words of the laws actually meant at the time they were written."

Such spin-offs from the parent dictionary are, for the moment, purely speculative. Similarly, it will be at least 18 months before anyone can buy the OED2 in computer form. A laser-disc version of the first OED, however, with software less powerful than the newest Waterloo innovations, has been commercially available for the past year.

In the meantime, Tompa and his colleagues have the technology to answer some interesting questions. Given the 2,435,671 quotations included in the OED2, which single author wins the citation sweepstakes? Most people would guess Shakespeare, and they would be right: 33,150 times. But who comes second? Tompa's keyboard clicks away, and the answer soon appears: Sir Walter Scott, 16,548.

Sir Walter Scott? But of course. The bulk of the new OED retains the stamp of the age in which it was born; it remains a triumph of Victorian duty and taxonomic zeal, of a century in which Scott was one of the most popular authors writing in English. Now that the text has become electronic and easier to revise, future OEDs may lose this 19th century bias. Not too soon, though, it is to be hoped. These handsome new books, containing a trove of information ! waiting to be mined, stand solidly between the past and future. They are an inexhaustible record of what we have written and said and the foundation for what we may yet come to invent.