On 29 March a large number of lexicographers and other drudges met at Claridge’s to celebrate the publication of the second edition of what was once the New English Dictionary on Historical Principles.* The occasion was properly festive, indeed a super-bash for the super-book. Before lunch the men who put it together in only five years told us how they did it. A wizard from IBM said how thrilled they were to have made the whole thing possible by supplying the computer, the computer know-how – and a lot of computer-generated money. Representing the constituency of the users, Malcolm Bradbury, gracefully grateful on behalf of the writing profession for so writer-friendly a dictionary, offered some innocent jests about bazooms – the last word actually defined in Volume One (A-Bazouki), though, as various persons complained from the floor, lacking any quotation from the Sun. (We were, however, assured by the editors that this paper does get into the great work under page-three girl.) Along the way Bradbury, though seeming very relaxed, dropped a reference to a work of his own, and also a parapraxis (‘my own mother, Rates of Exchange’) which Christopher Ricks, come all the way from Boston to perform an unjetlagged, nay virtuoso lexical dance before the ark, dexterously picked up and developed in a perfectly smooth improvisation.

It emerged that everybody loves the OED, not always as devotedly as Anthony Burgess, who was absent but often mentioned, but with a decent fervour all the same. Certainly all the speakers loved it, not least Daniel Boorstin, ex-Librarian of Congress, who came right out with it, in so many words. Could they have felt this way about a non-typographical database? Were we not loving, as it were by proxy, the dear old hot-metal version? I speculated that there would never be a third edition in book form, since soon we shall be able to access that ever-expanding database without recourse to printed pages. But an OUP man told me I was quite wrong: a conventional 21st-century third edition is already planned – books, he seemed sure, will continue to have their uses as information-retrieval systems, and people will find it easier to browse in them on a VDU screen. After lunch a speech of orotund whimsicality by Lord Jenkins sent the amorous symposiasts titubant into Brook Street, happy to defer for a while any serious consideration of the achievement they had been celebrating.

The historical principle still obtains, though no longer advertised on the title-page; and it calls for quotations to illustrate usage. By now nearly two and a half million are needed, all necessarily provided by human readers – the late Marghanita Laski, who single-handed supplied a quarter of a million, was properly lamented and celebrated. But without the computer the job might have taken fifty rather than five years. The editors had to combine the old dictionary, which reached its final form in 1933, with the four-volume supplement (1972-86), and then add about five thousand recently accumulated words or new senses of words. The computer has changed the typography, which looks good, though a keen eye might miss some of the beauties of hot metal.

Murray’s system of registering pronunciation has been replaced by the International Phonetic Alphabet. Foreign alphabets, except Greek, are now transliterated. These and other changes are explained in the scrupulous Introduction. Altogether the number of words has increased by 15 per cent, the size of the book by 34 per cent. I have, on a first cursory inspection, spotted only one error: a sort of triangle appears instead of an ‘A’ in Acold.

The whole story is of a century of hard labour, from the palaeotechnological ingenuities of Murray – perfectly recaptured in his granddaughter’s Caught in the Web of Words – to the machine-readable master-copy of today, with the whole invisible computer-speak book underlying the one you see. It is also a story of unforeseen delays and unexpected expenses. Despite all the volunteer labour and all the subventions, such a work couldn’t be expected to pay its way except (and it’s now a big exception) by providing the basis of profitable spin-offs. It seems an entirely deserved piece of luck that what began as a disinterested scholarly donation to the world should, because of the subsequent post-imperial imperialist history of the English language, become an asset to its publishers as well.

At £1500 the whole thing isn’t, compared with this and that – an academic monograph or a stall at Covent Garden – really expensive, but reviewers were understandably sent only the first volume. The original work for A-Bazouki was done a century ago, when the techniques of definition and arrangement were still being developed. (Years later the horrible word set, which occupies 18 large pages, still cost forty days’ work.) Yet A is little changed, except that it now additionally designates a first-class road, a blood type, a paper size, a record side and a social class (though this last usage dates from Charles Booth’s Labour and Life of the People of 1889, a moment too late for Murray and his readers), plus a whole string of modern abbreviations and acronyms (a word first recorded in America in 1943). Nine columns pass before we hit a as indefinite article and find the revisers content to meddle very little with the old account.

Since history still rules, words like abalienate and abandum, probably unused for centuries, remain, neighboured now by abalone and abangan (Syncretistic or nominal Muslims of Indonesia). The old book has algorism, dismissing algorithm as a learned confusion of this word with Greek arithmos, ‘a number’; the second edition repeats the explanation and the examples, but also gives algorithm as a word in its own right, distinguishing its use as a learned barbarism (= algorism) from the modern use (‘a process, or set of rules, usually one expressed in algebraic notation, now used esp. in computing, machine translation and linguistics’) – essential, in fact, to the production of this dictionary. There is an example from 1938, but the first true modern usage comes from Emile Delavenay’s Introduction to Machine Translation of 1960. (The Dictionary, having no space for gossip, cannot be expected to add that the same author wrote two books, one enormous, about D.H. Lawrence, and also fathered Claire Tomalin.) Since 1968, algorithm has also had a medical sense meaning a sort of diagnostic flow-chart. There is no mention here of Lacanian psychoanalytic theory, in which the word is famously used or abused. How we did without it for so long is a mystery. Anybody who has made a complicated timetable or syllabus presumably has an algorithm to his/her credit. (Which reminds me that there was some frivolous discussion at Claridge’s about the proper pronunciation of s/he.)

Alienation required expansion because of Marx and Brecht. It struck me as odd that in defining the sense of allay (as affected by ‘alloy’) nobody took advantage of the Donne revival to insert from ‘The Extasy’ an example clearer than any given (‘nor are dross to us, but allay’ – speaking of the effect of bodies on souls). Among words which might have been expected to occur in the old dictionary but appear for the first time here I note allotrope: recorded in a rival dictionary in 1889, it turns up trendily in the Listener of 1968, and with a rarefied allusiveness in the TLS for 1978, where it is written that Jonathan Miller has a variety of allotropes.

The big problems tend to come with the little words, such as all – eight and a half columns in the old text, more than ten in the new; but the old was good enough to provide the substance of the new. I was looking forward to any, but it occupies less than three columns as against the original two, the most interesting addition being the American usage meaning ‘at all’ – the examples go back as far 1735: ‘I have not traveled any this day.’ At, another challenge, has eight columns, no increase on the first attempt, indeed the magnificent definition is for the most part simply repeated. You could say of Murray, adapting what somebody said about a Procurator of Judea, that his little finger was thicker than other lexicographers’ loins. However, he didn’t know about A-OK, an astronautism (neologism, see OED iii) of 1961, and, more reprehensibly, missed astronaut, which fancifully occurred in something published in 1880.

My sample ran out, to my regret, before reaching by. The final entry, for bazouki, simply refers me forward to bouzouki, which I have not got; the penultimate word, as Malcolm Bradbury remarked, is bazoom, a vulgarism unknown to Murray. Before that there is bazooka, also unknown in those more innocent times. Bazoo, however, a sort of trumpet, was around in 1877, but Murray missed it. However, he has bazil, an obsolete form of ‘basil’, and with that word the concord-by-volume between the editions ends. The 1933 version managed to get the whole of B into the first volume.

The definition of Aufklärung is inadequate and that of Enlightenment, at which I peeked in the full set on display at Claridge’s, remains absurd. But the whole thing is virtually beyond criticism.

Once, when young, I visited Sir William Craigie at his Norwegian house in Oxfordshire. He was very old but still as precise as he must have been when doing this, that and the other dictionary – of English, of American English or of the Older Scottish Tongue. He apologised for quoting so many words beginning with Q, explaining that the lecture he was recalling had been written while, some thirty years earlier, he was compiling that letter for the Oxford dictionary. Later, as dusk fell on the summer garden, he bade me be silent and await the arrival of the fairies. These, he explained, were not to be thought of as resembling Cobweb, Mustardseed – Shakespearian diminutives. On the contrary, they were large fairies.

Familiar from my youth with large Manx fairies, bearing such names as ‘bugganes’ and ‘phynodderees’, I remained calm. But nothing happened, perhaps because I am as useless at fairy time as I would be at computing a dictionary; and anyway, he may have had in mind all those volunteer readers and their helpful slips, for large fairies, if, unlike bugganes, benevolent, sometimes do the housework.

The Random House Dictionary , which turns up at the same time as the second edition of the OED, is in a single volume, but a volume that weighs ten pounds. It is clearly meant to be placed on the kind of stand (normally occupied by Webster) that one sees in the studies of educated American households. It cost £4m. to assemble, and has 315,000 entries, about half as many as the new OED. It calls itself ‘unabridged’, which can only be an inexact description, merely meaning ‘pretty large’. It has about 2500 treble-column pages, plus a world atlas, a list of words commonly misspelled or confused, small French, Spanish, Italian and German dictionaries, the Declaration of Independence, and much other useful information, about, for example, wedding anniversaries, the classification of locomotives, volcanoes, the human skeleton, weights and measures, national parks, waterfalls of the world, signs of the zodiac and foreign alphabets. Like other American dictionaries, it continually aspires to the condition of an encyclopedia.

A few Random House comparisons with OED: A is much more rapidly despatched, abalienate and abangan are understandably absent, algorism and algorithm and all are more tersely defined, alienation misses the Marxist and Brechtian senses. The connection between allay and alloy is not made. Jonathan Miller is completely absent from the allotrope entry. Oddly enough, the specifically American usage is not recorded under any. Perhaps Murray’s gentle complaint that studious Americans, so informative about English English, seemed less interested in their own, is still valid. I find bousouki, defined as ‘bouzouki’, conjecturally derived from a Turkish word meaning ‘broken, ruined or depraved’, or from another Turkish word meaning ‘constricted or fretted’. Aufklärung is just Enlightenment, but Enlightenment gets a more rational definition than the wicked and aberrant one still offered by OED.

Send Letters To:

The Editor
London Review of Books,
28 Little Russell Street
London, WC1A 2HN


Please include name, address, and a telephone number.


Vol. 11 No. 13 · 6 July 1989

Would you be good enough to tell Mr Minter (Letters, 1 June) that it was Rehoboam not Jeroboam whose little finger would be thicker than his father’s loins?

Kiffin Rockwell
Marin, North Carolina

Vol. 11 No. 11 · 1 June 1989

Would you be good enough to tell Professor Kermode that it was not Pontius Pilate but Jeroboam whose little finger would be thicker than his father Solomon’s loins (1 Kings 12.10)?

R.A. Minter

Vol. 11 No. 12 · 22 June 1989

Frank Kermode (LRB, 20 April) says that the Second Edition of the Oxford English Dictionary (which, incidentally, reached us here before its publication-bash at Claridge’s) goes back only to the Listener of 1968 for the trendy use of allotrope. Perhaps it would be of interest to note that D.H. Lawrence made a more literary use of allotropic in his famous letter of 1914 to Edward Garnett: ‘There is another ego, according to whose action the individual is unrecognisable, and passes through, as it were, allotropic states which it needs as a deeper sense than any we’ve been used to exercise to discover are states of the same radically-unchanged element.’

A. Banerjee
Kobe College, Nishinomiya, Japan

send letters to

The Editor
London Review of Books
28 Little Russell Street
London, WC1A 2HN


Please include name, address and a telephone number

Read anywhere with the London Review of Books app, available now from the App Store for Apple devices, Google Play for Android devices and Amazon for your Kindle Fire.

Sign up to our newsletter

For highlights from the latest issue, our archive and the blog, as well as news, events and exclusive promotions.

Newsletter Preferences