The Oxford English Dictionary
Promised in two years, the first instalment arrived twenty-four years later and got as far as the word ant.

In May 1860 Herbert Coleridge, editor of a dictionary that did not yet exist, wrote an open letter about its progress. Barring some unforeseen accident, he said, “in about two years we shall be able to give our first number to the world.” The first number appeared in 1884. It ran to 352 pages and sold for twelve shillings and sixpence. It reached the word ant. The final instalment was published on 19 April 1928, sixty-eight years after Coleridge's estimate and sixty-seven after his own death at thirty.
Richard Chenevix Trench, then Dean of Westminster, had argued the idea behind that delay three years before Coleridge wrote. On the evenings of 5 and 19 November 1857 he read two papers to the Philological Society in London, setting out seven deficiencies in the dictionaries England already had. They admitted obsolete words by no consistent rule and left whole families of related words incomplete, and they failed to mark either the moment a word entered the language or the earlier senses it had since abandoned. Beneath all seven sat a single claim about what a dictionary is. Every difficulty of the next seventy years descends from it.
A Dictionary, then, according to that idea of it which seems to me alone capable of being logically maintained, is an of the language: much more indeed, but this primarily, and with this only at present we will deal. It is no task of the maker of it to select the good words of a language. If he fancies that it is so, and begins to pick and choose, to leave this and to take that, he will at once go astray.
The business which he has undertaken is to collect and arrange all the words, whether good or bad, whether they commend themselves to his judgment or otherwise, which, with certain exceptions hereafter to be specified, those writing in the language have employed. He is an historian of it, not a critic. The , on which so much, on which nearly everything in style depends, is a matter with which he has no concern.
There is a constant confusion here in men's minds. They conceive of a Dictionary as though it had this function, to be a standard of the language; and the pretensions to be this which sets up, may have helped on this confusion. It is nothing of the kind.
The lexicographer is making an inventory; that is his business; he may think of this article which he inserts in his catalogue, that it had better be consigned to the with all speed, or of the other, that it only met its when it was so consigned long ago; but his task is to make his inventory complete.
Richard Chenevix Trench, On Some Deficiencies in our English Dictionaries, papers read to the Philological Society, 5 and 19 November 1857 (second edition, 1860), pp. 4–5.
The Philological Society now had to turn that principle into a method, and the method created a labour problem no salaried staff could solve. An historian needs evidence, which here meant reading the whole of English literature, so the reading was given away. Volunteers were asked for in 1857, and the printed Proposal of 1859 went further, calling on Englishmen to “come forward and write their own Dictionary for themselves.” That August George Marsh issued a separate appeal to Americans, reserving the entire eighteenth century for their attention on the reasonable ground that older books were hard to come by outside Britain. When Oxford University Press took the project on and appointed James Murray editor in the spring of 1879, he asked within weeks for a thousand readers to give him three years. Every quotation came back on a slip six inches by four, carrying the sentence and its source, and the slips were filed under the word the reader had noticed. Five million of them were gathered for the first edition.
Murray was a Scottish borderer who had left school at fourteen and taught himself his languages, and he called the two corrugated-iron sheds in which the work was done the Scriptorium. The first stood at Mill Hill School, where he taught until 1885. The second went up in the garden of 78 Banbury Road in Oxford, lined with bookshelves and 1,029 pigeon-holes, and sunk into the ground so that it would not spoil the view of the next-door neighbour, the jurist A. V. Dicey. It was damp for the rest of its life and bitterly cold in winter, and Murray worked in it wearing an overcoat with his feet in a box against the draught. A visitor in 1882 found the incoming letters bound into volumes the thickness of a hand and counted more than five thousand of them, which was no surprise given that Murray was taking in thirty or forty a day. The Post Office eventually installed a pillar box outside the house for his convenience, and it is still in service.
He described the day's post himself in a lecture in 1910. “I write to the Director of the Royal Botanic Gardens at Kew about the first record of the name of an exotic plant; to a quay-side merchant at Newcastle about the keels on the Tyne; to a Jesuit father on a point of Roman Catholic Divinity; to the India Office about a letter of the year 1620 containing the first mention of Punch; to Lord Tennyson to ask where he got the word balm-cricket and what he meant by it.” A lexicographer who wants to be accurate, he concluded, has to be a universal enquirer about everything under the Sun, and over it.
Trench's rule demanded completeness, which forced the harder question of where the language stops. Murray answered it in the General Explanations at the front of the first volume, where he likened the vocabulary to a nebula with “a clear and unmistakable nucleus” shading off through zones of decreasing brightness “to a dim marginal film that seems to end nowhere.” Then the line everyone quotes: “the circle of the English language has a well-defined centre but no discernible circumference.” He drew a boundary regardless, since practical utility has bounds, admitting the common words of literature and conversation plus whatever scientific, technical, slang, dialectal and foreign words were moving toward common use, and stating plainly that the line would satisfy none of his critics. His reason for expecting that is the most generous sentence in Victorian lexicography: “no one man's English is all English.”
W. C. Minor, an American army surgeon confined at Broadmoor for a killing committed in Lambeth in 1872, was among the most productive of those volunteers, sending in thousands of quotations from his rooms in the asylum. His is also the story to handle most carefully. In the popular telling Murray travelled down to thank a man he took for the institution's physician and learned the truth only on the doorstep. Murray's own successors at the dictionary rebutted that account, and he had almost certainly known for some time before the two men met in January 1891. The visits then continued for twenty years. The collaboration is documented in the archive. The moment of discovery was supplied afterwards.
Murray died in 1915 at seventy-eight, having delivered on time a double section running from Trink to Turndown, and having edited more than half the English vocabulary: all of A to D, H to K, O to P, and nearly all of T. William Craigie and Charles Onions finished the remaining thirteen years' work. Bound in ten volumes, the first edition ran to 15,487 pages, defined 414,825 words, and printed 1,827,306 quotations. The celebration dinner at Goldsmiths' Hall on 6 June 1928 ran to nine courses and seven wines, from an 1896 Croft's to a 1907 Château Margaux, and Stanley Baldwin's speech went out live on the BBC. The Worshipful Company of Goldsmiths did not admit women to the hall, so the Press arranged for a few to sit in the minstrels' gallery above and look down on the dinner. Two of those offered a place up there were Rosfrith Murray, more than twenty years an editorial assistant, and Eleanor Bradley, on the staff since 1897. The word used for the arrangement in the Press's internal memos was skied. In that sense it is not in the dictionary.
The dictionary has never been finished. The reissue of 1933 put The Oxford English Dictionary on the spine for good and added a volume of material the first edition had no room for. A four-volume Supplement then ran from 1972 to 1986, a second edition folded the two together in 1989, and the first full revision began with the online edition in 2000 and is still under way, with some seventy editors working through the alphabet again. An OED entry is a dated evidence file, so the useful question to ask of any word is when it started meaning this, and who can be shown using it that way. Trench asked the Philological Society to make that question the trade's entire business, and a reader who carries it out of the dictionary has taken the better half of what is in there.