by John Simpson
By 1910 we realised that the word blueprint might have wider applications, and its domain expanded from the world of chemistry and design into general speech—anything could be a “blueprint” if it acted as a model or a template: a blueprint for the future, a blueprint for success. That original link between language and culture had weakened.
I had returned to my quiet dictionary editing, but over the next year Richard Charkin was orchestrating raging discussions in the higher echelons of the University Press about the future of the OED, and in particular about the possibility of computerising it. The OED was at the most alarming crossroads that it had seen for around one hundred years. Some senior members of the Oxford University Press and of the University of Oxford itself regarded the dictionary as a white elephant, and one that was stifling other exciting publishing projects. If the University Press could not agree on a computerisation plan, then this could mark the end for any future development on the dictionary, which would just be left to gather dust as an increasingly forgotten pillar of Victorian endeavour. But the momentum to transfer the full text of the original OED to computer, and in parallel the full text of the Supplement to the OED, was gaining ground. A tentative plan arose: that we might be able to transfer the text of the First Edition of the OED and the text of the Supplement to the OED into two separate large files, tagged to distinguish elements such as definitions, etymologies, etc., and then we might electronically merge the two files into a single, comprehensive, and enormous dictionary file, and the whole shooting match would be published on CD (the Internet not being available when these plans were drawn up).
There was no doubt that this computerisation project was a grandiose scheme, and a very risky one—almost certain to fail, given the University Press’s lack of experience in all aspects of this completely new style of publishing. If there was any hope of it becoming accepted, then the Press needed an editor to represent the dictionary in its computerisation discussions. After a while I became aware that Ed Weiner had been asked to transfer from his editorial work in order to fill this role. At this investigative stage there was no grand announcement. In one sense I was very disappointed that Ed had been offered this work and I had been left managing the New Words group. But at the same time I knew that it wasn’t what I was best at. Although I was more than curious about the idea of computerising the text, I’d always been better suited to keeping systems running—and improving them—than to building new ones from scratch. In any case, our jobs were two halves of the same egg: Ed was edging into a project to make electronic access to the dictionary easier, and I was spearheading a parallel project to update and simplify the editorial content. Everything was linked up. It would doubtless have taken me ages to get to grips with all the intricacies of the computerisation project anyway, and I would be well advised—if I so wished—to edge my way into that side of the project slowly.
What followed was a period of extensive activity of a type unfamiliar to shy, retiring lexicographers. On the computerisation front, the University Press knew it needed active partners from the international world of software analysis and data processing to get the project off the ground. It just didn’t have that level of expertise itself. The Press encouraged Richard to explore the options thoroughly, so that it might in due course decide on whether to give us the official go-ahead.
Ed was charged with writing a description of what we wanted to do, for circulation to any technologically minded organisation that might be brave enough to be inveigled into helping. Here he struck lucky, as the old OED editors had devised the dictionary as if they were just waiting for computers to be invented, and he uncovered a formidable information structure which in time yielded relatively easily to the fielded database structure of the computer. Databases like information to come in regular packages, and so do dictionaries. Dictionaries tell you whether you are looking at a definition, a pronunciation, or an etymology by their structure and by the way the page looks (by formatting—a change of typeface, size of print, special print characters, indentation, etc.). Because all of this information is presented regularly and repetitively, it proved to be relatively easy to pour the right pieces of information (the definition, etymology, etc.) into the right database buckets. Obviously it wasn’t that easy, and we didn’t have buckets. But you get the idea.
It soon became clear that my New Words group in St Giles’ would have a major role to play in the computerisation. In one of the numerous meetings I attended as the project began, I was asked whether we might be able to contribute some new entries for the computerised version of the dictionary. This enquiry ran counter to the general ethos of the project, by which we planned simply to republish what had already been published in the past, but in a new format. But there was some sense in the request, and I realised that it was an opportunity to start getting the old dictionary more up to date. I cavalierly agreed to contribute 5,000 completely new words and senses to add to the thousands already in the OED. As it happened, many of these were the very same words and meanings which the New Words group had been working on anyway (AIDS, computer disk, disinformation, black propaganda, etc., etc., ad infinitum). Others were new ones which we discovered ourselves as we continued our searches through the word files. And so at that point the project expanded from “put the full OED and its Supplement on computer” to “put the full OED and its Supplement on computer and add another 5,000 words and meanings whether there’s really time or not.”
We had another meeting a while later at which the Shark introduced a new factor: it would apparently be beneficial if we no longer showed Sir James Murray’s old handmade pronunciation system, but replaced it with the International Phonetic Alphabet. The Shark might be a management guru, but he also had his finger on the editorial pulse and knew a good idea when he had one.
To explain: the International Phonetic Alphabet (IPA)—which is the standard and neutral way of transcribing pronunciations—was devised just too late for the OED back in the 1880s. The first IPA system was published in 1888, four years after the first instalment of the dictionary had been published. In the absence of an acceptable phonetic transcription system, Sir James Murray had made up his own. In fact, it was roughly similar to the IPA, but in some cases too complicated (making vowels carry etymological as well as phonetic information), and in other cases too simplistic. As it was now a hundred years out of date and had been adopted by nobody except us, it seemed a good plan to replace it with the current industry standard. Obsolete words were not given pronunciations by the dictionary, but that still left us with about 140,000 pronunciation transcriptions to convert from the old system to the new one. In the end, we worked out a way to make these conversions automatically, by computer—most of the time. There was a problem, though. Words have stress—we stress the word preliminary on the second syllable. Murray’s old system placed the mark indicating the syllable carrying primary stress just before the stressed vowel (i.e., prel’iminary). The new system placed it before the stressed syllable (pre’liminary). So Ed had to spend a while working out how to tell the computer where syllables began. This isn’t always easy when you are talking to a machine.
And so the project expanded yet again to “put the full OED and its Supplement on computer and add another 5,000 words and meanings whether there’s really time or not, and then make sure that all of the tens of thousands of pronunciations are converted to IPA from the OED’s own home-grown system.”
There are often good reasons for avoiding meetings if you could manage it.
Despite all of this additional work, the prospect of being able, at some point in the near future, to revise the dictionary online was extraordinarily attractive to editors such as myself. But what intrigued me and my colleagues just as much were the possibilities that would be opened up by enabling people (ourselves included) to search the entire content of the dictionary instantly for information relating to the language. Since the dawn of dictionaries, access had only been through the word you looked up. If you wa
nted to know what strategy meant, you found your dictionary and turned to the letter S, struggled to find st-, etc., and then read the brief definition which (you hoped) enlightened you as to the meaning of the word. But supposing you suddenly developed wider interests. Supposing you were interested in finding all of the words in English that ended in -ology? You had to start on page 1 of the dictionary and read every entry until you reached Z. And there was quite a good chance that you would miss one or two as your eyes glazed over or the telephone interrupted your concentration. Once you’d spent an unconscionable amount of time discovering how many words in the OED ended in -ology (the dictionary currently knows of 1,011), you might then want to compare this with the number of words ending in -ography (508 this time). So then you would have to start all over again. Obviously this was far too difficult and time-consuming for the ordinary person to do, so it wasn’t done.
Hundreds of other questions which might have been asked about the language were not asked, or were only answered falteringly by considering just a sample of the data. What if you could dream up more or less any question you wanted about the language, ask it, and receive an answer seconds later? What if you could impose a historical dimension, too, on your questions? It was a dream worth working towards, and one that to most would have seemed impossible at the time.
When we think of words that end in -ology we are actually looking at a mixed bag of words with a mixed history. Historically, -ology words derive ultimately from Greek, where logos meant “word” or “discourse,” interpreted eventually as “field of study.” The very earliest word included in the OED with this termination is the word theology, recorded from the mid-fourteenth century. But we borrowed this word in its entirety: -ology was not at that early time a suffix which could be added to other words in English—we borrowed whole words which already ended in -ology (or its equivalent in their own language). In the fourteenth century, any word with a hint of scholasticism in it tended to enter English from French, as French and Latin were the languages of informed debate in Britain at the time. So it is not surprising to find that we hired the word theology from French théologie. It seems the word was quite new to French then, and that the French had received it from the Latin theologia, which in turn borrowed it from Greek theologia.
Other early -ology borrowings were taken wholesale, directly or ultimately, from the classical languages, and not forged in English from their component parts: the words entering English (from Latin and French) before 1500 were amphibology (ambiguous discourse; a quibble), astrology, etymology (word derivation), mythology, and tropology (figurative interpretation of the Scriptures), which gives you some hint of the subjects of discussion around the High Tables of Oxford and Cambridge in those years. Later formations on this model tended to be words from the sciences (which rather liked creating words from classical elements) and philosophy, which was clearly too rarefied at the time to use plain Anglo-Saxon terms.
But what we are looking for is the turning point: when -ology first became an English word-formation element, not one just carried in as part of borrowed words of foreign extraction. Demonology (1597 in English) is a possible example. Although it might have been formed in late Latin, we don’t yet have any evidence for this earlier than the first English use, and it can therefore be posited tentatively as created in English from demon and -ology (or perhaps demono- and -logy). It is not until the late seventeenth and early eighteenth centuries that we begin to find true English creations, and these are often playful ones, such as trickology (1723) and punnology (1744).
Through the eighteenth century we started to create new -ologies for branches of study. These were mostly formed on classical elements, but in compounds unknown to Latin and Greek. France was busy creating its own -ologies, and we also borrowed some of these, not liking to see them go to waste: entomology, morphology, etc. Nowadays we still create many -ologies on classical elements, but we also allow ourselves freedom to form them from English words (Egyptology, musicology, oceanology, reflexology). Next time you come across one of these, see which of the possible routes it has taken into English.
We put the word around in Britain and internationally in our search for partners for the new dictionary project—or at least those in serious positions of power in the Press put the word around. As usual, the silence resounded throughout Oxford and more widely in Britain, but it did eventually generate some interest in North America—which was at the time considerably more wired than the British Isles. And I carried on quietly writing entries for new words.
To capitalise on the North American interest, and to investigate how it might pull together to help the computerisation project, in November 1983, Ed, the Shark, and one of the University Press’s computer managers flew out to the United States and Canada for further discussions with potential partners. We had acquired three new friends: IBM, which offered equipment and personnel; International Computaprint Corporation (ICC), in Fort Washington, Pennsylvania, which undertook to have the whole text of the multi-volume dictionary keyed on to computer; and the University of Waterloo in Ontario, Canada, which became interested in the obscure question of “what sort of electronic database could most efficiently hold the mass of structured data of which the OED (theoretically, at this time) consisted.” Unbeknown to us, while we in Oxford had been busy publishing books on paper, the University of Waterloo’s Computer Science Department had been building a worldwide reputation for handling and processing electronic data. At the same time that the language was going global, and so—of necessity—was our search for project partners. The attraction to our partners was principally the size of the OED, eventually measured at around 67 million characters when stretched end to end, which was hard to do even then. For the time, this was big data. And if we couldn’t transfer this international interest in the dictionary into something that opened up the text to the masses, then we weren’t worth the paper we were written on.
Computer wasn’t really much of a word in the past. It didn’t make any serious inroads into the history of the English language until the second half of the twentieth century, though by then it had already been around for four hundred years. In terms of its development, it’s one of those words that moves from meaning a person who does something (i.e., who computes) to meaning a device that carries out an equivalent operation (i.e., that computes)—especially with the help of new technology. Printer and scanner followed the same route, as did typewriter, which in the late nineteenth century could mean both the machine on which a person typed and the person who did the typing.
For some reason I have a soft spot for words that entered English in the Early Modern period. That’s from, say, 1500 until the early 1700s—though my personal interest often wanes as we move into the eighteenth century. Computer (and its base, the verb compute) rings all the right bells for me. Except for an odd and very Latin-based meaning from the late Middle Ages, the verb really kicks into action in 1531; in 1613, without any fanfare, the noun computer slips into the language. But the meaning back then, before we had any electronic wizardry, was quite different. A “computer” was a person who computed, or made calculations. We didn’t know that machines could do that back then. The very first recorded occurrence of the word computer doesn’t fill you with any sense that this would be one of the biggest words of the late twentieth and early twenty-first centuries. Here it is (in its old sense of “a person who computes”):
I haue read the truest computer of Times, and the best Arithmetician that euer breathed, and he reduceth thy dayes into a short number.
That’s from the snappily titled Yong Mans Gleanings Gathered Out of Diuers Most Zealous and Deuout Fathers, written by the enigmatic “R. B. Gent. [i.e., Gentleman]” in 1613. This meaning carried on uninterrupted until the middle of the nineteenth century, when people started applying the name computer to devices that performed calculations, such as slide-rules (earlier alternative terms included the obvious calculating machine and the mechanical mill). It wasn’t
until the 1940s that developments in electronics meant that we could start applying the word to the sorts of things that we nowadays think of as computers.
Once we had international partners on board, things started to look more promising, and to move forwards faster. I, too, was becoming much more involved in the computerisation project, while still maintaining my new-words work. Although Ed and I could oversee anything to do with the content and structure of the dictionary, we recognised (as did others) that we didn’t have the expertise to manage the entire project ourselves. Fortunately other people in the University Press were good at that. So at the start of 1984 the Shark appointed an OED project director, Tim Benbow (affectionately known as the Admiral), to keep us on track, on the perceptive reasoning that if you show a lexicographer a track, he (or she) will wander off it without a director to keep them on the straight and narrow. Under Tim’s management, the University Press had by the end of 1984 given official approval to our risky and radical new project.
This achievement deserves a small round of applause. I was enormously pleased when the news came down the line that the Delegates of the Press had approved the project. At last we had a budget, and soon we would (we presumed) have a detailed plan—or at least more of a plan than we had etched out to achieve the project’s approval. No one knew if this decision heralded the end of the print tradition of the OED: opinions differed, but the right answer seemed to be “wait and see, and don’t call an end to print publication straightway.” If publishers can sell you the same thing in two or three different versions, it will make them happy. At the same time I took an A-level in computer science at night school in Oxford, on the reasoning that I didn’t want to be flummoxed by technical colleagues under pressure to curb our ideas. I discovered that when a software analyst said something wasn’t doable, my job was to call their bluff and to find the loose brick in the wall of obscurity they erected to prevent us from proceeding. It was the easiest course I ever took. I knew there would be dark nights ahead, but here at last was our chance to set the dictionary on the road to recovery and a viable future.