The Word Detective

by John Simpson

  All of this work required money. There was no extra money, however politely we asked. Then, to our surprise, our luck changed. As a result of arguments initiated by us, but eventually rumbling through the highest quarters of the University Press, we heard that we would receive a considerable enhancement of our annual budget to relaunch the OED Online in 2010 with much-improved functionality. At last everyone realised that ten years was too long for a website to remain unchanged in a rapidly changing digital environment. So then we just had to work out in detail what “improved functionality” might mean.

  It is never easy to shift people’s perceptions and expectations, but we wanted to re-envision the dictionary so that people could experience both word-based results to their enquiries and visualisations and animations. I’m not always the sharpest knife when I come to new ideas, but I remember becoming increasingly frustrated back in 2009, as we were discussing how a relaunched OED Online might look. Filtering was fine, but it still resulted in screenfuls of word-based output, leaving the user to do all the brainwork of analysis (which was itself often rather a dangerous thing to allow some users to do, as I had previously discovered).

  I was bold enough, at one of our terminally tedious meetings on the relaunch, to propose that we offer users visualisations of the results of searches alongside the traditional verbose, wordy views. I reasoned that traditional dictionary users were only familiar with word lookups, and didn’t appreciate the wealth of underlying information that was stored in the OED. If we could redirect the results of searches into timelines, graphs, animations, and the like, then perhaps some of these users would discover that there was more to their language than they had anticipated. I knew the lexicographers would be in favour of this, as some of us had pondered about the possibility already. I discussed these ideas with my colleague James McCracken, who was just the person to lead the computational experimentation.

  Meetings at OUP were polite. If you suggested something outrageous you would be told it was an interesting idea, but implicitly you would come to realise that, for whatever reason, its time had not yet drawn into the station. We had all experienced that many times before. But to give our chairman (the current successor to the Admiral) his due, he only gave me 50 percent of the normal response, and with a quizzical air wondered aloud if I might like to make a more formal presentation of my ideas at the next meeting.

  There wasn’t much time to put together a comprehensive document based on what I had learnt several years earlier was called “blue-sky thinking,” as we had our regular word deadlines to hit for publication. But I sensed that this opening was important, so I started jotting down a few ideas. Soon they had marshalled themselves into some sort of order. The basic idea was hardly ground-breaking in a real-world context, but it was enough of a divergence from normal dictionary thinking to masquerade as radical. Instead of outputting only streams of words to users, we would offer an alternative view by which the data was re-expressed in graphs, pie charts, timelines, and any other visual medium we could think of. (I don’t think I dared suggest sounds and colour, but they became part of the package.) In some ways this was returning to the discussions I’d had with Jeffery Triggs, when he was designing the prototype OED Online in the mid-1990s. We had tried to link entries together using whatever means we might have to hand—and the principal one now was a Historical Thesaurus of the OED, the result of a remarkable forty-year research project at the University of Glasgow, which we were engaged in binding into the main online dictionary. If we could achieve this, then we could actively start to promote the OED as an accessible repository for networks of words, not just of words existing purely in isolation.

  The Historical Thesaurus of the OED, the brainchild of Professor Michael Samuels at Glasgow, began its association with the dictionary in around 1965. Like the OED, it was a mad idea that worked. The idea itself was that most of the words and meanings in the OED would be written out on index cards, then sorted so that all the terms that had ever been used for the concept of, say, “an instrument for poking a fire” would appear side by side, in the order in which they were first found in English (purr, 1357/8; fire-purr, a1451; fire pike, 1483; poker, 1534; fire-pote, 1638; pote, 1638; teaser, 1839; kennedy, 1864; tickler, 1881; curate, 1891; fire stick, 1896). The same procedure was followed for all the other concepts in English (love, hate, continuous stationery, frogs) until eventually an enormous reference resource of links and cross-references existed which allowed you to skip around the main OED historically and semantically in quite new ways. The original OED took forty-four years to move from A to Z; not surprisingly, the work on the Glasgow Historical Thesaurus of the OED also took forty-four years from aardvark to zebu, or whatever the relevant concepts were. Nowadays, if you look up a meaning in the OED, you are likely to be offered a button taking you through to all the other words for the same concept, in chronological order of their appearance in English.

  By the time we arrived at our next online-planning meeting, I had primed a few likely types to support the plan to introduce visualisations into our search results, and we received a partial acceptance that we might perhaps investigate the possibility. This was as much as James needed, and he soon had a new working prototype in place which incorporated many of our ideas, and indeed several which sadly didn’t make it through to the final live version. We had charts showing, for instance, the gradual advent of Japanese and Chinese words into English from the late sixteenth century—it was so much easier to see the outline of the story from a picture.

  These charts made for a telling comparison. English traders and early tourists noted the Chinese litchi fruit, and as more Chinese words entered English, climbing steadily (rather than interruptedly, as in Japanese) to a peak in the late nineteenth century, we find many food terms (bok choy, chop suey, oolong, etc.), and music (san hsien, se, ti-tzu), but less “art” (decorative or martial) than in the equivalent Japanese set. The profiles of borrowings from these two Eastern languages are similar but distinct. As in the old and supposedly Chinese proverb, one picture is worth a thousand words.

  In fact, for any search you carried out on the dictionary data, you had the option of receiving the results either as a list or as a timeline visualisation. A clever short visual animation prepared by James and showing words entering English from 1150 to the present day on a world map, according to their country of origin, later provided a stimulating ride-through experience documenting on-screen the emergence of English as a world language: whenever I give a talk about the OED this is always one of the sections people want to discuss further. If only we could do that for French, or German, or Spanish, too. As usual, you had to view the visualisations with some caution, as there were often outlying results which had to be excluded for one reason or another. But the basic package was in keeping with the direction in which I wanted the dictionary to be moving. It was only a start, but the first step, as they say—to quote another trite proverb—is the most difficult.

  Offering visualisations was one way of engaging with a new, modern audience for the dictionary. But users were still fascinated by the idea that they might contribute to the dictionary, in much the same way in which thousands of people had contributed to Sir James Murray’s original OED back in the nineteenth century. This desire takes us to wikis, by way of crowdsourcing—harnessing the goodwill of people internationally to contribute to the story of English captured by the OED.

  We’ve made some advances in crowdsourcing since the early days of the first OED when readers in the nineteenth century sent in handwritten slips, but not enough. We are still presenting information to users without giving them the opportunity to make significant, wholesale contributions to the dictionary—for example, through wikis and other massive data-collection projects. It’s still something we need to work on, but we’ve travelled a little down that road.

  Around 2005, the dictionary formed a collaborative relationship with Jeff Prucher, an American author interested in improving the documentatio
n of the OED’s collection of science-fiction words. The result, apart from Jeff’s book, Brave New Words: The Oxford Dictionary of Science Fiction, was a website in the form of a wiki where readers could read details of the OED’s current coverage of the vocabulary of science fiction and enter their own findings (which were typically earlier references to specific terms: alien, earthling, space shuttle, etc.). Here’s an example. The OED includes the noun and the adjective anti-gravity from the 1940s (for example, Arthur C. Clarke, in Across the Sea of Stars). If you know about early science fiction, you know that the 1932 spring edition of Wonder Stories Quarterly contains a use of the term by James Morgan Walsh: “But the antigravity apparatus will have to be capable of generating a greater repulsive force than is required for ordinary interplanetary conditions.” And then if you know more about early science fiction than the OED editors, you’ll know to look out for the equivalent term counter-gravity (not yet in the dictionary, but with evidence waiting on the wiki site from 1950). The OED editors were then able to turn to this ground-breaking intergalactic resource when updating their terms. Yet again, it reinforced the lexicographer’s maxim that there are always people outside the dictionary offices who know more about any particular word than do the dictionary’s editors themselves. By sharing this knowledge, we have been able to improve the coverage of science-fiction vocabulary astronomically.

  This science-fiction pilot study suggested that there were at least two directions in which the dictionary might proceed. Firstly, we could establish wikis on a similar or enhanced model for a wide range of subject areas, collecting information from many pockets of interested users; secondly, we could open up a parallel OED in the form of a wiki where anyone interested would be able to suggest improvements to the editors over the whole range of vocabulary. I don’t think we have ever suggested going quite as far as Wikipedia and leaving the final product theoretically in the hands of others.

  Alongside all of these plans for the future, we still needed to keep the editorial chariot on track, maintaining our production targets and publishing more and more of the dictionary online at each of our quarterly website updates. We were progressing well through the alphabet by now, and we were producing remarkable entries, full of new and exciting information. Wherever the dictionary was going in the future, we knew we’d brought it through.

  THIRTEEN

  Becoming the Past

  By 2013 it turned out that, without really intending to, I’d been at the OED for over thirty-five years. In retrospect the time seemed to have flashed past in an instant, but that didn’t erase the fact that it had passed. Employment law was changing, and it was becoming possible to stay on beyond the grave, but that idea had never appealed to me. Ever since I had met that dapper military personnel director when I was first interviewed for a job on the dictionary back in the 1970s, I had felt we shared an unspoken agreement that when I arrived at sixty or so I would leave the building. I don’t think I’d told anyone else, and the Colonel had left decades earlier, but one of us still knew.

  There’s no best time to step off the OED roundabout, so I concluded that there was no worst time either. One of my earliest wishes, when we came to updating the big dictionary, was to ensure that, by the time I departed, the project should be far enough advanced not to be cancelled. It seemed to me now (pending a disaster) that it was as safe as it could be. We still had about 60 percent of the way to go to Z, but the direction of travel was firmly in place. I wouldn’t see the update through to its completion, but I’d come to terms with that some years back. Even Sir James Murray had not taken the dictionary through to Z himself: he had died while the original dictionary was working through the letter T in 1915.

  It seemed to me that the OED was entering a period of consolidation—without any radical changes on the horizon—and so it might be a good time to give someone else a chance to step into my editorial shoes. I hope I left it in good shape, and I certainly wish the new chief editor, Michael Proffitt, all the best as he and his colleagues take the dictionary forward into the future. I had never found it a problem to maintain my enthusiasm for the OED project, and so leaving it was in some ways hard. But when you’ve been doing something for over thirty-five years nothing’s too hard. You make the decision, and then you stick with it. There would be plenty of others things I could find to keep myself occupied.

  Words can change to the extent that we no longer recognise their original component parts. In the case of enthusiasm, there will be few people who can see the Latin or Greek words for “god” hidden in there, but in the old days that semantic image was what drove the word. Enthusiasm looks, on the face of it, like an unholy scramble of letters, which suggests it derives from Greek. According to the OED, there are 144 words in English that end in -asm, and most of them involve chasms, clasms, plasms, and spasms. There are no prizes for guessing that those words all stretch back to Greek, too. Enthusiasm is not medieval in origin, as you might think, but entered English in Shakespeare’s time, around 1600. The original senses of the word were rather negative. The nub of it is hidden in the first couple of syllables: enthus- means “possessed by a god” (there’s a theos, “god,” as in theology, behind the -thus-). So when the word was first used in English, enthusiasm meant “possession by a god,” or at least acting as if you were in a state approaching divine madness or frenzy. As the years passed (1660), enthusiasm came to mean any misdirected religious fervour: not a frenzy that seemed to take you over, but one that you intentionally adopted. By the early eighteenth century (1716), enthusiasm had started to drop its religious colouring, and to develop its modern meaning of rapturous intensity or passionate eagerness. Once we got there, we were so excited that it hasn’t really changed its meaning since then.

  I’m not allowed to criticise words, but perhaps by now I can be excused for saying that the barbaric enthuse dates from the early nineteenth century in the United States, and has sadly been going strong ever since. I’ve used it myself earlier in this book, just to show that I bear it no ill will.

  I do like a modicum of enthusiasm over words, but not too much. I come from a generation and a society where over-enthusiasm was deplored, and keenness was deprecated. Nonchalant, non-interventionist observation was the order of the day when I was growing up, and the perspective stuck. It turned out that that world-view worked well for historical lexicography, too—though it’s obviously not the only one that would. I found that it was easy for me to stand on a still point watching words stream past in all directions, just noting patterns, changes, peaks and troughs, accidents.

  If this propensity to monitor and report attracted me to the OED, I now realise that it’s what attracts me to other things I’ve become involved with over the years, whether it’s literary research, local history, or genealogy, and especially if it involves revealing long-standing misconceptions.

  These misconceptions are never far from the surface. After my father died, I inherited a large Victorian oil painting of an Italian lake scene, which had originally been bought by my great-grandfather. It was clearly very competently done, and the artist had made a sterling effort to catch the colour of the light playing around the old houses on an island in the lake, which formed the centrepiece of the picture. The painting is entitled Isola Bella, Lago Maggiore, in black lettering, on the frame. So Hilary and I thought it would be worthwhile using the scene as an excuse to go there—at which time I could confirm the historical details, and perhaps write up some notes on the place. Hilary had her annual misgivings over whether this technically constituted a holiday.

  We scanned the maps and found Isola Bella, one of three little islands along the side of Lake Maggiore, and we booked our tickets. It would have been useful to have had our elder daughter Kate along with us, as she was by then fluent in Italian, but she had other things to do. When we arrived at the lake, we immediately booked a trip out to the island. To our dismay, the scene that confronted us was nothing like my painting, and the island could never have looked
like that to the mid-Victorian painter.

  Nonplussed, we returned to the hotel and spoke to some locals. We had taken the precaution of bringing a photo of the painting. Eventually someone offered the opinion that the painting did not represent an island on Lake Maggiore, but one on the next lake up, Lake Orta. We didn’t have any better suggestions, so we arranged to travel over to Lake Orta and visit its island. From a boat a third of the way round the island, we saw the scene of our painting, almost precisely as the artist had depicted it—misnamed, misdescribed, and misadmired by my family for about a hundred years.

  The parallels with OED work were quite striking. The more closely you investigate the details of the past, the more you find that they have been altered slightly to suit the present, or have simply been forgotten or recorded in error. We—collectively—lose the information, for example, that a British English word comes originally from American English, and if we forget that, how can we remember from five hundred years ago that one word derives from French and another from Italian? Does it matter? I think it does, and that someone, and as many people as possible, should retain an awareness of the historical stages by which we have reached how we speak and live today. Small mistakes escalate: a misattributed date may cause us to misevaluate a larger variable, which in turn imposes a greater sense of inexactitude on our scaffolding of knowledge. This wasn’t something the Simpsons had ever discussed at home, but I think the wrongly labelled painting confirmed Hilary’s suspicion that with my family there was sometimes less than meets the eye.