The Word Detective

by John Simpson

  I had assumed that, on my first day, my handler would give me a relatively easy word to work on. That would seem fair. I’d heard that the department was at the time engaged on a thorough review of the letter Q. There was little chance of overstraining myself here, I thought to myself. Perhaps it was a good time to sneak into the department without being noticed, and gain some experience by editing up some quiet, out-of-the-way words in the byways of language. I thought maybe quack (you can’t go far wrong defining that, surely), or quadrennial (“once every four years”: surely no problems there for me), or maybe Quakeress, if such a word existed.

  But no, things didn’t work like that. There was a training scheme of sorts, and it meant not rolling up your sleeves and tackling real editing until you’d been around the track with all sorts of ancillary tasks designed to ensure that you were passingly confident with alphabetical order, the parts of speech, and the general content and shape of the OED. In retrospect that sounds sensible, but at the time I struggled to see the logic. I imagined that I already had majestic command of the parts of speech and the job each one performed; in truth, I was doubtless as casual about these as some of the young editors I’ve encountered since, who are initially equally foxed by participial adjectives, and absolute uses of nouns, and quasi-adverbs, and all the rest of the paraphernalia of slightly off-centre and old-fashioned grammatical categories that the original editors of the OED were happy to work with. Because, don’t forget, we weren’t writing a dictionary from scratch, we were labouring on a supplement, or long addendum. We were taking as read the venerable words written by Sir James Murray and his fellow editors back in the late nineteenth and early twentieth centuries and adding the new words and senses of recent days.

  I have a predilection for words that entered English in the seventeenth century. Well, we all have our preferences. Paraphernalia looks Latin, but like no Latin you are likely to have encountered. Not all words from Latin date from the days of the Roman Empire; many come from later versions of the language—religious Latin, scientific Latin, and by definition the languages into which Latin developed in the first millennium AD (for example, Italian, French, and Spanish—known nowadays as Romance languages not because of their suitability as the language of love, but because of their derivation from the language of Rome). Paraphernalia entered legal English, according to the OED, around 1650, directly from the legal Latin of the contemporary courtroom, though the word elements hark back to Greek. Para- means “besides” or “alongside”; -phernalia derives from the Greek word for a dowry. To the seventeenth-century lawyer, paraphernalia were those articles of a woman’s personal property (typically clothing and ornaments) which didn’t—like everything else—pass by law into her husband’s possession on marriage. By the eighteenth century we were using the term in a wider context—of any of the trappings or accoutrements belonging to a person. Curiously, it could take either a singular or a plural verb (“paraphernalia is” or “paraphernalia are”). Its onwards trajectory does not stop there. By the 1920s we associated the word with the apparatus of drug-taking, and then—more recently—in South-East Asia and the Philippines, it is used in the plural (paraphernalias) to mean any of the bits and pieces associated with any activity.

  To my disappointment, instead of setting me loose on the dictionary, my handler introduced me to the simpler aspects of the job. Instead of writing (which I thought I’d been recruited for), I was given some reading (which I thought I’d already spent years perfecting at university). And my chosen reading—that’s “chosen” by someone else, not chosen by me—was not even written by an English-language author. I immediately gained the sympathy of the two delightful women with whom I had the honour of sharing my first office when I was presented with the task of reading a translation of Film Language, an esoteric French text on the semiotics of film by Monsieur Christian Metz. Neither of my new friends could imagine curling up with such a dry text for a quiet bedtime read. What I should have realised was that the dictionary’s editor and his senior team had selected this modernist text as part of the project to modernise the dictionary. The dictionary was starting to move with the times, even though its Victorian and Edwardian scholarship harked back to earlier ages.

  Unsurprisingly, I had not previously read anything by Metz. According to the jacket copy, he was something of an expert in his field, but as the semiotics of film wasn’t a field into which I had ever stepped, that didn’t really help. Fortunately the book had been translated from French into English before I was asked to read it.

  The purpose of the read, I was informed, was to find words, meanings, and expressions that might prove useful to the OED, and then to write them out on index cards called, I’d soon learn, “slips.” These cards would be filed in the dictionary’s word dungeons, awaiting that future day when an editor needed them to help compile an entry. The task wouldn’t improve my knowledge of the semiotics of film: I wasn’t expected to retain any information after the book was read. English obtains many termes cinématographiques from French, so reading a translated French text was adjudged an excellent way of catching them at the moment they passed into English.

  There are people who argue that extracting example sentences from books is a poor way to collect information about words. They contend that when people are set to read a book for significant words, all they find are significant words (and they miss a lot of those, too), and so the result is skewed data: too much of what you don’t need—words towards the rarer end of the spectrum—and not enough of what you do—everyday but easily missable nuances of language. But at the time, (a) no one could really prove that the method was problematic, and (b) the Victorian editors had worked their fingers to the bone finding examples of everyday words to conceal the problem.

  So I continued with the task I had been set. I had my book, my slips, my ballpoint pen, and my desk facing the wall. The multiple volumes of the First Edition of the OED (1933 reprint) were ranged on the shelves in front of me. I needed to consult those fairly often, to check whether a word I’d found in my semiotically charged text was already suitably covered in the dictionary. There was no point in inundating the card files with slips containing words already perfectly well evidenced.

  It took me a long time to reach the end of Film Language, for reasons that will be obvious to many people. I excerpted about two hundred slips from the book over the following three weeks, which I’d later learn is about average for a medium-sized academic text.

  More than forty years on, you can miraculously still find evidence of my reading of Christian Metz’s Film Language in the OED. The dictionary documents everything, so when it defines a word, it presents a selection of example sentences containing the word from its card files (such as the ones from Metz’s book that I had culled). You will see a sentence from Film Language—collected by me—illustrating the entry for diegesis, meaning “the narrative presented by a cinematographic film or literary work,” and also another one transcribed by me illustrating non-chronological. There’s nothing particularly filmic about the latter word; it’s just a (nineteenth-century) alternative for the older (eighteenth-century) unchronological. Metz’s example was just available as a useful filler (doubtless not a role to which Metz would have aspired)—it wasn’t presented as the earliest example ever found of the term in English. In fact, in the OED Online today there are fifteen quotations from Metz’s Film Language, and you can bet your bottom dollar that they all came from my pen.

  The process of reading Christian Metz proved to me one central plank of the OED’s philosophy: that anyone who takes the time to read a text carefully can uncover information that is useful to the dictionary. It’s just that, normally, people keep this information to themselves, or don’t even bother to check whether they have made a discovery. You need to separate the collection of information from its analysis: not everyone can manage the second activity, but anyone can contribute to the information-gathering. That was one of the great discoveries that the dictionary m
ade when it was starting up, back in the 1860s and 1870s. So much new information about the language had to be collected that the early editors asked the reading public to help by noting down interesting words and usages they encountered in their reading, just as I had done when reading Metz. And it turned out that many people in those days had some leisure time and wanted to help. Right at the beginning, the OED discovered a hidden army of potential contributors.

  You will want to hear about my biggest Metzian successes: the words I found and carefully transcribed on to index cards which are currently printed as the first-ever known use of the terms in the English language. There are two of them: prefilmic and screening room. When I encountered these terms in Film Language, I didn’t know they were first usages. There were no entries in the dictionary back then for either prefilmic or screening room, so as a reader I was working blind. Word-collection relies almost exclusively on the intuition and experience of the reader. You often just card a word in the hope that it will prove to be useful, but you aren’t always right. It’s not an exact science, and you have to accept a certain amount of redundant data in the file. Some people develop the right historical radar and others don’t.

  But since we can research more efficiently these days—thanks to computers and enormous online historical databases—it may be worth checking whether either of those celebrated first uses would pass muster today. Let’s look at prefilmic first. A single search on Google Books rapidly brings up an example from 1967 (Sheldon Renan’s Introduction to the American Underground Film). So in the OED’s next pass-through, my brilliant first use will disappear. And what about screening room? That seems an odd English compound to be first encountered in a translation of a French text on semiotics. A moment’s search on the online historical databases, and we’re back to 1936 and the American magazine Popular Mechanics: “In the quiet of the screening room our little band of silent adventurers watched the film that had been automatically ground off by Jim’s motor-driven cameras.” Popular Mechanics was clearly not a source combed by the OED’s cerebral editors.

  So my two little first dates are destined for the scrapheap of history. They twinkled for a few years before new ways of researching caused them to be brushed aside.

  By this simple act of reading a book for the dictionary, I was joining a group of thousands of people who, over the previous hundred years or so, had contributed in smaller or larger ways to the dictionary’s word-store. These crowdsourcing “readers” have never been predominantly academics—they were typically just ordinary people who wanted to share with Oxford in the exploration of the English language. Old ledgers of these readers still exist: their names are mostly forgotten nowadays, but there you find reverend gentlemen and their wives, well-educated people concerned that knowledge of the language should be respected and treasured; members of London clubs with moments to spare before dinner or billiards; elderly spinsters from the Victorian leisured classes with afternoons stretching ahead of them; and sometimes literary figures—the novelist Charlotte Yonge, extremely popular in her day, would send the results of her reading to the editors in Oxford.

  Many of these readers had been attracted by short “Appeals Lists” of words “wanted” for the dictionary: lists of common and abstruse terms for which earlier, better, later, explanatory, or defining examples from printed texts or manuscripts were badly needed before the dictionary could make proper progress. These lists were drawn up by the dictionary’s editors and distributed by OUP throughout the world by means of the Press’s network of booksellers and other contacts. They were also published in OUP’s own in-house magazine, The Periodical, from where they might find their way into the national papers. Many readers lived in Britain, but concerted efforts were made to involve American readers as well as others from around the English-speaking world. The Philological Society had, very early in the life of the dictionary, invited American politician and man of letters George Perkins Marsh to drum up support across the Atlantic. He was not entirely successful, but later efforts produced a steady influx of American English material, and as the dictionary grew in size and acclaim, more and more readers became attracted to the work.

  “Reading” for the dictionary was all very well, and it helped to gather together a mass of material that might be useful in future years to the dictionary’s editors, but it didn’t do any good at all for my own ability to read. The process of reading text word by word, and then weighing up whether each word was worth carding for future reference, played havoc with my appreciation of literature. My estimate is that it would take the average person about five years of working on the dictionary and “reading” texts of all sorts before he or she came through the barrier and was able to read properly again.

  Put yourself in the reader’s shoes: You are reading Jane Eyre, perhaps not for the first time, but you’re enjoying it all over again. You’ve followed the narrative through its twists and turns, past the fire, towards the very end. You find yourself at the start of the last, cathartic chapter. What happens next? The chapter title is “Conclusion,” followed by “Reader, I married him. A quiet wedding we had; he and I, the parson and the clerk, were alone present.” What are the man-traps here for the budding lexicographer?

  Your growing lexicographical intuition stops you dead in your tracks. Firstly, there’s the word conclusion. You probably know that Charlotte Brontë didn’t invent it, and you probably have some inkling that it had been in English for many years. You would be quite right, as we borrowed the word from French towards the end of the Middle Ages. But would the OED be interested in this particular use of “conclusion” (dating from 1847) with reference to the conclusion of a story? How old is that use? Should you spend two minutes jotting all the information down on an index card for the dictionary’s files? You check in the big dictionary and find that even Geoffrey Chaucer used conclusion in roughly that sense. It takes a long time for new editors to develop a confident feel for the age of words, and that’s something which (though hard-fought) separates historical lexicographers from most other people.

  “Reader, I married him.” Your eye, closely followed by your brain, halts at “reader.” Again you know full well that reader has been around in English since the dawn of time (in this case from way back in the Anglo-Saxon period). But when did authors start addressing their readers directly as “reader,” or maybe “gentle reader,” in the text itself? Was that Victorian? Was Charlotte Brontë in the forefront? You reach up again for the relevant volume of the dictionary, and once again you sit down abashed. Any author might have addressed his (or her) reader in this way ever since the Middle Ages. William Cowper, in 1785, talks politely to “my gentle reader.” Charlotte Brontë is deep in the following pack and you don’t need to bother yourself with carding this use.

  You get the idea. For the new lexicographer, reading a historical text in search of material for the dictionary for the first time is an uncomfortable ride, full of pitches and tosses and jolts as you stop and start to check a reference which, as often as not, proves irrelevant to the job in hand. We can proceed a little further, if you like, and then I shall have to leave you to read the rest of Jane Eyre yourself.

  The next sentence reads: “A quiet wedding we had; he and I, the parson and the clerk, were alone present.” People can be quiet, dogs can be quiet, and even rooms can be quiet, but when did we start to think that weddings might be quiet, and what precisely does it mean? Is it a new sense, applied to something that wouldn’t normally speak or make a commotion? Again we consult the dictionary. Again we find we might have spent our time better: the OED already knows about this sense of “free from excess; moderate, modest; restrained” from the sixteenth century. Where have your lexicographical antennae gone? Will you ever get one right? You will, and you do, but not this time.

  It turns out, as we have just seen, that you don’t need to stop and consider any of those words. The OED had already taken care of them, but you don’t know that. So you check. And you spoil your
reading of one of the classic passages of English literature because you are too worried about the language. Well, good for you. You can read Jane Eyre again later, but you may not have the chance to think so carefully about the language of the early nineteenth century as you do now. And once you’ve checked those facts, you’ll stand in much better stead when you meet the same expressions next time round in another text from that period.

  There was more to lexicography than writing definitions: that much I’d discovered by now. But in my first weeks on the dictionary I found out that large-scale historical dictionaries such as the OED weren’t compiled by editors thinking deep thoughts at their desks and perusing scholarly books on language, but by editors grabbing hold of the evidence of the language that was racing past them, in the form of whatever text they could lay their hands on. Information had to be collected before it was analysed, and the scope of this collection influenced the scope of the dictionary. My first job had not involved learning how to write definitions, as I had expected that it would, but reading a book on the semiotics of film, in search of background material that might be useful to the dictionary at some unscheduled time in the future. And I learnt that we had to read a wide range of texts to hope to capture a representative sample of the language. If we only “read” novels, then we would be missing out on a stream of language swarming through newspapers; if we avoided poetry, we’d miss other types of language; and so it went on.

  The example quotations in the First Edition of the OED are often and (I think) rather ungenerously said to be drawn too much from famous, classic authors (Shakespeare, Milton, Swift, Pope, and—surprisingly high up on its ranking list—Sir Walter Scott). I don’t think the editors intended to privilege “highbrow” literature; it was just that these were the texts to which readers had easiest access, and which the publishing world of the day made most readily available. But I’ve always had the greatest respect for those early editors: they did a remarkable job with the resources at their command. At the same time they had left us ample scope to add new evidence from less formal sources (earlier evidence, informal idioms, technologies not popular with the original Victorian readers) and to show that language is never the exclusive preserve of the literary giants of the age. I had also found out that it wasn’t only editors who were competent to collect important dictionary information; anyone with an ear for language and history could make discoveries.