The Word Detective

by John Simpson

  To say that we started the revision of the OED at the letter M isn’t exactly true. We decided to leave the big entry M (containing all the abbreviations and initialisms beginning with M) until later, as that was a seriously irregular entry and needed to be done with other seriously irregular “initial-letter” entries in mind. The First Edition of the dictionary had followed the M entry with lots of other fractured pieces of toast: m’ and ’m and other very similar grunts and hums which we rightly thought we could leave for later. So without further ado, we ran our heads into M-A.

  You would have to have been very alert and to have concentrated very hard in school to know that there are—according to the OED—four nouns spelt M-A in the English language.

  M-A noun-one is the oldest word ma in English: it’s the letters M-A used as an abbreviation or shortening of the word master (like the modern “Mr.”) when it’s prefixed to the surname of a man (who might be a friend, a gentleman of higher status, the lord of a household, etc.). So instead of saying “Master Harvey,” might you, in the sixteenth and seventeenth centuries, say “Ma. Harvey”? You wouldn’t, as it’s only a graphic abbreviation—one used in writing: you could write “Ma. Harvey” if you wanted to, and no one would have thought you were talking about his mother. The OED’s revised entry has that fateful dagger hanging over it, the printer’s symbol used to indicate that the word is now obsolete or deceased. It has been deceased, as far as the dictionary knows, since 1602—the date of the last record of its use.

  There are thousands of these daggers (or “obelisks,” as the professional printers used to call them) dotted around the dictionary (†). They are visual marks on the page for anyone who can’t read and understand the abbreviation obs. (= obsolete). It is helpful to the reader to know whether a word or a meaning is still in use, and if it isn’t, the lexicographer stamps it with the dagger. In general the working rule is that the dagger is wielded if we have found no documentary evidence for a term for the past one hundred years. So sword-fencer, a seventeenth-century word for a “gladiator,” is marked as obsolete, whereas sword-hand (still current, though mainly in historical contexts)—the hand in which you wield your sword—is not, even though it dates from the sixteenth century, earlier in history than sword-fencer.

  Marking a term as obsolete if it no longer matches the criteria for active life is one of the most curious jobs that today’s OED lexicographers have to do. A term may have been current when the First Edition of the dictionary was written, but still not meet the hundred-year rule nowadays. Lots of old words are now being labelled “obsolete.” And just occasionally, we have new evidence for a word formerly regarded as dead, and we silently remove the dagger. One such example is nut, in the sense of “a revolving claw that holds back the drawstring of a crossbow until released by a trigger”: the original OED labelled this as obsolete on the basis of its evidence, but renewed interest in archery apparently brought it back into specialist use. We use one hundred years as the required period of disuse before a dagger is applied because of human lifespans. People might remember words from childhood into their dotage a hundred years later. In that case, we would be premature to mark it as defunct. But we don’t dismiss obsolete words from the OED altogether, because it is a record of the linguistic living and the linguistic dead. I don’t really like the concept of obsolescence, as it’s a false marker. Words can easily fall out of use and then return again. But readers like it.

  It turned out that our very first revised entry—for the word Ma.—was unbundled from another entry with which it should never have been associated. The original dictionary had placed the abbreviation Ma. = “master” in the same entry as another abbreviation: Ma. = “majesty.” They had no good reason for that, except the desire to save space. It is against all lexicographical principles to merge together two words from different origins. I think the old editors didn’t really think these broken-up abbreviations counted as real words, and so the normal rules could be dispensed with. Our principal heroics as far as Ma. = “master” was concerned were therefore to drag it out of the master/majesty jumble and set it up on its own account. We couldn’t find any earlier evidence, so we went with the old OED’s 1579 reference (the introductory epistle to Edmund Spenser’s Shepheardes Calender). In general we added all the trimmings: a proper etymology, a fuller definition, and a new illustrative example—that one from 1602 (from the playwright and pamphleteer Thomas Dekker’s Satiro-mastix).

  I have never heard anybody mention this entry outside the dictionary department, but for us it was a tiny classic. This was the very first time we had applied all of our new policies to an entry which was going to be published, rather than just scrutinised by a committee. All the policies had to work, even on an entry most people would regard as inconsequential. Because if they didn’t work here, they certainly wouldn’t work on anything more challenging. Like every other entry, it was a cameo with which the artist must be pleased or the effort is wasted.

  We then moved swiftly on to Ma. (number two) = “majesty”: first recorded in 1584; last recorded in 1679; a graphic abbreviation and a shortening of majesty; daggers drawn again to show that it is obsolete.

  The third M-A noun was ma = “mother.” Here at least we had something of real interest. It’s a word with a history, as in the nineteenth century it was ridiculed “as a mark of vulgarity”—as the OED carefully says (it would doubtless have preferred mother, mamma, or—if you had been to the right school—perhaps even mater). Early spellings include ma’ (the apostrophe showing that it was taken as a shortening of another word). It can be hard to research words which are essentially children’s words, because—by definition—children didn’t record their lexical usages when they were bawling their eyes out in the Victorian nursery. There is a suggestion that ma might have originated on the east coast of England, in Suffolk. But one little piece of evidence from 1823 doesn’t count as conclusive. Mum is earlier (late sixteenth century), and is probably a variant spelling of mam rather than a shortening of mummy—which wouldn’t work, because mummy appears to be the later term (mid-eighteenth century). American mom is later still (around 1850), shortened from momma, itself a variation of mama.

  These words live in word families. We shouldn’t really have worked on ma without editing pa. They both come from the same era: pa crops up first—as far as we can tell—in one of Ben Franklin’s letters, from 1773: “He . . . grew fond of me, and would not be contented to sit down to Breakfast, without coming to call Pa.” Obviously that won’t be the first-ever use of the word, but it’s the first we’ve found. Pa is short for papa; ma turns up a bit later (1829) and is short for mama. But bear in mind that these two common affectionate terms for mother and father don’t go back to the dawn of time.

  Why didn’t we edit the set ma, mum, dad, pops, etc. all at once? There are several reasons. Firstly, the editing system we had then only let us revise entries in alphabetical order (though we could insert new entries wherever we liked): it was the editorial equivalent of an ocean liner—you got on and cruised along without looking to stop at unscheduled locations. All the necessary checks and balances in the software depended on entries being completed in alphabetical order. It would have been, we were told, a “non-trivial” exercise to change this, even if we had wanted to.

  Secondly, these small sets can become stupendously large and unwieldly when you add in all the possible members: in this case we also would have had to consider mammy, marm, mater, maw, mums, mumsy, muvver; bab, bap, da, dad, dada, daddy, papa, and pops—and others, too, if you wanted to expand into informal names for siblings, aunts, and uncles. And thirdly, it is stultifyingly boring (and therefore leads to error) just to work on one type of word. One of my colleagues, Sara Kirkham, was asked to work on the names of all of the muscles in the human body. When she’d finished, she asked politely never to be scheduled to edit a close thematic sequence ever again, as repetition on this scale was exhausting and not conducive to the necessary levels of concentrati
on required.

  There was one final M-A noun for us to process, and this was one for our scientific colleagues. Back in 1925, a group of scientists was casting about for a name for chemical element number 43. These scientists were in Germany. The existence of the chemical element had been predicted over fifty years earlier by Dmitri Mendeleev, who perceptively noted a gap in his periodic table. Mendeleev called the element ekamanganese, not because he was a poet, but because he predicted it would be chemically similar to the element manganese. Our German scientists begged to differ, and called it masurium, after the name of Lake Masuria—now in Poland—near to where the family of one of the scientists had originated. So much for linguistic reasoning. The OED was involved because scientists internationally had decided to use Ma as the shortened name of masurium in formulas, in the periodic table, and wherever else they might need a neat abbreviated version. We were grateful, of course, as it gave us yet another M-A word. The scientists, it seems, sadly became less enamoured of the name masurium over time, and in 1947—looking around for an even uglier name—invented technetium as the name of our old element 43, and this one seems to have stuck. Needless to say, element number 43 has provided a fair amount of work for lexicographers, but presumably more for chemists.

  Tens of thousands of dictionary entries had to be edited in the same painstaking way we edited the M-A words. Even the simplest words require scrupulous attention to detail, though lexicographers become adept at making the right editorial decisions very fast. Nevertheless, even with our new recruits, we were not going to pull this off by the year 2000.

  Despite Ma (masurium), our most complex definitions were normally those for scientific terminology. Science had always been problematic for the OED. The members of the Philological Society and other senior academic advisers involved with the fledgling dictionary in the late nineteenth century had been firmly of the view that they should sit on both sides of the fence where scientific vocabulary was concerned. The dictionary, they blared loudly to anyone listening, was going to contain the entire English language. Not a jot or a tittle was going to be omitted, and when it was completed, the British Empire would have a dictionary to be proud of. “What about the vocabulary of science?” Well, mumble-mumble-mumble, yes, it’s very important but we certainly didn’t mean that the dictionary should include all of that incomprehensible verbiage describing things the average educated chap didn’t need to know about.

  The editors themselves were more enlightened, and favoured the systematic inclusion of as much scientific vocabulary as would be of interest to the educated reader, rather than the eccentric or even mad scientist. In the end they included more or less all the scientific vocabulary they wanted to include, but on occasion took to defining terms in ways that only other deeply involved scientists could understand. So a definition of a scientific word might read like this:

  acrolein: a colourless acrid liquid, of pungent irritating odour, formed in the destructive distillation of glycerin (from which it is derived by the abstraction of two molecules of water, thus, Glycerin C3H5(OH)3, Acrolein C3H4O″). It is the aldehyde of allyl, produced by the oxidation of allyl alcohol, and itself rapidly oxidizing to acrylic acid.

  It is reasonable to assume that complex chemical or biological terminology cannot always be defined in a way that makes sense to ordinary people, but this entry (from 1884) is on the end of the jetty as far as abstruse is concerned, and was typical of the issues confronting our modern-day editors.

  One day a while later, when revision had been proceeding for several years, the OED’s band of scientists working on the current revision met at one of their regular meetings to discuss science editorial policy, and came out of their huddle with the suggestion that they might consider making their scientific definitions more understandable to the general reader, and in fact that this should be the aim of their work. This notion ran counter to the drift of much scientific editing until this point in dictionary editing generally, but any suggestion of change had needed to come from within the fraternity to have any likelihood of success. Rather than let this suggestion fade, I encouraged it, and after a series of further meetings we developed a new science definition policy, driven through by a new breed of enlightened scientist led by my colleague Bill Trumble.

  In essence, the OED’s science definitions would no longer be quite so classificatory (at least when it led to absurd results); if we needed obscure information, we would extract it from the main definition and insert it in a small-print note for Nobel-level scientists. The purpose or use of a term would from now on always be prominent in the definer’s mind.

  An example is worth a thousand words. The substance rosenbuschite was defined in the Supplement to the OED as “a fluorine-containing alumino-silicate of calcium, sodium, zirconium, and titanium occurring as radiating groups of slender triclinic crystals of an orange or grey colour.” Most people would have picked up from this definition that this stuff (whatever it might be) was coloured orange or grey. As a result of the policy change, the comprehensible part of the definition was front-loaded to the beginning in the revised definition, and the most abstruse data, with the smaller readership, appeared last, in an associated note. The full definition now read: “A rare rock-forming mineral containing zirconium, occurring as slender orange or grey crystals, esp. in nepheline-syenites and pegmatites. [Note] Rosenbuschite is a fluorine-containing silicate of calcium, sodium, zirconium, and titanium, (Ca,Na)3(Zr,Ti)Si2O8F. Crystal system: triclinic.”

  Democracy invaded our scientific definitions from this point on. Simplification for the reader made things more difficult for the editor, naturally—which was one of the reasons why it had not happened before. To write a complex definition, the editor needed to understand—at some depth—a complex issue. It is much easier to explain the issue using the terminology of complexity. It is another thing entirely, and requires yet another stage of experimentation and thought, to reduce that complexity to something simple. But from my point of view the dictionary always had to look simple, whatever intricacies it was hiding.

  Hilary and I had been living in Oxford since I had started the job on the dictionary as a ragged-trousered young lexicographer back in 1976. But things were gradually reaching a critical point with Ellie. If events had run their normal course, she would have started attending the local primary school around now, as her sister, Kate, had done before her. Up until this point, we had looked after Ellie ourselves, and she had been cared for privately at home and at a regular day nursery.

  She had had numerous assessments over the years, and the end result of these was that we knew that her development had plateaued at an age of around eighteen months. Since then, there had been no measurable improvement. She was stuck, and we (along with the medical fraternity) had no idea what to do about it. But we did know that she was not “able” enough—in the jargon—to attend a “mainstream” school, even with extensive support.

  At this point we began one of our battles with the education system to find a “special” school suitable for her. We didn’t think of ourselves as the sort of parents who wanted additional support. We were quite able to cope on our own, and we didn’t need support groups and state intervention and extra help. Probably Ellie wouldn’t have minded where she went to school, as long as she was safe and loved.

  We set about investigating the special schools in the area and chose one that seemed to fit most closely with her needs. It also had a pioneering reputation: let’s try to give these youngsters a great start together—not apart, in a school that was “special,” but in a school that was as integrated as it could be with the “mainstream” school on the same site. For me this meant driving a few miles out of Oxford each morning, to deliver Ellie to school, before turning round and heading back to the dictionary.

  Time is different in the world of disability. You can’t just turn up at roughly the right time, and wave goodbye as your child enters the playground. You need to be there at a very precise time to hand her ov
er, and you need to observe the same time tyranny at pick-up time in the afternoon. In due course, we realised the only viable option was for us to move several miles out of Oxford to the village of Wheatley, where the school was located. That solved the deliver-and-collect issue, even though it threw up logistical problems for the rest of us. Fortunately, one of Ellie’s classroom assistants, Teresa, lived near to us, and she brought Ellie home with her each day from school and looked after her until we returned from work. I’m not sure what we’d have done without her. She stayed with Ellie until Ellie left the school at sixteen, and we gave her a flying lesson as a present. I think she landed okay.

  I hadn’t abandoned my attempts to communicate with Ellie verbally, but we had found it was more helpful if we provided additional context through actions and signals—directing her with an arm towards the table, or opening our hands if we wanted her to come to us. We became adept at recognising nonverbal signs—the sort of signs that normally accompany and consolidate language, but in Ellie’s case are all she has. She would push a person away if she didn’t want company. She might cry out briefly, but not in pain (we assumed—and still assume), when she wanted a change of activity. But she can’t tell you if she’s hungry or thirsty or has a headache. You just have to know. But maybe she didn’t need to communicate as much as we thought she did. She was happy, after all. At school the teachers would often write in her home diary “lots of smiles,” “Ellie loved the music,” “She just watched contentedly out of the window as we drove to the wildlife park.” She was responsive, but not interventionist. For me it was frustrating, and continually pointed up a failing in my own power to communicate. But no one else seemed to be able to offer advice. I was used to going to books to find answers, but there were none here. It wasn’t like language, where the obscurest point has been mulled over by linguists across the world. There was a resounding silence if I looked for mechanisms by which I could communicate with her. There was nothing. Compared to this, the dictionary work was easy. It used to remind me that often the meaning of the words we were defining was unclear without context: the surrounding words provided the equivalent of the gestures and nonverbal signs we had been developing with Ellie. But Ellie lived in a world with so little context that traditional meaning was largely irrelevant. In the end, it was more of a problem for us than for her.