language and words in the news language change and slang linguistics and lexicography

Cardiganed old duffers? A lexicographer responds

Lexicographer without cardigan!

The sad news that Chambers Dictionary is about to lose its lexicographic staff prompted a sympathetic article in the Times. Its author, Allan Brown, contrasted the efforts of Internet dictionaries (“pop-cultural hogwash”) with what he regarded as the work of “proper” lexicographers (“we know that our tongue is safe in their hands”). Very nice of him – except that his notion of what dictionary-makers do, and what we’re like, is wrong in every detail. For Brown, proper lexicographers are “white-haired, cardiganed index-carded old duffers … boffinish, pedantic and obsessed; for them the words disinterested and uninterested are as distinct as lions and tigers”.

Let’s deconstruct this. First, you apparently have to be old in order to be a real lexicographer (not like those young upstarts who create dictionaries in cyberspace). This will come as news to many of my colleagues. And cardigans are de rigueur: I have nothing against cardigans but have never owned one.

On to more serious points. It’s a common misconception that dictionaries exist to tell people what they should and shouldn’t say – to be prescriptive in other words – and the corollary is that lexicographers are hawk-eyed pedants ready to pounce on anyone who steps out of line. I have news for Allan Brown – well, not news exactly, because the notion that the lexicographer should be “an historian, not a critic” was first proposed by the Rev. Chenevix-Trench, one of the OED’s founding fathers, back in 1857. The whole point about a dictionary, Trench said, was that it should be an inventory of observable language use, and it isn’t the lexicographer’s job to select only “good” words.

Lots of people get steamed up about language, and the BBC switchboard gets jammed the minute any broadcaster splits an infinitive. But the interesting thing is that the complainers are invariably amateur linguists. Those of us who analyze language for a living are the very last people to make a fuss about the supposed difference between uninterested and disinterested (more on that subject next week).

Allan Brown may also be surprised to learn that the last time I used a “card index” was about 1983. We analyze massive language databases (“corpora”) using sophisticated software tools similar to those used by Google and other players in the language-engineering business. With these resources, we get an accurate picture of how people use written and spoken language in real communicative situations – and this is the raw material for our dictionaries (not a set of ill-founded prejudices about how words “ought” to be used). At a recent conference on e-Lexicography, the topics we discussed included search-engine optimization and the latest computational tools for mining language databases. The subject of correct usage didn’t come up once – most of us are rather disinterested in such matters. Of course, despite our increasing reliance on technology, lexicographers are passionate about language and want to learn more about the way it works. We make a conscientious effort to provide an evidence-based description of the way people use words, but the notion that the language is “safe in our hands” is misguided. We think it is perfectly capable of looking after itself.

About the author


Michael Rundell


  • There are many good online dictionaries, some of which are adapted from great offline dictionaries. Several of them are aggregated through OneLook, which makes it a very useful place to compare definitions. (I have sometimes wondered why Macmillan’s dictionary is not among these.) Wordnik, though not a traditional type of dictionary, offers a range of definitions alongside many interesting new word-related features and functions.

    So it seems unfair of Brown to choose HyperDictionary and Wordia to represent what the internet has to offer in the way of lexicography. I for one have never used either website, nor would I ever rely on Google for the purpose of ascertaining a word’s meaning — though no doubt many internet users do.

    By the way, de rigueur is spelt with two u’s. And in case you haven’t seen it, you might enjoy this cartoon.

  • Thank you for spotting that typo (corrected now for those who will read Stan’s comment and will wonder what he’s talking about). Fantastic cartoon. Love it!

  • Couldn’t agree more Michael. A quote I heard recently, forget where (Stephen Fry’s Radio 4 show I think) sums things up nicely: “A dictionary is an observatory, not a conservatory…”

  • The letter h is always silent in words like honest, hour, and (in American English) herb. And it’s sometimes silent at the start of a weak syllable. Historian falls into this category. History has primary stress on the first syllable, so the h is pronounced. But historian (also hotel and hotelier) have primary stress on the second syllables and some people prefer to say an instead of a in these cases. It’s maybe slightly old-fashioned these days, but it’s perfectly possible to study history and go on to become an ‘istorian.

  • I think this starts just north of Birmingham. I think Spanish pronunciation is creeping in due to holidays in Benidorm?

  • I’m sure that one of the best contributors to the writing of a dictionary was a Dr William Minor (1834 – 1920), an American confined to Broadmoor Hospital (an asylum for the criminally insane).

    Certainly no “cardiganed old duffer”.

    He was highly intelligent and, for over twenty years, he meticulously researched for words (mostly from rare books) to contribute to the New English Dictionary, which we now know as the Oxford English Dictionary.

  • Dear Michael,
    A wonderful piece, congratulations.
    Being a non-native English speaker (or rather, a listener), I am amazed to find so many variants of English. And as a member of the BBC Global Minds Forum, I can attest that it is an evergreen issue (with some 5,000 participants from all over the world, natives and non-natives alike), what should be considered as proper. English as a means of communication has surely become a ‘lingua franca,’ which sometimes is far from the accepted standards. Should you have some time to experience this medley, have a look at this website:
    By the way, one of the members of this forum has sent a link to observe the great variety of pronunciations ( – International Dialects of English Archive (IDEA). It is pure fun to hear so many variants.
    Some years ago (in my cardiganed period) I made a short excursion into lexicography when preaparing an updated version of an old Chambers into a medium-sized bilingual learners’ dictionary. It was a revelation to see what kind of new words and idioms have gained ground with the advent of ICT, with the changing habits of telecommunications, with the appearance of virtual reality.
    Being an enthusiastic user of MED, whenever I meet a new word, I feel a certain embarrassment while clicking on the MED icon. Is it included or too fresh to be allowed? Here are four words I have come across in the past month: troll, lurker, dashboard and scorecard. Two of them have a proper explanation. Not a bad result.
    Best regards.

  • A fine piece, and gets it right, I believe, in its general drift. However, the elephant in the room is the difference between English dictionary-makers and continental dictionary-makers.

    What the writer doesn’t mention is that the current mode in the English speaking world of compilation is indeed descriptive, whilst that in the continental world it is still that of ‘prescription’.

    As English teachers that is a barrier that we have to face. In English the precept that the description of usage is taken for granted constrasts severely with the tradition in Spain, France or Italy, where the use of certain elements of language must first be approved by some officially appointed body.

    In other words the author is fighting for a battle already won within English, but neglects to advise us about how to deal with cultures who see things in a different way.

    For example, you may say ‘I’m texting my friend’, and if you were lucky enough to have students of a level and curiosity which prompted them to ask ‘But is that a real word?’ you could reply ‘Well, it’s widely used’. Then you would get ‘Is it in the English dictionary?’ To which you would probably say ‘Yes, probably, since it is widely used’. The question that would follow would probably be ‘Yes, but what state-sponsored-official body said it was OK?’ and you would have to say ‘Well, none’ and try to explain the difference between descriptive and prescriptive, which is more of a cultural than a linguistic concept.

    Basically, latin countries are more used to being told what to do, and they accept it, than Anglo countries. At its worst extreme ‘You’re a zombie and I’m a free-thinker’.

    But what is the best way to transmit or explain that cultural difference when you’re on the ground? ‘Well, we all have different ways of doing things and one is as valid as another’ or ‘Sorry, but you are looking at language wrongly, it evolves and will happen regardless of the esteemed ‘Academia Real de la Lengua’, which preserves the dignity and correctness of our language?

    Please give us some advice on this much more thorny point.

    Sean Mitchell.

  • Having said that, taking it within the sphere of just the English language, there are grounds for grumping about the use of ‘disinterested’ as ‘uninterested’. The grounds that I refer to are not pedantic, but rather of the loss of a word or of a meaning. If disinterested becomes the same as uninterested then we have lost a means of expressing our thoughts. Unless, of course, one can think of a synonym for disinterested, and thereby render it superfluous. It might be ‘unbiased’ or might not. I tend to feel that it is slightly different.

    But if you can find a perfect synonym for ‘disinterested’ then I would accept its assimilation into untinterested (something quite different, on the face of it), but if you can’t then I would see it as not only a word but a concept lost to us.

    Again, Sean Mitchell.

    Do you feel that there is a difference between ‘disinterested’ and ‘unbiased’? If you do then perhaps we can

  • Sean Mitchell makes two very good points, which I’ll try to address. First, on the contrast between English (which has no academy) and many European languages (which do). I don’t feel qualified to offer advice about this, but Sean’s argument – that languages evolve and that language-change “will happen regardless of the esteemed ‘Academia Real de la Lengua’,” is demonstrably correct: the Académie française, for example, has proved more or less powerless to resist the tide of English loanwords coming into French. I agree there are sensitive cultural issues here, and interestingly these differences apply even within the English-speaking world: speakers of American English tend to be more conservative than us Brits; their dictionaries are more old-fashioned, and few if any are based on corpus data (and therefore can’t reflect real usage as accurately as corpus-based dictionaries like ours).
    On the “disinterested” question, I hope I can reassure Sean (and others) that the “unbiassed” meaning shows no signs of dying out. It has not been superseded by the “not interested” sense but the two co-exist quite peacefully. This is all pretty normal: many, if not most, words have more than one meaning but, thanks to context, true ambiguity (when a reader or listener genuinely fails to interpret a word correctly) is extremely rare. The word “bank” (a favourite with linguists) has at least two quite distinct meanings, but if I say to my wife “I’m just going to the bank”, she doesn’t stop me to ask which type of bank I mean. Similarly with “disinterested”: when we analyse the data for this in our corpus, it’s always clear which meaning is intended.
    Thanks to Sean and everyone else for their interesting comments!

  • There is actually more than one elephant in the room. In fact, the room is crawling with elephants.

    Thank you, Michael, for shedding light on the glamorous world of lexicographic fashion (both as a mode of dress and “the action or process of making, workmanship”). I fully agree with your deconstruction of the Times piece, of course.

    But my elephant carries a big poster, which reads: “Not only is there a lot of prescriptivism going on in the rest of the world, but, in fact, in some countries (let them remain nameless) index-cards are, alas, still not only de rigueur, but, in fact, a la mode.”

    Now, writing dictionaries or doing corpus linguistics or NLP for a big, well-established and well-resourced language is, really, like flying first-class: the food is excellent and the crew is very friendly. But there is not one, but actually many lexicographies being practiced around the world today. And some of it is plain ugly, dusty and extremely uncomfortable. Unfortunately, not everybody can fly first class. (And I am speaking from experience, after being stuck at Istanbul airport last week for two days).

    Even in the first-class cocoon, however, things are far from perfect. Not having Chambers around will be a serious loss, but the Janus-like nature of lexicography as both a scholarly and commercial activity will continue to cause serious theoretical and practical problems in the age of the “free-for-all” Internet.

    Now, I very much enjoyed your presentation in Louvain and your reassuring conclusion that lexicographers will exist — even if lexicography ceases to exist, at least in the form we know it today. It was a great start. But the majority of the papers at the conference did not actually come even close to addressing the big question of how lexicography should seriously deal with the Internet — by engaging with it and responding to it.

    I am talking here about major shifts, which I believe are necessary, if lexicography is to truly move forward: treating the dictionary as an embeddable web-service rather than a monolithic object and opening dictionaries to programmatic API calls so that they can be easily integrated with digital libraries and, really, any web content out there. (Thank God for Wordnik — the first API that was made public. The Transpoetika Project at the Belgrade Center for Digital Humanities that I am involved with is doing something similar for Serbian). The future of dictionaries, big and small, commercial and free-of-charge, will be, among other things, dependant on the advances in the semantic web technology.

    Today, we still go to certain websites to consult a dictionary. But that is really old-fashioned kinetics. Dictionaries should (and, I am certain, will, in the future) come to us. The goal of eLexicography is not so much to make dictionaries obsolete, as to make them invisible.

    So it’s still a long way to go — for everybody.

  • Dear Toma,
    Time will tell when and how you can fully rely on e-dictionaries, or what more, on ’embeddable web-service’, but for the time being, the state of the art is this: Your above (and highly important) sentences have become, well, a bit misleading, to say the least. Please forgive my pun of having these sentences

    ‘Now, writing dictionaries or doing corpus linguistics or NLP for a big, well-established and well-resourced language is, really, like flying first-class: the food is excellent and the crew is very friendly. But there is not one, but actually many lexicographies being practiced around the world today. And some of it is plain ugly, dusty and extremely uncomfortable. ‘

    been translated by Google translator into Hungarian, then back into English:

    “Now, writing dictionaries, or corpus linguistics or NLP is a large, well-established and well-equipped language, I really like flying first class: the food is excellent and the staff is very friendly. But there is not one, but in fact many lexicographies the impact of the world today. And some obvious, ugly, dusty, and extremely uncomfortable. ”

    Best regards.

  • Dear Mihaly,

    I love translation roundtrips — I remember using a rudimentary web application to translate back and forth Derrida’s opening sentence from Monolingualism of the Other: “I have one language, yet it is not mine”. That was long before Google Translate and the results were, much more ridiculous than what we get today.

    But actually I was not talking about automatic translation at all. That’s not the field that I work in or that I am personally passionate about. I am talking about “real”, human-made dictionaries that need to find new distribution paths… and become more readily available in the world of networked devices.

    Call me crazy, but I want dictionaries that can take the full advantage of fluid elctronic textuality.

    All best,

Leave a Comment