The Merriam-Webster Word Factory.

Jennifer Schuessler takes an enjoyable look (for the NY Times) at lexicographer Kory Stamper and her inside view of the workings of the Merriam-Webster empire (now sadly diminished, like all lexicographical enterprises, but still going strong). Stamper (whom I’ve posted about before, e.g., here) has a good attitude about language:

Ms. Stamper has no patience for self-styled purists who quail at “irregardless” — an actual word, she notes. (She is O.K. with ending sentences with prepositions as well as — brace yourself — split infinitives.) But she also describes being caught up in some higher-stakes fights.

One chapter takes an uncomfortable look at the racial assumptions baked into a Merriam-Webster definition of the color term “nude.” Another recounts the furor that erupted in 2009 when it added a subdefinition to its entry on “marriage,” noting uses to refer to same-sex unions that weren’t necessarily legally sanctioned. [..]

If dictionaries are a form of information technology, the building is in some ways a catalog of obsolescence. A downstairs gallery includes a 1934 poster advertising the second edition of the Webster’s New International Dictionary, billed as “one of the thickest books ever printed.” (The technology needed to bind it, Ms. Stamper said, no longer exists.)

There are also oddities like an asymmetrically bound Seventh New Collegiate from 1969, designed so it could hold itself up — an innovation that failed to catch on, probably because if you open it too far from the center, it falls over.

There’s a picture of that asymmetrically bound failure, as well as of the monstrously fat second edition of the New International and other related objects, and in general it’s worth a look. Thanks, Eric!

Addendum. Stan Carey reviews Stamper’s new book, Word by Word: The Secret Life of Dictionaries.

Comments

  1. That brings back a memory, but I don’t remember of what exactly: just an implausibly fat book, much thicker than its height. I do remember that Webster from various lbraries, with its dedicated lectern.

    I never liked the WNI: not the definitions, not the meager etymologies, not the ugly typefaces, all the things which the OED does so well.

  2. > […] an enjoyable look (for the NY Times) […]

    For a second, I thought this was an incredibly backhanded compliment. 😉

  3. Just what I thought.

  4. Heh. Well, it works either way…

  5. m-w.com defines nude in this sense as “having a color (as pale beige or tan) that matches the wearer’s skin tones” and gives the examples nude pantyhose and nude lipstick. I suspect this is a triumph of aspiration over facts: if there are still pantyhose out there labeled “Nude” (and maybe there aren’t), they probably still refer to the color of white people’s skin.

  6. David Eddyshaw says

    “… uncomfortable look at the racial assumptions …”

    I have an African folktale in which the protagonist is walking through the forest and discovers that he’s being followed by something; eventually he realises it’s “black, like a human being.”

    Assumptions surely aren’t ipso facto racist if they’re natural assumptions based on actual facts. Having said that, they can become racist if people persist in them in a sort of faux-naivety knowing that they are genuinely hurtful. (I shall never refer to “nude pantyhose” ever again.) It’s a bit like the “Eskimo” thing; it *doesn’t* mean “eater of raw flesh” and has no intrinsically pejorative sense, but it’s surely boorish to persist in applying it to people who are truly offended by it. Even if they are Canadian.

    Is “pantyhose” plural in American? Singular for me, though it’s not a word that I utter frequently. Pantyhoes, now …

  7. ə de vivre says

    Re: nude and skin

    Procuring appropriately flesh-toned shoes has also been a big problem for non-white ballet dancers.

  8. It seems more natural to me to say there are pantyhose than there is pantyhose. I think what happens is that I assimilate it to pants, which is a mass noun that’s plural in form (so-called pluralia tantum) and therefore takes a plural verb. Of course, when referring to them individually, both nouns take the classifier pair(s): one pair of pantyhose, two pair(s) of pantyhose, etc.

  9. “Assumptions surely aren’t ipso facto racist if they’re natural assumptions based on actual facts.”

    Interesting, I would not say that. I would say that an assumption based on facts can certainly be racist, if the assumption obscures my view of an individual whom it doesn’t apply to. Racial stereotyping right there.

    Machine learning systems readily make “neutral racist” assumptions like this. If you train a face detection system on a valid random sample of the US population, it will build a model for what faces look like. It tends to generalize, that’s part of what learning is. Most learning techniques will end up having better accuracy on majority faces than on minority faces, because they learn from the facts presented, and optimize for what they’re given. Is that racist? Well, by how it acts, it is. If you don’t want that result, that takes specific counter-action.

  10. Are data compression algorithms racist? By it’s very nature, data compression takes more frequent features (words, strings of 0s and 1s, whatever) and assigns them shorter sequences of the coded text. If we make a database of some characteristics of people that are correlated with race and nude race happens to be represented more heavily than ebony one then nolens volens the coded signal will be shorter for nudes (on a per person basis) than for ebonies. And the data for a minority group risks to be more prone to errors in transmission, longer times etc. For example, in streaming videos, images of ebony persons might be a bit more blurred or aliased as the result. Is there a reason to be outraged?

    Footnote: I am by no means an expert or even a knowledgeable amateur in streaming videos and my “example” is pure imagination (you may call it a euphenism). I have no idea whether there is such a problem in reality or not.

  11. David Eddyshaw says

    “I would say that an assumption based on facts can certainly be racist, if the assumption obscures my view of an individual whom it doesn’t apply to.”

    Valid point. In fact a good deal of purportedly dispassionate “scientific” racism (and sexism) is based on an illicit extrapolation from groups and averages to individuals (that’s even granting that their “facts” are factual in the first place.) Mark Liberman has made this point often over at the Log (with worked statistical examples …)

    I should have put it a bit more carefully. I *do* think it’s important to keep in mind that the question is fundamentally about facts, not courtesy: racism is repellent and morally wrong, but more important still, it is based on factual error. Failure by the good guys to keep this firmly in mind allows racists to pretend (perhaps even to believe) that inconvenient “truths” are being suppressed by political correctness.

  12. Trond Engen says

    Here’s an interesting interview from a couple of days ago on the economics of stratification.

  13. Interesting, but the insights seem blindingly obvious (except, apparently, to traditional economists). What struck me was the beginning of the first response: “So, I think the core of stratification economics […].” It’s a perfect example of the introductory “so” that’s become ubiquitous in recent years; I think I posted about it at some point, but it’s the kind of thing that’s hard to search for. (Dave Wilton at Wordorigins.org claimed a few years ago that it’s been around forever, but I’m quite sure he’s wrong, or at any rate that there’s been a huge surge in popularity.)

  14. An awful lot of words to say, or to tiptoe around saying, “we hate most savagely not those who have wronged us, but those whom we have ourselves wronged” (Evelyn Waugh). Still, his point about the anti-intellectualism of mainstream economics (which I would call an anti-science) is well taken.

  15. It’s very easy to search for the introductory “so”, just look at ones beginning with capital S. But if it happens mostly in spoken language…

  16. But, turning back that ecological fallacy to linguistics, how many people who are chatting idly about language (ahem) remember, for example, when discussing features of AmEng pronunciation, to make a nod toward AAVE vs. just making General American a stand in for the whole.

  17. David Eddyshaw says

    “To be frank, I think that the economics profession has a certain anti-intellectualism. That’s a pretty strong statement, but I mean that in the sense that if you think about intellectual activity as involving wide-ranging curiosity and also wide-ranging interests in research unbounded by disciplinary lines, I think the economists are very, very inclined to be somewhat incurious and to treat every problem from the standpoint of a fixed package of ideas.”

    This seems to be rather idiosyncratic use of the word “anti-intellectualism” to me. Intellectual incuriosity is hardly confined to the unintellectual, and nor are unintellectual people characteristically curious, at least in the sense he means.

    With regard to obviousness, it’s one thing for something to be intuitively obvious and another to come up with a coherent and fruitful way of modelling it in economics. I think the hope is that a lot of obvious things can be left out of economic models without them losing all usefulness; otherwise the models just get too intractable. The driver for trying to formally incorporate yet another piece of obviousness is failure of the existing models in some respect. An example is the idea that you can treat economic actors as being perfectly rational; this is obviously false as a matter of everyday observation (even, presumably, to economists) but it was justifiable to do this so long as the resulting model was useful. In this case, the motivation for coming up with a new model is that the existing ones would have predicted a decline in the degree to which “race” would predict prosperity, and that this hasn’t happened.

  18. Waugh seems to have been closely paraphrasing Tacitus: “Proprium humani ingenii est odisse quem laeseris.” Now I wonder who Tacitus was paraphrasing.

    And if machine learning algorithms and data compression systems make default racist assumptions that require specific counter-action, isn’t that exactly because they mimic brain processes?

  19. David Eddyshaw says

    Auden, source of all (well, some) wisdom:

    “Hard cases make bad law”, as the politician learns to his cost:
    Yet just is the artist’s reproach – “Who generalises is lost.”

  20. This seems to be rather idiosyncratic use of the word “anti-intellectualism” to me. Intellectual incuriosity is hardly confined to the unintellectual, and nor are unintellectual people characteristically curious, at least in the sense he means.

    Well, yes, but you have to allow for the euphemistic and waffling style. What he means is ‘dogmatic’. Science begins in curiosity about phenomena and an attempt to explain them, but in economics the explanations have taken over, and if the data contradict them, so much the worse for the data. Parts of other sciences suffer from this disease, as we know all too well from linguistics, but in economics it’s close enough to universal that radicals apparently have to dissemble their views.

    Yet just is the artist’s reproach

    “The artist” is Blake, an artist in both the narrow and the wide senses, but he was blunter about it: “To Generalise is to be an Idiot.” Indeed Auden’s epigrams of which that is one seem to be explicit imitations of Blake’s “Proverbs of Hell”. (This may be a commonplace of Auden criticism, about which I know nothing.)

    On investigation, though, “who generalizes is lost” appears as early as 1903 in a book by one Rupert Hughes, The Love Affairs of Great Musicians (which is just what it says on the tin). Even earlier, in 1897, it shows up in the North American Journal of Homœopathy.

  21. Of course, in homeopathic use the generalization is watered down until it is undetectable.

  22. January First-of-May says

    But, turning back that ecological fallacy to linguistics, how many people who are chatting idly about language (ahem) remember, for example, when discussing features of AmEng pronunciation, to make a nod toward AAVE vs. just making General American a stand in for the whole.

    To be fair, AAVE is its own dialect, and should probably be considered in a separate category from AmEng proper.

    I won’t be surprised if modern AAVE turns out to have less in common with General American than the latter has with Queen’s English (someone probably researched it, and for all I know I’m actually spewing linguistic nonsense… and IIRC the differences are more about vocabulary and grammar than pronunciation, anyway).

  23. David Eddyshaw says

    The major grammars treat British and American standard Englishes together without any great problems, just the odd subparagraph or footnote; but you wouldn’t be able to incorporate AAVE into the same framework without making the whole thing into a much more complex pandialectic study. The verb system especially would need a lot more than the occasional footnote to do it justice.

    That doesn’t absolutely settle the question, I suppose, though, given the complicated sociolinguistics of how far actual speakers accept variation as still falling within their own language. Even the linguistic gold standard of mutual comprehensibility is dependent on speaker experience, attitudes and preconceptions, of course. I’ve no doubt there are American standard English speakers who find AAVE much more comprehensible than even quite refined Glaswegian.

  24. David Marjanović says

    This seems to be rather idiosyncratic use of the word “anti-intellectualism” to me. Intellectual incuriosity is hardly confined to the unintellectual, and nor are unintellectual people characteristically curious, at least in the sense he means.

    What he means is that “the dismal science” is dismal and is not being done as a science.

    The verb system especially would need a lot more than the occasional footnote to do it justice.

    It’s on Wikipedia, and it’s quite scary.

  25. David Marjanović says

    From the interview:

    A new graduate student really has to make a very careful choice about which department to go to, and once there, who they should work with in that department. I would say that’s the research that needs to be done carefully, rather than telling people they shouldn’t go into economics.

    Yeah. He’s saying it’s not being done as a science anywhere or almost anywhere.

  26. Trond Engen says

    Many of the insights of mainstream economics are obvious, some of them in hindsight, others even before they were formulated, but even then, there is value in the formal description of the obvious, or in the forces behind it (“Yeah, planets orbit, big deal, Isaac”), Other insights are less obvious, and those are even more important. But even old ones like the paradox of thrift can be a hard sell to journalists, politicians, and the general public.

    So I don’t think the description of mainstream economics as anti-science is completely fair. Nor is it unfair. Too many have for too long been wasting their time on micro-founded supply-side model-playing with no calibration in aggregate-level real world data, and that has pretty obviously been because the predictable results from that approach have been in high demand, Professional discussion of macro economics consisted of reputed names each touting their own neat little model of the world without dicussing the assumptions it was built on, the ideological implications of those assumptions, and the resulting limits to its validity,

    Even so, a call to reject economics is more a call for a licence to reject its actual insights for the sake of popular prejudice than for better and broader solutions to complex social questions We need more and better economics, not less of it. As the saying goes, if we were to reject it, we would have to reinvent it with another name. After generations lost to voodoo, Important discoveries are waiting to be made on how human societies actually work. The economist’s contribution would be to reconcile the forces of individual decision-making with the real, observable aggregate effects on society. This has been going on even in the dark ages, but it’s now returning to the forefront of the profession — or at least that’s how it looks to this outsider. Some study how individual decision-making is actually done, others study huge aggregate effects in new ways (Piketty & Saez. Case & Deaton), and work like that of Darity & al. describes how unwanted aggregate effects can be the results of rational individual decisions.

  27. David Eddyshaw says

    What Trond said…

    There’s not nearly as much wrong with economics as a discipline as many people would like you to believe. The media-filtered view of it that most people encounter is about as good a reflection of how it really works as the media presentation of linguistics, i.e. there is reliable stuff out there, but swamped by reporting by journalists who don’t even know enough to know that they’re ignorant.

    With economics, this is vastly exacerbated by the fact that it deals with such politically salient material. On the one hand, this leads to genuine expertise being prostituted to political ends; on the other, to sustained and successful efforts by the powerful to rubbish and sideline expert consensus opinion whenever it is inconvenient. The proctocrats in power throughout most of the Anglosphere are much given to this.

  28. Yeah, I’m not gonna waste many tears on the poor misunderstood economists until they fight back a lot harder against the misuse of their (genuinely important) field. If you sell your soul to Caesar and Mammon, you have only yourself to blame. Stop trying to cozy up to the rich and powerful, stop pretending you’re a branch of the physical sciences, and start paying attention to how actual people actually interact in the economic realm. (I realize some are already doing that, and good for them. I also realize I know jack shit about it, but it is all of our god-given right to pontificate about stuff we know jack shit about.)

  29. The major grammars treat British and American standard Englishes together without any great problems

    Well, yes, but that’s because they necessarily ignore phonology and phonetics: “necessarily” because unlike most standardized languages, there is no standardized pronunciation of English. (I’m always amused when British linguists refer to RP as a “non-localized pronunciation”, a claim positively American in its unconscious arrogance.)

    We need more and better economics, not less of it.

    Sevenfold amen.

    Stop trying to cozy up to the rich and powerful

    Fred Foldvary, who among other things is a historian of economics (as distinct from an economic historian) says that eco-as-we-know-it results from a devil’s bargain between rentiers and the colleges they founded in the U.S. (notably Cornell) and the people who became university chairmen at those places: defend our dirty deeds with fancy footwork, and we’ll keep you funded.

  30. What stratification economics brings on the scene is a specific view of exactly with whom individuals are comparing themselves. — Who let the peever in the room? That took me four tries to parse… (Separating adverb from pronoun with a relativizer, not in my grammar you don’t).

    Yes, this stuff is blindingly obvious to any four-year-old, in fact kindergarteners spend a lot of time practicing the concepts. But the rational economic agent is such a convenient fiction, like a spherical cow in physics.

  31. Stephen C. Carlson says

    I’ve read the book now, and if lexicography is among your interests, it is a delightful read. As for the definition of “nude” that Cowan quoted, a chapter goes through the process of arriving at that definition, including her taking actual pictures for their citation files of a wide range of skin tones all labeled “nude” on various products.

  32. Yvy tyvy says

    I also realize I know jack shit about it, but it is all of our god-given right to pontificate about stuff we know jack shit about.

    The real problem is when people pontificate about stuff they know jack shit about, but don’t know they know jack shit about it, to an audience of people who also know jack shit about it, while not seeming like they know jack shit about it, when they actually do.

  33. David Marjanović says

    Couldn’t have said it better myself.

    (…Honestly, I couldn’t.)

  34. It’s a perfect example of the introductory “so” that’s become ubiquitous in recent years; I think I posted about it at some point, but it’s the kind of thing that’s hard to search for.

    Here’s that earlier post, and here’s a recent one. As I said here, it’s hard to talk about this development because sentence-initial “so” has been around forever, but used differently.

    This came to mind because the Log has a new post about it, and people are providing the same useless examples of other uses that have been around forever, and it annoys me.

  35. So I don’t really see a clear distinction between the different uses. I grew using initial “So” sporadically, and I do not perceive any particular rules to how it can be used. It can go at the beginning of pretty much any utterance.

  36. There’s a very clear difference between “so” as an implied continuation of preceding discourse and “so” as an empty marker introducing an answer.

Speak Your Mind

*