Language was a mistake?

« previous post | next post »

Today's Dinosaur Comics:

But someone else might conclude that this type of variable scope is actually a Good Thing.


  1. Philip Taylor said,

    May 25, 2021 @ 6:30 am

    "there's no oversights in your oversight" — does not standard English require "there are no oversights …" ?

  2. Cervantes said,

    May 25, 2021 @ 7:48 am

    Of course Lewis Carroll did this all the time.

    Twas bryllyg, and ye slythy toves
    Did gyre and gymble in ye wabe:
    All mimsy were ye borogoves;
    And ye mome raths outgrabe.

    Also Edward Lear, James Thurber. It's a favorite sport of many writers.

  3. Timothy George Rowe said,

    May 25, 2021 @ 7:58 am

    But of course, Lewis Carrol (through Humpty-Dumpty) did explain those words.

  4. S Frankel said,

    May 25, 2021 @ 8:10 am


    table (verb) – 'put on the table for immediate action' or 'remove from the table; put off indefinitely'

    presently – 'at once' or 'later'

  5. Cervantes said,

    May 25, 2021 @ 8:26 am

    Cleave is its own antonym as well.

  6. Bob Ladd said,

    May 25, 2021 @ 9:07 am

    @ S Frankel, Cervantes.

    Also: as far as I can tell and as near as I can tell

  7. RfP said,

    May 25, 2021 @ 9:37 am

    I could care less.

  8. Terry K. said,

    May 25, 2021 @ 9:44 am

    @Philip Taylor

    Depends perhaps on what's meant by standard English. Certainly more formal registers of English require "there are no oversights …" rather than "there's no oversights …" . But my impression is that the "there's" construction is, in American English, relatively standard, even if colloquial.

    [(myl) See "Ask Language Log: There's cookies involved", 3/16/2013.

    Also "Leading questions and frickin' cooks", 8/31/2005; "When 'there's' isn't 'there is'", 9/1/2005. ]

  9. Haamu said,

    May 25, 2021 @ 1:24 pm

    My admittedly limited understanding of hapax legomenon suggests that the concept makes sense when considering a defined or closed corpus, but is more problematic within an active linguistic community. So, T-Rex is going to have a problem (similar to the googlewhackblatt paradox) when people start reusing the term, even if only to note that "T-Rex claims grolitont is a hapax." Beyond that, descriptive linguists are going to tell him that he doesn't get to control what it means. But it's always fun to see his thinking evolve.

  10. Andreas Johansson said,

    May 25, 2021 @ 2:48 pm


    That's one of the perks of being an apex predator: when descriptive linguists tell him that, he eats them.

    I've heard various arguments to the effect of intelligence, consciousness, coming down from the trees, or leaving the water being mistakes, but I don't think I've seen language called so before.

  11. Rob Grayson said,

    May 25, 2021 @ 2:53 pm


    I don't know where you got that verse of Jabberwocky from, but it's wrong: like any competent writer, Carroll used "the", not "ye".

  12. Paul Garrett said,

    May 25, 2021 @ 3:05 pm

    But, Rob Grayson, isn't the common "ye" in modern transcriptions of older English a misrepresentation of a "thorn" or some other archaic letter, so that its pronunciation really should have been (some version of) "the" all along?

  13. Jerry Packard said,

    May 25, 2021 @ 3:13 pm

    To support Haamu's point, my understanding of hapax legomenon is that it is a word that only occurs once in a corpus, used e.g., as a way to estimate the probability that the next token will be one that we have never seen before. (From: Church, Kenneth and William Gale (1991) "A Comparison of the Enhanced Good-Turing and Deleted Estimation Methods for Estimating Probabilities of English Bigrams," Computer Speech and Language 5, 19-54).

    So, as soon as someone uses it again in the corpus (which didn't happen in the comic!), it is no longer a hapax legomenon.

  14. Richard Hershberger said,

    May 25, 2021 @ 3:34 pm

    Cervantes, re Cleave: Not really. First off, to divide something with a cutting blow is not really the opposite of adhering firmly to something. The second involves the person cleaving in a way that the first does not. Also, the adhere sense is a prepositional verb. Dictionaries don't make this distinction, but it is "cleave to," with the "to" mandatory.

  15. Michael said,

    May 25, 2021 @ 3:38 pm

    @Andreas Johansson: John Zerzan made the argument decades ago:

  16. Kenny Easwaran said,

    May 25, 2021 @ 4:44 pm

    I've never understood whether "cloven hooves" are understood as one hoof that is split in two, or two hooves that are stuck together. I think the fact that this one thing can be thought of in both ways helps explain how the word has both meanings.

  17. Chas Belov said,

    May 25, 2021 @ 4:50 pm

    @Kenny Easwaran: I alway took it that each hoof that was cloven constituted one cloven hoof, but very few animals, if any, have only one hoof, and if they have more than one hoof, the hooves are either all cloven or all not cloven. So, we speak of cloven hooves, not of cloven hoof, unless, perhap, being poetic.

  18. Chas Belov said,

    May 25, 2021 @ 4:51 pm


    *perhap (sic)

  19. Peter Taylor said,

    May 25, 2021 @ 5:00 pm

    @Jerry Packard, while hapax legomena certainly can have application in statistics, the term predates that kind of approach to corpora considerably. The oldest reference I can find in Google Books, which will surely be antedated by someone else very soon, is from the late 18th century when the point of interest was that the meaning is easily lost and not so easily recovered from context because you don't have multiple contexts to contrast and compare.

    Vocum in Bibliis semel tantum, aut rarius occurrentium, vulgo: hapax legomenon, aut, stylo Rabbinorum, dictionum, quibus non est socius, (schællo nimtza lahem chhabher) significatio hodie jam incerta; neque enim loca parallela his lumen fœnerari possunt.

    (From Vindiciae Vulgatae Latinae Editionis Bibliorum, Qua Ecclesia Romano-Catholica Utitur, Contra Assertam Hebraie, Et Graeci Textus Hodierni Absolutam Authentiam, György Czuppon. Yes, the main body of the text is in Latin, but I regard that as irrelevant).

  20. Stephen Hart said,

    May 25, 2021 @ 5:22 pm

    According to the Wikipedia entry for Jabberwocky:

    A decade before the publication of Alice's Adventures in Wonderland and the sequel Through the Looking-Glass, Carroll wrote the first stanza to what would become "Jabberwocky"…

    Carroll wrote the letter-combination ye for the word the in order to approximate the Middle and Early Modern English scribal abbreviation variant of the letter Þ (thorn) combined with the superscript form of the letter "e".

    Also of possible interest to Language Log habitués:

    Lewis Carrol's JABBERWOCKY as "recognized" by the Apple Newton

  21. Terry K. said,

    May 25, 2021 @ 5:27 pm

    Let's see if this works (using superscript markup).

    Twas bryllyg, and ye slythy toves

    Wikipedia has the above (and the rest of the verse similarly), with an explanation that talks about thorn. If you cut and paste that, then "ye" becomes "ye". Which is not what Lewis Carrol wrote.

    This page (link below) is the source for that in Wikipedia, and the modern typeface version has the normal "the". But the first version definitely has what looks like a y for the thorn.

  22. Terry K. said,

    May 25, 2021 @ 5:28 pm

    Nope… so, well, imagine the e in superscript, or go to

  23. JO'N said,

    May 25, 2021 @ 5:33 pm

    According to "Family Guy" producer Chris Sheridan:

    [Family Guy] has been on long enough that we did a bit once where we had the word ‘vagina’ in an episode. They said, ‘You can’t say vagina on the air.’ So we changed the word and we just came up with our own word and called it ‘cleman.’ We just swapped out ‘vagina’ with ‘cleeman,’ and it was like, ‘Oh, we’ll just say that. No problem.’ And then several years later, we did a joke, and we use the word ‘cleeman,’ and they’re like, ‘Well you can’t say that.’ And we’re like, ‘What do you mean we can’t say it?’ ‘Well no, it means vagina.’ I was like, ‘I know! We made it up.’ But it made its way into the Urban Dictionary, and then we couldn’t say it. I was like, ‘I think that’s completely unfair.’ So when a show is on that long, it’s just insane.

  24. Jerry Packard said,

    May 25, 2021 @ 7:35 pm

    @Peter Taylor, That is definitely cool.

  25. maidhc said,

    May 26, 2021 @ 1:00 am

    Wasn't hapax legomenon originally applied to words in the Bible? Or possibly Homer? At any rate, texts that are basically fixed for all time, barring some kind of sensational archaeological discovery.

    It doesn't make sense to use it in reference to a corpus that can still be modified.

  26. Peter Grubtal said,

    May 26, 2021 @ 2:24 am


    thanks for the Zerzan link: you've expanded my consciousness considerably.

    I was fully convinced the Zerzan text was a brilliant spoof, a Sokal 3, until I looked him up.

  27. Michael Watts said,

    May 26, 2021 @ 3:49 am

    No, it doesn't make sense to use hapax legomenon in reference to a living corpus. But it also doesn't make sense to use it in reference to a particular text; it is of no interest that a word is hapax legomenon within the Bible if it also occurs frequently in other contemporaneous texts. I expect it to be used to indicate that a word occurs just once within the entire corpus of e.g. all surviving Greek literature.

  28. Philip Taylor said,

    May 26, 2021 @ 3:54 am

    Jabberwocky in a type-style closer to that which appears at the beginning of the Wikipedia article thereon —

    Twas bryllyg, and yᵉ slythy toves
    Did gyre and gymble in yᵉ wabe:
    All mimsy were yᵉ borogoves;
    And yᵉ mome raths outgrabe.

  29. Keith said,

    May 26, 2021 @ 6:53 am

    @Kenny Easwaran

    "Cleave" in the sense "cling or stick to": preterite and past participle "cleaved".

    "Cleave" in the sense "split or divide": preterite and past participle "cleft", "cloven" or "cleaved".

  30. ktschwarz said,

    May 26, 2021 @ 2:30 pm

    And that's because the two "cleave"s go back to different Germanic verbs, which stayed different in modern German: kleben and klieben. The first is a weak verb and goes back to the same Proto-Indo-European root as "cling", "clay", and "glue". The second is a Germanic strong verb in the same class with "freeze" and "choose".

  31. ktschwarz said,

    May 26, 2021 @ 5:28 pm

    … and the "cleave" that means split is cognate with "glyph", which was originally a mark that was carved.

    @Richard Hershberger: Some dictionaries do say explicitly that "cleave" meaning stick must be followed by "to"; even lists "cleave to" as a phrasal verb.

  32. JPL said,

    May 26, 2021 @ 9:30 pm


    "And that's because …." You forgot to mention the main difference maker for Kenny Easwaran, namely that the lexeme 'cleave' glossed as "divide in two" is transitive, while the historically different lexeme with the homonymous citation form 'cleave' glossed as "cling" (roughly) is intransitive; the expression "cloven hooves" is a passive construction (as would be "cleaved hooves").

  33. Michael Watts said,

    May 26, 2021 @ 10:29 pm even lists "cleave to" as a phrasal verb.

    Huh. I thought a phrasal verb was a verb that — itself — included a particle, like "throw away", "eat out", or "heave to". "Cleave to" is not like that — the "to" is not a particle that is part of the verb; instead, the object of the verb must be marked by the preposition "to".

    E.g. you can "throw something away", but you definitely cannot "cleave something to". You can only "cleave to something".

  34. Amanda Adams said,

    May 27, 2021 @ 10:42 am

    I know I'm "wrong" but I cannot resist mentioning than you can cleave someone's head [in two] to his breastbone….

  35. Philip Anderson said,

    May 27, 2021 @ 5:46 pm

    @Amanda Adams
    “Cleave in twain” is the usual idiom I think.

  36. R. Fenwick said,

    May 28, 2021 @ 7:58 am

    @Kenny Easwaran: I've never understood whether "cloven hooves" are understood as one hoof that is split in two, or two hooves that are stuck together.

    Folk analysis at least in the West pretty universally treats it as a single hoof that is split in two, these usages no doubt influenced by Bible translations from the Hebrew הַשְּׁסוּעָ֑ה הַפַּרְסָ֖ה ha-parsāh ha-šəsū‘āh "the split hoof, the cloven hoof" (Deuteronomy 14:7). Interestingly, Klein's Hebrew etymological dictionary derives פַּרְסָ֖ה parsāh itself from the verb פָּרַס pāras "to divide, to cut, to cleave" – indeed, later in the same verse the Hebrew explicitly uses this cognate verb in פַרְסָה֙ לֹ֣א הִפְרִ֔יסוּ parsāh lō hiprîsū "they do not have cloven hoof".

    (Anatomically, there's not really any such thing as a cloven hoof per se. Even-toed ungulates have two distinct weight-bearing digits, each terminating in its own hooflet. In fact, because the digits are completely separate, hoof infections in cattle can often be effectively treated by simply amputating the infected digit.)

  37. SusanC said,

    May 28, 2021 @ 12:03 pm

    With a slight nod to Wittgenstein’s private language arguments … if a word has only ever been used once, like in the comic, does it mean anything at all?

    On the other hand with historic texts … you have words that clearly once were used lots of times by a community of speakers, but now you’ve only got one surviving instance, which possibly isn’t enough to figure out what it used to mean.

RSS feed for comments on this post