Dogged by an etymological shape

« previous post | next post »

[This is a guest post by Martin Schwartz]

The following is just an idle speculation for which I have no answer, but somehow I don't think mere coincidence is really a factor.

A number of Old World languages of different groups show a word for 'dog' or a doglike beast of the type affricate/sibilant plus /a/ (plus vowel) plus l/r.:
Basque txakur /čakur/ 'dog'; see Wiktionary, where under "descendents" Romance and Turkish(!) comparanda are given.
Kartvelian (grosso modo) dzaGHl- 'dog'.

Sanskrit śRgāla- etc.(R syllabic r)  'jackal' (hardly with Wiktionary conn. with the 'lion' words MidPers. šaGHr, Khwar. sarGH, Sogd. šarGHu etc.), but somehow with (Mid)Pers. šaGHāl 'jackal'.
(using 6 for Semitic 'ayn:) Yemenite Arabic THa6al, Aram. ta6(àl-ā, Heb. šū6āl 'fox', Arab. THa6lab, Acc. šēlebu 'fox' (-b, cf. PSem. *kalb 'dog' and *∂i'b 'wolf' ? I have considered, and rejected, that *-ab is a PSem. animal suffix.
That's it.

Selected readings



  1. Annie Gottlieb said,

    September 21, 2023 @ 11:35 pm

    "Etymological shape," eh? Is that what you would call my observation that the words for outsiders—gajo, gaijin, goy(im)—are so similar in Roma, Japanese, and Yiddish?

    Of course, all three words could originate from the same source in Europe, and could have been carried to Japan by the Portuguese, like "tempura" (but apparently not "arigato").

  2. martin schwartz said,

    September 22, 2023 @ 12:25 am

    @Victor Mair: Thanks for being the post host with the most
    (doggone, that "Of dogs…" is one long post indeed, into which
    I see I [ahem] em-barked sufficient times).
    @Annie Gottlieb: Naw, I wouldn't sayl your 'outsider' words
    have the same or a similar "etymological shape" (ood phrase, that!).
    The Japanese word is from a Chinese compound on which there are for more learned heads that I on this bLog (hmm, bLog looks Tibetan); Romani 'gadjo' may be from a Midddle Indic derivative of
    Sanskritic garhya- or gārhya- 'domestic', or so some say–I dunno–
    and Yiddish goy (parallel to 'gentile' < Lat. gentilis, and cf. Lat.
    paganus 'rustuc') is from Heb. 'nation' (as in the prophet's
    'nation shall not lift up sword against nation'); in the plural, 'the [other] nations' (gõyīm) gets to mean 'the Gentiles'. From the large
    Jewish community of Georgia in the Caucasus, goyim-i gets to mean
    'a bumpkin' in general colloquial Georgian; I've noticed the pl.
    goyim used for sg. in some varieties of Judeo-Iranian jargons.
    Martin Schwartz

  3. martin schwartz said,

    September 22, 2023 @ 12:27 am

    I meant "ODD shape".

  4. AntC said,

    September 22, 2023 @ 4:18 am

    a word … of the type affricate/sibilant plus /a/ (plus vowel) plus l/r.

    That seems a wide net. As a comparable metric, how many words in those languages fit that pattern, irrespective of meaning?

    /čakur/ , dzaGHl

    Those examples seem to have not merely (plus vowel) but (+ velar + vowel) ??, so that widens the net further. Can we be sure they're not related?

    And how about the languages whose word for dog(-like) _don't_ fit that pattern — like errm 'dog', 'hound', 'canis'/ svan- / *kwon- ?

  5. Chris Button said,

    September 22, 2023 @ 8:43 am

    A while back on LLog, I suggested a possible link with 豺 EMC dʐəɨj “jackal, dhole”. It reconstructs in OC with a rhyme -rəɣ in accordance with its phonetic speller 才, but a rhyme -rəl would have given the same EMC reflex. Depending on when the Chinese character was coined, it could reflect that intermediary stage of convergence.

  6. Cervantes said,

    September 22, 2023 @ 8:52 am

    Could just be cherry picking. There are innumerable concepts to be rendered as nouns, you will always be able to find purely coincidental similarities.

  7. Pamela said,

    September 22, 2023 @ 9:21 am

    I think that's interesting, and appreciate the spirit in which it is offered–musing has its place, and can be powerful. Now, what about a semantic shape for "other people"? It is very common for autonyms to just mean "people," but how does one construct non-people? I'm responding to Anne's note that "gõyīm" means "other nations/people" and I think immediately of my childhood when certain adults in our Mennonite community referred to non-Mennonites as "anner leite" (andere leute), just meaning "other people." English speakers among us called the people "English" and today they are called by police Mennonites "the public." It is a "gentiles" semantic shape that differs from a "barbarians/dogs/devils/corpses" semantic shape, surely?

  8. Windowless Monad said,

    September 22, 2023 @ 5:33 pm

    And in Australian languages:

    "…Mbabaram word for "dog" was in fact dúg, pronounced almost identically to the Australian English word (compare true cognates such as Yidiny gudaga, Dyirbal guda, Djabugay gurraa and Guugu Yimidhirr gudaa, for example[3]). The similarity is a complete coincidence: there is no discernible relationship between English and Mbabaram." (Wikipedia)

  9. martin schwartz said,

    September 23, 2023 @ 1:46 am

    Yes there are coincidences, and I am aware of many, beginning with
    Persian and English [bæd] 'bad', and in the "doggophonic" realm
    the fact that Dagon is a god both in the Nuristani and Semitic realms; the respective history of the words in each set show
    that there is no etymological relationship involved. What I set up merely for discussion are words,in numerous languages hat are broadly similar in their TRiconsonantal form, and it is possible that the domesticated dog provided a Kulturwort that became widely distributed (the 'jackal' and 'fox' words may be a different story). I
    strongly suggest, before judging, that those seriously interested
    have a look at Václav Blažek's remarkable "Hic sunt leones".
    I thought that Language Log would be a good forum for the matter,
    because of earlier postings on words for 'dog' to which Victor Mair
    calls attention above.
    @AntC: "Can we be sure they're not related?" No, we can't–my point. But then you imply that real relationship between words
    is contradicted by words of other phonic shapes–HUH???? Btw
    have a look at earlier discussions on LL as to the problematic phonology of Lat. canis vis-à-vis Gr. kúōn, Skt. śvan-, etc.
    @Pamela: The remark about Heb. gōyīm 'nations' becoming
    'other nations, outsider was mine (as was the typo õ) . Both you and Ms. Gottlieb should understand that I was talking about
    phonetic shapes, not semantic shapes. But accounts of words for ethnic outsiders is an interesting topic–I think of Armenian otar,
    odar, which has, I think, a Parthian origin, and of course Spanish gringo–no doubt Wiktionary has something interesting on
    those. And to return to the original subject, sure, we could be dealing with synchronous accidents; I have no dog in this race.
    Martin Schwartz

  10. martin schwartz said,

    September 23, 2023 @ 1:49 am

    I meant "synchronistic", i.e. randomly accidental (if that's not
    a pleonasm).

  11. Yves Rehbein said,

    September 23, 2023 @ 10:38 am

    The described etymological "template", as some would call it, loosely matches PIE *h₂ŕ̥tḱos better than *ḱwṓ. Compare ursus, ἄρκτος (árktos), خرس‎ (xers), ऋक्ष (ṛ́kṣa). This comparison is a des-aster, a kat-astro-phe, of cosmic proportions.

    Compare Ursa major:

    > Ursa Major (/ˈɜːrsə ˈmeɪdʒər/; also known as the Great Bear) is a constellation in the northern sky, whose associated mythology likely dates back into prehistory.

    Ursa minor:

    > The ancient name of the constellation is Cynosura (Greek Κυνοσούρα "dog's tail"). The origin of this name is unclear (Ursa Minor being a "dog's tail" would imply that another constellation nearby is "the dog", but no such constellation is known).

    An analysis of the various assembled folk believes is, of course, not in my power.

    Since you have mentioned PSem *∂i'b 'wolf', eg. Arab. ذِئْب‎ (ḏiʔb), Egyptian Arabic ديب‎ (dīb), the least I can say is that the so-called Big Dipper is reminiscent of i. "α Ursae Majoris, known by the Arabic name Dubhe ("the bear"), ii. Běidǒu 北斗 "the northern dipper".

    On another note, since goy (foreigner) and gaijin (big man, i.e. foreigner) were mentioned and I was just looking at Russian gav-gav as well as Indonesian guk guk, Vietnamese gâu gâu and an infinity of different words which translate woof ("the sound of a dog barking"), I find it remarkable that Akkadian barbarum ("wolf") and Sumerian /urbarak/

    > From (ur, “dog”) +‎ (bar-ra /⁠barak⁠/, “of outside”), literally “outsider dog”

    resemble Greek βᾰ́ρβᾰρος "barbarian"

    > Onomatopoeic: from the perceived βαρ-βαρ (bar-bar) sounds incomprehensible to Ancient Greeks and spoken by foreigners

    Well duh!

  12. martin schwartz said,

    September 23, 2023 @ 11:50 pm

    @Yves Rehbein: Thanks for illuminating cynosure, a word about which I was fuzzy; its etymology is not a sinecure.
    BUT: I'm not sure how to take your etymological conSIDERAtions.
    You seem to be saying that the PIE 'bear' etymon looks more like
    the phonetic shape of my very putative 'dog' Wanderwort than the latter look like the PIE *k'won- 'dog' (I wish people did not follow Wiktionary's bad decision to cite words by nominative form rather than stem). But I prefer to keep my already questionable beast on a tight leash and not have it mate with an unbearable bear. Your
    stellar ursalities are out of my sphere . Last I knew, though,
    the Arabic word for 'a bear' is dubb. Entertainingly learned is
    the online Phililogical Crocodile on (alleged) bear taboo;it mentions
    Hittite hartugga, from which Wiktionary shies away.On your
    other note, I note that, as reported by Elias Petropoulos' Kaliardá,
    a book about artsy erstwhile(?) Greek homosexual café argot,
    in Kaliardá, dzungloGHuGHulfú = Metro-Goldwyn-Mayer;
    this analyzes as a word with -ú Kal. institutional suffix = Gr, -ía;
    dzunlo- = jungle, GHúGHulfos = dog (< Eng. 'wolf' x GHav GHav,
    the sound of barking); jungledog ='lion'; company of the lion
    = MGM. I doubt that BéoGHhoolf has a better kenning.
    Martin Schwartz

  13. Sophie said,

    September 26, 2023 @ 10:25 am

    Martin, Fascinating post! In “goyim used for sg. in some varieties,” does “sg.” mean “something “?

  14. Alexei Savchenko said,

    September 30, 2023 @ 10:35 am

    I have noticed that the East Iranian dog, *kuta- does not have a reliable etymology. Steblin-Kamensky explains kuč-kuč in Wakhi as a call addressed to a dog; Abayev assumes the same for kut-kut in Ossetic. That seems to me Volksetymologie, given dog’s close relationships with man since the beginnings of history. Also, I have met at least three different ways to call a dog in different corners of Tajikistan.
    Am I missing something harvesting the grapes in my village?

    Alexei Savchenko

RSS feed for comments on this post