I always assumed burnoose was from Arabic, and so it is: “From French burnous, from Arabic بُرْنُس‎ (burnus).” But where is that from? Aha:

Via some Aramaic form (Jewish Babylonian Aramaic בורס‎, Classical Syriac ܒܢܪܘܢܐ‎) from Byzantine Greek βίρρος (bírrhos), from Latin birrus, from Gaulish, from Proto-Celtic *birros (“short”). Doublet of بِرْنِيطَة‎ (birnīṭa).

(That last transcription should read birrīṭa; someone who can edit Wiktionary should fix it. Also, Proto-Celtic *birros gives Welsh byr ‘short,’ as DE will be the first to tell you.)

Anthony Ossa-Richardson, who sent me the link, said “now that’s a well travelled word,” and I can’t disagree.


  1. CuConnacht says

    Latin birrus is also behind Italian (and English) biretta and French (and English) beret.

  2. David Marjanović says

    Awesome, but how did -rr- come to be reinterpreted as -rn-, and how did -īṭa happen?

  3. That last transcription should read birrīṭa

    Eh? No it shouldn’t, that would be بِرِّيطَة‎.

  4. Woops, of course you’re right.

  5. David Eddyshaw says

    The Welsh Substratum strikes again!

  6. PlasticPaddy says

    Acc. to Wiktionary birnīta is a later borrowing from Italian beretta and there is an alternative birrīta. So if the form was originally *bir(r)eīta, then an n could be inserted between e und ī to prevent vowel collision, alternatively the e could be dropped, giving *bir(r)enīta and birrīta, with *bir(r)enīta simplifying to birnīta.

  7. David Eddyshaw says

    I would imagine that the most successful word of ultimately Celtic origin in our current world is (by some margin) car, having long since ousted cheval and its Romance cousins from the top spot.

  8. Interesting. Not sure about the intermediary though; it’s striking that neither of the Aramaic forms cited can provide the missing link between the Latin/Greek and Arabic. I can’t find בורס‎ in Sokoloff to begin with, and I think the Syriac form is a typo; the closest I can see in CAL is:

    byrwn, byrwnˀ (bīrōn, ābīrōnā) n.m. toga, cloak. also: ܒܪܘܢܐ‏
    1 toga, cloak Syr. Pel 10.12 : ܟܘܬܝܢܐ ܕܣܥܪܐ ܘܒܪܘܢܐ ܕܥܡܪܐ‏ .
    Apparently from Latin birrus, birrum through Greek βίρρον “a cloak to keep off rain”; see A. Butts, Hugoye 19(2016): 145..

    A lot of North Africans seem very attached to the idea that this word reached Arabic through Berber abeṛnus. The distribution suggests otherwise, but in the absence of a decent Aramaic source it can’t be ruled out entirely.

    Looking at CAL, I notice gullā “cloak, hood” and glīmā “cloak, mantle”; it’s tempting to take that as the origin of Maghrebi Arabic gəlmuna “hood”, Tuareg tagəlmust “head-cover”, and other North African forms. But I wouldn’t have expected the g to correspond to g…

  9. There is a typo in the Wiktionary entry! For those who can’t read Syriac, the Syriac form should be ܒܝܪܘܢܐ ⟨byrwnʾ⟩ “toga, cloak, patriarch’s chlamys” and not ܒܢܪܘܢܐ ⟨bnrwnʾ⟩ as it is currently spelled in the Wiktionary. (Syriac ܝ y and the medial form of ܢ n, which differ mostly in height of the “tooth”, were confused.)

    The Mishnaic Hebrew form בִּירוֹס and the like have their own problems. See these entries in Jastrow’s dictionary:



    Sokoloff in A Dictionary of Jewish Palestinian Aramaic of the Byzantine Period (1992, p. 101) also gives ביר “a type of cloak”.

    Butts’ article mentioned in Lameen’s post is here:


    On page 146, he accounts for the appearance of the n in Syriac by taking it from the Greek accusative.

    I wonder, did Arabic burnūs arise from a crossing of an Aramaic *birōnā (the Syriac type) with an Aramaic *birōs(ā) (the Mishnaic Hebrew בירוס type)?

    The Greek etymon has a geminate r. I wonder if an intermediate Aramaic *birrōs could have undergone dissimilation of an original rr to rn within Arabic. For this possible sporadic change, which not everyone accepts, possibly compare ذروح ðurrūḥ- “Spanish fly, cantharides” with by-form ذرنوح ðurnūḥ- (among other variants) and Arabic خروب ḫarrūb- “carob” with by-form خرنوب ḫurnūb- (among other variants).

    I wish I could write and research more, but my arm is injured and I am typing with voice recognition and my left hand.

  10. gullā glīmā

    Must be κάλυμμα (Wiktionary).

    (words for articles of clothing have all reasons to be similar, and of course Greek contacted with “Romance” North Africa, but I can’t obtain tagelmust from κάλυμμα either)

  11. David L. Gold says

    Since it is highly unlikely that Jewish dressways have influenced the dressways of non-Jews in Arabic-speaking countries or in countries with Muslim majorities, Jewish Aramaic of any kind probably does not figure in this etymology.

    A few weeks ago I mentioned Federico Corriente’s significant contributions to the study of Arabic influence on the languages of the Iberian Peninsula.

    His publications on the subject not being within easy reach at the moment, I see that the entry for Spanish albornoz in the dictionary of the Royal Spanish Academy, where the etymologies of Arabic interest are his, offers this etymology:

    Del ár. hisp. burnús o barnús, este del ár. clás. burnūs, y este del gr. βίρρος bírros.

    Thus, < Hispanic Arabic burnus or barnus (both finally stressed) < Classical Arabic burnūs < Greek βίρρος [bírros].

    A link headed Maghrebine Arabic should presumably be added between the Hispanic and Classical Arabic ones.

  12. Given the early and widespread attestation of the word in Arabic, the generally unconvincing Aramaic connections, and the fact that the best of the latter is in Mishnaic Hebrew, my bet would have to be on a direct transmission from Greek to Arabic somewhere near Jordan, perhaps in a Nabataean context. The only seeming advantage of postulating an Aramaic intermediary is to explain the nasal dissimilation, and, as Xerîb points out, that can be accounted for adequately Arabic-internally. Wonder if it turns up in Safaitic…

  13. David Marjanović says

    I wonder if an intermediate Aramaic *birrōs could have undergone dissimilation of an original rr to rn within Arabic.

    Long consonants are expected in certain morphological conditions in Semitic. (I have no idea which ones.) Could that be the reason?

  14. It occurs to me that the plural of burnoose should be Bernice.

  15. Maghrebi Arabic gəlmuna “hood”, Tuareg tagəlmust

    I hope LH readers can see p. 1058 in Federico Corriente (2017) Dictionnaire du faisceau dialectal arabe andalou on these words under qlms and qlmn here in Google Books:


    Dozy, Supplément aux dictionnaires arabes, p. 400, on these words:


    Dozy refers to Simonet, Glosario de voces ibéricas y latinas usadas entre los mozárabes, p. 77 here:


    For the forms with -s, the other candidate that Corriente mentions is represented by Syriac here in CAL:


    Jastrow on Post-Biblical Hebrew כְּלָמוּס (< χλαμύς):


    Apologies for the brief comment and profusion of links, but I have to research by voice recognition because of an injury. Maybe others can locate more recent accounts with this.

  16. All those links work for me; thanks very much, and guéris vite!

  17. I just remembered that


    means Bathrobe.

  18. Stu Clayton says

    One transitional form not discussed so far could be burrito. Asses and eats were common features of many civilizations.

  19. January First-of-May says

    One transitional form not discussed so far could be burrito. Asses and eats were common features of many civilizations.

    Wiktionary says that the asses in question were named for their brownish color, (root-)cognate with English fire. I’m not sure how plausible that is.

  20. David Eddyshaw says

    The derivation of burnoose form a word meaning “short” seems, on the face of it, a bit lucus a non lucendo. Presumably βίρροι were midi-length …

  21. >One transitional form not discussed so far could be burrito. Asses and eats were common features of many civilizations.

    There’s already the well known derivation djellaba < jellybean across the same semantic fields.

  22. I think the form בורס in the Wiktionary etymology that no-one could find is a conjecture that the editor of the Wiktionary entry took out of context from the top of page 51 in Fraenkel (1886) Die aramäischen Fremdwörter im Arabischen, here:


    This is the word handled in the entry for בִּרְסִין in Jastrow that I linked to above:


    Maybe someone who is familiar with the Wiktionary tagging system can repair the etymology there.

  23. “Presumably βίρροι were midi-length ”

    @David Eddyshaw, the main Russian word for anything that you (1) take off when enterign a house (2) does not cover your knees and (3) is not made of wool is kurtka.

    Куртка, is you want to google images (but the image included in Russian Wikipeadia page is a modern Wikidedia borrowing. 🙂 )

  24. Куртка being “Derived from Latin curtus (‘short’).”

  25. If so, then Polish (they borrow from Latin).

    French Wiktionary proposes ‘persan کُرتا, gurtā.”. They must mean کرته (Wiktionary) kurta (-pedia).

    And Vasmer/Trubachev:

    ку́рта (Радищев 28), укр. ку́ртка, ку́рта. Заимств. через польск. kurta, kurtkа – то же – из лат. curtus “короткий”; см. Маценауэр 230 и сл.; Брюкнер 284; Бернекер 1, 649 и сл. Ср. диал. короты́шка – то же, смол. (РФВ 62, 213), под влиянием коро́ткий. Едва ли правильно произведение из тур. kürtä “короткое платье” (Мi. ЕW 148; ТЕl. 2, 113; Доп. 2, 161; Тиктин 3, 1398; Горяев 176; Доп. 1, 24). Тур. слово, согласно Радлову (2, 1461), заимств. из русск. ••

    [Но ср. еще осет. kyræt / kuræt “бешмет”, согд. kwrδ᾽k, qwrtу “рубашка”, перс. kurtа “tunica interula”, лезг. kurt “короткая шуба”, возм., ир. происхождения; ср. κόρτην “парфянская одежда” (Гесихий); см. – без приведения русского слова – Абаев, Ист.-этимол. словарь, I, стр. 609. – Т.]

    Hm. It never occured to me to look towards Iran. I only knew this word from its modern use and hardly any old texts, so my first idea was “French”. Then the logical question is since when it is attested in Slavic.

  26. Just for refrence, برنس in Arabic dictionaries (I do not know maybe there are more convenient versions of such collections in HTML. ) The one in English is Lane, it is a compilation of others: ” [some say] it is from البِرْسُ, meaning “cotton,” and the ن is augmentative: or, accord. to some, it is not Arabic. (TA.) “

  27. Trond Engen says

    Except for breeches, all names of clothes derive from words meaning “cut”.

  28. David L. Gold says

    @ Xerib “I think the form בורס in the Wiktionary etymology that no-one could find is a conjecture that the editor of the Wiktionary entry took out of context from the top of page 51 in Fraenkel (1886) Die aramäischen Fremdwörter im Arabischen, here:”

    Fraenkel’s conjecture is misprinted in Fraenkel 1886.

    The top of page 51 in Fraenkel 1886 reads “Eine entsprechende aram. ist erhalten in בורכסין [here follows a bibliographical reference], wofür als Singular בורס anzusetzen ist.”

    That is, ‘A corresponding Aramaic form has been preserved in בורכסין , the singular of which is to be presumed to be בורס’.

    When the masculine pluralizer ין- is removed from בורכסין, what remains is not four letters (בורס) but five (בורכס).

    בורס is therefore a misprint and no wonder that nobody can find it in any primary or secondary source other than Fraenkel 1886 .

    To the list of secondary sources mentioned in earlier posts that do not have an entry for what we now know to be a misprint should be added Jakob Levy’s Neuhebräisches und Chaldäisches Wörterbuch über die Talmudim und Midraschim (with additions by Heinrich Leberecht Fleischer), 4 vols., Leipzig, F. A. Brockhaus.

  29. בורכס
    Googles as a variant for bourekas. Which returns us to burrito.

  30. Tim May says

    It occurs to me that the plural of burnoose should be Bernice.


  31. David Marjanović says

    There’s already the well known derivation djellaba < jellybean across the same semantic fields.

    Hey, every djellaba is jellybean-colored.

    (And so is everything else.)

  32. That is, ‘A corresponding Aramaic form has been preserved in בורכסין , the singular of which is to be presumed to be בורס’.

    I don’t understand what you are talking about. The top of p. 51 in Fraenkel 1886 reads בורסין as far as I can tell. In poor scans, the samekh come out looking like a kaf. Here is a good scan:


    The בורסין is quite clear. If there were a kaf in the word, Fraenkel’s discussion wouldn’t make any sense. Here, at the bottom of page 138, is Sachs’ discussion, which Fraenkel cites:


    Did these forms really only catch scholars’ attention with Zuckermandl’s 1882 edition of the Tosefta?
    Levy/Fleischer 1876 still treat this form under בדס with ד (sic, rather than ר ) here, volume I, p. 194:


    If I interpret it correctly, Jastrow gives בּוּרְסִין as the reading for the Jerusalem Talmud, where the Babylonian Talmud has ברסין :


    There is some word-play involving b and r and s between ברסין “birrus blankets” and ברדסין “Brindisian blankets” it seems.

    Apologies for not being clearer or more elaborate—I can’t write more because of an injury.

  33. The derivation of burnoose form a word meaning “short” seems, on the face of it, a bit lucus a non lucendo. Presumably βίρροι were midi-length …

    Not all skirts and shirts are still short… :^D

  34. Oh. HOW? I mean, how come I did not think about them:/ But let us add shorts.

    P.S. Except for breeches, all names of clothes derive from words meaning “cut”.

  35. David L. Gold says

    @Xerib. You are right. The better scan you have provided shows a samech.

  36. Thanks once again to Xerîb for great contributions. I love the idea that the Tuareg tagelmust is etymologically the Greek chlamys, though I still find the g puzzling…

  37. Not all skirts and shirts are still short…

    Neither are kirtles/kyrtlar.

  38. John Cowan says

    Hat is not related to any word meaning ‘cut’ that I know of. The Welsh cognate is caddu ‘provide for, ensure’; that is, hats protect the heads they sit upon.

  39. In a play by A.Ostrovsky with Nevezhin Whim (Блажь) there is a strange line, Это значит из шляпки бурнус делает = It means, makes a burnous out of a [woman’s] hat. It is used as a common saying, but I cannot figure out the meaning. It is not a “mountain out of a molehill” and doesn’t seem to be “silk purse out of a sow’s ear” maybe just the opposite.

  40. PlasticPaddy says

    Хотела я из своей кацавейки жилетку скроить: куда! и половины не выходит.

    Н.А.Некрасов. Двадцать пять рублей. 1841.

    В упомянутом выше рассказе Некрасова просторная кацавейка оказывается недостаточной для того, чтобы сшить жилет…Такой прием был широко распространен в русской бытовой культуре и отражен в художественной литературе прошлого века, когда из какого-нибудь предмета одежды предлагается сделать заранее невозможное. У А. Н. Островского, например,— «из шляпки бурнус сделать» («Светит, да не греет», 1881).

    Source: http://costumer.narod.ru/text/o/katsaveyka.htm

  41. The context is cattle, why they have 1/3 of what they used to have.

    Сарытова. Он переменяет породу.
    Бондырева. Скажи лучше — переводит. Это значит из шляпки бурнус делает! Пропадешь!


    “He changes breed”.
    “You-say better – [he]-переводит [it]. [It] means, [he] makes out-of hat-DIM burnous. You-will-get-lost”.

    Переводить “lead over” can mean today (!!!):

    – to translate
    – [about a resource, say, sugar] to waste
    – не перевелись ещё рыцари (настоящие мужики, … )! “are not extinct yet knights (real men, …)!”
    – in technical use in the context of breeds it is spoiling.

    Выводить (вы- “out”) is fighitng with cockroaches (or stains on a dress:).

    Завести is the opposite of the above, it is to start it: you can завести a freind or a dog.
    Завестись (reflexive) is when animals appear and begin living on your land (or parasites in your hair).
    It is a funny verb: “something living appeared out of nothign and now lives here” – this is how it feels.
    Заводчик – is a professional breeder

    Разводить (раз- “spreading out”) to breed.
    Also to increace in number/volume (can be applied to “mess” as in your room).

    пере- here is reducing in number of quality, making few out of many. Either it was used technically as in today, in the sense of reduction in quality (which explains fashionable woman’s hat – [crude?] burnus) – or in the sense of reduction in quantity, which I think is the original meaning of пере- in the verb that is applied to living creatures in many meanings. This usage matches the context.
    Or both.

  42. Ah, and I forgot to mention!

    In 19th cenutry Russia burnous was a familiar women’s (and men’s too) sort fo coat)))))

    I maybe was wrong about [crude?] above.

  43. Кирсанова Р.М. Костюм в русской художественной культуре 18 – первой половины 20 вв
    М. 1995

    p 55 of the pdf file

    (a book about dress in Russian art with pictures)

  44. PlasticPaddy says

    Thank you for explaining the difficult (for me, probably not for languagehat) Ostrovski text. For the Nekrasov line only the quantity interpretation is possible, N is explicit about this (even if he were not, the reduction is from katsaevika to zhiletka, and zhiletka sounds more fashionable, because French gilet)

  45. David Eddyshaw says

    The Welsh cognate is caddu ‘provide for, ensure’

    Welsh caddu is a figment, as far as I know.
    Cadw, perhaps?


  46. PlasticPaddy,

    I think “переводит” means reduction of cattle in quality or quantity. I have no idea which one.

    But burnous and shlyapka (diminutive of “hat” that is, women’s hat, lighter than men’s. Cf. also trusiki girly ‘panties’, < trusy, underpants. Or “panties”) both were fashionable. I thought “crude” because “burnous” sounds heavy in Russian.

    ًWhen I google it, I see many (really) girls “in a hat and burnous“. Jeans and T-shirt: me. Hat and coat: a spy ????️. Shlyapka and burnous is apparently a girl.

    But burnous is very large and expensive compared to a hat (especially straw hat). I think, the opposition is [negligible thing with diminutive suffix] [substantial thing], “making a coat out of hat”, both things familiar, both are what a girl puts on, and both are what a girl looks like form ditance. So you and D.O. must be right.

  47. Благодарю тебя очень за купальный бурнус лучшего и полезнейшего подарка ты мне сделать не могла. 27. 4. 1877. Ф. Кони – А. В. Каировой.

    “Thank you very much for bathing burnous, you could not make a better and more useful gift”. ( F. Koni to his wife, a war correspondent, by the way*.

    bathing burnous:/ **

    *…что такое «специальный корреспондент» русской газеты… это несчастнейшее существо в мире, нравственная тряпка, обязанная «ловить момент» и сообразоваться со всем на свете, кроме своего личного убеждения, а подчас и истины. Я не виню за это русские газеты. Может быть, они и сами подчиняются не своей воле, а вынуждены грустным положением своим требовать от корреспондентов не того, что есть, а того, что в данную минуту желают, чтобы было. Но нам, корреспондентам, от этого не легче. Мы-то все-таки поставлены под двойной гнет двойной цензуры и вынуждены говорить, когда хотелось бы молчать, и молчать, когда совесть велит говорить…

    if it is a variation of “bathing khalat” (modern Russian for “Bathrobe”), why not.

  48. ** if it is a variation of “bathing khalat” (modern Russian for “bathrobe”), why not…

  49. David Marjanović says

    hats protect the heads they sit upon.

    Mein Hut heißt Hut, weil er mich behütet!

  50. Welsh caddu is a figment, as far as I know.

    Yeah, I think so… See pages 67-68 in J. Loth (1925) “Notes étymologiques et lexicographiques”, Revue Celtique 42:


    You can find caddu in Ascoli (wherever he took it from) in his handbook Lezioni di fonologia comparata from 1870. It was also perpetuated in Walde’s Lateinisches etymologisches Wörterbuch and in Walde-Pokorny too:


    (This even after Loth’s identification as a ghost word!) From Walde-Pokorny I imagine got got sucked up into Sergei Starostin’s online databases, which were then used by Wiktionary editors.

  51. David Eddyshaw says

    Ah, a misreading of a Middle Welsh spelling kadu for cadw. Makes sense.

    It’s interesting (and perhaps facilitated the error) that the -d- is unexpected if the word derives from *kadʰ-.

  52. David Marjanović says

    But then, the existence of *kadʰ- is deemed improbable on pages 4, 5 and 20 of this paper. (The passages in question would be short enough to cut & paste, but the paper is scanned as a picture, so I can’t.)

  53. David Eddyshaw says

    Makes even more sense. So the Welsh simply continues the root *kat-.

    I notice that in the same place the paper also mentions the root *kad- behind the Welsh cas “hateful”, that I mentioned in passing when we were talking about cesair “hail” the other day.


  54. David Eddyshaw says

    The other ever-popular Brythonic *kat- root is of course the one in cad “battle”, that may or may not be present in Catraeth/Catterick. Is there anything it can’t mean? I’m beginning to wonder if Marr should have added it to sal, ber, yon, rosh. Such is the power of the “Hat” etymon.

  55. PlasticPaddy says

    In Middle Irish there is the verbal noun cadádh meaning covering in the sense of wrapping. But the ca there is supposed to be from *kom. Is this possible in Welsh, or would that give, say *cydw instead of cadw?
    There is also (unrelated) scáth meaning shadow or protection;
    cáid meaning holy (you seem to use sanctaidd or glan, we have naofa for sanctaidd, glan means clean);
    cath meaning battle/fight –cathugu ( modern cathú ) also means temptation.
    The best is cais, corresponding to Welsh cas. eDil explain this as “Expld. both as love and hate , and in texts capable of either meaning acc. to context:”

  56. David Eddyshaw says

    The Welsh equivalent of *kom- is cyn-; cf cynnadl “conference” (nowadays cynhadledd) where the second component is dadl “debate, argument”, the same etymon as the Irish dáil. Welsh doesn’t do the wholesale scrunching of originally word-internal vowels that Irish does, so the syllable after cyn- keeps its full vowel, eg cyngad “conflict”, from *kon-katu.

    What is the second component of cadádh? Presumably the -d- derives from *-nt- or *-nd-, in the usual Irish fashion?

  57. David Eddyshaw says

    (Actually, not *-nd-; that gives -nn-, of course.)

  58. David Eddyshaw says



    Verbal noun of con-toí (*com-to-so-) “turn, convert”, it says. I can’t think of any Welsh cognate. Welsh uses troi for “turn, convert”, but this also has the possible meanings “curl, wrap around”, so there is at least a parallel in the semantics.

    Glân is primarily “clean” in Welsh, too, but extends as far as “utter, complete” (like “clean forgot” in English) and “pure”, hence also “holy.”

  59. PlasticPaddy says

    toí is churning. While I cannot really exclude a connection with Welsh troi, I think the latter is related rather to a word for (the curved part of) the foot, troedd Welsh, trácht Irish. The word treo (direction) is probably borrowed from the Welsh. But there is also the preposition tre (modern trí) which means through, and you could possibly try to make a noun from it meaning direction or a verb meaning convert. But i have not seen any attempt to explain any of these tr words that way.

  60. David Eddyshaw says

    Welsh troed “foot”, pl traed, has lost medial /ɣ/ (as normal) since the Old Welsh period: it’s plainly related to Old Irish traig, genitive traiged. I don’t know what the deal is with the vowel in the Welsh singular. Irish trácht must be a different formation. The Welsh equivalent of that would have been traeth.

    GPC doesn’t hazard a guess about the etymology of Welsh troi beyond Brythonic; Wiktionary (along with some astonishing nonsense about Troy) suggests a connexion with τρόπος etc, which is an interesting thought; I think it works phonologically.

    Wiktionary claims Welsh trwy “through” is from PIE *terh₂ “cross over”, and thus ultimately related to traws “across” and Latin trans. It’s unproblematically the same etymon as the Irish tre/trí, anyhow.

    There is a Welsh verb toi “roof, thatch, cover, hide”, which is etymologically transparent enough: it’s obviously from the *teg/*tog root which turns up all over, eg. Latin toga and the Welsh , Old Irish tech “house” etc etc. To say nothing of “thatch” …

  61. Which came up here.

  62. PlasticPaddy says

    Tràcht/tráig is a doublet in Middle Irish; both can mean “beach” (Welsh traeth) or “part of the foot”

  63. Verbal noun of con-toí (*com-to-so-) “turn, convert”, it says. I can’t think of any Welsh cognate.

    For a reflex of the same root, PIE *sewh₁-, in Welsh, I would check the GPC under amheuaf (amau):


    Also here:


  64. David Eddyshaw says

    Thanks, Xerîb. I must say that amau would never have occurred to me as being cognate with con-toí … (even if I hadn’t foolishly overlooked the fact that the to in con-to-so- is a second prefix and not part of the root at all. I should have remembered about Old Irish verbs …)

  65. PlasticPaddy says

    imm-soí in Middle Irish. Judging from some of the MI reflexes, I think the Modern Irish reflex corresponding to Xeribs Welsh word is iompaigh-to move (in various ways). Wiktionary agrees.

  66. David Eddyshaw says

    Amau is “doubt, suspect”, which I suppose can be derived from “twist about” with enough imagination.

  67. David Eddyshaw says

    Tràcht/tráig is a doublet in Middle Irish; both can mean “beach” (Welsh traeth) or “part of the foot”

    That is both odd and very interesting: interesting in how the metaphorical uses of body-part words sometimes diverge surprisingly between languages.

    The bank of a river in Kusaal is its “mouth”, perhaps the less odd as there aren’t any actual rivermouths in Kusaasi territory, though I’m told that some southern Ghanaian languages use the same metaphor.

    The side of a lake in Kusaal is its “forehead.” Hills have backsides rather than feet.

    I met a mad German anthropologist in Ghana who collected such things. He would draw a bottle and ask people what its “neck” and “bottom” were called.

  68. I once learned that the rectangles on either side of a newspaper masthead (“All the News that’s Fit” etc.) are called the ears.

  69. David Eddyshaw says

    Looking up the Irish Wikipedia page for the Tractatus* (by a natural connection of ideas) I see that Old Irish felsub “philosopher” is now fealsamh. This** seems Very Wrong to me, and has probably got something to do with the Young People of Today. Really, nothing has been as it should be since 1453, if you ask me …

    * sef Traethawd, gan yr athronydd Llewelyn ap Siarlys Gwytgynstain

    ** the -mh instead of -bh.

  70. @Y: In the business, the title atop the front page on a newspaper is the “flag.” The “masthead” lists the publisher and principal staff of the paper and is usually found on atop one of the other early pages; using “masthead” in place of “flag” is fairly common, but it would be considered a solecism in an American newsroom. I don’t think I’d ever heard of “ears,” although it is evidently a real term.

  71. “Ears” was used at the college newspaper where I went to school.

  72. John Cowan says

    the –mh instead of -bh.

    No worse than Caerdydd for Caerdyf > Cardiff, though.

  73. Wiki: masthead (BrE) = nameplate (AmE) = “the flag”; masthead (AmE) = imprint (BrE)

  74. “ears”


    1. ear
    2. (archaic, dialectal) edge, border
    3. handle


    koski ‘(river) rapids’, thus, Koskenkorva ‘(a village) near rapids’

    From a dictionary: Kalat kerääntyivät kosken korvalle. ‘Fish gathered at the rapids.’

  75. David Marjanović says

    imprint (BrE)

    Impressum in German, hidden at the bottom of some later page.

  76. PlasticPaddy says

    I was trying to see in eDIL how latin (really Greek!) ph was reflected in old borrowings. Intitial ph is f in Old Irish (philosopher, pharaoh, phoenix, etc.) except perhaps for
    (a) limb, member, organ (used of any portion of the body): …Ml. 44d8 . innam (inna, MS.) b.¤ tuisten gl. locorum genitalium, 58b2
    Naah it couldn’t be. Or maybe it was and the wicked Sassenach took the word from us and misapplied it.
    What I first suspected about felsub was that it might be a contraction of feallsamhnaib, but now I am not so sure. For prophet eDIL has fáith (compare Latin vates). There is even saebfháid = false prophet (I suppose saeb corresponds to German schief). The word I knew for prophecy is tairngreacht and that also goes back to Old Irish, the tairng- looks very like a word for nail (modern tairne) so tairngir looks like “nail man” but I do not know (and hope prophecy was not aided by driving a nail into, say the spine of, a sacrificial victim).
    camphor is compora
    aphorism is afforaise.
    antigraphum is angraib but graphum is graif or graibh. I would have slight suspicions of these as evidence for ph > b, because the inventor would know Latin scribere and the Irish borrowing with b. And there is also a form engraifib, so again angraib/engraib might be a contraction.

  77. David Marjanović says

    I suppose saeb corresponds to German schief

    Looks like it.

  78. David Eddyshaw says

    I think ball is cognate, rather than borrowed. It’s only non-initial Latin/Greek f/ph that can turn up as Old Irish b in loans. Fáith is a cognate too, rather than a loan. The corresponding Welsh word has come down in the world:


Speak Your Mind