In an early LH post I trotted out my favorite example of coincidence in language:

…even though English and Persian are related languages, and even though Persian “bad” means the same thing as English “bad” and is pronounced almost identically, there is no historical connection whatever.

Imagine my joy at discovering an entire webpage of such coincidences, from Arabic/Mongolian akh ‘brother’ to Japanese yabanjin ‘person from the wilderness’/Turkish yabanci ‘person from the wilderness.’ For lagniappe there’s a table of “Yin/Yang reversals,” eg Catalan alt ‘high’/Turkish alt ‘low.’ I’m certainly not vouching for the accuracy of all the forms, but it’s a great idea, and I applaud Yahyá M for putting so much work into it! (If anybody notices any howlers, I’d appreciate hearing about them.)


  1. What I find strange about this site is that in addition to the list of coincidences it contains a credulous section on Proto-World.

  2. Good heavens. I didn’t investigate that far. That is extremely strange. The human mind is a bizarre thing.

  3. I wonder … who is to say that something is a “coincidence”? Take for instance,
    “Hungarian béka ‘frog’ — Sanskrit bheka ‘frog'”
    Now, I know that there are a lot of Romani speakers in Hungary, and Romani is linguistically related to Sanskrit…

  4. “Hungarian béka ‘frog’ — Sanskrit bheka ‘frog'”
    It could exist in Persian, and absorbed by the Magyars from Persians met during their travels. That happened with a few words.
    The biggest mystery for me about Indian words in Europe is how Indo-Aryan “maidan” entered Ukrainian.

  5. it contains a credulous section on Proto-World
    Not to mention a poem in Nostratic.

  6. Having looked around his site more, I’m afraid Yahya is a kook, and I suspect by “coincidences” he means “words whose similarity proves the Proto-World hypothesis.” Sigh. Well, it’s still a useful list for saner purposes.

  7. My favorites are “man” as in “Turcoman”, which means something like “man”. It’s not a hybrid Turko-English monster.
    “Horde” from the Mongol and the Germanic “hoard” are completely unrelated in every way, but in English I think that there’s pollution, both seeming to mean something like “a shapeless heap of objects or men”. The Mongol horde was very well-disciplined, so a different fake etymology from “order” is recommended.
    Likewise the Anglo-Saxon “gumman” means “man, human” as does the Mongol “kumun” (umlauted).
    Isn’t misleading scholarship so much more fun than valid scholarship?

  8. how Indo-Aryan “maidan” entered Ukrainian
    there’s nothing too irregular about it. Early Slavs were in contact with some Iranian tribe. Mostly religious lexicon was borrowed, so for example bog ‘god’. One group of Slavs has more borrowings than others, namely the Poles. This says certain things about the chronology of Slavic movements, i.e Poles probably stayed back east longer and had more contact with our Iranian tribe. Majdan is also in Polish. I cannot tell you at the moment how the borrowing went, because I just packed all my books. 🙁

  9. Vasmer says Russian got maidan from Turkic (Tatar &c), for what that’s worth.

  10. Speaking of maidans, are you familiar with Tychyna’s
    priceless, На майдані коло церкви?

  11. Nick Wiseman says:

    My favourite antonym is Potuguese puxe – pronounced “push” but means pull. Even Portuguese and Brazilians that speak fluent English have to think twice when you shout “push!”.

  12. One that I was surprised not to see in the near misses is English “ear” and Mandarin Chise “er3”.

  13. Since I mentionned Etiemble in an other post, I thought I’d quote one of his favorites: Chinese shamo 沙漠 (“desert”) and French chameau (“camel”). Another one, if I remember correctly is Spanish caracol (“snail”)/French caracoller (I don’t have any English dictionary here, but it is, of a horse, something like “to run very fast, in the leading position of the race”).
    In fact, I am afraid that many of the so-called similarities on that webpage are more graphic (which is completely arbitrary) than phonological. I mean, French “ça y est” and Urdu “sahih hai” (‘that’s right!’) may look the same when romanized that way, but is that similarity audible? Same for Fr. “toi” and Vietn. “tôi”, and many others (not to mention transliterated/transcribed items in “Ancient Greek” and “Old Chinese”). I know it’s only a matter of poetical leisure (for us, obviously not for the authors of the page), but even in that case, that weakens a lot the potential fun (but I do have a pro-speech vs. writing bias, except for Chinese, admittedly a very particular case).

  14. “French “ça y est” and Urdu “sahih hai” (‘that’s right!’) may look the same when romanized that way, but is that similarity audible?”
    Yes: sa i e v. sahih hE (epsilon), and I understand h is often replaced with breathy voice in India, making it even more similar.
    What I note is the converse problem: not all of them are in fact coincidences… Hebrew awir, for instance, really is an Indo-European loanword (from Koine Greek, I understand.) And of course, whether you accept Jakobsen’s idea or some sort of historical explanation, the mother and father words are unlikely to be just coincidences.

  15. Middle English: “devel” – devil (from Greek “diabolos”)
    Romani: “devel” – god (from Old Indo-Aryan “devata”.)
    now imagine trying to refute accusations of devil-worhiship given such a linguistic set-up.

  16. I believe the nominative form is del, not devel (o Del si yek, o Del si baro). But of course other forms are devla, devlesa.

  17. i got “devel” from “Romani: a linguistic introduction” by Yaron Matras. i figured, if you can’t trust Cambridge University Press, who can you trust in this world? 🙂

  18. i wonder if this could be just difference between dialects… i mean, i don’t think it’s vocative, because “dEvl-a” is supposed to be vocative…

  19. Yeah, must be a dialect thing. I checked my Greek Romani dictionary and my Russian Kalderash book, and they both gave del, but of course there’s a lot of dialect differentiation.

  20. The great cormorant (Phalacrocorax carbo) is called kawau in Japanese and Maori:
    The Japanese word 川鵜 is a compound, made up of kawa 川 (river, brook) and u 鵜 (cormorant), and pronounced ka-wa-u.
    There is also an umiu 海鵜 (sea u, aka Japanese cormorant):

  21. 西瓜 xīgua ‘watermelon’ & σικύα ‘bottle-gourd’. Herbert Giles says it isn’t a coincidence, but he quotes the smaller Liddell Scott, instead of the larger one, which is more specific.

  22. David Marjanović says:

    Now, I know that there are a lot of Romani speakers in Hungary, and Romani is linguistically related to Sanskrit…

    But why would the word for “frog” get borrowed?

Speak Your Mind