I recently ran across the Wikipedia article Qurabiya:

Qurabiya (also ghraybe, ghorayeba, and numerous other spellings and pronunciations) is a shortbread-type biscuit, usually made with ground almonds. Versions are found in most countries of the former Ottoman Empire, with various different forms and recipes. […]

There is some debate about the origin of the words. Some give no other origin for the Turkish word kurabiye than Turkish, while others have given Arabic or Persian. Among others, linguist Sevan Nişanyan has given an Arabic origin, in his 2009 book of Turkish etymology, from ġurayb or ğarîb (exotic). However, as of 2019, Nişanyan’s online dictionary now gives the earliest known recorded use in Turkish as the late 17th century, with an origin from the Persian gulābiya, a cookie made with rose water, from gulāb, related to flowers. He notes that the Syrian Arabic words ġurābiye/ġuraybiye likely derive from the Turkish.

A typical Wikipedia etymological mishmosh; can anybody say what’s most plausible? (Xerîb?)

Also, courtesy of Trevor Joyce, a brief YouTube clip in which Werner Herzog regrets having been forced at gunpoint to speak French.


  1. I wonder if it’s related to several Russian/Slavic terms also dealing with some type of bread/pie/biscuit.

    Kolob, Kolobok, Coulibiac- none of them really have convincing etymology.

    Why not Persian?

  2. From the link I learned that “Baku kurabye”, an “endangered but extant species” of Moscow’s confectionary stores of my days, was originally “Persian kurabye” until the Stalinist expunging of all “foreign” adjectives from the food names. This sweet was one of several semi-extinct ethnic sweet biscuits, notably including Yiddish-insipred zimmelakh and kikhelakh.

  3. BruceBrown says:

    There’s an Indian sweet called gulab jamun. It’s great, like an intensely saffrony cheesecake in a crisp fried coating.

    Wikipedia says gulab is from Persian gol (flower) and āb (water), referring to the rose water-scented syrup, and traces the recipe to Turkic invaders. So 15th century?

    But then, maybe that’s not much help? There are lots of Middle Eastern sweets with rosewater in, and they might be completely unrelated.

  4. January First-of-May says:

    “Baku kurabye”, an “endangered but extant species” of Moscow’s confectionary stores of my days

    Moscow’s confectionary stores in my days (mid-2000s) still regularly offered курабье бакинское, and I’ve seen it sporadically later; I suspect that the lack of my recent encounters with it is just because I don’t visit confectionary stores much any more.

    That said, the recent versions I’ve had were nowhere near as good as the tasty version from a decade as a half ago. And I’m sure that back in your time they were even better.

    Kolob, Kolobok, Coulibiac- none of them really have convincing etymology.

    …that’s колобок and кулебяка, right? I can’t figure out what the first one is.

    I always thought that the колобок was somehow related to the “wheel” root, but in retrospect it does make sense that they could be related to each other, and кулебяка does sound like it could be Turkic. But I’d probably have to look up the accepted etymology first.

  5. We discussed kołacz and its etymology before, with the regular wheel-word explanation.

    I don’t visit confectionary stores much any more

    Right, the sad truth. I was just musing about it yesterday after an Afro-Argentine tango friend confided about mean people causing problems with her baklava business (she apprenticed with the Greek friends in her younger years, and her baklava is incredible). Some White Americans pestered her with “Make African food, not Greek food” bullsh*t, while a bunch of aggressive Turkish commenters on the Instagram accused her of misappropriating Turkish heritage. And I was like, wish I could support you as a customer too, but I’m not young enough to buy baklava by a tray 🙁

  6. That reminds me of the sad fate of the Cappadocia Restaurant which opened on Broadway in Astoria, NYC, in the summer of 1996. It served wonderful Turkish food but was run by Kurds who had had to flee Turkey; I got to know them and lent them a good history of the Kurds I happened to have. I decided to have dinner there in early December but found it closed with no obvious reason; the next day I called and got a “The number you have dialed has been temporarily disconnected” message. Eventually I asked at a nearby store and was told there had been a “bad fire”; they hinted, but did not outright say, that it had been burned out by local Greeks (it was then a largely Greek neighborhood) wanting to drive away the invading “Turks.” Just another reason to despise nationalism.

  7. PlasticPaddy says:

    In a case of a restaurant fire that is not accidental, there is also the possibility of (a) insurance fraud by owner or (b) failure of owner to pay protection (or help the protector launder money, hide stolen items, etc). While in case (b) in a “Greek” neighbourhood the protectors might be “Greek”, the arson would be done strictly for business reasons.

  8. Sure, all those are distinct possibilities. I loved Astoria, but it had a lot of unsavory stuff going on if you pulled up the rug a bit.

  9. John Cowan says:

    Eventually I asked at a nearby store and was told there had been a “bad fire”; they hinted, but did not outright say, that it had been burned out by local Greeks (it was then a largely Greek neighborhood) wanting to drive away the invading “Turks.”

    My enemy’s enemy is indistinguishable from my enemy. Because I’m an idiot.

    Pretty much what happened to Saladin, whom WP calls the most famous Kurd in the world. Except he managed to die in his bed.

  10. Charles Perry says:

    My money is on ghurayyibah, a diminutive of Arabic gharibah, “marvelous, extraordinary.” It is often flavored with rosewater, orange blossom water or a mixture of the two, but the salient thing about this shortbread is its delicate texture, not a rosewater flavoring (which it shares with countless other Middle Eastern sweets). The Turkish form would then be explained by metathesis.
    BTW, this pastry appears in Greek pastry shops under the barely recognizable name kourambiedes (though doubtless fairly recognizable to Language Hat readers).

  11. If the Arabic etymology is not right (and I suspect it has a strong claim), then we can at least say that in Arabic it gravitated towards an association with the Arabic root. Because a simple borrowing would not have ended up gh-. Not my linguistic bailiwick, but looks more likely to have been borrowed from Arabic into languages that don’t have gh than vice versa.

  12. Andrej Bjelaković says:

    FYLSC has gurabija, and I love them!

  13. In Algeria they’re ġʷṛaybiyya – one of my favourite sweets, though a bit dry. The Arabic etymology seems reasonable, but I’d want to check the medieval cookbooks before reaching any conclusions…

  14. Charles Perry says:

    Lameen — I find no mention of this pastry in the medieval cookbooks, which suggests that ghreybeh is probably a relatively recent development. No shame in that.
    BTW, in case anybody was wondering about the k in kurabye, the form gurabye was also used in Ottoman.

  15. “gulab” = “rose” traveled far – here it is in thai: https://en.m.wikipedia.org/wiki/Suankularb_Wittayalai_School

  16. A typical Wikipedia etymological mishmosh; can anybody say what’s most plausible? (Xerîb?)

    Here are some random notes to add to what has been said above. Two views of the development of Turkish kurâbiye were noted:

    1.) The family of Persian گلابی gulābī “confection made with rose water”, Ottoman Turkish كلابيه gülâbiye “wafer, cookie” is original and Turkish kurâbiye is an alteration of this original. Arabic ġurayyiba and its many variant forms are loanwords from Ottoman Turkish and have undergone some sort of folk-etymology alteration. (Here is a link to the word gülâbiye in the 1780 revision of Meninski’s important Ottoman lexicon (1680). The 1780 revision edition reads كل بيه , which must be a simple printer’s error for كلابيه . The original 1680 version, which I have in front of me but couldn’t readily link to online, reads ⟨ كلابيه g’ülābije⟩ correctly with Meninski’s own transcription. Here is gurabiye in gurabiyeciyan (perhaps written ⟨غرابیەجیان⟩ in the original text?), plural of gurabiyeci “kurabiye-maker” in a modernized edition of Evliya Çelebi’s Seyâhatnâme “Book of Travels” from the mid-to-late 17th century.)

    2.) Arabic ġurayyiba, ġurayba, etc., are original and Turkish kurâbiye is an alteration of these. Arabic ġurayyiba, means “marvellous little one” and explains itself as a designation for a cookie.

    It’s harder to motivate some aspects of the first view—why would an original Persian گلابی gulābī or Turkish كلابيه gülâbiye “confection flavored with rose water” end up with a ġ- and -r- in Arabic? Particularly when we consider that Persian گلاب‎ gulāb “rose water” came across fine as Arabic جلاب jullāb, julāb (cf. English julep). It’s not like Arabic is completely intolerant of loanwords that have no transparent root derivation—on the contrary, the language is full of such words from Persian and Turkish, particularly in the culinary sphere. Why should gülâbiye have cried out for accommodation and been folk-etymologized to the root ġ-r-b? In addition, we can note that Modern Repubican Turkish has ku- with a back vowel in the modern form kurabiye anyway (and Azeri has back qurabiyə too…). This back vowel would have to be explained as colloquial vowel-harmonic alteration in this scenario. This is not out of the question, but it is an extra step—in the other scenario, the back vowel just comes for free. I’d be interested to know Nişanyan’s reasons for changing the etymology in his dictionary. (Building on this… Could the كلابيه gülâbiye “wafer, cookie” in Meninski be a learnèd alteration of an earlier ġurābiye, kurābiye, after gülāb “rose water”, so commonly used as flavoring for sweets in the Near East?)

    I think the second view is reasonable—Turkish kurabiye could be an irregular alteration (influenced by the many Turkish nouns in -iye < Arabic -iyya?) of Arabic ġurayyiba “shortbread cookie”, found in variant forms around the Arab world. In form, Arabic ġurayyiba is the feminine of ġurayyib, the regularly-formed diminutive of ġarīb “foreign, strange, unusual, rare, exotic, marvelous, extraordinary”. (Perhaps in Ottoman times, there could also have been mutual influence and reborrowing between Ottoman and Arabic, and reaccommodation to Arabic morphology, contributing to the variety of forms in Arabic.)

    For parallels to the formation in Turkish -iye and similarly phonologically irregular colloquial adaptations of foreign words into Turkish, we can point to şehriye “noodle” (from Arabic šaʿr “hair”) and pişmaniye “Turkish-style cotton candy” (from Persian pašmīna “woollen; a kind of confection (cotton candy?)”, from pašm, “wool”).

    For a phonological parallel to the otherwise unusual change of an Arabic ġ- into Turkish k-, we can point to Turkish kalabalık. This is from Arabic غلبة ġalaba “superior power, prevalence, predominance, ascendancy” (root ġ-l-b, “conquer, defeat, gain ascendancy”). Arabic ġalaba became colloquial Turkish kalaba “crowd, throng”, and with the addition of the Turkish noun and adjective suffix -lık, the common, everyday Turkish word kalabalık “crowd, throng” (< “a predominance by number or amount”) as a noun and “crowded, congested (as a street)”, as an adjective. (The usual thing is for ġ- to show up as Turkish g-, as is galiba “probably” from Arabic غالباً ġāliban “preponderantly, mostly, most often” from the same root.) Note also Azeri qərib “stranger” with q from Arabic ġarīb (cf. qurabiyə above).

    For a semantic parallel to such a name ġurayyiba (“little exotic one, extraordinary little one, marvellous little one”) in confectionery, perhaps consider French nonpareille (English nonpareil) “pinhead-sized candies used in quantity as decoration; a chocolate drop covered with such candies”, but literally meaning “peerless, unequalled, without parallel”…

    And for the use of ġarīb in Arabic in the description of food, and specifically for a cookie, it is interesting to consult Nawal Nasrallah’s 2007 translation, Annals of the Caliphs’ Kitchens: Ibn Sayyar al-Warraq’s Tenth-Century Baghdadi Cookbook (mentioned in an article linked to earlier this month on Language Hat):

    Proficiency in the art of cooking was therefore a common pursuit in which the public participated alongside professional chefs. It was one of the desirable accomplishments of the ‘Abbasid man,’ especially the aspiring boon companion, who wished to win the favors of his superiors. A cookbook in the mix would definitely be a bonus in his credentials. This also gave rise to a genre of books that dealt with the etiquette adab of dining and wining with one’s superiors such as Adab al-Nadīm by Kushājim. The perfect nadīm ‘boon companion,’ for instance, was expected to perfect at least 10 exotic dishes… [In] a cuisine in whose language the epithet gharīb (exotic, unusual) was a high complement and whose society valued boon companions nudamāʾ by the number of exotic dishes gharāʾib al-ṭabīkh they mastered, incorporating some foreign unfamiliar elements into its already rich repertoire was a good thing. This was made possible by active trade and Baghdad’s cosmopolitan nature during the Abbasid period.

    One of the items called ġarīb in al-Warraq is found in the following:

    صفة خشكنانج غريب واثقي لأبي سمين

    ṣifatu ḫuškanānajin ġarībin wāṯiqiyyin liʾabī samīnin

    “Description of an unusual ḫuškanānaj of al-Wāthiq belonging to Abū Samīn” (= “Recipe by Abū Samīn for an unusual ḫuškanānaj made for al-Wāthiq.”)

    Abū Samīn (literally, “Father of the Fat Man”?) was a professional cook in the palace kitchens of the Abbasid caliph al-Wāthiq (reigned 842–847) according to Nasrallah. The word ḫuškanānaj “shortbread, cookie” in this recipe is from an Iranian source equivalent to New Persian خشکنانه ḫušknāna “biscuit, cookie” (from خشک ḫušk “dry” and نان nān “bread”). Nasrallah translates the recipe for this ḫuškanānaj ġarīb as follows (p. 418):

    Grind 3 raṭls (3 pounds) refined sugar and sift it in a fine-mesh sieve (munkhul ṣafīq). Add 1½ raṭls (1½ pounds) fine samīdh flour (high in starch and bran free). Mix them well. Add ¼ raṭl (½ cup) sesame oil and knead mixture the way you usually do with flour dough.
    Put the mixture in a mortar (hāwan) and pound it to crush ingredients into each other and help them bind.
    Take a small bowl (uskurruja), the smallest you have, or anything similar in shape such as a wooden or brass ḥuqqa (bowl) with a rounded base and a wide rim. Stuff the bowl tightly with some of the sugar-flour mixture and turn it over onto a khiwān (wide low table). Do this with the rest of the mixture.
    Prepare a large level pan with low sides and arrange the molded pieces, leaving a space between them.
    Lower the pan into a slow-burning tannūr. Let cookies bake until they are golden brown. Take the pan out and take the cookies out of the pan with a thin spatula (isṭām raqīq). You carefully slide the spatula underneath each cookie and transfer it to a clean platter. Arrange the pieces in one layer (yuṣaff), God willing.

    This is patently a recipe for shortbread and what would nowadays be considered a kind of kurabiye. Can we see ancestor of kurabiye in the ġarīb of this ḫuškanānaj ġarīb? Nasrallah seems to suggest as much:

    The cookies in this recipe are reminiscent of the ones known nowadays as ghureyyiba in most of the Arab countries, and shakarlama in Iraq. Al-Warrāq calls it gharīb.

    Or is this just coincidence? In any case, Nasrallah’s translation of this cookbook is really fun to read—I recommend it to anyone with an interest in the history of human institutions or a kink for experimenting in the kitchen.

    From the Arabic root(s) ġ-r-b come an amazing number of derivatives with meanings as varied as “depart”, “west”, “the tree Populus euphratica”, “drinking cup”, “the two edges of a camel’s haunches”, “stranger”, both “marvellous” and “miserable”, both “crow” and “silver”, and words referring to the notions of both “black” and “white”… Here is the entry for the root in Lane’s dictionary.

    The word ġarīb, “strange, foreign” and “a stranger, a foreigner”, from this root has many resonances, shading into “rare, exotic, marvelous” on the one hand, but particularly “forlorn, miserable, poor, wandering, longing for someone absent, all alone in the world” in Kurdish xerîb and Turkish garip—a word we’ve already encountered in Language Hat in the post The Garip Manifesto a few years ago.

    For lagniappe: I particularly like this wonderful performance of a Turkish classic about the fate of the garip, the stranger, the foreigner, the wanderer, the forlorn.

  17. Wow, thanks for that thorough response!

  18. I remember that Father’s Lebanese graduate student called it something like buqrawiya, some 40 years ago.

  19. And my maternal grandmother’s cousin was a great expert at baking them. Her husband was sickly, and so they had to move to the Crimea, where she learned the craft. In 1944, they were deported along with the Crimean Tatars (a Tatar is a Tatar, no matter Crimean or not) to Uzbekistan.

  20. Always buqrawiya tomorrow, never today!

  21. And it was the cousin who lived in the otrub, and her eldest daughter stayed with Grandmother’s when she started school, otherwise it would be too far to attend it.

  22. Juha’s comments reminded me of a point that I didn’t add to my previous message:

    If in fact Ottoman گلابيه gülâbiye “rose-water shortbread” is the original etymon of the forms in the Neo-Arabic languages and a quite late creation, then a problem arises when the word is borrowed into the Neo-Arabic languages—what to do with the Turkish گ /g/ sound in Arabic? This might explain the introduction of غ /ɣ/ as the initial—and then subsequent folk etymologizing and perhaps alternation of rand l (not unheard of in the Upper Mesopotamia linguistic area and cross-linguistically common) might explain the rest.

    As an aside relating to food words and the problem of adopting foreign /g/ into Arabic… Egyptians can handle /g/ fine (written with ج jīm, pronounced /g/ in that variety), but the other varieties of Arabic need different strategies. And this is featured in a recent Burger King ad campaign:


    The gist of this can be understood I think even if you don’t understand Arabic…The guys drive up to Burger King and begin arguing about the correct way to pronounce برجر “burger” in Arabic. The guy with the moustache wants to pronounce it as /burd͡ʒar/ as it is written with the letter ج jīm in the sign برجر كنج Burger King, which has two jīm‘s in it. The guy behind the wheel insists on an Anglicising pronunciation /burɣar/ with ghayn غ /ɣ/… They argue and are transformed into professors of Arabic philology, until they ask the drive-up window employee to settle the matter. But the employee says “It’s called a Whopper. Whopper!”—with a /p/ that poses even more problems for Arabic orthography and pronunciation.

    The ad campaign continues with the proposal for the introduction of the letter ڃ with two dots in place of the ج for the name:


    (The letter ڃ is in use in Sindhi and other languages, as explained here in French and here in German.)

  23. In Bulgarian it just means cookie :shrug:

  24. I’m reading Kaverin’s Перед зеркалом, and I just got to this, in a description of Yalta circa 1923:

    Высокий человек лет пятидесяти, с усами и черной бородой, в толстовке с галстуком, быстро ходит по городу.

    Через плечо — голубой ящик на ремне, на ящике надпись: «Горячие пирожки Тодорского». Время от времени он останавливается и громко кричит: «Пира-а-жки! Кура-абье!» Это — в прошлом широко известный юрист Тодорский. Курабье я не пробовал, пирожки у него покупал.

    (A former lawyer is now selling kurab’ye and pirozhki.)

  25. PlasticPaddy says:

Speak Your Mind