Polyglot Vegetarian has another superb post, this one on the linguistic history of Persian پنیر panir ‘cheese,’ which like many Persian words has spread throughout Western Asia (it will be familiar to many as the “paneer” of Indian restaurants). I wasn’t going to blog it, because I could easily wind up just serving as a PV reprint service, since pretty much everything there is worth telling people about and I assume that anyone who reads LH will have bookmarked it by now anyway. But then I got to the part where he mentions the Etymological Dictionary of the Persian Language being prepared at Yerevan State University and the Этимологический словарь иранских языков [Etymological dictionary of the Iranian languages] that’s so far published two fascicles (up through d) and links to the 1890 Grundriss der neupersischen Etymologie by Paul Horn on Google Books, and I couldn’t resist passing that along. And of course there are the usual side trips into things like Armenian proverbs and the Bhagavata-purana (भागवत पुराण) and the egregious (in every sense) Sir Richard Burton:

Reading The Lake Regions of Central Africa, Burton of course has much to say about the native cuisine. But in particular for this topic, he mentions (p. 52),

The mutunguja (the Puneeria coagulans of Dr. Stocks,) a solanaceous plant, called … by the Baloch panír, or cheese, from the effect of the juice in curdling milk, …

and again (p. 464-5),

Milk is held in high esteem … mtindi (curded milk), the laban of Arabia, and the Indian dahi. … [T]hey consider cheese a miracle, and use against it their stock denunciation, the danger of bewitching cattle. The fresh produce, moreover, has few charms as a poculent among barbarous and milk-drinking races … On the other hand, the curded milk is every where a favorite … [They] do not … make their dahi …, like the Arabs, with kid’s rennet, nor like the Baloch with the solanaceous plant called panir.

Much of this is the usual Victorian racism, though the notion of milk-drinking races survives a bit in the conclusion that a difference between cheese and tofu cultures is genetic lactose intolerance in Asia. The observation here is that there is a plant called panīr used as a vegetable rennet. Vegetable rennet is important to lacto-ovo vegetarians. It can be tricky to discover how the cheese in prepared cheese foods was made and the conservative assumption always has to be that “enzymes” means animal rennet. Commercial vegetarian rennet is made from molds, but there are plant alternatives.

Keep up the tasty work, MMcM!


  1. Egregious, eminent and outstanding. I assume that’s what you meant.

  2. Cool stuff. It seems the botanically correct name for the plant is now Withania coagulans.

  3. I learnt to make paneer from my mother who must have learnt etc etc
    Boil whole cream milk till roiling – at least a gallon else you don’t get enough solid matter to make it useful – you can now use a tablespoon or two of white vinegar or lime juice to “break” the milk, which immediately [boiling continues but not roiling, turn heat down a bit] begins to seperate into whey and solid fresh curds. Turn off heat. Drain through clean muslin cloth [mom has used dad’s handkerchiefs on occasion] and then tie the curd into the cloth [you can tie a knot at the ends] and place in sink, on slanted cutting board to drain with heavy flat weight [paneer will be flat, firm, tofu like texture ] or hang from tap for a more “grits” like texture you can scramble like eggs – yum. [infini]

  4. Of course, that’s practically identical to the manner in which tofu is made from fresh soymilk. I usually use a colander to contain the curds for the pressing part, though.

  5. Could I just say, I love the word “roiling”.

  6. that’s practically identical to the manner in which tofu is made from fresh soymilk.


    Tofu skin, beancurd skin, beancurd sheet, or beancurd robes is a food product made from soybeans. During the boiling of soy milk, in an open shallow pan, a film or skin composed primarily of a soy protein-lipid complex forms on the liquid surface.[2][3] The films are collected and dried into yellowish sheets known as tofu skin.[4][5] Since tofu skin is not produced using a coagulant, it is not technically proper tofu; however, it does have a similar texture and flavor to some tofu products.

    Tofu skin’s use was first documented in written records in China, Korea, and Japan in the sixteenth century. It is widely used, fresh, fermented, or dried, in Chinese, Korean, and Japanese cuisine.

  7. Lars Mathiesen says

    I wonder if you could collect enough of the “skin” that forms on boiling cow’s milk to make a meal of it.

    The savage and milk-drinking Norwegians are of course to blame for myseost which is basically whey boiled to surrender and partial caramelization. (A similar idea but much firmer than dulce de leche; you do need one of those cheese slicers to cut it, your cheese wire will snap if you try). It’s slightly cheating in this context since a coagulant has been used to remove most of the other milk solids before boiling, leaving only lactose.

  8. Kaymak, Sarshir, or Qashta/Ashta (Persian: سَرشیر saršir) (Arabic: قشطة Qeshta or Arabic: قيمر Geymar ) is a creamy dairy food similar to clotted cream, made from the milk of water buffalo, cows, sheep, or goats in Central Asia, some Balkan countries, some Caucasus countries, the countries of the Levant, Turkic regions, Iran and Iraq. In Poland, the name kajmak refers to a confection similar to dulce de leche instead.[1]

    The traditional method of making kaymak is to boil the raw milk slowly, then simmer it for two hours over a very low heat. After the heat source is shut off, the cream is skimmed and left to chill (and mildly ferment) for several hours or days. Kaymak has a high percentage of milk fat, typically about 60%. It has a thick, creamy consistency (not entirely compact, because of milk protein fibers) and a rich taste.[2]

  9. The easiest way to make dulce de leche is by putting an unopened can of sweetened condensed milk in a pan of boiling water for a couple of hours. Just make sure that there’s enough water to keep it covered or floating, and let it cool down before opening it.

  10. Trond Engen says

    Lars M.: The savage and milk-drinking Norwegians are of course to blame for myseost

    The closest Norwegian relative of dahi et al may be gomme. Boil milk. Add rennet (or nowadays kefir milk) to separate whey and curd. Keep simmering slowly for hours, stirring only rarely between your daily chores, until the free fluid is bolied out and it’s all porridgey. Let cool. Serve as a dessert in a bowl or as spread on lefse.

    Modern versions add sugar (early), raisins (late) and cinnamon or other spices (early or after serving).

  11. Lars Mathiesen says

    @Y, that is indeed the neatest magic trick. The kids loved it.

  12. Simmered and curdled cream (old fashioned clotted cream) is the starting point for Stilton cheese.

  13. There is another root etymology for Persian پنیر panīr, Middle Persian 〈pnylpanīr, available. Here is the root as presented by J. Cheung (2007) Etymological Dictionary of the Iranian Verb (the pa- in Persian panīr being the reflex of the preverb *pati-, “against, on(to), in(to), at, beside, in return for, etc.” according to Cheung):

    *naiH² ‘to churn (butter)’
    •KHOTANESE: LKh. ñye, ñe ‘buttermilk’, LKh. nīyaka- ‘fresh butter’
    •NWIR: Kurd. nivīšk, Bal. nēmaġ ‘butter’, Abyan. nīmešk ‘fresh butter’, Tr. nīmešk ‘churned butter’, Tal. niyə ‘churn’ || (+ *pati-) NP panīr ‘cheese’
    •NEIR: Rosh. nay-/nid, Sariq. nεy-/nůd, nïd ‘to churn’, Sh. nīm-δōrg ‘churnstaff’,
    Sangl. nī́δū̆k, Ishk. nuduk ‘buttermilk’, M. nī̆yo, Yi. nī̆ya ‘sour milk’ || (+ *pari-) Wa. pərnə́c ‘churn’
    •SANSKRIT: nīta- ‘fresh butter’ (ĀpŚS), navanīta- ‘fresh butter’ (Kh+), et al. EWAia II: 25 f.
    ◊ The verbal forms appear to be attested in a few Pamiri languages. This IIr. root may have an IE provenance, as it has a correspondence in Baltic. This can hardly be regarded as mere “convergence”, considering the very specific nature of the meaning.
    •PIE *neiH- ‘to make butter, churn’ LIV: – | Pok.: –
    •IE COGNATES: Latv. sviestu nīt ‘to make butter’, pa-nijas, pa-nĩnas ‘buttermilk’

    I have not traced this root etymology for panīr further back than H.W. Bailey (1979), Dictionary of Khotan Saka, p. 184, column b, under nīyaka-, here, but I have also not looked very hard.

    For the particulars of the relationship between buttermilk and cheese in the Iranian world and the realia behind the possible derivation of a word for cheese from a verb for churning, scroll down to the section Acid cheeses in the article “Cheese” in the Encyclopædia Iranica.

  14. The kaymak of Mardin in Upper Mesopotamia is quite distinctive. Thin layers of clotted cream are lifted off the heated milk successively and laid flat, one by one on top of each other on little plates, slowly building up layered pile no more than a centimeter deep. The texture of the product is light and flaky, sometimes almost dry. Mardini people usually put a pink-colored flavored sugar syrup called şerbet over it, but I prefer honey. It goes rancid very quickly, but that is never a problem—if someone in my house goes out to the kaymakçı in the morning and buys some for breakfast, it will all be gone by noon. There is a description in good description in this blogpost:

    In Mardin paper-thin, almost translucent sheets of kaymak made from goats milk resemble no dairy product I’ve ever seen. Arranged in layers on shallow plates like yufka or phyllo awaiting transformation into a borek or baklava, they drew us into a kebab shop where we ordered one to share (and ended up with two).

    The post also has a very representative picture.

  15. On making butter from yoghurt, which was mentioned as the first step in the process of making acid cheese in the Iranica article I linked to, there is this webpage with a nice embedded video from Instagram:

    I should add that perhaps the sense of the preverb *pati- in the precursor of Persian panīr was “back, again”, in that the yoghurt, having yielding butter and become buttermilk, is treated again to yield curds pressed into cheese.

  16. The post also has a very representative picture.

    Looks delicious! (And the etymology is intriguing.)


    They consistently spell ayran as aryan. Spellchecker? Or is it because it’s about Indian practices?

  18. David Marjanović says

    Yoghurt butter has been available in supermarkets over here for a decade or two. It isn’t good, but then I find yoghurt thoroughly disgusting – much more so than yoghurt butter, though.

  19. I only learned about kajmak in Serbia. I was somewhat surprised that we do not know it.

    We have cream and we have smetana… don’t Turkic-speaking populations of Russia have their own traditional products different enough from those to motivate borrowing of the word?

    Much of this is the usual Victorian racism, though the notion of milk-drinking races

    A person writing about 19th century texts in her own language is ready to dismiss them because of the word “race”. How the same person will interpret texts by barbarous and milk-drinking peoples?

    (Of course Victorian society was more racist than the modern society which is still racist and of course the tone of the quoted passage is snobbish. But as far as I can see, “race” here means “population”, it simply is not a part of the snobbery. By learning to use new words you don’t become any less snobbish).

  20. David Eddyshaw says

    I think that’s true; Burton was probably rather less racist (in modern terms) than the norm for his time and place, and he certainly doesn’t actually mean by “race” what a modern writer would mean by the word. The relevant pseudoscience had not yet reached its full development in his day, after all. (The modern Approved Expression for what he does mean would be “ethnic group.”)

    To this day, the arable-farming majority in West Africa certainly regard their dairy-farming neighbours are Not Quite the Thing, even if not as outright barbarians and milk-drinkers. (The Fulɓe fall on both sides of this division, but the Islamic scholars, Jihadists and empire-builders belong to the non-milk-drinking side of the family.)

  21. January First-of-May says

    milk-drinking races

    …indeed the Aryans, both on the historical NW India side and on the misunderstood North Germanic side. (A few others too, but AFAIK the North Germanics get the highest scores.)

    IIRC blonde hair and/or blue eyes have fairly high correlations as well…

  22. We have cream and we have smetana… don’t Turkic-speaking populations of Russia have their own traditional products different enough from those to motivate borrowing of the word?

    Kumis. (Russians, including Lev Tolstoy, used to take trips beyond the Volga to hang out with Turkic-speaking populations and drink kumis for a chunk of time to improve their well-being.)

  23. David Eddyshaw says

    barbarous and milk-drinking races

    It seems pretty clear to me that Burton was in fact deliberately winding up his readership in the passage in question. He was certainly well aware of all this himself (and taking the piss out of his contemporaries’ attitudes to race is exactly the sort of thing he’d delight in.)

  24. Turkic-speaking populations of Russia have their own traditional products


    Chal, or shubat (Kazakh: шұбат, pronounced [ɕʊˈbɑt]), is a Turkic (especially Turkmen, Uzbek and Kazakh) beverage of fermented camel milk, sparkling white with a sour flavor, popular in Central Asia — particularly in Kazakhstan, Uzbekistan and Turkmenistan.[1] In Kazakhstan the drink is known as shubat, and is a staple summer food.[2] Due to preparation requirements and perishable nature, chal has proved difficult to export.[3] Agaran (fermented cream) is collected from the surface of chal.[4]

  25. Shubat is very rich in fat and I feel full after drinking two cups.

  26. @LH, juha, thank you. I actually meant anything called “kaymak” specifically, but I did not know chal, shubat and agaran!

    I know kumis since ever (as everyone maybe) and I learned ayran (and accrodingly tan) later. My ignorance is maybe in part due to that I rarely drink/eat diary products other than cheese (and butter).

    @juha, is there a recipe distinct from generic сливки known to Turkic-speaking people of Russia (or maybe USSR) as “kaymak”? Wiktionary has каймак, but it tends to translateit as сливки/cream…

    P.S. if “agaran” is “fermented cream”, it can be related to kaymak story too somehow…

Speak Your Mind