A medieval Chinese cousin of Eastern European cherry pierogi?

Language Log 2024-08-14

As a starting point for pierogi, here's a basic definition:

Pierogi, one or more dumplings of Polish origin, made of unleavened dough filled with meat, vegetables, or fruit and boiled or fried or both. In Polish pierogi is the plural form of pieróg (“dumpling”), but in English the word pierogi is usually treated as either singular or plural.

(Britannica)

Now, turning to Asia, we are familiar with the Tang period scholar, poet, and official, Duàn Chéngshì 段成式 (d. 863), as the compiler of Yǒuyáng zázǔ 酉陽雜俎 (Miscellaneous Morsels from Youyang), a bountiful miscellany of tales and legends from China and abroad.  Yǒuyáng zázǔ is especially famous for including the first published version of the Cinderella story in the world, but it also contains many other stories and themes derived from foreign sources.

Knowing of my interest in such matters, Zihan Guo called my attention to a terse mention of a culinary item called "yīngtáo bìluó 櫻桃饆饠" ("cherry [?something?]") in Yǒuyáng zázǔ.  One look and I was hooked.  "Yīngtáo 櫻桃" doesn't present any major problem of its own; that's just the usual word for "cherry" in Sinitic, though, if I had all the time in the world, I would do an etymological and botanical study on its origins.  As for bìluó 饆饠, I could tell from the extreme rarity of the characters and the word, plus the fact that it is disyllabic, that it almost certainly had a foreign origin.  That suspicion is reinforced by the additional fact that it has a variant orthography, viz., bìluó 畢羅.  The Middle Sinitic reconstruction of both written forms is pjit la.

I'm not the only person who suspected that bìluó 饆饠 / 畢羅 had a foreign derivation.  Since it bore a superficial resemblance to "pilau", i.e., "pilaf", many scholars jumped at this equation, to the extent it has become more or less a commonly accepted etymology (as in Wiktionary and zdic) that the Sinitic word comes from Persian پلاو (pelâv).  But there are three (actually more) strikes against such an assumption.  First, and most obvious, the Middle Sinitic reconstruction doesn't work as well as the Modern Standard Mandarin.  Second, neither of the variant orthographies of bìluó 饆饠 / 畢羅 use the "rice" radical mǐ 米 (Kangxi 119).  Instead, the first variant just uses the general "eat" radical shí 食 (Kangxi 184) on both characters.  Third, although cherry pilaf is possible as a dish, cherries are not one of the usual ingredients for pilaf, and it's not likely in any case that it would be so important an ingredient that a pilaf dish would be named after it.  Fourth, other, more detailed, Tang recipes for bìluó 饆饠 / 畢羅 indicate (so Zihan tells me) that it is made of baked wheat dough and has meat filling.

With "pjit la" in the back of our mind (more about that later), we have to look elsewhere.

The first thing I thought of was "pi[e]rogi".  It is made of wheat and has a filling (can be either meat or fruit).  Pierogies are usually boiled, less often pan fried, but they can also be baked.  Moreover, the sound of pjit la is vaguely similar to that of pierogi.  Now we have to dig deep into the history and nature of pierogi.

As a matter of fact, when we were examining the background of that delicious Nebraska nosh, runza, we looked into pierogi a bit.  

A runza (also called a bierock, krautburger, or kraut pirok) is a yeast dough bread pocket with a filling consisting of beef, cabbage or sauerkraut, onions, and seasonings.  Runzas can be baked into various shapes such as a half-moon, a rectangle, a round (bun), a square, or a triangle.

The runza sandwich originated from the pirog, an Eastern European baked good or more specifically from its small version, known as pirozhok (literally "little pirog"). In the 18th century, Volga Germans (ethnic Germans who settled in the Volga River valley in the Russian Empire at the invitation of Catherine the Great because of their skill in farming), adapted the pirog /pirozhok to create the bierock, a yeast pastry sandwich with similar savory ingredients.

(source)

This is encouraging.

Having come this far, I have not been disabused of the notion that bìluó 饆饠 might somehow be related to pierog.

A few more preliminary remarks before plunging headfirst into the philology of this conundrum.  Coming to the crux of the matter, we have to decide whether the word "pierogi" has a Slavic or a Turkic basis, which has been endlessly debated.  Here I must say that I come down on the side of the Slavicists and will explain why in a moment.

Borrowed from Polish pierogi, the plural of pieróg (dumpling), which ultimately is derived from Proto-Slavic *pirъ (party” [VHM:  also "banquet"]). Unrelated to Turkish börek. Doublet of pirogi (from Russian), pirohy (from Czech and Slovak), and pyrohy (from Ukrainian).

(Wiktionary)

To say that "pierogi" is "unrelated to Turkish börek" may sound somewhat abrupt, because börek may in some way have been influenced or cross-fertilized by Slavic pierogi.  Still, for all of the many reasons I am about to list, I must declare that I am in agreement with the Wiktionary editors.

I have discussed this matter at great length with my learned Turkologist colleagues, but, in the end and after devoting much time and thought to the problem, I have decided that the case for a Turkish derivation of "pierogi" is not nearly so compelling as one that it is from Slavic.  First, and above all, we can trace the Slavic word back to Proto-Slavic..  Second, if we attempt to connect the Slavic word to Turkic börek, we run into all sorts of difficulties, phonologically and semantically.  Third, chronologically  it is hard to demonstrate that Turkic börek was present in Eastern Europe by the time pierogi were popular there.  Borek derives from Ottoman Turkish (14th-20th centuries AD).  Fourth, the etymology of Turkic börek is confused and contested, with some authorities tracing it back to an old Turkic word for kidney, from the supposed shape of the pastry to the organ.  Fifth, pierogi are dumplings, usually boiled in water (see below), whereas börek are flaky, crusty pies made of filo dough that are savory and baked or fried in a pan.  Sixth, börek are savory and usually have vegetable ingredients such as spinach and potatoes or meat and cheese. I am unaware of fruits being used as filling for börek.

There are numerous different, conflicting theories about the origin of the word börek.  Many of them are given in the "Origin and names" section of the Wikipedia article on börek, from which I offer just the first:         

According to lexicographer Sevan Nişanyan, the Turkish word börek is ultimately originated from Turkic bögrek, from böğür (meaning 'kidney'). Nişanyan noted that the word is also used in Siberian Turkic languages such as Saqa as börüök.

In contrast, Marcel Erdal holds:

The pastry name börek is first documented in two Mamlûk sources from Egypt, both  early 14th century. There is not the slightest reason to think that it ever had the shape of a kidney. The kidney is called böbrek in Turkish and it seems quite certain that the phonetic similarity is a coincidence. 
The 'kidney' word (originally bögür ~ bögüräk, related to Mongolic böere 'loin') is discussed in Doerfer's TMEN, 2nd vol., pp. 353-354.
 
(personal communication)

All things considered, Occam's razor impels me to choose the Slavic origin of "pierogi" over a Turkic one.  Aside from the Slavic semantics being simpler and neater than the Turkic as the source of "pierogi" meaning dumpling with fruit or meat (plus vegetable) filling, the phonetics of Slavic words like Czech and Slovak pirohy and Ukrainian pyrohy match better than Turkic börek as the source of Polish (> English) "pierogi".

A few miscellaneous, yet relevant, matters to clear off the table.

Carol Kennedy tells me:

My grandmother (who was a Jew from Odessa, who spoke Russian and Yiddish) always called them “vareniki", which just means “boiled”. She never called them “pierogi”.

The same is true of other, diverse parts of Eastern Europe, where boiled dumplings are also called "vareniki".  Thus, cherry pierogi = cherry vareniki.

Eastern Europe, where pierogi / vareniki originate, was not too far from Tang China either, and even closer to medieval Eastern Central Asia (ECA).

Since we have textual evidence of what conceivably may be considered the equivalent of cherry pierogi in medieval Sinitic, namely yīngtáo bìluó 櫻桃饆饠 (Middle Sinitic [ca. 600 AD] 'eang daw pjit la), where the second pair of syllables is evidently a transcription of a borrowing, I'm led to consider the possibility that the original language may have been related to an early version of the Slavic word "pierogi", though not necessarily Slavic itself. 

Don't take the final -t of pjit too literally because it is there only to indicate an entering or checked "tone".  It's not really a tone in the phonetic sense, but rather a type of syllable that ends in a stop consonant or a glottal stop.  There are two other so-called "entering tones", -p and -k, hence the holy trinity of so-called "entering tones" (rùshēng 入聲 [a calque]) "-p, -t, -k".

What I discuss next are archeologically discovered pastries made of wheat dough from medieval ECA.  They are not dumplings per se, but evidently had an open filling of some sort (quite likely fruit jam) in the center of a dough pastry that was made to look like a cherry / plum blossom.

For photographs and descriptions of such archeologically recovered pastries, see:

Victor Mair, ed., Secrets of the Silk Road:  An Exhibition of Discoveries from the Xinjiang Uyghur Autonomous Region, China (Santa Ana, CA:  Bowers Museum, 2010).

p. 123, 23-1 "Plum Blossom Shaped Dessert", 7th-9th Century, Excavated from Astana, Turfan; Xinjiang Uyghur Autonomous Region Museum Collection

p. 125, 23-4 "7-peteled Flower Dessert", 7th-9th Century, Excavated from Astana, Turfan; Xinjiang Uyghur Autonomous Region Museum Collection

Since we have archeological evidence of pastries made of dough in the shape of cherry / plum blossoms and with what may have been fruit filling in the center recovered from the medieval Astana cemetery (42.882°N 89.529°E) near Gaochang 高昌 (Khocho, Karakhoja, Qara-hoja, Kara-Khoja, Karahoja, Chotscho, Khocho, Qocho or Qočo), we are within the ballpark of those "yīngtáo bìluó 櫻桃饆饠" ("cherry [?something?]") documented in the celebrated 9th c. Yǒuyáng zázǔ 酉陽雜俎 (Miscellaneous Morsels from Youyang); 34°15′40″N 108°56′32″E  incidentally, I know personally from having visited Uyghur families in their homes that such pastries are still popular in that region today, so I wonder what they are called in recent and contemporary Uyghur and other ECA languages.

 

Selected readings

  • "Respect the local pronunciation: runza and Henri" (6/13/24)
  • "Pork floss Beckham" (8/10/21) — Chinese nosh
  • "Beijing Noshery" (10/23/15)
  • Reed, Carrie E. (2003). A Tang Miscellany: An Introduction to Youyang zazu. New York: Peter Lang. ISBN 978-0820467474  Under the name Carrie Wiebe, the author of this book has written many other scholarly studies and translations based on Yǒuyáng zázǔ.
  • Beauchamp, Fay (2010). "Asian Origins of Cinderella: The Zhuang Storyteller of Guangxi" (PDF). Oral Tradition. 25.2: 447–496.
  • Victor H. Mair, tr., “The  First  Recorded  Cinderella  Story,” in Hawai’i  Reader in Traditional Chinese Culture, ed. by  Victor H. Mair, Nancy  Steinhardt, and  Paul R. Goldin (Honolulu:  University of Hawai’i Press), pp. 362-67.

[Thanks to Mehmet Olmez, Juha Janhunen, Peter Golden, Peter Hoca, Marek Stachowski, Marcel Erdal, and Sattar Salam] ]