Yuan Dynasty Imperial-Style Thirst-Water
|Full Recipe Name|
|Yuan Dynasty Imperial-Style Thirst-Water|
|The Compendium of Essential Arts for Family Living|
During Lord Þórfinnr Hróðgeirsson 's investigation of Yuan dynasty (1271–1368, led by Mongols) Chinese cuisine, he came across a selection of “thirst water” recipes in The Compendium of Essential Arts for Family Living. It also calls them “sherbets,” and the majority of them, like their relatives in Central Asian cookbooks, are sweet fruit syrups. This one, however, is a spiced short mead.
Sherbets likely arose in the Islamic world, and spread to China in the 13th century. A 12th century Egyptian cookbook contains quince and sugar syrups, while a 14th century Arabic cookbook contains a lemon-quince-rosewater sherbet. Other examples exist.
Old Stuff from the Martial Forest by Zhou Mi (1232-1298) names fifteen “thirst waters” flavored with fruits, herbs and spices. It seems that the Mongols had a particular affinity for them: the 18th century Corrections to the Compendium of Materia Medica tells us that 14th century scholar Wu Lai records that the Yuan dynasty planted an orchard of 800 probably-lemon trees for the purpose of making “sherbet” by cooking their juice with boiled-down honey. This, and the name of this recipe, implies to me that sherbets were drunk by the upper strata of society, although there is some evidence for a broader culture of sweetened beverage drinking in earlier Song China.
Essential Arts lists 7 sherbets, five of which are boiled down fruit juices with honey or rock sugar, and often rosewood, borneol and musk, then dissolved into water to drink. One recipe uses “pine sugar” and some herbs, and is drunk diluted into ice water. The first recipe, however, is fermented, and therefore it is fundamentally more interesting than the others.
For about 1 gallon. The original recipe is fifteen times this, fermenting 50 liters of water.
- 5.33g each of cassia cinnamon, clove, osmanthus flowers, cardamom (shelled) and Amomum villosum (shelled)
- 10.67g each of fine yeast cakes (see supplemental yeast cake documentation) and malted barley (I used six row)
- 21g hops (I used pellet Saaz due to my LHBS not having any leaf noble hops, and figuring that leaf Cascade would be farther from period than pelletized Saaz)
- 422g wildflower honey
- Brush yeast cakes clean, and break off an appropriately sized chunk, about 1” square.
- Bring 2.4L of water to a boil in a pot. Add the hops, and boil until it is reduced to 1.67L, about an hour. Let cool until bearable.
- While boiling, grind spices, yeast cakes, and malted barley to a fine powder in a mortar.
- Strain hop liquor through a fine cloth. Discard hop residue.
- Add another 1.67L water to the hop liquor, and the honey. Mix well.
- Place ground ingredients in a heavy silk bag. Pour wort into a 1 gallon wide-mouth fermenter, add the bag, and let ferment for five days at room temperature.
- Remove bag and serve immediately.
In my test batch, this yielded a sweet beverage with a distinct bitterness and bitter aftertaste at about 2.7% ABV, having gone from 12 to 9 apparent °B. The batch presented started at 11.2 °B and used a different honey. The spice blend is very similar to chai teas from India, which is unsurprising because they have the same primary ingredients: cinnamon, clove and cardamom. The sweetness matches well the expectation that this is a refreshing beverage to drink on a hot day.
Being drunk young, it’s quite cloudy. The bag serves to reduce that somewhat, but not completely. Being transported to Brew U didn’t help either.
Deviations from Period Practice
This beverage was prepared using modern equipment, but that should have little influence on the final flavor. Period cookware would have been iron or pottery vessels over a stove, and so would have contributed little flavor and had a similar cooking experience. A new silk bag was not used because StarSan was likely good enough sanitization. While the use of modern sanitizers does exclude some wild microbes, the yeast cakes and spices have enough to make up for it. The modern fermenter reduces oxidation as compared to an urn, but a five day fermentation does not offer much time for oxidation to happen.
The spices and are probably more potent than period ingredients, as they have not suffered from rough transport and storage.
“Malt” in the recipe could be wheat or barley. Six row was used because it’s somewhat more common for European malts, but the author does not have a good understanding of barley varieties in China. Because the quantity is so small, and no mashing happens, the impact should be small.
The honey was raw, centrifuged and filtered, rather than a boiled and skimmed refined honey as directed. This probably introduced some more complex honey flavors, but they should be swamped by the clove and cardamom.
The big issue is the hops. As discussed in the footnotes to the close translation, wisteria is a more conservative reading of “vine flower,” but hops were used because they were easy to access, because they’re plausible, and because the “boil until reduced by one third” reminds me strongly of medieval European recipes, and will isomerize the hops well. The bitterness adds to the refreshing quality of the beverage as well.
Hops were used in beverage production in Medieval China. They appear in four yeast cake recipes in Song Dynasty (960-1279) brewing manual The Wine Canon of North Hill, with an undated annotation in the text saying that a synonym for hops is “spicy mother vine,” matching the “vine” in our text. However, the author did no not come across boiled hops.
The author does not have a rich understanding of what hop varieties exist in China, but there are several wild hop species in East Asia including Humulus lupulus and Humulus japonicus. Noble hops are as similar a commercial hop as reasonable, but the wild hops probably differ significantly. The pellet hops used are somewhat more potent than leaf hops, but not by much.
This beverage is presented you in a porcelain stem cup based off an 8th century example, poured from a bottle which mimics Song dynasty celadon ware, on a lacquer tray, mimicking high-status late Song presentation: “wine [was] served in cups of fine porcelain placed on lacquer trays.”
Thirst Waters are called in foreign lands sherbets.
Imperial Style Thirst-Water
Cinnamon [Cinnamomum cassia], clove [Syzygium aromaticum], Osmanthus flowers [Osmanthus fragrans]
Fine yeast cakes, wheat/barley malt [possibly barm], each four liang.
Grind fine with a stone. Take half a jin of vine [Dalbergia parviflora? Wisteria? Hops?] flowers, ten jin of well-refined honey. Sixty jin of newly-drawn water. Take the vine flowers, and after a moment in the wok, simmer it to 40 jin. Strain through a new dense plain-weave silk cloth into one small-mouthed jar. Fill a new dense plain-weave silk cloth bag with the above seven flavor powders. Then add it to the jar. Then, add 40 jin of new water. Then add the already-refined honey. Seal the jar’s mouth. In summer, five days. In fall or spring, seven days, in winter, ten days. If you have leftovers, warm it in spring or autumn, cool it in summer, and heat it in winter.
- Kanz al-Fawa'id fi tanwi al-mawa'id, in Medieval cuisine of the Islamic World by Lilia Zaouali p. 133-134.
- The Description of Familiar Foods (Kitab wasf al-at’ima al-mu’tada), in Medieval Arab Cookery translated by Charles Perry p. 442-443.
- 吳萊 (1297-1340)
- 舍里別 Middle Chinese “syae li pjet,” while also giving “thirst-water” as the Chinese name.
- 攝里白 Middle Chinese “syep li baek,” primarily calling them “thirst-waters.”
- Gernet, Jacques, Daily Life in China, on the Eve of the Mongol Invasion, 1250-1276. Translated by H. M. Wright. New York: The Macmillan Company, 1962. Page 49.
- These plant identifications are primarily based on Paul Kroll’s A Student’s Dictionary of Classical and Medieval Chinese, 2015.
- Amomum villosum is a bit more obscure. It’s a cardamom relative. This identification is based on cross-referencing with a number of modern pharmacological sources where the name is still used, and with David Bensky and Andrew Gamble’s Chinese Herbal Medicine Materia Medica, 1989. It was hard to find, but the vendor I eventually bought it from (tsemporium.com) labeled it as 砂仁, the same name.
- Because this is an all-season recipe, I reason that the spices and flowers must be dried.
- If this were “purple vine flower” or “white vine flower” it would be unambiguously wisteria. Without that it’s not clear, although wisteria is the most conservative reading. 藤 itself means “climbing or trailing vines, including various species of wisteria or liana” and is very vague.