Sunday, December 4, 2016

Su Dongpo Is A Tease

Last time, I translated a mead recipe from Zhang Bangji, which he claimed was the one Su Dongpo used and wrote poems about.  Apparently unbeknownst to Zhang Bangji, but knownst to us, Su Dongpo wrote a huge book of literary spew where he talked about random topics, Dongpo's Forest of Footnotes.  One of those footnotes, the twelfth entry on scroll eight, is about mead.

This is a received text, a scan of an 18th century copy made during an attempt to catalog and reproduce every book ever written, which means that there's no punctuation since it hadn't become popular in China yet.  I'm going to lean a bit on the punctuation from the blog post that led me to this text, but the blog post was missing some phrases so I've taken a more rigorous crack at translating it than I did in my first blog post.

My method for mead: stir together in pure water:
For every dou of grain, 2.5 liang of steam-cake flour, and “cake seeds” 1.5 liang just like the normal way for taking the liquid from un-pressed wine.
Add another liang of steam-cake flour.
Ferment for three days.
Observe the flavor - is it appropriate?  If it’s harsh or hard, then take another dou of cooked grain and add it.
If it’s sweet and soft, then with every addition also add yeast cakes and [other] cakes, one half liang each.
Wait another three days and it will be done.
When fermenting, always be aware of how the brew will expand and contract.  The less water you add, the better.

This recipe has some problems.  There are questions about what "steam-cake flour" and "cake seeds" are, but I think those are just different kinds of yeast cakes, ground up.  But worse, where's the honey?

The text could be corrupted.  It's hard to know if that's the case, so I'm going to ignore that possibility.

One interpretation of the text as it stands is that this isn't a mead recipe but a "honey wine" recipe - wine that tastes like, or is somehow otherwise reminiscent of honey.  I don't think that's very convincing unless it's unrelated to the author's poem about mead, which is definitely about mead since it mentions bees several times.

A possibility raised by my colleague Song Zidie is that the “steamed cake flour” and “cake seeds” contain the honey.  “Cake seeds” may actually mean “pancakes.”  This strikes me as unlikely, but it’s definitely not outside the realm of Chinese brewing (see: mutton wine).  Scroll 7 contains a reference to “steam cakes” that says something like “Some years ago there was a flood in [place] [...] and when the waters receded, [the river] was silted up, and was not able to be deep.  It was called “steam cake silt,” and the court hated it.”  I don’t see that too much can be drawn from this text, though.  Other references to “steamed cakes” really do seem to mean the food, cakes that are steamed, not yeast cakes so I’m confused here.

Another interpretation is that I've got something wrong with that first line.  It's weird grammatically which might mean that I'm parsing it wrong.  Let's break it down.  When a character has multiple meanings in A Student's Dictionary, I'll number them, and omit meanings I think don't fit.
予 "1. I, me"
作 "3. devise, create, make."
蜜 "1. honey, sweet; possibly pre-Han loanword from the Tocharian B mit [this doesn’t help but is super cool].”
酒 “generic term for alcoholic beverages produced through fermentation.”
格 1. “Frame(work), structure, scaffold(ing).  2. Established custom or law, precedent, protocol.  3. Bring to (proper) pattern, systematize; frame, put together, coordinate.”
與 “2. Join with”
真 “1. b. Pure, perfect.”
水 “1. Water, liquid.”
亂 “1. b. trouble(d); confuse(d); blur(red); jumble(d), mix(ed) up.”

I think 予作蜜酒格 is the “topic” of the sentence: “The framework [I use] when I make mead.”  Which leaves 與真水亂.  真水 is pretty clearly “pure water” - I don’t see any other plausible readings.  與...亂 suggests “Join x and the following, and mix” but it’s funny diction.  There are other words for mix that I would expect in this context (混, 攪 for example).  So maybe it’s saying to join the following ingredients with pure water mixed with something?  And maybe that something is implied to be honey?  It’s a stretch.  I’m not sure.

I'm disappointed that this recipe isn't very useful as it stands, but we do have the other mead recipe from Zhang Bangji, so I think I'll start by trying that one out.

Creative Commons License

A Mead Recipe

Recently, I posted about my search for a Chinese mead recipe, and some potential leads.

Today, I cracked open the English-language textbook on Chinese food and beverage production, Science and Civilisation in China Volume 6 Biology and Biological Technology Part V: Fermentation and Food Science, and it contains a section on mead (page 246)!

It backs up my research so far, although it doesn't include the really early archaeological finds because they were excavated after it was published.  There are a few mentions of honey used for various things in the classical period, but it isn't until the Tang dynasty in about 650 that we start to see references to mead.

The Song dynasty author Zhang Bangji (張邦基) recorded in his book Random Notes from the Scholar's Cottage (《墨莊漫錄》) what he claims is Su Dongpo's recipe for the mead mentioned in the poem I translated last time.  Science and Civilisation has a translation, but I decided to make my own before I read theirs.  It's in the fourth scroll, section 68:

It being the case that Dongpo’s nature was to like to drink, when he drank he did not drink many different things [possibly just “did not drink much”].
In Huangzhou he experimented with using honey to brew, and also wrote The Song of Mead, but people rarely share his method [for brewing].
For every four jin of honey, purified fully, put it into boiling water and dissolve it, to make up one dou.
Take two liang of good wheat yeast cakes, and one liang of southern baijiu yeast cakes.
Grind until fine, put them in a newly-woven juan cloth [dense silk tabby] bag, and put everything in a single vessel.
Seal it tight.  In great heat cool it and in coolness warm it.  If the weather is cold, then heat it.
In one or two days it will bubble, and after several more days the bubbling will stop.  When the wine is clear you may drink it.
At first, it will keep the taste of honey, but after half a month of clarifying, it will be fine strong wine.
Again adding half a liang of purified honey, cold, just when it begins to bubble is marvelous.
I tried this myself, and the flavor was sweet like rich-tasting shortwine, but experienced drinkers may not enjoy it as much.

My translation agrees pretty closely with the Science and Civilisation translation.

The recipe mixes weights and volumes, which means we need to be careful with the amounts we assign to them since it's not strictly proportional.  Fortunately, Wang Li's dictionary has historical information on quantities.  We want to know how big a jin, a liang, and a dou are, around 1100 CE, during the Northern Song dynasty.  Su Dongpo wrote a little earlier than Zhang Bangji, but they were not so far apart.

During the Song dynasty, Wang Li gives us the following sizes:
1 jin is 16 liang, and based on archaeological finds a jin is about 633 g, so a liang is just under 40 g.
1 dou is 9.5 L

So, redacted, and rounding slightly, our recipe is:
  • 2.4 kg honey for the primary addition
  • 20 g honey for the secondary addition
  • Water to make up 9.5 L
  • 80g good wheat yeast cakes
  • 40g southern baijiu yeast cakes
And following the steps above, which are pretty straightforward.  This is about a 1:4 honey:water ratio, which will make a medium strength mead at about 10-13% abv, give or take.  That explains why the author thinks it's nice, but not strong enough for drinkers used to 15-20+% wine produced by staggered additions of rice.

Next time, we'll take a look at a recipe which might be from Su Dongpo himself!

Creative Commons License

Wednesday, November 16, 2016

Mutton Wine

While reading Science and Civilisation about mead, I found this passage talking about the medicated wines in The Wine Canon of North Hill (北山酒經):
Two wines with animal components are included, the venerable tiger-bone wine and the estimable mutton wine.  The tiger-bone is still going strong, but the lamb appears to have lost its appeal. (page 237)
For a brief moment I felt safe in knowing that I didn't have a copy of the source text.  These were simpler times, full of innocence and happiness.  Then I though, "well, I should really make sure it hasn't been digitized yet" and lo and behold, has it.  Ugh.  It's in the third chapter ("下")


White Mutton Wine


In the twelfth month take thirty jin of the meat of a particularly tender wether.
Within the thirty [jin?] should be ten jin of fat.
Add six dou of water to the bones,
Put it in a wok and boil the meat, and make it extremely soft.
Strain out the bones, split and break down the meat fibers , and retain the broth.
When you are steaming the rice for the wine, evenly scatter the fatty meat on the surface of the mixed rice, and steam it until it is soft.
Mix it in a common bowl, use a full six dou of the meat broth, to completely spread the savory flavor throughout.
Steam it again, a good long time to get rid of what’s on the table top, and spread it to cool.
Select some good dregs, and mix it in according to the previous recipe.
Again take two sheng of meat broth, and put it together with what’s already on the table and the first liquid to be pressed out.
Pass a few days following the usual method for big wines, but only use yeast starters from the middle of the cakes [?].
One method is to distribute the dregs, and only put the boiled meat into the additions of grain.
Take dregs at one time and mix them into the weng.

The wine canon was published in 1117 during the Song dynasty, and according to the dictionary 王力古漢語字典, during the Song dynasty one dou was ten sheng, and one sheng was 670 mL.  One jin was 633 grams.  That makes the overall recipe call for 18.99 kg of meat and 4.154 L of broth.

I'm not making this because I don't eat sheep.  But maybe someone will.

Science and Civilisation goes on to point out that this wasn't just this one book.  Mutton wine continues on after North Hill, with books in 1235, 1330, sometime in the Yuan dynasty, and 1596 all having recipes.

Creative Commons License

Chinese Mead?

Carolingia, the Boston-area SCA group is planning an event in March called the Laurels Prize Tourney, where laurels (people who are Officially Very Good At Arts and Sciences) put out challenges and people try to meet them.

One of the challenges calls for well-researched mead, and I wondered if I could find any Chinese mead recipes.  I know that grain wines were far and away the dominant thing, but there's got to be something out there, right?

Well, I read through 46 pages of search results for every mention of "honey" after the Han dynasty (206 BCE - 220 CE) in the big online repository of Chinese texts and came up with nothing.  Rats.  I did learn that they preserved plums in honey, which is kind of cool I guess. doesn't have everything though.  And while I was taking a break around page 20, I decided to see what Google could turn up.

English wikipedia mentions this paper about a 1000-2000 BCE find supporting a mixed honey, rice and grape fermentation, but having just been disappointed with the Mijiaya beer (there's no recipe post because it tasted like hot mouth woe) I'd rather take something with a bit more evidence behind it than isotope and residue analyses.

Chinese wikipedia has a more relevant view of history but no citations.  We get a single relevant bullet in their "history" section:

  • Tradition has it that one of the first mead making regions in China was in Gongyan, during the Western Zhou dynasty in 780 BCE, where there was a record of mead making, which also flourished in the Tang (618 - 907 CE) and Song (960 - 1279 CE) periods.
And in the "Uses" section,
  • The Compendium of Materia Medica (本草綱目, first published 1578) thinks that mead can treat some various diseases.
Carrying on with the broad search, there's a poem by Su Shi, who lived from 1037 - 1101 CE.  I've translated it poorly:

There is a daoist in the west of Shu named Yang Shichang who makes great mead, very rich and strong.
I, having just obtained his recipe, wrote this song to bequeath it.
Pearls of rice water make sweet-wine, and in July the field-laborer's sweat runs bright.
But it doesn't compare to the spring urn's self-issuing fragrance, bees ploughing and weeding flowers to make grain.
One day, a small boil, with foam like fish spittle, the second day, foggy becomes clear bright and alive.
The third day, when you open the urn, the fragrance fills the room, I quickly pour it into a silver bottle without waiting for it to scatter.
A hundred coins a cup and thick without sound, sweet dew with no turbidity as clear as ghee.
The lord does not see the bees in the south garden picking flowers like rain, who Heaven taught first to brew wine and inebriate the gentleman.
Over the past years the gentleman has become poor to the bone, did he ever beg for grain?
All things in the world truly is unclear and uncertain, but the honeybee's great victory keeps watch over the river.

This is way more of a recipe than I expected from a poem.

Something like:

  1. Boil honey and water (how much?) and place in an urn.
  2. At some point it will clarify.
  3. After clarifying, it'll become fragrant.  Bottle it and sell it for $$$.
There are a bunch of medium quality articles on the Chinese internet that I'm reading through as well.

This one (which seems to be ripped from this paper which I can only read the abstract of.  Chinese internet yaaaay) claims to have a few recipes for mead, including the one mentioned in the above poem.

They say, and I translate,

In 1082 CE, Su Shi [aka Su Dongpo] obtained the secret mead recipe from the Western Shu Daoist Yang Shichang, which used sticky rice and honey as ingredients.  I personally brewed an "open fermenter and the fragrance fills the walls" mead. 
About the method for making "Dongpo mead," A Collection of the Records of Dongpo has this recorded: "My making of mead is built on [true one water disordered?], and for every dou of grain, use two and a half liang of steamed flour, like the normal method, take dregs from wine, and also put one liang of it into the steamed cakes to ferment them, and after three days, if the flavor is very hot and hard, and then add one dou of cooked grain to it; if it is sweet and soft, then every time you add cooked grain, add an additional half a liang; after another three days, add more, and it'll be done.  If all during fermentation the fermenter is full to the brim and you lose it from expansion, adding less water is advisable."
Tang dynasty pharmaceutical texts say that wine was from "millet, other millet, rice, honey, grapes and others" and that "Making wine or short-wine always requires yeast cakes, except honey and grapes, which alone do not. 

There are some leads here.  Hopefully I can get enough evidence to make a real recipe.

Creative Commons License

Monday, November 14, 2016

I Found a Better Name

I took a trip to the Metropolitan Museum of Art in late September because I thought there was an exhibit on Chinese laquer ware that was going to close soon.

I had the year wrong, it's opening and closing in 2017.  Whoops.

Instead, I took a stroll through the Chinese collection, on the lookout for drinking cups I could replicate, and took a ton of photos for my friends with research interests.  There was a lot of debate about what exactly this woman

is holding happening in the SCA China facebook group, and I happened to be walking past the statue at the time so I took some video and helped answer the question.

I also took a look at the scrolls section since I have friends who do beautiful calligraphy and are always looking for ideas.

I came across this beautiful scroll:

Which is titled, politely, "Drinking and composing poetry."  Except it isn't.  Here's the title colophon

Read right-to-left, these are foot-high characters that read


I don't think a better name for the blog exists, so I'm taking it.

Creative Commons License

Where We Stand, Where We Go From here

Pennsic was a whirlwind of activity, and despite me promising to update this blog "real soon I promise" life got in the way.

I think this is a good time to take inventory of what I've figured out over the past year or so.  There are a lot of unanswered questions, but I've made a ton of progress nailing down this process.

Yeast Cakes

I've got a recipe for "exceptional" yeast cakes that worked once, last year.

I tried to make it a month ago and it rotted 😑

But hey, it worked once.

I also learned that the warning about "don't store this in a sealed container or it'll turn black and get gross" wasn't kidding, since my cakes did just that.  Mr. Jia wasn't joking, apparently.

The recipe as I handed it out at Pennsic is posted at the bottom.


I made one rice wine for Pennsic that was delightful.  Sweet, dark, saucy, it was great.  I didn't get to panel it, sadly.  I made one short-wine a while ago that also worked well.  So I have a process that works sometimes.

Everything else has, frankly, been a disappointment.  There's a real challenge to growing the yeast cultures correctly so that the timing works, and I think the degraded yeast cakes are a hindrance.  I also think I still haven't steamed the millet right.

What you get when it doesn't work is either a sour, pale wine, if you had not-steamed-enough-millet.  The failed rice short-wines were... stranger.  Because the rice cooked fully, but didn't mash (converting starch to sugar), I was left with sour, starchy water with a viscous consistency.  It's drinkable, but not actually pleasant in any way.

Mouth-woe, in other words.

The basic rice wine recipe (YE1W1 for those following along at home) I passed out at Pennsic is at the bottom of this post.


I think I'm going to give up on millet for a while.  It's clear that I need polished millet, and figuring out how to do that seems impractical.  Later wine books only use rice, and I think there are flavor reasons for that besides the difficulties in steaming and polishing.

I need a new batch of yeast cakes, so I'll start that soon.  One question that's worth testing is whether I can use commercial whole wheat flour for the raw 1/3 of the cakes.

I also plan to iterate more on the rice wine recipes.  On suggestion of someone from the brewers' guild, I bought a 1 gallon widemouth pickle jar, and my local brew shop had a lid for it which can take a bubbler.  This will be a lot nicer than using a bowl, and hopefully let me watch the progress of waking up the yeast.

I've got more translations coming down the pike, but brewing is the bottleneck here, and I hesitate to post them without trying them.  The ones with pasteurized wine are super tempting, though.

I also want to congratulate my friend Vika on her induction into the Order of the Maunche, a big fancy award for being awesome at arts and sciences in the fun-medieval-times group.  She blogs about clothing and is pretty awesome, so check her out. 


Exceptional Yeast Cakes #1

Makes about 25

  • 3 pounds whole wheat berries
  • oil for frying
  • water
  • optionally, commercial yeast balls, which can be found in Chinese supermarkets
  1. Place 1 pound of the berries into the steamer and steam them until they are soft.
  2. While the berries are steaming, stir-fry another pound of the berries in a large frying pan, stirring constantly until they are yellow and fragrant but not burnt.  Immediately remove from the heat.
  3. Grind all three pounds of berries well.
  4. Combine all portions.  If using, powder and mix in the commercial yeast balls.  Form the mix into flattened balls about three inches across and one high, adding just enough water to hold them together.
  5. Place the balls in a clean box with limited but not completely restricted air flow.  They'll drip, so make sure it's on a plate.
  6. Appease your gods, and make sacrifices to statues made of the paste as necessary.
  7. After seven days, flip the cakes.
  8. After seven days, gather the cakes in a pile.
  9. After seven days, take the cakes out.  When dry enough, thread twine through the middle of them using a clean needle, and hang them to finish drying.
  10. Store in an open container.

Exceptional Yeast Cakes #1 Wine #1

Scaled to 100 mL of yeast cakes, which uses 2.1L of millet or 1.8L of rice and makes about 2 L of wine.
  • 100 mL powdered exceptional yeast cakes, about one cake
  • 2.1 L dry white millet or 1.8 L dry white sticky rice
  • water
  1. Dry the yeast cakes completely, and scrub them clean.
  2. Crush and grind the cakes into powder.  Soak them in 500 mL water for three days until they start to bubble, and add them to your fermenter.
  3. Rinse and soak 300 mL of dry grain overnight.
  4. If using millet,
    1. Steam the millet for 20 minutes in a steamer lined with cloth.
    2. Spread the millet to cool.
    3. When cool, steam for another 20 minutes or until fully cooked.
  5. If using rice,
    1. Steam the rice in a steamer lined with cloth just until steam issues from the steamer
    2. Take the rice out, and cover it with boiling water.
  6. Spread the grain to cool, and then add to the fermenter.
  7. Add water.  It's unclear how much to add.
  8. The next day, repeat steps 3-8 with 300 mL of grain.
  9. The next day, repeat steps 3-8 with 1L of grain.
  10. If using millet, three days later repeat steps 3-8 with 300 mL of grain.
  11. Several days or weeks later, once the mash has finished fermenting, transfer it into a cheesecloth or strainer bag and press it to extract the wine.  You will need a good amount of force to separate the liquid.  Fully pressed mash is about the consistency of cooked cornmeal, like a tamale.
  12. Optionally filter a second time through silk.  Liner silk works.
  13. Transfer to a closed container, let settle, then decant and bottle.

Monday, June 6, 2016

Mijiaya, or Going Retro

On the outskirts of Xi'an, the Chan River flows.  It eventually meets the Wei River, which flows into the Yellow river.  On a dusty terrace above the river is a site called Mijiaya, where excavations in the 2000s uncovered pottery dating from 3400 BCE to 2000 BCE.

Some of the earliest potter, from 3400 BCE to 2900 BCE, the late Yangshao period, includes parts of an amphora, an urn and a funnel.  This May, some archaeologists at Stanford published an analysis of residue on the pottery:

It's malted barley, along with millet, something called Job's tears, and some other starches.

I think this is awesome!  We know that alcohol production in China predates writing.  The Records of the Grand Historian (史記) from 50 BCE discusses King Zhou, the last king of the Shang dynasty, who reigned 1075–1046 BCE:
[He] made a pond of wine, hung the trees with meat, made men and women chase each other about quite naked, and had drinking bouts the whole night long. source
 Science and Civilisation in China includes pictures of brewing vessels - amphorae, urns, funnels - from that era, which have residues of Foxtail millet.  Liquid wine was found in a bronze Shang dynasty vessel at Tianhu Village in Hebei but no analysis was done on it, and one kind of alcoholic beverage, 醴 (lǐ) is mentioned on Shang oracle bones.

So we know that alcohol production was going strong by around 1000 BCE.  The myths surrounding the invention of alcohol put it firmly in "definitely fictional prehistory" territory (Du Kang, this blog's namesake, was the mythical inventor, and he supposedly worked for the Yellow Emperor himself).  Some historical texts push "documentable" wine back to the early Xia dynasty, but I don't trust historians writing about events that happened 1500 years prior to them very much.

The pottery evidence in Science and Civilisation does go back to 3000 BCE and beyond.  The authors suspect malted grain as the sugar source, as well as fruits.  There are a variety of different words for alcohol with different significances but that's not worth digging into here.

Science and Civilisation suggests that these beverages were made by mixing cooked grain (millet, rice) with malted barley.  There's a long argument about when the fungal mash I've been using in the rest of this blog came about, but by 200 BCE it was all anyone was using.  Qimin Yaoshu has instructions for malting wheat and barley, but only uses it to make sugar.

This is a bit outside of what I've been working on, but it should be a fun break while I work on making enough different wines for Pennsic.

Stay tuned for an attempt at a recipe!  I've got whole barley, Job's tears and more millet standing by.

Creative Commons License

Monday, May 23, 2016

Successfully Cooking Rice

I posted about rice cooking methods a little while ago, with mixed success, but I think I've got it down now.  The secret?

Soak the rice overnight.
Delicious and sticky after steaming for 20 minutes
I had initially thought that I could measure hydration of the rice by how much it expanded, and since it's done expanding after about an hour or so I stopped soaking there.  But what a difference soaking overnight makes!

Soaking didn't seem to affect the millet much, which might be why Qimin Yaoshu says to steam it twice.  It's also probably why I didn't have as much trouble with millet wines.
Note how some grains are fluffy and some are still dark
I tried steaming a second time since I was steaming the millet anyway, and poured a little bit of cool water over the rice and millet before returning it to the steamer.  The rice was 100% cooked throughout, as was the millet, although the millet feels a bit dry to the touch.

Creative Commons License

Mistakes Happen

I'm trying to get a short wine recipe to work, but it's tricky.

One of the problems is how long to incubate the yeast cakes before using them.  The recipe says:

Soak the yeast cakes until they bubble with bubbles the size of fish eyes.

And so I waited for bubbles.  They never came, and one day when I checked on the brew...
Sadness.  It's covered in a slick of fungus.

Creative Commons License

Monday, April 4, 2016

Methods for Cooking Rice

Previously, I'd struggled with adequately hydrating millet and rice while steaming.  You'd think this wouldn't be a problem, but once the grain is at the temperature of the steam, no more steam will condense on it.  If it's not already wet, it gets hot, but not really cooked.

Steaming grain twice (letting it cool in between) helped, as did wrapping the grain in cloth, but not enough.  I was missing something.

Reading through Science and Civilisation in China, Volume 6, Part 5, Fermentations and Food Science, which is far and away the most complete and well-researched English-language book on the topic of Chinese brewing, I've spotted a hint.

A line from a poem in the Shi Jing, the Classic of Poetry, this one from the 10th to 9th century BCE:


The given translation for this is:
We pound the grain, we bale it out,
We sift, we tread,
We wash it, we soak it,
We steam it through and through.
The poem is talking about how food is prepared for sacrifices, but the cooking methods are probably the same for normal food.

I'm not totally convinced by this translation.  My dictionary gives the words in question as an onomatopoeia for washing rice, so it'd be something like "we wash it washy-washy, we steam it fluffy-fluffy."  However.  Soaking seems to make a big difference.

I took 200mL of white sushi rice to make a short wine, washed it, and soaked it for three hours.  It swelled to 300mL and may have swelled more if I had let it soak longer.  That's a big difference in water content.

After steaming for 30 minutes (starting from cold water), the rice is nearly cooked enough.  The grains on top are a little al dente, but the rest are fluffy and fully cooked.  Soaking worked.  I steamed the rice a second time to make sure that it was thoroughly cooked throughout, although it still absorbed water after steaming and I needed to add more to get it fermenting.

I soaked more rice for about 12 hours and it expanded about the same amount: to 1.5x volume.

This helped with hydrating the rice, but wasn't really enough.  I needed to add more water to get it to ferment.  I might have been rushing it though, since I was trying to make it in time for Mudthaw.  It's possible that the alcohol produced by fermentation, combined with the freeing of the water of hydration bound up in the starch, will add enough liquid to allow the rice to continue breaking down.

Creative Commons License

Wednesday, March 30, 2016

Pressing and Straining

Chinese grain wines are fermented on the mash, which means that if you want to get something drinkable out the other end, you need to separate the grain from the liquor.

I initially was assuming this was happening without firm evidence, because the fermented mash is not wine by any means.
I've used cheesecloth with some success to filter it, but it's a little difficult because you have to squeeze it pretty hard to get the wine out, or else sensuously massage the bag of mash with your hands to allow it to continue draining.  I think that it's hard to drain because the grain particles are pretty fine, and clog up the filter.  Beer mashes can run into this problem occasionally, but don't usually because they're less porridge-like since the grain isn't cooked.

Straining through cheesecloth has been what I've been doing, but you still end up with a lot of sediment coming out of the bag.
Some of this sediment will settle out, particularly from millet wines, but not all of it, so you have to either decant carefully or deal with the sediment, which changes the flavor profoundly.  It's also why I poured pancake batter at Pennsic last year.

It does clear out in millet wines, though:
I suppose this is why it's called "yellow wine"
Rice wines so far have had much more sediment that's mysteriously immune to gravity's inexorable pull.

But!  I've discovered some more evidence in the Qimin Yaoshu.  

When the wine seems finished, press it out [using a bamboo screen over a trough, and a board], and decant it so that it is clear [ed. note: scroll 8 “如壓酒法,毛袋壓出” has more information on pressing].

This is in the recipe for YE4W1, 神麴酒方:A method for making wine using exceptional yeast cakes.  Most recipes don't specify pressing but I think it's implied.

The editor's note is a little wrong.  In scroll 8, there are a number of recipes for making vinegar which start by making wine.  A few of them mention the text included above, which is "a method for pressing wine: press it out through a hair/wool bag."

We also have this line from YE3W2, a short-wine, about processing the yeast starter culture:

Take a wool bag [described in scroll six, under “raising sheep,” called a 毛堪酒袋] and use it to strain out the yeast cake sediment, and then additionally filter it using juan [plain, tabby-woven silk], and put the liquor into the weng, and then add the rice.

Which mentions a two-step process.  This is what I think we can tell so far:

  1. It's reasonable to assume that most wine was pressed through a hair/wool bag or something equivalent.  You don't really get wine otherwise.
  2. Even pressing that way doesn't get you clear wine, as my experience has shown.
  3. Using a second stage filter of silk is documentable for the starter culture, and is probably a place to start for filtering finished wines.
The first question that arises is what kind of bag are we talking about?  Carded or combed wool?  Is it felted?  What was Chinese wool technology like?

I spent some time reading Science and Civilisation in China, Volume 5, Part 9, Textile Technology: Spinning and Reeling hoping to get some insight into wool use, but it doesn't mention wool at all.

The Qimin Yaoshu does mention sheep rearing, and wool handling, though. It specifically mentions felt production, but has no other discussion of woolen fabrics. This fits with my understanding that wool in China was not really raised, but that there was a small amount of exchange with herding people to the north, and the Qimin Yaoshu is from enough of a border region to include that kind of information. There are also recipes for cheese!

In the wool section, it talks mostly about "white sheep," and then talks about "black sheep" which need to be sheared later in the season.  There's an annotation in the text which follows.  I don't know when or who the annotation is from, but it's all we have right now:

[black sheep] hair is suited for wine sacks, and when doubly-plied [as rough yarn?] has this benefit: it is much sleeker than white sheep[‘s wool].

This doesn't make a lot of sense on its face until you realize that 羊 means both sheep and goats. I think they're talking about mohair.  I got to feel a sample of a coarse-woven mohair fabric and I think it would do the initial filtering well.

As for the second stage of filtering, I have yet to get my hands on the right kind of silk.  I tried using nylon bags that modern brewers use to strain out grain, with little success: they are not fine enough to catch the tiny particles polluting my rice wine.  They might be a good initial filter, though, since they seemed to clog less than the cheesecloth.

To do: get some mohair fabric, or fibers and sweet-talk some spinners and weavers into making it into fabric, and to get some tabby-woven silk that's dense enough to filter well.

Creative Commons License

"Boiled" Millet Wine

I brought one new wine to King and Queens Arts and Sciences : a "boiled" millet wine.  This is the first wine for the second "...