Wednesday, February 21, 2018

"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 "exceptional" yeast cake recipe in Qimin Yaoshu - the yeast cakes I'm actually using at the moment.  It's got something weird going on.  Let's dive in.

YE2W1 造神麴黍米酒方:To make proso millet wine using exceptional yeast cakes

Finely file the yeast cakes, and air them in the sun to dry.  For one dou of yeast cakes, use nine dou of water, and three dan [30 dou] of grain.  If you need to make more, keep the ingredients in this proportion.  Use an appropriately sized weng.
When the mulberry trees lose their leaves [Chinese months 9 or 10, mid October - mid December].  First, add one dan of grain, then add five dou, then four dou, and then three dou, adding each after the grain has dispersed [note that this only adds up to 2.2 dan, not 3], making sure to add it at the right time.
Once the flavor is complete, boil it until it is completely cooked.  Although the flavor may be right, don’t stop boiling, or else you will not stop the fermenting power of the yeast, and you should once again add grain.  If you do not add more grain, then the wine will taste bitter, and be thin.  If you hit the spot, the wine will be light and fragrant, and better than most.
If when you first make this wine, and is clearly too thin, then what?  Think about it similarly to how you think about using common yeast cakes: you probably used too little grain, so that the fermentative power of the yeast wasn’t exhausted, and that’s why it was thin.
Don’t let chickens or dogs see [you making this wine].  If you focus on something while making this when the mulberry leaves fall, focus on making sure the millet is extremely cool.

Redacted a bit:
Clean, dry and powder 1 part yeast cakes and soak them in nine parts water until they bubble.  Add cooked millet according to the following schedule, waiting until the grain has dispersed before adding more: 10 parts, 5 parts, 4 parts, 3 parts.  Once it's done fermenting, "boil" it, and then add 8 parts cooked millet.  Let sit for a few days, press, filter and let clarify before serving.

Millet at left, before boiling.
Weird.  Why are we boiling our wine?  The first time I did this, I left it in the pot for a bit figuring it was sterile.

It wasn't!

It rotted and smelled something awful.

This time around, I tried something different: what if "boil" means more like "heat until lots of bubbles come out?"  Here's my theory:

If we're boiling it in a pottery vessel over a stove as the technology of the time likely was, and that vessel is anything like full, you've got a pretty deep bed of liquid going on.  Because it's a thick porridge, it won't convect much, plus you have a TON of nucleation points.  So the bottom will boil long before the top gets hot enough to.

I tried this in a modern kettle and while I wasn't the most rigorous with it, I got something like a temperature of 165 °F when it was "boiling all over."  Now, the recipe exhorts us to keep boiling, but what if we stopped once it was good and bubbly?  165 °F is plenty hot to sterilize it and kill the yeast... but readers familiar with beer brewing will notice that it's exactly the temperature at which barley amylase works ideally, and not nearly hot enough to denature it significantly.

I added the last, large batch of millet.

I waited a few days, and then performed a starch test with iodine.  If my theory is right ,we should see no starch, with all of it converted to sugar over those few days.

I cleaned my kitchen after I realized how dirty it looks in this photo
The iodine stayed brown, which means that there's no starch.  This is from a liquid I dumped half a liter of millet into three days before.

I think what's intended here is that this will produce a sweeter wine.  By killing the yeast off, we stop further fermentation, and we could add sugar, but we've already got a ton of amylase, and grain is cheap.  I didn't really get that effect, but the starch is gone.  There was an almost imperceptible sweetness, but I think I still have other millet cooking issues to tackle.

I'll have to revisit this theory in the future.  Maybe brew a batch using rice to test the amylase theory better.

Creative Commons License

They liked me so much, they got me a ROCK!

Painted image copyright this person whose real name I don't know, although if you have a burning desire to make a copy just copy the thing it's based on, honestly.
This past weekend was King and Queens Arts and Sciences championships in Mt. Kisco, NY.  I had entered it two years ago when this research was in a much, er, less tasty state, and I was still figuring out some of the process details.  There were ~40 entrants, and, as I did two years ago, I made the second round of judging, but did not win, which is totally fine.  I love these events because A&S is the thing I like the most about the SCA, and the best of SCA A&S is on display there.  Unfortunately, I didn't get to dig into the other entrants as much as I would have liked to because I got to talk to so many people!

A neat thing happened there: I apprenticed to Magnús, which in SCA land means something like "we're officially science-friends."  This is cool because it means I get to bug him with questions more than usual before he starts complaining.

I also got an award at the event which is really meaningful to me: a Maunche, which is for doing a lot of cool A&S stuff.  I guess I've done a lot of cool A&S stuff now.

It's tradition to get a "scroll" for each award you get.  Apparently it was decided that I would get this scroll back in September,1 and my friends decided that a parchment scroll would be lovely but that they could do better - my persona is approximately viking after all.  So my friend Vika wrote a very nice praise poem in Kviðuháttr meter, which Magnús transliterated into runes and Feilinn painted them on a rock.  This is awesome and I'm floored.

The rock includes slightly modified images from the Stora Hammars I picture stone:
Copyright Wikipedia user Berig, CC By-SA 3.0 and GNU FDL
A different nearby stone depicts Odin as a raven stealing mead and this is amazing.

I didn't really brew anything you haven't seen before, (well, except for the millet wine which you can read more about here - it wound up being pretty standard) but I'm gradually refining my technique.  I brought three wines:

  1. The headliner, Rice Wine à la Yuan Pushe (YE1W2).  This is the steam-and-pour-over rice wine that has become dependable for me.  Still sour, but not bad.  I actually enjoyed drinking this one.
  2. YE2W1, a boiled millet wine.  I'm getting better results gradually, but something is still really wrong with my millet processing.
  3. YE3L1, the non-glutinous rice short wine I made at Pennsic.  Producing this to hit a specific date is a PAIN, and I went back and forth between "oh no I'm far too late" and "oh no I'm far too early."  I wound up being perhaps a day early - it was less sweet than I wanted.
One thing I was proudest of was my documentation.  Of course, I have access to sources that nobody else is using, which is great, but I spent a lot of time writing, and got a lot of help editing, so that it's clear what my thought process is and how it led me to the conclusions I reached.

Take a look!  Please let me know if you find it helpful, or if you think there is something I can improve on for next time.

I also thought my display worked well.  I took heavy advantage of my work-sponsored trip to Japan and brought back pottery, lacquerware and some textiles which I had bought specifically for this purpose.

Photo copyright Dayna Tarabar, used with permission

Left-to-right on the tray I had the millet wine, the rice wine, and the short-wine.  In front I have the yeast cake, millet, sticky rice, soaking rice, cooked rice, and unpressed short-wine.  This way, I was able to give a good overview of all the steps in the process.  I also had a flask of water for rinsing.  The tablecloth is a somewhat-accurate reproduction of an 8th century textile held in Japan.

1. This is a "kingdom-level" award, so it's decided by the King and Queen, who rule for 6 months, and made their decisions early on in Their reigns.  Then comes the business of finding the right event to give it to the person, with as many of their friends there as possible, without actually telling them ahead of time.  It takes a ton of back-channel communication, but the effect is tremendous, and the recipient's friends get to giggle about the surprise in private for a good long time.

Creative Commons License

Friday, September 22, 2017

Survey of Vinegar Recipes

Scroll 8 in Qimin Yaoshu has twenty-three vinegar recipes in it.  I've now translated them all, and rather than posting them fully, I thought I'd offer a summary, and a few of them in particular.

I haven't (successfully, ahem) made any of these yet, and when I do I'll post separately about them, with the full recipe included.

Science and Civilisation in China notes that vinegar was a relatively late addition to Chinese cooking, replacing the use of Prunus mume (a kind of apricot) as a souring agent in the Spring and Autumn period (771-476 BCE).


Most of these are millet vinegars, produced like a millet wine.  But they also don't all use yeast cakes.  Most are using what I'm translating as "wheat grains" (麥䴷), which are also called "yellow steam" (黃蒸).  To make this:

作黃蒸法:To Make Yellow Steam

In the middle of the sixth or seventh months, take fresh wheat, and finely grind it.
Soak it in water, and then steam it, and when thoroughly steamed, take it out, and spread it to cool.
Set it out, cover it, and treat it like wheat grains.
Again, do not winnow it; consider how it will diminish.

We also see a few other kinds of millet, barley, millet husks, and water, and some vinegars that just use wine as the primary source of fermentable.


Most of these are more or less just mixing together the ingredients and letting it go, although they sometimes have specific timing instructions, or cooling instructions.  It seems to have been a big concern not to ferment too hot, and to cool the fermenter with cold water on the outside.

Most recipes call for cover with a silk floss cloth.  This will allow in the air needed to turn the alcohol into vinegar.

作大酢法:To Make Great Vinegar
1 part wheat grains
3 parts cooked foxtail millet
3 parts water
秫米神酢法:To Make Exceptional Glutinous Millet [Sorghum?] Vinegar
1 part wheat grains
3 parts glutinous millet (maybe sorghum) or glutinous proso millet
10 parts water
又法:Another Method
1 part wheat grains
3 parts cooked foxtail millet
10 parts water
又法:Another Method
1 part wheat grains
9 parts cooked foxtail millet
9 parts water
粟米、麴作酢法:To Make Vinegar From Foxtail Millet and Yeast Cakes
1 part common (previously, “lumpy”) yeast cakes
10 parts cooked foxtail millet
10 parts water
秫米酢法:To Make Glutinous Millet Vinegar
1 part powdered yeast cakes
10 parts glutinous millet
Soured water from rinsing the grain, enough to make a thin congee
大麥酢法:To Make Barley Vinegar
1 part wheat grains
1 part finely milled barley, 0.05+ parts foxtail millet
3 parts water
燒餅作酢法:To Make Vinegar From Roasted Cakes
1 part wheat grains
Some roasted wheat cakes
3 parts water
迴酒酢法:To Make Vinegar From Turned Wine
1 part powdered yeast cakes, 1 part wheat grains

50 parts unpressed wine (that spoiled before pressing), 10 parts water
動酒酢法:To Make Vinegar From Spoiled Wine

10 parts wine (that spoiled after pressing), 3 parts water
又方:Another Method
1 part wheat grains
6 parts cooked proso millet
20 parts wine
神酢法:To Make Exceptional Vinegar
1 part yellow steam
3 parts steamed bran
Enough water to cover
作糟糠酢法:To Make Vinegar From Dregs and Husks

1 part wine dregs, 1 part foxtail millet husks

酒糟酢法:To Make Vinegar From Wine Dregs

Wine dregs not pressed to dryness, crushed grain

作糟酢法:To Make Vinegar From Dregs

4 parts cooked foxtail millet
20 parts water-diluted wine dregs
《食經》作大豆千歲苦酒法:To Make 1000 Year Great [soy?] Bean Vinegar from the Classic of Food

1 part great beans
Unfiltered wine
作小豆千歲苦酒法:To make 1000 Year Small Bean Vinegar

5 parts small beans, steamed proso millet to cover
30 parts wine
作小麥苦酒法:To Make Vinegar from Wheat

3 parts cooked wheat
20 parts thin wine
水苦酒法:To Make Water Vinegar
2 parts glutinous rice yeast cakes, or wheat grains
2 parts coarse grain
10 parts clear water
新成苦酒法:To Make Newly-Formed Vinegar
1 jin [unit of weight] roasted crushed yeast cakes
1 dou proso millet, but also 1 dou foxtail millet if you want it to be good.
5 dou water
烏梅苦酒法:To Make Vinegar From Smoked Prunus mume

1 part smoked Prunus mume
5 parts vinegar [makes a sort of dried vinegar]
蜜苦酒法:To Make Vinegar From Honey

1 part honey
10 parts water
外國苦酒法:To Make Vinegar as Foreigners Do

3 parts honey, cilantro leaves
10 parts water

There's another vinegar recipe in chapter 34, the chapter on fruit trees, which is made entirely from windfallen overripe peaches.

One thing I find interesting is that some of the later recipes (starting with the great bean vinegar) are 
coming from a now-lost text, the Classic of Food, and for one, Newly-Formed Vinegar, the author of Qimin Yaoshu haas commentary which reads, "I tasted this vinegar, but it was not good.  I added one dou of foxtail millet, and fouteen days later, it was clear and beautifully heady, but not better than the great vinegar [the first recipe]."

I think it's really interesting to see the author give more than just dry recipes - he clearly has a standard to which he compares the others, and thinks the recipe he's inherited isn't so great.

Sunday, September 17, 2017

Evidence in Funny Places

I'm working on the vinegar recipes in Qimin Yaoshu, and one of the recipes contains a line that offers some useful insight into the wine making business:

When it has become the tenth month, press it with a wool bag as if you were pressing wine, and then store it.

Seems like I've been on the right track using cheesecloth and nylon bags, although I would like to get a bag made from mohair or something like that eventually.

Wednesday, August 2, 2017

Wine that tastes good!

Having just made new yeast cakes, I put them to good use in a batch of millet wine and rice wine following the recipe I call YE1W1 and YE2W2, as they're the first and second wines (W1, W2) listed under the first exceptional yeast cakes (YE1).  They came out really good!

The recipe:

Millet or Rice Wine à la Yuán Púshè

This volume will fit in a 1 gallon wide-mouthed fermenter, barely.

Ingredients and Tools

  • ¼ cup powdered exceptional yeast cakes, about one cake.  But do measure.
  • 5.25 cups dry millet or 4.5 cups dry white sticky rice
  • Water
  • 1 gallon, wide-mouthed fermenter - cooked grain can’t go through a narrow mouth
  • Cheesecloth
  • Optionally, liner silk or some other filter fabric finer than your cheesecloth
  • Mortar and pestle


  1. Dry the yeast cakes completely and scrub them clean several times.
  2. Grate, and then grind the yeast cakes in a mortar.  Soak them in 1.25 cups water for three days until they start to bubble.  Add to your fermenter.
  3. Thoroughly rinse and then soak ¾ cups of dry grain in water overnight.  If using millet, polish it in a food processor first.
  4. If using millet,
    1. Steam the millet for 20 minutes, lining your steamer with cheesecloth.
    2. Spread the millet to cool.
    3. When cool, steam for another 20 minutes.  The millet should be fully cooked.
  5. If using rice,
    1. Steam the rice just until steam comes out of the steamer.
    2. Cover the rice with boiling water.
  6. Spread the grain to cool, and then add to the fermenter.
  7. Possibly add extra water, no more than just to cover the grain.
  8. The next day, repeat steps 3-8 with 1.25 cups of grain
  9. The next day, repeat steps 3-8 with 2.5 cups of grain
  10. If using millet, three days later repeat steps 3-8 with 300mL of grain.
  11. The mash will take a few weeks to ferment.  Wait until all the floating grains (“ants” in period) settle.  Transfer it into a cheesecloth or strainer bag and press it to extract the wine.  You can use your hands, or a board in a trough.  To use your hands, twist the bag and massage the mash.  Fully pressed mash is about the consistency of cooked cornmeal.
  12. Optionally filter a second time through densely-woven silk.
  13. Transfer to a closed container, let settle, and then decant and bottle.

This worked much better than in the past because I polished the millet, because I added enough water to let the millet ferment, and because I'm using wide-mouthed gallon jars which let me ferment on the grain for weeks.  This makes a huge difference.

I'm pretty sure I wasn't letting any of my past batches actually ferment to completion before pressing them, leaving them sour because a lot of the starch got left behind and not turned to sugar.

Commentators and poems about wine in period describe wine that isn't done yet as "muddy" or "pearly floating logs" or "floating ants."  This matches what happens at least in the rice wine very well:
Left: millet.  Right: rice.
The floating rice grains on the right are the logs or the ants, and there's a thin foam floating with them.

I would have liked to wait until it was fully settled, but Pennsic calls and I'm heading off.  So I pressed the wine, filtered it, put it back in the fermenter and chilled overnight to settle out what sediment I could, and then bottled it.
Left: millet.  Right: rice.
The rice pressed easily and filtering through silk wasn't awful.

The millet, not so much.  Again, I had to deal with a lot of small grain bits that gave it a pancake-batter feel, and fouled the silk repeatedly so I had to stop filtering and wash out my cloth.  I got 3 wine bottles worth of the rice wine, but only one of the millet.

No wonder millet fell out of favor!  I suspect that with more water and a longer fermentation the millet, too, might digest better and not be as hard to press.

Look at the color difference from the first time I made this same wine recipe with millet:
Some of this is because the first round of yeast cakes had some caramelization happen during curing that leaves some color, but I think that this is more due to the millet hulls leaving color.  With polishing, they're partially removed so they don't color the wine.

This stuff actually tastes really good.  Way, way better than older batches.

Tasting Notes

As always, these are subjective.

Millet Wine


Nearly white, with a pearly cloudiness.  Will probably clarify more as it sits.


Mostly soured grain notes with some nuttiness and a bit of alcohol.


Lightly sour but not overly so.  Balanced with a faint sweetness and a medium-light mouthfeel.


Nutty, sake-like aroma.

Rice Wine


White with a milkly cloudiness.  Also will probably clarify more.


Pears!  Some alcoholic notes and a hint of grain.


Lightly sweet, just a hint of acidity on the swallow, and a bit of a burn.  Medium body


Pear notes again, mixed with some sake character.

I really like the rice wine.  Catch me at Pennsic for a taste!  My class is Tuesday at 2:00 in A&S 12.

Creative Commons License

New Yeast Cakes

My first batch of yeast cakes is running out, and I think they had some process issues, so I wanted to make a new batch.

I decided to try the second yeast cake recipe:

YE2 又造神麴法:Another method for making exceptional yeast cakes

Take equal quantities of steamed, stir-fried and fresh wheat, and join them as in the previous recipe, but without the north-south cross, the offering of alcohol and dried fruit/meat, the cooked noodles, offering to the “yeast cake king” or having children shape the cakes.
Prepare the three kinds of wheat as before, join them and finely grind them.  On the third day of the seventh month, make the yeast cakes.  Wet them so they are “firm [more water than “just add a little”].”  Pound the mixture until it is evenly mixed and fine.  Make it into cakes using a round iron mold, 5 thumb-joints across, and one and half thumb-joints thick.  On top of a level board, have a strong person really stomp on it.  Use a peg to pierce a hole in the middle of the cake.
Clean and sweep a room with an east-facing door, and spread the cakes over the floor.  Block the windows and doors, using mud to seal any cracks, so as to not allow wind to enter.  Flip the cakes after seven full days.  After a second seven days, gather them, and both times re-seal the room.  After a third seven days, take them out, and air them in the sun until dry.  The cakes are finished.
Whenever you please, hang the cakes for storage in a high, dry place.  These cakes do not keep if stored in a weng.  Cakes that are stored in a weng turn black and vile, and around the hole they will be black and rotten.  If you want to have people help you, you must at least make sure that the three types of grain are equal, and don’t let three dan of grain be a limit.
One dou of these yeast cakes can ferment three dan [thirty dou] of grain, while one dou of common yeast cakes can ferment six dou of grain: whether use a little or a lot decidedly depends on this.  Use Qixi burnt wheat yeast cakes or spring wine yeast cakes like common yeast cakes.

This is nearly identical to the first one, but presses the molds into cakes.  I used a cheese press I had lying around.
Mixing the three kinds of grain in a food processor.


I'm using this metal tray, and using sushi rolling mats to keep the bottoms from getting too damp.

Flipped after a week.

Gathered after another week


These are a little grosser than the first ones.  I scrub them before using, since the recipes say to, and it seems to make sense.

Grating it on a box grater reveals the inside.

Brushing these gave me a mild allergy attack, so I'm going to do it outside from now on.

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 "...