Brewing with Egil: Holy Hell, It Doesn’t Suck! (Or: Crossing the Streams)

Last time, I talked at length about my latest failure in reproducing Viking-era ale: concentrated essence of vodka-soaked hot dogs.

Can you taste it yet? Can you imagine the cornucopia of flavor in your mouth? The delicious salty smokey flavor of highly-processed extruded meat product, coupled with the mouth-puckering taste of soured honey water, wrapped in the delicate aroma of a half-digested lunch?

My friends, science is dangerous. It removes that part of the brain where “common sense” usually resides, and causes us to put unwise things in our mouths.


But today is not a day for unwise mouths. Today is a day of triumph! The hot dog wars are won! And I’m going to tell you about it!

It’s funny every damn time.

So my latest attempt at making this stuff saw a revisit of my approach. After considerable additional research and discussion with various people, I envisioned a two-part production system for Viking boozemahol: part one involves producing an iteration of a non-alcoholically fermented nutritive/medicinal grain beverage, and then using that product to start a larger quantity of sweet liquor.

I added oil seeds back to a dried malt biscuit (made using peat malt) at a rate of 1:1 by volume, to mimic the ratios documented in the Talmudic description of “zythum” (1/3 grain, 1/3 oil seed, 1/3 “salt” or probably “brine”). The biscuit was already salted, so it was a matter of tossing in some oil seeds. I used a 1:4 biscuit/seed mix (1/2 cup crushed biscuit and 1/2 cup flax seeds), mixed into a quart of water and heated slowly on the stove. The product allowed to cool, and then poured into a Mason jar. Add yeast and wait a while (about a week), then strain the liquid out.

The nutritive beverage essentially serves as a yeast production medium, allowing the yeast to multiply without actually producing significant alcohol. They also digest the grains and assist in extracting oil from oil seeds in the process, creating a sort of nutritive “liquid bread” that could also be used as a medicine. This actually makes a lot of sense; yeast require unsaturated fatty acids (read: plant-derived oil) during growth to synthesize their cell walls, and some nerds have run experiments confirming that, indeed, adding oil to a yeast starter greatly promotes yeast growth – more than even oxygenation.

The experiment above was an interesting one: the filtered liquid had the consistency of oil, was slightly carbonated, had no discernible alcohol (despite having gone for a week), and tasted fucking awful and smelled even worse.

Imagine vomit plus slightly rotting grain and a vague hint of olives. Then put it in your mouth.

That, my friends, is science.

Those who have been following along for a while may remember that my initial research already pointed me at this conclusion – that is, two different production streams for grain beverages in the ancient world: an unfermented or lightly fermented medicinal/nutritive beverage, and a strong alcoholic drink. I had previously thought those two streams were separate, but shared common ingredients. This latest production method essentially involves crossing the two streams at the exudate of the medicinal beverage.

“When someone asks you to cross the streams, you say ‘Back off man, I’m a scientist!'”
I believe this constitutes the greatest density of social currency ever dropped in one place.

In other words, the processing method for making this medicinal beverage would also produce useful byproducts that could be employed for producing other things. Recall that Pliny describes a use for Egyptian “zythum” (or rather, its “spuma”) as a cosmetic applied to the face. Given the oil content of the product I produced, I can see this being rather plausible. Vegetable fats plus vitamins from growing yeast would likely make an excellent facial moisturizer or similar skin treatment. My previous post also discussed my reasoning for seeing this as a precursor to an alcoholic beverage; Anglo-Saxon vocabularies and leechdoms indicate that “gruit” might mean something like “dregs,” both “zythum” and “beor” carry warnings against consumption by pregnant women (indicating that common ingredients may exist in both), and the 14th century Le Ménagier De Paris mentions using “leveçon de cervoise” to start an alcoholic drink.

With all that in mind, I tried it out. I mean, the big question is: does it work?

BwE - The Jug of Stuff BwE - Things In and Around My Mouth

I had ~400 ml of weird oil stuff from the above experiment. I mixed that with 1.6 L of honey-water (which itself was concocted from 320 mL of honey and 1280 mL of water), and have been letting it do its thing for about 5 days now. I literally tasted it an hour ago, and it’s awesome. It’s light, pleasant-to-drink, slightly carbonated, somewhat alcoholic, and has a nice balancing character that I can only presume is added by the medicinal compound. Oddly enough, it smells something like salted olives (which isn’t too surprising, given that it contains both salt and unsaturated vegetable fat); there’s no olive flavor, though there is definitely a savory component that is balancing the sweetness one would normally encounter in such a beverage.

I omitted fruit this time around (so as to not have too many flavors mucking things up), but plums or polar berries would go quite well in this beverage.

And so, I am pleased to report tentative success! We’ll see how this product develops as it continues to age – though I suspect that it would’ve been drunk relatively early on in the Viking era.


Restating the Process

I’ve been told by some that navigating this work is an onerous task, and trust me – it ain’t that easy for me either. So I’m going to attempt a written recap with links to my evidence/prior work; down the road, I hope to construct a sort of “roadmap” that diagrams all the connections between my various findings and pieces of evidence, to help people (including me!) navigate the murky waters of speculation.

I. Medicinal/nutritive grain beverage (MNGB):

Both Pliny and the Talmud discuss the product called “zythum,” which is described as a combination of grain, oil seeds and/or herbs (unclear from the text), and salt or brine; a medicinal use is indicated. Pliny discusses it in conjunction with other grain products (one of which is “cerevisa”), and outside of the context of alcohol. He indicates a similar use for all of the products. Pliny also discusses numerous other remedies derived from barley, and discusses the grain/herb/salt/oil seed compound “polenta” and its uses as a nourishment/medicine.

The Anglo-Saxon vocabularies compiled from the 9th to 12th centuries indicate that the word “eala” is cognate with “cerevisa,” which is one of the grain beverages also discussed by Pliny in his mention of “zythum.” He indicates similar uses between the various products, which I say serves to demonstrate possible similarities between the products. The AngloSaxon leechdoms indicate several medicinal preparations, many of which make use of “eala/ealu/ealo,” all of which are synonymous.

Thus, I conclude a degree of continuity between “zythum,” “cerevisa,” and “eala.” All three appear likely to be forms of a nutritional/medicinal grain drink. The probable ingredients for the nutritional/medicinal grain beverage are grain, oil seeds, salt, and possibly herbs. Pliny’s mention of “spuma” likely indicates that yeast were a factor in the drink, though it was probably not an alcoholic beverage.

Archaeological evidence shows that Vikings had small unleavened “breads” consisting of grains, oil seeds, and local herbs.

II. Strong (likely alcoholic) drink

Beór is seen in the vocabularies as being cognate with “hydromel” or “mulsum.” The leechdoms indicate that pregnant women should avoid drinking it, or else they will give birth prematurely. This is listed separately from precautions against drunkeness, indicating that beór is special in some regard. “Zythum” carries a similar warning, despite being glossed with a different drink (“cerevisa”); it is plausible that the two drinks share some kind of linkage.

Beór is glossed differently than eala, and the two are listed as separate options for certain preparations in the leechdoms. This indicates that beór and eala are separate products with different considerations. In some rare cases, the two appear connected by the gloss “sicera,” indicating that while they are separate products, they are probably related in some capacity. Beór is the etymological root of the modern word “beer.” It appears repeatedly in Anglo-Saxon literature, and is defined in Bosworth-Toller as “strong drink.” This gives it a separate context of use from “cerevisa,” primarily associated with celebrations.

It appears that beór is different from, but related to, “cerevisa” and consequently “zythum.”

The poem “Alvissmal” seems to indicate a relationship between the Norse “ol” and the Anglo-Saxon “beor.” In the literature, “ol” is similarly associated with celebrations.

III. Using “gruit” from MNGB to make a strong drink

Bosworth-Toller indicates that “grut” is a remnant of another product: “condimentum cerevisae.” It also means “fine meal,” and is glossed accordingly (with “pollis” in the vocabularies). “Grut” is the etymological root of “gruit,” commonly understood today to indicate a mixture of herbs used for flavoring beer.

In the Treatise of Walter de Bibbesworth, we find a word glossed with “grout” that is meant to be of wheat or barley, used in conjunction with malt to produce an alcohol. This falls in line with the meaning of “grut” as “fine meal,” and also indicates its use (somehow) in fermentation.

The 14th century Le Menagier de Paris talks about leveçon de cervoise used to start an alcoholic beverage.

My conclusion is that “gruit” was likely a grain byproduct that remained after the fermentation of some other beverage. The leavings or some other exudate would be used to produce an alcohol. Given that the word occurs in conjunction with words related to “cerevisa” quite frequently, I speculate that “gruit” is derived from “cerevisa” and/or “zythum.” This would mean that “gruit” also contains herbs and residual oils from the oil seeds in the product; the herbal connection is supported by inclusion of “gruit” in the leechdoms, and further serves to explain how “gruit” came to be associated with a purely herbal product later on.

Given that “beor” and “zythum” carry similar pregnancy restrictions, it seems plausible that one is used to make the other. More than likely, the initial medicinal product is filtered and used for its purposes, and the remaining dregs are used as a starter for an alcoholic beverage. “Zythum” uses safflower oil, and the Viking-era finds have flax seeds; both can be abortiofacients in sufficient quantity, so there seems to be an element of truth to that.

The use of a “starter” beverage to make a “strong” beverage is reflected throughout history and the modern era.

Likely, a specialized vessel was used for preparing the medicinal beverage, as I previously speculated. The strong drink could be derived from honey, fruit, malt sugars, or any combination thereof. Merryn Dineley is researching the equipment Vikings may have used to convert grain starches to sugars.


So there we have it, I think. A precursor medicinal/nutritive product leaves behind dregs that can be used to make alcohol. The medicinal product was probably manufactured in a specialized vessel containing a persistent yeast strain – the yeast would absorb into the wood, allowing the leavings (or perhaps top-cropped yeast) to be used to make alcohol. This would be a very efficient system of production, and would also likely fall to a few specialized people.

My most recent experiment uses the whole product – whether or not this was done is unknown. Future experiments will utilize only the dregs to attempt a fermentation.

Alright, that’s enough for tonight. Digest! Read! Drink! Be merry! Go forth and appreciate that you can just buy a goddamn beer instead of having to write a paper detailing its production.

And I know you probably think I don’t have to write a paper about beer to have beer – but then again, you’re probably a normal, reasonable person.

I’m a scientist. We don’t do things the easy way.


Brewing With Egil: Revisiting the Past

Huzzah! The move is completed! Life has begun to settle back down, and now I think I can return to a normal-ish update schedule: every other week, on Sunday.

We shall see how long that lasts.

I’ve written a fair bit about the necessity of being wrong and on the need to occasionally revisit your work because of that. This is all good and well. So today, I’m going to revisit some of my earlier research and share with you what I’ve learned since. I’ll also report on my most recently-completed Viking-era brewing experiment (preview: it was awful), and document the next attempt I started literally today.

That’s right, I’m reporting this to you live.

“Gruit goes in, ol comes out, never a miscommunication. You can’t explain that.”
To be fair, they couldn’t either.

On the Meanings of “Gruit” and “Mealt”

A bit ago, I talked about my speculations on the exact meanings of words such as “gruit” and “mealt,” and how they may have actually been implemented. I’ve been revisiting my conclusions, and inspired by some other evidence, have been winding down a slightly different path.

Previously, I argued that both “mealt” and “gruit” referred to an herbed grain mixture. This still holds for “gruit,” I contend; between the meaning of “grut,” the connections to herbal remedies in the Anglo-Saxon leechdoms, and the method described in the Talmud (which is also medicinal), that connection is well-established.

“Mealt” is a little stickier, though. I never had a solid connection to herbs, and sort of presumed it by its connection to “gruit.” Recently, I discovered a 19th-century translation of the Senchas Márone of the texts compiling ancient Irish law. In it, on pages 241 – 243, we find a description of both the production and the “testing” of malt – which is called “braich.” This may be a general term meaning “grain;” I’m not familiar with Gaelic, so I will accept the translation as presented (for now). There are two critical things about this method: 1) if valid, this is probably the earliest written record of a full production method for ancient malt; and 2) the method describes the stuff being made into “cakes,” but never being mixed with herbs.

There is a possible route of cultural transmission from the Northern Irish to the Western Scots; there are shared linguistic and genetic roots between the two groups. It is conceivable that such processing methodology was passed from the ancient Irish to the people who would become the Scots. We know that the Norse later purchased “malt” from the Scots – so it’s conceivable that they were actually using “malt” that was produced in a manner similar to what is described here.

Given that it was turned into “cakes,” it is plausible that the stuff was still sour-leavened – the rest of my arguments regarding “malt” would hold true. The use of cakes for transportation and storage makes sense; loose grain requires a very solid piece of fabric for sacking material, which would be likely difficult to produce in the era. Cakes could be carried in the equivalent of netting, a more utilitarian form of container that is not as hard to produce. The cakes would probably still be dried in a manner not unlike the “gruit” cake, as the drying would help preserve the grain.

So I am now considering “mealt” and “gruit” to be completely distinct products with some similarities.

Hops: The Debate Rages

There are many impassioned arguments about the usage (or not) of hops in ancient beverages. I’m not going to spend a lot of time on this one, because that’s a much longer to discussion. To whit:

1) There is no physical evidence directly linking any plant product to any brewing activity in the Viking era – because we really don’t have complete products. All such connections are necessarily speculative, based on plants finds in proximity to sites.

In other words, there is no more evidence that any herb was used in Viking-era brewing than there is that hops were used in Viking-era brewing.

2) There is physical evidence that, among other plants, hops were present on Viking-era celebratory and burial sites.

3) There is documentary evidence that hops were being used in brewing as early as 822 CE (via decree of Abbot Adalhard of Corby).

4) There is physical evidence (the Graveny boat) that Vikings were importing hops.

5) Hops have a traditional use in herbal medicine – which I have already thoroughly connected to Viking-era brewing.

6) There is no evidence that hops were ever used exclusively in brewing in the Viking era.

My most reasonable conclusion I can draw: the above-mentioned “gruit” may have contained hops, but probably did not contain hops exclusively. It would have likely been present in a mixture of other locally-grown herbs.

This should piss some people off, I’m sure of it.


OK, OK, The Vats Weren’t Made of Sheep

I mean, it was fun speculation and all, but I’m pretty sure the vats were made of wood. And I’ve hit on a very specific vat.

Previously, I argued that “skap-ker” was a reference to a combination working vat/serving vat. The same vessel that was used to hold the fermenting beverage was also used as a display item. I argued that a particular brewing vessel type (the “Buddha bucket”) served as an example of a type of bucket that could fit such a purpose.

I’m starting to believe that there may have been multiple vessels involved in a multi-path brewing process. I also believe I have speculatively identified a bucket that could have actually served as a brewing vessel.

This bucket, figure 94 in the second volume of Osebergfundet vol. II, is very interesting. It’s made of fir, is roughly 5.5 gallons in capacity, is of a “wet” use type, and is fairly plain. It’s bound by 9 beech hoops which are further secured by iron tacks. It’s probable that the hoops were tacked down to ensure that the bucket doesn’t fall apart; “wet-use” buckets typically need to stay wet in order to stay together, as the swollen wood provides the needed tension. The tacks would allow the hoops to maintain their pressure even if the bucket dried out.

It is also inscribed with runes of ownership – a rare thing in Viking-era finds, and a culturally significant phenomenon.  Runes were often inscribed on items that were of extreme importance and significance to the owner – items without which they simply could not be. A warrior might, for example, inscribe ownership runes on his sword.

The runes translate to “Sigrid owns [me].” Sigrid is a female name, and she has inscribed runes on this relatively plain wooden bucket (other Oseberg finds are far more elaborate than this). No other bucket is so marked.

So we have a “wet” bucket, with additional securing measures, inscribed with runes of ownership by a woman. The runes and extra securing measures seem to indicate that the bucket is of extreme importance to the owner. That it’s owned by a woman indicates that it’s for a job that women historically performed. That it is of a “wet” type further tells us that the type of work involves liquid.

My conclusion: this is the working-vat for brewing. The wood may contain a native yeast culture, which would serve to explain the ownership runes – that particular bucket was essential for Sigrid’s work, because it probably contained her personalized yeast strain. It’s also conceivable that it was a bucket for medicinal preparation – but then again, medicine and brewing are already tightly connected.

A Method Revised

From this, I conclude a two-part brewing system: one involves the production of “ealu” or “brauð” using “gruit” in a specialized vessel, and the second part involves mixing the first product with a quantity of sweet liquid to make “öl.” The sweet liquid could be honey-water (“hydromel,” glossed as I’ve discussed before), or could possibly be a liquor derived from “mealt” (which I now understand to be distinct from “gruit”).

The ingredients for “gruit” are probably still the same as I’ve described before: grain, oil seeds, herbs, and saltwater. This essentially serves as a yeast culture medium; the grains provide nutrients, while the oil seeds provide a mild antimicrobial effect. The herbs would provide flavor and whatever medicinal properties they may have. When dumped into the designated bucket, the liquid becomes a medium for yeast growth as the trapped yeast cells now have a food source. Fermentation proceeds.

The result of that fermentation could also be combined with other sugars as I mention above. Anglo-Saxon writings advise pregnant women to avoid consuming beór, and the Talmud’s “zeethos” carries a similar warning. It’s worth noting that both flax seeds and safflower seeds have been associated with miscarriages and spontaneous abortions; it seems reasonable that both beór and zeethos shared oil seeds as a significant ingredient. This helps corroborate the use of “gruit” as a precursor ingredient.

“Gruit” is also used to indicate the “dregs” of brewing in the Anglo-Saxon leechdoms, again seeming to indicate that it is fermented prior to being used to make beór.

Such a technique is still in use today – many modern breweries will make a “small” beer, and then use the leftover yeast to start a “big” beer. Running through an easy fermentation ensures that you have a healthy, thriving yeast population; pitching healthy thriving yeast ensures a rapid and clean fermentation. For many high-gravity beers, pitching actively growing yeast is the only way to get them to start – and given that “beór” is glossed with “hydromel,” I’m willing to bet that it was a high-gravity brew.

So how am I making this stuff now?

This tasted like beef gravy. Have you ever fermented beef gravy? Here’s a hint: DON’T.

Somewhat interestingly. The above-pictured experiment is from a modification of my original mucking about; here, I omitted the flax seeds (because I had them packed up – but as I indicate above, they were probably always in use) and used peated malt as my base grain. Finds of corn-drying kilns from Scotland indicate a variety of woods as the possible fuel sources, as well as peat. As you get farther north in Scotland, peat becomes more prevalent; this is most likely the type of fuel that would have fired a kiln in the Orkneys in the Viking era.

So I took peated malt, herbs, and salt, and made them into sourdough biscuits as I’ve done before. I kept the same ratio as in my initial experiment: 4 biscuits (~1 cup crushed) per quart of water. Heat slowly to just under a boil. Mix with honey, and ferment. This time around, I also added some fruit to the mix – plums, as they’re found all over the place in the Viking era (as are crab apples and polar berries – other good candidates for additives).

It took me a while to place the flavor that resulted, but after a friend of mine tried it and nearly vomited in disgust, the answer was found:

Hot dogs.

The formula  of 1 part biscuit, 1 part honey, 3 parts water, and brewer’s yeast produces something that tastes exactly like the water leftover from boiling hot dogs, albeit highly concentrated.

Concentrated hot dog water. With booze.

It was not good.

This concerned me, because I really didn’t want to make something terrible. It also didn’t really taste sweet at all, so I figured that was a good enough excuse to say “NOPE! BACK TO THE DRAWING BOARD!”

A normal, rational person would make hot dog water ONCE and then learn from it. But I’m a scientist: I have to do this at least three times before I can accept that it’s a bad idea.

So, obviously, I have to add back in the flax seeds (since they’re probably a very major component of gruit) – because if there’s anything that can improve hot dog water, it’s flax seed. This time, however, I’m altering the ratios.

Given that I’m looking at a two-stage fermentation in two different vessels, it stands to reason that different mixing steps might apply. The 1:4 fermentable:water ratio is reflected all throughout brewing history; I’ve decided to stick with that quite firmly, and apply it twice.

So the glop in the above-pictured jar is 1:4 biscuit/flax:water. 1 cup of a 50/50 mix crushed peated malt biscuit and flax seeds, and 4 cups of water. Heated slowly as I’ve done before. The whole mess is getting fermented by some dry Nottingham yeast.

Whatever liquid results from this will be mixed in a 1:4 ratio into some honey water which itself was prepared in a 1:4 ratio. So if I have 500 ml of fermented hot dog/flax water, I’ll add that to 2 L of honey water (with some fruit thrown in because hey it can’t get worse) which is made from 400 mL of honey and 1.60 L of water. That hot dog water is derived from ~250 mL of biscuit added to 1 L of water.

That should severely reduce the impact of the peated malt (taking it from 20% of the contents to something like 5%, which is far more normal for peated malt use) and bring this much more into the realm of a Scotch ale style. I’m considering using a strong ale base instead of honey water – that’d make it much more like a wee heavy, and given the prospective path of true malt production, it’s much more feasible than before.

Experiments are afoot! I’ll report back when I have findings, but for now: I’m going to stop all this typing and get back to reading.