Collections: Bread, How Did they Make it? Part III: Actually Farming

As the third part (I, II, III, IV, A) of our look at the basic structure of food production in the pre-modern world (particularly farming grain to make bread) we’re going to finally look at how one actually farms grain to make bread. Now that we have all of our farmers in place, both the big ones and the little ones, we can get down to the earth and the mechanics of making some food. As a result, this part is going to be a bit more descriptive and a bit less analytical.

This is not going to be a how-to guide (farming is complicated), but I want to give a sense of what our cereal farmers are doing at each stage of the year, the labor and tool demands of those tasks, and how they impact farming overall, along with the major processing stages (threshing, milling, baking) and the sorts of people who were involved with them. I should note that our look at processing especially is going to be much more closely connected with the ancient and medieval Mediterranean, as a function of where I feel like I have a decent grasp on the structure of milling and baking as industries.

As always, if you like what you are reading here, please share it. If you want updates whenever a new post appears, you can click the button below for email updates or follow me on twitter (@BretDevereaux) for updates as to new posts, as well as my occasional ancient history, foreign policy or pop-culture thoughts. And if you want to ensure that I can continue to buy bread (and/or a farm) you can support me on Patreon.

((Also as a reminder, the selected bibliography for all of this is back in the first post.)

Farming in Season

Since we have our farmers in place, let’s look at the activities of a single farming season in sequence. Again, just to point out the necessary caveats,: different crops have different steps (these are the steps for wheat and barley), but also this is just a general overview and I am sure there are any number of important tasks that a farmer will spot that I’ve missed.

Now as I noted in the first post, the planting and harvest times move depending on if winter wheat was used (planted in autumn, harvested in early summer) or spring wheat (planted in spring, harvested in fall). These are actually different strains of wheat (triticum aestivum). Local soil conditions, especially seasonal moisture levels, are going to determine which variety of wheat is used. Barley likewise can be a summer crop (especially in temperate areas) or a winter crop (in very warm areas; barley tolerates salty soil and dry conditions, but does poorly in cold winters). I am going to follow the timings, roughly, for winter wheat (because it was the most common in the ancient Mediterranean, where my expertise lies). When a spring wheat or barley is used, it is sown in the spring, with weeding in the summer, and with reaping, threshing and winnowing in the fall, so if you like, you can follow along, converting over to spring wheat timings.

Via Wikipedia, farmers using a plow on an Akkadian cylinder seal (seal to the left, imprint of the seal to the right), c. 2200 BC, now in the Louvre.

But with winter wheat, that means our cycle begins late in the summer, getting ready for the planting season. We have already pulled in and processed the previous harvest (in early summer), had our harvest festivals, banqueted our neighbors and are now preparing for planting. The first step is plowing. The plow – a wood or metal board pushed into the earth – carves grooves in the soil (making those distinctive little ridges in farmland) and turning up the lowest soil and putting it on top. Plow design is actually something we can observe archaeologically; plows tended to improve over time, particularly their ability to deal with heavy, dense soils (though taking a heavy, deep-soil plow into lighter shallow soil is perilous, as various farmers have found at points in history); I won’t get into the details here, but Wikipedia actually has a pretty capable list of forms. That said, while plow technology did steadily improve (only occasionally fast enough for our sources to notice, e.g. Plin. Nat. Hist. 18.172), the average peasant farmer is going to use the plow that is common in their region, which may or may not be the best for the task from a global or diachronic inventory of plow-types. Technological change for subsistence farmers was slow, so even unsuitable plows might last in regions for decades or centuries before a more suitable design was developed or imported. Whatever design you had, you were largely stuck with it, at least in a time-frame measured by decades instead of generations or centuries.

(An aside: being American, I am using the American spelling ‘plow.’ The British spelling, which you may see, is ‘plough.’ As is common, the British spelling has no particular claim to antiquity – the Old English is plōh, which in turn gives the Middle English word, spelled alternately as ‘plow,’ ‘plouh’ and ‘plough,’ among other variations. British ‘ou’ for ‘o’ spellings are frequently either Latin or Old English words originally spelled with just the ‘o’ (e.g. Latin, honor, color and here Old English plōh), which acquired their ‘u’ with the admixture of French. Not that there is anything special about Americanisms; merely to note that just because a usage is British does not make it old or original, as far as I can tell in any case the American word is about as likely to be the original older spelling as the British one)

A field was generally plowed more than once in order to get an ideal soil situation. The Romans seem to have assumed a normal field would be plowed three times before planting (with denser soils potentially requiring more runs): once to break up the soil (which will have spent the growing season compressing and hardening; you want it light and loose), then a run ‘cross-plowing’ (plow-lines perpendicular to the first effort), before the third plowing on the original direction (‘ridging’) leaving the distinctive farming ridges in the dirt. In some cases, a different plow might be used for the steps, with the first step (or first two) carried out with an ard-plow (sometimes called a ‘breaking’ plow), which has no mouldbord and so merely cuts the earth but does not invert it and the last step carried out by a mouldboard plow, which has a mouldboard designed to basically ‘flip’ the soil as it moves, exposing the clods of dirt in the lower layers to the air. The plowings could be spaced out; the advice for a fallowed field (one where there was no crop growing in the previous year) was to plow once each in spring, summer and autumn. The schedule might be compressed for rich land continually cropped.

Via Wikipedia, a modern reconstruction of a mouldboard plow (the mouldboard is the large wooden element at the base); the mouldboard serves to up-turn the soil as it moves.

Now as we’ve noted, the very poorest of farmers are not going to have regular access to a plow-team. While they may have used human power to move a plow, the more common expedient here was hoe-farming using hoes, digging sticks or small ‘hard ards.’ That was slow, inefficient and backbreaking work, so getting access to a plow-team was important if at all possible. Even for an animal-drawn plow (with a plow-team, normally oxen) there’s quite a lot of muscle work to be done to keep the furrows (the grooves left by the plow) straight – the straighter and more even the better – and to keep the plow down in the earth (pushing down on it may be necessary). Remember: you are not only moving the weight of the plow, but the weight of all of the earth it is turning up. This is one of the advantages of an animal plow teamoxen, being big, strong things can pull a much deeper plow. Of course a lot of land needed to get plowed and in a relatively constrained time window; as you might imagine, the fellows who might own the plow teams would thus be in a position to make sure their lands were plowed during the most favorable time windows.

Plowing does a number of important things to enable farming. The thing that everyone generally has in mind is that it upturns the earth, bringing deeper soil up; since crops exhaust the nutrients in the soil, plowing ‘shuffles the deck.’ But it also crucially cuts through the root-systems of weeds and disrupts the life-cycle of various pests (both by disrupting larva or eggs, but also by killing the plants those pests might feed on). Removing the weeds by plowing was especially important in dry climates because otherwise the roots of those weeds would bring moisture out of the ground into the leaves of the weeds where it would evaporate; killing the weeds at the roots conserved the moisture for the crop. Finally, plowing served to loosen up the dirt to allow the seeds to germinate better.

Planting and the Growing Season

Things get busy as we get to the sowing season. If the field is going to be treated with manure, that gets laid down before the final plowing (which would come around September for an autumn-planting/summer harvest), with seeds being sown with that last plowing. Manuring before the final plowing (with a mouldboard plow that upturns the soil) ensures that the nutrient-rich manure ends up buried and mixed into the soil; if left on the surface the manure will dry out and decompose, wasting the nutrient-value.

From the ‘Tomb of Sennedjem” in Deir el-Medina, Egypt, a mural showing a farmer plowing his field with a pair of oxen as a plow-team, with his wife sowing behind him, dating to c. 1200 BC (New Kingdom). Apologies for poor quality of the image, getting a decent picture that is of the painting itself and not a reproduction is difficult; this one is from wikipedia.
Here we can see sowing by the broadcast method, done by the woman on the left. Planting, as one of the peak periods of labor demand, seems to have brought women into the fields even in societies were agricultural labor was normally gendered male.
This is also, I should note, an ard-plow as it lacks a mouldboard.

There are a number of ways to sow seeds, which vary by the available technology and ground conditions. Ideally, the farmer wants all of the seeds to be evenly and fairly densely placed, but the raw number of plants that need to be planted typically prohibits hand-placement of individual seeds. Generally the best method is for seeds to be dropped into shallow holes and then covered, called ‘seed drilling‘ – basic seed drills existed in the Bronze Age Near East (c. 1400 in Babylon) and in second century China, but never reached Europe or apparently the broader Mediterranean; the Romans seem not to know of them. Instead, seed-drilling in Europe only arrives in the early modern period. Seed-drilling allows for fairly regular lines of plants and optimal seed-depth, but wasn’t technologically an option in much of the world for most of the pre-modern period.

In places where seed-drilling devices weren’t available, seeds were sown by the broadcast method. The ground was plowed, then the seeds were thrown out over the ground (literally cast broadly; this is where our term broadcast comes from); the ridges created by plowing would cause most of the seeds to fall into the grooves (called furrows; thus a ‘furrowed’ brow being one scrunched up to create ridges and depressions that looked like a plowed field), creating very rough rows of crops once those seeds sprouted. Then the land is then harrowed (where our sense of ‘harrowing‘ comes from – seriously, so much English idiomatic expressions are farming idioms, for obvious reasons), typically with rakes and hoes to bury the seeds by flattening out the ridges (but not generally entirely erasing them) in order to cover the seeds over once they had been placed with very loose clods of earth. Farming was a skill and our sources (e.g. Plin. Nat. Hist. 18.197) are clear that a farmer is likely to have a very refined, very intentional practiced motion for broadcasting the seed, both to manage the distribution of the seeds in each cast but also the rate of sowing to ensure that the seed is evenly distributed over the whole field.

The labor demands of the plowing and planting process were pretty high (but of course varied with tool, plow and animal availability). Columella, a Roman agricultural writer, estimates roughly 4 labor days per iugerum (c. 6 days per acre) to plow and sow, compressed into the period between the September equinox (22nd or 23rd) and early December. The schedule was more demanding when we remember that weather (heavy rains especially) might ruin multiple days. After the seed was in the ground, the labor demands decrease quite a lot, but do not go away; Columella gives five days of labor per iugerum for post-sowing, pre-harvest tasks (harrowing, hoeing, weeding). Now it is important to stress that the labor demands of farming are conditioned by the need to grow plants in bulk; this may seem a strange thing to note but for many modern green-thumbs, there’s a bit of a fundamental disconnect on how different it is to grow a few dozen or even few hundred plants compared to a farmer who might be (using Columella’s figures; modern figures differ but not by nearly as much as you’d expect) sowing more than one-and-a-quarter million seeds per acre (which may neatly explain why individual seed placement is impractical!). Tending to each plant individually was simply not an option.

Via Wikipedia, illustration from a Flemish Book of Hours (early 16th century), now Bayerische StaatsBibliothek Clm 23638 showing the labors of September. In the foreground, two horses draw a harrow. In the middle right, a man sows seed using the broadcast method, while at the top left another man drives a plow, showing all three stages of the sowing process.

What is happening in the growing period? In dry climates (including most of the Mediterranean), hoeing is an important task which breaks up the upper-layer of soil above the crop, creating a layer of loose, dry earth which doesn’t conduct heat or water well and thus shields the seeds from the drying heat of the spring (keeping in the moisture it needs to grow most effectively). This had to be done with care as hoeing was best done as the tops of the crops were nearing the tops of the ridges; care had to be taken not to disturb the crop’s root systems or damage the young stems. Practices varied, but Columella suggests that wheat was generally hoed twice, but barley only once (one of many ways in which barley was less labor intensive; he also assigns it fewer days for plowing and weeding; barley was just a more tolerant crop. This was offset by it being both less tasty but also less nutritious, something that even the Romans were already aware of). This process is also going to (hopefully) disrupt the growing of weeds.

Weeding in earnest begins in the spring. Here ridging bears more advantages: while the crop sits (hopefully) at the base of the furrows, the weeds will often grow atop the ridges, where they can be easily removed. Weeding was a continual task and done by hand, removing the offending plants from the crop one by one. Weeds were a real problem for the farmer: they soak up moisture and nutrients which would otherwise go to the crop, so their removal is an important task.

I’ve given the timetable so far for a Mediterranean winter-wheat planting from autumn to summer (our harvest is just around the corner). The basic steps in North Chinese wheat agriculture were the same, but the seasonal timing was different using a spring -wheat, with plowing and planting in spring, weeding in summer, harvest in fall and storage over the winter. Which at last gets us to:

Harvest Time!

While the harvest required less total labor than plowing, it took place over a much narrower time-frame, about a month and a half for the entire harvest process (including threshing) beginning in June or July depending on the climate. This was in many ways the most vulnerable moment of the process: the grain was ripe in the fields, easily spoiled by pests, knocked down by windstorms or heavy rain, pillaged by armies and so on. Until all of that ripe, delicious wheat (and ripe, bland barley) was reaped, threshed, winnowed and stored all of those dangers remained. A quick harvest was thus very much desirable.

Reaping – the cutting down of the wheat and barley plants – was typically done with a sickle or scythe, though as I understand it the sickle was more common for grains, the scythe being more usually used on hay-grasses. There’s quite a bit of regional and temporal variation in technologies; K.D. White (Roman Farming (1970)) notes that the 19th century English scythe-using farmer could reap twice as fast as his first century, sickle-bearing Roman equivalent (with the rate of the latter supplied by Columella, though it must be noted that Columella is universally pessimistic in his statistics for wheat as he wants to convince you to farm grapes instead).

Via the British Museum, a print from a c. 1580 French woodcut series showing the months. This month – August – is marked by the harvest. On the left you can see men working in the fields cutting down grain and bundling it. A woman, with her skirts gathered up, hauls the grain to the threshing room (center right). As with the planting, women often worked in the fields to get the harvest in on time, even in societies where they generally did not do agricultural labor.
In the back right, you can see a man inside the building threshing grain with a hinged threshing flail.

Interestingly, exactly how the reaping was done differed by area due to local conditions. Varro has this fascinating note that there were, in his day (first century BC) three methods in Italy. In Northern Italy, the stalks were cut right under the ears (the seed-bearing part) of the plant, and the rest of the plant left in the field to dry and only being cut later for use as thatch (since fodder and grazing was easy to come by in the region). Meanwhile in Southern Italy, where grazing and fodder stocks for animals were far more scarce, grains were cut close to the base, then cut again under the ears once they were down with the stalks being bundled in the field for use as fodder for animals. In Central Italy, a hybrid of these two methods was practiced. It’s pretty clear in this case the question was fodder availability, but I bring up this example just to note how much local regional variation there could be, even just within Italy, as farmers respond to local conditions. Interactions with animals can be especially complex – for instance in parts of the Roman world, it was suggested that animals should actually be grazed in wheat fields very early in the growing season, to prune down the wheat stalks, which causes them to ‘bush out’ at the base. All of which is to say, consult your neighborhood primary source for local variation before making assumptions about local systems!

There would also be stubble in the field – the roots and bits of plant below the cut of the sickle. In wetter climates, that stubble will keep growing a bit and can provide useful pasture for animals to nibble on, but in dry climates the stubble, exposed to the heat of summer (as in the Mediterranean) parches rapidly and dies. It seems that it was common in the Mediterranean to actually burn the stubble, in order to return its nutrients to the ground, but it might also be plowed under to similar effect.

Once the stalks are cut down and the ears separated, you have the straw (used, again, as fodder or thatch and thus here exiting our story) and the ears of wheat and barley themselves. These are full of seeds (the thing you want) but also a bunch of plant matter you do not want. Which brings us to the first processing step: threshing and winnowing, which I’m including here because it was typically done on the farm during the harvest (often literally simultaneously, with part of the household reaping the crop and another part of the household threshing the wheat as it came in).

Via the British Museum, a print of a Flemish woodcut (1565) showing almost every stage agricultural work in order from back to front (and one idler that this woodcut is quite disapproving of). Halfway up the frame in the center, two individuals drive a team of horses pulling a wheeled plow (a rather late agricultural development). In the next closest field to the far left, a man sows seed using the broadcast method. Still closer in the mid-ground center, three men and one woman are threshing the wheat using threshing flails (notice again how it is in this stage that women are shown working in the fields). At the very foreground right, a man is winnowing wheat with a winnowing basket; the chaff is being collected, presumably to be fed to animals.

Threshing comes first: this is the process of getting the edible grain out from the inedible parts of the plant, by a number of different processes that, to the untrained eye, appear to consist of simply beating the everliving hell out of the grain until it gives up and begs for mercy. There is, of course, more to it than this. Methods vary. For wheat and barley, the most common pre-modern method was threshing with flails, either unjointed (which is to say, a club) or jointed flails: the cut grain was put on the threshing floor and beaten with the flails which knocked the grain and much of the chaff (the dry, protective casing the grain-seed itself sits in) loose. The actual grain, being small and dense, falls to the base of the threshing floor; once all of the grains are free, the straw may be gently raked away, leaving just the grains and some of the chaff (much of the chaff may blow away if there is a breeze as it is very light). On larger farms, the same method might be achieved by spreading the material on the threshing floor and having it either trampled by farm animals (especially oxen) to provide the same percussive effect as the flails, or having a ‘threshing-sledge’ (a heavy wooden board with flint teeth on the underside) drawn over it by animals. Of these, as best I can tell, threshing with flails seems to have been the most common, at least in Europe and the Mediterranean where my knowledge is best.

(As an aside: truly manual threshing, where the stalks (complete with their seeds) are bundled together and beat over a surface to knock the seeds loose doesn’t seem to my knowledge to have been common with wheat or barley, but does appear to have been one of the standard methods with rice. At the same time, as far as I can tell, threshing flails were not generally used with rice. Threshing by trampling was used with both, however.)

Threshing was a long, tedious, labor-intensive process which would begin often even as the harvest was coming in and continue potentially even into the winter for families that were short on labor. Consequently, this is the moment we are mostly likely to see marginal laborers – children and in some cultures women – employed directly in agricultural labor. Even in places (like Greece and Rome) where we rarely see women in the fields (we are mostly relying on representational evidence to assess this, but textual evidence also often suggests that women engaging directly in most agricultural labor was seen as strange and a sign of severe financial distress) they frequently figure in harvest and threshing. This was an all-hands-on-deck part of the year, though – this is my purely subjective sense of the sources – in a good year it tends to be reckoned as a busy but happy time, since the period directly after the harvest generally held the year’s biggest festivals (the part where you banquet your neighbors). At the same time, we can easily imagine, in the event that the harvest has come in under expectations, the steady dread of the adults in farming families who are likely already calculating in their heads the steps they will need to take to survive until the next harvest.

But we’re not quite done with the harvest. After threshing, the grain we have is still has all sorts of things which are not grain seeds in it – potentially including pests, lots of bits of chaff, rocks off of the threshing floor and so on. That all has to be removed from the actually edible grains. One of the main worries is the grain weevil (genus curculio, after the Latin name for the bug) which might spoil the grain if not removed.

Via the British Museum, a mid-19th century painting from India showing the entire harvest process. In the background, three men are reaping the grain. At the front left, a man drives oxen over the wheat for threshing, while in the center front another man is using a winnowing basket (or ‘winnowing fan’) to winnow the grain.

So we move on to winnowing. This can be done in two main ways, before the advent of modern machinery (winnowing machines appear in China as early as the 14th century but not in Europe until the early 18th). In the first method a winnowing basket (sometimes called a winnowing ‘fan’) is used: the threshed grain is placed in the basket, which is then shook (often raised overhead). The chaff and other not-yummy-grain-seed-things are all less dense than the grain seeds and so rise to the top of the pile as it is shaken; the basket is tilted slightly so that in rising, the chaff falls over the edge and out of the basket as it is shaken, eventually leaving just the grain on the bottom. Alternately, a ‘winnowing shovel’ can be used to throw the threshed grains up into the air during a light breeze; the straw and chaff and other impurities, being light, will be blown clear, while the heavy grains fall straight back down and form a pile at the winnower’s feet. As you’ll note, both of these processes require the grain seeds to already have been knocked loose from the chaff and straw (so that the only problem is that they are intermixed), which is why you have to both thresh and winnow, in that order. In some cases, winnowed grain might then also be passed through a sieve, particularly if it was intended for long storage, to try to weed out any lurking pests.

An interesting note here: intentional selective breeding of all kinds of staple cereal crops happens very early (we can tell because of how dramatically the strains of farmed wheat and barley change compared to their wild ancestors), but we actually have references to it in the Greek and Roman agricultural tradition: it is suggested that the heaviest grains, which will make their way to the bottom in the winnowing process, be selected as the seed grains for the following year.

Sale and Storage

What all of that process leaves us with is a pile – ideally a very large pile – of grain seeds, which are what contains all of the useful (for humans who can’t break down cellulose) nutritional value. I have always found it striking that for all of the plant in those vast fields of wheat, it is only the small batch of seeds up at the top that we actually eat.

Via Wikipedia, cereal grain seeds. Clockwise from the top-left, they are wheat, spelt, oats and barley. This is the end objective of all of that farming work.

Now we’re going to deal with (very broadly) extraction and markets next week, but I want to note that this is generally the first point in the process where we would see them. Unsurprisingly, tax collectors and merchants are uninterested in buying all of the straw and chaff: they just want the edible grains. In the entire ground-to-bread chain, this is also the best stage for long-term storage and long-distance transport: basically all of the extraneous material has been removed, but the seed is still in its little shell, which improves storage life. Now we’ll get into all of the elements of getting this stuff to non-farmers next week; I just wanted to note that this is the stage, if you are going to sell or be taxed in grain, that this would happen. This is also the point where grain is moved to the cities – milling and baking is generally done relatively close to the point of consumption. So almost any kind of trade in cereals is trade happening in threshed and winnowed (but not milled or baked) grains.

But whether the grain is now in the hands of the farmer or the merchant (or the tax-man) it is going to have to be stored in a granary. Granary design varies wildly from place to place based on local conditions and building materials, but the basic demands are constant: grain needs to be kept moderately cool, out of direct sunlight and mostly importantly very dry in order to discourage mold and pests from spoiling it. Complicating this problem, the grain is going to be bringing a fair bit of moisture with it. Thus, perhaps somewhat counter-intuitively, grain can’t just be sealed up inside solid containers because without air-circulation, the moisture already in the grain will lead to the growth of mold (which will in turn, due to its biological processes, heat the grain which leads to fermentation, which in this case is not desirable). Air-flow is thus essential, but has to be balanced with the avoidance of direct sunlight and keeping out pests. Thus, while grain was sometimes stored in ceramic vessels (amphora, pithoi, pots) it was (and is) most commonly kept in bags or sacks.

Via Wikipedia a Greek box (c. 850 BC) made in the shape of a set of granaries (five, top), now in the Ancient Agora Museum in Athens. This design of granary – the ‘beehive’ shape – was quite common in the Near East. Notice how each granary is suspended above ground (see the gaps in the ‘feet’ of the granary such that they don’t sit flush to the ground). They all also have the opening at the top to allow airflow to avoid retaining moisture.

Consequently, some of the most common ways to build granaries is for them to be suspended slightly up off of the ground so that cooler, drier air can enter from below. A hole or window near the top of the granary allows warm air to escape; because warm air carries more moisture leaving than cold air entering (warm air can dissolve more moisture into it than cold air can), this keeps the moisture level low. This is why many traditional granaries are tall rather than wide, even in areas where space is not at a premium – a taller, thinner structure maximizes this air movement. Elevating the base of the granary also serves to discourage animals and pests from getting at the grain. While modern granaries are sometimes built on large concrete foundations to get them up off of the ground, pre-modern designs tend to use some sort of stilts.

Within that basic scheme there is a lot of variation. Roman agricultural (writers suggest the use of plaster, amurca and cement to carefully seal the granary against pests, along with high vaulted ceilings to enable the necessary air movement (with high windows facing north, presumably to avoid letting in direct sunlight). Wooden granaries elevated on ‘staddle stones’ were common in Britain, but same design features working to capture the same advantage of an elevated granary shows up in a wide variety of places, for instance the rangkiang elevated rice barns of Indonesia.

Via Wikipedia, two rangkiang rice granaries from Sumatra, Indonesia (c. 1895). Note that the granary structure itself is elevated off of the ground and how the granaries are tall, narrow with opening at the top. The cereal crop here is rice, but the basic concerns about temperature and moisture lead to similar granary construction.


Grains still require processing in order to be eaten. Raw unmilled barley and wheat can be made edible by roasting, or better yet turned in to porridge by boiling the grains in equal parts grain and water for roughly an hour or so. For folks without the ability to mill grain into flour – particularly ancient war parties or armies on the move – that was enough to render the grains edible, at least. But turning grain into bread has nutritional advantages, breaking down some of the harder to digest plant compounds and cell-walls, thus allowing the human body to derive more nutrition from the bread itself than it would from the grains. Consequently, more logistically sophisticated ancient and pre-modern armies often brought portable milling and baking equipment with them so that they could readily process captured grain (which is going to be stored, as noted above, as grain seeds) first into flour and then into delicious, nutritious bread.

Milling is a mechanical process by which the grains are crushed down to a powder, called flour. As with farming above, milling is one of those processes that seems simple and easy until you remember the scale it has to be done at. As you may recall from the first part, a single five-person family might have consumed around 1,750 liters of grain (by volume, of course) which might in turn mean something on the order of twenty-million seeds which need to be crushed and ground into powder. Obviously, this needs to be done en masse. A simple hand mortar-and-pestle isn’t going to get the job done. Instead, much larger stone grinders are used to pulverize the grains into flour.

Via Wikipedia, a Neolithic (3700-3500 BC) saddle quern from England. The grains would be placed in the depression and then crushed by pushing the stone above down and then moving it back and forth against the base.

The earliest method used to do this seems to have been the saddle quern (terminology note: the word ‘quern’ can mean both an entire small grinding assembly or merely the bottom, non-moving element); a saddle quern consists of a larger, elongated bowl-shaped rock (so that there is a depression to place the grains in) and a second stone shaped to fit that depression which can then be pushed up and down the long depression in the quern-stone to grind the grain underneath it.

By the iron age, we see the appearance of rotary querns (Wikipedia actually has a wonderful video of one of these, a Nepalese rotary quern, in use). Also made of two stones, the lower of which (the stationary quern-stone) does not move. The second stone, which typically fits into a pivot on the quern-stone is called the muller, millstone or (occasionally) grindstone and is typically quite heavy. The lower stone is usually ever so slightly concave and the upper stone ever so slightly convex. Typically the a rough-surfaced, somewhat porous kind of stone (like basalt) was used in order to aid the grinding action. The grain is put between the two stones (often through a conical hopper in the center of the grindstone) and the millstone is rotated (usually by means of a hand-grip lever). The rotary motion, combined with the weight of the grindstone, is what grinds the grain down to flour – the big advantage over the saddle quern is that whereas the crushing force in the saddle quern is supplied almost entirely by the operator, in the rotary quern, much of that energy comes from the weight of the millstone pressing down and the operator merely needs to provide rotary motion to agitate the milling action. Moreover, a rotary quern may be operated continuously; the millstone is shaped so that the ground flour is pushed out at the edges, while unmilled grain can be continually fed in through the top.

Via Wikipedia, on the left you can see a ‘beehive’ rotary hand-quern. The hole through the center of the upper stone (through which they are running a wooden rod to support it) would be where grains would be passed through to enter the grinding action.

It is also far, far easier to scale up the size of a rotary quern, because the only limit to the grinding area (dictated by the radius of the quern-stone and millstone) is the ability to supply energy to turn the increasingly massive and heavy stones against each other. Initially this was still muscle power: by attaching longer beams to the grindstone (which in these larger designs is generally a heavy circular disk), many humans or animals could push the grindstone around at once. The underside of the millstone for these larger mills was ‘dressed’ by carving furrows into it, typically arranged so that ground flour is pushed to the outside of the mill. The actual operation of a mill is, I should note, more complex than what I am describing; the miller has to control the separation of the stones and the rate at which grain is fed into the mill. But we’re here for the basics.

Via Wikipedia, a larger millstone being dressed with grooves to assist in grinding the grain and shifting the resulting ground flour to the edge of the millstone.

The next major change in milling technology came with the application of a power source other than human or animal power. Water mills first become widespread in Europe during the Roman Imperial period and Vitruvius, a Roman architect and military engineer, describes their operation to us. The technical challenge of using water efficiently was in a gearing system to allow a vertical water-wheel to power a horizontal millstone. Windmills first appear in Central Asia in the 9th century and spread out from there.

(As an aside, you may ask: wouldn’t this process result in small bits of the millstone being ground down by the grinding process, breaking off in tiny bits and ending up in the flour? And the answer is yes! One of the key ‘quality’ markers of bread was the likelihood of you finding a bit of mill in it when you bit in.)


Now what the mill’s grinding has produced is flour, which is still not really directly edible (although you can also make porridge out of it). To turn that into bread, we need to take that flour, add water (and normally yeast, if we want the bread to rise) and then expose it to some heat. Early baking probably used airborne yeast spores, with dough leavening naturally when left out. Beer production might also provide yeast (by skimming the foam produced) for bread-making, which seems to be how yeast for baking was acquired in Egypt in the Bronze Age (to the point that a bakery and a brewery were seen as linked industries) and according to the Roman Pliny the Elder, also what was done in pre-Roman and Roman Spain and Gaul. In much of the rest of the Mediterranean, a paste of flour or wheat-bran mixed with wine or grape juice in the process of fermentation might be used as a source for yeast. Naturally, you can also use dough that already has yeast in it as a starter, as with modern sourdough.

Via Wikipedia, a model of a bakery and brewery (c. 2000 BC); the brewery is on the left, the bakery to the right. In the bakery we can see the dough being kneaded in the top and baked in the center, with one man carrying a round bread-loaf in the middle.

Baking could, of course, be done in the home on the family hearth. The dough was kneaded and then could be cooked in an oven or exposed to the embers of a fire. The exact style of ovens or indeed the shape of the loaves of bread varied from place to place and period to period. The first ovens of the style we would easily recognize – fire below, doors in the front, pre-heated – seem to be Greek (M. Toussaint-Samat, A History of Food (2009), 202). We see molds used to shape bread as early as the 25th century BCE in Egypt, but the most common bread-shape in antiquity is still the round, fairly flat loaf (like the French boule) one gets when the dough is kneaded into a ball before being baked. This was a shape so ubiquitous and easily recognizable that it often appears as a simple symbol for bread more commonly (the way a modern bread-slice is sometimes used) – for example making a circle with the thumb and fingers was the sign of bread “because bread is customarily round” in Cluniac monasteries in the 11th century.

In the ancient and medieval Mediterranean, household bread production was normally a strongly gendered activity, with kneading and baking being done primarily by the women of the household (whereas, in contrast, commercial baking was generally done by men – a trend we will see repeat in industry after industry where the household version was gendered female but the commercial version was gendered male and often of higher social status).

Mass-scale commercial baking is, unsurprisingly, connected with intensifying urbanism. Bakeries in major cities, like Athens or Rome, could be large commercial operations, selling food to the general population (although not generally to the elite; a Roman large landholder likely had his own bakery and mill on his vast estate – buying food in the market was sometimes called out as shameful for such men, since it indicated a lack of confidence in their own estate’s produce and an inappropriate reliance on markets. More on that next week).

Via Wikipedia, a commercial bakery illustrated in an early 15th century Italian manuscript of the Taqwin as-Sihha, known in Europe as the Tacinum Sanitatis, an 11th century Arab medical treatise on health which had several Latin translations and was quite popular in the later medieval Europe.

Millers and Bakers

Now I want to pause for a moment here because we’ve introduced a whole bunch of non-farmers in our millers, mill-workers (you’ll see why I separate them in a moment) and bakers. And I think it is worth talking about how they fit into these societies generally. My focus here is going to be fairly narrowly in ancient and medieval Europe, because that’s where I know the evidence best, but my sense is that the same patterns replicate more generally.

Let’s start at the bottom with the mill-workers, by which I mean the fellows who – especially when mills are still muscle powered – are supplying the muscle power. The social status of mill workers was often very low. In the Ancient Mediterranean, when most mill-power was still supplied by muscle power, being ‘sent to the mills’ to work was a severe punishment, often meted out to the very poor or enslaved persons. More than one Latin student has been profoundly confused by enslaved persons in Roman literature (especially Roman comedy, with heavily features enslaved protagonists) being threatened by being sent in pistrinum, which gets translated as “to the bakery” by the hapless student (Roman milling and bakery was often done in the same facility, so ‘bakery’ is often the first translation a dictionary offers for ‘pistrinum‘), when what is meant is “to the gristmill” to work turning the millstone. Such labor was brutal and backbreaking; moreover the continual inhalation of pulverized grain dust damages the lungs. Being thus sent in pistrinum was essentially a deferred death sentence, much like being sent to work in mining. Consequently, in large, muscle-powered mills, the fellows actually turning the millstone were likely to be captives, criminals or enslaved persons (often being punished for some transgression), or else very desperately poor; in any of these cases they probably had sadly short lives. Animals seem to have been more frequently used than enslaved persons, however – particularly donkeys, to the point that the word molarius in Latin (which literally just means ‘a mill-thing’) came to mean “a mill-turning donkey” because the association was so common. Of the beasts of burden, it seems worthy to note, the donkey was also among the lowest in status; so even the animal mill-workers weren’t well off! We get a description of the treatment of a molarius in Apuleius’ Metamorphoses (often known by the title The Golden Ass) and it is hardly humane animal treatment, even by ancient standards.

At the same time, owning a pistrinum could be quite lucrative; one need only look at the famous Tomb of the Erysaces the Baker (pictured below) outside of Rome to see that. Millers, the fellows who owned and operated the mills – as distinct from mill-workers (the poor sods turning the millstone) – were often commoners of somewhat elevated status and wealth. The pistores in Rome – which included bakers and millers – had their own guild (called a collegium, from where we get college and colleague); their essential role in food production gave them a degree of political power. In medieval Europe, millers were usually a fair bit wealthier than the peasants around them, leading to sayings like something being “worth a miller’s thumb.” As noted briefly above, actually running the mill – making sure the grain was fed at the right rate, controlling the separation of the millstones, and so on – was a highly skilled task if you wanted to ensure high quality flour. The same, of course, was true of baking the bread itself. These were skilled occupations and as a result enjoyed a certain level of status, albeit still belonging generally to the commons. The miller or the baker was perhaps a head above the peasantry, but usually only a head and still very much a member of the lower classes in other respects, except in the rare cases he might become truly wealthy.

Via Wikipedia, the Tomb of Eurysaces the Baker, just outside of the Porta Maggiore, Rome. The Tomb dates to the Late Republic (c.50-20 BC); the small circular holes on the front side are meant to evoke ovens. Eurysaces was likely a freedman (that is an enslaved person who had obtained freedom) and clearly a successful businessman as a baker, given the extravagance of the tomb.

That said, state power often tended to interact quite a lot with both mills and bakeries. The food supply was, of course, of crucial interest to whoever held political power. Consequently the pistores of Rome, for instance, had (from Augustus onward) a special political appointee, the Praefectus Annonae overseeing them, himself an official of sufficient importance that we actually know the names of quite a lot of them. These praefects were very much not bakers but members of Rome’s wealthy, leisured landholding aristocracy serving in political office for political gain. During the European middle ages it was common, though by no means universal, for the local mill and/or bakery to be directly controlled or owned by the local lord; in many places it thus became illegal to mill grain anywhere else and millers were sometimes empowered by local political authorities to destroy the millstones of any illegal mills that challenged the lord’s monopoly (which in turn enabled him to extract resources from the countryside)!

For the actual millers and bakers, this must have often been a double-edged sword: on the one hand, that close political attention often meant greater political influence, but it also meant more direct intervention in their business by far more powerful individuals in the society. The impact of that second thing should not be overstated: the social status gap between a Roman praefectus or a Scottish baron and an individual miller or baker was vast and the former could bring down tremendous power on the latter if they wanted to. Roman emperors and medieval kings could and did fix prices for grain and bread, for instance, in ways that could not have been fantastic for the fellows doing that work. I feel the need to point out that price-fixing schemes of these sort do not actually work. Nevertheless, again, we ought to imagine the average miller or baker to be in a better position relative to those powers, certainly, than the badly exploited small farmers of last week.


Where I really want to bring the focus of this post to at the end is not the processes – although we’ve talked a lot about the processes, by necessity – but the people. As we’ve discussed before, there is a tendency when popular culture represents the past to erase not merely the farmers, but most of the commons generally. Castles seem to be filled with a few servants, a whole bunch of knights and lords and perhaps, if we are lucky, a single blacksmith that somehow makes all of their tools (we’ll dispense with the solitary master-armorer in a later series).

But the actual human landscape of the pre-modern period was defined – in agrarian societies, at least – by vast numbers of farms and farmers. Their work proceeded on this cyclical basis, from plowing to sowing to weeding to harvesting and threshing to storage and then back again. Religious observances and social festivals were in turn organized around that calendar (it is not an accident how many Holy Days and big festivals seem to cluster around the harvest season in late Autumn/early Winter, or in Spring). The uneven labor demands of this cycle (intense in plowing and reaping, but easier in between) in turn also provided for the background hum of much early urban life, where the ‘cities’ were for the most part just large towns surrounded by farmland (where often the folks living in the cities might work farmland just outside of the gates). People looked forward to festivals and events organized along the agricultural calendar, to the opportunities a good harvest might provide them to do things like get married or expand their farms. The human drama that defines our lives was no less real for the men and women who toiled in the fields or the farmhouses.

And of course all of this activity was necessary to support literally any other kind of activity.

Next week, we’ll finish our look at cereal farming and bread by looking at those folks – all of the other non-farmers – and how they fit into this system.

105 thoughts on “Collections: Bread, How Did they Make it? Part III: Actually Farming

  1. Ah, broadcasting. . . .

    When writing, it’s wise to avoid metaphors that your characters wouldn’t use. Like, no “strong suit” before the invention of cheap paper, and hence of playing cards, and hence of bridge.

    But the real fun arises when your characters would use it, but your readers won’t believe you. A farmer could speak of broadcasting a rumor long before radio, but readers will be jolted.

    Liked by 1 person

    1. The famous “Tiffany” problem. Fantasy English is a very interesting constructed dialect of stuff that SOUNDS old but is easy to understand. It’s kind of fascinating what works could as proper fantasy English and which don’t and how that dialect has been constructed and changes.

      Liked by 2 people

    2. Another funny example is “looks like meat’s back on the menu, boys!” which could be interpreted in a number of different ways: either (a) it’s an anachronistic reference by the real-world LotR scriptwriter to modern-style restaurant dining menus, of which the Uruk-Hai obviously wouldn’t have had any cultural knowledge; or (b) it’s a reference to some actual element of Uruk-Hai social structure like crude army ration lists, translated for convenience as “menu” by the in-universe LotR storyteller, i.e. Frodo sitting in the comfort of his home in the postwar Shire writing the Red Book of Westmarch; or (c) my favorite, extremely overthought interpretation, what we might call “the Last Ringbearer interpretation,” that it is a reference to dining menus like those at modern-style restaurants, which are part of Uruk-Hai culture, and the alleged barbaric savagery of orcish/Uruk/goblin social structure is itself a much deeper and more pernicious in-universe propaganda myth designed to help the “free peoples of the West” racialize and demonize their wartime enemies.


      1. The line of course is movie-only, not Tolkien.

        As himself notes, we basically only see the ‘soldier’ mode of orcs.

        I see people say that sit-down restaurants are pretty modern, though digging on Wikipedia pushes it back to French cabarets before the “restaurants” proper, and also an 1962source claim of Song Dynasty restaurants and menus much earlier. That said, buying food is much older, Roman if not earlier, and I think it likely that some Roman taberna offered a choice of available foods, probably on a chalkboard, which we could call a menu even if nothing was being made to order. (E.g. “bread and cheese”, “falafel”, “sausage”.)

        I have no sympathy with “The Last Ringbearer” school of thought.


  2. There was a standard English legal proceeding called ‘suit of mill’ – enforcing compulsory use of the lord’s mill, and a famous case where the Abbot of St Albans confiscated all the hand-mills in the town and paved the abbey floor with them.


  3. Typo hunt, with the usual caveat that I may be wrong:

    A woman, with her skirts gathered up, hauls the grain to the threshing room (center left). -> (center right)

    In the back left, you can see a man inside the building threshing grain with a hinged threshing flail. -> back right

    and the rest of the plant left in the field to try -> dry


    1. Since you mention a few of them, I’ll add my list of proofreading corrections here:
      already pulled in an processed -> already pulled in and processed
      design as developed -> design was developed
      lot of land need to get plowed -> lot of land needs to get plowed (or change get to be)
      creating a very rough lines -> creating very rough lines (delete a)
      in order to over the seeds -> in order to cover the seeds
      seed is even over -> seed is evenly distributed over (suggestion)
      assigns it less days for plowing -> assigns it fewer days for plowing
      the straw made be gently raked -> the straw may be gently raked
      standard method with rice -> standard methods with rice
      both, however) -> both, however.) (missing period added)
      and ground in powder -> and ground into powder
      bread if customarily -> bread is customarily
      is worth taking about -> is worth talking about
      in Latin(which literally -> in Latin (which literally (insert space)
      to saying like something -> to sayings like something
      members of Roman’s wealthy -> members of Rome’s (or Roman) wealthy


    2. A bit late, but I noticed a couple as well.
      “which has no mouldbord” -> “mouldboard” (though see comment at the end)
      “oxen, being big, strong things can pull a much deeper plow” -> “things, can” (probably stylistic)
      “disrupting larva or eggs” -> “larvae”
      “even in societies were agricultural labor” -> “where”
      “so much English idiomatic expressions” -> “many”
      “the upper-layer of soil” -> “upper layer” (hyphens are the worst. Grain of salt.)
      “Roman agricultural (writers suggest” -> “agricultural writers”
      “better yet turned in to porridge” -> “into”
      “Typically the a rough-surfaced, somewhat porous” -> “Typically a”
      “was quite popular in the later medieval Europe” -> “in later” (as written implies that the 11th century was not contemporary with medieval Europe, which I think isn’t true?)
      “Tomb of the Erysaces the Baker” -> “Tomb of Eurysaces” (both “the” and missing a “u”)
      “schemes of these sort” -> “of this sort”
      I’m also curious that almost immediately after defending American spellings of words you started consistently using the spelling “mouldboard”, when I would expect the American to be “moldboard”. Any reason for this?


  4. Note: although often not the case in the past, modern brewing yeast and baking yeast are quite different strains. Using beer yeast for bread will be a bit slower to rise but should be OK, using baking yeast for making booze is going to give you problems as won’t taste very good at all.


      1. I first learned the meaning of that phrase (and and much of the rest I knew of pre-industrial grain farming prior to this article) from Sunday school.


      2. I agree that the phrase still mostly exists in a religious context. The lessons typically briefly explain the context (especially for children), so the religious analogy makes sense.


  5. Based on your two previous articles and the differences between the large estates and the small subsistence farmers, how much of the agricultural product that went on to feed the cities, either as tax or as commercial product came from the small farmers and large estates proportionally? Do we have any idea of the split? You mentioned taxes of about 10% of the harvest for the small farmers but without an idea of the scale of the total tax take its hard to estimate the importance of the small farmers in feeding the cities.


    1. It was emphasized in the first article that small farmers produced little extra food and most of the extra went into throwing a feast for the neighbors to build community bonds so that they would help in times of trouble.

      So it would be the big farms that provided 90+% of the food for the cities.


      1. whats more, those big farms would exist in the first place because of that trade to the larger urban areas. whether they were created specifically for them, or were the result of farmers growing their operation to fill a lucrative trade void, the big farms would live or die on the trade to urban areas. since unlike subsistence farming, operating their farms would involve money, needed to pay workers, maintain the less common tools of the trade, etc.
        money which they can only get through the sale of their cash crops, which are only in demand in urban areas where there isn’t any subsistence farming going on to support the populace. and in turn, those urban areas are going to be where the craftsmen who make and maintain the more complex tools of the farmer will be located, since urban areas also support enough other trades that also need specialized tools from those craftsmen that they are constantly busy.

        you aren’t going to see many large organized farms without urban areas nearby.

        Liked by 1 person

        1. I don’t think this is true. A medieval manor is a big farm, and wasn’t necessarily cash oriented. You can have a big farm that’s food oriented, with the owner skimming off a lot for feeding meat and war animals, plus the greater nutrition of the owner (especially as part of a warrior class.)

          Wiki pages said a Greek hoplite commonly owned 5 hectares (I assume worked by the hoplite family and a few slaves), but a Spartan citizen would have 10-18 hectares (worked entirely by slaves.) Not huge farms yet, but showing the direction things can go in.


        1. Well the “little extra” is after taxes and tithes. Also the small subsistence farmers will tend to have a more people per acre than the sharecroppers which changes the equations.


        2. AIUI small farmers here means people who own small farms. Sharecroppers function economically as labour for large farmers.


    2. From the Conclusion:
      “where the ‘cities’ were for the most part just large towns surrounded by farmland (where often the folks living in the cities might work farmland just outside of the gates)”

      How would you count subsistence farmers who lived in the cities and worked the nearby farms? Is that export to that cities?


  6. I would argue that setting a price ceiling on milling can work and is actually desirable. Demand for flour is inflexible and constant: people must eat and they prefer freshly baked bread. If a miller or millers guild has a monopoly, they will have high incentive to set the price too high. But bread is essential so making it as affordable as it can be is good both for the general population and for the elites who want to have a stable social structure.


    1. True, but mills were often an elite monopoly. One article I read noted that in the north of England, where communal mills were fewer, the mills charged one-thirteenth of the grain while in the south it was one sixteenth. Presumably this fed into local markets or went to the lord’s household.


    2. Or put another way, if an efficient mill is a natural local monopoly, that’s the sort of situation that calls for regulation even in orthodox economics. If you make sure to allow competition from hand mills or visiting a nearby mill (how far apart would they be) that helps, but if there’s one well-build (capital-intensive) watermill sitting on the only good location for such, regulation may still be a good fit.

      Price controls are usually a bad idea in competitive markets but that requires effective competition.


      1. It would seem that many mills were un-natural monopolies, so price controls would be a case of the important people trying to solve a problem that they had created themselves. Or at least, trying to be seen to solve a problem they had created themselves.


    3. When high prices result from monopolists taking advantage of their customers’ lack of options, you’re right. However, rising prices may reflect an actually reduced ability to provide the product, e.g. grain prices go up when there’s less grain to sell, so when that happens, by imposing a price ceiling you’re taking away the usual method of resource apportionment (i.e. give resources to the people with the most money), and you have to replace it with something else. Given that the government is usually made up of the elite, who would have enough money to spend on grain in a free market and who are unlikely to adopt regulations that make their own lives much harder, I’m not sure that price ceilings would do much more than make grain more affordable for the elite and less available to everyone else. If the government actually cares about the well-being of all its subjects, you are probably right, but I doubt that that was true in most premodern states.


      1. I can only make sense of this if you’re imagining a ceiling on the price of flour coming out of the mill. The more logical ceiling is on the rate the miller can charge the person bringing grain. Scarce grain will mean expensive flour and bread but you can limit the effect a mill monopoly has, to no one’s loss except the miller’s.


      2. Effective pre-modern states were often concerned to limit the impact of grain shortages, and forbade hoarding, set prices and gave grain away to the poor. In medieval England and France this sort of intervention was usually at the local level. In China and India the state maintained granaries to the same purpose. The great sufferers in lean times were the urban poor and the landless peasantry. The landless could look to the manor, the urban poor to the town.


        1. The Romans also had a bread dole, given to hundreds of thousands of people in Rome, complete with associated goddesses: Annona and Empanada.


      3. Food prices rising above what the urban population can easily afford is one of the classic factors in civil unrest, coups, revolutions, and suchlike violent changes of government. Functional governments take care to prevent it happening, at worst by transferring food shortages to the countryside where insurrections are more dispersed.


        1. It’s one of the most enduring lessons of the French Revolution. If civilization as you know it is between the mob and the bread then the mob will tear it apart brick by brick to avoid starvation.


  7. How do legumes and other non grain crops fit into this (if they did fit in much)? I hear vaguely of crop rotation, and also of nutritionally how useful they are, but it seems some of these steps would work differently.


    1. While looking up the history of the seed drill I found out that Georgics was still in widespread use as a prescriptive text in the seventeenth through nineteenth centuries. That doesn’t conclusively prove that his techniques were in use in his period but does suggest they basically worked, which would be odd if he pulled them out of his ass.


  8. Fascinating, thanks for writing this! I’m assuming the flour and bread were what we would today call whole wheat, am I right? Do your sources mention anything about the various grades of flour and their relative prices?


    1. Yes, usually.
      White flour is made by repeatedly sieving wholemeal to remove the bran.
      In preindistrial times, this was a labour-intensive step.
      This made white bread more expensive than wholemeal. The whole issue ended up getting kind of nasty during the French Revolution with the prohibition of breads other than ‘equality bread’.

      Liked by 1 person

  9. I remember reading on Sam[]zdat that grain was easier to tax than other crops. Is that because of the post-winnowing stage, where all the grain is in a state that is easy to see and weigh or is that because of all the hardware necessary to eat it as anything other than porridge?

    If a desperate peasant wanted to hide some of their grain from the taxman, would the process be any different than hiding potatoes or beans? I’m asking mostly from a writer’s perspective, because that sounds like an interesting subplot to include.


    1. My understanding is that, primarily, grain transports well. It doesn’t spoil, and is as dry as foodstuffs get. Try to extract the same nutritional value* from the farmers in the form of fruits instead, and you end up having to transport several times as much stuff (most of the difference being water). Collection is also a factor: farmers did have to collect all grain, whereas some other crops (especially root crops) can be left in the field. If you want them, dig them up yourself, taxman!

      *: Nutritional value is what matters. This is what determines how many people’s work you can buy with the tax.


      1. Root crops are also higher moisture! Charles Mann wrote about potatoes having 25x the weight yield of 1600s wheat, but 4x the nutritional yield given potatoes being 80% water and dry what 20%. Which means more bulk again.

        Nuts would seem as dry and denser than cereals but we don’t see mass nut agriculture.


    2. If you want to hide potatoes, don’t you just put them in the ground? Or leave them in the ground until you need the? I do not know.


    3. Probably both the post-winnowing stage AND the hardware.

      Compounding both, remember what was said about how grain has to be stored in a cool, dry place and kept *very* dry to remain edible for long. Typically this means that you have to store your grain in a built structure of some kind, something well-prepared and adequately maintained. You can’t just drop the grain sacks in a hole in the ground in the woods and expect them to be edible when you come back for them.


  10. A bit of nitpicking here:

    As is common, the British spelling has no particular claim to antiquity – the Old English is plōh, which in turn gives the Middle English word, spelled alternately as ‘plow,’ ‘plouh’ and ‘plough,’ among other variations.

    The British spelling is more archaic in a sense, since the “gh” in British English “plough” and the “h” in Old English “plōh” represent the /x/ sound that existed at the end of the word; American English “plow” represents its loss in Modern English.

    Not that etymology is a good guide for spelling; for example, “doughter”, not “daughter”, is the etymologically correct form.


  11. You state that a Mediterranean farmer could expect two crops a year; how far north did this extend? How did people even do agriculture in regions with winters too cold to expect a crop?


    1. Plant in the spring, ripen and weed in the summer, harvest in the autumn, like the man said. Agriculture could be inherently a bit more precarious in colder climates, and more so as the climate grows colder, of course. But these are annual plants; they evolved to ripen and spread seeds every year during some recognizable growing season. It’s doable.


  12. My mother grew up on a mid sized (for the UK, am guessing fairly small for the US) arable and dairy farm. As recently as the 1980s they still burned the stubble every year – we have a great photo of her as a teenager standing on black still-smouldering earth, pitchfork in one hand and black hair down round her shoulders, wearing jeans and a tee and grinning like hell’s most wholesome demon.

    I can’t speak for all modern practices (which are no doubt to some extent as variable as ancient ones in that the local microclimate, soil conditions, and markets are going to be highly variable) but many of the other practices you mention are still running on the family farm, which is still owned by my uncle. Stubble burning was made illegal (in the early 90s I think) to preserve air quality – I only have the haziest memory of seeing it myself. But manure from the dairy herd is still collected and used on the land, as an organic farmer my uncle uses crop rotation, leaving fields fallow. He grows a mixture of winter wheat and summer barley, as well as rotations of other crops (oil seed rape, hay and/or kale for animal fodder, nitrogen fixers to revive the soil). Planting time and harvest time are still weather dictated, high intensity, unpredictable events.
    What has changed most is the number of people needed to do all this work: when my grandma ran the farm she had a little troop of farm workers, but far fewer than would have been needed a generation before. My uncle gets by with a couple of highly skilled employees. The difference is of course the tools available – tractors to pull ploughs; combine harvesters that (as the name suggests) combine harvesting, threshing and winnowing the grain and do all that work in the time it takes you to drive them back across the field so that they can tip the grain straight into a waiting trailer; milking machines that allow one person to milk fifteen or twenty cows at once.
    Modern granaries can have powered ventilation systems – the grain sits on a huge grill and enormous fans can push air through from underneath, helping keep the crop dry.

    The modern equivalent of the big landowner (in terms of dictating flexibility for the farmer) is the big supermarket – farmers rarely get to sell their products directly to end users, and the food market is heavily weighted towards a handful of big brands.

    I’d hate to be a medieval subsistence farmer. It looks stressful enough farming nowadays, when personal starvation is not always on the cards.


  13. Farming was clearly a skilled occupation. Medieval peasants were not the brainless clods their social superiors portrayed. They couldn’t afford to be. They may have been illiterate and highly conservative in their methods but they had to plan and reason out optimal strategies on the fly.

    Liked by 1 person

    1. Their problem was that their many skills were quite common, and easily transferred from place to place. The Gallic slave bought by the Roman farmer was likely already conversant with agricultural tasks, and Flemish peasants readily adapted to farming in Poland in the Middle Ages. A ruthless elite could suppress the peasantry knowing that more peasants could be had cheaply.

      Liked by 1 person

  14. ‘…Consequently, this is the moment we are mostly likely to see marginal laborers – children and in some cultures women – employed directly in agricultural labor…’

    I’ve heard that here in the UK, as late as the middle of the twentieth century, schools in some regions had to work around the need for ‘all hands on deck’ as you put it of the hop harvest.


    1. Canada too. I don’t know if it’s still the case, but as late as the 90s kids in rural areas would often miss a few weeks of school because their families needed all hands on deck to help with the harvest, even with all the technology they had on hand to assist in the process.


      1. Also true in the US.
        It has only been in the past 25 years or so that some school systems began to change their yearly schedule. Until then (and of course now, with COVID they may discover a whole different schedule) schools all over the country began in early September and ended in early to mid-June. The origins being the need to officially release not just the students but also the teachers for harvesting labor rather than try to teach to empty classrooms. In SoCal holiday schedules have been extended because so many students traveled with their families across the border and did not return “on time,” leaving low attendance, which leads to less tax dollars.


        1. Even today in Iowa many of my coworkers go back to their brother’s farm at night and on weekends to help out during harvest. It isn’t unusual either for someone to have a week of vacation on the schedule with a note that if it is raining they will reschedule. Harvest is a busy time of the year for farmers and there is a real problem finding skilled labor to help out. (remember they only need this labor for a couple months of the year). That family is often working for meals and friendship is a nice bonus, but most farmers have plenty of money to pay help.


  15. This is quite interesting; thank you for sharing this information in an easily understood form like this.

    I have a question about your section on millers and mill workers. You mention that “being ‘sent to the mills’ to work was a severe punishment, often meted out to the very poor or enslaved persons … the fellows actually turning the millstone were likely to be captives, criminals or enslaved persons of extremely low status, or else very desperately poor….” Why would people who had been enslaved, rather than those who were born as slaves, be more likely to be sent to work in the mills? It seems likely that people who had been used to freedom earlier in their lives would be more likely to run away or otherwise resist being kept as slaves than those who had been slaves for their whole lives, and thus would more often be punished with more dangerous and unpleasant work like this; is that the main explanation?


    1. I think that sentence is supposed to be understood as “slaves whose value is basically the strength of their limbs, not the skill of their fingers or minds” rather than “people who were just captured instead of people who were born into servitude”. From what I understand of slavery in Rome (2nd century BCE to 1-2nd CE), there was a bit of a hierarchy based on the slave’s skills. Someone who could read and write skillfully could be purchased as a tutor for one’s children or as a secretary. Skilled slaves were more likely to be treated well, and had greater chances of becoming freedmen, I imagine (hopefully Brett can respond yea or nay!) A captive in war is less likely to have mastered the language of its captors. The easiest way to get one’s money’s worth is to use them on unskilled, backbreaking work (that is an incredibly awful thought), why waste additional money on education when you send them to the mills or the mines?

      To sum up what I’ve tried to say (rather badly) is that captives were more likely to be lumped in with criminals and less valuable (less skilled/educated) slaves due to the fact that they were either unable to communicate their skills effectively, or because as former enemies of the state they were perceived as being as valuable as criminals, unskilled, and the poor.

      I’m finally one step above a McJob. I’d be screwed in Ancient Rome.

      Liked by 1 person

      1. Also because their skills were likely to be incompatible with the jobs at hand. British slaves were particularly decried as being stupid and untrainable.


  16. This is probably outside your bailiwick, but any thoughts on the feasibility of treenut-based agriculture, to presumably cut down the plowing and weeding time? Crop yields sound comparable if not superior to cereals, especially accounting for caloric density. My guesses for why we don’t see it are

    * humans don’t want to wait 10 years for harvest (especially if some army burned your ‘crops’)
    * harvesting is a problem, unless you’re good at keeping the trees dwarfed
    * lots of similar trees together make pests like squirrels very happy
    * some trees may not reliably produce a large crop every year (preferring feast or famine cycles) and it’s hard to breed trees

    For fantasy worlds (like elves) this seems where some light-touch biological magic could make a big difference: ward off pests, controlling yield times (making it yield early is high magic to me breaking energy budgets, but delaying until a convenient time is different.) And Tolkien elves are adept in trees.


    1. Also on the fantasy front: I’ve seen people speculate about magic to increase crop yields (predict or control weather, ward against pest or disease), which could help general reliability a lot, but I’ve noted wouldn’t necessarily greatly change the % of labor force on farms: you still need to sow and harvest (unless you have magic for *that*, but “harvest crops” is a lot different from “bless crops” in magic feel.) If you increase crop yield per land, that may just mean needing (and feeding) more labor per land, not increasing crop per labor.

      (Also if you have more food and people but not more textiles and wood, you can have higher population density that’s well fed but poor in clothing and timber/firewood… like early 1800s Ireland, in well-fed poverty thanks to potatoes?)


      1. The problem is, as I see it, unless magic is ubiquitous in a fantasy setting, anyone able to tell the mundane natural laws to bend over backwards and prepare to be spanked can (and probably does) charge a lot of money for their time and services – more than most farmers are going to be able to afford.
        So it probably requires a magician with a patron specifically interested in agriculture, for any kind of magical resources to get pointed at farming. (Well either that or a magician who has made their fortune, retired, and is now amusing themselves with running an estate for their own fun or out of a sense of religious or social obligation.)
        (EXCEPTION: I think one of TSR’s worlds in the 1990’s had a pantheon with a deity specifically interested in farming and agriculture, who had priests able to carry out divinely sanctioned magic and with the church’s specific focus being assisting farmers and the like; they probably still charged – especially non-worshippers – for their services, mind you.)

        Quasi-sentient undead are a game changer in terms of agricultural labour. (As for which matter are golems, other constructs, and elementals.) You suddenly have a tireless workforce, which is isn’t going to fall ill or (usually, depending on what you’re using) want paying, and is probably very low maintenance.


        1. ‘Ubiquitous’ is certainly possible.

          * The King’s Peace has the land gods providing useful magic to everyone. (Contraception and healing, specifically.)
          * Village priests might have it.
          * Elves, amongst themselves.
          * The Exalted RPG had widespread thaumaturgy, of just the sort of power level of “ward pests and influence biorhthms”
          * If magicians are rare and powerful, well, lords are interested in the volume and security of their food supply! As long as magicians aren’t so powerful as to live on their own magic or conjured food, they’ve got potential interest.


        2. Much depends on how large scale the spells can be made. If the king can decide that haying will be the last week in June, and the weather wizard obeys, and all the peasants know it, it can help enormously.


          1. Not to beat my own drum, but I collaborated in the development of an RPG where low-level useful magic use was ubiquitous – it does indeed give a different landscape and social system when you work your way into it. I ended up writing some stories set there (now three novels), where this is part of the background. Broadly, extreme inequality is difficult, gender counts less, but ecological constraints are tighter (don’t upset the dryads!).


          2. The fun thing is that most “medieval” RPGs would require ubiquitous low-level useful magic use. Too many women frolicking about in occupations they would never have been allowed to train for in that number if infant mortality wasn’t way low — “Making a fortune and bearing thirteen children — no human being could stand it.” as Virginia Woolf put it, and large numbers of women had to have thirteen children because others died in childbirth with the first one and so didn’t contribute to the average children that seldom made it to adulthood.

            Liked by 1 person

      2. ‘…which is isn’t…’ in my previous response should read ‘…which isn’t…’ Somehow an extra ‘is’ slipped in there. Sorry about that. (embarrassed)


      3. Biological pest control, companion planting, and other things exist and may provide exactly the feel you want.
        Invisible pheromones, viruses, bacteria, fungi, nematodes, and very hard to see insects. Before microscopes, who can tell you that mycorrhiza aren’t magic?
        This naturally extends from blessings to curses as well. Sow ergot in your enemy’s grain fields! Kill their apple orchards with fire-blight (a real disease)!
        Or go with something as simple as terra preta and/or mulching.

        Much less magical, but if you have multiple types of crops, consider that people will grow them even if they give somewhat lower yields and no reduction in risk, simply because they need to be sown/harvested at a different time from the main crop. This is (partly) how turnips made it into the four-field rotation.

        Random remark: I get a feeling that people would to some extent see draft animals as low-status people rather than non-people. In a mill, if you have one donkey and one slave, you put both to work, each on their own beam. Both complain about the dust in their own way. The donkey doesn’t speak, the slave isn’t worth listening to. If you had that donkey for several years, you might be more emotionally invested in it (e.g. protectiveness toward the loyal servant) than in a recently-bought slave.
        (Power output: one horsepower is approximately 750 W. It’s more of an engineer joke, but you can hear sometimes that one donkeypower is 250 W. And a human working with his legs, according to Wikipedia: “Maximum sustained power levels during one hour range from about 200 W (NASA experimental group of “healthy men”) to 500 W (men’s world hour record).”)


      4. Many if not most substance farming societies IRL had people *widely believed* to be capable of casting “bless crops” type magic. Its probably worth looking into how they worked sociologically, for anyone planning on writing that type of story.


    2. Well in pre-Columbian America acorn cultivation was very widespread along the eastern seaboard (which caused deer to have a big population boom since they had a lot of acorns to eat and few people hunting them after smallpox etc. did horrible things to the local population.

      Similarly there used to be a HUGE amount of chestnut trees in 19th century America which could provide a big part of people’s diets before the blight killed them off. You get similar things in Corsica. So what you’re talking about is certainly possible.

      The main problems with tree-based agriculture is that if you get a blight come through you’re really really fucked. Same with an enemy determined enough to kill off your food-producing trees since those take a LOT longer to recover than grain.


    3. I believe some recent work suggests parts of the Amazon supported large populations on tree-based agriculture before western intrusion.


  17. Quick notes about fermentation which was touched on briefly in the main post:

    1. I wouldn’t call mold infecting the grain fermentation per se, bread with mold all over it isn’t fermented. HOWEVER, mold can break down the longer starches of the grain that are hard for yeast to eat and then allow the yeast to come in and get to work which does cause fermentation. Usually grains are fermented by malting and mashing them which is a pretty involved process so in some places you get fermentation that’s intentionally kickstarted by mold. This is done today with Korean makgeolli (sour rice beer, really good stuff if you get the traditional version, not the commercial versions with artificial sweetener) and was probably done in a lot of other places before people perfected malting and mashing.

    2. Yeast is naturally found on the skins of many fruit so using fruit as a starter makes sense. Yeast also has an easier time eating fruit than grain since fruit sugars are simple and easier to digest.

    Liked by 1 person

    1. Wow, that is seriously impressive. Do you know of any resource, that lists the labor demand per crop over the years?


  18. 3 questions:

    1. Signal-boosting Dillon’s question above – how do legumes fit into this? Would farmers be rotating (say) wheat and chickpeas, and if so, did these crops have the same planting and harvest seasons?

    2. Was there any double-cropping in the Mediterranean? Cf. double-cropped rice in Guangdong and Southeast Asia.

    3. You bring up crosswise plowing as an intermediate step. Would that change with ox teams? Medieval manors (and Early Modern seigneuries) had long, narrow plots, optimized for ox teams that were a lot better at straight plowing than at turning.


  19. Can I just be a pedant myself (as well as giving away my day job as a Biology lecturer) and point out that in binomials the generic name should be capitalised (e.g. Triticum aestivum not triticum aestivum )


  20. Very nicely done! Also, now I really feel a need to go plug in my electrically-powered breadmaker, pour in my already-milled-for-me flour and a few other ingredients brought to me by automotive power, and make some bread! I probably won’t appreciate it quite as much as the poor farmers and mill-workers who did all of the above, but I will also probably not get any millstone in mine.

    Liked by 1 person

  21. I’m seconding/thirding/fourthing whatever number we’re at now the request for how legumes fit into this. But I also have a different question.
    What was the difference between wealthy and poor peasants? Animals, sure, but you note that peasants generally don’t have access to their own draft team. Was it that rich peasants had more poultry/sheep/maybe a milk cow or donkey? And was it often that rich peasants could afford to devote some land to luxury or non-grain crops(I’m reminded of the bible story of a king promising every man who served in his army would have his own fig tree…)


  22. Are you aware of / what are your thoughts on James C. Scott’s contrary takes on ancient farming?

    I have only read a summary of his most relevant book, Against the Grain, but I’ve read enough of his other work (which typically involves agriculture) to give a brief summary:

    Subsistence farmers typically prefer crops which do not have to be planted and harvested all at once, like beans or root crops, over grains. Grains are wildly over-represented in historical sources compared to what people actually ate.

    Grains are preferred by large institutions, especially the state, because they are especially easy to assess and collect surplus. Having everything harvested at once makes it easy for the tax collector or absentee landowner to come by once a year, see how productive the farm was (large piles of grain are hard to hide), and appropriate some. It makes much more work for the elite to take their cut of crops harvested over the course of months from the farmers.

    Large scale grain cultivation almost always occurred in exploitative relationships, by the nature of the crop itself. Since the people represented in history are typically these elites, most of our sources about farming focus on grains.


  23. One thing about millers were that they were often quite unpopular among the peasants. This likely stemmed from the business-model – take a part of the flour as payment, in a way that wasn’t particularly transparent, and was open to either real or perceived abuse.

    A Dutch saying goes “A usurer, a miller, a banker, and a publican are the four horsemen of the apocalypse”, and it’s hardly a coincidence that the Miller in Canterbury is specifically a miller.


  24. I’m reading Scott’s Against the Grain, and just hit this, relevant to the legume question:

    “One might imagine that ancient domesticated legumes, say—peas, soybeans, peanuts, or lentils, all of which are nutritious and can be dried for storage—might serve as a tax crop. The obstacle in this case is that most legumes are indeterminate crops that can be picked as long as they grow; they do not have a determinate harvest, something the tax man requires.”

    That raised more questions, which may help answer.


    1. ‘The work of civilization, or more precisely the state, as we shall see, consists in the elimination of mud and its replacement by its purer constituents, land and water. Whether in ancient China, in the Netherlands, in the fens of England, in the Pontine Marshes finally subdued by Mussolini, or in the remaining southern Iraq marshes drained by Saddam Hussein, the state has endeavored to turn ungovernable wetlands into taxable grain fields by reengineering the landscape.’


  25. Unrelated thought: I think of porridge as “primitive” compared to making bread, but then realized that boiling without metal pots (to conduct heat without burning) seems challenging. I know there were methods — heat rocks and drop them in — but that seems like maybe as much work as grinding and making flatbread. So maybe porridge, or the infamous fantasy stew, presupposes a lot of infrastructure for peasants to have a metal pot. (There’s copper, but it’s toxic with acid foods; tin lined copper is basically the ingredients for bronze, which wasn’t cheap…)


    1. Ceramic cooking pots apparently go back to pre-Neolithic times. The potters wheel goes back 6,000 to 8,000 years. Therefore metal pots aren’t really needed.


  26. “millers were sometimes empowered by local political authorities to destroy the millstones of any illegal mills”. My understanding is that lords would generate income by auctioning licences to exclusively engage in a particular business in an area. I once saw the claim that the English civil war, by breaking down this system of licensed monopolies, contributed to economic growth through increased competition. Is that true?


Leave a Reply

Fill in your details below or click an icon to log in: Logo

You are commenting using your account. Log Out /  Change )

Google photo

You are commenting using your Google account. Log Out /  Change )

Twitter picture

You are commenting using your Twitter account. Log Out /  Change )

Facebook photo

You are commenting using your Facebook account. Log Out /  Change )

Connecting to %s