We should start with the sort of supplies our army is going to need. The Romans neatly divided these into four categories: food, fodder, firewood and water each with its own gathering activities (called by the Romans frumentatio, pabulatio, lignatio and aquatio respectively; on this note Roth op. cit. 118-140), though gathering food and fodder would be combined whenever possible. That’s a handy division and also a good reflection of the supply needs of armies well into the gunpowder era. We can start with the three relatively more simple supplies, all of which were daily concerns but also tended to be generally abundant in areas that armies were.

For most armies in most conditions, water was available in sufficient quantities along the direction of march via naturally occurring bodies of water (springs, rivers, creeks, etc.). Water could still be an important consideration even where there was enough to march through, particularly in determining the best spot for a camp or in denying an enemy access to local water supplies (such as, famously at the Battle of Hattin (1187)). And detailing parties of soldiers to replenish water supplies was a standard background activity of warfare; the Romans called this process aquatio and soldiers so detailed were aquatores (not a permanent job, to be clear, just regular soldiers for the moment sent to get water), though generally an army could simply refill its canteens as it passed naturally occurring watercourses. Well organized armies could also dig wells or use cisterns to pre-position water supplies, but this was rarely done because it was tremendously labor intensive; an army demanded so much water that many wells would be necessary to allow the army to water itself rapidly enough (the issue is throughput, not well capacity – you can only lift so many buckets of so much water in an hour in a single well). For the most part armies confined their movements to areas where water was naturally available, managing, at most, short hops through areas where it was scarce. If there was no readily available water in an area, agrarian armies simply couldn’t go there most of the time.

Like water, firewood was typically a daily concern. In the Roman army this meant parties of firewood forages (lignatores) were sent out regularly to whatever local timber was available. Fortunately, local firewood tended to be available in most areas because of the way the agrarian economy shaped the countryside, with stretches of forest separating settlements or tended trees for firewood near towns. Since an army isn’t trying to engage in sustainable arboriculture, it doesn’t usually need to worry about depleting local wood stocks. Moreover, for our pre-industrial army, they needn’t be picky about the timber for firewood (as opposed to timber for construction). Like water gathering, collecting firewood tends to crop up in our sources when conditions make it unusually difficult – such as if an army is forced to remain in one place (often for a siege) and consequently depletes the local supply (e.g. Liv. 36.22.10) or when the presence of enemies made getting firewood difficult without using escorts or larger parties (e.g. Ps.-Caes. BAfr. 10). Sieges could be especially tricky in this regard because they add a lot of additional timber demand for building siege engines and works; smart defenders might intentionally try to remove local timber or wood structures to deny an approaching army as part of a scorched earth strategy (e.g. Antioch in 1097). That said apart from sieges firewood availability, like water availability is mostly a question of where an army can go; generals simply long stay in areas where gathering firewood would be impossible.

Then comes fodder for the animals. An army’s animals needed a mix of both green fodder (grass, hay) and dry fodder (barley, oats). Animals could meet their green fodder requirements by grazing at the cost of losing marching time, or the army could collect green fodder as it foraged for food and dry fodder. As you may recall, cut grain stalks can be used as green fodder and so even an army that cannot process grains in the fields can still quite easily use them to feed the animals, alongside barley and oats pillaged from farm storehouses. The Romans seem to have preferred gathering their fodder from the fields rather than requisitioning it from farmers directly (Caes. BG 7.14.4) but would do either in a pinch. What is clear is that much like gathering water or firewood this was a regular task a commander had to allot and also that it often had to be done under guard to secure against attacks from enemies (thus you need one group of soldiers foraging and another group in fighting trim ready to drive off an attack). Fodder could also be stockpiled when needed, which was normally for siege operations where an army’s vast stock of animals might deplete local grass stocks while the army remained encamped there. Crucially, unlike water and firewood, both forms of fodder were seasonal: green fodder came in with the grasses in early spring and dry fodder consists of agricultural products typically harvested in mid-summer (barley) or late spring (oats).

All of which at last brings us to the food, by which we mostly mean grains. Sources discussing army foraging tend to be heavily focused on food and we’ll quickly see why: it was the most difficult and complex part of foraging operations in most of the conditions an agrarian army would operate. The first factor that is going to shape foraging operations is grain processing. [S]taple grains (especially wheat, barley and later rye) make up the vast bulk of the calories an army (and it attendant non-combatants) are eating on the march. But, as we’ve discussed in more detail already, grains don’t grow “ready to eat” and require various stages of processing to render them edible. An army’s foraging strategy is going to be heavily impacted by just how much of that processing they are prepared to do internally.

This is one area where the Roman army does appear to have been quite unusual: Roman armies could and regularly did conduct the entire grain processing chain internally. This was relatively rare and required both a lot of coordination and a lot of materiel in the form of tools for each stage of processing. As a brief refresher, grains once ripe first have to be reaped (cut down from the stalks), then threshed (the stalks are beaten to shake out the seeds) and winnowed (the removal of non-edible portions), then potentially hulled (removing the inedible hull of the seed), then milled (ground into a powder, called flour, usually by the grinding actions of large stones), then at last baked into bread or a biscuit or what have you.

It is possible to roast unmilled grain seeds or to boil either those seeds or flour in water to make porridge in order to make them edible, but turning grain into bread (or biscuits or crackers) has significant nutritional advantages (it breaks down some of the plant compounds that human stomachs struggle to digest) and also renders the food a lot tastier, which is good for morale. Consequently, while armies will roast grains or just make lots of porridge in extremis, they want to be securing a consistent supply of bread. The result is that ideally an army wants to be foraging for grain products at a stage where it can manage most or all of the remaining steps to turn those grains into food, ideally into bread.

As mentioned, the Romans could manage the entire processing chain themselves. Roman soldiers had sickles (falces) as part of their standard equipment (Liv. 42.64.2; Josephus BJ 3.95) and so could be deployed directly into the fields (Caes. BG 4.32; Liv. 31.2.8, 34.26.8) to reap the grain themselves. It would then be transported into the fortified camp the Romans built every time the army stopped for the night and threshed by Roman soldiers in the safety of the camp (App. Mac. 27; Liv. 42.64.2) with tools that, again, were a standard part of Roman equipment. Roman soldiers were then issued threshed grains as part of their rations, which they milled themselves (or made into a porridge called puls) using “handmills”. These were not small devices, but roughly 27kg (59.5lbs) hand-turned mills (Marcus Junkelmann reconstructed them quite ably); we generally assume that they were probably carried on the mules on the march, one for each contubernium (tent-group of 6-8; cf. Plut. Ant. 45.4). Getting soldiers to do their own milling was a feat of discipline – this is tough work to do by hand and milling a daily ration would take one of the soldiers of the group around two hours. Roman soldiers then baked their bread either in their own campfires (Hdn 4.7.4-6; Dio Cass. 62.5.5) though generals also sometimes prepared food supplies in advance of operations via what seem to be central bakeries. This level of centralization was part and parcel of the unusual sophistication of Roman logistics; it enabled a greater degree of flexibility for Roman armies.

Greek hoplite armies do not seem generally to have been able to reap, thresh or mill grain on the march (on this see J.W. Lee, op. cit.; there’s also a fantastic chapter on the organization of Greek military food supply by Matthew Sears forthcoming in a Brill Companion volume one of these years – don’t worry, when it appears, you will know!). Xenophon’s Ten Thousand are thus frequently forced to resort to making porridge or roast grains when they cannot forage supplies of already-milled-flour; they try hard to negotiate for markets on their route of march so they can just buy food. Famously the Spartan army, despoiling ripe Athenian fields runs out of supplies (Thuc. 2.23); it’s not clear what sort of supplies were lacking but food and fodder seems the obvious choice, suggesting that the Spartans could at best only incompletely utilize the Athenian grain. All of which contributed to the limited operational endurance of hoplite armies in the absence of friendly communities providing supplies.

Macedonian armies were in rather better shape. Alexander’s soldiers seem to have had handmills (note on this Engels, op. cit.) which already provides a huge advantage over earlier Greek armies. Grain is generally (as noted in our series on it) stored and transported after threshing and winnowing but before milling because this is the form in which has the best balance of longevity and compactness. That means that granaries and storehouses are mostly going to contain threshed and winnowed grains, not flour (nor freshly reaped stalks). An army which can mill can thus plunder central points of food storage and then transport all of that food as grain which is more portable and keeps better than flour or bread.

Early modern armies varied quite a lot in their logistical capabilities. There is a fair bit of evidence for cooking in the camp being done by the women of the campaign community in some armies, but also centralized kitchen messes for each company (Lynn op. cit. 124-126); the role of camp women in food production declines as a product of time but there is also evidence for soldiers being assigned to cooking duties in the 1600s. On the other hand, in the Army of Flanders seems to have relied primarily on external merchants (so sutlers, but also larger scale contractors) to supply the pan de munición ration-bread that the army needed, essentially contracting out the core of the food system. Parker (op. cit. 137) notes the Army of Flanders receiving some 39,000 loaves of bread per day from its contractors on average between April 1678 and February of 1679.

That created all sorts of problems. For one, the quality of the pan de munición was highly variable. Unlike soldiers cooking for themselves or their mess-mates, contractors had every incentive to cut corners and did so. Moreover, much of this contracting was done on credit and when Spanish royal credit failed (as it did in 1557, 1560, 1575, 1596, 1607, 1627, 1647 and 1653, Parker op. cit. 125-7) that could disrupt the entire supply system as contractors suddenly found the debts the crown had run up with them “restructured” (via a “Decree of Bankruptcy”) to the benefit of Spain. And of course that might well lead to thousands of angry, hungry, unpaid men with weapons and military training which in turn led to disasters like the Sack of Antwerp (1576), because without those contractors the army could not handle its logistical needs on its own. It’s also hard not to conclude that this structure increased the overall cost of the Army of Flanders (which was astronomical) because it could never “make the war feed itself” in the words of Cato the Elder (Liv 34.9.12; note that it was rare even for the Romans for a war to “feed itself” entirely through forage, but one could at least defray some costs to the enemy during offensive operations). That said this contractor supplied bread also did not free the Army of Flanders from the need to forage (or even pillage) because – as noted last time – their rations were quite low, leading soldiers to “offset” their low rations with purchase (often using money gained through pillage) or foraging.

Of course added to this are all sorts of food-stuffs that aren’t grain: meat, fruits, vegetables, cheeses, etc. Fortunately an army needs a lot less of these because grains make up the bulk of the calories eaten and even more fortunately these require less processing to be edible. But we should still note their importance because even an army with a secure stockpile of grain may want to forage the surrounding area to get supplies of more perishable foodstuffs to increase food variety and fill in the nutritional gaps of a pure-grain diet. The good news for our army is that the places they are likely to find food (small towns and rural villages) are also likely to be sources of these supplementary foods. By and large that is going to mean that armies on the march measure their supplies and their foraging in grain and then supplement that grain with whatever else they happen to have obtained in the process of getting that grain. Armies in peacetime or permanent bases may have a standard diet, but a wartime army on the march must make do with whatever is available locally.

So that’s what we need: water, fodder, firewood and food; the latter mostly grains with some supplements, but the grain itself probably needs to be in at least a partially processed form (threshed and sometimes also milled), in order to be useful to our army. And we need a lot of all of these things: tons daily. But – and this is important – notice how all of the goods we need (water, firewood, fodder, food) are things that agrarian small farmers also need. This is the crucial advantage of pre-industrial logistics; unlike a modern army which needs lots of things not normally produced or stockpiled by a civilian economy in quantity (artillery shells, high explosives, aviation fuel, etc.), everything our army needs is a staple product or resource of the agricultural economy.

Finally we need to note in addition to this that while we generally speak of “forage” for supplies and “pillage” or “plunder” for armies making off with other valuables, these were almost always connected activities. Soldiers that were foraging would also look for valuables to pillage: someone stealing the bread a family needs to live is not going to think twice about also nicking their dinnerware. Sadly we must also note that very frequently the valuables that soldiers looted were people, either to be sold into slavery, held for ransom, pressed into work for the army, or – and as I said we’re going to be frank about this – abducted for the purpose of sexual assault (or some combination of the above).

And so a rural countryside, populated by farms and farmers is in essence a vast field of resources for an army. How they get them is going to depend on both the army’s organization and capabilities and the status of the local communities.

Bret Devereaux, “Collections: Logistics, How Did They Do It, Part II: Foraging”, A Collection of Unmitigated Pedantry, 2022-07-29.