Posted by Bret Devereaux

https://acoup.blog/2025/08/22/collections-life-work-death-and-the-peasant-part-iva-subsistence-and-a-little-more/?utm_source=rss&utm_medium=rss&utm_campaign=collections-life-work-death-and-the-peasant-part-iva-subsistence-and-a-little-more

https://acoup.blog/?p=31013

This is the start of the fourth part of our series (I, II, IIIa, IIIb) discussing the structures of life for pre-modern peasants, who made up the majority of all humans who have ever lived. In the last few sections, we’ve looked broadly at how mortality, marriage and childbearing patterns shape the households these folks live in: high mortality, particularly child mortality shapes marriage and childbearing patterns, however even under extremely high mortality regimes, some level of fertility control is required to avoid more rapid population growth than we generally see in the evidence. As an aside, that should also help explaining why these populations ‘bounce back’ from catastrophe so quickly: in a situation where the land and resources to support rapid growth are available, these communities are capable of fairly rapid population growth.

Most of all, those models give us a broad sense of what these households might look like. Of course within that sense there is variation, but within a range – these households are more like each other (excepting the late/late European marriage model) than they are like most households in the modern industrialized world.

This week, we’re going to start to take that understanding of household structure and see how it shapes and defines the daily activities of these peasants, which is to say labor. Part of the reason I wanted to write this series was to debunk the utterly silly idea that people today work more than medieval or ancient peasants. In the context of the very low labor productivity that peasants faced, it took a tremendous amount of labor just to manage bare subsistence and even more to obtain ‘respectability’ in terms of material needs. That isn’t to say peasant life was joyless or lacked free time, but there is a reason that the moment industrial life in cities became available, millions of peasants flocked to it.

Finally, before we dive in, I want to note that since the school year has started back up and I am again teaching, I am going to be breaking this part into a lot of smaller components, working through various shaping concerns of peasant labor and how we might model them. We’re going to start this week by just thinking through what a peasant household might need and then in subsequent weeks we’ll look at the labor they have to put in to get it.

But first, if you like what you are reading, please share it and if you really like it, you can support this project on Patreon! While I do teach as the academic equivalent of a tenant farmer, tilling the Big Man’s classes, this project is my little plot of freeheld land which enables me to keep working as a writers and scholar. And if you want updates whenever a new post appears, you can click below for email updates or follow me on Twitter and Bluesky and (less frequently) Mastodon (@bretdevereaux@historians.social) for updates when posts go live and my general musings; I have largely shifted over to Bluesky (I maintain some de minimis presence on Twitter), given that it has become a much better place for historical discussion than Twitter.

Subsistence

Paul Erdkamp framed the economic goals of peasant small farmers as “subsistence – and a little bit more” and I think that is a sharp way to understand the aims these households have.¹ On the one hand, bare subsistence – just enough food and clothing to survive – isn’t enough: our small farmers will need some goods they cannot produce themselves and on top of that will be forced to support non-farmers through things like taxation. So ‘just subsistence’ wasn’t enough.

On the other hand, the gains to be had by working for more than subsistence were slim. We’ve discussed this before, but stockpiling money was of only limited use for our peasants. Banking as a service was simply not generally available in most of these societies for the poor and not necessarily reliable where it was available.² Holding cash was remarkably risky: it could be stolen or lost and it provided a ready supply of easily extractable wealth for elites or the state to tax.

Worse yet, the moment where our farmers might really need their savings – a harvest failure – was the precise moment that food would be so scarce no amount of money could buy it anyway.

Storing up goods had equal problems. Some amount of moveable property might be worth keeping safe as security against catastrophe or as a supply for payments – fabric was a good option for this, being immediately useful, valuable and portable – but there were sharp limits to this too. Of course the primary production of the farm – food, mostly in grains – doesn’t keep forever. There was no way to, for instance, stockpile enough grain ‘to retire on’ because grain spoils, it gets eaten by pests and so on. As noted, fabric is a little better for value preservation, but only a little and once you have a lot of it, it ends up being exposed to the same theft-or-taxation risks as money.

Mostly importantly, there’s almost no way in these societies for a farming household to work its way out of the peasantry. These are, almost invariably, remarkably low social mobility societies by modern standards and indeed they expect low social mobility as part of their generally communal attitude: you are supposed to fill the same role in the community as your parents did, in the context of a community whose social and economic order is not supposed to change. But equally as a matter of finance, escape from the peasantry is nearly impossible, because the price of productive capital (land, work animals, tools and in many of these societies, enslaved laborers) is very high, while the value of labor – the thing a peasant has to give – is very low. Do peasants sometimes work extra hard and save carefully to buy an extra field or keep a plow team? Of course. But buying one’s way out of the peasantry doesn’t mean going from a 3 acre farm to a 3.5 acre farm, but from a 3 acre farm to a twenty or thirty acre farm, to have the amount of lend necessary to no longer need to fully take part of the daily back-breaking labor yourself. The returns to agricultural labor were simply not typically high enough, compared to the typical cost of land, to enable that kind of growth even over a whole lifetime – and that’s without the existence of Big Men, landlords and the state all of whom will want to have some claim on all that wealth being built up.

In short, few peasants could hope, by dint of ‘hard work’ to work their way into something other than being a peasant and so as a result there was very little material comfort extra labor could buy them beyond a certain level. I should note this is a big difference from modern societies, where the high productivity of labor means that a extra labor often can bring greater material comforts or financial security.³

So much like with fertility, our peasants are aiming to work a lot, but not – generally – for maximum labor. Instead, they’ll aim for a workload that ideally puts them securely above subsistence – although as we’ll see, the unpredictability of agriculture means this security is always, always precarious – and ideally approach an invisible line of “subsistence – and a little more.” That “and a little more” is representing access to the tools and comforts that might be available to the working peasantry, keeping in mind that in these structures, the kinds of comforts available to the elite – or even just the urban merchant and trade classes – are essentially forever out of reach (or at least, forever impractical) for the peasant. Beyond that point, it makes little sense for our peasants to do even more labor (as we’ll see, by that point they’ll be laboring a much lower marginal returns, so each new hour invested earns less and less) because they can’t realize much in the way of benefits from it.

So in understanding peasant labor we also have to understand peasant subsistence, because these households will work hard to a point, after which the incentives for more labor go down substantially. Thus understanding what subsistence – and “subsistence – and a little more” – are is going to help us to understand both what a household needs to do to survive but also how it is going to allocate the labor time it has. But to explore that, we’re going to need some model households to think with.

Some Model Households

Now, as we’ve discussed, our peasants do not meet their economic needs as individuals, rather they do so as households. So we need to begin by proposing some households. As we noted back in the first post, the average household size for a peasant community under conditions of high mortality is going to be between four and five, but since there are a lot of households and larger households contain more people, the average person lives in a household of around six or seven people. And now, of course, we’ve explored the mortality, marriage and fertility patterns that are going to produce that sort of household size (again, there will be variation – within limits – both regionally and temporally, but we’re dealing here with a relatively high degree of abstraction, a general rule from which specific places may deviate to a greater or lesser degree). And then, of course, there are very small households (less than three persons), which will come back into our story a bit later.

So let’s propose for our model three households along those lines, keeping in mind what we know about plausible marriage and fertility patterns. First, a household on the smaller side with four individuals, meet the Smalls: Mr. Smalls (M, 40), his wife Mrs. Smalls (F. 32) and their two children, John (M, 14) and Jane (F. 6) Smalls.⁴

Then, we’ll have a slightly larger, but still broadly typical multi-generational household of six members – meet the Middles: Mr. Middles Sr. has passed away but is survived by his wife Widow Middles (F. 46), with the nominal head of household being her son Mr. Middles Jr. (M. 27), who a few years ago married his wife Mrs. Middles Jr. (F. 22), with whom he has two daughters, Fanny Middles (F. 4) and Freida Middles (F. newborn).⁵ Widow Middles had two other children surviving to adulthood, a daughter now married (and thus in another household) and a younger son, Freddie Middles (M. 16), not yet married and still in the household.

Finally, we’ll have a larger household, something sitting basically on the mean for multi-family household size in the Egyptian data, with ten members. Meet the Biggs: Mr. Biggs Sr. was married twice (first wife lost in childbirth) and recently passed away, leaving behind his second wife Widow Biggs (F. 50). The household is instead run by Mr. Matthew Biggs (M. 43, from the first marriage) and his wife Mrs. Maddie Biggs (F. 33). They have three children, Mark (M. 16), Matilda (F. 12) and Mary (F. 8) Biggs. Matthew’s half-brother, Mr. Martin Biggs (M. 28) is also married to Mrs. Martha Biggs (F. 22) and they have two children, Michael (M. 4) and little Melanie (F. 1) Biggs.

I know that was hard to follow, so here’s a chart (beneath each name is the relation to the male head of household, just to help keep track):⁶

Now as you might imagine, these households are going to have radically different amounts of labor available but also different kinds of labor because of the differing ages and genders of their members. But it also is going to impact their resource demands as well, because while children are resource intensive, they’re not as resource intensive as adults. Which gets us back to the question of what “subsistence and a little more” is.

Bread and Baskets

So we are looking to establish two ‘baselines’ of economic survival here: a ‘subsistence’ baseline and the ‘a little bit more’ baseline, which we’ll call ‘respectability‘ for reasons that will become clear in a moment. The subsistence baseline is the line below which the household is actively in shortage. A household can dip below this line temporarily, relying on the charity of neighbors or on belt-tightening, but it cannot operate below this line permanently without eventually running short on the essentials of life. By contrast, ‘respectability’ reflects something closer to the ideal amount of material comfort a peasant could access. We’ll come back to the respectability line in a moment, but let’s return to subsistence.

In calculating subsistence, one handy thing is that a peasant household can actually produce almost everything it needs for subsistence itself. What is going to dominate these subsistence requirements are two major concerns: food and clothing. For food needs, when modeling past societies, historians often resort to a simplifying assumption: since grains (wheat and barley, generally in the form of bread) make up the large majority of all calories these folks are eating, we can work in ‘wheat equivalent’ to simplify our understanding of food demands, even though our peasants will be growing a somewhat wider range of crops in order to feed themselves.

Determining the wheat requirements is itself, however, a bit tricky. There are two approaches. The first is to assume that the nutritional needs of human beings haven’t changed that much over time – we are, biologically, mostly the same humans we’ve been for the last 10,000 years – and so using modern nutritional estimates, most often the World Health Organization (WHO) and Food and Agricultural Organization (FAO) statistics.⁷ The tricky this is these figures come in meaningfully higher than the figures we find in our sources and that makes sense – the WHO and FAO are suggesting ideal not minimum standards and it’s just very clear that ancient and medieval peasants did not survive on modern ideal nutritional standards. The alternative is to use figures derived from ancient or medieval sources, as for instance Paul Erdkamp does,⁸ though these can be tricky to use because they generally come not in calorie counts but in ancient units of dry measure, with all of the complications of back-filling to a caloric measure.

Now because I believe that the best solution to any situation in which there are two or more unsatisfactory standards is to develop a new standard, I am going to use the approach I use in my book project – which if you want updates on the progress of that project, patrons get monthly updates – which more or less splits the difference between Erdkamp’s ancient-derived figures (which struck me as too low, being inter alia below ancient figures we have for military rations) and modern figures from the FAO/WHO, etc. Because I am a Roman historian, I am going to do a lot of the background calculations of model here in Roman units (iugera rather than acres and – relevant here – modii rather than dry litres), but my split-the-difference approach gets us roughly the following food requirements, assuming all food intake is wheat (with the activity level and then the food requirement in rough annual kilograms of wheat (with the original calculation in modii in parenthesis) and then the estimates calories per day below):

So that gets us our rough totals for a modest minimum nutritional demand for each household. Again, I should note, while I am using the FAO figures as a guide for adjusting for age and gender, my calorie estimates here are generally around 10% less (or so) than the FAO figures because relatively poor farming laborers in the past do not seem to have eaten quite so well as the FAO or the WHO (or I) would like.

The other immediate major survival need the family has is clothing. Here we have no nutrition figures to anchor our estimates on and clothing demand is very clearly context sensitive. Households in cold climates will need more, but equally the question of minimum social expectations will differ substantially in the amount of expected coverage. Meanwhile, while human beings have a clear limit to how much food they can really eat, there is no such limit to how much clothing they can have, as demonstrated by all of our overstuffed closests, any one of which would have embarrassed your average rich peasant. Clothing demand thus could ‘expand to fill the space’ – the family could certainly consume whatever production was available.

But we’re interested in minimums. We’ve actually tackled this problem before. We get a suggestion of the absolute minimum for survival from a writer like Cato the Elder (De Ag. 59), well known to be a cruel and miserly master towards his enslaved workers, who recommends each worker get a new long tunic and cloak each year. That comes out to something in the neighborhood of 21,650cm² per year, per person. Roman soldiers seem to have been issued two sets of clothing per year, which might suggest that something a bit more normal for a well-off peasant household (from where Roman soldiers were recruited). Our peasants might be somewhat less than that: if we assume something like a single new complete set of clothing (a bit more ample than Cato’s allotment and every year, not every other) per year (I am going to calculate assuming Roman clothing – these values would need to go up for colder climates), we might figure something in the very rough neighborhood of c. 50,000cm² per adult, perhaps half as much per adolescent child and a quarter as much for very young children. That might suggest the total fabric need of the family as follows (expressed in square meters), along with a very rough estimate of the amount of raw fibers (wool, flax) required:

Of course there are going to be some other requirements. Our household already has land and a farmhouse (almost certainly inherited) but the farmhouse will demand regular cleaning and maintenance to remain livable. The family also needs heating and to a limited degree, lighting. For now, we’re going to assume our bare-bones subsistence family is generating its heating needs with wood fuel gathered from something like a common forest.⁹ And of course keep in mind that our calculations for food above are in ‘wheat equivalents’ but we understand they represent a wider range of foods including beans and other legumes, possibly vegetables out of a small garden, potentially a modest amount of meat from the keeping of a small number of animals¹⁰ and so on. These other foods might not be as land and labor efficient as wheat, so to represent their requirement for basic survival – enough protein and vitamins to not get sick and die – we might raise the wheat requirement by perhaps 10% over our raw calculation (which was derived from wheat rations that would have been supplemented by other foods in any case).

We also need to adjust for taxes. After all, even if these peasants only farm on land they own free and in the clear – neither sharecropping the land of a Big Man nor having to farm the land of the lord’s manor – chances are they still have to pay taxes. The amount of these taxes vary tremendously, based both on local productivity – the more productive the local land, the heavier the taxes can be – and the political economy. In societies where the large mass of small farmers wield real political power – the Roman Republic and some Greek poleis jump to mind – taxes are often kept somewhat low. By contrast, in societies where the peasantry is broadly disarmed and politically irrelevant, kings and temples can (and do) tax the peasantry down to the subsistence line. This is another topic for the book project, but I suspect one of the reasons Rome, and, to an extent, Macedon and the Greek poleis were relatively good at producing a lot of heavy infantry is that their political structures made it harder – not impossible, but harder – for elites to use taxes and rents to drain away all of the wealth of the peasantry, which in turn resulted in peasants with enough wealth to equip themselves as soldiers.

In any case, taxes on agricultural production were ubiquitous. Tax rates varied, but a tithe – a 10% tax on agricultural production, which the Romans called the decumae – is about as low as they go. So to survive, our peasants need to exceed their basic subsistence needs (which, you will recall, we’ve raised already by 10% to account for non-wheat foods) by enough to pay taxes and still not starve, so we need actual farming production to be high enough that 90% of it covers our subsistence needs to simulate a low taxation environment for our peasants. That produces an estimate of bare subsistence for each family that looks like this:

Man Can Not Live On Wheat Equivalent Alone

But of course, those figures are pretty close to bare minimum: again, families could (and did) slip below those figures from time to time, but they could not operate long-term much below them without starvation, health complications from exposure to the elements, or angering the taxman.

But what of the “a little bit more” beyond subsistence?

That, of course is a lot trickier, because we’re no longer dealing with basic necessities and minimums. One tool historians use to think about this is a “respectability basket” – a collection of goods that we suppose, based on our sources, added up to what was understood as a respectable lower-class or working-class living.¹¹ Generally scholars use these respectability baskets as a way of converting needs into currency figures which can then be assessed against the wages for labor.

But we don’t want to assess against labor, because our interests are peasants who (generally) have a bit more to offer than labor. Our peasants, after all, are defined not merely as laborers but also as possessing some rights – either outright ownership, or some sort of attachment – to land, which is to say capital. That puts them in a different economic space than the urban tradesman or unskilled laborer of the sort whose wages might be reported in our sources. As a result, we do not want to convert into money, which is limited direct usefulness for our peasants, but instead to try to stay in units of what they do produce: grain and fabric in our simplified model here.

What makes that tricky, of course, is that a lot of the things that we’re going to add for our respectability basket – fancier foods, wine, lamp oil and so on – are things that the household is not going to produce itself, but must acquire from others. Of course fundamentally the household has to acquire these goods in exchange for the things it can give, sell or exchange, which are agricultural goods (here still simplified to ‘grain’), textiles or unskilled labor.¹² Our peasants might, for instance, sell their surplus grain, or the surplus fabric made by their spinners and weavers, to afford things they cannot produce themselves, but I should note ‘market exchange’ is not the only way they could do this. They might also exchange with other households, often on credit and favors (rather than barter) and may also be involved in vertical systems of banqueting and gift-exchange with the Big Man, all of which provide non-market ways to effectively exchange grain, textiles or labor for things they cannot produce themselves.

What can help us here in thinking about how much our peasants would need to produce to basically satisfy all of the ‘optionals’ of their material needs is looking at the relationship between the costs of a ‘bare bones’ subsistence basket that produces a minimum caloric value and the fancier ‘respectability basket’ which reaches the same caloric value (along with some material comforts) in a substantially more expensive but more pleasant way. The ratios between the two ‘baskets’ can give us some sense of how high above ‘subsistence’ the line for ‘comfort’ was.

So, for instance Robert C. Allen¹³ proposes an early modern Northern European respectability basket consisting of bread, beans/peas, meat, butter, cheese, eggs, beer, soap, linen, candles, lamp oil and fuel. Our bare subsistence model has effectively already accounted for bread, beans/peas, linen and fuel which represent 46.7% of the full basket’s spending, along with (remember our 10%-for-other-foods – this is that) 5 out of the 26kg of meat and 3 of the 5.2kg of the butter, along with half of the candles and lamp oil (as these are included in Allen’s own bare bones model), which is another 9% of total respectability basket spending (out of the 26% spend on the full basket’s meat, butter, candles and oil). All of which is to say we might imagine our bare bones subsistence total of food and fabric represents something like 56% of the respectability basket following Allen’s Northern European respectability basket (put another way, to achieve respectability, our household needs to produce something like 178% of its bare subsistence production).

That said, Allen’s early modern basket is useful to think with but tricky in one immediate regard which is how small a portion of the budget is in fabric – five square meters of linen make up just 5.3% of the total – a product of the substantially greater textile productivity of a post-spinning-wheel,¹⁴ post-horizontal-frame-loom¹⁵ making fabric a lot less labor intensive than it was for most of the ancient and medieval peasants we’re focused on. Because the adoption of those technologies increased textile worker productivity, potentially several times over, they create a ‘discontinuity’ in the structure of household budgets we need to be wary of. Now Allen does also run numbers for antiquity, but I think we’re better off relying on Walter Scheidel’s approach to the same math a year later, tweaking some of the numbers.¹⁶

And here Scheidel does us a remarkable favor: while Allen calculated a ‘bare bones’ early modern subsistence basket, he didn’t detail its monetary cost, but Scheidel makes his two baskets (respectability and bare bones subsistence) directly comparable not only in calories but also in direct cost, which makes our task here a bit easier. Scheidel breaks down ancient price data into two periods, 1-160AD and 190-270AD (there was a substantial bought of inflation between them) and he has to adjust for goods which we know might be consumed but for which we simply do not have prices, like cheese; I am using these “adjusted totals” here. In period 1, respectability’s cost was 249 drachma at the lower end, compared to a bare bones cost of 112 (44%); in period two the respectability cost was 535 compared to a bare bones cost of 266 (49%).¹⁷

Now as you can tell there is significant chronological and regional variation in these figures: what defines respectability and what it costs varies place to place. However, for the sake of our model we can make a decent ballpark assumption, going off of these figures that the complete ‘respectability’ package of material comforts reflected achieving something like double the basic bare-bones subsistence requirements. That doesn’t mean the family is necessarily eating twice as much food – or even wearing twice as many clothes – but that they might be producing double subsistence and then trading away (market sale, gift exchange, etc.) the surplus in order to acquire things they cannot produce themselves.

Assuming that ‘roughly double,’ we can now, at last get a sense of the material needs for our three model peasant households:

Now in practice of course, our peasants, like everyone else are going to experience diminishing marginal returns as they push over subsistence: each extra bit of production is going to buy a bit less comfort, so they may decide to stop doing more labor well before reaching the full respectability basket, but the overall picture here gives us a decent sense of what the upper-end of ‘aspirational’ for our peasant families would be.

So we now have our model families with their model members and thus we have a sense of how much labor the family has and what the family needs from its labor. We are at last ready, in the next few weeks, to explore how these families might deploy the labor they have to try to meet those needs and what the implications would be for how much they work and how they live.

https://acoup.blog/2025/08/22/collections-life-work-death-and-the-peasant-part-iva-subsistence-and-a-little-more/?utm_source=rss&utm_medium=rss&utm_campaign=collections-life-work-death-and-the-peasant-part-iva-subsistence-and-a-little-more

https://acoup.blog/?p=31013