schema

Posted by Bret Devereaux

https://acoup.blog/2025/09/19/fireside-friday-september-19-2025-on-the-use-and-abuse-of-malthus/?utm_source=rss&utm_medium=rss&utm_campaign=fireside-friday-september-19-2025-on-the-use-and-abuse-of-malthus

https://acoup.blog/?p=31446

Hey folks, Fireside this week! Next week we should be back to start looking at the other half of labor in the peasant household, everything that isn’t agriculture. Also, here are some cats:

For this week’s musing, I want to address something that comes up frequently in the comments, particularly any time we discuss agriculture: the ‘Mathusian trap.’ Now of course to a degree the irony of addressing it here is that it will still come up in the comments because future folks raising the point won’t see this first, but at least it’ll be written somewhere that I can refer to.

To begin, in brief, the idea of a Malthusian trap derives from the work of Thomas Robert Malthus (1766-1834) and his work, An Essay on the Principle of Population (1798). In essence the argument goes as follows (in a greatly simplified form): if it is the case that the primary resources to sustain a population grow only linearly, but population grows exponentially, then it must be the case that population will, relatively swiftly, approach the limits of resources, leading to general poverty and immiseration, which in turn provide the check that limits population growth.

As an exercise in logic Malthus’ point is inescapable: if you accept his premises and run the experiment long enough you must reach his conclusion. In short, given an exponentially growing population and given resources that only grow linearly and given an infinite amount of time, you have to reach the Malthusian ‘trap’ of general poverty and population checked only by misery. So far as that goes, fine.

The problem is assuming any of those premises were generally correct in any given point in history.

I find this comes up whenever I point out that certain social and political structures – the Roman Empire most notably – seem to have produced better economic conditions for the broad population or that other structures – Sparta, say – produced worse ones: someone rolls in to insist that because the Malthusian trap is inevitable the set of structures doesn’t matter, as a better society will just produce an equally miserable outcome shortly thereafter with a larger population. And then I response that Malthus is not actually always very useful for understanding these interactions, which prompts disbelief because – look just above – his logic is airtight given his premises and his premises are at least intuitive.

Because here’s the thing: Malthus was very definitely and obviously wrong. Malthus was writing as Britain (where he wrote) was beginning to experience the initial phases of the demographic transition, which begins with a period of very rapid population growth as mortality declines but birth rates remain mostly constant. Malthus generalizes those trends, but of course those trends do not generalize; to date they have happened exactly once in every society where they have occurred. Instead of running out of primary resources, world population is expected to peak later this century around 10.5 billion and we already can grow enough food for 10.5 billion people. The next key primary resource is energy and progress on renewable energy sources is remarkable; at this point it seems very likely that we will have more power-per-person available at that 10.5 billion person peak than we do today. Living standards won’t fall, they’ll continue to rise, assuming we avoid doing something remarkably foolish like a nuclear war. Even climate change – which is a very real problem – will only slow the rate of improvement under most projections, rather than result in an actual decline.

So while Malthus’ logic is ironclad and his premises are intuitive, as a matter of fact and reality he was wrong. Usefully wrong, but wrong. The question becomes why he was wrong. And the answer is that basically all of his premises are at least a little wrong.

The first, as we’ve noted, is that Malthus is extrapolating out a rate of population growth based on an unusual period: the beginning of rapid growth in the second stage of the demographic transition – and then he is extrapolating that pattern out infinitely in time in every direction. And that is a mistake, albeit an easy one to make: to assume that the question of population under agrarian production is an effectively infinite running simulation which has already (or very soon will) reach stability.

Here’s the thing (this is a very rough chronology): human beings (Homo sapiens) appeared about 300,000 years ago. We started leaving the cradle of Africa around 130,000 years ago, more or less and only filled out all of the major continents about 15,000 years ago. The earliest beginnings of agriculture are perhaps 20,000 years old or so, but agriculture reached most places in the form Malthus would recognize it much later. Farming got to Britain about 6,500 years ago. Complex states with large urban populations are 5,000 or so years old. Large sections of the American Great Plains and the Eurasian Steppe were grazing land until the last 150 years.

In short, it is easy to assume, because human lives are so short, that the way we have been living – agrarian societies – are already effectively ‘infinitely’ old. But we’re not! Assuming we do not nuke ourselves or cook the planet, in the long view pre-industrial agriculture will look like a very brief period of comparatively rapid development between hundreds of thousands of years of living as hunter-gathers and whatever comes after now. To Malthus, whose history could stretch no further back than the Romans and no further forward than the year in which he wrote, his kind of society seemed to have existed forever. It seemed that way to the Romans too. But we’re in a position to see both before agrarian economies and also after them; we’re not smarter, we just have the luck of a modestly better vantage.¹

In short, while we might assume that given infinite time, exponential population growth will outpace any gains made to production but you shouldn’t assume infinite time because we are actually dealing with a very finite amount of time. Farmers, whose demographics concern us here, appear around 20,000 years ago and begin filling up the Earth, spreading out to bring new farmland under the plow (displacing, often violently, lower population density societies as they did so) and that process was arguably nearing completion but not yet complete when the second agricultural and first industrial revolutions fundamentally changed the basis of production. As we’ve discussed, estimates of global population in the deep past are deeply fraught, but there is general agreement that population globally has increased more or less continuously since the advent of farming; it never stalled out at any point. In short, the Malthusian long run is so long that it almost doesn’t matter.

But if we limit our view to a specific region or society, that changes things. We certainly do see, if not Malthusian traps, what we might term ‘Malthusian interactions’ apparent in history. Rising population density and trade connectivity help spread disease, which lead to major downward corrections in population like the Antonine Plague, the Plague of Justinian, the Black Death and the diseases of the Columbian Exchange. Notably though, these sudden downward corrections are at best only somewhat connected to population growth and resource scarcity: lower nutrition may play a role, but travel, trade lanes, high density cities and exposure to novel pathogens seems to play a larger role. It’s not clear that something like the Black Death would have been dramatically less lethal if the European population were 10 or 15% less; it seems quite clear the diseases of the Columbian exchange cared very little for how well fed the populations they devastated were. Still, we see the outline of what Malthus might expect: downward pressure on wages before the population discontinuity and often upward pressure afterwards (most clearly visible with the Black Death in Europe).

So does Malthus rule the ‘small print’ as it were? Perhaps, but not always. For one, it is possible, even in the pre-modern world, to realize meaningful per capita gains in productivity due to new production methods like new farming techniques. It is also possible for greater connectivity through trade to enable greater production by comparative advantage. It is also possible for capital accumulation in things like mills or draft animals to generate meaningful increases in production. And of course some political and economic regimes may be more or less onerous for the peasantry. Any of these things moving in the right direction can effectively create some ‘headroom’ in production and resources. Some of that ‘headroom’ is going to get extracted by the tiny number of elites at the top of these societies, but potentially not all of it.

This is what I often refer to as a society moving between equilibria (a phrasing not original to me), from a state condition of lower production (a low equilibrium) to a stable condition of higher production (a high equilibrium).

Now in the long run when just thinking about food production, the Malthusian interaction ought to catch up with us in the long run. The population increases, but the available land supply cannot keep pace – new lands brought under the plow are more marginal than old lands and so on – and so the surplus food per person steadily declines as the population grows until we’re back where we started. Except there are two problems here.

The first is that can take a long time even in a single society, region or state because even under ideal nutrition standards, these societies increase in population slowly compared to the rapid sort of exponential growth Malthus was beginning to see in the 1700s. It can take so long that exogenous shocks – invasion, plague, or new technology enabling a new burst of ‘headroom’ – arrive before the ceiling is reached and growth stops. Indeed, given the trajectory of pre-modern global population, that last factor must have happened quite a lot, since even the population of long-settled areas never quite stabilizes in the long term.

All of which is to say, in the time frame that matters – the time scale of states, regimes, economic systems and so on, measured in centuries not millennia – some amount of new ‘headroom’ might be durable and indeed we know it ended up being so, lasting long enough for us to get deep enough into the demographic transition that we could put Malthus away almost entirely.

The second thing to note is that not all material comforts are immediately related to survival and birth rates. To take our same society where some innovation has enabled increased production: the population rises, but no new land enters cultivation. That creates a segment of the population who can be fed, but who need not be farmers: they can do other things. Of course in actual pre-modern societies, it is most the elite who decide what other things these fellows do and many of those things (warfare, monumental construction, providing elite extravagance) do very little for the common folks.

But not always. Sometimes that new urban population is going to make stuff, stuff which might flow to consumers outside of the elite. We certainly seem to see this with sites of large-scale production of things like Roman coarseware pottery. Or, to take something from my own areas, it is hard not to notice that the amount of worked metal we imagine to be available for regular people for things like tools seems to rise as a function of time. Late medieval peasants do seem to have more stuff than early medieval or Roman peasants in a lot of cases. Wages – either measured in silver or as a ‘grain wage’ – may not be going up, but it sure seems like some things end up getting more affordable because there are more people making them.

And of course some of that elite investment might also be generally useful. Of course as a Roman historian, the examples of things like public baths and aqueducts, which provided services available not merely to the wealthy but also the urban poor, spring immediately to mind. And so even if the amount of grain available per person has stayed the same, the number of non-farmers as a percentage of the society has increased, making non-grain amenities easier for a society to supply. And naturally, social organization is going to play a huge role in the degree to which that added production does or does not get converted into amenities for non-elites.

In short it is possible for improvements to provide quality of life improvements even if a new Malthusian ceiling is reached. It is the difference between getting 3,000 calories in a wood-and-plaster building with a terracotta roof, a good collection of coarseware pottery and clean water from an aqueduct versus getting 3,000 calories in a wood-and-mud hut with a thatched roof, no pottery at all and having to pump water at the local well. In a basic Malthusian analysis, these societies are the same, but the lived experience is going to be meaningfully different.

Notionally, of course, you might argue that if population continued to rise we’d eventually reach the end of those fixed resources too: we’d run out of clay and metal ores and fresh water sources and so on, except that of course there are 8.2 billion of us and we haven’t yet managed to run out – or even be seriously constrained – by any of those things. We haven’t even managed to run out of oil or coal and again, at the rate at which renewable energy technology is advancing, it looks like we may never run out of oil, so much as it just won’t be worth anyone’s time pulling the stuff out of the ground.²

None of which is to say that Malthus is useless. Malthusian interactions do occur historically. But they do not always occur because the sweep of history is not infinitely wrong and developments which produce significant carrying capacity ‘headroom’ actually happen, on balance, somewhat faster than societies manage to reach the limit of that capacity.

On to Recommendations:

First off, the public classics project Peopling the Past has turned five! Congratulations to them. Peopling the Part runs both a blog and a podcast both highlighting the ways that scholars, especially early career scholars, study people in the (relatively deep) past, with an emphasis on highlighting interesting work and the methods it uses. It’s a great project to follow if you want a sense of how we know things about the past and the sort of work we continue to do to understand more, with an especially strong focus on archaeology.

Meanwhile over on YouTube and coinciding a bit with our discussion of Malthus, Angela Collier has a video on why “dyson spheres are a joke,“³ in the sense that they were quite literally proposed by Freeman J. Dyson as a joke, a deliberate ‘send up’ of the work of some of his colleagues he found silly, rather than ever being a serious suggestion for science fiction super-structures.

Where this cuts across our topic is that Dyson, writing in 1960, explicitly cites “Malthusian pressures” as what would force the construction of such a structure and it serves as a useful reminder that until well into the 1980s and 1990s, there were quite a lot of ‘overpopulation’ concerns and it was common to imagine the future as involving extreme overpopulation and resource scarcity. I wouldn’t accuse Dyson of this view (he is, as noted, writing a paper as satire), but I think it is notable that these panics continued substantially on the basis of assumptions that the demographic transition – which was already pretty clearly causing population growth in Europe to begin to slow significantly by the 1950s and 1960s – was, in effect, a ‘white people only’ phenomenon, fueling often very racially inflected fears about non-white overpopulation. You can see this sort of racist-alarmist-panic pretty clearly in Paul Ehrlich’s The Population Bomb (1968), appropriately skewered in the If Books Could Kill episode on it.

Of course as noted is that what actually happened is that it turns out the demographic transition does not care about race or racists and happens to basically all societies as they grow wealthier and more educated – indeed, it has often happened faster in countries arriving to affluence late – with the result that it now appears that the ‘population bomb’ will never happen.

For this week’s book recommendation, I am going to recommend Rebecca F. Kennedy, C. Sydnor Roy and Max L. Goldman, Race and Ethnicity in the Classical World: An Anthology of Primary Sources in Translation (2013). Students often ask questions like ‘what did the Greeks and Romans think about race?’ and the complicated answer is they thought a lot of things. That can come as a surprise to moderns, as we’re really used to the cultural hegemony of ‘scientific racism’ and the reactions against it. But it is in fact somewhat unusual that a single theory of race – as unfounded in actual reality as all of the others – is so dominant globally as an ideology that people either hold or push against. Until the modern period, you were far more likely to find a confusing melange of conflicting theories (advanced with varying degrees of knowledge or ignorance of distant peoples) all presented more or less equally. Consequently, the Greeks and Romans didn’t think one thing about race, but had many conflicting ideas about where different peoples fit and why.

That makes an anthology of sources in translation an ideal way to present the topic and that is what Kennedy, Roy and Goldman have done here. This is very much what it says ‘on the tin’ – a collection of translated primary sources; the editorial commentary is kept quite minimal and the sources do largely speak for themselves. The authors set out roughly 200 different passages – some quite short, some fairly long – from ancient Greek and Roman writers that touch on the topic of race or ethnicity. Those passages are split in two ways: the book is divided into two sections, the first covering theories and the second covering regions. In the first section, the reader is given examples of some of the dominant strains of how Greeks and Romans thought about different peoples and what made them different – genealogical theories, environmental theories (people become different because they are molded by different places), cultural models and so on. The approach is a brilliant way to hammer home to the reader the lack of any single hegemonic model of ‘otherness’ in this period, while also exposing them to the most frequent motifs with which the ancients thought about different peoples.

Then the back two-thirds of the book proceed in a series of chapters covering specific regions. Presenting, say, almost 20 passages on the peoples of ‘barbarian’ Europe (Gaul, Germany, Britain) together also helps the reader get a real sense of both the range of ways specific regions were imagined but also common tropes, motifs and stereotypes that were common among ancient authors.

The translations in the volume are invariably top-rate, easy to read while being faithful to the original text. The editorial notes are brief but can help put passages in the context of the larger works they come from. The book also features reprints of a series of maps showing the world as described by the Greeks and Romans, a useful way to remember how approximate their understanding of distant places and their geographic relations could be. Overall, the volume is useful as a reference text – when you really need to find the right passage to demonstrate a particular motif, stereotype or theory of difference – but is going to be most valuable to the student of antiquity who wants to begin to really get a handle on the varied ways the Greeks and Romans understood ethnic and cultural difference.

https://acoup.blog/2025/09/19/fireside-friday-september-19-2025-on-the-use-and-abuse-of-malthus/?utm_source=rss&utm_medium=rss&utm_campaign=fireside-friday-september-19-2025-on-the-use-and-abuse-of-malthus

https://acoup.blog/?p=31446