Did seasonality drive the invention of agriculture?

Andrea Matranga has a job market paper (pdf) which is speculative but interesting:

During the Neolithic Revolution, seven populations independently invented agriculture. In this paper, I argue that this innovation was a response to a large increase in climactic seasonality. Hunter-gathers in the most affected regions became sedentary in order to store food and smooth their consumption. I present a model capturing the key incentives for adopting agriculture, and I test the resulting predictions against a global panel dataset of climate conditions and Neolithic adoption dates. I find that invention and adoption were both systematically more likely in places with higher seasonality. The findings of this paper imply that seasonality patterns 10,000 years ago were amongst the major determinants of the present day global distribution of crop productivities, ethnic groups, cultural traditions, and political institutions.

Here is his home page.


One factor to consider is how global explanations for the recent emergence of agriculture relate to the great filter.

The authors suggest that the earth's shallow axial tilt, low orbital eccentricity, combined with the earth's axial precession during the last ice age worked to moderate seasonal extremes.

We can speculate that a range of planetary factors like the above may prevent even those species cognitively capable of developing agriculture and civilization from progressing.

In fact, other animals farm! Humans are just the only vertebrates that do so. http://en.wikipedia.org/wiki/Ant%E2%80%93fungus_mutualism

The more or less simultaneous emergence of agriculture at widely separated sites strongly suggests that there was some common underlying ecological driver.

An Ice Age ended.

The Holocene nominally started about 10,000BC, and the warming started maybe ten thousand years before that. By assimilating various bits of pop-science, my guess about events is:

* Long ago humans looked more or less modern, but were not mentally our equals.
* Some fairly rapid improvements occurred between 100-50 thousand years ago -- this was firmly during the last ice age. Around the same time, clever humans spread out of Africa.
* Then the ice age ended, for the first time ever, rapid climate change was happening while many parts of the world were populated by new-model clever humans.
* In a few of these places, the response to the climate changes was to invent agriculture.
* Agriculture spread from those nuclei.

I don't know what the magic "right" kind of climate change was. Perhaps this paper explains it.

"agriculture was independently invented at least seven times within a 7,000 year period (the Neolithic Revolution)."

I have a problem with the assumption that something that happened over a 7,000 year period is considered 'simultaneous'. A very large amount of travel and migration can happen over a 7,000 year period. Our indications of what happened going that far back (12,000 years ago) are guesses at best.

Of course one thing you never see mentioned in any of these discussions on the invention of agriculture - not in any supposedly liberal "politically correct" Jared Diamond book, not in this guy's paper, not anywhere is the absolutely critically important role of affordable labor in the development of economically profitable agricultural systems. Due to the lack of strict borders during neolithic times labor was able to flow into different regions therefore limiting the rise of wages and allowing agriculture to become an effective alternative to hunter-gathering. Today this lesson has been buried by establishment thought-leaders and we face a future of stagnation as labor monopolists crowd out competition through draconian closed border policies.

0/10 try harder

You need to start referring to your immigrant girlfriends more often

Doing the ’jobs Americans won’t do.

Kapow! A pile of JAMRC ashes is all that's left.

Thats more along the lines of Ray Lopez's beat

It's wonderfully convenient for his hypothesis that the invention of agriculture in PNG somehow doesn't count. I take it that somewhere near-equitorial doesn't have an especially seasonal climate, though I dare say that there's a bit of a monsoon affect.

P.S. What about seasonal places where agriculture wasn't invented?

Yeah, it seems like they left it out. But they discuss rainy seasons as a spur to agriculture in the tropics, so I'm not sure it would change much.

That depends on whether the rainy season is a big thing in the PNG highlands. I have some atavistic notion that monsoon and the like are mostly coastal phenomena, but I have no idea if that intuition is correct.

"...I argue that this innovation was a response to a large increase in climactic seasonality."

I think he probably means climatic seasonality, but I don't have a PhD, so what do I know?

Huhn. At first I thought you were wrong - so I looked it up. Turns out you are correct.

Climactic, derived from climax

Climatic, derived from climate


this explains why all of us lloyds were born in december

There's zero argument among people who know anything about the topic that climate was a major driver. But climactic seasonality? That one only works if you include a lot of caveats and have already conditioned on much more important variables. This is because seasonal fluctuations for millennia were a big part of the reason why agriculture *did not* take off:

"The last glacial period was arid and extremely variable compared to the Holocene. [...] The intense variability of the last glacial carries right down to the limits of the nearly 10-year resolution of the ice core data. [...] [Ice core data] shows millennial- and submillennial-scale temperature fluctuations from 60-18 thousand years ago with an amplitude of about 8°C, compared to fluctuations about °2C in the Holocene epoch." (Boyd and Richerson, The Origin and Evolution of Cultures, pp.340-42.)

Climate was obviously important. But two major things happened in the Holocene: It got warmer, and climatic fluctuations became much less pronounced than they used to be. Less pronounced, not more. Here's what the author of the paper writes in his introduction:

"I propose a new theory for the Neolithic Revolution, construct a model capturing its intuition, and test the resulting implications against a panel dataset of climate and adoption. I argue that the invention of agriculture was triggered by a large and exogenous increase in climatic seasonality, which made it hard for hunter-gatherers to survive during part of the year."

Yes, it was certainly a big increase that we went from an amplitude of about 8°C to an amplitude of about 2°C. I stopped reading when I read that sentence, thinking 'yet another paper written by a clueless economist who doesn't know nearly enough about the topic to be coming up with models explaining anything.' I may be wrong, but these people are too common for me to risk spending time on a paper like that one. If you're curious, some more quotes from Boyd and Richerson's coverage of this topic is available here. Those two are certainly very far from clueless. They argue that low productivity and huge fluctuations made agriculture impossible before the Holocene, and that it was partly the smaller fluctuations which eventually made agriculture possible. Which would be a very different conclusion from the one in the paper. In the context of smaller fluctuations and higher average temperature/productivity perhaps areas with slightly larger fluctuations than average were more conducive to agriculture. But that's a very different conclusion from the one the author seems to argue in favour of.

Doubtful, and it's not about who invented farming or when, which we can't know, it's about where and when it developed

From Wikipedia on the History of Africa:

"The Fertile Crescent of Western Asia, Egypt and India were sites of the earliest planned sowing and harvesting of plants that had previously been gathered in the wild. Independent development of agriculture occurred in northern and southern China, Africa's Sahel, New Guinea, parts of India and several regions of the Americas.[3]"

I'd say that Northern China sounds like the place with the most seasonality on that list.

In general, agriculture developed at lower latitudes and slowly moved north, which is why ancient power centers were more southerly than today. E.g., Babylon was a bigger deal relative to the rest of the world in its day than Baghdad is today.

It's interesting, isn't it, that agriculture required a level of cognitive ability that apparently all of contemporary Homo sapiens happened to possess.

The fact that agriculture was developed by diverse populations in the Quaternary that had already diverged some 40,000 to 70,000 years ago suggests that humans had the cognitive equipment to transition away from hunter and gatherer economies a very long time ago.

Humans would have noticed that plants grew from seed long before they started farming. They may even have practiced gardening of fast-growing plants for millennia. In fact there's evidence that they cultivated the bottle gourd across much the (non Arctic) world well before the Americas were settled.

Agriculture was imported into India from elsewhere (wheat from the Middle East; rice from East Asia). This maybe connected with the arrival of the Dravidian speakers from the west and the Munda speakers from the east.

I’d say that Northern China sounds like the place with the most seasonality on that list.

I think it would be more about how seasonality contributes to plant growth cycles of actual plants you can actually store and eat and feed to animals.

E.g. ancient Russia had a lot of seasonality (continental climate - swings from hot to cold), but the only thing that is practical for hunter gatherers to do make a living there is kill horses or other big dumb animals.

So maybe people will domesticate a horse eventually or something, but you're not going to get much of a build to agriculture out of it.

Similarly in Mesolithic Europe or the Pacific Northwest, there's a lot of seasonality around fish harvesting, but that doesn't get you to plant agriculture or animal domestication.

In general, agriculture developed at lower latitudes and slowly moved north

Not really so much on the North China plain. In India, agriculture moved south over time (and Southerners there are the smartest and best farmers I think). I'm unsure about Native American agriculture.

Agriculture in West Eurasia is unusual in terms of the best domesticates (plant and animal) being far from the most well watered regions with the best soil (although they are cold and have big seasonal swings in day length). This isn't a general rule though - not the case or weakly the case in China and Mesoamerica.

Seasonality (i.e., a short growing season due to northern latitudes) was a massive problem for would be agriculturalists. One of Jared Diamond's points is that most agricultural staples emerged at lower latitudes (e.g., corn/maize in Mexico) and had to be laboriously adapted to the shorter growing seasons of higher latitudes. For example, when the Pilgrims landed in Massachusetts in 1620 and Squanto advised them on corn-growing techniques, corn had only been successfully grown that far north by Indians for about two generations. The low population density of North America in 1491 relative to Meso-America and the Andes had a lot to do with the shorter growing season in what's now the United States.

I tend towards the Carl O. Sauer explanation: It was an accident. It was reluctantly adopted.
As Sauer wrote, people on the verge of starvation are not going to go through the fits and starts that accompany the development of agriculture. So it had to be a real well fed group that started horticulture (small gardens) that eventually blossomed into agriculture. Then when you look in the work to leisure ratio, hunter-gathers have a much much better life than agricultural societies. So people didn't adopt agriculture because the wanted to, they adopted it because they had to; either forced into it (slavery) or simply because the pastoral pathways were blocked by hostile forces.

Or large game was unavailable due to over-hunting and the disruptions of the climate at the end of the Ice Age. The Neolithic began with an enormous episode of global warming.

Your quality of life argument applies only at the Malthusian limit. A group of hunter-gatherers who have an opportunity to add some horticulture to their lifestyle would see the nice garden greens as a marginal improvement to the lifestyle.

Growing greens won't tie them down as true farmers; you can grow a lot of leaf veggies and even some root vegetables in the course of a couple months.

I mean there is a sequence of incremental steps, each rational under some common conditions, which take you from being a hunter-gatherer to being a peasant farmer. Note also, that at the Malthusian, something must be limiting population. So some groups might have taken up (incremental steps towards) farming when the difficulty of finding food the old way had risen high enough to balance out the costs of farming.

Try actually hunting and gathering, compared to farming, and see how that argument stands up. A domestic pig is guaranteed bacon. Wild pig or deer? Not so much.

I do not see why you need a well fed group to try it. People observed that their seeds grew the following year. Dogs were domesticated long before agriculture, other animals are not that far off. Gardens can be tended while the kids are napping and the men are out hunting.

But what do you need in order to feed domesticated livestock? Grain. Where do you get that from, given that agriculture is not invented yet? Hunting vs. farming pigs today has nothing to do with the ease of hunting in prehistoric times.

I took one look at Matranga's photo on his website and dismissed him as a pseudo-retro conformist trying to dress like he's in a Bertolucci movie. Then I read his CV and found that he is actually an Italian and went to university in Milan. More importantly, the title of one of his working papers quotes Monty Python. I say hire him.

Where they did it is probably less important than why.

The main problem early humans faced was other early humans trying to kill them. A good way to solve that problem was to kill the other early humans first. A good way to kill the other early humans first was to develop agriculture to support an army to kill the other early humans before they developed agriculture and sent their own armies. A good way to develop agriculture was to enslave lots of other early humans and put them to work growing stuff.

And that paradigm pretty much takes us right up to the modern era.

No one just stands and passively waits to be killed, each group is able to defend itself. Killing the other group is not feasible because an invasion will be resisted and the invaders may get killed themselves. Violence is expensive.

The reason violence decreases is the law of association; work performed under the division of labor is more productive than working in isolation. People gradually realize that the activities of both sides will be more profitable, if they cooperate and trade instead of fighting.

Furthermore, no one at the time could even envision "supporting an army", there were very few people at the time and no such thing could possibly exist. Armies were formed long after the development of agriculture with much higher population numbers. This flinstonization of such topics and poorly thought out comments like yours make me sick.

I thought it was beer. All of the first crops were beer ingredients. It was just a fluke that other uses were found for the crops.

It's not unlikely that yeast was manipulated first for beer then for bread, but growing grain for beer then discovering it could be eaten seems a bit dense.


This is the beer guy...though seeing his web page makes me think he is a little kooky.

Squanto? Slavery? Seriously guys?

There are so many issues with this paper/some of these comments that I don't even know where to begin... It may be instructive to at least browse the abstracts of the modern archaeological/anthropological literature to understand the current conversations surrounding the adoption of agriculture (Not Jared Diamond, please, please stop talking about Jared Diamond...). I'll give you some hints: a) it isn't about seasonality, and b) whoever wrote that the main problem early humans faced was murder by other humans.... I don't even know what to do with that.

First of all, most archaeologists now accept the fact that the 'Neolithic Revolution' was - to adapt a phrase from Sally McBrearty's work on earlier such transitions - "A Revolution that wasn't." Domestication of crops, like animals, goes through several stages from harvesting/stewardship through full blown Monsanto - over looooong periods of time (the longue durée). A lot of what these comments and this paper are assuming -incorrectly- is that agriculture was an end game, some sort of pinnacle of energy catchment (Caused by Malthusian pressures of course!!). It wasn't. They weren't just waiting until the climate was seasonal so they could plant corn. The development of cultures do not follow a path to modern, industrial, agricultural, 'enlightened' societies. We ditched these theories 50 years ago. Prehistoric communities around the globe experimented with different stages of domestication, as with many other technologies, over thousands of years. Often dropping them, forgetting them and possibly retrying them a few generations later. Or not. History, despite what hindsight might imply, is not a linear process. The first few 'farmers', much like the first few city dwellers, would have had a horrible experience. Hunting/gathering/foraging takes very little energy or time. Agriculture, on the other hand, is labor-intensive and is fraught with problems (drought/floods, pests, etc.). Why would they continue? Good question.

There are so many variables - yes, even intangibles - that would have gone into the "invention" (lol) of agriculture. Attempting to explain any phenomenon in terms of a singular causality is, obviously, ridiculous. But even more so here because, despite J***d D*****d, the conversations happening in the discipline moved beyond environmental determinism several decades ago. Yes people interact with their physical environments. Yes cultures change within the context of changing environmental conditions. Do people simply bend to the will of the weather? No.

I have neither the space nor will to continue with this comment. Again, for more comprehensive understanding of crop domestication, please do not read an economics paper that is clearly pandering to an audience that is swept up in the current climate-determinism fad (I'm not saying climate change is not real and our fault). And, leave Jared Diamond at the airport bookstore where he belongs. Read some of the actual research - it's easy, just google.scholar it.

‘yet another paper written by a clueless economist who doesn't know nearly enough about the topic to be coming up with models explaining anything.’ Indeed.

whoever wrote that the main problem early humans faced was murder by other humans…. I don’t even know what to do with that.

Probably the best thing to do is to learn the statistics on homicide rates in early humans, and study what is known of the military history of early humanity.

Hunting/gathering/foraging takes very little energy or time. Agriculture, on the other hand, is labor-intensive and is fraught with problems (drought/floods, pests, etc.). Why would they continue?

The possibility of being killed or enslaved is an effective incentive.

The evidence is actually pretty unambiguous on this. There's still some "noble savage" promoters out there, but they're not supported by the evidence. You can find some isolated societies with lower homicide rates, but they're the exception.


"Several archaeologists and anthropologists now argue that violence was much more pervasive in hunter-gatherer society than in more recent eras. From the
!Kung in the Kalahari to the Inuit in the Arctic and the aborigines in Australia, two-thirds of modern hunter-gatherers are in a state of almost constant tribal warfare, and nearly 90% go to war at least once a year. War is a big word for dawn raids, skirmishes and lots of posturing, but death rates are high—usually around 25-30% of adult males die from homicide. The warfare death rate of 0.5% of the population per year that Lawrence Keeley of the University of Illinois calculates as typical of hunter-gatherer societies would equate to 2 billion people dying during the 20th century.

At first, anthropologists were inclined to think this a modern pathology. But it is increasingly looking as if it is the natural state. Richard Wrangham of Harvard University says that chimpanzees and human beings are the only animals in which males engage in co-operative and systematic homicidal raids. The death rate is similar in the two species. Steven LeBlanc, also of Harvard, says Rousseauian wishful thinking has led academics to overlook evidence of constant violence."

Hmm. Interesting that - despite my suggestions above - you went straight to the economist (an article one that quotes J***d D*****d within the first 2 paragraphs no less). Again, perhaps, it would be instructive to google some actual articles by actual archaeologists. That is probably the best thing to do.

I'm sorry, but no, the evidence is certainly not 'unambiguous' in favor of the war hypothesis. As most of us that are doing the research on these topics will tell you. The researchers cited in this article are scholars who, many other archaeologists feel, have sadly clung too tightly to their views of war as the main catalyst for pretty much everything. They are otherwise good researchers who took the ride on the pendulum a bit too far from earlier ideas. Prof. LeBlanc has done great work on the art and population movements of the Mimbres culture in southwest US. However once you start getting ideas from him that all preliterate societies were the same and would have the same reaction to 'stresses' (his favorites are drought and human induced environmental degradation) you start to understand that he just wants it to be true too badly. Modern archaeological research has been highlighting the dynamic responses to stresses human societies have developed time and time again. Not just turning to cattle raids. The author also could have included the work by the Posanszys in Peru and also Chagnon the most (in)famous (mostly for purchasing weapons for his subjects and inciting them to fight neighboring tribes. Well, that and pathologically taking advantage of local women sexually). But these researchers, as far as war goes, are on the fringe. Also many researchers like David Watts at Yale, are doing pretty well at fighting back against the old ideas of Cimps=war Bonobo =sex (our two closest ancestors representing the two sides of our id).

This article is clearly not written by someone that is current on the conversations happening within the archaeological/anthropological disciplines (well, at least for the past 10 years or so).

Look, no one here is arguing for the noble savage trope that was popular in the sixties, and to a lesser extent, the early to mid seventies. Also, it is pretty clear that we are currently living in the most peaceful time in our history. But the idea that war and the fear of being enslaved (or the environment, for that matter) were the main causes for the widespread adoption of agriculture has almost no support within the disciple. They are simply re-imaginings of the old 'conscription' theory. Simple, reductionist understandings of how technologies emerge and cultures develop will get us nowhere.

While I agree that the "Nobel Savage"" is romantic BS, the notion that we can calculate murder rates for people who lived millennia ago with evidence that is (to say the least) sparse, is also worthy of considerable skepticism.
And calling occasional violence "war" is stretching the word a lot (and ignoring the possibility that some, perhaps most, such deaths were due to internecine violence, not conflict with outsiders).

Wow. Just wow.

If you have ever been out trying to hunt and gather your own food (like deer or bird hunting), you know how much easier it is when the food comes to you. Every time I come back empty handed I am glad I don't have to feed my family that way (its called "hunting" not catching for a reason). Seasonality! LOL. Someone needs to be shaved with Occams Razor and try hunting for their own food.

I don't need a model to know that. This is not an innovation spurred by seasonality. Agriculture and domestication of animals happened because they are good ideas relative to hunting and gathering. Guaranteed food production. Try it.

As for the simultaneous, not sure that anything that happens over 8000 years can be called simultaneous. Fire and metal smelting technology also spread.

If there is any timing coincidence with the ice age, people could move around more, and there was more arable land. People moving around take their ideas with them.

Funny, you never hear about the simultaneous invention of a bunch of knuckleheaded ideas 8000 years ago that did not work.

It's guaranteed food production today, but you're leaning on 10,000 years of plant/animal domestication and breeding, not to mention thoroughly modern notions of property rights. Try living on teosinte instead of corn, when you're the only one growing, the concept of "ownership" is vague, and lots of people around you are hungry.

I am not sure what you mean property rights were "vague." In 10,000 BC you kill all the people on the land and take it over. It's yours if you can defend it. Your descendants inherit it. Chimps, wolves, and other predators engage in war for territory too. Wolves, for example, will defend a territory sufficient to ensure enough prey and move during a food shortage. Humans (even hunter-gathers) are not particularly special in that regard.

Probably a big motivator for migration as the ice age ended.

Now, you are right that early agriculture was not 100% guaranteed like it is today. But, people have been using dogs and flax for 30k years. Somewhere in there some people noticed that: If I keep the sheep within my defensible territory; and/or ensure that the seeds of the grasses (like flax) they prefer are re-spread for next year, I will have better luck hunting those sheep. It does not take much to manage a proto-farm, which really just ensures your prey stays close within the territory you can defend.

A lot of commenters here seem to imagine agricultural society developed immediately from paleolithic society. There was an intermediate stage called the mesolithic, in which people lived in large groups near rich hunting sites, such as large animal river crossings. Think of pacific northwest indians and their salmon supply. In other words societies had already achieved a fairly good level of food security before agriculture. The problem with this "seasonality" theory is that the most seasonal places are the most northerly, whereas agriculture developed first in the temperate zone. Sure, southern Turkey and Palestine are more seasonal than the tropics, but not as seasonal as Germany or Russia.

There also seems to be a lot of confusion around the term "simultaneous invention". The idea is that two people at the same time without help from each other invent the same thing. There's no way to know who did or didn't share ideas 10,000 years ago. Certainly all the earliest agricultural sites are close enough to each other that they could have shared ideas. There's no evidence of simultaneous invention.

What we do have are some independent inventions of agriculture, such as in the Americas, and many separate domesticaton events, where people in different locales each domesticated their local wild varieties, which is not evidence of independent invention of agriculture.

Comments for this post are closed