The dawn of human civilization is often pinned down to the rise of farming. As food production grew, so did human populations, trade, and tax.

Or so the prevailing story goes.

Economists have now put forward a competing hypothesis, and it suggests a surplus of food on its own was not enough to drive the transition from hunter-gatherer societies to the hierarchical states that eventually led to civilization as we know it.

Instead, multiple data sets covering several thousand years show this reigning theory is empirically flawed.

Even when some parts of the world adopted farming and began producing a surplus of food, it did not necessarily lead to complex hierarchies or tax-levied states.

Only when humans began farming food that could be stored, divvied up, traded, and taxed, did social structures begin to take shape.

That's probably why cereal grains like wheat, barley, and rice – rather than taro, yams, or potatoes – are at the root of virtually all classical civilizations. If the land was capable of cultivating grains, evidence shows it was much more likely to host complex societal structures.

"The relative ease of confiscating stored cereals, their high energy density, and their durability enhances their appropriability, thereby facilitating the emergence of tax-levying elites," the authors of the hypothesis write.

"Roots and tubers, in contrast, are typically perennial and do not have to be reaped in a particular period, but once harvested are rather perishable."

In parts of South America, for instance, perennial root crops like cassava can be harvested all year round. Unfortunately, however, cassava rots easily and is difficult to transport.

Researchers suggest this is why hierarchies beyond chiefdoms did not emerge in any societies dependent on cassava, even if there were more than enough roots to feed everybody.

On the other hand, the Maya were one of the most dominant and sophisticated civilizations in Central America, yet this ancient society did not rely solely on root crops. Instead, this civilization was heavily dependent on maize.

The same goes for Incas in the Andes.

The type of food grown by farmers was clearly more important to society than how much of it was produced.

The different social effects of root crops versus cereal crops could help explain why some civilizations became more complex, while other societies remained as local communities or chiefdoms. It could also clear up why a surplus of food in a hunter-gatherer society did not necessarily lead to the development of civilizations, either.

Farming was obviously a necessary step to improve food production, but researchers suspect only those crops that could be easily confiscated led to the rise of an elite class.

If a powerful echelon of society began collecting tax in the form of grain from farmers without extra food on their hands, then farming communities would not be able to support as large of a population. As a result, their numbers would likely shrink, thereby creating a surplus of food to bestow to more elite classes.

If these farmers didn't protect the elite, the elite wouldn't protect their food stockpiles from bandits. Stealing grains is, after all, much more valuable than stealing perishable foods.

"Thus," the authors of the new hypothesis write, "we concur with conventional productivity theory that farmers in hierarchical societies produce surplus, but our contention is that rather than surplus generating the elite, the elite generate the food surplus on which it can flourish, once the opportunity to appropriate rises."

An atlas of human cultures, for instance, shows that the largest number of wild relatives of cereal crops can be found in the Fertile Crescent, which is often said to be the cradle of human civilization.

Historical societies that did not practice any type of agriculture, meanwhile, are found in North-West America, Central Asia, Australia, and South-West Africa. These societies also lacked complex hierarchical structures.

Data from the United Nations Food and Agriculture Organization provides further evidence for the new hypothesis. It suggests regions where cereals are more productive than roots and tubers are more likely to be organized as states with higher rates of taxation.

Roots and tubers, meanwhile, are uncorrelated to more complex social hierarchies, even when they are grown on highly productive farming land.

"Using these novel data, we were able to show that complex hierarchies, like complex chiefdoms and states, arose in areas in which cereal crops, which are easy to tax and to expropriate, were de-facto the only available crops," explains economist Luigi Pascali from Pompeu Fabra University in Spain.

"Paradoxically, the most productive lands, those in which not only cereals but also roots and tubers were available and productive, did not experience the same political developments."

Pascali's co-author, the economist Joram Mayshar, calls it a "curse of plenty". Without a type of food that can be hoarded and protected by elite individuals, there is no ranked society of givers and takers, controlled by law and order.

Ultimately, Mayshar says, this reliance on root crops seems to have prevented the emergence of statehood and economic development in some parts of the world, like islands in the south pacific.

None of the empirical investigations laid out in the recent paper can fully prove or disprove the new hypothesis. But the authors argue their results are robust enough to "cast doubt on the prevailing productivity-and-surplus explanation for the emergence of hierarchy." They found no evidence for this oft-cited hypothesis.

"Only where the climate and geography favored cereals, was hierarchy likely to develop," says Mayshar from the Hebrew University of Jerusalem.

"Our data shows that the greater the productivity advantage of cereals over tubers, the greater the likelihood of hierarchy emerging."

The old adage 'we are what we eat' could hold more truth than we thought.

The study was published in the Journal of Political Economy.