2.1. Multilevel selection and group-level traits

Despite some resistance to MLS, there is a growing interest in applying the concept more broadly to include human social evolution (Boehm Reference Boehm1997; Reference Boehm1999; Reference Boehm2012; Boyd & Richerson Reference Boyd and Richerson2002; Gowdy & Krall Reference Gowdy and Krall2013; Reference Gowdy and Krall2014; Hodgson & Knudsen Reference Hodgson and Knudsen2010; Reeve Reference Reeve2000; Richerson & Boyd Reference Richerson and Boyd2005; Smaldino Reference Smaldino2014; Turchin Reference Turchin, Richerson and Christiansen2013; van den Bergh & Gowdy Reference van den Bergh and Gowdy2009; Wilson Reference Wilson1997; Wilson & Gowdy Reference Wilson and Gowdy2013; Reference Wilson and Gowdy2015). Boyd and Richerson (Reference Boyd and Richerson2002) argue that social learning in humans leads to gene-culture coevolution and selection for group-level traits. Social scientists and biologists have acknowledged the importance of the coevolution of genes and culture (Richerson & Boyd Reference Richerson and Boyd2005; Wilson Reference Wilson1997). Smaldino (Reference Smaldino2014) has explored the emergence of group-level traits through, among other things, evolutionary competition between groups. Caporael (Reference Caporael1997) and Caporael and Garvey (Reference Caporael and Garvey2014) use the concept of “core configuration” – the scaffolding of smaller to larger social and economic units – to examine the emergence of group-level traits. The group selection approach has been fruitfully applied to the evolution of cooperation (Sober & Wilson Reference Sober and Wilson1998), the evolution of state societies (Spencer Reference Spencer2010), and the role of warfare in early agricultural societies (Choi & Bowles Reference Choi and Bowles2007; Turchin Reference Turchin2006a). Campbell (Reference Campbell, Gruter and Bohannan1982, p. 161) provides guidance as to how group selection works in human societies:

While it is clear that culture can promote homogeneity and cohesion within groups, it leaves ambiguous the common evolutionary process at work for both humans and insects. One cannot reasonably argue that insects have culture in the way that humans do. The explanation for homogeneity and cohesiveness of the insect colony is usually attributed to genetic relatedness, while that of humans is most often attributed to culture. A more complete parsing of the MLS literature with regard to the common evolutionary matrix at work for both social insects and humans who become ultrasocial is warranted.

2.2. MLS1 and MLS2

The multilevel selection literature divides selection into two types: multilevel selection 1 (MLS1) and multilevel selection 2 (MLS2). As described by Okasha (Reference Okasha2006), MLS1 is a case where certain traits (such as altruism, to use D. S. Wilson's trait group example) may decrease individual fitness but enhance the competitiveness of groups that practice it. Therefore, groups having many altruists have a competitive advantage over groups that have few, and altruism gets reproduced (Okasha Reference Okasha2006, p. 178). Altruism is an individual (particle) level trait, but the group dynamic enhances the probability that it will be reproduced. With MLS1, natural selection still operates at the individual level and also at a group level, but both levels of selection work on “a single evolutionary parameter” – in this case, altruism (Okasha Reference Okasha2006, p. 178). With MLS1, altruism is still an advantage to the individual because, by enhancing group survival, it also enhances individual survival.Footnote ⁷ Most examples of group selection in the literature fall into the MLS1 category – “traits that can be easily measured in individuals but require group selection to evolve because they are locally disadvantageous” (D. S. Wilson Reference Wilson2010, section XIX).Footnote ⁸

2.3. Emergent characteristics in humans and social insects

The evolutionary picture becomes more complicated when the play of selection at the group level is not on a single trait but on a cluster of emergent characteristics that define the group and make it a whole. Okasha (Reference Okasha2006, p. 178) says it succinctly: “In MLS2, individuals and groups are both ‘focal’ units, and the two levels of selection contribute to different evolutionary changes, measured in different currencies.” According to Okasha (Reference Okasha2006, p. 112): “Emergent characters are often complex, adaptive features of collectives, which it is hard to imagine evolving except by selection at the collective level.” Okasha (Reference Okasha2006, pp. 229–230) accepts the standard view of MLS2 but points out that it applies only to the later stages of an evolutionary transition (after the collective units have formed and are replicating). In the early transitional stages, cooperation must spread among the particles so that they will eventually give up their individuality and form discrete collectives. Thus, the “emergent characters” present difficulties for multilevel selection because it is a chicken-and-egg problem. This is described by Okasha (Reference Okasha2006, p. 113) in reference to work by Williams (Reference Williams1992) and Sober and Wilson (Reference Sober and Wilson1998): “the emergent character requirement conflates product with process.” This may be true, but it does not constitute an intractable problem if sufficient attention is paid to both process and product, that is, the emergence and configuration of the group-level trait. MLS2 needs to capture the process that forms an altered group that then has sufficient cohesion and force to become a focal point of selection. The process must account for the emergent characteristics which configure the group that then becomes a unit of selection.

The commonality in the structure and dynamic of ant, human, and termite societies that develop large-scale agriculture is not merely coincidental. MLS2 does not sufficiently sort through the process/product problem and it delineates the fitness criteria as successful collectives producing more offspring collectives. This does not capture a basic attribute of the fitness of a food producing group, namely, its internal expansionary dynamic. Fitness is not simply that more new colonies form. It is that any single colony can grow to an enormous size.Footnote ⁹ With agriculture the collective is extended over generations. Some ultrasocial social insects develop new collectives from existing collectives when a queen flies away and starts a new nest. The cumulative effect is different for humans who do not start new colonies with the same ease and sharp delineation as agricultural insects. Again, we argue that, although the details of expansion may differ between humans and social insects, the economic drivers of that expansion are similar.

Smaldino has expanded the discussion of group selection with his refinement of MLS2. His discussion helps guide our thinking about the common evolutionary story of insect and human ultrasociality. The group is defined by Smaldino as the emergence of group-level traits that “involve organized collections of differentiated individuals” (Smaldino Reference Smaldino2014, p. 243). These within-group traits constitute between-group differences that become the focus of selection. Smaldino is clear: “In order to explain group-level traits, the emergence of differences among individuals within a single group, and the subsequent organization of those differentiated individuals and their coordinated behavior must be accounted for” (Reference Smaldino2014, p. 249). In this framework we must identify the commonalities in both the process of differentiation and the commonalities in the reconfiguration of that differentiation into a whole if we are to explain the common structure and dynamic of societies of ants, humans, and termites. The configuration of the whole then defines its group-level trait and the unit of selection.Footnote ¹⁰

Smaldino (Reference Smaldino2014, p. 248) specifically discusses the homogeneity of the group using the example of social ants. Collaboration among workers, soldiers, drones, and queens is promoted through genetic relatedness and the colony becomes an extended phenotype of the queen. The conclusion is that differentiation within the colony comes about through environmental stimuli that trigger phenotypic differences and cohesion is largely a matter of genetic relatedness, although positive assortment through group structure may also play a role. Smaldino contrasts this to the situation found in humans where between-group differences, and within-group cohesiveness, are “triggered culturally rather than genetically requiring different explanations for their emergence and evolution.” Differentiation and cohesion are culturally mediated, but placing too much emphasis on the different mechanisms in social insects and humans that bring about differentiation and cohesion is a diversion from identifying the commonality among these diverse lineages. The organization of agricultural production promotes a more elaborate differentiation of individuals (division of labor). In a sense it does not matter how species attain differentiation as long as they do so. In all cases the extensive division of labor around the active management of food production creates a profound interdependence. The extreme differentiation of occupations brings about a much greater cohesion and co-dependence in production and rewrites the boundaries between the individual and the group.

3.1. Actively harnessing the inputs to food production

With agriculture it was possible for humans, ants, and termites to interject themselves directly into the food production process. Humans could actively capture solar energy in crops that replaced other vegetation and tap into stocks of agricultural inputs like fertile soil and water for irrigation. The economist Georgescu-Roegen (Reference Georgescu-Roegen and Georgescu-Roegen1976) pointed out that stocks of inputs like the nutrients in fertile soil have an advantage over flows because they can be used at any rate and can therefore make possible a larger and more complex production process (Georgescu-Roegen Reference Georgescu-Roegen and Georgescu-Roegen1976).Footnote ¹³ More energy and other resources can be directed to food production. E. O. Wilson (Reference Wilson1987) points out the importance to ant agriculture of tapping into resource stocks:

This gave them an almost unlimited supply of nutrients to support their agriculture. Hölldobler and Wilson (Reference Hölldobler and Wilson2011, pp. 2–3) write: “The leafcutter ants are partly comparable in their achievement to that of human agriculturalists. And they have attained a breakthrough of organic evolution: by using fresh vegetation on which to grow their crops, they have tapped into a virtually unlimited food source.” Perhaps forest floor vegetation is not technically a stock, but it essentially amounts to a stock for ant agriculturalists. Attine ants commandeer vegetation away from species that would otherwise use it.

Tainter et al. (Reference Tainter, Allen and Hoekstra2006, p. 52), in discussing energy transformations in complex societies, argue that the evolution of ant agriculture shows two major resource transitions. The first was the adoption of agriculture itself – based on actively growing fungus on high-quality insect droppings. This transition was highly successful, but the size of the colonies was limited by the scarcity of quality insect droppings. The second transition was a shift from collecting droppings to growing fungus on much more abundant leaves and other organic material. Each step in ant social evolution not only paved the way for the next step, but also came with increasing complexity and regimentation of ant society, which itself enhanced the ability to tap into the stock.

3.2. The complex division of labor

An important economic driver of ultrasociality and its evolution is the expansion and sharpening of the division of labor. Both biologists and economists have extensively studied the division of labor. Biologists have identified the benefits of division of labor in social insects (Beshers & Fewell Reference Beshers and Fewell2001; Franks Reference Franks1987; Oster & Wilson Reference Oster and Wilson1978; Wilson Reference Wilson1971). Hölldobler and Wilson (Reference Hölldobler and Wilson2009, Ch. 5) give numerous examples describing its advantages. In the case of ants, the number of total tasks increases with the total number of ants sampled in a colony (Hölldobler & Wilson Reference Hölldobler and Wilson2009, p. 125). Holbrook et al. (Reference Holbrook, Barton and Fewell2011) and Hölldobler and Wilson (Reference Hölldobler and Wilson2009) found that the division of labor (the number of distinct roles) increases with colony size, although it is hard to separate cause and effect. Phenotypic variation enables ants to differentiate according to their productive role or their role in supporting productive activity (as in defense).

It should be acknowledged that the division of labor is common in the animal world and is not by itself a distinguishing characteristic of ultrasociality. For example, a division of labor based on care of the young is common among animals. It spontaneously appears in normally solitary queen ants when the queens are forced to associate (Fewell & Page Reference Fewell and Page1999) and similarly in normally solitary bees. Solitary sweat bees alternately dig nesting holes and guard the nest. When two are put together, one will specialize in excavation and the other will guard the nest entrance, resulting in efficiency gains in both tasks. According to Holbrook et al. (Reference Holbrook, Clark, Jeanson, Bertram, Kukuk and Fewell2009, p. 301), “Paired individuals performed more per capita guarding, and pairs collectively excavated deeper nests than single bees – potential early advantages of social nesting in halictine bees.” Even in a simple society of two individuals there is an advantage to a larger scale (from one to two), permitting a division of labor. The spontaneous appearance of the division of labor in these simple cases is remarkable and may hold keys to its development in ultrasocial societies. But the extent of the division of labor in ultrasocial species is unique in its complexity and interdependence.

The active management of agricultural crops is a particularly powerful impetus for a more complex division of labor. Consider the differences between a eusocial honeybee colony and an ultrasocial attine ant colony. Honeybees survive by foraging for pollen from plants but they do not actively manage the sources of pollen. Honeybees have a division of labor, for example, cleaning, feeding the brood, receiving the nectar, foraging, and defense, depending on the age of the individual bees (age polyethism). Yet, there are only three physical castes: workers, drones, and queens. Workers are of only one physical type. Some attine ants, by contrast, have many castes who actively manage the production of various species of fungi which feed the colony. Castes are based partially on size, which “coarse-tunes” rather than “fine-tunes” the phenotypes to perform a wide variety of tasks (Oster & Wilson Reference Oster and Wilson1978). Ants within each size caste can perform a number of further-refined, highly specialized tasks. For example, there is a caste of tiny attine ants whose job it is to ride atop the much larger leaf carriers and defend them from attacks by parasitic flies (Hölldobler & Wilson Reference Hölldobler and Wilson2011). Large soldier ants have jaws so specialized for defense that they cannot feed themselves. There are even “untouchable” ants whose job it is to remove wastes and pathogens from the fungal gardens. The point is that the active management of food production, and the defense of food surpluses, calls forth a much more complex division of labor requiring many more tasks and a much greater degree of coordination than does mere foraging. Ferguson-Gow et al. (Reference Ferguson-Gow, Sumner, Bourke and Jones2014) found a significant positive relationship between the complexity of agriculture systems in attine ants (from lower attines to leafcutters) and complexity of the division of labor.

Another factor facilitating the division of labor under agriculture is mutualism. The fungi that ants and termites live on could not survive without active management by the ant colony. This requires complex tasks including using antibodies to control the bacteria that attack the fungus (Aanen & Boomsma Reference Aanen and Boomsma2006; Mueller & Gerardo Reference Mueller and Gerardo2002). It is true that flowers are pollinated by (non-agricultural) honeybees, but there exist other pollinators and the plants could survive without the presence of honeybees (as they are now doing in many areas because of honeybee die-off).

The advantages of a division of labor are a central tenet of economic theory. They were recognized by Adam Smith in The Wealth of Nations (Reference Smith1776/1937, Book 1), where he presented his well-known pin factory example of the role of the division of labor in rationalizing the production process. One person working alone, he wrote, could scarcely make one pin a day. But when the enterprise is divided into several sub-tasks, productivity increases dramatically.

Smith also pointed out that the division of labor is limited by the extent of the market. The larger the market, the more specialization is possible, which enlarges the market still more in a kind of “virtuous circle.” Although he was writing about market capitalism, Smith was onto something more fundamental in the sense that he clearly saw that there is a system at play that places a premium on capturing the efficiencies inherent in expanding the division of labor.

Active intervention in food production called forth a more extensive and interdependent division of labor in humans as well as social insects. In hunting and gathering societies the division of labor was relatively simple, primarily based on gender. A more interdependent and extensive division of labor began to take form in pre-agricultural societies with the harvesting of wild grains. There were more kinds of tasks to perform. Access to the good stands of wild wheat was essential, and it is likely that defense became more important in order to lay claim to a prime spot. As the active cultivation of crops increased, so did the necessity for coordination of production in planning, preparing the soil, planting, cultivating, harvesting, processing, storing, and distributing the agricultural output. As well, reproductive rates increased in humans, so there was more work to be done in birthing and child rearing. Women's roles became more narrowly circumscribed around these activities. Thus, agriculture expanded the division of labor and increased the interdependence between people for day-to-day sustenance.

The division of labor in humans does not entail the same phenotypic (morphological) differentiation as in ants and termites because humans have greater recourse to cultural, institutional, and technological extensions of themselves. But occupational differentiation in humans is also made possible to some extent by genotype-phenotype plasticity. Human brain plasticity allows for a remarkable degree of differentiation in terms of the ability of individuals to adapt to different cultures and behavioral patterns (Frith & Frith Reference Frith and Frith2010; Wexler Reference Wexler2006). This is not to say that specific instances of occupational differentiation are genetically determined, but brain development plasticity gives humans the flexibility to perform a variety of functions. The extent of the detailed division of labor becomes so great in ants and in humans that individuals become tied to a very narrow productive role in society. This creates an interdependence that secures and strengthens the group as a self-referential entity.

3.3. Increasing returns and the competitive advantage of larger group size

Larger group size may also be more metabolically efficient because of economies of scale in energy use. Hou et al. (Reference Hou, Kaspari, Vander Zanden and Gillooly2010) and Shik et al. (Reference Shik, Hou, Kay, Kasari and Gillooly2012) demonstrate this in their studies of ant colonies and use Kleiber's Law (the rate at which an organism processes energy increases at a rate that is approximately equal to the ¾ power of that organism's body mass) as the explanation. Larger colonies have lower rates of per capita energy use (Bruce & Burd Reference Bruce and Burd2012). But there seems to be an upper limit on leaf cutter ant colony size due to the fact that colonies will eventually reach a limit where the returns to increasing foraging territory is not profitable (Bruce & Burd Reference Bruce and Burd2012).

Among the many tasks that developed with agriculture, defense was one that gave an advantage to larger group size. In human agricultural societies, because of the need to lay claim to property and because of the time lag between planting and harvesting, there was a need to defend property. Although defense may have been necessary even while harvesting wild grain, it took on added importance with the investment in managed agriculture. Human societies that were larger and better able to organize warfare, and develop war-making innovations, outcompeted others and expanded rapidly (Matthew & Boyd Reference Matthew and Boyd2011; Turchin Reference Turchin2006a). Eventually, warfare became prevalent as larger-scale state societies began to form. As Larsen (Reference Larsen2006, p. 17) puts it: “The record strongly suggests that population size increases associated with food production provided conditions conducive to the rise of organized warfare and increased mortality due to violence.”

There are other advantages to larger size and scale. Unlike solitary insects, social insects can perform a number of tasks in parallel as opposed to in sequence, an advantage described in Adam Smith's pin factory example. As Georgescu-Roegen (Reference Georgescu-Roegen and Georgescu-Roegen1965) pointed out for the human economy, if the scale of operation is large enough, idle factors of production can be eliminated and thus larger-scale systems can entail a more efficient and productive use of resources as long as they are abundant enough to support the larger scale. Also, time is not lost in moving from one task to another. An ultrasocial society can also take advantage of a spatial distribution of labor, allowing for more risk taking in foraging than would be possible for individuals acting alone.Footnote ¹⁴ In general, larger scale and the division of labor are mutually reinforcing. A larger scale of operation can also take advantage of a spatial distribution of labor, again allowing for more risk taking in foraging than would be possible for individuals acting alone. Also, with a larger scale of operation, individuals can specialize and become more proficient in their tasks.

Human hunter-gatherers lived off the flows of nature – from the solar energy directly captured by living plants and indirectly present in the flow of animals feeding on those plants. They had limited control over these subsistence flows. If hunter-gatherers, like any large carnivores, took too many animals or harvested too many wild plants, this resulted in immediate shortages; thus, resource management became necessary and this tended to keep human societies in ecological balance. These societies also had leveling mechanisms to promote an egalitarian distribution of wealth and power. With the adoption of agriculture, human society and the relationship between humans and the natural world changed dramatically.

4.1. The origins of human agriculture

There is no consensus as to the origins of agriculture. Price and Bar-Yosef (Reference Price and Bar-Yosef2011, p. S168) summarize: “There is as yet no single accepted theory for the origins of agriculture, rather, there is a series of ideas and suggestions that do not quite resolve the questions.” Our intention here is not to provide the definitive explanation for this transition (which likely varied from place to place; see McCorriston & Hole Reference McCorriston and Hole1991), but rather, to offer in broad outline a plausible story for the transition from hunting and gathering to settled agriculture and the concomitant social and environmental consequences. Agriculture gave our species the ability to control and expand its supply of food, and this was an evolutionary advantage (as measured by total population), even though it apparently made the average individual worse off. According to Larsen (Reference Larsen2006, p. 12), “Although agriculture provided the economic basis for the rise of states and development of civilizations, the change in diet and acquisition of food resulted in a decline in quality of life for most human populations in the last 10,000 years.” The archeological record substantiates Larson's claim. After agriculture, humans became shorter and suffered from more debilitating diseases, from leprosy to arthritis to tooth decay, than their hunter-gatherer counterparts (Cohen & Crane-Kramer Reference Cohen and Crane-Kramer2007; Lambert Reference Lambert2009). It is only in the last 150 years or so that longevity once again reached that of the Upper Pleistocene. The average human life-span in 1900 was about 30 years, and for Upper Pleistocene hunter-gatherers it was probably about 33 years.Footnote ¹⁸ Only in the last century or so has the well-being of the majority of humanity improved dramatically. It remains to be seen whether or not these improvements can be maintained. Care must be taken not to see the achievements of the very recent past as representative of the consequences of the agricultural transition.

Humans had extensive knowledge of wild plants long before the adoption of agriculture (Cohen Reference Cohen1977; Zvelebil & Rowley-Conwy Reference Zvelebil, Rowley-Conwy and Zvelebil1986), so there must have been some experimentation with planting during our lengthy hunter-gatherer history. Flannery (Reference Flannery and Meggers1968) observed: “We know of no human group on Earth so primitive that they are ignorant of the connection between plants and the seeds from which they grow.” Yet understanding and observing and collecting wild plants is one thing, while domestication is another. What pushed humans to adopt agriculture? Convincing arguments have been made that the Holocene provided a period of climate stability that was necessary for successful agriculture. Richerson et al. (Reference Richerson, Boyd and Bettinger2001) have demonstrated that the possibilities for agriculture were severely limited before the Holocene because of unpredictable climate fluctuations. There were several periods of warming after the evolution of modern humans but none except the Holocene led to agriculture. Climate data indicate that prior to the Holocene, changes in temperature as great as 8°C occurred over time spans as short as two centuries (see Bowles & Choi Reference Bowles and Choi2012, supporting online material, p. 4). Ice core and pollen records indicate that centuries-scale abrupt climate events occurred regularly during the Pleistocene and that it was not until the Holocene that a protracted period of warming occurred. In the Late Pleistocene, plant productivity was low because of reduced CO₂ levels (about 180 ppm, compared to 250 ppm at the beginning of the Holocene (Shakun et al. Reference Shakun, Clark, Marcott, Liu and Otto-Bliesner2012). Beerling (Reference Beerling1999) estimates that the total amount of stored organic land carbon was 33% to 60% lower in the Late Pleistocene compared to the Holocene.

One popular argument is that population pressure drove the adoption of agriculture (Binford Reference Binford, Binford and Binford1968; Cohen Reference Cohen1977). But others have pointed out that there is little evidence for population pressure in the areas where agriculture first appeared (Price & Bar-Yosef Reference Price and Bar-Yosef2011). However, the Holocene warming may be related to population pressures in some areas. McCorriston and Hole (Reference McCorriston and Hole1991, p. 49) noted that the effect of the Holocene warming in the Levant was to prolong summer aridity and that this would have affected the availability of agricultural inputs. This likely affected the availability of viable locations for groups of humans. The drier climate also put more pressure on access to water and probably concentrated human and animal populations in areas with ample water. The population picture is further complicated by the fact that even incipient agriculture may have resulted in a more sedentary life, which in turn increased fertility rates. Even the gradual and marginal use of wild grains might have altered population dynamics if it made people more sedentary and more fertile, thus creating a positive feedback path reinforcing the need for more agriculture as pressure on wild food sources and hunting became more challenging.

4.2. A plausible story of the human agricultural transition

A plausible scenario of the human agricultural transition can be sketched out (Gowdy & Krall Reference Gowdy and Krall2014). Mobile hunter-gatherers moved through places where wild grains thrived, and these grains provided a significant portion of hunter-gatherer diets. As the climate warmed and became more stable, wild grains became more reliable and more important as a food source. People began to sow wild seeds to enhance grain growth, and they began to store the grain they collected. As they sowed, they also selected for desirable characteristics. Storage, especially amenable to annual grains, was a good subsistence strategy since there was variability in production from year to year. This enhanced and concentrated food supply led to a more concentrated population. Perhaps a portion of the population began to stay behind in the seasonal migrations in order to manage the wild crops. Selective planting and harvesting of crop varieties eventually led to managed agriculture and populations more and more dependent on intentional food production.Footnote ¹⁹

This plausible story fits with what we know of the agricultural transition in the Levant (which includes parts of modern Palestine, Syria, Israel, and Jordan) beginning around 10,000–12,500 years ago (Bar-Yosef Reference Bar-Yosef1998, p. 162; McCorriston & Hole Reference McCorriston, Hole, Kiple and Ornelas2000b).Footnote ²⁰ The Levant is the area of the world where the advent of agriculture is the best documented. A key feature of the Holocene warming in the Near East was that it created conditions favorable to the development of annual grains. The climate of the Levant became more stable and seasonal differences became greater. Around 10,000 years ago the pre-agricultural Natufians began to rely more heavily on wild grains like wheat and barley (McCorriston & Hole Reference McCorriston, Hole, Kiple and Ornelas2000a; Reference McCorriston, Hole, Kiple and Ornelas2000b). McCorriston and Hole (Reference McCorriston and Hole1991, p. 61) claim: “In our view, the wild annual plants had never been available in densities comparable to those of the Early Holocene when seasonality reached unprecedented extremes and favored annual over perennial adaptation.” Annuals have an advantage in places with enhanced seasonality, especially where there are hot, dry summers and strong seasonal rainfall variation. Annuals store their reproductive ability in seeds which can wait (sometimes years) for rain to germinate. Annuals also have unique characteristics that may allow rapid coevolution to develop between humans and plants (Cox Reference Cox, Ceccarelli, Guimaraes and Weltizien2009). Annual grains can also be stored, and so a greater reliance on them meant a greater capacity to secure a surplus. Evidence exists for food storage at about 11,000 years ago, about 1,000 years before domestication and large-scale settlements, in the form of purposely built granaries (Kuijt & Finlayson Reference Kuijt and Finlayson2009).

One mutation of wild wheat that would have been important was non-shattering, a mutation that allows for seeds to hang on and not fall to the ground as quickly. This is apparently a rare mutation but might have been noticed by those seed gatherers who were accessing wild stands of early wheat varieties. The time period for harvesting wild wheat was short – three days to a week before shattering occurs depending on weather conditions. After seeds shatter, they must be harvested from the ground, which is more time consuming. If people reached a stand of wild wheat after shattering had taken place, the only seeds still standing would have been the mutants, which would have been noticed. Also, using a sickle would have been a particularly good technology for non-shattering seeds. In this way mutations in wheat interfaced with human intervention to create a selection bias for non-shattering seeds or any other trait that was noticeable and desirable.

Domestication of grains may have initially been unintentional in the sense that wild varieties would have been eliminated and replaced more systematically with plants that required human intervention (Bar-Yosef Reference Bar-Yosef1998, p. 167). With cultivation, there gradually developed a specialized active management of grains where control of production was more concentrated within human groups (Flannery Reference Flannery and Meggers1968; McCorriston & Hole Reference McCorriston, Hole, Kiple and Ornelas2000b; Rindos Reference Rindos1984). Active management also required more complex and integrated tasks (in both humans and insects), and thus there is a connection (mutualism) between the characteristics of the crops and the species managing the crops.

As the climate improved and stabilized during the Holocene, wild grains such as triticum monococcum, a wheat-like grass, became more plentiful. Evidence suggests that the Natufians intensively harvested wild cereals using sickles (Bar-Yosef Reference Bar-Yosef1998, pp. 164–65). There is also evidence of heavy wear of teeth, presumably due to consuming coarsely ground cereals (Smith Reference Smith1972, pp. 236–37). To reiterate, annuals might have been increasingly abundant and amenable to the climate conditions of the Holocene. Hunting did not cease as the use of wild grains increased, but there was likely a shift in hunting strategies and the importance of hunting in the diet. Hunting may have become less reliable as the climate change of the Holocene would have changed the range and concentrations of wild animals (McCorriston & Hole Reference McCorriston and Hole1991).

Some mention should be made of the effect of the Younger Dryas – a sudden cooler and drier period occurring between 12,800 and 11,500 years ago – on the cultures of the Near East. Belfer-Cohen and Bar-Yosef (Reference Belfer-Cohen, Bar-Yosef and Kuijt2000) argue that it was the stress of the Younger Dryas that pushed the adoption of agriculture. But others point out that there is no evidence for more intensive resource use or food stress in the late Natufian (Munro Reference Munro2003). In fact, during this period the population densities and settlement patterns of early Natufian culture reverted back to pre-Natufian levels. It may be that the Younger Dryas interrupted the transition to agriculture rather than encouraged it.

The Natufians were followed by the fully agricultural cultures of the Neolithic, referred to as Pre-Pottery Neolithic A (PPN-A). PPN-A sites are much larger than the Natufian sites, with storage bins for grains, ceremonial structures, and a rich lithic industry. The best known PPN-A settlement is Jericho (about 10,000 years ago), thought to be the world's first known town, with a population of about 2,000–3,000 people. The Jericho site shows the first known domesticated cereals: emmer wheat and two-row hulled barley (McCorriston & Hole Reference McCorriston and Hole1991, p. 51). Storage technology is found abundantly in the Pre-Pottery Neolithic. Makarewicz (Reference Makarewicz and Eren2012, p. 217) writes: “The Pre-Pottery Neolithic A marks a major shift in human approaches to subsistence from plant gathering to the consistent practice of plant cultivation, where wild plant resources were augmented by a more predictable food source in the form of managed plants, particularly cereals and legumes.” Eventually the point was reached where the human population in the Levant could not survive without the grains, and the grains could not survive without human intervention.

4.3. Managing crop production: The economic drivers kick in

The active management of crops fundamentally altered human economic organization. Like our ant and termite distant relatives, humans began to actively engage in the primary production of their food supply. Economic life was no longer a matter of living off the day-to-day flows from nature. It now involved a direct intervention into food production.Footnote ²¹ Many species use social organization to get food, as in cooperative hunting, for example. But direct intervention in producing food is categorically different, and the economics of production take center stage.

First of all, actively producing food requires a more complex and interdependent division of labor. This increase in complexity had its beginnings in the harvesting of wild stands of grains. Harvesting must move quickly, or most of the grain may be lost. Protecting access to the good stands of wild grain was essential. The reward was a greater control over food supply for the group as a whole, but the repercussions if things failed were formidable. As long as storage was possible, production of surplus was an insurance against lean times.

A second feature of emergent agricultural society is an impetus to expand because of the productive advantage of larger group size. More food allows expansion, and expansion captures the efficiencies of a greater division of labor and economies of scale. Also, many tasks are required that do not directly contribute to food production itself. Those who directly produce food must provide for those not actively engaged in agricultural production. Part of the dynamics of agriculture is the phenomenon of increased reproductive rates caused by sedentary life. Increased reproductive rates also provided one of the most important resources for successful agriculture, a large supply of laborers. But for a time at least, this requires some investment in nonproductive individuals, that is, young children who cannot yet work. The result is again that those engaged in the direct provisioning of food must produce enough to feed more nonproductive individuals. Greater population creates a greater need for agricultural output, and greater agricultural output creates a greater need for population.

Turchin (Reference Turchin, Richerson and Christiansen2013) points to increasing returns to scale in warfare as a major driver of ultrasociality. Turchin et al. (Reference Turchin, Currie, Turner and Gavrilets2013) argue that agriculture increased the payoff for aggression, which, in turn, necessitated more food production to feed the expanding military. Groups with more soldiers eliminated or absorbed smaller groups. Warfare has been suggested as a key to the development of state societies (Carneiro Reference Carneiro1970; Tilly Reference Tilly1992; Turchin Reference Turchin2006a). By contrast, a strong case has been made that warfare did not exist in hunter-gatherer societies (Culotta Reference Culotta2013). Violence certainly existed in these societies, but most lethal events resulted from personal disputes (Fry & Söderberg Reference Fry and Söderberg2013).

Finally, the ecological consequences of annual grain agriculture may have also encouraged the expansionary dynamic. Annuals had a greater capacity for seed production, and the rewards of active management were greater, but they also had a greater potential for ecological damage, especially soil erosion and soil disturbance during planting (Cox Reference Cox, Ceccarelli, Guimaraes and Weltizien2009). Expansion was also one way out of the ecological problems caused by grain agriculture, although in the long run they exacerbated the ecological problems as agriculture expanded into forested areas and deforestation brought down silt.

4.4. Surplus and hierarchy

Surplus production was perhaps initially a response to the favorable climate of the Holocene in the Near East, but surplus production and expansion increasingly became necessary to accommodate population growth, the growth of the proportion of nonproductive individuals, and the problems of seasonal variation, and to counter the ecological degradation of grain agriculture. Because of the premium placed on maximizing output, it was important to control production and maximize the output of individual laborers. During key periods such as planting and harvesting, work was necessarily intensive and repetitive. The regimentation of work and the productive benefits of a division of labor and attendant economies of scale promoted the success of agriculture but also increased the bureaucracy necessary to manage the complexity, the distribution of surplus, and the associated ecological problems. In this environment, relying on stored surpluses to carry societies over seasonal periods of low production made maximizing surplus production in any given year all the more critical. It also increased the stakes associated with management of production and with dissemination of the surplus.

With storage of surplus, complex technology, and the ability to physically control food surplus also came hierarchical societies. Control itself took on added importance, moving society increasingly in the direction of hierarchy and property. Year-round storage also meant that early farmers could settle in larger groups throughout the year, and gave society the flexibility to support specializations like craftspersons, administrators, warriors, and religious leaders (McCorriston & Hole Reference McCorriston, Hole, Kiple and Ornelas2000b). The need for efficiency and control set up a more mechanistic and interdependent configuration of production and distribution. The survival of the society depended on the payback from the management of production. Those at the top benefitted from pushing the envelope of agriculture intensification. The successful production of agricultural surplus expanded the material and cultural possibilities of society. But the fact that most individuals had no other option to secure the material necessities of their lives other than to participate in directly producing food placed the human propensity to cooperate at the disposal of both direct and indirect coercion.

Antecedents to hierarchical organization exist in non-agricultural societies, which gives credence to the importance of the management of surplus and the active control of production in the creation of hierarchy. Woodburn (Reference Woodburn1982) distinguished between “immediate return” and “delayed return” hunter-gatherer societies, with the latter possessing a more elaborate technology (capital) such as nets and boats, and showing some evidence of social stratification. In delayed return hunter-gatherer societies, people have rights over valued assets of various kinds. A commonly given example of a hierarchical culture without agriculture is that of the original inhabitants the northwest coast of North America. A village on Keatley Creek in northwest Canada was occupied between 2,500–1,100 years ago by hunter-gatherers who lived on the abundant seasonal salmon runs. The village consisted of more than 115 pit houses and other structures with log and earth roofs. The peak population of the village was about 1,500 people. Evidence indicates wide disparities in living standards, which may have originated in unequal access to prime fishing grounds (Prentiss Reference Prentiss2012; Pringle Reference Pringle2014). Richerson et al. (Reference Richerson, Boyd and Bettinger2001) point out that the transition from the beginning of agriculture to the development of state societies took around 2,000 years. Thus, a dynamic that started out modestly and benignly locked early agriculturalists into a perhaps irreversible process that led to hierarchical state societies.

5.1. The social conquest of Earth: Ecosystem domination, sustainability, and collapse

Ultrasociality evolved in only a handful of species, yet those species dominate the ecosystems within which they occur and indeed the entire planet (Wilson Reference Wilson2012). Sanderson et al. (Reference Sanderson, Jaith, Levy, Redford, Wannebo and Woolmer2002) calculated that over 80% of the global terrestrial biosphere is under direct human influence. Astonishingly, the total dry weight human biomass (about 125 million metric tons) is over 12 times the weight of all other vertebrates combined (Smil Reference Smil2013). Social insects also dominate their ecosystems to an amazing degree. In a survey of a patch of rainforest in the Brazilian Amazon, social insects comprised about 30% of the entire animal biomass and 75% of the insect biomass (Fittkau & Klinge Reference Fittkau and Klinge1973; Wilson Reference Wilson2008, p. 6). Wilson (Reference Wilson2012, pp. 116–17) “crudely” estimates the total number of ants to be 10¹⁶, or 10 thousand trillion. If one ant weighs one-millionth of the weight of a human, the total weight of the world's ants is about the same as the total weight of all humans. In subtropical and tropical ecosystems, termites can make up as much as 95% of the soil insect biomass (King et al. Reference King, Warren and Bradford2013; Korb Reference Korb2007, p. R998).

Ants, in effect, have re-engineered the ecosystems they are a part of. Folgarait (Reference Folgarait1998) discusses some of the major functional roles of ants, including soil modification through physical and chemical changes, changes in nutrient and energy fluxes, and changes in vegetation. Other species have coevolved to accommodate themselves to the presence of the numerically dominant ant and termite colonies. By contrast, humans after agriculture have had a predominately negative impact on ecosystems and biodiversity.Footnote ²²

The archeological and historical record of early agricultural societies is characterized by rapid expansion, followed by collapse and social disintegration (Diamond Reference Diamond2005; Weiss & Bradley Reference Weiss and Bradley2001; Weiss et al. Reference Weiss, Courty, Wetterstrom, Guichard, Senior, Meadow and Curnow1993). Examples include the Akkadian empire, Old Kingdom Egypt, the Classic Maya, and the Harappan of the Indus valley. These civilizations disintegrated because of a variety of factors, including the loss of soil fertility, erosion from reliance on annual plants, soil salinization, water mismanagement, and the inability to withstand prolonged droughts. Climate change in particular is increasingly accepted as a driver of past societal collapse and disruption (Rosen & Rivera-Collazo Reference Rosen and Rivera-Collazo2012; Weiss & Bradley Reference Weiss and Bradley2001). The basic problem is that ultrasocial societies are expansionary, that is, they have a constitutional proclivity to expand; and because of their tremendous interdependence, they are particularly difficult to disengage before they reach the point of collapse either due to ecological limits (that might be exacerbated by climate change) or due to internal conflicts between classes, in the case of humans, regarding the distribution and use of surplus.

Of course, social insects have had tens of millions of years of evolutionary trial and error to hone sustainable cultures. It is quite possible that the early history of ultrasocial ants and termites is also littered with unsuccessful experiments with agriculture. Can humans learn something about sustainability from insect farmers? Aanen and Boomsma (Reference Aanen and Boomsma2006, p. R1016) write:

Ants and termites have practiced monoculture successfully for tens of millions of years, while the short history of human management of a few crops shows a pattern of recurring instability (Benckiser Reference Benckiser2010). One reason social insects have achieved sustainable agriculture is that their agricultural practices have more successfully harnessed the benefits of mutualism – the advantages of cooperation between ants and termites and the fungus they raise and feed on. Our mutualism around annuals is more ecologically problematic due to soil erosion, pest control, and a number of other challenges. Cooperation between unrelated agents can have great benefits, but there is always tension between the benefits of cooperation and the benefits of defecting – the classic Prisoner's Dilemma (Trivers Reference Trivers1971). Social insects have largely overcome Prisoner's Dilemma–type reproductive conflicts by developing mutualism because one-to-one cooperation is more stable (easier to enforce) than cooperation among several agents. Ant feces provide critical nutrients (especially nitrogen) without which the fungus could not survive. Not only is mutualism between agriculturalists and their crops taken to the point where one could not survive without the other (as with humans), but in ants it is taken to the point where their physical bodies have adapted to the physical needs of the fungus. Foster and Wenseleers (Reference Foster and Wenseleers2006) show that three key factors are important to mutualism evolution: (1) high benefit-to-cost ratio, (2) high within-species relatedness, and (3) high between-species fidelity. The economic drivers behind the success of managed agriculture likely raised the benefit-to-cost ratio of mutualism.

Regarding sustainability, an important difference between human and insect societies is social instability. Unlike insect societies, human groups are characterized by recurrent and sometimes calamitous within-group conflicts. Georgescu-Roegen, following Lotka, attributes this to the difference between “endosomatic” and “exosomatic” organs (Georgescu-Roegen Reference Georgescu-Roegen1977a; Lotka Reference Lotka1925/1956). The occupations of social insects are determined in large part by their phenotypes – for example, door keeper ants have large heads (endosomatic) which serve no other purpose than to block the entrance holes to the colony. Humans, by contrast, depend on the objects of material culture (exosomatic), which can be appropriated according to cultural rules that may or may not be accepted. An ant is born to be a soldier; a human is not. There is no biological reason for one human to be homeless and another a billionaire.

5.2. The loss of individual autonomy in ultrasocial societies

Another basic feature of ultrasociality is the subjugation of the individual for the evolutionary success of the group. Subjugation can be understood partly as an artifact of the division of labor in agricultural societies. An important commonality in human and ant division of labor in ultrasocial agricultural systems is that individual behavioral complexity and flexibility are in general not as great compared to those societies relying on hunting and gathering. The behavior of individuals, in the context of an elaborate division of labor, is simpler in ultrasocial societies even as the society grows more complex. In the case of ants, increasing social complexity is not associated with increasing individual behavioral complexity (Holbrook et al. Reference Holbrook, Barton and Fewell2011). Adam Smith (Reference Smith1776/1937, p. 734) recognized the danger for humans of labor specialization and the mental toil on individuals who endlessly perform the same tasks:

Commenting directly on Smith's observation, Moffett (Reference Moffett2010), in reference to ants, writes: “This deficiency can be observed for large ant societies as well, in which specialized workers are incapable of accomplishing much without the cooperation of nestmates” (p. 70). Likewise, Anderson and McShea (Reference Anderson and McShea2001) found that “individuals of highly social ant species are less complex than individuals from simple ant species” (p. 211). They found that individual ants in more complex ant societies with a high degree of division of labor exhibit “low individual competence” and “low individual complexity.”

Increasing social complexity in ants is associated with a loss of brain size. Riveros et al. (Reference Riveros, Seid and Wcislo2012) tested the association between brain size and sociality across 18 species of fungus-growing ants and found that increased colony size was associated with decreased relative brain size.Footnote ²³ With agriculture, human brain size began to decrease dramatically. According to Hawks (Reference Hawks2011), the decrease in brain size during the last 10,000 years is nearly 36 times the rate of increase during the previous 800,000 years. There is an important distinction between the “social brain” and “social intelligence” and “collective intelligence.” We fully agree with the “social brain” hypothesis that human intelligence evolved to facilitate within-group cooperation, empathy, and mind reading (Frith & Frith Reference Frith and Frith2010; Wexler Reference Wexler2006). Collective intelligence, on the other hand, refers to the ability of groups to solve complex problems far beyond the capabilities of any individual within the group. Collective intelligence can increase while individual intelligence declines.

Individual simplicity may be an advantage in collective decision making. In fact, the standard economic model of extreme rationality may apply more to ant colonies than to humans. For example, a number of experiments show that humans are susceptible to the fallacy of irrelevant alternatives. A choice between alternatives A (a fully paid 10-day vacation to Paris) and B (a fully paid 10-day vacation to Rome) should not be influenced by the inclusion of an irrelevant unrelated choice C (a paid vacation to your least favorite nearby city). Humans are consistently susceptible to this fallacy (Ariely Reference Ariely2008), but ants are not. Edwards and Pratt (Reference Edwards and Pratt2009), in an experiment involving the choice of an ant colony between nesting sites, showed that the colonies are not influenced by irrelevant alternatives. They surmise: “We suggest that immunity from irrationality in this case may result from the ants' decentralized decision mechanism. A colony's choice does not depend on the site comparison by individuals, but instead self-organizes from the interactions of multiple ants, most of which are aware of only a single site” (Edwards & Pratt Reference Edwards and Pratt2009, p. 3655). In some species of ants, the colony solves very complex problems of economic organization, and in fact it may outperform humans (Edwards & Pratt (Reference Edwards and Pratt2009). Collectively, ants have developed complex strategies to manage agricultural production. Like humans, they have developed a number of herbicides to control weed molds that attack the fungus they rely on, they have elaborate manuring regimes that maximize harvests, and cultivars are shared between distantly related ant colonies (Mueller et al. Reference Mueller, Rehner and Schultz1998).

The social insects demonstrate that collective intelligence can be quite impressive without a corresponding level of individual intelligence. Off-loading of tasks in complex human societies is one explanation given for the decline in human brain size after the Pleistocene (Geary & Bailey Reference Geary and Bailey2009). People may not have to be as smart to stay alive. Cognitive scientist David Geary refers to this as the “idiocracy theory” after the 2006 film Idiocracy (McAuliffe Reference McAuliffe2010). Geary and Bailey (Reference Geary and Bailey2009) and Mithen (Reference Mithen2007), among others, argue that the complex material culture that came with agriculture allowed humans to offload some cognitive requirements, allowing the energy-expensive human brain to decrease in size.Footnote ²⁴ Off-loading individual intelligence to the “environment,” “technology,” or the “social brain” is not necessarily a good thing from the point of view of an individual human. Social intelligence increased, but individual intelligence may have declined (see Barrett et al. Reference Barrett, Henzi and Rendall2007). This hypothesis fits nicely with our point that a major consequence of ultrasociality is a loss of individual autonomy and, possibly, cognitive capabilities. According to Mithen (Reference Mithen2007, p. 705), even the social brain may have deteriorated with agriculture and civilization:

A basic categorical mistake is to conflate the collective accomplishments of civilization with the understandings and intelligence of the average human – as in “we” formulated the theory of relativity, or “we” put a man on the moon. Scientific understanding of the origins of the universe and our species, the works of Shakespeare and Mozart, space exploration, and so on, are equated with human intelligence even though most people on the planet are unfamiliar and little affected by these achievements.

Intelligence, both social and collective, may be related to group size. Dunbar (Reference Dunbar1993, p. 681) suggests that cognitive constraints imply a consistent group size for effective human communities: “There is a species-specific upper limit to group size that is set by purely cognitive constraints.” Effective group size is limited by the maximum number of individuals with whom a person (or animal) can maintain social relationships by personal contact. For humans this maximum number is somewhere around 150–200 individuals. Naroll (Reference Naroll1956) presented evidence for a “critical threshold at a maximum settlement size of 500, beyond which social cohesion can be maintained only if there is an appropriate number of authoritarian officials” (quoted in Dunbar Reference Dunbar1993, p. 687). The size of the neocortex increases with group size – but only up to a point.Footnote ²⁵ Dunbar's argument does not contradict the evidence that human brain size, and by implication cognitive ability, decreased after agriculture.

5.3 Control without hierarchyFootnote ²⁶

Ultrasociality channeled the existing propensity to cooperate in a new direction (Gowdy & Krall Reference Gowdy and Krall2014). For example, sociality, caring for others, and cooperation with non-kin are defining characteristics of the human species (Frith & Frith Reference Frith and Frith2010). These traits not only made it possible for humans to flourish and survive the extreme environmental changes of the Pleistocene, but also they fostered sustainable use of environmental resources and equalitarian social arrangements (Boehm Reference Boehm1997; Pennisi Reference Pennisi2014). In non-ultrasocial groups, these traits worked both for the benefit of the group and for individuals within the group. Small-scale human societies have developed myriad forms of social organization to minimize group conflicts and to ensure that one individual or one small group of individuals cannot dominate (Boehm Reference Boehm1997; Wilson et al. Reference Wilson, Ostrom and Cox2013). Woodburn (Reference Woodburn1982, p. 438) writes of immediate return (simple technology and material culture) hunter-gatherers:

But sociality and cooperation take on a different character with ultrasociality. Cooperation and coordination of activities in ultrasocial societies subjugate the individual to further the needs of the ultrasocial entity. The emergence of ultrasociality leads directly to a loss of autonomy at the individual level because autonomy interferes with the coordinated functioning of the group (Anderson & McShea Reference Anderson and McShea2001, p. 219). There is general recognition that selfish behavior can subvert the common good. In evolutionary terms, adaptation at any given level in the MLS hierarchy tends to be undermined by what happens at lower levels (Wilson Reference Wilson2013). What is not generally recognized is that what is good for the higher level may be not be good for entities at the lower level. For humans, the most social of human activities is the reproduction of material life, but the productive configuration of society changed with the active management of crops. In a sense cooperation was co-opted with agriculture and rigidly structured around narrowly defined productive roles (Gowdy & Krall Reference Gowdy and Krall2013). With human ultrasociality the terms “prosocial” and “contrasocial” lose definition. When food production becomes the organizing principle of a society, the “good of the group” becomes “the good of the ultrasocial entity,” not the good of the average member of the group. What is good for the higher-level entity may be bad for individuals at the lower level. Of course, “good” and “bad” are human value judgments that cannot be applied to ant societies.

When the evolutionary leap to ultrasociality is made, individual survival (and by extension to humans, individual well-being) becomes secondary to the survival of the group as an evolutionary entity. The selective “pull” of the group over the individual becomes greater with increasing complexity. In an ultrasocial system, there is no reason why specific individuals should be more likely to survive. Like cells in a body or bees in a hive, particles are there to serve the collective. In ants and termites, serving the collective has implications for reproductive fitness. In the most extreme examples, there are sterile castes. In humans, the implications are different because sacrifice for the group is not expressed reproductively.Footnote ²⁷ Rather, it is measured in the extreme interdependency that delineates and reduces individual life to the role that enables the system itself to be reproduced. When the group begins to take on a life of its own and actively begins to shape its environment,Footnote ²⁸ individuals are expendable. The group is the organism subject to the imperative to survive. Bees sting to defend the nest, thereby sacrificing themselves for the good of the group. In some species of ants, soldiers have such large mandibles that they cannot feed themselves. Individuals are expendable for the good of the group.Footnote ²⁹

In both ultrasocial human and insect societies there is a loss of totipotency, defined as “the potential, throughout life, to express the full behavioural repertoire of the population (even if never actually expressed)” (Crespi & Yanega Reference Crespi and Yanega1995, p. 110). The term is meant to capture the range of behavior in a society (occupations, for example) compared to the range of behaviors available to a particular individual. It has been noted that workers in complex insect societies tend to be less totipotent (Anderson & McShea Reference Anderson and McShea2001). Studies of social insects and colonial marine invertebrates (reef shrimp) show a negative correlation between colony size and totipotency (Burkhardt Reference Burkhardt1998). In humans, the loss of totipotency is more complex. It is expressed in the more extreme specialization that attends ultrasocial social organization, but it is also expressed by the fact that for the majority of humans, agricultural life became narrowly focused around a single economic purpose.

Interestingly, Adam Smith (Reference Smith1776/1937, p. 735) also discussed the loss of what might be called human totipotency in complex societies as compared to what he called “barbarous” societies:

Agriculture entailed an altered organization of economic life that changed the relationships among individuals within the group and the relationship of both the individual and the group to the biophysical world. Once in place, the economic factors driving efficiency in production gave a competitive advantage for those species that could best capture them. The evolution of state societies was reinforced by a process of “downward causation” (Campbell Reference Campbell, Ayala and Dobzhansky1974; Sperry Reference Sperry1969). Campbell (Reference Campbell, Ayala and Dobzhansky1974, p. 180) writes: “Where natural selection operates through life and death at a higher level of organization, the laws of the higher-level selective system determine in part the distribution of lower-level events and substances.… All processes at the lower levels of a hierarchy are restrained by and act in conformity to the laws of higher levels.” Those societies having group traits most favorable to surplus production outcompeted other groups. The needs of the higher-level entity began to mold the behavior, organization, and functions of lower-level entities. Some possible consequences of this for present human societies are discussed in the next section.