Part 1 of A Crisis in the Neolithic, by Robin Dunbar [WHY Religion Evolved]
CHAPTER 8
A Crisis in the Neolithic
From around 12,000 BP, new developments in the Middle East and elsewhere presage the arrival of the Neolithic, or New Stone Age. Settlements begin to appear in the form of clusters of stone-lined pit houses with clay floors. Initially, these settlements were small in size, but over a matter of a thousand years or so they increased in size dramatically. By 11,000 BP, Jericho amounted to some seventy dwellings, suggesting a population of around 300–400, and by 9,000 BP Çatalhöyük in Turkey was large enough to accommodate a population of 5,000–7,000 people.
Although the very earliest settlements do not suggest any significant change in style of religion, excavations at Göbekli Tepe and Çatalhöyük in Turkey suggest that by 9,500 BP, major new developments were under way. Some parts of these sites seem to have a religious or ceremonial function rather than being residential. At Göbekli Tepe, for example, there are several hundred massive pillars weighing 10–20 tons, many engraved with reliefs of animals. It has been estimat- ed that it would have taken a community of 500 or more to move the pillars the 500 metres from the quarry to their final position. One does not go to these lengths for conventional dwellings. At Çatalhöyük, the dead were regularly buried under the floors of the houses. The walls were hung with the plastered skulls of animals, and some of the walls were painted with scenes of hunting and other activities. Stone-carved or clay-baked animal heads and human figurines are common.
The subsequent pace of development was remarkably rapid. By 3000 BC (5,000 years ago), the Naqada culture had developed into a formal state uniting Nubia and Upper Egypt, and the Akkadian empire had been established in Mesopotamia; the first of the Chinese empires, the Xia Dynasty, was established in central China less than a millennium later. This dramatic change in the demographic organization of populations would have had very significant effects on how individuals related to each other, and it is this, I argue in this chapter, that triggered the rise of doctrinal religions. To understand how and why this happened, we first need to consider the consequences of moving from a dispersed hunter-gatherer lifestyle to an urban one.
The Transition to Village Life
Traditionally, the Neolithic Revolution is associated with the mastery of cultivation, and a settled lifestyle has usually been seen as necessary for the provision of a workforce for these purposes. This is almost certainly wrong, for three reasons. First, large workforces were not needed for agriculture until industrial-scale farming to service external markets developed, and these did not emerge until many thousands of years later. Subsistence-scale farming requires only the labour of a single family, even now.
Second, archaeologists now agree that the earliest evidence for crop-growing predates the first Neolithic settlements by several thousand years. More importantly, during the early phase of settlement, a hunting and gathering economy was still the norm. Aurochs (wild cattle), gazelle and other wild animals were hunted and wild cereals gathered, although these practices declined over time with the domestication of cereals and livestock.
Third, despite the widespread assumption that agriculture was good for you because it yields a significant improvement in nutrition for less effort, it seems that the opposite was in fact true. Evidence from physiological stress signs in skeletons and calculation of the energy throughput for hunter-gatherer and early farmer populations living in the same region show rather starkly that the farmers were under much greater nutritional stress than the hunter-gatherers. Farming was extremely hard work and subject to the vagaries of climate as well as the depredations of wildlife and pests. It seems these populations had little choice: they could not pursue a hunter-gatherer life even if they wanted to but instead were obliged to live in settlements and put up with the significant costs of doing so.
In short, people did not live in villages in order to develop agriculture, but developed agriculture in order to live in villages, at least once these settlements achieved any significant size.
Seeing the sequence this way round raises the obvious question as to why people made this sudden transition, and why they did so as rapidly as they evidently did. The answer appears to be defence against raiders. Human populations seem to have been growing rapidly just before and into the early Neolithic, especially in the northeast corner of Africa and the Near East. That raiding by neighbours was a problem is obvious from the fact that most settlements were built on hilltops or rocky promontories that provide a good all-round defensive position. In addition, the entrances to the houses in these early Near East settlements were almost always from the flat roof (a practice that persisted well into biblical times in the Middle East), with just a few small, mainly inward-facing windows at ground level. Some of these settlements had substantial brick-built surrounding walls – recall the famous walls of Jericho which stood some 12 ft (3.6 m) high and 6 ft (1.8 m) wide at the base. These were costly to build in terms of manpower and hardly likely to be purely decorative.
The use of defensive positions was widespread throughout the Neolithic and into historical times. Well-known historical examples include the thousands of Iron Age hill forts that dot the landscape of northwest Europe, the mesa-top villages of the Hopi and other pueblo Indians of the American Southwest, and the conquistador-period Inca hilltop cities like Machu Pichu in Peru.
Although archaeologists have sometimes been surprisingly coy about admitting that these sites had anything to do with defence, in fact it is pretty obvious that this was why these settlements were built where they were. In fact, we know this is so from the historical record. At the end of the first millennium AD, the ancestral Anasazi Indians of the American Southwest deliberately moved their villages from the rather open, vulnerable positions they had previously occupied to more defendable positions on nearby mesas or ledges in canyon walls in direct response to increasingly intense raiding by Comanche and Navajo as these northern tribes began to expand southwards into their territory. Many of the thousands of pueblo ruins in the area are on ledges in canyon walls that could only be accessed by rock climbs or the use of ropes, with overhangs that made them all but impregnable from three sides as well as above. We know from Spanish accounts that, during the Pueblo Revolt against their Spanish overlords in 1680, the Zuni abandoned their six large scattered villages and relocated to the 7,300 ft (2,200 m) Dowa Yalanne mesa (Corn Mountain), and remained there for three years until a peace treaty had been signed. The Spanish records tell us that they had used this same refuge a century earlier in 1540 when the conquistador Vázquez de Coronado undertook an expedition to annex the lands north of their Mexican territories.
Similar observations come from Africa. Sir Kenneth Bradley noted that, while a young administrative officer in Northern Rhodesia (now Zambia) in the 1920s, a Chewa headman then in his eighties told him that his uncle, who had been headman many decades earlier, had moved their villages from the more fertile riverine plain where they had previously lived into the hills above because of marauding Nguni impis (war parties). After suffering a catastrophic defeat at the hands of Shakar’s Zulu confederation during the Zulu civil wars of the 1830s, the Nguni Zulu clans fled north of the Zambezi River in 1833 and spent the following decades ter- rorizing the local tribes in what is now Zambia and Malawi. Many of the Chewa were enslaved by the Nguni. The neighbouring Tumbuka had also fled into the hills in the face of Nguni marauders and did not reoccupy their ancestral lands in the plains until after 1907 when the newly established British colonial administration put an end to Nguni raiding and slaving. Likewise, the Fali and other peoples of the Mandara Mountains on the Nigeria/Cameroon border had retreated from the richer plains into the less productive mountains for protection from the horse-based slaving parties of the Fulani, Wandala, Shuwa Arab and Kanuri that continued to harass them (for both domestic slavery and export across the Sahara to the slave markets of North Africa and Arabia) until as late as the 1920s – fully a century after the Atlantic slave trade had been suppressed by the Royal Navy.
We tend to think of our prehistoric ancestors as living an idyllic life, quietly going about their business with only the occasional hunting expedition to excite them. In fact, nothing could be further from the truth. There has been a constant movement of peoples on every continent for at least the last 70,000 years, and as time progressed these migrations increasingly resulted in conflict between different tribal groups. That these interactions were serious is evidenced by the Crow Creek massacre that took place in South Dakota during the 1320s AD, when an entire village of Caddoan Central Plains Indians was massacred by their Mandan neighbours. Of the 486 bodies of men, women and children that have been recovered by archaeologists from this site, over 90 per cent had been scalped (as evidenced by cut-marks on the sides of the skulls). In addition, many had had their lower arms and tongues removed (another common form of trophy-hunting among Native Americans that persisted well into modern times). The Caddoans had been recent invaders from the south, and had clearly felt sufficiently threatened by the local Mandan that they were in the process of repairing their village’s defensive dry moat fortification when they were overwhelmed.
Similar massacre sites have been discovered in many parts of Central Europe, dating to periods around 7000 BC. Most appear to involve massacres of entire bands of early European hunter-gatherers by farmers working their way westwards from Anatolia. The burial sites include men, women and children, with clear evidence of violent death from blunt instrument trauma to the side or back of the skull and casual rather than respectful disposal of the bodies. In around half the skeletons at one site, shin bones had also been deliberately broken before death (presumably to prevent escape).
At many of these burial sites, there is a conspicuous absence of young women, suggesting that these had been separated out and taken away as sex slaves – a practice that has been, and continues to be, characteristic of societies all over the world. Women of reproductive age were a major prize of raiding among Amazonian Indians even as late as the 1960s. More remarkable is the fact that we can sometimes detect the genetic echoes of these events many centuries later. Eight centuries after Genghis Khan’s conquests of Central Asia and Eastern Europe, around 7 per cent of all males currently alive within the geographic range of the thirteenth-century Mongol empire have the Y-chromosome genetic signature of the Khan and his generals. Similarly, the genetic signature of the fifth- century Anglo-Saxon invaders is still evident in the contemporary English population: while the women on the east side of England have native Celtic mitochrondrial DNA, the men typically have Anglo-Saxon Y-chromosome (paternal) signatures. We know from the historical records that the Anglo-Saxons operated a ferocious form of apartheid that, until the legal reforms of King Alfred the Great four and a half centuries later, actively discriminated against the Romano-British natives, including denying them legal status and allowing Saxon men to take both the property and women of the British natives with impunity. Similarly, while 85 per cent of living Icelandic men have the expected Norse male genetic signature of the population that originally colonized the island, 50 per cent of Icelandic women have a Celtic maternal signature, reflecting the historically well-documented fact that the Norse regularly collected Irish and Scots slave girls on their way over to Iceland. Indeed, slaving on the British mainland by Irish raiders continued until well into the twelfth century, notwithstanding explicit attempts to end it at the Council of London in 1102 AD.
In other words, protection from attack by raiders has been a very serious issue for a very long time, and living in settlements provided the best defence. The problem, however, is that living in groups of any size is stressful.
From time to time, the San Bushmen feel the need to dispel what they call ‘star sickness’. This is a mysterious force that takes over a group of people, causing jealousy, anger, quarrels and the failure of gift- giving. These pressures pull people apart and damage social cohesion. Trance dancing mends the social fabric because it assuages hostility. In addition, the stresses created by social friction can have a dramatic effect on female menstrual endocrinology, causing temporary infertility. This is a general problem for all mammals and is a major factor limiting the size of the social groups of mammals in general, primates in particular and, by extension, humans.
In an analysis of the causes of disputes among San in Namibia, the most common causes of complaint were the behaviour of inconsiderate trouble-makers, jealousy over possessions, persistent stinginess or failure to share, inappropriate sexual behaviour and the failure to fulfil obligations towards kin. Ostracism was the ultimate, albeit rarely used, sanction that, in such societies, was equivalent to a death sentence since individuals rarely survived when living alone. Most cases of ostracism involved behaviour, and especially sexual behaviour, that was considered socially disruptive. Under pressure from criticism by other members of the community, one woman who had had frequent sexual relationships with neighbour- ing Bantu men left the band and subsequently died. Another case involved the expulsion of a family in which the wife was frequently drunk and had promiscuous sexual relations with Bantu men; and, to cap it all, the children were unruly. Eventu- ally, after the entire community had ganged up against them, the family left to live elsewhere. They were allowed back only after the woman had died.
In any society, these kinds of petty, or not so petty, disagreements eventually boil over and lead either to direct violence or to accusations of witchcraft, which in turn invite retributive justice. Nearly a third of all recent historical deaths among the Gebusi horticulturalists of Papua New Guinea were homicides. While Gebusi men were more likely to be killed during intergroup disputes and raids, women were more likely to be killed as a result of within-community disputes, many of which involved accusations of witchcraft. In hunter-gatherer societies, the proportion of deaths that are due to violence increases linearly with living-group size. In groups of fifty, violent deaths account for almost half of all mortality. No population can cope with demographic pressures on this scale. The flexible social arrangements of hunter-gatherers normal- ly allow them to manage this so as to prevent things getting completely out of hand: if relationships become too fractious, a family can leave their camp group and move in with another group with whom they have connections. It may, hence, be no accident that the upper limit on living-group size in contemporary and historical hunter-gatherers is around fifty.
Because settlement results in the whole of the community or tribe being concentrated in one place, living in permanent villages and towns exacerbates this problem. Farmers living in villages of several hundred individuals experience significantly higher rates of homicide than hunter-gatherers, much of it due to within-community disputes. In small-scale communities, fighting is much more likely to get out of hand because there is no police force to step in and resolve disputes. Once things get to this stage, views polarize rapidly along lines that have little to do with right and wrong and much more to do with personal and family loyalty. In a famous analysis of an axe-fight among Yanomamö Indians in Venezuela, the American anthropologist Napoleon Chagnon found that villagers rapidly sided with one or other of the main protagonists in a dispute and did so mainly on the basis of whom they were most closely related to or whom their own close relatives had sided with. Fights force people to take sides once they become serious enough to threaten injury or death, and that inevitably ruptures the delicate bonds that hold the community together.
The stresses of living together often emerge in conflicts between the women in the group. While studying the Siuai on Bougainville Island in Micronesia during the 1950s, the anthropologist Douglas Oliver was told that villages invariably underwent fission when they exceeded nine households, mainly as a result of tensions created by quarrels among the women. The effects of this can even be detected within polygamous households. In tribes that practise sororal pol- ygamy (all the wives are sisters), the women commonly live together under the same roof; however, when the women are not related, each wife usually has her own hut in the husband’s compound. If they do not, relationships between them become too fractious.
The need to find ways to reduce the levels of stress and internal violence as villages increased in size was crucial if humans were to live in larger settlements. Tribal societies that live in large settlements typically exhibit a variety of strategies to hold these disruptive forces in check. These include increased frequency of community bonding activities (dances, feasts), more formal marital arrangements involving the ex- change of bride-wealth between contracting families, a shift from a democratic social style to male hierarchies and formal leadership roles . . . and a switch to doctrinal religions with more explicit rituals, formal religious spaces and religious specialists.
The Rise of the Doctrinal Religions
Ritual spaces, symbolic representation of religious ideas, a priestly caste, gods and moral (as opposed to social) codes are all absent from modern hunter-gatherer societies but all appear very rapidly in urban settings during the course of the Neolithic.
Some buildings at early sites like Jericho and ’Ain Ghazal (about 50 km east of Jericho) have been interpreted as having ritual functions on the grounds that they have unusual apsidal or circular shapes, are much smaller (sometimes just 2–3m in diameter) than the other domestic buildings, and contain what appear to be basins and altars, as well as under-floor channels (to allow blood from sacrifices to drain out?). At ’Ain Ghazal, four of these ‘ritual’ buildings have been found spanning a 500-year period, only one of which was in active use at any one time.
Most of the domestic dwellings in these early Levantine Neolithic settlements are also associated with under-floor burials, with the bodies sometimes standing vertically, some- times trussed into a crouching position. Some of these burials seem to have been so old that later householders were un- aware of them and broke through, damaging the bones, when digging a grave for a later burial. Throughout the Levant as far east as Iraq there was a distinctive ‘cult of skulls’: in some cases, for example, skulls were removed from corpses and ar- ranged around a central pit in a mortuary space on the edge of the village or in niches within a building. Some skulls had their faces reconstructed in plaster in eerily lifelike masks.
It is difficult to be sure when the various components of doctrinal religions first emerge. What we do know is that by 2000 BC in Sumeria, as well as around the same time in Egypt’s Old Kingdom, we have evidence for a priestly caste. In some cases, these might be priestesses, many of whom were of noble birth, who often functioned as sacred prostitutes at temples. For example, Enheduana (the earliest poet known to us by name) was the High Priestess of the goddess Inanna and the moon god Nanna-Sin in the Sumerian city-state of Ur in the decades before 2250 BC. She composed the 153-line ‘Nin-me-Šara’ (Exaltation of Inanna) as well as a body of hymns of varying length, of which forty-two have survived. Enheduana was of royal blood, and seems to have been one of the naditu priestesses (temple slaves) recruited from high-born families. Inanna was worshipped in the Sumerian city of Uruk from as early as 4000 BC, but came to particular prominence and popularity as Ishtar, the goddess of love, beauty, sex, war, justice and political power, after the establishment of the Akkadian empire around 2300 BC. Known as the ‘Queen of Heaven’, she was not only the highest ranking deity in the Assyrian pantheon but continued to be worshipped until as recently as the eighteenth century in some parts of Mesopotamia.
This period also seems to coincide with the appearance of formal rituals. One characteristic of formal rituals is that they have to follow a prescribed script so as to be performed in exactly the same way on every occasion – otherwise they lose their ‘power’. One of the earliest known examples of this are the so-called ‘Pyramid Texts’ that were carved onto walls and sarcophagi in the pyramids at Saqqara during Egypt’s Fifth and Sixth Dynasties (2686–2181 BC). They specify prayers, spells and instructions to ease the deceased’s way into the afterlife, many intended to be uttered ritually by priests. It has been suggested that the wording of some of these texts may even date back as far as the Second or Third Dynasties (2890–2613 BC).
The earliest forms of ritual seem to be associated mainly with propitiating gods whose interest in human affairs is often capricious and vindictive. In many traditional religions, the practice of making libations, offering farm produce or sacrificing animals is near-universal. Animal sacrifice was common in First and Second Temple Judaism right up to the time of Jesus, as was offering of vegetables and grains – something that, arguably, persists in modern Christianity in the form of harvest festivals where rural churches are decorated with produce from the field.
Ritual sacrifice may even extend to human victims. Sometime during the first century AD, Celts in the northwest of England gave a young man in his twenties a last ritual meal of charred bread, then garrotted him and cut his throat, finally depositing him naked in what at the time was a remote bog now known as Lindow Moss. Since the discovery of Lindow Man (or Pete Marsh as he is often irreverently known) in 1984, a total of 140 bodies have been recovered from bogs in Britain, all dating to around the same period. A large number of similar ‘bog people’ have also been found in Denmark, indicating that it was a widespread practice in northwest Europe at the time. Perhaps significantly, this period coincided with the beginning of the Roman occupation of Britain and continental Europe north of the Rhine, raising the possibility that these sacrifices might have been an attempt to propitiate the gods in the face of major societal disruption.
Religion and the Rule of Law
In the earlier sections, I suggested that there is a natural progression from informal religions in small-scale societies to formal religions in large-scale societies as a way of managing the stresses involved. This development can be traced in some detail in South America. A reconstruction of village sizes on the Taraco Peninsula in Lake Titicaca, Bolivia, between 1500 and 250 BC (the period leading up to the formation of the Tiwanaku state) found that, in the early phase when the economy was largely subsistence agriculture, average village size was about 127 people and villages tended to split up when they reached 170 in size – at almost exactly the same size as they did three millennia later among the Hutterites in North America. By 1000 BC, average village size had increased to around 275, with as many as a quarter of the villages containing more than 400 people. By about 800 BC, this had given rise to a fully fledged religious complex known as the Yaya-Mama religious tradition. This included a novel form of ceremonial public space (a plastered sunken court), decorated serving bowls, ceramic trumpets, incense burners, and a distinctive style of stone sculpture.
The American Plains Indians provide us with another example because some of them switched from one state to the other and back again on a regular annual cycle. For most of the year, these tribes lived in small mobile bands of half a dozen families (perhaps 30–35 people), each with its own chief but little or no formal ceremonial. During the annual buffalo hunt, however, all the bands of a tribe came together in a single encampment of several thousand individuals. During this period, and only during this period, the chiefs of the individual bands formed an all-powerful council, with one member elected as the paramount chief. In addition, there were men’s secret societies, religious ceremonial and a formal police force responsible for enforcing strict adherence to rules of behaviour. None of these occurred, or were needed, during the dispersed phase.
While the imposition of top-down discipline by a police force is certainly effective in managing disruptive behaviour among community members (especially the young males), it is always more effective to do this bottom up by making the individual feel committed to the community. In essence, this is how religions work at their best – by creating a sense of belonging through shared beliefs and shared rituals. In effect, the Seven Pillars [discussed in an earlier chapter] constitute a totem pole at the centre of the village on which everyone can voluntarily hang their hats. If to this we add both Moralizing High Gods who monitor what humans are up to and religiously justified moral injunctions (such as the Ten Commandments), then we create a very powerful carrot-and-stick effect.
In his analysis of the social and political correlates of Moralizing High Gods, Dominic Johnson found that these kinds of gods (the kind found only in doctrinal religions) are significantly more likely in societies that live in large communities and those that have ‘jurisdictional hierarchies above the community level’ (politically more complex societies in which an elite rules over the community). In other words, organized religions seem to have been part of the machinery used to keep the lid on fractiousness so as to allow larger communities to exist.
Attempts to identify when and why Moralizing High Gods (that is, those that take an active interest in human behaviour) emerged have focused mainly on either economic or demographic contexts. As we noted [earlier in the book], Hervey Peoples found that belief in Moralizing High Gods was more or less confined to tribal groups with a pastoralist or agricultural economy – in other words, property-owning societies. In her view, Moralizing High Gods emerged as an adjunct to allow elites within society to maintain control over the product of others’ labour, and this only became feasible with herding (most pastoralists typically have very large flocks and herds because they extol herd size as a measure of status) and intensive agriculture.
However, this explanation seems to put the cart before the horse: for most pastoralists, herding is a family business not a community arrangement and can be managed within the normal obligations of family relationships. The biggest problem that herders face is defence against raiding by other pastoralists. The mobility of stock animals means that pastoralists the world over face a constant threat of rustling that invariably results in significant loss of life (and women) as collateral damage.
Banding together in communities large enough to provide sufficient numbers of warriors to protect and avenge these raids is the only viable solution. It is to ensure community cohesion for mutual protection that High Gods are necessary, not to arrange herding duties or prevent thefts within the community.
A number of recent analyses have focused on whether belief in High Gods arose before or after the emergence of political complexity. Many studies differentiate between two separate phenomena: supernatural punishment and Moralizing High Gods. Supernatural punishment is usually associated with rituals to appease capricious deities who have little interest in the moral behaviour of humans, but require sacrifices and other forms of invocation designed to allay their wrath. An example would be the religion of the Aztecs, which required very large numbers of victims (typically war captives or slaves) to be sacrificed (sometimes by having their hearts torn out while alive) to deities such as Huitzilopochtli, the bloodthirsty god of war. The Aztec rain god Tlaloc was particularly fond of child sacrifices as he required their tears to make rain. In contrast, Moralizing High Gods usually take an active interest in the wellbeing and behaviour of their human worshippers, rewarding those that adhere to the rules they lay down (usually via some kind of revelatory process) and punishing (now or in the world after death) those who don’t. The Abrahamic religions provide the classic examples.
Religions with capricious deities tend to view the performance of the relevant rituals as a communal duty, with divine punishment being meted out indiscriminately on the whole of society. Moralizing High God faiths tend to view divine retribution as a personal matter, reflecting the individual’s performance over the course of their life.
The close correlation between the emergence of human sacrifice and the suite of rituals associated with it, and the switch from simple to stratified societies in the Austronesian cultures that I mentioned in [an earlier chapter] provides an example of this process in action. Cultures that adopted sacrifice as a new ritual complex were able to switch to having stratified social structures, which in turn allowed them to increase the size of their populations. The analyses were very clear: sacrifice emerged before stratification. In effect, sacrifice and the rituals that make up doctrinal religions provide the gateway for increasing social complexity (and hence population size). In part, this comes about by imposing draconian discipline on those who transgress against the rules. If you don’t develop the necessary rituals, your social groupings will necessarily remain small and dispersed.
In this context, human sacrifice seems both to raise the psychological ante for being a member of the community and acts as some kind of transcendental threat (our god is very demanding, so beware lest you become the victim).
An alternative approach is to examine the actual sequence of historical changes through time at a particular location. This provides a different perspective on the question because it allows us to map the acquisition of different rituals and beliefs onto the changes in the demographic and economic state of the community as it developed from a relatively simple settlement through a classic city state to a large empire. Three recent studies have used this approach to try to determine when High Gods emerged, and why.
One study collated data from over 300 societies covering the last 10,000 years and mapped the changes in social complexity along with the appearance of beliefs in supernatural punishment and High Gods (or, as they term it, ‘moralizing supernatural punishment’). Social complexity was indexed by a rather eclectic mix of traits ranging from population size, hierarchical and legal structures, infrastructure (such as canals and roads), a calendar, writing and metal coinage. They found that Moralizing High Gods typically appear around 300 years after a society reaches its peak of structural complexity. They argued that there was no direct relationship between the two variables, but rather that both were a response to the rise of more sophisticated military technology (notably cavalry) and agriculture (including, to a lesser extent, pastoralism). The unanswered question here, however, is whether having a sophisticated military forces you to have a complex religion or whether militarization and religion are a correlated response to being constantly attacked by others.
More importantly, perhaps, there are two separate causal routes to any causal hypothesis: complex societies can only evolve if they first evolve complex religions or complex religions evolve after a complex society has evolved as a means of stabilising society. Ideally, what we need to know is whether those states that failed to acquire a complex ‘High God’ religion broke up significantly faster than those that did.
Either way, the data clearly point to a population of around one million people as the key threshold at which High Gods appear, indicating that they are associated with empires rather than city-states and so are likely to be associated with managing very large scale sociopolitical stresses.
A smaller-scale study of the co-evolution of High Gods and social stratification provides some further light on this process. This study suggested that, in Austronesian societies, supernatural punishment is an immediate precursor of the evolution of social stratification, with Moralizing High Gods appearing after social stratification. Few of the societies in this study had more than 100,000 members, compared to the figure of a million associated with the appearance of Moralizing High Gods suggested by the previous study; indeed, few of them actually had Moralizing High Gods.
This suggests that there may well be a series of demographic glass ceilings at the high end just as there is at the low end. At each stage, a new mechanism is needed to allow the society to break through the constraint and increase its size to the next level. In these kinds of analyses, it is crucial to get the causal logic the right way round.
The causality really has to be societal complexity driving the evolution of Moralizing High God religions as a stabilizing force, not High God religions driving the evolution of complex societies as some kind of inevitable outcome. In evolution as in real life, solutions come after the problem, not in anticipation that there might be a future problem.
The handful of major religions that now dominate the world all have their origins in the Yellow and Yangtze River region in China, the Ganges valley in northern India and the eastern Mediterranean. Why all these large religions should have appeared at about the same time (the period known as the Axial Age during the first millennium BC) in these geographically widely separated locations has puzzled historians for a very long time.
In fact, a third study suggests that the best predictor of the switch to the Axial Age state is per capita annual energy production (essentially agricultural output), with a threshold at around 20,000 kcals per person identifying the transition point. This probably represents the level of production that creates agricultural surpluses sufficient to feed large cities at the centre of an empire (for example, Rome). The next-best predictor was population density (itself a correlate of energy surplus, of course). Although population size, the size of the principal city and the geographic size of the state individually had negligible effects, in reality these almost certainly form a single interrelated complex of variables.
[Footnote: Baumard, Nicholas, et al. (2015). ‘Increased affluence explains the emergence of ascetic wisdoms and moralizing religions’. Current Biology 25.]
The authors of this last study argued that increasing affluence spurs people on to develop more prosocial attitudes in order to protect the ownership of property. They suggested that the resulting emphasis on societal cooperation and the inhibition of selfish tendencies (for example, to prohibit stealing from others) precipitates a switch from a largely lawless society to one in which the rule of law applies and ideologies develop that encourage citizens to behave prosocially. These, they suggest, are reflected in the large-scale doctrinal religions that characterize the Axial Age, with their emphasis on good behaviour and individual responsibility for one’s actions.
One problem with this claim is that it looks suspiciously like group selection: why would the downtrodden masses voluntarily acquiesce in protecting the wealth of the elite? Evolutionary biologists, as we saw in [an earlier chapter], are very nervous of any explanation that, intentionally or inadvertently, invokes group selection.
Moreover, agricultural production alone cannot be an evolutionarily viable explanation. Evolution doesn’t occur simply in order to allow more food to be produced. Production is either a proxy for something else, or a solution to a constraint that allows more sophisticated ideologies to emerge that maximize individual fitness.
Looking back over these three studies, two main conclusions suggest themselves. First, Moralizing High Gods were a very late development. Most of these transitions occurred in the first millennium BC, the period known as the Axial Age. All are associated with a dramatic increase in socio-political complexity, and this is invariably associated with a dramatic increase in the size of the population to a million or more individuals, with correspondingly greater difficulty in managing the stresses of living together. These stresses, as we saw in the previous section, include increased levels of violence, as well as theft, abuse and argumentativeness.
Second, preceding this, there was often a long phase of ritual complexity and propitiatory worship based around belief in divine retribution that gave rise to a constant need to mollify the deities involved. This seems to have been associated with societies that number around 100,000 individuals. The earliest of these (in Anatolia and the Levant) date to around 6,000 BC, but most predate the appearance of High Gods by around 2,000 years (that is, they appear around 2,500 BC).
Although little is made of this, in almost every case the appearance of these forms of ritual coincide with a distinct upswing in social complexity soon after, as a result of an escalating increase in population size and its concentration in a handful of large centres (towns and cities). The increase in population size seems to have something to do with the climate and conditions for growing agricultural surpluses in these regions, while the rapid growth of cities is almost always fuelled by economic opportunities at the seat of wealth and power sucking in people from the rural provinces – an effect that continues right into the present day.
Added religious complexity seems to be necessary to allow community size to increase to these levels. What we seem to be seeing here is evidence for a series of glass ceilings. Finding a solution at each step allows community size to increase to the next glass ceiling rather than imploding on itself through lawlessness and civil war; failure to find a solution means either that community size remains stuck at the lower level or that community fragmentation results in regression to a lower stable point.
One final relevant observation concerns the circumstances under which a priestly caste (religious specialists) appears in a society. In an ongoing project, Joseph Watts has used comparative methods to analyse ethnographic data on the presence/absence of religious specialists in different contemporary hunter-gatherer societies drawn from all around the world. He was mainly interested in whether religious specialists appeared when the environment was especially unpredictable – a context in which diviners and shamans might have played a role in predicting and controlling environmental catastrophes. His analyses suggest, however, that environmental unpredictability is not itself directly related to the presence of religious specialists.
Instead, it seems that the factor that most directly influences their appearance is the presence of food stores. Food storage is indicative of food surpluses, implying that religious specialists appear when a society is sufficiently well-off to be able to set aside food for some individuals who can devote their time to being religious specialists. This suggests that a priestly caste – as opposed to individual shamans or healers – emerges only when agricultural production is high enough to give rise to populations of sufficient size to produce food surpluses. Of course, when populations reach this size, they are also likely to experience serious social and demographic stresses.
[You have now reached the end of this Social Science Files exhibit.]
[MGH: This rather long chapter arrives at this useful conclusion and then continues with a further fascinating argument, which I will prepare for a Part 2 ..]
The Source of today’s exhibit has been:
Robin Dunbar, How Religion Evolved, And Why it Endures, Oxford University Press 2022
Social Science Files displays multidisciplinary writings on a great variety of topics relating to evolutions of social order from the earliest humans to the present day and future machine age.
‘The Heller Files’, quality tools for Social Science.