Can models of evolutionary transition clarify the debates over the Neolithic Revolution?

Abstract

The ‘Neolithic Revolution,’ sometimes referred to as the emergence of agriculture at its earliest in the southern Levant, is the most significant shift in human history, shaping the world we live in today. Yet, after 100 years of study, its major cause, tempo (gradual or revolutionary), and impact of human intentionality remain disputed. Here, we examine the research potential of an evolutionary transition in individuality (ETI) to clarify this dramatic shift. Applying an ETI research perspective reveals how different causes and conditions lead to the same result, enabling a holistic view rather than a reduction of ‘Neolithic' to ‘agriculture,' or to one major climatic condition, inheritance system or standard evolutionary model, thus allowing us to clarify and bypass some of these heated, unresolved disputes. Additionally, unlike current archaeological emphasis on ‘where,' ‘when,' ‘why' and ‘how' questions, the ETI perspective offers a productive path for resolving a fundamental preliminary anomaly: why and how could the Neolithic lifeway evolve at all, given the selfish interest of individuals in a hunter–gatherer group? We do not intend to solve the shift to Neolithic lifeways, only to offer a fresh lens for examining it, emphasizing the relevance of tracking within and between group differences.

This article is part of the theme issue ‘Human socio-cultural evolution in light of evolutionary transitions’.

1. Introduction

The emergence of the Neolithic lifeway, often referred to as the emergence of agriculture, happened worldwide in different regions at different times. Its earliest evidence is observed from the Levant in West Asia, where the onset is dated to about 11 650 calibrated years before present (cal BP; [1]). For over 100 years, archaeologists, ethnographers and physical anthropologists, as well as palaeozoologists, archaeobotanists and palaeoclimate scientists have been studying the ‘Neolithic Revolution.' Excavations yielding ample finds have led to a variety of interpretive theories, ranging from Marxism to climate determinism to genetic determinism [2,3], and to conflicting biological and cultural evolutionary models [4]. In this paper, we discuss the emergence of the Neolithic in the southern Levant as a possible case of evolutionary transition in individuality (ETI). Our goal here is not to solve the riddle of the Neolithic but to suggest a fresh and hopefully productive research perspective. An ETI occurs when individual entities that were capable of independent selection and response to selection can now only undergo these processes as part of a more complex individual entity. For example, from independent prokaryotic cells to organelles within eukaryotic cells, from free-swimming eukaryotic cells to organ-specific cells within multicellular organisms, or from independent organisms to members of a distinct cultural group. Specifically, we will ask about a particular dramatic change: from nomadic aggregates of hunter–gatherers to sedentary, interdependent and large agricultural communities. We will not examine to what degree and in what respects the Neolithic Revolution fits the ETI criteria ([5]; and see papers in this issue). Rather, we examine here an essential preliminary question: whether and how implementing an ETI perspective contributes new and relevant insights for explaining this dramatic change in human history, thus hopefully enriching both the archaeological and the ETI literatures.

The shift to the Neolithic is typically referred to as ‘the emergence of agriculture,' emphasizing the importance of agriculture, practically reducing the term 'Neolithic lifeways' to the term 'agricultural lifeways'. Further reduction is seen from ‘agriculture' to ‘domestication' (yet see [6]). We argue that applying an ETI perspective avoids such reductive overtones and can assist scientific advancement. Moreover, it does not methodologically foreground reduction to a single cause (climate, genetics, economy etc.) nor a single essential characteristic (domestication) and, in that sense, allows one to follow non-exclusive explanatory pathways. This short opinion paper does not profess to comprehensively review or resolve the debate over the Neolithic Revolution but rather to suggest a new evolutionary perspective for its study.

(a) . The evolutionary transition perspective

ETIs hold an important role in evolutionary literature [7]. Szathmáry [8] models these major events in the history of life via two dimensions: the evolution of new levels of individuality (e.g. cells, multicellular organisms, linguistic communities) and the evolution of new informational systems (e.g. genetic, epigenetic, linguistic inheritance). Szathmáry divides each major transition into three phases: origin, maintenance, and transformation to a higher-level unit, and assigns a crucial role to group selection in the third phase of transformation. West et al. [9] add an important emphasis, that whenever moving from the second to the third phase, ‘the same problem arises at each transition: how are the potentially selfish interests of individuals overcome to form mutually dependent cooperative groups?' [9, p. 10 112]. This very question arises in the Levant during the shift to the Early Neolithic: simpler lifeways of egalitarian hunter–gatherer bands are transformed into complex and hierarchical agricultural communities. During such a process, each rational individual in the group is expected to prefer to hunt and gather while their neighbour labours longer hours in the fields and waits anxiously for the autumn rains [10]. Therefore, one is bound to ask: why and how could the Neolithic lifeway evolve at all against the interest of each individual? We try to untangle this seeming anomaly by using the ETI research perspective.

The ETI's three phases are described here based on Griesemer's [11] process-oriented view of development (the 3Cs), Michod's [12] result-oriented view of fitness (Darwinian dynamics) and Black et al.'s [13] patchy-environment experiment (ecological scaffolding). Each of these models could independently explain ETIs, yet their combined usage increases their explanatory power. However, before application, the practical meaning of ‘individual' and ‘group' in this context needs to be clarified. Since any new level of individuality is by definition a reproductive group of former individuals [13], then ‘individual' and ‘group' are used here in different senses according to context. During an ETI, the lower-level ‘individual' can denote a token entity (e.g. a particular human in a particular group), or a type representative of a sub-group (e.g. a ‘gatherer' who ideally represents all gatherers in their group). Each such individual is composed of former independent individuals (on an evolutionary timescale), and such a nested hierarchy also applies to groups. A group is composed of groups and is also part of a group (a group of hunter–gatherers at a site is composed of a group of gatherers and a group of hunters, yet also part of a larger tribe). Here, the relevant context of ‘individual' is synonymous with ‘group member', so an individual is typically an idealized type that directs research attention to within-group differences, whereas a ‘group' here focuses attention on tracking between-group differences.

Using a within-group–between-group lens, Griesemer [11] tracks an ETI process by its three developmental stages of interaction: from a coordinated collection of individuals in the first stage, to a cooperative connected collection in the second, to a collaborative community in the third. Hence the model's name—3Cs. Specifically, and most important for the Neolithic Revolution, the 3Cs research perspective allows one to describe and explain the origins of such a radical transformation in terms of incremental change. At the coordination stage, individuals suffice with organizing their activities so as to avoid collisions. The resulting entity is a collection of independent individuals. ‘Coordination' means the ordination—putting in order in space and time—of the work of group members in a place and time they co-occupy. This ordering may be self-organized, or it could be imposed by ecological constraints or forced by selection from outside the group. Such coordination is expected to benefit mostly those coordinating individuals and be transmitted only by them, so any group success or inheritance is a mere epiphenomenon. Hence, at the first C stage, the existing group is mostly an aggregative sum of its individual members. Cooperation appears even at this first stage (e.g. large game hunting in nomadic bands) yet as an occasional rather than constitutive interaction.

At the second C stage, interactions of within-group cooperation increase. Individuals not only coordinate but also cooperate and thus form a socially connected collection. ‘Cooperation' means operating together. The literature uses this concept as practically synonymous with a shared ‘partnership' at the one end and with ‘coordination' at the other. To avoid conflation, we use the concept of cooperation in its narrower, individually based meaning. It need not be for a shared purpose, or even for mutual individual benefit, though such reciprocity may be a by-product of self-benefited cooperation and conflict mediation. Such increased cooperative interactions also increase within- and between-group trait differences, in turn facilitating within- and between-group selection. In such a connected collection of cooperating individuals, ‘group fitness' is no longer a mere aggregate of ‘individual fitnesses' ([7]; for detail on group versus individual fitness, see [12,14]). The group's expected success in reproducing its group characteristics (i.e. group fitness) now causally depends not only on the number and average reproductive success of its individual members (i.e. individual fitness), but also on its nexus of organizing its intra-group variance. For example, during competition over some limited resource, an initially larger group could shrink and eventually dissolve (i.e. group fitness equals zero even if no individual died), whereas a more cooperative, yet smaller group, may increase in numbers and eventually split into similar cooperative daughter-groups (i.e. group fitness equals 1 even if the average fitness of its individual members is different). Note that individual and group fitness are still strongly and positively correlated at this stage. This occurs because existing group traits keep changing relatively rapidly via individual mechanisms such as biological or cultural migration, which is accepted at the group level; and existing group traits are inherited only via individually based mechanisms: biologically by individual reproduction and culturally by individual communication, imitation, learning or teaching. At this stage, a group is no longer an epiphenomenal aggregate but its cooperative characterization still largely depends on the individual level both for its selection and for its inherited response to selection.

At the third, collaborative, stage, a community emerges from its network of individual connections, via collaboration between its different interacting inhabitants. ‘Collaboration' means co-labouring toward a shared goal and, in that respect, it differs from cooperation. Community collaboration is typically organized via some mechanism of division of labour, either morphological as in ants or cultural as in human societies. Distinct individual roles largely reduce intra-group conflict and are expected to remain constant even if costly for the individual, as long as inter-group exceeds intra-group selection pressures. Group selection depends on group variance, which develops from ‘within' by increased within-group interactions, and/or is ecologically scaffolded from ‘outside', by a constraining patchy environment. Either way, at this collaborative stage, intra-group and inter-group differences are not only selected at the group level but also inherited at the group level, via additional information systems—e.g. cultural and ecological inheritance—over and above genetic inheritance. That is, individual-based mechanisms of inheritance mentioned previously still exist, but group-level mechanisms are added and become more dominant, such as specific pheromones to mark each ant colony or a canonized artefact to mark each distinct human group. In that sense, our framework follows those extending the standard evolutionary Neo-synthesis [6,15,16]. The canonical group mark itself need not be practically beneficial [8]. It could be a non-practical artefact, as is often the case with symbolic markers, yet its group-rallying function provides an adaption at the group level, and is culturally inherited by its daughter groups. Moreover, and most important at this third and final collaborative stage, group fitness is no longer correlated with individual fitness, but decoupled from it [12]. In the next section, we review the archaeological evidence and debates regarding the Neolithic Revolution in the Levant, followed by an attempt to implement the 3Cs ETI perspective, and examine its relevance to this dramatic shift in human prehistory.

(b) . What happened in the Levant? Data and major debates

The Levant is the eastern portion of the Mediterranean coast, including modern south Turkey, Syria, Lebanon, Israel, Palestine, Jordan and parts of the Sinai Peninsula (Egypt). The late prehistory of the Levant, known as the Epipalaeolithic (25 000–10 000 cal BP), is when the earliest shift on record to the Neolithic lifeways happened (table 1). It began in the relatively harsh climatic conditions of the Last Glacial Maximum, continuing into the still existing interglacial Holocene. During these ca 10 000 years, changing environmental conditions generated climate fluctuations, the nature, amplitude and impact of which are in debate [18]. What researchers believe to have been rather severe climatic changes are the background to the dramatic shift to the Neolithic way of life, but so are the changes in social organization, mobility and economic mechanisms, which are all similarly in dispute [19].

Table 1.

Chronological framework for the Mediterranean zone of the southern Levant (modified after [1,17]).

period (major cultures)	environment	calibrated years before present (cal BP)	duration (years)
Early Epipalaeolithic (Masraqan, Kebaran)	Last Glacial Maximum – relatively harsh conditions	ca 25 000–19 000	ca 6000
Middle Epipalaeolithic (Geometric Kebaran)	post-glacial warming	ca 18 500–15 000	ca 3500
Late Epipalaeolithic (Natufian)	relatively favourable conditions	ca 15 000–11 650	ca 3500
PPNA—pre-pottery Neolithic A (Khiamian, Sutanian)	end of younger Dryas and Early Holocene	ca 11 650–10 600	ca 1050
PPNB—pre-pottery Neolithic B	Holocene	ca 10 600–8350	ca 2250
Late Neolithic 1 (Yarmukian)	Holocene	ca 8500–6400	ca 2100

At the beginning of the Epipalaeolithic, humans were living in small groups of nomadic hunter–gatherers, using tools and survival strategies similar to the ones they had used during the past ca 2.5 Myr. However, evidence from a short time thereafter, from the Early and Middle Epipalaeolithic, indicates that these groups were already collecting massive amounts of seeds and fruits, applying sophisticated fishing and hunting methods, and living in sedentary village-like settlements, only in huts made from vegetation [17] rather than the Late Epipalaeolithic stone buildings. Some scholars have even suggested that the botanical remains excavated from the site of Ohalo II on the shore of Lake Kinneret (23 000 cal BP) is evidence for full-scale agriculture [20]. Given the large distances covered by these nomadic groups [21,22] and the ample evidence for their inter-group communication, agriculture—full or partial—was an existing option—unadopted—long before the Neolithic.

Over the course of the Epipalaeolithic, archaeological evidence indicates reduced mobility, changes in territorial range and increased group size, accompanied by dramatic increase in cultural diversity, both within and between settlements [1]. Stone tools became smaller (microliths) and more typologically well-defined [23]. The best examples for this process are the emerging distinctive lithic traditions (or material culture traditions) defined in the arid zones of the Levant versus the Mediterranean areas [24–26]. This process reached its peak with the emergence of the final culture of the Epipalaeolithic—the Natufian [27]. The Natufian culture marks a time of technological advancement. Natufians produced sophisticated and highly diverse stone and bone tools, developed food storage techniques and domesticated the dog [28]. They had long-distance trade relations, evidenced by obsidian tools from Turkey found in the Hula Valley site of Eynan [29]. Their artistic representation increased dramatically—ornaments, carved stones and bones, sculptures and more [27]. Many of these objects were found in burial context. Hundreds of burials were excavated from Natufian sites, including individual and group burials of all ages and genders. The deceased were buried in various positions, some decorated with jewellery or artistic objects, and, in two cases, humans were buried with a dog [28,30]. At Hilazon Cave in the Galilee, an old, crippled woman was buried with the wing of an eagle, the tail of a bull, the bones of leopard and wild boar, a broken basalt bowl, and over 80 tortoise shells, evidence for symbolic and religious behaviour, an emerging ‘shaman' [31,32]. The Natufian lived in large houses of stone arranged together. Archaeologists debate whether ‘village' is the relevant term given organizational implications. Significantly, their heavy stone implements, immovable devices and the appearance of commensal animals such as house mice [33] are evidence that sedentary lifestyle [34–36] pre-dated agriculture (cf. [37]). However, the Natufian economy was still based on hunting and gathering and not on domesticated farming [27].

The shift to an agricultural way of life is documented to the end of the next period, the pre-pottery Neolithic A (PPNA). The locus of change moved to the northern Levant where the first ‘mega villages,' almost ‘towns' emerged [38]. In the southern Levant, the widely spread Natufian culture was replaced by the distinct Khiamian and Sutanian cultures (table 1). These two entities differ from each other primarily in their stone tool traditions but the sites of both cultures are spatially large, evidently home to hundreds of people. Large public buildings such as the famous tower of Jericho first appeared in the PPNA [39], suggesting complex social organization [40]. Almost all PPNA sites display symbolic cult and religious activity in the form of temples and burial customs [41]. The cultural differences typifying the PPNA dissolved during the second stage of the pre-pottery Neolithic (PPNB), when all across the Levant quadrate house architecture replaced the earlier circular structures and extensive evidence is found for plant domestication and animal husbandry [42]. By the end of the PPNA and during the PPNB, the people of the Levant had fully completed the transition to Neolithic lifeways. They lived in very large, sedentary, culturally complex and agricultural communities. Their lifeway was based on a non-egalitarian—and prima facia hierarchically stratified—economy and society, represented by distinctly unequal house sizes, burial goods and public/private architecture. Within each site, distinct social roles existed (e.g. farmer, shaman, artisan) within social institutions that were physically grounded (‘religion' is practised in temples, ‘politics' in large public buildings and ‘agriculture' in large fields outside the ‘village' area), and unique artefacts provided a visible ‘signature’ for each individual community [43].

How do researchers explain such a dramatic transition? Disagreements mostly engage the broad spectrum of ‘when,' ‘where,' ‘how' and ‘why' questions. Falling in the range of C¹⁴ chronology, researchers are answering the ‘when’ questions more precisely as radiometric chronology advances. However, they are still deliberating between an abrupt, in prehistoric timescale, tempo of change and a gradual process over thousands of years, placing the actual term ‘revolution' in question [42]. The timing of the emergence of agriculture depends, in part, on the definition of both agriculture and Neolithization and the consequent interpretation of the archaeological findings. Are the grains of wheat found in Jericho domesticated [44]? Is the young animal buried at Eynan a domesticated dog or a wolf [30]? Therefore, even if the dating of both Jericho and Eynan is established, when and where an agricultural revolution in the Levant occurred can still be rationally debated. The search for genetic data on the evolutionary lineage of wild animals and plants is a current research focus (e.g. [32,33]). However, genetic information alone, although necessary, is far from sufficient to explain the primary riddle of an evolutionary transition: how could a higher-level unit evolve if it goes against the interest of its individual members [8]? A shift from an aggregate of small individual bands of nomadic gatherers to complex Neolithic communities causally depends on much more than a single trait found in wheat grains, inherited by a single genetic sequence, and selected by a single dominant environmental setting (for details see discussion on the explanatory limits of the evolutionary Neo-synthesis and its extended synthesis alternative in [15]). Therefore, the recent archaeological attention to evolutionary models is a step in the right direction, yet its focus on explaining many different, and highly complex, cultural phenomena by tracking the genetic data of a single trait is somewhat naive according to the extended evolutionary synthesis.

Even when answering the more complicated questions about the Neolithic, most scholars still tend to seek one explanatory model that denotes one major casual factor for answering two complicated questions: first, Why and how did this lifeway originate?; and second, How and why did it survive, defeating alternative strategies despite its detriment to the individual farmer? Primary factors heatedly argued as the major cause of the Neolithic shift include climate changes and their resulting environmental stress on the community, sometimes called ‘climate determinism' and intensive population growth due to improving subsistence measures [45]. Agricultural technological development is frequently cited in the process for becoming farmers. Agriculture manifested a technological bottleneck of challenges to full agricultural subsistence, including harvesting, ploughing, and storage and processing of crops [46]. Some argue that the emergence of private property drove establishment of social institutions, which brought about agriculture. The resulting socio-economic inequality created a stratified community that, in turn, maintained cultivation lifeways [47]. And, importantly, not a few scholars attribute the dramatic lifeway change to a conceptual cognitive shift via concepts like ‘religion' [41] or the delineation of ‘private versus public' [48]. A change in the symbolic way people perceived their world could justify the emergence of the ‘domestic sphere' which became dominant in the Neolithic culture [49].

Related to cognition is the debate over the intentionality of the Neolithic Revolution. At one pole, some claim that human thought and intentions had almost nothing to do with the evolution of domesticated species. They argue for a non-intentional process in which human activity such as intensive gathering of wild plants contributed to the selection of specific mutations among these species, which, in turn, led to their domestication. A frequently used example is the mutation that prevents the seeds of wheat from naturally falling from the spike. This is a fitness disadvantage for the wheat in natural habitats, yet a highly advantageous mutation for the human gatherer, who loses fewer grains. Therefore, such mutation, if present in a given sowed field, would take over very fast even without goal-oriented selection by the farmer (cf. [6]). Others argue that human understanding is essential, but claim that ‘innovating agriculture' was a series of random, almost accidental discoveries that led to plant domestication rather than a planned process (e.g. the ‘dump heap' scenario). On the opposite pole are scholars who argue for a fully intentional process in which knowledgeable people, armed with thousands of years of experience, deliberately conducted trial-and-error experiments with different species in search of domesticable plants [42].

In these important debates, it should be noted that two points are frequently pre-assumed. First, the Neolithic lifeway is synonymous with the agricultural revolution, which is synonymous with plant and animal domestication (yet see [6,15]), hence the conclusion that domestication, in effect, explains the Neolithic Revolution. Second, explaining is best done by identifying a—single—major cause within a universal and simple evolutionary model. We believe that these two preferences are non-accidentally connected, and that both can be avoided by employing an ETI research perspective. In the next section, we demonstrate how the ETI perspective destabilizes both pre-assumptions and offers a novel mechanism for studying the Neolithic riddle that avoids most of the longstanding debates.

(c) . Applying an evolutionary transition in individuality perspective to the Neolithic Revolution in the southern Levant

To examine whether an ETI perspective can explain the emergence of the Neolithic lifeway, we will address the debates over its primary causal factor: was it climate change, population growth, technological innovation, the economy or a cognitive shift? Was domestication intentional or accidental? Gradual or (relatively) instantaneous? And, finally, since agriculture is clearly costly to the individual compared with hunting and gathering, why and how did it survive as a subsistence strategy and spread so successfully?

To be sure, agriculture and/or domestication per se is not a sufficient or even unique characteristic of the early Levantine Neolithic. As noted, plant domestication may have existed much earlier, in the Early Epipalaeolithic, some 23 000 years ago, practised by semi-nomadic groups [20]. Later, during the Late Epipalaeolithic sedentary lifeways of the Natufian (15 000–11 650 cal BP), dogs were buried beside humans. The lack of direct botanical evidence for agriculture in Natufian sites can be attributed to the absence of well-preserved assemblages. Alternatively, or in conjunction, it can be attributed to a marginalized place in subsistence notwithstanding agricultural knowledge transfer. That is, given the wide-ranging communication throughout the Epipalaeolithic, it is safe to assume many sites possessed knowledge of agriculture-based economy and could obtain its technology and practices. Why did the Natufian not shift to full-scale agricultural subsistence? From the individual's perspective, the answer is clear: Farming is more laborious and riskier than the hunter–gatherer alternative, and it becomes a winning strategy (e.g. supporting more children, personal wealth and territory) only after it eradicates its less-costly alternatives. Therefore, the ETI perspective challenges us to explain: Why was it adopted at all?

Using the ETI model as a heuristic guide, we follow processes not often discussed by archaeologists, tracking selection and inheritance at the individual and group level for interpreting the Natufian and PPNA empirical evidence. For an ETI to occur, the 3C developmental stages—coordination, cooperation and collaboration—and an ecological scaffolding all need to be examined. At the final collaborative stage, group selection is expected to override individual selection and pass down the culturally selected group characteristics via cultural group inheritance mechanisms. We demonstrate that all the above are indeed found in the archaeological data, and that using an ETI research perspective can help clarify the Neolithic literature.

The first stage, coordination, best describes the Early and Middle Epipalaeolithic when hunter–gatherer group boundaries are ephemeral, composed of small bands of 20–25 people [14]. At this stage, group-life leads to a situation in which most differences in fitness-related traits, e.g. level of physical strength or artisanship, are between individuals rather than groups, and group traits are no more than the average of individual traits. The result is a clear dominance of individual selection over group selection. Most interactions are individual coordination at the physical level, such as at times avoiding the hunting tracks of another band and at times meeting it for reproduction within a larger gene pool.

During the Late Epipalaeolithic, sedentism emerges. Year-round settlements strengthen group boundaries and distinctions; thus, material cultural diversity increases at both the individual and group levels (see examples below) and facilitates fitness differences at both levels. Fitness differences are also plausibly affected by ecological scaffolding, generated by changing world climate, with the collapse of the last ice age regime during this time. At the group level, settlements now differ by population size, function of settlement (habitation, symbolic centre), mobility expressed in sedentary or semi-sedentary lifeways, and aspects of material culture [1, p. 40]. Fitness-related diversity between groups generates a certain degree of group selection. At the individual level, i.e. within specific Natufian groups/settlement, the emerging diversity of new social roles and institutions, burial ceremonies, and unequal property produces a flare-up of individual selection pressures. To address such an increase in social challenges, coordination interactions do not suffice. According to the ETI model, the second C stage now emerges, at which cooperative interactions replace coordination as the dominant mode of interaction, both within and between groups. Such individually based cooperation also increases the effectiveness of individual communication, thus increasing the effectiveness of millennia-old cultural inheritance via individually based mechanisms of imitation, learning and teaching. Such inheritance mechanisms transmit the unique cultural traits from one individual to another and, via individual commerce and migration, also facilitate—as a by-product—the spread of new technology to new groups. As a result, intra- and inter-group diversity increases.

Indeed, the archaeological evidence reveals an explosion of technological, artistic and symbolic diversity in the Natufian, mostly within but also between Natufian groups: 'An obvious example of this is the use of the microburin technique (mbt) in some Natufian sites, while absent from others. Indeed, we believe that the mbt – used systematically [in] the [groups of] Nebekian, Nizzanan and Mushabian (and including in the Negev/Sinai and eastern Transjordan) to shape microliths, but unknown in the Mediterranean [groups of] Kebaran and Geometric Kebaran, was now introduced to some Mediterranean Natufian communities' ([1], p. 40).

This fits nicely what the ETI model predicts at the second phase. A new technology unevenly spread will help some groups grow faster than others, and within each group, some toolmakers will succeed more than others; yet, at least at this stage, individual and group fitness are expected to be correlated rather than decoupled. In addition, since at this second stage, the group's trait/culture is still rather fluid and not based on some fixed group ‘canon', then most individuals will tend to freely change what was transmitted to them and teach their own style to others. As a result, no rigid ‘group signature' exists. The bone fish-hooks excavated from the site of Jordan River Dureijat on the banks of the Upper Jordan River are a good example. Each of these highly sophisticated tools is different from all others in size, style and features (e.g. barbs). Excellent technical knowledge is shared by all group members, but each executes this knowledge in a specific way [50]. The ETI model expects this high diversity, also at the group level, yet predicts relatively low fidelity in transmitting group culture, i.e. low group inheritance. Since evolution requires both selection and response to selection (i.e. inheritance), then cultural evolution at the group level cannot occur at this cooperative stage.

A necessary condition for an ETI cultural transition into the third, collaborative, stage, is clear cultural distinctions between groups that are also inherited through group inheritance mechanisms. This condition can arise via several non-exclusive causal pathways: via patchy ecological conditions that scaffold different cultures at different localities; via inter-group selection overriding individual selection to the point that group and individual fitness are decoupled, or via in-group social institutions that maintain division of labour, canonic group signature/identity, and a cultural bias against other groups. Given the above, learning and teaching how to replicate one's cultural group ‘signature'. and meticulously replicating that group identity from generation to generation, provide cultural inheritance at the group level. At this point, a complex and stable community has evolved.

Indeed, when reviewing the archaeological data from the Natufian to the PPNA: 'It seems that there was a need to deal both with the ‘external' changes in groups' subsistence resources (necessitated by changes in territory sizes and population densities), as well as with the ‘inner' social alterations … [as] the changes affected some groups more than others' [1, p. 40].

In fact, the ETI model seems to unknowingly function as a simplifying ‘inference to the best explanation’ tool for organizing the extensive data from the Neolithic Revolution. For example, see Belfer-Cohen & Goring-Morris [1, p. 41]: ‘Thus one can observe for the Natufian the following:

• 1. As communities became increasingly sedentary, there were opportunities to accumulate more ‘wealth' (i.e. material belongings).
• 2. As communities increased in size, there was a necessity to develop novel social mechanisms to regulate the increasingly complex nature of interpersonal, intra-community and inter-community interactions.
• 3. The growing intra- and inter-group densities dictated more marked distinctions among the various communities comprising [sic] the Natufian entity as a whole. Most immediately, many of these phenomena were expressed in the archaeological record, including investments in the decorative and symbolic aspects of both non-utilitarian and mundane artefacts. The latter was [sic] expressed, among other ways, by the distinct stylistic ‘signatures' of specific communities. The processes enumerated above continued into the PPNA.'

A specific example is found in the fact that the Natufians occupied various sites without clearly distinct material cultural identities, then, as expected by the ETI model, during the Levantine PPNA such group distinctions emerged between the Khiamian and Sultanian cultures. These material cultures emerged (at least in the early stage of the PPNA) next to sedentary Natufian ‘villages' [51], demonstrating ‘solid evidence for local, in situ continuity between the two cultural manifestations [Late Natufian and Early Neolithic]' [1, p. 37]. When competing over physical or cultural resources, the larger number of inhabitants in a Neolithic site became an asset, as did social organization. As field labourers, children are an asset; a farmer typically sustains more children than a hunter–gatherer and gains more wealth. This advantage could have been a motive to become a farmer and for groups of farmers to increase. In this sense, the trait of group size is highly correlated with average individual reproduction, yet deceiving at this stage. A farmer's wealth and status are still lower than a shaman's or a warrior's even if the farmer sustains more children. More importantly, a community without a shaman is culturally—and as a result physically—doomed (unless ‘the shaman' is not a specific person but has already became embedded in routine group maintenance and/or dispersal mechanisms). Therefore, during the PPNA, the trait of group structure seems to potentially affect group size more than average individual reproduction and, in that sense, group fitness is decoupled from individual fitness—at least for some crucial traits—during the Early Neolithic.

As the PPNA ends, only one Neolithic culture remains, and spreads across the Levant [43]. The selective benefit for the group is clear, but why would the individual farmer not revert to gathering when facing a bad wet season? First, if under the Neolithic social structure being a ‘farmer’ means you follow the ‘shaman's' or ‘ruler's' orders, personal preference gives way to enforced group pressure or personal internalization of groupthink or group-identity. Second, if one's ‘profession'/'caste' and/or economic status ‘rich'/'poor' has a smaller effect on one's reproductive success than the result of competition with a neighbouring group, group selection overrides individual selection. Therefore, what seems like a farmer's non-rational altruism becomes rational when group fitness is decoupled from individual fitness. At this point, the farmer's own fitness depends primarily on group membership; hence, the cost of altruism that still exists compared with his non-farming neighbour is actually lower than the selfish alternative of reverting to hunting and gathering.

An ETI also requires an inherited response to such a combined intra- and inter-selection process. Szathmáry expects an in-group bias for imitating one's group traits, typically non-functional ones [8]. Similarly, as noted above, Belfer-Cohen & Goring-Morris describe the inheritance of artefacts of group signature [1]. Eventually, the combination of cultural group inheritance, inter-group selection, and changing ecological conditions resulted in a single canonic and highly complex Neolithic culture across the Levant [43].

Throughout the PPNB stage, the coordinated, cooperative and collaborative inner-group interactions became evermore deeply and systematically inter-connected and mutually dependent. That is, group characteristics such as economic and social stratification, religion, and larger group size and territory that theoretically could exist separately, and during the Natufian perhaps did exist separately, can no longer do so in practice. Moreover, once farming exists as the individual's lifeway, for whatever reason (e.g. favourable ecological conditions, short-term economic benefits, religious duty, enforcement by the group leader(s), individual curiosity) it becomes entrenched in one's personal and community lifeway. At the individual level, one's social status, property, religion, nutrition, family size, and territory are constituted by one's farming lifeways. At the group level, organized division of labour and advances in competing with other groups will dissolve if that function breaks down. Therefore, returning to hunting and gathering threatens one's self as well as one's community as a whole—culturally and biologically, owing to displacement into less favourable areas—and thus is too costly for the individual tempted not to farm. In short, the ‘point of no return' has passed.

Hence, neither a grand technological leap nor extreme irrational altruism, nor fierce group competition, nor climatic disaster was necessary to entrench farming (although all could occur). A gradual process—developmental, selective and ecological—alongside cultural group inheritance could lead to the same result. Moreover, as increased human interactions also increased group selection during this third, collaborative stage, even mild and short climatic conditions (e.g. a 2-year drought, leaving no evidential trace) or small intra-group developmental pressures (e.g. charismatic shaman or ambitious ruler coalescing group identity) may have driven such changes.

2. Conclusion

We have established the potential relevance of the ETI model for explaining the highly complex Neolithic Revolution. The ETI perspective first allows one to bypass a series of empirically un-answerable major debates, such as locating the major cause of the shift to farming and deciding whether it was instantaneous or gradual, conscious or coincidence. Different answers to such ‘why,' ‘how,' ‘what' and ‘when' questions are not contradicting options that require resolution. On the contrary, the model explicates the embedded complexity in this major prehistoric change; hence, it clarifies that different answers can coexist in various contexts. It also explains why and how such a transformative process could (and did) occur in other regions at different times in human history, propelled by different causative factors that need not have been dramatic.

Second, an ETI perspective demonstrates that the ‘Neolithic culture' should probably not be perceived as synonymous with ‘agriculture' nor the latter synonymous with ‘domestication.' It explicates and foregrounds why the Neolithic lifeways so deeply depended on collaborative interactions, in-group social structure, inter-group selection and cultural group inheritance, all as crucial to Neolithization as agriculture. In addition, agriculture itself is not a simple concept. It occurred much later than 11 650 years ago in regions like South America, does not necessitate a sedentary lifeway (as many nomadic cultures demonstrate today), nor plant or animal domestication (as Kenyan rubber tree plantations demonstrate). Not all these traits and factors are required under an ETI model, though they exist in the process of Neolithization. As a historic phenomenon, the advantages of agriculture for the farmer become relevant only after it has already won. Thus, its evolution is problematic to solve, yet many of the solutions in the literature are so immersed in collecting large quantities of empirical detail—the majority of relatively low empirical resolution—that they overlook this challenge and may provide circular explanations. The ETI perspective offers a general and relatively simple and useful heuristic that directs us toward individual- and group-level dynamics. Observing the forest to see its many different trees. To conclude, an evolutionary ETI perspective may offer archaeologists a non-conventional and useful perspective for looking at their concepts, data and primary debates, and that is all we seek.

Data accessibility

This article has no additional data.

Authors' contributions

A.S.: conceptualization, methodology, writing—original draft, writing—review and editing; G.S.: conceptualization, methodology, writing—original draft, writing—review and editing.

Both authors gave final approval for publication and agreed to be held accountable for the work performed herein.

Conflict of interest declaration

We declare we have no competing interests.

Funding

We received no funding for this study.