Chapter 4: Transition 1: Modelling the migrations and peopling of new geographic areas by Homo sapiens
Entrées d’index
Keywords : emergence, migration, modelling, multi-agent system, Out of Africa theory, simulation
Texte intégral
1The first transition studied in the framework of the programme TransMonDyn is the most ancient: it concerns the diffusion of our species, Homo sapiens, on the surface of the planet. This expansion is marked in particular by the departure of our ancestors from Africa roughly 100,000 years ago, then again, some 30,000 years later. These migrations led our species into Asia and Europe, but also to the conquest of lands previously unoccupied by human beings: Australia and the Americas.
2We present this transition here, as well as an attempt at modelling, in order better to identify the different factors which have conditioned its spatial and temporal characteristics.
3We first introduce the context of the transition, together with its different dimensions and possible causes. These elements underlie the features of the model we present later. The model involves simulation and allows testing of the hypotheses concerning the way in which our ancestors have peopled new regions by migration. We describe the different components of the model and, finally, present the results of the simulations and several interpretations.
Homo sapiens conquering the Earth
The emergence of modern Man in the context of human evolution
4About 150,000 years ago appeared the anatomically modern man, Homo sapiens, the last step in a long evolutionary process. This process started seven to eight million years ago within the subtribe1 Hominina, which has contained all human, as well as all pre-human or para-human, species, since its split with the subtribe Panina, which today contains chimpanzees. In the course of this long period, numerous species succeeded each other and sometimes co-existed. Among them, Homo habilis was born about 2.6 million years ago. His name and his status as the first representative of the human genus come from the first stone tools with which he is classically associated, even if that relation is questioned today.2 Following this major cultural innovation, nearly 800,000 years passed before a successor, Homo ergaster, left Africa to spread across Eurasia.
5The appearance of our species is thus to be situated within a bushy evolution, both in space and time. In particular, the first Homo sapiens were not the only human beings on the planet, and for a while, kept company with their ‘cousins’ both in Africa and, later, in Asia and Europe. The nature of their relations is still poorly understood, but breeding took place between Homo sapiens and two other species: the Neanderthals and Denisovans.3 4The genes, however, do not inform us precisely concerning the nature and frequency of contacts between groups of different species. Even if one is beginning to understand, for instance, the role that the transmission of certain genes might have played in Homo sapiens’ adaptation to various environments, the existence of cultural transmission is still debatable.
6Time-scales are sometimes deceptive, and it should be recalled that the appearance of Homo sapiens is much earlier than the great cultural innovations which distinguish many modern societies. In fact, agriculture and sedentarisation appeared about 10,000 years ago and writing ‘only’ 6,000 years ago. The earliest paintings are more ancient, but even if they are dated back almost 40,000 years, more than 100,000 years separate the first ones from the appearance of our species. During the greater part of its existence, therefore, it has been characterised by cultural and demographic structures very different from those we commonly observe today.
The ‘Out of Africa’ theory
7The presence of Homo sapiens on virtually all land surfaces has engendered numerous theories about the peopling process connected with our species. Especially with regard to the most ancient phases of peopling, the multiregional theory and that known as Out of Africa rivalled each other during the 1980s, before the second gained the advantage.
8The multiregional hypothesis, initially formulated in 1984,5 is opposed to the idea that different species may be distinguished within the human genus. On the contrary, it defends the existence of one sole species, which appeared before the first departure from Africa and which includes the more ancient populations of Homo erectus, Homo neandertalensis, etc., as well as the more recent ones of Homo sapiens. The passage from archaic populations to those more modern would then be produced in continuous fashion over the course of time, a point that distinguishes this theory from a polygenetic origin of several human species, that is, from their emergence in several distinct places6. Simultaneously maintaining the existence of a single species at the global level and of regional differences between its members is not straightforward. Indeed, it requires a subtle balance between homogenising transfers of genes and local phenomena of genetic drift and selection.
9In opposition to a ‘continuous’ reading of human evolution, the replacement hypothesis currently labelled Out of Africa considers that early human populations in Eurasia have been replaced by a more modern population originating in Africa7. Our species would have appeared in east Africa between 150,000 and 200,000 years ago, before spreading out across the continent, then across others, along different migratory routes8. This theory has the approval of the greater part of specialists today and receives support both from paleoanthropological9 10and genetic data.11 12Researchers in different disciplines are currently seeking to identify the migratory routes taken and the dates that can be associated with them. The study of the geographical distributions of different haplogroups,13 and especially those carried by mitochondrial and Y chromosome DNA, reveals among other things the various migrations of our distant ancestors.
10The history of our species taking shape today is constituted, therefore, of ancient migrations within Africa and a possible precocious first movement in the direction of the Near East, although without ever going beyond that region14. Thus, it is a second later migration, around 60,000 to 70,000 ago, that marks the veritable debut of the expansion of our species across Asia, then the rest of the world. Observation of a genetic bottleneck at this period suggests that the founding population was small.15
11Following this migratory wave, which led Homo sapiens to south and south-east Asia, Australia and New Guinea, the dispersion of our species in Europe, for its part, took place later, about 40,000 years ago16. The easternmost part of Asia has also been colonised since this period. The eastern side of the Bering Strait, in particular17, was reached about 30,000 years ago18 by populations that later rapidly colonised the Americas a little less than 15,000 years ago19.
12With regard to the expansion of about 70,000 years ago, a debate exists as to a passage by the Horn of Africa and the Red Sea and/or a migration along the Nile, then across the Near East, but various evidence points rather in the direction of the former route – genetic data20, but also archaeological discoveries in the Arabian Peninsula, which suggest an occupation more than 120,000 ago21. Following the first passage, small population groups would have traced the coasts of south Asia to reach Indonesia, then Australia, about 50,000 years ago. To establish with confidence the ancient migration routes is particularly difficult because of the evolutions in climate and sea-levels during the last Ice Age, which began about 120,000 years ago and culminated 21,000 years ago. The expansion of ice entailed a significant drop in sea-levels – from 50m or 70m 70,000 years ago to 120m during the maximum Ice Age. This very probably facilitated the passage of the Strait of Bab-el-Mandeb in the Red Sea (see figure 1), with the presence at that time of islands today below the surface. The colonisation of the Sahul, the body of land connecting Australia and New Guinea during the greater part of the last Ice Age, is especially significant. Many archaeological sites in Australia confirm the presence of Homo sapiens on the island about 50,000 years ago22. Palaeographic studies of the region of Wallacea, between Sunda – the name given to the landmass that had then emerged at the south-east extremity of Asia – and Sahul, permit the conclusion that a number of maritime crossings were necessary, one of them extending over at least 90km23 24. The mastery of navigation needed for these long and difficult crossings could have developed progressively along the coastal route leading from Africa to Sahul, as is suggested by the very ancient occupation of the Andaman Islands in the Indian Ocean, itself necessarily effected by sea-travel25.
A first major transition in the peopling process
13The migration of Homo sapiens from Africa is a major transition in the peopling of the Earth. It is interesting to situate it on the basis of preceding scientific elements before considering what its motivation might have been.
14Two regimes, one preceding and one following the transition, can be defined so as to bring out the common points and the differences. The first corresponds to the presence of Homo sapiens limited to Africa more than 100,000 years ago. The second was formed following the series of migratory episodes that led to the colonisation of Europe, Asia and the Americas. It can be estimated that this second regime was well in place about 15,000 years ago. The colonisation of the islands of the Pacific is much later28 and can be treated separately, as can the cultural evolutions that accompanied or resulted from the discovery of agriculture: sedentarisation, new religious practices, etc. The transition which concerns the passage from the first to the second of these regimes is therefore characterised by the different migratory waves of our species on the surface of the planet, in Africa and on the other continents.
15In analysing the two regimes, pre- and post-transition, we can first concentrate on the ‘ways of inhabiting’ and the ‘occupation of space’ (cf. figure 4, chapter 2).
16The two regimes are characterised by an identical local demographic structure: human populations structured in small nomadic groups of a couple of dozen people, likely thirty on average29. The densities of the population, if they increased slowly over time, were then extremely feeble: the global population is estimated at more or less one million individuals 40,000 years ago, and from six to seven million individuals 8,000 years ago.30 31.
17Locally, numerous archaeological sites suggest that our ancestors’ places of habitation were not chosen by chance but responded to different criteria: shelter on the height of a promontory offering good views, protection against predators, proximity to sources of water, etc. The data are less precise for the more ancient periods, but the evolutions which could have taken place over the course of time are not necessarily linked to the transition. Subsequently, at a global level, the difference between the two regimes is important, since before the transition, the modern human population was confined to one continent, whereas it afterwards covered several, and in particular areas previously lacking all human presence, Australia and the Americas.
18If one focuses on the ‘forms of mobility’ in the two stationary regimes, apart from the nomadism already mentioned, it is possible to signal a general evolution towards lesser constraint on displacements as a result of environment. Indeed, if the most ancient migrations of Homo sapiens better respected the topography and the biotope, the migratory routes along the coasts of the south of Asia, then towards Australia from the Asiatic south-east, suggest the overcoming of formidable obstacles: arid or swampy regions, mangroves and the deltas of great rivers32, and especially the vast expanses of water in the region of Wallacea, crossed to reach the Sahul. The passage to the Americas by the Bering Strait later on is another striking example of adaptation to difficult environments. Thus, even if evolutions in climate have partly determined human migrations,33 it appears clearly that our ancestors have liberated themselves by stages from their initial ecological niche. Thus, in the particular case of navigation, a plausible evolution is the passage from the crossing of short distances, perhaps accidentally, to longer crossings, intentional and planned.34 It is possible that particular motives underpinned the development of these new forms of mobility, but seeking to identify them remains a speculative exercise.
19Our limited view of the past allows us to evaluate only partially the failures35 of certain migratory episodes: many attempts at migration may have failed without leaving any trace, and it is difficult to know if a territory was difficult to ‘conquer’. The discussions concerning the colonisation of the Sahul illustrate this point, with opposition between scenarios of accidental and intentional peopling. According to the first, some individuals canoeing in proximity to the shores of Wallacea were carried by the wind and the currents as far as Sahul. Such accidental scenarios, however, run up against the slim chances of survival of humans caught in such a predicament, especially if they did not have water and food for the time of the crossing. They are also problematic with regard to the number of individuals theoretically necessary for the successful peopling of a new territory.36 To these, therefore, various authors37 38prefer ‘intentional’ scenarios, according to which the crossings were not a matter of chance but were planned within human groups and made by sight, with preparations considerably increasing the chances of survival. Even with such precautions, it is not improbable that numerous human beings would have lost their lives in such an enterprise, and that the colonisation resulted from a number of successful crossings spread out over time.
20With regard to their ‘means of sustenance’ and ‘relationship to their environment’ (see figure 4, chapter 2), the two regimes, before and after the transition, are identical, in that our ancestors’ subsistence was dependent above all on hunting and gathering. However, during the transition, forms of progress can be noted that denote greater understanding and exploitation of the environment. Besides the occupation of different ecosystems, archaeological sites in New Guinea suggest the cutting down of certain plants to benefit the growth of other edible ones almost 50,000 years ago39. In Australia, the evolution of the local climate and the monsoon 40,000 years ago indicate an increase in fires, probably then used to orient prey during hunts40. Analysis of the remains of hunting and eating, moreover, shows variations of diet, and in particular the hunting of certain animals according to their seasonal presence and pattern of movement41. One notes as well the development of fishing42, which may have been particularly significant for the groups that followed the Asian coasts in the direction of south-east Asia: the coastal navigation route, at the interface between two milieux, marine and terrestrial, could have favoured the development of fishing techniques43 or the gathering of shellfish.
21The ‘forms of social interactions’ are difficult to identify. Interactions between groups were limited until recent times, even if the constraints of genetic mixing and the spatial distribution of lithic material suggest that various exchanges must have taken place44. Estimating their frequency is difficult because of the lack of data. One can, however, give a rough quantitative approximation thanks to simulations: frequency was on the order of the year rather than the month or the decade45. It is not until about 15,000 or 20,000 years ago that the archaeological record contains clear traces of exchanges over long distances or the existence of regional cultural units, for instance as attested by personal ornaments46. Within the groups, it is difficult to go beyond general distinctions in the division of labour, whether between the sexes or according to expertise. One can nevertheless note, beginning with the first regime, the presence of symbolic traces on tools or decorative objects, which are probably the tangible reflection of social relations at the time47. In the course of the transition, the ever greater specialisation of tools suggests activities constantly more diversified and probably the development of differential skills among members of the group. The pronounced development of paintings, sculptures and other objects with a strong symbolic value surely goes along with a densification of representations and cultural activities: belief in supernatural entities, funerary or initiatory rites, etc.
22In the final analysis, the ‘forms of power’, whether between or within groups, are unknown to us until very recent periods. Only the projection of modern ethnographic referents – with, for example, the role played by individuals such as shamans – can provide landmarks in this respect.48
23It thus appears, from analysing the broad characteristics of the regimes before and after the transition, that the spatial expansion of our species over the course of time is a key element of the transition. Different developments accompany this expansion, but these are gradual, and it is not easy to say if they are specifically the consequence of the migrations (with, for example, the encounter with other species or other environments) or marks of the more general and autonomous development of our species.
The causes of the first transition
24The characteristics previously described of the pre- and post- transition phases can be placed in sequential relation with one another. It is possible, however, to go further in shedding light on the possible causes of the transition.
25Climatic and environmental evolutions may explain in part the migrations of human populations. In particular, more difficult conditions could have led to the extinction of populations, or at least to their displacement, with phenomena of geographical contraction and abandonment of certain regions as too cold, too arid, etc. This is postulated, for example, for Homo sapiens in the Near East 75,000 years ago49, but other demographic responses to the environment are also observed in the course of stage 3 of the last ice age (from 60,000 to 25,000 years before the present) for Neanderthals and modern humans.50 Nevertheless, as has already been pointed out, Homo sapiens developed greater autonomy in relation to the environment, something made possible by significant cultural and technical development and better understanding of its functioning. These advances attest to the existence of increased cognitive capacities, compared with those of more ancient human species. One can see in them a driving force, if not the essential one, of the transition. Different cognitive evolutions have been given prominence: development of the working memory,51 cognitive fluidity permitting the transfer of analytical functions from one cognitive domain to another,52 etc. Such evolutions, especially considered together, can account for a number of the evolutions observed over the last 100,000 years.
26Even if researchers such as Klein53 proposed in the 1990s that the major cognitive evolutions took place about 50,000 years ago, shortly before the cultural explosion of the Upper Palaeolithic, the speciation event that brought Homo sapiens to light is the best moment for envisaging such transformations. Modifications of the cerebral structures may indeed be the physiological basis of such changes. Still, the lag must be explained between the appearance of our species and the belated character of certain behavioural manifestations, such as the cave paintings. Here, various elements should be taken into account. The first is that our image of Homo sapiens before his departure from Africa is still very much incomplete. While European soil, more densely excavated and for a longer period than African, furnishes us with numerous sites for recent periods, the accumulation of discoveries in Africa is progressively pushing back the beginnings of human cognitive and cultural modernity54. A second important point is that the possession of cognitive aptitudes does not necessarily signify their application. The case of writing is a good illustration: while every human being can learn to write, numerous human communities have recourse only to oral communication. Thus, one can speak of ‘cognitive potential’, and the passage from potential to actual realisation depends on particular cultural, social and environmental conditions, such as, perhaps, the encounter with other species or the need to deal with new environments. Biology, cognition and culture are therefore interconnected in a complex fashion, with interactions that may be reciprocal. The causes of human migrations out of Africa thereby form, in all probability, a complex difficult to resolve into component elements.
27Among the cognitive acquisitions of our species, language in its modern form occupies an important place. The palaeoanthropologists agree in attributing modern language to Homo sapiens at the time of the cultural explosion that took place 45,000 years ago, especially in Europe, but the question is not as clear-cut for earlier epochs. The colonisation of Australia a little more than 50,000 years ago, with the construction of sturdy boats and the possible planning of crossings, constitutes another interesting point of anchorage. The technical mastery of the navigators bears witness to the development and transmission of skills over time, to a large degree of cooperation, and to the elaboration of a shared project within a group, namely taking to the sea to arrive at a distant land. All this seems scarcely possible with the aid only of rudimentary signals linked to the emotions or the satisfaction of immediate needs. It appears from this that communication of an essentially modern kind was then already well in place. More generally, one wonders whether modern language played a decisive role in the success of the migrations out of Africa by supporting more complex interactions between individuals by means of expression.
28Determining the causes of the transition does not make possible precise specification of the properties of the migrations which are its spatial signature. The geographical and temporal expansion of our species, if it seemingly stems from new cognitive capacities, must also be studied, to be understood, from the angle of the processes of interaction between human groups and their environment. The goal of the modelling developed in this chapter is thus to envisage how the conquest of new spaces with heterogenous and limited resources could have taken place. The model was developed in two stages. In an initial phase, the authors, ‘geographer-modellers’, took their inspiration from the narrative of their ‘linguist-thematician’ colleagues regarding the great migrations of Homo sapiens to develop the model HU.M.E (HUman Migrations and Environment). This model’s objective is to simulate modes of peopling new territories in a context of environmental perturbations. The main point is to explore the role of mechanisms of innovation and diffusion concerning the manner in which human colonisation evolves over the long term. In a second phase, this general model was confronted with, and adapted to, the specific question of the movement out of Africa described in the first part of this chapter55.
Modelling the peopling of new territories
29A number of questions taken up previously can be illuminated by a modelling approach: how did the human groups spread out in Africa and Asia? At what speed? Models corresponding to different philosophies make it possible to engage these questions, and one can, notably, distinguish the models aiming to reproduce a past reality and those exploring a thought experiment.56 57
30In the first case, the objective is to reconstruct as faithfully as possible the phenomena as they actually took place, and to test hypotheses concerning the processes which caused them. In the case of the prehistoric migrations, this is all the more difficult because the period concerned is remote and/or because the events, such as the restoration of water levels mentioned earlier, have destroyed the archaeological traces. In certain cases, the genetic data58 5960allow one to mitigate this insufficiency of archaeological data in order to estimate the extent and the density of the peopling of different places and at different epochs. The emblematic model of Ammerman and Cavalli-Sforza61 concerning the increase and the dispersion of populations in ancient times, a model known as ‘reaction-diffusion’, has given rise to a substantial literature and been applied to many cases: diffusion of the Neolithic in Europe62 and America, migrations of hunter-gatherers in the Palaeolithic, etc.
31The models in the second category are more exploratory, indeed experimental. The purpose is no longer to reproduce empirical facts but rather to explore a thought experiment by simulation. Thus Premo63 defends the idea that a model, for instance an agent-based one, is a tool for constructing alternative histories by multiplying scenarios of the ‘what if’ variety. He proposes to replace the question, ‘What happened in region X during period Y?’, by this one: ‘How likely is it that behavior Q or trait Z would evolve in the population in region X during period Y given a wide range of plausible environmental conditions and alternative histories?’. Such models are productive and enable reflection, even when the data are meagre, on the functioning of different processes and the plausibility of different changes. The simulations then serve to ‘generate’ hypotheses, rather than to ‘test’ them as in the first category64, and to explore different plausible alternative modes of evolution65. The model HU.M.E. subscribes to this approach.
Models of migration and peopling of new spaces: a brief state of the art
32The transition modelled by HU.M.E. relates to the process that leads a space initially void of all human presence to become peopled as the effect of a wave of migration. More precisely, the model explores the possible history of human colonisation of an unoccupied space in the context of environmental perturbations. The objective is to explore the effects of different parameters and different initial situations on the spatial configuration and the rhythm of peopling. The central process is that of the migration, and we wish to propose a brief review of the literature on its modelling in prehistoric times.
33The models differ first by their level of abstraction of the geographic space. Young66 considers agents representing individuals or groups which move in an abstract space, homogenous and isotropic (this model is set out in chapter 3). Numerous studies, on the other hand, anchor their model in a geography approximating reality: for example, to model the diffusion of the Neolithic within the European space, Parisi et al.67 reconstitute the expansion of farming in this region on a fine scale (cells of 70 km²). The level of abstraction, moreover, concerns the mechanisms modelled. The more abstract models represent the movement as a simple random walk (chapter 3). By contrast, other models explicitly formalise the mechanisms leading an individual or a group to move according to decisions which one can characterise as rational. In Parisi et al.68, for example, people move when the ‘carrying capacity’ of the cell they occupy becomes insufficient, that is, when their consumption needs exceed the available resources. In that case, a fraction of people migrates to an adjacent cell, chosen at random from among the unoccupied cells at proximity presenting a large enough agricultural potential.
34The most abstract models are the most generic: by varying a small number of parameters, a single model can reproduce the broad outlines of a process of colonisation in quite distinct spaces and at different periods. For example, in Young’s model, certain combinations of rates of demographic growth and of migratory propensity lead to a front of colonisation, such as is observed in Europe in the Neolithic, while others, by contrast, issue in a diffuse colonisation over the whole available space, similar to the colonisation of Australia in the Pleistocene period69. More realistic perspective is obtained, whether by introducing environmental data for the space considered70 71or more sophisticated behaviour for the agents72.
35The model HU.M.E. was developed with a multi-agents system (MAS) in which groups of hunter-gatherers are formalized as ‘agents’ (cf. chapter 3).
The principal components of the HU.M.E. model
36The HU.M.E model studies, by means of simulations, the peopling of an unoccupied space by groups of hunter-gatherers (cf. inset 1). It is of an intermediate level of abstraction, with highly stylised rules of behaviour, and a space likewise stylised but heterogenous. The migrations of the groups are essentially motivated by the search for resources. They exploit their environment with a certain level of expertise (formalised in the model as a set of techniques), which may evolve and ensure more efficient exploitation. The goal of the model is to assess the respective roles of three interacting factors – demography, resource availability and technical innovation – in determining the form, the speed and the durability of the peopling of a new space.
37Two types of entities are defined (figure 2):
the agent-group, representing a group of hunter-gatherers comprising thirty or so individuals. It exploits the resources of its environment (i.e., of the cell which it occupies) in keeping with its technical level and stores energy as a result. It is capable of innovating in order to exploit the resource more efficiently and to imitate neighbouring groups possessing more elaborate techniques of exploitation. In case of migration, the group-agent moves to a randomly chosen neighboring cell.
the cell, corresponding to a portion of territory. The exploitable cells are characterised by a level of resources which can be considered as a biomass. This regenerates itself with time and can be affected by exogenous perturbations, such as various climatic or environmental changes.
38Two types of interactions are associated with these entities and serve as a driving force in the evolution of the system:
between the group-agent and its environment: when the resources in the latter are insufficient, the agent-group leaves the cell;
between the different group-agents situated on the same cell: relations of competition for the resources and of imitation, with the possibility of transfer of the technical level.
Block-by-block construction of the model
39The presentation of a model’s functioning can take different forms. We propose here an incremental approach in order to introduce the generic mechanisms of the model and highlight the role of each. This progressive construction issues in four versions of the model (indicated as V0 to V3), from the simplest (figure 3.a) to the most complete (figure 3.b).
40In the version V0 of the model, only two mechanisms exist, migration and demography (figure 3.a). Migration takes place from one cell to another according to a random walk: the agents have neither prior knowledge nor preferences. The environment is neutral: the agents have no particular ‘reason’ to move; the decision to migrate is determined simply by a parameter fixing the probability of moving at each time-step. The demographic increase is also determined by a single parameter, which assigns at each time-step the probability that a group will divide into two groups and thereby give rise to a new group-agent. With only these two mechanisms, the model reproduces the two forms of colonisation observed by Young: i) with strong demographic growth and weak mobility, a broad front of colonisation (figure 4.a); ii) with a low birth-rate and strong mobility, a diffuse colonisation of entire space (figure 4.b).
41The second version (version V1) consists in introducing mechanisms employing the notion of ‘resources’. These resources are partly ‘external’, derived from the environment, and whose exploitation is necessary for the survival of the group-agents (food, tools). The resources can also be ‘internal’ and result from a process of accumulating the benefits brought by a period of sedentariness: rest, possibilities for making preparations and, in general, the development of a good physical and relational state, etc. These ‘internal resources’ have been formalised in the model by a variable we chose to metaphorically name ‘energy’. Energy increases at each time-step when the group remains on the same cell and diminishes with each migration. This variable thus intervenes in the rules formalising the movements of the group-agents (movements implying an expenditure of energy). It serves as well to regulate demography, since too low a value entails the disappearance of the group. In this version, constraints connected to interactions with the environment (scarcity of the resource), with energy and with competition between the groups are the motive forces of the migration. Movements ‘for no reason’ (unrelated to the environment) are nevertheless still maintained, with a low probability, to take account of other logics than response to environmental constraints (cf. chapter 6 for a more detailed approach to these alternative logics).
42The third version (version V2) of the model integrates the mechanisms of innovation and imitation that take into account more elaborate behaviour of the group-agents. Innovation permits a group to acquire a new technique, more advanced, which permits more efficient exploitation of the resource. If the resources of the cell on which it finds itself are insufficient, the group may respond to that pressure with a technical innovation. Thanks to this innovation, it will be able to take care of its needs without supplementary resources73. The innovation thus constitutes an alternative to migration, with a consequent increase in the energy allowed by a longer period of sedentarisation. A group may also innovate by imitation: when two groups find themselves on the same cell, the group with the weaker technological level may acquire, through imitation, the level of the other group. In figure 4c, which corresponds to this version V2 of the model, the groups are represented by squares whose size is proportional to their technical level. A differentiation between the groups emerges rapidly, and the most advanced technical levels appear in the zones where the concentration is greatest, that is, those which are closest to the first places colonised.
43Finally, in the most complete version of the model, V3, fortuitous perturbations occur and affect the resources. Figure 4d shows how one of these has affected the level of resources of a central zone of the space, creating an obstacle to the migratory advance of the groups. Only those that have accumulated sufficient energy will succeed in crossing this ‘inhospitable’ stretch of land.
44All these mechanisms interact and give rise to varied situations. Furthermore, overall behaviour is made random, so that no configuration emerges in deterministic fashion. This version V3 of the model is used in the explorations presented in the following section.
Exploring by simulation the conditions for success in peopling a new territory
45The model has been used to explore the conditions for success versus failure in peopling an empty space. The point is to evaluate the type of situation to which each simulation leads and to compare different simulations in terms of success or failure. To do this, three indicators characterising the state of the peopling system at each time-step have been constructed:
the survival of the population, measured by the total number of group-agents: the more numerous they are, the more successful the colonisation is taken to be;
the extent of the space peopled, measured, on the one hand, by the average distance between the group-agents’ positions and the starting point of the colonisation (figure 5.a) and, on the other hand, by the number of cells occupied. This is a second dimension of the success of the colonisation;
the capacity to cross the sea, measured by the percentage of groups having reached the island (figure 5.a). The discussion above of the colonisation of the Sahul stressed the difficulty of such crossings. Their success, with the consequent opening up of new territories, contributes to the overall success of the colonisation.
46Exploration of the output of the simulations was carried out using the visualisation platform VisuAgents74. Figure 5.b, for instance, represents the case of a successful colonisation, and one can follow over the duration of the simulation the evolution of the number of group-agents as a function of the average distance from the starting point. The intensity of the colour of the point reflects its position relative to the duration of the simulation: bright colours represent the states of the system at the beginning of the simulation, dark ones the states at the end of the simulation. The number of groups begins by diminishing, while the distance to the starting point increases rapidly; then one observes a point of reversal: the number of group-agents begins to increase at the same rate as the distance from the starting point. Finally, after about 400 repetitions, the distance stabilises. The totality of the space is then peopled.
47Calculating these indicators for the final situation (fixed at time-step = 20,000)75 makes it possible to characterise the state of the system and to evaluate the degree of success of the spatial colonisation.
48An experimental design was conceived in order to explore the results of the simulations and test their sensitivity to the variation of four parameters associated with the key mechanisms of the model. These are the probability that a group will divide (demography), the quantity of energy accumulated in case of sedentarism, the cost in energy of a migration and the probability of innovation. The point is to explore the relative importance of the different processes involved with the help of the concluding indicators defined above. Table 1 presents the parameters entailed with their respective domains of variability.
Table 1: Mechanisms and key parameters of the model.
Mechanisms | Key parameters | Detailed name | Explanation | Reference value and domain of value |
Demography | PR | Probability of division | Reproduction of groups (demographic growth) | 2.10-5 (from 2.10-5 to 16.10-5, by increments of 2.10-5) |
Migration | ED | Energy linked with movement | Expense of energy entailed by a migration | 0.12 (from 0.10 to 0.20, by increments of 0.02) |
Innovation | PI | Probability of innovation | Probability that a group may improve its technical level | 0.055 (from 0,03 to 0,06, by increments of 0.005) |
Accumulation of energy | EA | Energy accumulated | Quantity of energy accumulated at each time-step when a group remains on a cell | 1 (from 0.5 to 1.5, by increments of 0.25) |
49In order to illustrate the results of the simulations, we first present the output for the eight experiments corresponding to the variations in the first parameter, the probability that a group will divide (PR in table 1). These experiments are classified from E1 (PR=2.10-5) to E8 (PR=16.10-5) and permit exploration of the variability of the results, with all the values of the other parameters set at their value of reference. The simulations cover 20,000 time-steps, with a record made every 100 time-steps. These records make it possible to represent different trajectories of an indicator, or of a set of indicators, and furnish us with several examples for illustrative purposes.
50Figure 6 thus crosses, for the experiment E3, the total distance travelled on average by the groups (on the X-axis) with the standard deviation of that indicator (on the Y-axis). The latter gives information about the degree of coverage of the peopling (the case of a high value) or, on the contrary, of the concentration of the ensemble of groups in the space (the case of a low value). This graph thus makes it possible to follow the dynamic of the colonisation and its spatial pattern. The cloud of points represents the trajectory of the system of peopling, with a total distance covered which increases in the course of the simulation. The position on the Y-axis takes into account phases of contraction (a low standard deviation of the distances from the starting point) and dispersion of the groups (high standard deviation). These phases are seen to be cyclic. This is linked to the perturbations which, by diminishing the resources in a portion of the cells, compel the groups to come closer together.
51Figure 7 illustrates the difference between the experiments E2 (low PR value) and E7 (high PR value, in b). These figures cross the number of occupied cells (on the Y-axis) with the average technical level of the groups (on the X-axis). Thus, for a low PR value, the colonisation is twice as diffuse as for a high value (the number of occupied cells reaches 140 for (a) and only 70 for (b)), and the final average technical level is lower. This result is in part linked to the fact that with a greater density of groups is associated a better diffusion of technical levels. One observes in the two cases that the technical level may change abruptly over time, because for close values of time (represented by the intensity of colour), technical level varies significantly (from left to right on the plots). These variations can be explained by the perturbations, as they lead to a contraction of the peopling process which favours the diffusion of technical innovations among group-agents.
52The preceding results are produced by simulations using the same random seed, in order to highlight the effect of a change in the value of a parameter. Since the model is stochastic (the actions of the group-agents being determined by probabilities), it is nevertheless interesting to vary this random seed and conduct several replications (300 in our case) to explore more systematically the results of the simulations in terms of success or failure of the colonisation. A comparison has thus been made between the experiments corresponding to eight values in the “rate of reproduction” parameter tested. Table 2 shows that the relation between the increase in this parameter and the total number of groups at the end of the simulation is not as simple as one would suppose intuitively. The relation is indeed regular for experiments E1, E2 and E3, the number of groups at the end of the simulation increasing with the rate of reproduction. This regularity is then broken, and experiments E5 and E6 lead to lower values. The most successful colonisations, then, are associated with rather low rates of reproduction. These results suggest an interaction (not explicitly introduced into the model) between competition for resources and demography. Indeed, a greater number of groups fosters both a better diffusion of the innovation and greater pressure on the resources. It thus appears that for E5 and E6, the advantages of growth are cancelled by the lack of resources that renders the system vulnerable. In this situation, a perturbation may lead to the extinction of a significant number of groups (E5 et E6), and elevated demography is needed to counteract this phenomenon (E7 et E8).
Table 2: Average number of groups (Nb.), for 300 replications, at the end of the simulation for eight values of parameter PR (probability of reproduction of the group)
| E1 | E2 | E3 | E4 | E5 | E6 | E7 | E8 |
PR | 2.10-5 | 4.10-5 | 6.10-5 | 8.10-5 | 10.10-5 | 12.10-5 | 14.10-5 | 16.10-5 |
Nb. | 100 | 120 | 140 | 90 | 10 | 25 | 40 | 70 |
53A systematic exploration of the space of the parameters has shown that certain sets of parameters lead the total of 300 replications to the same result; certain issue in the emergence of peopling that is durable and covers the entirety of the space, while others lead systematically to the extinction of all the groups. We have rejected these sets of parameters as improbable. Indeed, even if the situation is still imprecise, we know that Homo sapiens succeeded in colonising virtually all the planet, despite our awareness of the likely difficulties or failures at certain epochs: extinction in the Levant some 75,000 years ago, the probable difficulty of reaching and peopling Australia, the challenges of crossing the Bering Strait (cf. the first section of this chapter).
54The choice has therefore been made to concentrate on those groups of values of the parameters producing a certain variability in the results, that is, allowing a greater role and visibility to random processes. Some sets of parameters thus lead, in the course of the replications, either to a failure or to a complete colonisation of the region, featuring, moreover, certain groups that succeed in overcoming the obstacle posed by the sea. The classification of the 300 simulated trajectories with the set of parameters chosen as a reference has made it possible to bring to light four kinds of trajectories (figure 8.a): success of the colonisation with a constant increase of the number of groups (red curve), failure of the colonisation (green curve) and two intermediate situations (blue and orange curves). The figure at the right shows the relation between three indicators of success of the colonisation –the extent of the space colonised (on the X-axis), the capacity to cross the sea (on the Y-axis) and the number of groups (size of the circles) at the end of the simulation. Notable is the emergence of three principal types of colonisation: the red and orange points illustrate the combination between a significant demography, a colonisation extended to the whole continent76, and a small proportion of groups on the island (Type I); the blue and green points correspond to a feeble demography (small size of the circles) and to a colonisation of slight spatial extent, whether on the island (the case of the points at the top, Type II) or on a small perimeter of the continent, as reflected in the wide dispersion of the ‘average distance’ indicator (Type III). These three types may be interpreted as the three principal ‘attractors’ of the dynamics of the peopling system. Certain trajectories lead to intermediate situations (about 20% are outside the circles in the broken line of figure 8.b), something that shows a certain heterogeneity of behaviour surrounding the principal attractors.
Discussion and perspectives
55The approach through simulation presents a real interest for exploring the possible dynamics of the system of peopling in the past, in a context where observations are infrequent and highly heterogenous in time and space. It is not possible, then, to calibrate a simulation in precise fashion, since the data are deficient or imprecise, and validation is dependent on the judgements of experts. By contrast, the model constitutes a tool for assisting reflection and permits exploration of the effects of different hypotheses concerning the behaviour of groups of hunter-gatherers in the face of the environment and their interactions, making it possible to say ‘how’ the peopling of a new territory took place. The approach aims to be stylized, and the model does not purport to reproduce the historical reality accurately. On the other hand, its interest is to suggest non-linear effects, not always intuitive. The results of the HU.M.E. model have made possible reflection on concrete questions, for example: how many times has a colonisation been attempted before succeeding? What are the roles of stochasticity and known processes in explaining success and failure?
56Certain mechanisms introduced in the model are classic. Such is the case of the scarcity of resources as a triggering element in a migration. Others are more specific, such as the mechanism of diffusion of innovation by geographic proximity. This mechanism, classic in geography, has no empirical basis for the periods dealt with here. Similarly, the mobilisation of energy does not correspond to a concept classically employed for this period. On the other hand, the work of Nonaka et Holme77, highlighting the interest of a multi-agent system (MAS) to study the movements of agents seeking resources in a heterogenous landscape, accords a central place to energy. Their model does not concern human groups, but, on the conceptual level, the situation is the same as in the HU.M.E. model (heterogenous resources which regenerate themselves and mobile predators that consume them). In the context of the hunter-gatherers of 70,000 years ago, there is no storage, and the concept of energy is used rather in a metaphorical way to translate the hypothesis that one group remaining a certain time in the same geographic zone may have benefited from a time of preparation – to organise the crossing of an obstacle, for example.
57Beside movement linked to the scarcity of resources, there exists in the model a slight probability of ‘migration without reason’. This rule has been suggested by the thematicians of the group in order to accommodate other logics of migration, very likely present but difficult to determine with precision. These different logics have not been forgotten, and have been grouped, for simplicity’s sake, in the same category, whose label, ‘departure for no reason’, signifies simply, ‘without the reason of scarcity of resources’. A multitude of good ‘reasons’ can indeed be imagined in the ‘real’ world.
58The HU.M.E. model lends itself to other developments. In its current version, it will be used to explore in greater detail the role played in the success of a colonisation by the form of diffusion of technical innovation. Additionally, the model can be enriched by endowing the agents with genetic traits, thus making it possible to trace their genealogy (in order, among other things, to frame the synchronic distributions of genes in a diachronic structure of transmission) and to confront the results of the simulations with those furnished by the geneticists (and thereby highlight the possible mismatch between reconstructed phylogenetic trees and actual filiations).
Notes de bas de page
1 In biological classification, a subtribe is a taxon situated immediately above genus.
2 Harmand Sonia, Lewis Jason E., Feibel Craig S., Lepre Christopher J., Prat Sandrine, Lenoble Arnaud, Boës Xavier, et al., « 3.3-million-year-old stone tools from Lomekwi 3, West Turkana, Kenya », Nature, 521(7552), 2015, pp. 310–315.
3 Prüfer Kay, Racimo Fernando, Patterson Nick, Jay Flora, Sankararaman Sriram, Sawyer Susanna, Heinze Anya, et al., « The complete genome sequence of a Neanderthal from the Altai Mountains », Nature, 505(7481), 2014, pp. 43–9.
4 Reich David, Green Richard E., Kircher Martin, Krause Johannes, Patterson Nick, Durand Eric Y., Viola Bence, et al., « Genetic history of an archaic hominin group from Denisova Cave in Siberia », Nature, 468(7327), 2010, pp. 1053–1060.
5 Wolpoff Milford H., Wu Xin Zhi, Thorne Alan G., « Modern Homo Sapiens Origins: A General Theory of Hominid Evolution Involving the Fossil Evidence from east Asia », The Origins of Modern Humans: A World Survey of the Fossil Evidence, New York, Liss, 1984, pp. 411-483.
6 Wolpoff Milford H., Hawks John, Caspari Rachel, « Multiregional, not multiple origins », American Journal of Physical Anthropology, 112(1), 2000, pp. 129–136.
7 Stringer Christopher B., « Modern human origins: progress and prospects », Philosophical transactions of the Royal Society of London: Series B, Biological sciences, 357, 2002, pp. 563–579.
8 Lahr Marta Mirazon, Foley Robert, « Multiple Dispersals and Modern Human Origins », Evolutionary Anthropology, 3(2), 1994, pp. 48–60.
9 Lahr Marta Mirazon, « The Multiregional Model of modern human origins: a reassessment of its morphological basis », Journal of Human Evolution, 1994.
10 White Tim D., AsfawBerhane, DeGusta David, Gilbert Henry, Richards Gary D., Suwa Gen, Howell, F. Clark, « Pleistocene Homo sapiens from Middle Awash, Ethiopia », Nature, 423(6941), 2003, pp. 742–747.
11 Cann Rebecca L., Stoneking Mark, Wilson Allan C., « Mitochondrial DNA and human evolution », Nature, 325(6099), 1987, pp. 31–36.
12 Underhill Peter A., Passarino Giuseppe, Lin Alice A., Shen P., Lahr Marta Mirazon, Foley Robert A., Oefner Peter J., Cavalli-Sforza Luigi Luca, « The phylogeography of Y chromosome binary haplotypes and the origins of modern human populations », Annals of human genetics, 65, 2001, pp. 43–62.
13 In molecular anthropology, those genes strongly tied together have a high probability of being transmitted together in the course of generations. This shared heritability and rare mutations make it possible to distinguish different populations on the basis of their genes and to study their history. In this context, a haplotype is a set of alleles (i.e., variants) of genes which are transmitted together, and a haplogroup is a set of haplotypes which defines a population and the shared ancestor of its members.
14 Liu Hua, Prugnolle Franck, Manica Andrea, Balloux François, « A geographically explicit genetic model of worldwide human-settlement history », American journal of human genetics, 79(2), 2006, pp. 230–237.
15 Liu H. et al., « A geographically explicit… », op.cit., pp. 230–237.
16 Mellars Paul, « Going East: New Genetic and Archaeological Perspectives on the Modern Human Colonization of Eurasia », Science, 313(5788), 2006, pp. 796–800.
17 The Bering Strait is the land-passage that connected Siberia and Alaska when the the sea-level was low.
18 Pitulko Vladimir Victorovich, Nikolsky Pavel A., Girya E. Yu, Basilyan Alexander E., Tumskoy Vladimir E., Koulakov Sergei A., Astakhov, Sergei N., Pavlova E. Yu, Anisimov Mikhail A. « The Yana RHS Site: Humans in the Arctic Before the Last Glacial Maximum », Science, 303(5654), 2004, pp. 52–56.
19 Kitchen Andrew, Miyamoto Michael M., Mulligan Connie J., « A Three-Stage Colonization Model for the Peopling of the Americas », PLoS ONE, 3(2), 2008, p. e1596.
20 Quintana-MurciLluis, Semino Ornella, Bandelt Hans-J., Passarino Giuseppe, McElreavey Ken, Santachiara-Benerecetti A. Silvana, « Genetic evidence of an early exit of Homo sapiens sapiens from Africa through eastern Africa », Nature genetics, 23(4), 1999, pp. 437–441.
21 Armitage Simon J., Jasim Sabah A., Marks Anthony E., Parker Adrian G., UsikVitaly. I., Uerpmann Hans-Peter, « The Southern Route “Out of Africa”: Evidence for an Early Expansion of Modern Humans into Arabia », Science, 331(6016), 2011, pp. 453–456.
22 Allen Jim, O’Connell James F., « The long and the short of it: Archaelogical approaches to determining when humans first colonised Australia and New Guinea », Australian Archaeology, 57(57), 2003, pp. 5–10.
23 Birdsell Joseph B., « The recalibration of a paradigm for the first peopling of Greater Australia », dans Sunda and Sahul: Prehistoric Studies in Southeast Asia, Melanesia and Australia, éd. par J. Allen, J. Golson, et R. Jones, London, AcademicPress, 1977, pp. 113-167.
24 Coupé Christophe, Hombert Jean-Marie, « Les premières traversées maritimes: Une fenêtre sur les cultures et les langues de la préhistoire », dans Aux origines des langues et du langage, éd. Par J.-M. Hombert, Paris, Fayard, 2005b, pp. 118–161.
25 Coupé C., Hombert J.-M., « Les premières traversées… », op.cit.
26 Amante Chris, Eakins Barry W., « ETOPO1 1 Arc-Minute Global Relief Model: Procedures, Data Sources and Analysis », NOAA Technical Memorandum NESDIS NGDC-24, 2009.
27 Quantum GIS Development Team. Quantum GIS Geographic Information System. Open Source Geospatial Foundation Project. 2015.
28 Irwin Geoffrey, The prehistoric exploration and colonisation of the Pacific, Cambridge, Cambridge University Press, 1992.
29 Hassan Fekri A., Demographic Archaeology, New York, Academic Press, 1981.
30 Biraben Jean-Noël, « The rising numbers of humankind », Population & Societies, 394, 2003, pp. 1–4.
31 Bocquet-Appel Jean-Pierre, « When the World ’s Population Took Off: The Springboard of the Neolithic Demographic Transition », Science, 333(July), 2011, pp. 560–561.
32 Field Julie S., Petraglia Michael D., Lahr Marta Mirazon, « The southern dispersal hypothesis and the South Asian archaeological record: Examination of dispersal routes through GIS analysis », Journal of Anthropological Archaeology, 26(1), 2007, pp. 88–108.
33 Stewart John R., Stringer Christopher B., « Human Evolution Out of Africa: The Role of Refugia and Climate Change », Science, 335(March), 2012, pp. 1317–1322.
34 Coupé C., Hombert J.-M., Les premières traversées …, op.cit.
35 The term ‘failure’ is used here to indicate that a migration actually effected at a given moment did not give rise to long-term peopling. Let us stipulate, however, that we do not know the objectives of these groups and that it is a simplification to designate their attempts in terms of success or failure.
36 McArthur Norma, Saunders I. W., Tweedie R. L., « Small population isolates: A micro-simulation study », Journal of the Polynesian Society, 85(3), 1976, pp. 307–326.
37 Bednarik Robert G., « The origins of navigation and language », The Artefact, 1997, 20, p. 16.
38 Coupé C., Hombert J.-M., « Les premières traversées … », op.cit.
39 Summerhayes Glenn R., Leavesley Matthew, Fairbairn Andrew, Mandui Herman, Field Judith, Ford Anne, Fullagar Richard, « Human adaptation and plant use in highland New Guinea 49,000 to 44,000 years ago », Science, 330(6000), 2010, pp. 78–81.
40 Johnson Beverly J., Miller Gifford H., Fogel Marilyn L., Magee John W., Gagan M. K., Chivas Allan R., « 65,000 Years of Vegetation Change in Central Australia and the Australian Summer Monsoon », Science, 284(5417), 1999, pp. 1150–1152.
41 Niven Laura, Steele Teresa E., Rendu William, Mallye Jean-Baptiste, McPherron Shanon P., Soressi Marie, Jaubert Jacques, Hublin Jean-Jacques, « Neandertal mobility and large-game hunting: The exploitation of reindeer during the Quina Mousterian at Chez-PinaudJonzac (Charente-Maritime, France) », Journal of Human Evolution, 63(4), 2012, pp. 624–635.
42 Hu Yaowu, Shang Hong, Tong Haowen, Nehlich Olaf, Liu Wu, Zhao Chaohong, Yu Jincheng, Changsui Wang, Trinkaus Erik, Richards Michael P., « Stable isotope dietary analysis of the Tianyuan 1 early modern human », Proceedings of the National Academy of Sciences of the USA, 106(27), 2009, pp. 10971–10974.
43 O’Connor Sue, Ono Rintaro, Clarkson Chris, « Pelagic Fishing at 42,000 Years Before the Present and the Maritime Skills of Modern Humans », Science, 334(6059), 2011, pp. 1117–1121.
44 Duke Christopher, Steele James, « Geology and lithic procurement in Upper Palaeolithic Europe: A weights-of-evidence based GIS model of lithic resource potential », Journal of Archaeological Science, 37(4), 2010, pp. 813–824. Elsevier Ltd.
45 Coupé Christophe, Hombert Jean-Marie, « Polygenesis of linguistic strategies: a scenario for the emergence of language », dans Language Acquisition, Change and Emergence: essays in evolutionary linguistics, éd. par Hong J. Minett et W. S.-Y. Wang, Kong, City University of Hong Kong Press, 2005a, p.153-201.
46 Vanhaeren Marian, D’Errico Francesco, « Aurignacian ethno-linguistic geography of Europe revealed by personal ornaments », Journal of Archaeological Science, 33(8), 2006a, p. 1105–1128.
47 Bouzouggar Abdeljalil, Barton Nick, Vanhaeren Marian, D’Errico Francesco, Collcutt Simon, Higham Tom, Hodge Edward, et al., « 82,000-year-old shell beads from North Africa and implications for the origins of modern human behavior », Proceedings of the National Academy of Sciences of the United States of America, 104(24), 2007, p. 9964–9.
48 Lewis-Williams James David, Clottes Jean, « The Mind in the Cave the Cave in the Mind: Altered Consciousness in the Upper Paleolithic », Anthropology of Consciousness, 9(1), 1998, p. 13–21.
49 Shea John J., « Transitions or turnovers? Climatically-forced extinctions of Homo sapiens and Neanderthals in the east Mediterranean Levant », Quaternary Science Reviews, 27(23-24), 2008, pp. 2253–2270.
50 Bocquet-Appel Jean-Pierre, Demars Pierre-Yves, « Population Kinetics in the Upper Palaeolithic in Western Europe », Journal of Archaeological Science, 27(7), 2000b, pp. 551–570.
51 Wynn Thomas, Coolidge Frederick L., « The implications of the working memory model for the evolution of modern cognition », International journal of evolutionary biology, 2011.
52 Mithen Steven, « From domain specific to generalized intelligence: a cognitive interpretation of the Middle/Upper Palaeolithic transition », dans The ancient mind. Elements of cognitive archaeology, éd. par C. Renfrew et E. B. W. Zubrow, Cambridge University Press, 1994, pp. 29-39.
53 Klein Richard G., « The Human Career », Human Biological and Cultural Origins, Chicago et London: The University of Chicago Press, 1999.
54 Mcbrearty Sally, Brooks Alison S., « The revolution that wasn’t: a new interpretation of the origin of modern human behavior », Journal of human evolution, 39(5), 2000, pp. 453–563.
55 A different procedure has been adopted in the following chapter, where the model has been co-constructed by the five authors, ‘linguist-thematicians’ and ‘geographer-modellers’, in order to respond to a question of peopling of a territory in a more specific context (cf. chapter 5).
56 Lake Mark W., « Trends in Archaeological Simulation », Journal of Archaeological Method and Theory, 21, 2014, pp. 258–287.
57 Phan D. (dir), Ontologies et modélisation par SMA en SHS, Hermes-Lavoisier, Londres-Paris, 2014, 558p.
58 Ammerman Albert J., Cavalli-Sforza Luigi Luca, « The Neolithic Transition and the Genetics of Populations in Europe », Princeton University Press, 1984.
59 Eriksson Anders, Betti Lia, Friend Andrew D., Lycett Stephen J., Singarayer Joy S., von Cramon-Taubadel Noreen, Valdes Paul J., Balloux François, Manica Andrea, « Late Pleistocene climate change and the global expansion of anatomically modern humans », Proceedings of the National Academy of Sciences of the USA, 109(40), 2012, pp. 16089–16094.
60 Schneider Stefan, Excoffier Laurent, « Estimation of demographic parameters from the distribution of pairwise differenced when the mutation rates vary among sites: application to human mitochondrial DNA », Genetics, 153, 1999, pp. 1079–1089.
61 Ammerman A. J., Cavalli-Sforza L. L., The Neolithic Transition… op. cit.
62 Fort Joaquim, « Demic and cultural diffusion propagated the Neolithic transition across different regions of Europe », Journal of The Royal Society Interface, 12(106), 20150166, 2015.
63 Premo Luke S., « Exploratory Agent-based Models: Towards an Experimental Ethnoarchaeology », dans Digital Discovery: Exploring New Frontiers in Human Heritage, éd. par J. T. Clark et E. M. Hagemeister, CAA, 2006, Computer Applications and Quantitative Methods in Archaeology, Proceedings of the 34th Conference, Fargo, USA, April 2006, Budapest, Archaeolingua, 2007, pp. 29-36.
64 Lake M. W.,Trends in Archaeological…, op.cit.
65 Chevrier Benoît, « Ni espace ni temps en Préhistoire ancienne: “Out of Africa” ou le paradigm de la flèche », Mappemonde, 2012, p. 106.
66 Young David A., « A new space-time computer simulation method for human migration », American Anthropologist, 104(1), 2002, pp. 138–158.
67 Parisi Domenico, Antinucci Francesco, Natale Francesco, Cecconi Federico, « Simulating the expansion of farming and the differentiation of European languages », Evolution of Languages, 2008, pp. 1–41.
68 Parisi D., et. al., « Simulating the…», op.cit.
69 Bowdler Sandra, « The coastal colonization of Australia », dans Sunda and Sahul: Prehistoric Studies in Southeast Asia, Melanesia and Australia, éd. par J. Allen, J. Golson, et R. Jones, London, Academic Press, 1977, pp. 205-246.
70 Heckbert Scott, « MayaSim: An agent-based model of the ancient Maya social-ecological system », Journal of Artificial Societies and Social Simulation, 16(4), 2013, p. 11.
71 Parisi D. et. al., « Simulating the…», op.cit.
72 Barton Michael C., Riel-Salvatore Julien, « Agents of change: modeling biocultural evolution in Upper Pleistocene western Eurasia », Advances in Complex Systems, 15(1-2), 2012.
73 In other words, the group succeeds in extracting more energy from the same quantity of resources.
74 Cura Robin, Boukhechba Mehdi, Mathian Hélène, Le Néchet Florent, Sanders Lena, « VisuAgent – Un environnement d’exploration visuelle de données spatio-temporelles issues de simulation », SAGEO, Grenoble, 2014.
75 t=20 000 was determined by experimentation; it is the threshold above which the number of groups evolves no further.
76 Given the form of the virtual continent and the placement of the island (inset 1), the average value of the indicator ‘average distance from the start’, on the hypothesis of a uniform occupation of the space by the agent-groups, is 39 for the continent, i.e. with no group-agent on the island, and 46 for the island, i.e. with all group-agents on the island.
77 Nonaka Estuko, Holme Petter, « Agent-based model approach to optimal foraging in heterogeneous landscapes: Effects of patch clumpiness », Ecography, 30(6), 2007, pp. 777–788.
Auteurs
Department of Linguistics, The University of Hong Kong, Hong Kong.
Dynamique Du Langage, UMR 5596 CNRS et Université Lyon2, Lyon.
Université Gustave Eiffel, Laboratoire Ville Mobilité Transport, Marne-la-Vallée, France.
UMR 5600 EVS, CNRS-Université de Lyon, Lyon.
Géographie-cités, UMR 8504 CNRS-Université Paris 1, France.
Le texte seul est utilisable sous licence Licence OpenEdition Books. Les autres éléments (illustrations, fichiers annexes importés) sont « Tous droits réservés », sauf mention contraire.
Quatre ans de recherche urbaine 2001-2004. Volume 2
Action concertée incitative Ville. Ministère de la Recherche
Émilie Bajolet, Marie-Flore Mattéi et Jean-Marc Rennes (dir.)
2006
Quatre ans de recherche urbaine 2001-2004. Volume I
Action concertée incitative Ville. Ministère de la Recherche
Émilie Bajolet, Marie-Flore Mattéi et Jean-Marc Rennes (dir.)
2006