Inequality during the Iron Age in France.
Tracing the archaeological record

The Iron Age (730-125 BC) in France is of particular interest for examining the issue of inequality because societies evolved over these seven centuries through a nonlinear process of nonlinear complexity, from simple chiefdoms to archaic States. These were politically autonomous entities, which marked quite strong social distinctions through the wealth of funerary deposits and more or less monumental tombs. As we know, burial practices are not true reflections of social organization; we must also include the evidence for settlement hierarchy. Spectacular results have been achieved over the last thirty years due to extensive fieldwork, at last enabling the proper study of large settlements of this period, from simple farms to urban areas.

The archaeological data complement the information provided by Greek and Latin textual sources that describe various organization strategies: social elites, ordinary free people and slaves, age – and gender-based categories in different groups, accumulation of economic surplus, and exclusion. A brief description of the history of ideas on the evolution of inequality helps us to understand the current State of the debate, the mainstream ideas of which have fluctuated between evolutionist and anti-evolutionist approaches since the eighteenth century. Challenging new perspectives on this issue clearly favor an evolutionary perspective. Archaeology has provided an impressive amount of data studied with more efficient methods of analysis, which suggest processes of unequal growth that are better documented than before. The theoretical implications are drawn to finish highlighting a blind spot of research on stateless societies: the question of forms of political regimes.

Brief history of ideas on the evolution on inequality

In the mid-eighteenth century primarily, reflections on the causes of increased social inequalities multiplied. We find in most of these works a similar evolutionary pattern emphasizing a fundamental tension between the limited environmental resources and the universal conquering nature of the human psyche¹. This constant insecurity would have stimulated the capacities of technical innovations that enabled the increase in production and, therefore, the increase in population. It would have resulted in more numerous and more serious conflicts, which would have required coordinated ways of policing, thus, greater political complexity. From the start, there are two opposing interpretations to explain the process of increasing complexity: one offers a consensual process, the community accepting the authority of one or a few for a better management for the benefit of all², the other a contentious process imposed to consolidate the gains made by a few³. This conception, free from metaphysical beliefs, was strengthened by Charles Darwin’s theory, which gave an overall powerful materialist logic to the evolution of living things and, indirectly, of human societies⁴. Major works synthesizing the history of mankind, suggesting the first social typologies and obviously necessary for any comparative work, were completed. Two of the most notable of these were Morgan’s three stages of increasing complexity: (1) savagery, (2) barbarism, and (3) civilization⁵; and Engels’s five types of societies: (1) hunter-gatherers with egalitarian economies, (2) pastoralists with kind leaders, (3) farmers and metallurgists with surplus labor and social classes, (4) specialized warriors and craftsmen in tribal confederations, and (5) States⁶.

Between 1890 and 1930, evolutionary approaches generated a strong outcry from the believers whose intimate religious convictions and conservative political movements were shaken, and who were very concerned about the social criticisms formulated by K. Marx and F. Engels, who worked from an explicitly evolutionary approach⁷. In the academic world too, the mainstream line of thought required adhering to description and classification, and banishing any speculation. Among social anthropologists, evolutionism was rejected on all sides in the Western world. Some felt that it established a hierarchy among peoples and, therefore, supported racist attitudes⁸. Those who collaborated with the colonial administrations reckoned it offered nothing⁹.

An evolutionist current reaffirmed itself in the United States under the leadership of Leslie White, and then Julian Stewart. The main advances of this neo-evolutionism are the inclusion of the environmental dimension and the demonstration of perfectly compatible multilinear approaches¹⁰, in contrast to the rigid caricature of unilinear and mechanical perspectives denounced by opponents. The archaeologist V. Gordon Childe continued an evolutionary approach by borrowing elements of Marxism without falling into Stalinist dogma¹¹. He highlighted the revolutionary importance of the adoption of a production economy by humanity, and then of urban organization, by treating western Asia and Europe integrally. Evolutionism found itself at the same time, validated by the discovery of the genetic structure of the living (DNA). The result was that random mutations and natural selection do not operate at the individual level, but on that of genes¹². The evolutionary sequence of neo-evolutionists, however, has not fundamentally changed compared to those generated by their predecessors¹³. It was mainly enriched by details of the evolutionary sequence’s continuation by insisting on a growing complexity of task specialization and on processes of the diffusion and expansion of social innovations. Clarifications were also proposed for the factors of change. On the growth of resources, great importance was always given to technical innovation, but also to the effects of climate change. On the internal conflicts that might lead to the disintegration of societies, diverging economic interests between the dominant and the dominated, according to Engels, remained a force, but was now combined with other factors. According to R. Carneiro, the name environmental area emphasized the critical importance of war and the submission of conquered populations¹⁴. Various means for legitimizing social hierarchy have been proposed based on concrete examples from archeology or anthropology: a privileged relationship with the supernatural, control of resources, trade and/or work, the formation of alliances, information processing, risk management, inter-societal domination. The systemic perspective has become commonplace, accompanied by a positive feedback effect. It is, as it turns out, often caused by the possibilities offered by a more hierarchical and better-coordinated organization, which is able to overcome resource limitations (conquest of new lands, major development work, etc.). In general, the conceptual evolution is then characterized by the growth of the temporal, spatial, and demographic dimensions of social groups, which should be taken into account in order to be more relevant and not to remain focused on highly localized communities anymore¹⁵. Additionally, some authors have stressed the benefit of ethology and asserted the idea of a nature/culture co-evolution¹⁶. The intellectual debate, however, escalated sharply in this regard with the 1975 publication of Sociobiology, The Vew Synthesis, by Edward O. Wilson. In it, this entomologist borrowed promising theories from several researchers: William D. Hamilton’s kin theory¹⁷, Robert Trivers’s theory of reciprocal altruism¹⁸, and John Maynard Smith and George R. Price’s game theory in dynamic populations¹⁹. Wilson devoted his last chapter to human societies, hence the scandal. Reading the text carefully, without excessive prejudices, we found that he unambiguously affirmed that selective and environmental pressures had led to an “additional mental growth and a social organization that had made hominids cross the threshold leading to a more interesting autocatalytic phase of evolution”, and that mental and social change came to depend more on an internal reorganization and fewer direct responses to environmental characteristics. Social change, in short, acquired its own engine.

In 1982, Ian Hodder’s Symbols in Action²⁰ launched a new movement in archaeology called postprocessualism. Designed in opposition to the new archaeology, this movement was the archaeological version of postmodernism, which stood against the desire to draw general laws of social change, against the belief in truly objective theories, independent of the context, and against the importance of ecological factors. This conceptual nebula is characterized by two basic principles. The first is a philosophical skepticism and even a fundamental metaphysical doubt about the existence of the world outside our sensations – it would only be the product of our imagination, our representations – relying on the assumption of an absence of universal reality; current knowledge is considered a matter of perspective. Then, there is a further critique of Western society by means other than Marxism and structuralism, by isolating new structures of oppression: race, gender, culture, affirming that there is no unique truth in the past, only stories developed to gain more power in the present. These principles rejected big universalizing theories, and all those considered politically incorrect (e.g., the barbarians’ dependence on Mediterranean civilizations, or a pre-state centralized and hierarchical political organization), and declared the preference for religious and symbolic dimensions of culture, local education, microsocial dimensions, prominence of the individual actor, plurality of meanings, and relativism. Until around 2000, this mainstream movement dominated the human sciences, resulting in sterilized research by enclosing the majority of researchers in the deconstruction perspective; however, to deconstruct to the extreme leads to the dissolution of the studied object, and therefore to a total loss of significance and widespread confusion. In Theory and Practice in Archaeology²¹, Hodder tried to advocate a more nuanced approach, avoiding binary oppositions, but the damage was done. Postmodernism was the mainstream movement the conventional majority of the archaeological community followed.

New perspectives on the evolution or inequality

Around 2000, evolutionary approaches again became dominant after making very significant progress, supported by the achievements of cognitive science and ethology. For our purposes, they led to the answer to a crucial question that had previously been unexplored: How do humans manage to cooperate, within large political entities, with thousands of unrelated individuals? Two mechanisms led to a better understanding of the cooperative evolution trend among an animal species: kin selection²² and reciprocal altruism²³. The first is the fact that natural selection acts on genes and not on individuals, and a behavior that might contribute to the transmission of genes has been selected, even if it might affect the survival of the individual, especially when it comes to saving an offspring or siblings. But we also found evidence for cooperative behavior between unrelated individuals among humans and many primates. Robert Axelrod has shown that cooperation can be established between unrelated individuals if they repeatedly interact. Individuals can cooperate conditionally, that is to say, until their partners become uncooperative²⁴. They may not, however, establish trusting relationships with a large number of members, because they have a limited memory and a limited time.

Samuel Bowles and Herbert Gintis proposed a solution to this problem with the theory of strong altruism. Cooperation with many unrelated individuals implies a willingness not only to cooperate conditionally, but also to punish uncooperative behavior, even if it involves costs for the individual and lack of future gain. The higher the proportion of strong reciprocal altruism there is in a group, the more caution is required toward opportunistic individuals. If the species evolved in a context of competition between groups for the same ecological niche, groups with a higher proportion of strong reciprocal altruism could flourish more and impose themselves in the population through a group’s selection process. The arguments are mainly derived from game theory²⁵. However, strong reciprocal altruism does not explain why the formation of larger human societies, with village or urban agglomerations, was necessarily accompanied by greater social differentiation and political centralization. Weak or strong, altruism is sensitive to group size, as noted by Benoit Dubreuil²⁶ in response to Robin Dunbar’s observation according to which the average group size in primates is proportional to the relative size of the neocortex, that is to say, the ability to store information on “who’s who” and “who did what to whom²⁷”. He took up the idea that language is a form of “remote grooming”, an economical way to stay abreast of “who did what to whom²⁸”. The size of the human neocortex, compared to that of other primates, leads one to suppose that we should form groups of about 150 members. It is likely that human groups reach an optimal size, pushed up by the benefits associated with increased social pressure, but tend to peak due to difficulty for humans to monitor the actions of their peers in too big groups. Group size would be subject to a cap because of rising costs associated with social sanction; the emergence of villages and towns would be made possible by the creation of institutions that allowed a tight control of costs related to the punishment of deviant behaviors.

Christopher Boehm examined the behavior of chimpanzees and bonobos to better understand ancient human societies and, specifically, their apparent resistance to the emergence of very heightened inequalities²⁹. He derived from these comparisons the idea that archaic humans had also lived in small exogamous groups of individuals capable of selfishness, aggression, and nepotism. Their social behavior was governed by rules and standards such as the priority given to some, for the access to food, by their implicitly subordinate companions. These rules potentially benefited all members of these groups, as they imposed a social order and prevented chaos. Alpha males of archaic human groups were suitably equipped to ensure that subordinate members adhere to the rules. But at this level of reasoning there arose, he said, a fundamental question: How had these dominant males been led to resist the temptation to intimidate the subordinate members of the group or to monopolize, in excess, the resources for themselves?

He assumed that mechanisms had existed for subordinate group members to put pressure on the controlling behavior of Alpha males. He thought that, as in primate groups, subordinate members of archaic human groups might form coalitions and unite against Alpha men to limit their power and access the resources. He concluded that hunter-gatherers were engaged mainly in altruistic behavior for sharing meat, rejecting tyrannical individuals, and repressing free riders, cheats, and disloyal members. They defended equality and fairness carefully through promoting and implementing sanctions. Social deviants, bullies, and fraudsters were reprimanded, excluded, stigmatized, and often murdered. To explain this peculiarity, he stressed that the most important differences between chimps and humans were the hunter-gatherers’ language skills. Language allowed humans to limit selfishness and reward altruism by acting on the reputation of members of their groups through gossip. These people were not good or bad by nature; like us today, they were both. Although humans are naturally willing to give free rein to their selfish interests by cheating and parasitizing, they are also naturally disposed to dominate these drives when they expect they might produce side effects, such as corporal punishment and damage to their reputation. From many examples, the anthropologist has shown, moreover, that the amount of generosity and selfishness in hunter-gatherer societies is correlated with the abundance of food and other resources. Satiety promotes altruism. In times of moderate scarcity, nepotism worsens. And in the case of famine, it becomes every man for himself. It is the specific skills of language, reasoning, and planning that would have allowed early humans to establish contracts and social norms, identify deviants, and agree on ways to punish them.

This idea seems compatible with the major theories of social anthropology and historical data on the global evolution of societies toward greater inequality. Village societies of hundreds of individuals all came to create complex kinship structures (often hierarchical), with sanctioned divisions of labor: in a lineage or clan special duties are assigned to punish relatives, often to meet a higher standard of preserving honor (honor killing). Societies of hundreds of thousands of individuals have experienced a universal differentiation between a political center and a periphery accompanied by the appearance of chains of command where the cost of punishment was controlled by a hierarchy. A process of bureaucratization, militarization, and monopolization of political power accompanied the appearance of the primary State everywhere (Mesopotamia, Egypt, China, India, Peru, and Mexico). Cognitive constraints do not account for all the reasons these institutions appear, but explain the correlation between demography and social organization³⁰.

Along with the integration of knowledge about our cognitive system, the need was felt again imperiously to expand the vision beyond the local, regional, and national areas to embrace the scale of a globalized world. In line with the work of Immanuel Wallerstein³¹ and Fernand Braudel³², with their respective concepts of world system and world economies, global history has turned the page from the hostile postmodernist interlude to broader comparative approaches, to grand narratives necessarily considered accomplices of capitalism and imperialism. Christopher Bayly, in The Birth of the Modern World 1780-1914, rightly notes that “postmodern work generally ignores the ‘metanarrative’ specific that underlies them, which is both political and moralistic by its origins in its implications. Thus, many of these works seem starting from the idea that a change towards a better world could emerge if the historical engines generators of domination phenomena such as the unifying State, patriarchy; or even the rationalism of Western Enlightenment were not as powerful³³”. This comprehensive history examines transfers between different cultural zones in a systemic and comparative manner. He freed the constrictions that had been placed on the long-term macrohistorical prospects over vast areas articulating societies of unequal developmental levels, such as Southwest Asia and Europe.

In Europe, the growth in inequality is made visible by traversing the sequence of social organization types proposed by Allen W. Johnson and Timothy Earle³⁴ (fig. 1). As elsewhere in the world, this trajectory has followed linear growth neither in time nor in space. Besides the importance of these fluctuations, it should be noted that Neolithic, urban, and State appearances were systematically secondary processes in Europe, proceeding as inhabitants, goods, and ideas extended from the Near and Middle East. For Europe, strictly speaking, it is necessary to distinguish three distinct areas where political complexity and growing inequality have developed in a very staggered manner by answering different combinations of factors. The adoption of a State organization is generally considered the transition to a higher level of social inequality.

The emergence of Minoan States around 2000 BC and Mycenaean toward 1600 BC, continued by primarily borrowing an Oriental organizational model facilitated by the same type of Mediterranean bioclimatic environment, the same palatial organization policy, and participation in the same system of economic relations around the eastern Mediterranean basin; a system in which the general collapse of the 12th c. BC led to the disintegration of the Aegean States.

The Greek and Etruscan city-States, formed in the 8th c. BC, quickly spread their political organization in the form of settlements and trading posts along the Mediterranean and Black Sea coasts. Their trading partners in the European hinterland accentuate their own level of organizational complexity, but do not to take the form of State organization outside the Mediterranean climate zones. This suggests that, in humid temperate regions, agro-pastoral farming techniques were not able to produce sufficient amounts regularly enough to supply urban areas that State organizations require.

The emergence of more extensive States in humid temperate zones and steppes, took place in the 2nd c. BC promoting their integration into the economic system run by Rome and an intensification of agricultural production made possible by the widespread use of iron tools.

The emergence of States in Denmark and around the Baltic, that is, in the northern areas of the humid temperate European hinterland, farther from the Mediterranean, took place in the eighth century through an increase of maritime contacts with the southern most States that had previously been integrated into the Roman Empire during the first four centuries of the Common Era (Viking colonies).

Europe adopted technical and organizational innovations, including the State, developed in the Middle East. This secondary process cannot, however, be explained using a simple diffusion model. Specific difficulties had to be resolved for States to appear across the whole of Europe. Various combinations of factors, adapted to local social and environmental conditions, had to be implemented. They give us precious keys to understand the general issue of growing inequality. It is considered that the Iron Age began in Western Europe, thus in France, around 800 BC. Social change has occurred here in five distinct stages.

800-730 BC (Gündlingen Level) and 730-625 BC (Hallstatt C)

Compared to the end of the Bronze Age, around 800 BC, changes appear to have mainly resulted from economic factors, what we have called a “new deal³⁵”. The main centers of power have indeed changed location, favoring the most topographically remote areas, but closer to dense forests (for fuel), and deposits of iron and rock salt, and where it was also possible to develop a breeding program to produce exportable cured meats (pigs, cattle) and luxury textiles (woolly sheep). This new situation may have resulted from climatic degradation and disorganization of the bronze exchange networks to which the answer would have been the transfer of economic investment to the best assets, including other resources. However, no structural change is noticeable politically, the picture is blurry because it is composed of contradictory elements. Some elements suggest a fundamental continuity (no major break in material culture, sustainability of the basic agropastoral occupation with scattered farms and hamlets, continuation and even an increase in the practice of individual burial mounds for male elites). Other elements point to significant evidence of rupture (a break in the practice of terrestrial hoards, abandonment of many fortified and unfortified agglomerated settlements). Simple chiefdoms probably continued, a dominant chief or a council controlling local networks (lineages) and external exchange networks. The rapidly growing number of rich funerary deposits compared to non-funerary deposits reproduced the change, first demonstrated in Denmark³⁶, but that proves quite general between these two deposition categories. This alternation suggests that there were cultural choices expressing two different political ideologies: one favoring unity and solidarity of the group by more egalitarian burial practices and the ceremonial expenditure of wealth in form of votive deposits, and the other showing the distinction of elites in a more or less ostentatious and individualized manner.

628-460 BC (Hallstatt D)

During Hallstatt D, more extensive attempts at political and hierarchical integration occurred in some areas. Complex chiefdoms were formed, as a result of agropastoral productivity, which benefited from improved weather conditions and intensified connections via external exchange networks. These local and global factors combined to provide the means of enriching and developing power, taking particularly conspicuous forms during the sovereign’s funeral. This strengthening of social hierarchy has long been demonstrated in burial data. This political and economic complexity is largely confirmed by the prioritization and diversification of institutions. Through an increase of field operations on larger surfaces than before-made possible by preventive archeology; especially in the Paris Basin – we have data that reflect much more reliably the level of complexity of ancient societies than grave data alone.

The top-level sites suggest fifty-kilometer radius modules for politically autonomous territories between western Bavaria and Berry. The climax of this phenomenon, traditionally called “princely”, lasted about three generations (53-460 BC). During this period, settlement typologies diversified, with family farms being open, enclosed, or fortified and larger or smaller in size. Agglomerated settlements were open or fortified and of various dimensions, culminating with the famous « princely residences », some of which, such as Hundersingen “Heuneburg”, Vix “Mont Lassois”, and perhaps Bourges, have even approached an urban pattern. This growth in political and economic complexity has been confirmed in recent years in the group of southwestern cultures of the North Alpine cultural complex³⁷. This test of urban and State organization has nevertheless remained unfinished, the scale and the level of integration regaining the previous dimensions; those simple chiefdoms were very constrained by a difficult threshold of a day’s walking distance back and forth, that is, in politically autonomous territories of about 25 to 30 kilometers on flat land. Beyond this dimension, the cost incurred in travel time and energy for the chief and his escort made centralized management and control more difficult and fragile, requiring the ability to relay supreme power in more remote areas. However, the loyalty or fidelity of these subordinate leaders has always been a major problem for the unity of such political entities. Even territorial States larger than city-States have long been threatened by defections of great vassals, which constantly called their unity into question.

460-325 BC (La Tène A-B1)

Disintegration of politically complex chiefdoms was accompanied by a decrease in ostentations display. Even in the western margins such as Berry, Middle Loire, Aisne-Marne, Saarland, Hunsrück-Eifel and Hesse, the richest tombs, many containing Etruscan bronze vessels, are less impressive than the Ha D princely grave. Their monumentality is inferior, their funerary deposits are less impressive, and corresponding settlements are less rich and monumental. Rich tombs at corresponding institutions show a very different geographical distribution of previous “princely” centers: more dispersed with much shorter distances to the nearest neighbor. Despite maintaining privileged relationships with networks of northern Italy, it seems that the size of political entities remained equivalent to simple chiefdoms where a leader could intervene anywhere in half a day to regain control, conduct arbitration, and restore order.

The reasons for the disintegration of great Hallstatt D “principalities” remain difficult to explain. As with most things, it is probably a combination of factors that triggered collapse. Active exchange networks with the Etruscans might have become substitutes for Greek partners, upsetting established alliances with “princely” elites. In addition to this are the effects of a less favorable climate and increasingly controversial social relations, with the peasantry supporting too-despotic potentates with greater difficulty.

It seems that the solution to these difficulties of continuity was finally found in the organization of orderly migration, as suggested by textual sources from the beginning of La Tène B, around 400 BC. These migratory movements continued for a little more than a century, aiming for more southern areas. Climatic degradation might have accentuated this heliotropism, but the most striking of these movements lies in the attraction of North Alpine Celts for the more developed societies with which they previously entered into indirect contact. They clearly sought to invest in zones that put them in direct contact with Etruscans in northern Italy and Greece, through the Balkans, along the Black Sea, and to the Greek cities of Asia Minor, by crossing the Dardanelles Strait.

325-150 BC (La Tène B2-C)

Networks of cross-cultural exchange of prestige goods were most likely interrupted by migratory turbulence. Intercultural systemic relationships have, however, continued. They have changed in nature. Trade and diplomatic relations, during Ha D, with Etruscan and Greek city-States were made very attractive to the North Alpine Celts, for example, due to the orientation of their migration in northern Italy, in the Carpathian basin, in the Balkans between the Black Sea and Asia Minor’s Greek cities, and to their massive presence in mercenary armies recruited by the various Mediterranean countries³⁸.

At the end of 4th c. BC, simple chiefdoms of the North Alpine Celtic world show signs of reconstruction, including constructed shrines, with a major role as territorial markers. They illustrate increased importance for members of these political entities and their neighbors, a more sustainable symbol of their collective identity. Social cohesion, promoted and celebrated regularly in and around these sanctuaries, seems to have been additionally cemented by frequent military conflicts between competing chiefdoms, battles with weapons; and even vanquished corpses proudly displayed in these religious monuments. Deterioration of the previous level of political complexity, which had been characterized by authoritarian centralization and autonomous territories of unprecedented size, probably created a stimulus for economic competition.

We note that from the end of 4th c. BC (LT B2) major technological changes occurred in the iron and steel field. One of the most important was the production of iron agricultural tools, especially plowshares³⁹. Widespread use of iron in the ordinary tools of the peasantry implies a sharp rise in raw material production. This is a crucial qualitative and quantitative leap, well attested by the discovery of large iron mines, as in La Bazoge “Petites Rouilles” (between Normandy and Touraine)⁴⁰ and in the rapid increase in the number of forge sites in the Paris Basin⁴¹. This, logically, resulted in a strong intensification of farming production. An increase in technical efficiency, enabled by the use of iron for plowing, facilitated the opportunity for more agricultural productivity. There is, indeed, by the evidence of aerial surveys and the excavations preceding the construction of highways and railway tracks, the habit to investigate general areas previously unoccupied, especially uplands⁴². This gain in productivity, probably supported by continuous global warming, could give growers more confidence, as suggested by the gradual abandonment of the practice of planting two or three different species in the same plot. The adoption of monospecificity in this period was discovered by V. Matterne⁴³ while analyzing the composition of grain reserves in the Paris Basin. This is also based on the assurance of having a broad and well-organized solidarity; further indicative of organizational complexity in net growth.

We note, finally, the appearance of villages in the 3rd c. BC. We expect more specific publications, but it seems that on the surface of several dozen hectares, fairly small and densely occupied enclosures were identified. Various indications of craft production are concentrated, but internal organization, the composition of resident groups, and the chronology of their development are difficult to grasp. An increase in productivity, however, occurred. It was also logically accompanied by other technological improvements such as the rotating grinding wheel.

150-25 BC (La Tène D)

In the middle of the 2nd c. BC, a new attempt at social complexity reached its fulfillment with city and State emergence. Spatial organization quite clearly illustrates social organization. These are, in fact, significantly hierarchical networks that appear as archaeological finds. Economic and political integration levels are logically structured by distance, that is to say, the length and difficulty of moving. At the head of these networks is the capital city. At lower levels, there are towns and villages, and finally hamlets and farms. Two types of networks are combined within each politically autonomous formation:

• Settlements with a primarily agropastoral function, which come in several hierarchical levels, depending on their area and the investment level required for their construction, which likely reflect strong inequalities in land ownership.
• Settlements with more diversified economic functions, controlled by the family that owns land on which they were located.

Open agglomerations, which began to grow in the 3rd c. BC, grew and multiplied in the first half of the next century. The very pronounced aristocratic and military characteristics of these societies suggest that local aristocracy owned the land in question, exercised patronage on these large villages or concentrated on handicraft production, to ensure minimum safety. The trend emerges gradually, and suggests a meeting or merger of major branches of these two types of networks to form large fortified settlements. This merger took place, it seems, by the fortification of a wide area around the original lowland village, or by transferring the population of the original town from a nearby point, also with fortifications, such as Levroux⁴⁴. In many cases, this fortified urban area was created on a site where there was also a quadrangular ditch sanctuary, such as Gournay-sur-Aronde⁴⁵. Thereafter, for about a century, during the period from LT D1 to LT D2a, most low-lying fortified settlements were gradually abandoned in favor of new fortified towns on hills, for example, Villeneuve-Saint-Germain in favor of Pommiers⁴⁶.

These very large towns fulfill all the criteria of urbanization⁴⁷. Their urbanized societies also have characteristics of archaic States. The main archaeological criteria are the use of a currency and writing. A combination of the two became common in the 1st c. BC with bronze coins bearing the name of a king or chief magistrate. This implies that at least a fraction of the population had the ability to read these names. Several styles have also been discovered in different materials. Gauls certainly did not have lapidary inscriptions, however, they wrote on wooden tablets coated with wax. Caesar also refers to these, unambiguously stating that, on the one hand, writing was outlawed within the religion, but, on the other, that 368,000 migrants, soldiers, women, children, and old people (263,000 Helvetii, 36,000 Tulingi, 14,000 Latovices, 23,000 Rauraci, 32,000 Boii) he stopped at the beginning of the Gallic Wars carried tablets on which were displayed their full civil status in the Greek alphabet⁴⁸. Writing was therefore used for administrative management necessitated by the organizational complexity of any State-level political entity.

Assuming that there is little doubt that medieval dioceses had recorded the areas of the Gallic civitates, then we should be able to see changes in the size of integrated territories compared to previous periods. Eight civitates were part of a circle with a 25 to 30 km radius (2500-3600 km²), that is to say, a difficult threshold for managing simple chiefdoms, which became widespread in the mid-2nd mill. BC. It was logically the average size of pagi, these subsets of civitates. A second category of around 28 civitates had about a fifty kilometers radius (10,000 km²) and consisted therefore of federations of three or four pagi, this was the average size of complex chiefdoms during the 6th and 5th c. BC, traditionally called “principalities”. A third category consisted of 16 civitates within 75 km (22 500 km²), thus uniting 6 to 12 pagi (fig. 2). Caesar said, however, that the Helvetii State consisted of four « cantons », which would correspond to an average pagus with a 37 km radius, that is to say, a little larger than a day of walking round trip. Several signs suggest that apparent discrepancies between the sizes of certain medieval dioceses and those of previous civitates resulted from the integration of some pagi into different civitates. There is nothing surprising here, as States are historically full of variations of this kind; territorial subsets often follow a strategy to promote their own interests first. And this might explain the surprising attachment of the Mandubii, with Alesia as their capital, to Aedui State, in defiance of logic related to topographic and spatial factors, rather than their likely integration into the civitas of Lingons; the power of Aedui could have led them to expect a higher profit in terms of economic benefits and military alliance. Some pagi felt, conversely, that their interest was to resume their political independence in the course of the 1st c. BC, as in the case of the Meldi separating from the Suessiones⁴⁹. The best strategy for pagi, as for civitates, was based on various factors, likely including their cultural history and geostrategic location, as suggested by their mapping. The larger archaic States were juxtaposed within a wide band barring Gallic space diagonally from northeast to west-central, while most smaller States lined up along a strip parallel to the Channel. This bipartition strongly echoes the previous cultural millennium during which the respective Atlantic and North Alpine cultural complexes faced each other while maintaining a marked contrast in artifact styles, in residential building forms, or in burial practices. Smaller States also tend to be found along the boundaries between different sets of political and economic levels: firstly, the boundary between the Roman world, Narbonensis, and central Gaul, then that between the most integrated Gallic States and those who were less so, where aligning States are the size of single chiefdoms, a pagus or an average city-State (Eburovisques, Parisii, Meldi, Viromanduens).

Theoritical implications

The changing scale of territorial integration during the Iron Age in humid temperate Europe accords with what happened all over the world. This fluctuating type of integration scale has been clearly demonstrated in various regions in central Mexico between 800 BC and AD1700, in the Andes between 1400 BC and 1700 AD, in Mesopotamia between 4500 and 1500 BC, or in Egypt between 3700 and 300 BC⁵⁰. These developments are interspersed with sometimes drastic phases of disintegration, followed by stages of growth that always end, overall, by exceeding the highest level previously achieved. This fundamentally nonlinear change is a universal feature of globalization, at least since the Neolithic; a condition that is nearing completion today.

In Europe and elsewhere, various factors have combined to cause the collapse of this level of integration, which logically include climate changes. To avoid confusion, the subtitle of Jared Diamond’s book about societal collapse driven by climate change should be: How Societies Choose to Leif or Succeed⁵¹. These are, indeed, social organizations that are traditionally incompetent, blind, and focused on short-term management and are incapable of changing their lifestyle to adapt better to hazards. It was probably the same during the Iron Age. Climate improvements clearly correspond to the stages of growth of the scale of political integration (fig. 3).

Another factor played a major role in these changes: technological progress. Among them, the most crucial were, of course, those which were likely to increase agropastoral productivity; the level must be sufficient to maintain communities where many residents do not produce their own food. Perhaps it was not a failed urbanization attempt in the middle of the 1st mill. BC, which would explain, in part, its ephemeral nature. Dramatic growth in steel production, which transformed agricultural tools and made them available to all users, most likely provided the means for subsequent urban and State development. This implies some correct conclusions from the valuable History of World Agriculture. Neolithic to the Contemporary Crisis⁵². It focuses, however, for far too long on slash-and-burn farming systems, and ignores the widespread use of the iron plow, coulter, and scythe from the 3rd c. BC. The authors do not distinguish between uncultivated systems and light animal traction, and those in fallow and plow culture, where there was a fundamental stage of development that led to social changes of great magnitude. They called the “agricultural revolution of the Middle Ages” the adoption of the moldboard plow, which is considered the most significant attribute of plow culture. There are many reasons to believe that iron agricultural tools had already created conditions for revolution through agrarian economy by the beginning of the last quarter of the 1st mill. BC.

Another factor has been underestimated or even ignored: the change of political regimes. It is, moreover, one of the shortcomings of social typologies, such as that of Johnson and Earle⁵³. Relevant foundations have been laid by Greek and Roman thinkers from Antiquity. Thus, the theory of Anacyclosis by Polybius⁵⁴ proposed a six-phase cycle of political regimes: monarchy, tyranny, aristocracy, oligarchy, democracy, and ochlocracy within which there would be a wait for the right man who would renew the monarchy. Behind the cyclical, very mechanical aspect of the proposal lies a relevant structure that distinguishes three types of government: government of one, government of several (or minority), and government of all (or a majority). Three types might have two forms each: one just concerned with the public interest guided by reason – respectively, monarchy, aristocracy, and democracy – the other deviant, subject to the special interests and passions –respectively, tyranny, oligarchy, and ochlocracy (fig. 4). It seems that the evolution of political systems in Europe, from the 8th c. BC does not conform only to Johnson and Earle’s typology in terms of organizational complexity, but also to the typology of political regimes. In terms of simple chiefdoms, the head is a small monarch, a wren, which can already only exercise power. At the level of complex chiefdoms, a king may be more inclined to rule in a despotic, even tyrannical, manner. And this risk is even higher in archaic States. Over the course of these eight centuries, the textual sources clearly demonstrate such changes. Julius Caesar stated that the Gallic hereditary kings tended more and more to make room for judges elected among aristocrats in each archaic State, but in an extremely tense political atmosphere, until the outbreak of revolutions similar to those that overthrew and expelled the Tarquin in Rome or Codrides in Athens. These tyrannical deviant monarchies were transformed into the “aristocracy”.

The Arverni renounced their royal family in 121 BC. It is likely that the Aedui State was suspicious of monarchy even earlier; it seems that, for the Gauls, royalty was most unfavorable, and that they took the most exacting provisions against tyranny. The Aedui had accurate and skillful laws, which Polybius might have found to be as good as those in Roman law. In peacetime, the Aedui people obeyed a single magistrate, elected annually, that the Celts called vergobret. He was very like a dictator or consul of primitive times, for he had, except in title and duration, all the royal authority. He was a sovereign judge, who held the power of life and death. His power was likely to become even more dangerous, in that one would not have hesitated to attract to this sovereign position young, bold, and ambitious persons. But many precautions were taken against institutionalizing the vergobret. his magistracy lasted only one year; he could not cross borders of State; he could not appear at the head of armies, unless he was out of office: if he gave them orders, it was as a minister of war who would not fight, and his authority was as a commissioner, during campaigns, to one or more military commanders of infantry or cavalry. The Aedui shared with all Greco-Roman towns fear of a coup d’État and suspicion of tyranny⁵⁵.

Some salient points can be made in conclusion. First, and in general, it is confirmed that protohistorical societies were more complex, economically and politically, than expected and that their jagged evolution corresponds to a standard of universal history, as had plainly been seen by Stewart with his multilinear evolution and cultural ecology approaches⁵⁶. Then, during the 8th and 7th centuries BC, a progressive rupture occurred, marked by a relocation of many high-level settlements, but despite the still too rare documentation, it seems that the political integration module did not change after 800 BC. It experienced a strong growth in the 6th and early 5th c. BC, especially in the southwestern part of the north-alpine zone. The model of  princely residences” is indeed validated broadly; some princely capitals have even proven much large than expected, although they did not reach the urban stage. There is a decline in social complexity from the middle of the 5th century BC, the causes of which remain poorly understood. Of these cases, the military seat of power and the exceptionally ostentatious funeral practices of ruling dynasties, related to their power to install a woman of their family on the throne, suggest that despotic management might have caused a spike in revolts. A reprisal of the political complexity appeared from the 3rd c. BC on a more solid economic base. It led to a further growth of the integration scale among unstable federations of pagi, allowing the spread of what Caesar has called civitates, meaning, archaic States, including Northwestern Gaul, that reached the territorial dimension of principalities from the previous four centuries, while those located farther southeast and nearer to the Roman world reached significantly larger territorial dimensions. More complex in their political organization, these States were also places with great social inequalities. Caesar cited powerful Gallic aristocrats with a clientele of thousands of people, and even a hired army. Currently available funerary data, however, does not demonstrate strong social inequalities; they are significantly less ostentatious than during the “princely phenomenon” in the middle of the 1st mill. BC. However, the major agropastoral areas uncovered recently by preventive archaeology operations leave no doubt about the degree of inequality of these societies. This paradox is due to the fact that the funerary practices fail to accurately reflect not only the social hierarchy – if the monumentality and the rich funerary deposits unambiguously reveal a hierarchical society, uniformly simple graves do not necessarily indicate an egalitarian society – but also the fundamental difference between political inequality and economic inequality. It has long been assumed that the inequalities of law and resources tended to grow correspondingly in human societies until the adoption of democracy, with which the link weakens. This was a mistake. It should now be considered that at all levels of organizational complexity, societies might vary between forms of political regimes oriented toward the general interest, on one side, to the special interests of the powerful, on the other. There is a growing number of indications that human societies have always been aware of the risks of abuse and the violation of freedoms by rulers, who have always been potentially tempted by despotism and tyranny. Social evolution, as made perceptible by archeology, during the Iron Age in France shows that among the factors that combined to explain the observed changes, political regimes probably played a major role.