CC BY
64
Early State Today
Henri J. M. Claessen
Leiden University Renee R. Hagesteijn
Netherlands Organization for Scientific Research Pieter van de Velde
Leiden University
The end of this special edition of Social Evolution and History is a suitable place to go into some of the issues raised by the foregoing articles, some of which are in support of the views presented many years ago, others are more critical and suggest a different approach, and again others offer suggestions for improving or refining the ‘canonical’ texts. Indeed, based on new insights and data, several of the statements formulated thirty years ago, are in need of reformulation.
We will now give our comments to the articles in the same alphabetical order as they appear in this volume.
In his substantial contribution, after emphasizing that the state not only is a set of political institutions but ‘first and foremost’ a type of society, Dmitri Bondarenko discusses amongst others the three types of early states distinguished in The Early State (also Bondarenko 2006). In principle a state is non-kin based, while all other societies are essentially kin-oriented. States, and thus also early states, are based on territorial ties. The application of this principle, however, is not so simple. Early states are phenomena in evolution, which means that their development is gradual and, because of this, the drawing of a clear dividing line between an early state and its tribal predecessor often is difficult. In The Early State it was already pointed out that
We simply do not know when or where the decisive step - if there is justification for speaking of any decisive step - was taken in each individual case. The notion of a gradual development gives us more reason to think in terms of fairly inconspicuous processes. These took place very slowly and unnoticeably, until the anthropologist or historian is able, in the retrospect, to observe the sudden presence of specific institutions that are characteristic of the state (1978: 620-621).
Social Evolution & History, Vol. 7 No. 1, March 2008 245-265 © 2008 ‘Uchitel’ Publishing House
The corollary of this gradual development is that even though in say, the political structure the characteristics of an early state are found, in other fields, say, the kinship structures, the old behaviour can still be dominant. Or, to refer again to The Early State (1978: 621): centralizing tendencies are found often already long before the emergence of the early state, and ‘To this we must add here that many tendencies that are characteristic of pre-state phases (such as for instance, communal ownership of land, allegiance to family or clan heads) did not disappear after state formation.’ The occurrence of such uneven developments was already established by Carneiro, who termed this type of development ‘differential evolution’ (1973: 104-108). It is in these rather nebulous surroundings that the demarcation lines between early state and non-state have to be drawn (e.g., Kurtz, this volume). An example of such a ‘nebulous' situation is found in Barth's The Last Wali of Swat (1985) in which the consolidation of the early state of Swat is described. During the life time of this last Wali, the borderline between state and non-state is crossed several times in both directions, until finally in the late 1940s definitively. Even in the rather absolute thirteenth century state of France there were many corners of the territory where the understanding of being part of a large, well-organized whole, hardly was found. Montaillou (Le Roy La-durie 1975) is a good example of such a simple, kin based society within a mature state. It is the scholar who makes the choice, and Bondarenko, following his strict logical approach, draws his borderline different from the one drawn in The Early State, where emphasis was laid on the occurrence of a specific political organization and less on the continuation of pre-state social structures. In last instance the choice is a matter of selection or preference by the scholar(s) involved. We refer here to Early State Dynamics (1987: 3):
The distinctive criteria used for classification are analytical tools that are related to specific theoretical views and research premises and are not inherent in the phenomena studied... Any meaning infused into the particular categories flows from the theoretical framework behind the construction, not from the data.
The construction of the early state (for it is a construction) is based upon comparative analysis. The basis for the construction was laid in Claessen's thesis (see Claessen, this volume), in which five early states were compared in great detail. These findings found ample confirmation in the comparisons in The Early State and in the cases later added to the sample (see Bondarenko's notes 11, 12, and 15). Differences of opinion on this subject will remain, but Bondarenko certainly is right when he concludes that the labelling of a society as either a complex chiefdom or as an early state ‘does not either expand the compendium of evidence on this society or increase our possibilities to study it’. By any other name the rose will smell as sweet; we could not agree more.
Gary Feinman first places The Early State and other publications of the Project in a wider context and then presents his views on the necessity and the possibilities of expanding the research of states and state institutions over disciplinary boundaries obstructing such research thus far. In his opinion scholars deprive themselves already too long of many useful data and insights by restricting themselves to only their own (limited) field of study. Historians hardly ever consult the work of anthropologists, political scientists do not use the insights gathered by archaeologists - and so on. Such borderlines really do exist and up till recently persisted. All three authors of the present article experienced negative comments or questions by colleagues from other disciplines when they trespassed. Claessen was informed by a historian that he asked questions no historian would ever ask, and produce answers no historian was interested in. In a review of Hagesteijn's Circles of Kings (1989) the reviewer characterized the work as written by an outsider in Southeast Asian history, working too much with political anthropological models (Reynolds 1992). Van de Velde was told by a functionary of his Department of Archaeology that his work on early states could not be accepted here as scholarly - as it was not archaeology. Thresholds between disciplines can be crossed only with difficulties, and fools rush in where angels fear to thread. Yet, following Feinman, by combining forces wide vistas of knowledge and insight would be opened to us all. This perspective must be inspiring - to fools as well as to angels!
Experience in the Early State Project has taught us that the identification and invitation of capable and interested scholars is not difficult. To have them accept the concepts and definitions necessary for comparative studies, is another matter. Most of them are prepared to think and work in a comparative framework - until the moment that their own special tribe or culture is classified as just another case, then reservations become manifest. An experience every colleague working with comparative analyses will be familiar with (cf. Kobben 1967, 1973). The goals identified by Feinman, however, fully deserve the attention of the Early State community. Hagesteijn's article (this volume) is an effort to gain new insights by crossing boundaries.
Leonid Grinin presents a stimulating and refreshing discussion of a subject that was sadly neglected in earlier early state studies, namely the phenomenon of the mature state. It is true, in several places in our publications the mature state was mentioned, but nowhere was it made the subject of a thorough analysis. Grinin gives an overview of the places where the term ‘mature state’ turns up, and he suggests possible reasons why it never got the attention it deserves. In The Early State (1978: 22) the mature state was mentioned as being the phase after the early state - there had to be something into which developing early states could arrive, a phase foreshadowed by the transitional early state. A logical, but rather
poor argument (see also 1978: 633, where the transition from the early to the mature state is formulated). Perhaps our neglect of the mature state was mainly caused by our lack of knowledge; we were already happy to have brought together some convincing cases of early states, so why bother with developments that came after these? The best we did was suggesting a number of possible mature states, such as the Roman Empire, medieval France, Japan of the Shoguns, the dukedom of Burgundy, the caliphate of the Abbasids, the China ruled by Kublai Khan etc. (Claessen 1991: 184-185; also Bargatzky 1987).
Grinin, however, does more than only pointing to a neglected subject. He goes deeply into the matter, and shows that the dichotomy ‘early-mature’ is an incomplete developmental sequence. According to him there are structural differences between the cases suggested by us, and states with a capitalistic and industrial background (imperial Germany, the United States, and Russia etc.). He therefore proposes a sequence of three types, early states, developed states and mature states. In this way he places a separate category between the early and the mature state; this developed state is characterized by him as ‘the formed centralized states of the Late Antiquity, the Middle Ages, and the Early Modern Period. They are politically organized societies with distinct estate-class stratification’.
We greatly appreciate the way in which Grinin augmented and improved our work. With this classification a serious gap in the evolution of the state is closed. The only critical comment we have - had we ever thought to investigate the problem ourselves - is a terminological matter. We would have reversed the terms developed and mature because in the few places we used ‘mature’ the term was always placed immediately after the early state. Since Grinin's publications such a reversal of terms is no longer feasible.
With the introduction of the Frailty States Renee Hagesteijn adds a new dimension to the field of Early State studies (cf. Feinman, infra). She points out that a comparative study of Frailty States and Early States opens a road to a better understanding of the phenomenon, for though the Early States and the Frailty States operate on different evolutionary levels, her comparisons show that in many respects the socio-political structure of Early States was much more developed than the structure of the Frailty ones. In her opinion the very same factors that were positive in the formation of Early States, were also crucial in the development of the Frailty States. Here, however, the factors worked in a negative way: too many wars, too few economic resources, no sufficient legitimacy of the leaders, and so on. Interestingly, several of the Frailty States had a glorious past, during which the nowadays run-down polities had strong, legitimate leaders, flourishing economies, and the countries knew a peaceful existence. Though it is currently en vogue to blame all degeneration and poverty on colonialism, the findings of the Early State Project show
that many Early States had collapsed already, or were conquered by stronger polities (e.g., Imperial Rome, Imperial China), long before the coming of European colonizers. The data, presented by Hagesteijn make clear that the worst developments occurred in the years of Decolonization. The once existing indigenous socio-political structures had disappeared in the colonial time (Trouwborst, infra; Tymowski, infra) and the new foreign structures that had been introduced instead failed - in most cases inevitably so.
Nikolay Kradin discusses ‘Early State Theory and the Evolution of Pastoral Nomads’. In his contribution we discovered two highlights: suggestions concerning the theory of the early state, and his description of the pastoral nomads.
After some introductory statements Kradin proposes to revise our typology of Early States. First by eliminating the category of inchoate early states, by calling them chiefdoms. The category of transitional early states he suggests to call ‘the existing mature pre-industrial or traditional states’. According to Kradin the only form of the early state is the typical early state. He adds to this that in the Early State Project's texts chiefdoms are repeatedly presented as ‘very unstable and inclined to collapse’. These assertions ask for some comments.
As regards the characterization of chiefdoms, neither in the pages of The Early State (1978: 22), nor in those of The Study of the State (1981: 491) referred to by Kradin, expressions indicating instability or decay of chiefdoms can be found. Neither can they be found in any of the other publications of the Project. Chiefdoms were defined on the basis of Service's definition of 1975 (Service 1975: 16). The fact that early states as well as chiefdoms are vulnerable to fission was mentioned - but this is not the same as instability; nor is it the same as collapse. That there have been chiefdoms as well as early states that were unstable is true, but it is not an inherent characteristic of either of them (Tymowski, this volume): chiefdoms as well as early states are known to have existed for many centuries. A good example of such a stable polity is the Tonga Islands in Western Polynesia. Archaeologist Burley (1995, 1998) follows its history from a tenth century chiefdom up to a full blown early state in the eighteenth century. Cases of instability are found in Hagesteijn's Circles of Kings (1989), in which she analyses a number of polities in early continental Southeast Asia, which in the course of a few centuries several times reached the level of an early state, to fall back to the chiefdom level a little later, a process that continued until new legitimizing ideologies were introduced (Hagesteijn 1987, 1989, 1996).
In his analysis of Early States Kradin actually uses only characteristics we assigned to transitional early states. So he sees the appearance of written codes of laws as ‘one of the most important aspects of the early state’ (see on the written word: Khazanov 1978a: 89-90; Wormald 1977),
and he adds to this the ‘availability of juridical specialists, who consider the majority of legal questions’. Not surprisingly these features are absent in inchoate or typical early states. This approach does not seem very fruitful to us. Then he turns to the category of special functionaries, defined in The Early State (1978: 576-581) as ‘functionaries, whose governmental activities are restricted to only one aspect of government administration’. As most other kinds of functionaries, including general functionaries can also be found in chiefdoms, he makes the existence of special functionaries the decisive criterion for statehood, implying that polities where no special functionaries are found cannot be considered as states. The data in The Early State (p. 576-581) show that eighteen of the twenty-one cases knew special functionaries, the exceptions being: Iberia, Mongols and Scythia. These cases were left out of consideration in Table XVII ‘because of lack of data’ (1978: 580). According to us it is not realistic to conclude on this basis alone that such polities could not be (early) states. As we see it, the decision to characterize a polity as a state or not should not depend on one single criterion but rather on a complex of criteria (see e.g., Grinin, this volume).
In Kradin's views on the Mongol polity, his second major point, he shows himself a well-acquainted specialist in this field. In a clear way he brings together various views on the Mongols, and adds examples from polities all over the world. He makes a useful - and necessary - distinction between the political organization created by Chinggis Khan, and the one developed by his successors when the state organization expanded too much and finally fell apart in separate polities. Thirteenth century Kubilai Khan, basing himself on conquered China, ruled over a really mature state. There is no reason, however, to assume that pastoral peoples cannot be analyzed in terms of the Early State Project, which is shown convincingly by Elgin Kursat in her thesis Zur fruhen Staaten-bildung von Steppenvolkern (1994), and Anatolii Khazanov on early states among pastoral peoples (Khazanov 1978b: 425-439; 1981, 1984).
In bold strokes Donald Kurtz rightfully points out first that the activities, usually attributed to ‘the state’ in fact ‘emanate from government and not, as social scientists commonly assert, from an anthropomorphized state’ (Kurtz 2001, 2006), and second, that governments of early states -rulers, ministers, functionaries - ‘attempt to subvert real or perceived threats to their authority from local level organizations that comprise their nations by entrenching their authority vertically into the nations' communities and institutions’. After clarifying his concepts, he illustrates his views with three cases of early states, the African Zande, the South American Incas, and the Aztecs of Mexico, representing respectively the inchoate, the typical and the transitional type of the Early State.
Regarding his first thesis, namely that governmental activities always emanate from people, we cannot but agree. In many cases, however, it is
not easy to identify the individuals responsible for specific activities, especially when decisions are based on discussions in councils, or on advice by ‘anonymous’ functionaries. This last category has been identified in The Early State (1978: 581-584) and to it belong members of the rulers' family, courtiers, priests, military leaders, and royal consorts (and concubines; think of Madame de Pompadour c.s. [Hanken 1996]). In last instance however, it is always the ruler who issues the commands (Kurtz 2006: 100). Especially in early states where the position of the ruler is strong, the commands, orders, laws and directives are issued by the king as can be demonstrated for the courts of Buganda (Speke 1863), Dahomey (Burton 1864) and the Incas (Cobo 1653/1983).
His second thesis, that governments always try to eliminate the role and influence of local (and, we would add, regional) organizations is the main thrust of the article. The general idea behind this policy is that central governments consider local or regional groups, regardless of whether they are based upon kinship, on political or religious considerations, or on feelings of ethnicity, as a threat to their existence. In this connection the legitimacy of the ruler is of great importance. Problematic can be when there are people who, because of their descent have the same rank as the ruler and have similar claims to legitimacy. This not only holds for close relatives, but also heads of mighty families with pedigrees as long as the ruler's are dangerous competitors. Under these circumstances the sapa Inca Huayna Capac once tried to create a new religious legitimation, by announcing that the Sun was only a hard working servant of a much higher God, with whom he claimed a relation. This was a conscious effort to redefine his sacred status (Toland 1987: 151) in order to cut out eventual competitors. There is no reason to think that Huayna Capac created this strategy to impress his humble subjects; it was rather a move on a higher political level.
The cases Kurtz describes are clear examples of early state governments trying to maintain their ascendancy over local and regional groups. Sometimes such efforts consisted of military actions, but more effective, according to Kurtz, were the less violent strategies found in the Realm of the Incas. In the Aztec state local and regional groups became split by economic differences; social and economic classes slowly formed and horizontal connections came in the place of the earlier ethnic or political interests and loyalties.
Struggles with regional lords, local groups or towns are found in virtually all early states. In the Polynesian kingdom of Tonga the heads of noble families competed with the tui Tonga for status and power - a situation noted and described by James Cook in 1777 (Cook 1967: 174; cf. Burley 1995). In medieval France the Carolingian rulers, from Charlemagne up to Charles the Bald, struggled hard to keep the regional lords -dukes, counts and bishops - in line (McKitterick 1983; Nelson 1992) and the same can be said for the Capetian rulers, who not only had problems
with (the very same) dukes, counts and bishops, but even more so with the kings of England, who - by a lucky marriage - had become lords of the western half of France (Fawtier 1974, especially chapters 7, 8 and 9). Only by fully using their rights as feudal overlord, and after several military victories, the French kings succeeded to break the English influence (Hallam 1980).
While Donald Kurtz presents an analysis on a meta level, Jean-Claude Muller describes in great ethnographic detail the fascinating rituals that the West African chiefs of the Dii and the Rukuba have to endure before they are acknowledged as such. These rituals are not simply continuations of the rites de passage imposed on ordinary people. The chiefly rituals are different, and only candidates for chiefship have to undergo the prolonged and painful rituals. These rituals centre on circumcision, connecting the chief with fertility. Among the Dii the chiefly ritual was a kind of re-circumcision, which made him ‘the sole complete initiate of the chiefdom which, under his government, can enjoy a true civilized social life’. By giving a big feast to which all the neighbouring village chiefs are invited the new chief demonstrates ‘that he can provide abundance’. The connection of a chief (or king) with productivity, fertility, and good harvests is a general feature of African chiefship and the literature on traditional African chiefs and kings abounds with references to this phenomenon (Claessen and Oosten 1996: 3-4). The form in which this connection was expressed could vary (e.g., Miller 1976: 46ff., 62, 165; van Binsbergen 1979).
In the case of the Rukuba, Muller emphasizes that the chief after his inauguration becomes a person outside and above his society. He has become a scapegoat ‘who is expelled in cases of serious trouble’. A scapegoat has a dual status, good and bad at the same time and the person chosen to become chief is always someone who is already in an ambiguous position because of his descent. A crucial part of his inauguration consists of transgressions of the norms of the society, the most important of which is his eating the flesh of a new born baby. This completes his transmogrification into a scapegoat, who can be expelled when things go wrong. In his publications on the Rukuba, especially in his Le Roi Bouc Emissaire (Muller 1980) he has elaborated these views.
In several early states in Africa the concept of a scapegoat king is found, albeit in slightly different forms (Simonse 1992). An old king, it is generally believed, is no longer capable to guarantee fertility and the well-being of his people, so he has to die and be replaced by a young and vigorous man. Among the Shilluk, a Nilotic people, some near relatives or his consorts kill the king when he is very ill (Evans-Pritchard 1948 denies this; Simonse 1992: 423-424 confirms the tradition). Among the Jukun, a Sudanese people, the king is obliged to commit suicide when old and weak (Meek 1931); in Interlacustrine Bunyoro, the ruler when seri-
ously ill is killed by his consorts (Beattie 1971, but he has his doubts); Nahimana (1981: 123) says that the ruler of Rwanda (also an Interlacus-trine Early State) must be killed when he no longer is capable to bring rain. In other cases rituals are found to prolong the life of the ruler by killing a substitute. In West African Dahomey, once a year a boy, dressed in the robes of the akhosu, is killed (Bertho 1946), and in Buganda the Nankere ritual was performed, during which a son of the Nankere priest, as a substitute of the king, was killed (Roscoe 1911: 210).
Outside Africa the phenomenon of a scapegoat king does not play a role, though Frazer in his Golden Bough presents wonderful stories some of which could be interpreted in this direction, but the credibility of his examples is limited. The connection between rulers and fertility, however, is found all over the world.
The late Albert Trouwborst addresses the problem of the end of the early state. As an example he analyzes how the Early State of Burundi in the Interlacustrine Region came to an end when colonizing European powers took over. Burundi was for some time part of German East Africa, became after the First World War a mandate of the League of Nations, and was since 1945 a Trust Territory of the United Nations under the supervision of Belgium. Not surprisingly after so many years of colonial rule the old political structure had collapsed completely. The changes began shortly after the German occupation, when the competing factions were suppressed and the mwami, the king, became considered as the sole indigenous ruler. Under the Belgian rule the situation changed, for they introduced the policy of ‘indirect rule’, which led to more influence of the local and regional chiefs. The king ‘had become a puppet in the hands of the Belgian rulers who also strictly controlled the chiefs who had become civil servants’. It will be clear that Burundi no longer was an ‘independent’ socio-political organization - or early state, but a strictly controlled province of Belgium.
Apart from external (colonial) influences, more can be said on the ‘end’ of the Early State. In his explanation of the decline of the early state of Burundi, Trouwborst does not apply the nowadays so popular theory of ecological deterioration, popularized by authors as Diamond (2004) but uses political change as an explanation. It is true, some early states (and chiefdoms) did collapse because of exhaustion of the soil and additional ecological disasters. Well-known cases are Easter Island and the Marquesas Islands in Polynesia (Claessen and van Bakel 2006), several Maya polities (Culbert 1988) and some more examples mentioned by Diamond. Yet, for a general explanation the ecological theory is neither sufficient nor necessary. Many more early (and developed) states fell because of colonial conquests: Dahomey, Buganda, Asante, and Congo in Africa, Aztecs, and Incas in the Americas (though more factors played a role here, see Conrad and Demarest 1984). Another frequent factor in the end
of early states was internal difficulties. The early state of Ancient Egypt collapsed for similar reasons several times (Morris 2006), and the same holds for the ancient Indonesian kingdoms of Srivijaya, Mataram, and Majapahit (cf. Wisseman Christie 1995), and Southeast Asian Angkor (Hagesteijn 1987).
In an elaborate article Michal Tymowski compares ‘the processes of limiting the operating scope of tribal organizations and state emerging processes in medieval Europe and pre-colonial Africa’. The first part of his text has some resemblance to aspects of Kurtz's article (this volume); his approach, however, is different. By his combining anthropology and history, Tymowski is nearer to Feinman's ideas (Feinman, this volume). He begins with the developments in Europe in the early middle ages. In all cases the tribal organizations were taken over or destroyed by more powerful states. The influence of Christianity, which legitimized the new political order, was an additional factor in these processes. For quite some time this model of development was believed to be universal. Tymowski's analysis of the African data shows that different transformations were possible.
Some African regions knew more state formation than others, (e.g., West and Central Africa south of the Sahara, the Nile valley and delta, the interlacustrine region), but even here tribal organizations continued to exist. Moreover, many states in Africa disintegrated and collapsed due to internal processes. The disintegrating states then fell apart in a number of tribes.
Apart from the coexistence of early states and tribal organizations, Tymowski stresses that in many African states (and not only in Africa) tribes continued their existence within the framework of states. The differences with Europe are great. Clovis, the first ruler of the Franks (about 480-510) spent much energy in eliminating tribal chiefs and kings and taking over their rights, a policy continued by Carolingians and Capetians as well (see above). ‘Elimination of the tribes made the existence of the states irreversible. Hence most of the states formed in Europe in the Middle Ages have survived until this very day’. In African early states tribal organizations continued to exist. The tops of these organizations were often used as lower management levels, which therefore did not contribute to the stability of these states. A similar policy was followed by some colonizers too (Trouwborst, this volume).
Some of the factors mentioned by Tymowski to explain the differences between the European and African developments were already noted by him in earlier publications (esp. 1987), others are new. So was the diversity in languages, customs, law systems etc. in Africa much greater than in Europe, where neighbouring tribes often had similar cultures. This made the unification in European states easier. Moreover the productivity of European agriculture was higher than in Africa, which
made the rulers in Europe relatively rich. We are not so sure that this difference already played a role in the early middle ages of Western Europe. Regarding the Frankish kingdom, several historians stress the poverty, the low production, and the difficulties of transport in the times of Merovingians and Carolingians - a situation which only gradually improved when population growth, the development of trade and the emergence of towns began to play a role under the Capetians (Duby 1968; Hallam 1980). Only then the central government was able to strengthen its position. The same holds for the role of writing in ‘communication, management and transmission of culture’. However, though reading and writing are crucial in the exercise of government, it should be borne in mind that in the early Middle Ages only few people were literate, and they were nearly always monks or priests (cf. Wormald 1977). And yet, European rulers were able not only to limit the role of the tribes, but, what is more, to make the centre indispensable for the regional and local leaders, by providing services that none of the lower order centres could equal (Bargatzky 1987). While the central authority thus became more complex, the component parts of the state became simpler and were less able to act independently any longer.
That Christianity afforded medieval rulers a strong legitimation is certainly true, but the Church exerted also much influence on governmental activities. For example, the Christian urge for conversion provided an excuse for Charlemagne's campaigns against the Saxons, but also for actions to suppress the Jews and the Kathars (the Albigensian crusade); and for the Crusades against the Mohammedan peoples. Though in Africa some general aspects of kingship can be identified - sacrality, influence on fertility, etc (The Early State 1978: 555-559) - these never were so developed that a general ideology could rise here.
Edward van der Vliet opens his article on Early Greece with a number of questions, all of them related with the problem of whether the Greek poleis were early states or not, and if not, why not. He also asks why early Greece was absent in The Early State. The obvious answer to the last question is that the editors of that volume had not found a scholar willing to tackle the Greek problem, for Greece was then - and still is - a problem as appears from the many studies devoted to a clarification of its precise political character (van der Vliet 2005; Berent 2006).
Already in the beginning of his article van der Vliet points to a crucial difference between Early States and the Greek poleis. ‘The focus of the Early State is its centre at the top, where we find a ruler. In the polis the centre is empty, in the midst of a circle of citizens, who are equals’. It is the citizens around whom the Greek polis turns, but only free men qualify as such. Women, children, slaves, visitors, foreigners all are excluded from this blessed status and political influence (though perhaps some haetairae could exert a secret influence; van Dolen 1997). This limits the number of citizens to no more than twenty percent of the inhabitants.
Then also among the male citizens a distinction was made between a small minority of wealthy and a large majority of middle range or poor men. In practice it was the wealthy few, if only because they had the time and the money to spend, who dominated political life.
When discussing the evolution of the Greek poleis, van der Vliet points out that difficulties begin already before the poleis emerged, for there is - in spite of Homer - not much known about the political formations preceding them.
With this background, and full of questions and critical comments, van der Vliet entered the Dutch Early State Society of which he became a prominent member, contributing papers, and organizing meetings. But, in the background of all these activities was lurking the critical question ‘in how far the Early State concept can be applied to the early Greek polis’. To complicate this question but few Greek poleis could eventually qualify as an early state, for most of them were too small and not sufficiently developed; they were mere villages. The larger ones, however, ‘were in effect states’. These Greek states he defines as organizations with the ‘presence of a political (administrative) organism that disposes of the structurally legitimated power which enables it, if necessary, to use an ultimate monopoly of force to coerce the people into obedience’. In early states kings (rulers) symbolize the authority of the state. In the Greek case, however, one finds instead a group of citizens representing the state. In large towns as Athens, government was in fact a matter of delegation, and boards and councils were entrusted with aspects of government Theoretically all citizens could partake in meetings of the Council - but most were seldom able to attend to them.
It is here that we start wondering about the evolution of early Greek society. Why had they become so different? Agreeing that there is not much known about the forefathers of the Greeks, one can safely say, for instance, that in Mycenae (1400 B.C.) powerful men have ruled, strong enough to build an impressive town. Whether they were chiefs, big men, or kings is not relevant; evidently they were powerful rulers. Moreover, Greece was surrounded by powerful, centralized states: Persia, Babylonia, Assyria, and Egypt; and Mycenae - be it on a smaller scale - falls within this pattern (cf. De Blois and van der Spek 1994: 58-59). Now the question is: why and how did the Greek of later centuries become so different? It must be assumed that the concept of the polis with its concomitant form of democratic organization emerged only there and not in other regions. The thought of a connection with the rather specific physical milieu comes to the mind. A rugged country and an orientation towards sea and trade may have caused the splitting up of Greek society in numerous separate and independent groups, some of which were bigger than others. Yet, they all were characterized by ‘a single ideological model’, to quote Muller's felicitous phrase (Muller 1981) in which government was the task of citizens and not of supreme rulers. If such were the case, a com-
parison with other democratic commercial centres such as medieval Venice, or capitalist Amsterdam might be productive.
Only under certain circumstances the difference between the Greeks and all other peoples fell away, namely when tyrants came to power. In the eyes of the Greeks, however, such rulers were a strange or different phenomenon. In terms of an Early State approach, these tyrants are most promising.
Assuming that for the evolution of early state structures a certain level of development is necessary (Bondarenko, this volume, and above) the Greek data are rather difficult to interpret. There are insufficient archaeological data to enlighten the past or to support evolutionary hypotheses. This limitation holds especially where the ideological aspects of Greek society are concerned. The development of legitimation, laws, justice, and democracy are difficult to establish with the limited archaeological data available. With the help of the Complex Interaction Model (Claessen, this volume) van der Vliet tries with some success to overcome these difficulties. By connecting developments in the format of the society with economic and ideological aspects some of the lines leading to the formation of the poleis have become a little clearer.
Jianping Yi poses the question whether early Chinese rulers should be correctly considered as autocratic. He thinks that this was not the case and that on the contrary in many early states in China, especially in those that were relatively small, non-autocratic forms of government were dominant. He opposes to Marxist authors, from Karl Marx to Karl Wittfogel (1957, 1969), who consider historical Asian states as despotic (for a critical Marxist view: Hindess and Hirst 1975: 207-220). Also many Chinese historians believe, according to Yi, that ‘the political organizations in the whole ancient history (ca. the 21st century - ca. the 16th century B.C.) were despotic’. Others however, think that before the state came into being, political organizations were democratic. This is, as Yi explains, more a matter of theory, than of research, for they only apply the concept of the ‘military democracy’ developed by Morgan on the basis of his study of the Iroquois (1877). The latter combined the military attitude of a society with its administration by elected and removable chiefs. Khazanov (1974) demonstrates that this political construction, as far as it ever existed, was not a general evolutionary phenomenon and that the concept itself is a contradiction in terminus.
Apart from the dubious hypotheses based on the ‘military democracy’, Yi states that there are more and better indications for the occurrence of non-autocratic forms of government. He points to the mode of succession in the period of the (mainly legendary) Wudi (the 26th to the 22nd century B.C.), where the ruler handed over his position to a capable person. Only later a succession from father to son developed. The historian Wu Yujiang (1993, quoted by Yi) elaborates this mode of succession
and sketches how the ruler chose assistants, who in their turn were to choose assistants, to ‘unify all the thoughts of the people’ enabling the ruler to reign in accordance with the desires of his people. Though the historicity of such stories is doubtful, Yi thinks that they may reflect traces of past practices. A problem is the scarcity of historical data on the Xia and the Shang dynasties, who ruled over the central part of China from the 21st to the 11th century B.C. Yet, Yi thinks it possible that ‘at least the political organization of the Shang could not be completely or simply characterized by the concept of autocracy’. A traditional history tells that a Shang ruler consulted diviners, and accepted the views of the majority of them, which means that neither the views of the ruler nor those of the principal diviner were dominant. To this non-autocratic behaviour can be added that there were found also two councils, one of the people, and one of the elders (senate). They played a role in the making of decisions in state matters such as succession or war. This all suggests that in the early periods of the state there were rules and regulations which had to be heeded also by the ruler. Further Yi refers to Mengzi (Mencius), a philosopher in the 4th century B.C., who was consulted several times by the ruler of the early state of the Qi how to behave towards his ministers, many of whom were members of his family. Theoretically they had the power (or obligation) to dispose of him after misbehaviour. Tradition holds that several times rulers were sent into exile by their ministers, and some even seem to have been put to death. In each case it seems that bad behaviour towards the people lay at the basis of such a punishment. In later years, when bigger states began to suppress smaller ones, the non-autocratic principles lost influence.
Several of the non-autocratic customs Yi describes, such as the role of councils, the appointment of rulers by notables, and the influence of philosophers and diviners, can be found also elsewhere, for example, among the Carolingian rulers. If we are prepared to replace the philosophers and diviners by abbots and bishops (‘functional equivalents’, see The Early State 1978: 536), there is plenty of evidence for their influence on Charlemagne, Louis the Pious and Charles the Bald, while also the role of councils under these kings can be pointed to (McKitterick 1983; Nelson 1992). Clovis was accepted as ruler by his nobles by acclamation, and when Pippin III wanted to become king instead of the then ruling Meroving, he needed the agreement of the Pope, and the acceptance by his nobles (Wood 1994). This is not so strange. There are no rulers who govern alone; they always need supporters: councillors, ministers, military, henchmen, and civil servants. It is the variation in the influence they have that makes a ruler autocratic or non-autocratic.
FINAL REMARKS
Since the appearance of The Early State in 1978, several attempts have already been made to amend some of its shortcomings and add new in-
sights. For a good understanding of the Early State Project these additions are essential; they present a number of corrections and a broader perspective. Apart from several publications in Dutch (Claessen 2000), we refer to The Study of the State published in 1981 (edited by Claessen and Skal-mk), in which several of the statements made in The Early State, have already been amended or re-formulated (e.g., the contents of the ideological and the economic factor; more attention to the role of kinship and family, the reduction of the role of war in the formation of states, etc.). In 1985 Development and Decline was published (edited by Claessen, van de Velde, and Smith). In this volume problems of development and decline of socio-political organizations were discussed, and a model of socio-political evolution, the Complex Interaction Model, was suggested. This model was reiterated in Early State Dynamics (edited by Claessen and van de Velde, 1987), containing several articles (e.g., Bargatzky, Shifferd, Tymowski, van der Vliet, Toland, and Hagesteijn) which were influential in many of the later publications. While in Early State Dynamics emphasis was laid on the functioning of early states, the dynamics of government, and aspects of decision-making, in the next volume, Early State Economics (edited by Claessen and van de Velde, 1991) the political economy of the Early State was the central issue. In it the generation of a sufficient income, the way in which state finances were spent, and the relation between the economy and early states were discussed. After Early State Economics it was deemed useful to complete the ‘series’ with a volume on ideology. A compilation of articles on Ideology and the Formation of Early States appeared in 1996, edited by Claessen and Oos-ten. In it the relation between ideology - often religiously based - on the one hand and government and politics on the other was analyzed. Especially legitimacy and legitimation, crucial for the functioning of government, appeared to be narrowly connected with the ideological background. It also came to the fore that the emergence and development of early states was only possible when a suitable ideology was available. Shortly before, in 1994, a volume was published on changing cultural identities in Early States, Pivot Politics (edited by van Bakel, Hagesteijn and van de Velde). These remarks show that the study of early states is a dynamic undertaking.
This overview seems also the place to mention with gratitude the intellectual debt we owe to the works of scholars as varied as Engels, Fortes, Evans-Pritchard, Sahlins, Service, Carneiro, Weber, White and numerous others.
In addition, and from a different angle, our Russian colleagues introduced the concepts of ‘alternatives’ and ‘analogues’ of the early state (summarized in Grinin et al. 2004, and Grinin et al. 2006). These sociopolitical types are according to Grinin ‘complex stateless systems’ which ‘often coped with problems comparable with ones encountered by states,
they are quite comparable with early states by the range of their functions and level of their structural complexity as well as causes and prerequisites for their formation. So it is incorrect to consider them as pre-state structures.’ (Grinin et al. 2004: i). The introduction of these concepts makes it necessary that a distinction is made between the alternatives and analogues on the one hand, and early states on the other. From 2006 onward Claessen and Bondarenko, in the course of a long e-mail correspondence looked for the distinction. It was concluded finally that only by applying ideological criteria a demarcation could be formulated. The other criteria, often used to distinguish states from non-states, such as modes of production, number of people, and a specific territory could apply to all these types (cf. Claessen 2002). Only when a certain society has or accepts an ideology which explains and justifies a hierarchical administrative organization and its concomitant socio-political inequality, one can conclude to the existence of an early state - provided that the other criteria (number of people, economic structure, and territory) are also found here. These considerations led to the introduction of some changes in the definition of the early state, which should read now:
An early state is an independent, centralized three-tier (national, regional, local level) socio-political organization for the regulation of social relations in a complex stratified society, divided into at least two basic strata or emergent social classes - viz. the rulers and the ruled - whose relations are characterized by political domination by the former and the obligation to pay taxes of the latter, legitimated by a common ideology of which reciprocity is the basic principle.
NOTE
Walter de Gruyter of Berlin, successor publisher to Mouton's still has some copies of The Early State volume (ISBN 90-2797904-9) available.
REFERENCES
Bargatzky, T.
1987. Upward Evolution, Suprasystem Dominance, and the Mature State. In Claessen and van de Velde 1987: 24-38.
Barth, F.
1985. The Last Wali of Swat. An Autobiography as Told to Frederik Barth. Oslo: Norwegian University Press.
Beattie, J.
1971. The Nyoro State. Oxford: Clarendon.
Berent, M.
2006. The Stateless Polis: A Reply to Critics. Social Evolution & History 5(1): 141-163.
Bertho, J.
1946. Les indigenes du Dahomey et le monde invisible [The Natives of Dahomey and the Invisible World]. Grands Lacs. Revue generate des Missions d'Afrique 61: 60-62.
Binsbergen, W. M. J. van
1979. Religious Change in Zambia. Exploratory Studies. Ph.D. thesis. Amsterdam: Free University.
Blois, L., de, and van der Spek, R. J.
1994. Einfuhrung in die Alte Welt [Introductions to the Old World]. Stuttgart: Steiner.
Bondarenko, D. M.
2006. Homoarchy. A Principle of Culture's Organization. The 13th - 19th Centuries Benin Kingdom as a Non-State Supercomplex Society. Russian Academy of Sciences. Moscow: KomKniga.
Burley, D. V.
1995. Mata'uvave and the 15th Century Ha'apai: Narrative Accounts and Historical Landscapes in the Interpretation of Classical Tongan History. The Journal of Pacific History 30: 154-172.
1998. Tongan Archaeology and the Tongan past, 2850-150 B.P. Journal of World Prehistory 12: 337-392.
Burton, R. F.
1864. A Mission to Glele, King of Dahomey. With Notes on the So-Called ‘Amazones’, the Human Sacrifices, the Present State of the Slave Trade, and the Negro's Place in Nature. 2 vols. London: Tinsley.
Carneiro, R. L.
1973. The Four Faces of Evolution. In Honigman, J. J. (ed.), Handbook of Social and Cultural Anthropology (pp. 89-110). Chicago: Rand McNally.
Claessen, H. J. M.
1991. Verdwenen koninkrijken en verloren beschavingen [Disappeared Kingdoms and Lost Civilizations]. Assen: van Gorcum.
2000. Structural Change. Evolution and Evolutionism in Cultural Anthropology. Leiden: CNWS Press.
2002. Was the State Inevitable? Social Evolution & History 1: 101-117.
Claessen, H. J. M., and van Bakel, M. A.
2006. Theme and Variations. The Development of Differences in Polynesian Socio-Political Organizations. Bijdragen tot de Taal-, Land- en Volkenkunde 162: 218-268.
Claessen, H. J. M., and Oosten, J. G. (eds.)
1996. Ideology and the Formation of Early States. Leiden: Brill.
Claessen, H. J. M., and Skalnik, P. (eds.)
1978. The Early State. The Hague: Mouton.
1981. The Study of the State. The Hague: Mouton.
Claessen, H. J. M., and van de Velde, P. (eds.)
1987. Early State Dynamics. Leiden: Brill.
1991. Early State Economics. New Brunswick: Transaction.
Claessen, H. J. M., van de Velde, P., and Smith, M. E. (eds.)
1985. Development and Decline; the Evolution of Sociopolitical Organization. South Hadley: Bergin and Garvey.
Cobo, B.
1653/1983. History of the Inca Empire. An Account of the Indian's Customs and their Origin, together with a Treatise on Inca Legends, History and SocialInstitutions (ed. by R. Hamilton). Austin: University of Texas Press.
Conrad, G. W., and Demarest, A. A.
1984. Religion and Empire. The Dynamics of Aztec and Inca Expansionism. Cambridge: Cambridge University Press.
Cook, J.
1967. The Voyage of the Resolution and Discovery, 1776-1780. (ed. by J. C. Beaglehole) Cambridge: Cambridge University Press. [Hakluyt Society, Extra Series 34].
Culbert, T. P.
1988. The Collapse of Classic Maya Civilization. In Yoffee, N., and Cowgill, G. L. (eds.), The Collapse of Ancient States and Civilizations (pp. 69-101). Tucson: The University of Arizona Press.
Diamond, J.
2004. Collapse. Los Angeles: North Point Press.
Dolen, H. L. van
1997. Hetairen: vrouwen voor het plezier [Hetaeras: Women for Pleasure]. Hermeneus 60: 108-117.
Duby, G.
1968. Rural Economy and Country Life in the Medieval West. London: Edward Arnold.
Evans-Pritchard, E. E.
1948. The Divine Kingship of the Shilluk of the Nilotic Sudan. Cambridge: Cambridge University Press (Frazer Lecture).
Fawtier, R.
1974. The Capetian Kings of France. Monarchy and Nation 987-1328. London: Macmillan (11th edition).
Frazer, J.
1957. The Golden Bough. London: Macmillan (abridged edition in 2 vols).
Grinin, L. E., Carneiro, R. L., Bondarenko, D. M., Kradin, N. N., and Korotayev, A. V. (eds.)
2004. The Early State, its Alternatives and Analogues. Volgograd: Uchitel Publishers.
Grinin, L. E., Bondarenko, D. M., Kradin, N. N., and Korotayev, A. V. (eds.)
2006. Rannee gosudarstvo, ego al'ternativy i analogi. Volgograd: Uchitel Publishers. (Russian edition of Grinin et al, 2004, with some different articles).
Hagesteijn, R. R.
1987. The Angkor State: Rise, Fall and in between. In Claessen and van de Velde 1987: 154-169.
1989. Circles of Kings. Political Dynamics in Early Continental Southeast Asia. Dordrecht: Foris (Verhandelingen van het Koninklijk Instituut voor Taal-, Land- en Volkenkunde 138).
1996. Lack of Limits - Cultural Aspects of State Formation in Early Continental Southeast Asia. In Claessen and Oosten 1996: 187-204.
Hallam, E. M.
1980. Capetian France 987-1328. London: Longman.
Hanken, C.
1996. Gekust door de koning; Over het leven van koninklijke maitresses [Kissed by the King; the Life of Royal Mistresses]. Amsterdam: Meulenhoff.
Hindess, B., and Hirst, P. Q.
1975. Pre-Capitalist Modes of Production. London: Routledge and Ke-gan Paul.
Khazanov, A. M.
1974. ‘Military Democracy’ and the Epoch of Class Formation. In Bromley, Y. (ed.), Soviet Ethnology and Anthropology Today (pp. 133-146). The Hague: Mouton.
1978a. Some Theoretical Problems of the Study of the Early State. In Claessen and Skalnik 1978: 77-92.
1978b. The Early State among the Scythians. In Claessen and Skalnik 1978: 425-440.
1981. The Early State among the Eurasian Nomads. In Claessen and Skalnik 1981: 155-176.
1984. Nomads and the Outside World. Cambridge: Cambridge University Press.
Kobben, A. J. F.
1967. Why Exceptions? The Logic of Cross-Cultural Analysis. Current Anthropology 8: 3-34.
1973. Comparativists and Non-Comparativists in Anthropology. In Naroll, R., and Cohen, R. (eds.), A Handbook of Method in Cultural Anthropology (pp. 581-596). New York: Columbia University Press.
Kursat-Ahlers, E.
1994. Zur fruhen Staatenbildung von Steppenvolkern (On Early State Formation among Steppe Peoples). Berlin: Duncker and Humblot [Sozialwis-senschaftlichen Schriften, Heft 28].
Kurtz, D. V.
2001. Political Anthropology. Paradigms and Power. Boulder: Westview Press.
2006. Political Power and Government. Negating the Anthropomorphized State. Social Evolution & History 5(2): 91-111.
Le Roy Ladurie, E.
1975. Montaillou, village Occitan de 1294 a 1324 (Montaillou, an Occitan Village between 1294 and 1324). Paris: Gallimard.
McKitterick, R.
1983. The Frankish Kingdoms under the Carolingians, 751-987. London: Longman.
Meek, C. K.
1931. A Sudanese Kingdom. An Ethnographic Study of the Jukun-Speaking Peoples of Nigeria. London: Kegan Paul, Trench, Trubner.
Miller, J.
1976. Kings and Kinsmen. Early Mbundu States in Angola. Oxford: Clarendon.
Morgan, L. H.
1877. Ancient Society, or Researches in the Lines of Human Progress from Savagery through Barbarism to Civilization. Cleveland: World Publishing Company (ed. by E. B. Leacock).
Morris, E.
2006. ‘Lo, Nobles Lament the Poor Rejoice’ - State Formation in the Wake of Social Flux. In Schwartz, G. M., and Nichols, J. D. (eds.), After Collapse. The Regeneration of Complex Societies (pp. 58-71). Tucson: The University of Arizona Press.
Muller, J.-C.
1980. Le roi bouc emissaire. Pouvoir et rituel chez les Rukuba du Nigeria Central (The Scapegoat King. Power and Ritual among the Rukuba of Nigeria). Quebec: Fleury.
1981. Divine Kingship in Chiefdoms and States, a Single Ideological Model. In Claessen and Skalnik 1981: 239-250.
Nahimana, F.
1981. Les principautes hutu du Rwanda septentrional (The Hutu Principalities of Northern Rwanda). In La civilization ancienne des peuples des Grands Lacs. Colloque de Bujumbura (pp. 115-137). Paris: Karthala.
Nelson, J. L.
1992. Charles the Bald. London: Longman.
Reynolds, C. J.
1992. Authenticating Southeast Asia in the Absence of Colonialism: Burma. Asian Studies Review 15: 141-151.
Roscoe, J.
1911. The Baganda. An Account of their Native Customs and Beliefs. 2nd edition. London: Cass.
Service, E. R.
1975. Origins of the State and Civilization. New York: Norton.
Simonse, S.
1992. Kings of Disaster. Dualism, Centralism and the Scapegoat King in Southeastern Sudan. Leiden: Brill.
Speke, J. H.
1863. Journal of the Discovery of the Source of the Nile. Edinburgh: Blackwood and Sons.
Toland, J. D.
1987. Discrepancies and Dissolution. Breakdown of the Early Inca State. In Claessen and van de Velde 1987: 138-153.
Tymowski, M.
1987. The Early State and after in Precolonial West Sudan. In Claessen and van de Velde 1987: 54-69.
Van Bakel, M. A., Hagesteijn, R. R., and van de Velde, P. (eds.)
1994. Pivot Politics. Changing Cultural Identities in Early State Formation Processes. Amsterdam: Het Spinhuis.
Van der Vliet, E.
2005. Polis. The Problem of Statehood. Social Evolution & History 4(2): 120-150.
Wisseman Christie, J.
1995. State Formation in Early Maritime Southeast Asia. A Consideration of the Theories and the Data. Bijdragen tot de Taal-, Land- en Volkenkunde 151: 235-288.
Wittfogel, K. A.
1957. Oriental Despotism. A Comparative Study of Total Power. New Haven: Yale University Press.
1969. Results and Problems of the Study of Oriental Despotism. The Journal of Asian Studies 38: 357-365.
Wood, I.
1994. The Merovingian Kingdoms, 450-751. London: Longman.
Wormald, P.
1977 Lex Scripta and Verbum Regis: Legislation and Germanic Kingship, from Euric to Cnut. In Sawyer, P. H. and Wood, I. N. (eds.), Early Medieval Kingship (pp. 159-180). Leeds: School of History, University of Leeds.
