Social Boundary Networks and the Vertical Entrenchment of Government Authority in Early State Formations

Social Boundary Networks and the Vertical Entrenchment  of Government Authority in Early State Formations Social Boundary Networks and the Vertical Entrenchment  of Government Authority in Early State Formations
Author: Kurtz, Donald V.
Journal: Social Evolution & History. Volume 7, Number 1 / March 2008


One purpose of this paper is to use a reconceptualization of the idea of the state that I have developed elsewhere (Kurtz 1993, 2001, 2006) to explain accurately the location of the politics and practices social scientists so commonly attribute to the state. I contend that these practices emanate from government and not, as social scientists commonly assert, from an anthropomorphized state. The problem with the idea of the state as an anthropomorphized political agent is that it diminishes analytic insight because it allows scholars to gloss over government practices by simply attributing their outcomes to the state.

Another purpose of the paper is to use these ideas to explore the hypothesis that governments of early state formations 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 (Cohen 1969a). This may be accomplished in different ways. Governments may eliminate the threatening organizations, co-opt them, or transfer the loyalty and allegiance of those committed to them to the state's government. I will rely on data from three early states to demonstrate this hypothesis: the Zande of Africa, Inca of South America, and Aztecs of Mexico1. Before exploring the hypothesis I will consider the interlocking ideas – state, government, and nation – that are important to my analysis since they differ some from other ideas on these topics.

State, Government, Nation

I think of a state2 as a hierarchical structure of abstract offices at the apex of which a single office exists that is vested with specific powers to command the nation's military forces, execute the government's laws, and manage its revenues (Kurtz 1993, 2001, 2006). Beneath this apical office a structure of lower offices3 exists that is organized and empowered to regulate and control political, economic, social and other concerns that are relevant to the time and place in which the state exists. A state conceived as an abstract structure of offices cannot act as a political agent. That role is executed by the government of the state. But the state may, and often does, serve as a symbol and metaphorical referent for the nation's citizens and government.

The problem with the approach social scientist bring to the state is that too often they do not separate it from the idea of government (Easton 1953; Smith 1960; Abrams 1988; Kurtz 2006)4. As a structure government is comprised of the incumbents that occupy the offices that constitute a state. In practice a government represents an organization of incumbents that occupy state offices and engage in processes concerned with managing the public affairs of a nation through administration and politics. Administration is concerned with how the incumbents of state offices conduct public business and coordinate political activities. Politics is concerned with how the incumbents of those offices, individuals, especially leaders, and collectivities – councils, committees, senates, parliaments and the like (however they might be designated culturally) – compete over access to material and ideological resources of power to pursue public and personal goals (Smith 1960; Kurtz 2001).

A nation may be defined in different ways. One common definition refers to the populations and communities which inhabit a more or less firmly demarcated territory over which the government of the state exerts authority. The population of a nation thus conceived either may be relatively homogeneous or vary considerably with reference to its social structures, languages, cultures, organizations of relations, specialized occupations, religions, and the like (Cohen 1969a).

When considering the vertical entrenchment of government authority, another way to think about a nation may be more rewarding theoretically. In this context a nation may be conceived as a maximal network of relationships which may be either firmly or loosely bounded. The idea that social relations represent socially bounded networks is not new. Murdock suggested that a socially bounded network, such as a lineage, is ‘a structured system of relationships in which individuals are bound to one another by complex and ramifying ties’ (1949: 91–92; also see Engels 1942; Gluckman 1955; Fallers 1965; Cohen 1969b). These networks may be associated with any sphere of social activity in which people engage, such as specialized economic production units (guilds, unions), political organizations (parties, bureaucracies), kinship associations (lineages, clans), religious cults (Moonies, Wiccans), and the like.

The nation conceived as a social boundary system is constituted internally of other social networks which also may be more or less firmly bounded. Firmly bounded networks include individuals that give their allegiance and loyalty to the goals of the network. Loosely bounded networks include individuals who do not concede their loyalty to the network. As a result loosely bounded networks allow individuals to pass through the network without identifying with its goals (Cohen 1969b). A major goal of a nation's government is to reduce the loyalty its citizens give to firmly bounded local networks and replace them with a firmly bounded nation under a government to which all citizens give their primary allegiance.

The state as a structure of offices, government as incumbents of those offices, and a nation as socially bounded networks of relationship that is subject to a state's government authority constitutes a state formation. Within that formation the relationship of state governments and their local socially bounded networks creates a dialectic of central government control and local autonomy (Giddens 1979). As a working hypothesis a nation will be either firmly or loosely bounded in an inverse relationship to the firmness or looseness of the boundaries of the social networks that comprise the nation, and these boundaries can contract or expand through pressures exerted internally and externally to the network. The ideal boundary system from the point of view of a government is a firmly bounded nation in which all citizens give allegiance to the government. A nation's government attempts to accomplish this ideal by subverting the integrity of socially bounded networks within the nation which may threaten it (Cohen 1969a, b). The extent to which such networks exist is a matter of degree that is influenced by myriad factors.

The vertical entrenchment of government authority (theory)

Perhaps the major challenge to any state government, certainly the governments of early states, is to acquire the support of the citizens that abide within the nation. In early state formations the vertical entrenchment of government authority is a response to the perception, real or not, that the allegiance and loyalty a nation's population gives to its local institutions, values, and ideologies presents a threat to the authority and, perhaps, existence of the government and the integrity of the state formation it represents. I take it to be axiomatic that all states, governments, and nations are to some degree inchoate, works in progress which are never completed and only rarely attain the support of all their citizens. Here I will focus on how governments of early state formations attempt to overcome the inchoateness of their nations through the vertical entrenchment of their authority.

The idea of the entrenchment of government authority (Cohen 1969a) complements the idea of the legitimation of government authority (Kurtz 1978, 1981, 1984, 2001; Claessen 1988). But entrenchment and legitimation operate from different principles and engage political agents in different practices. The legitimation of the authority of an early state's government requires support from two of a nation's social categories: the ruling class and the common people. Machiavelli identified the dialectic embedded in these two levels. He pointed out that

a prince who attains his principality with the aid of the nobility maintains it with more difficulty than he who becomes prince with the assistance of the common people, for he finds himself a prince amidst many who feel themselves to be his equals, and because of this he can neither govern nor manage them as he might wish (Machiavelli 1952: 63).

On one level then, the governments of early states have to be concerned with threats to their legitimacy from the ‘nobility’ or ruling class from which the heads of state are recruited (Claessen 1988). State governments deal with threats to their legitimacy at this level by keeping close watch on those who might inspire a coup d'état and commonly respond to such threat with considerable force. On the other level governments seek legitimation of their authority by striving to acquire the support of their nations' populations. At this level legitimation is a complex and protracted process that involves reciprocity between rulers and ruled as well as the exercise of benevolence and coercion by the rulers (Kurtz 1978, 1981, 1984, 2001).

The entrenchment of authority, on the other hand, is a response to the dialectic of local autonomy and centralized government control (Giddens 1979). Whereas legitimation is a concern of every state government, the entrenchment of state authority is a process that transpires most commonly in early, new, or otherwise inchoate and marginally legitimate governments. The early state formations that are the focus of this paper emerged as the result of the conquest and incorporation by one population of neighboring populations in environments that were either socially (Zande) or environmentally (Inca, Aztec) circumscribed (Carneiro 1970). State formations that emerge in this way are represented by governments that share much socially and culturally with their nations' citizens. They are similar in their cognitive orientation, language, technology, and ethnicity, and the symbolic gap between the two is narrow to begin with. The governments of these formations confront populations within which established socially bounded networks, especially those identified with kinship associations, demand the loyalty and allegiance of their members at the expense of loyalty and allegiance to the governments of the state formations in which they exist (Cohen 1969a, b). To govern effectively – at least from a government's point of view – rulers need to establish some distance from those they rule (familiarity breeds contempt). They begin to accomplish this by entrenching their authority in the nation's local institutions and distinguishing more sharply the culture of the ruled from that of the rulers. Each of the state formations under scrutiny complies with this model, and the degree of social, technological and cultural homogeneity each manifested is important to the strategies by which their governments pursued their entrenchment.

Claessen (1978) established a typology of early state formations which he identified in ascending order of complexity as inchoate, typical and transitional. This taxonomy depends on 6 criteria concerned with trade, markets, succession to offices, private ownership of land, remuneration of functionaries, and the degree of development of the judicial and taxation system. Each of these criteria was less developed in inchoate states and most developed in transitional states (Claessen 1978).

My problem is that the criteria in this typology seem to be most compatible with practices of governments and organizations of relations within the nation. My analysis will rely on data related to state formations depicted by this typology: Zande (inchoate), Inca (typical), and Aztecs (transitional). However, in this analysis the designations ‘inchoate’, ‘typical’, and ‘transitional’ do not refer to types of states exclusively. Instead they represent the qualities of nations and their governments, for each of these formations was characterized by significant inchoateness; that is, each was constituted of socially bounded networks of relationships which, from the point of view of each state's government, were inimical to the government's idea of what constituted an acceptable nation.

In some instances the governments of these nations did determine that firmly bounded networks, such as secret societies or conquered communities, posed real and imminent threats to their right to rule. In other contexts, it is difficult to determine when and to what extent these governments actually determined that lineages or other firmly bounded kinship associations – clans, moieties, joint and extended families – did not serve the government's interests. But nearly all the governments of early state formations – and more recent ones – seem to establish policies aimed at subverting the allegiance and loyalty individuals give to their lineages (Fallers 1965; Geertz 1963; Fried 1967; Cohen 1969a). In this paper I will pay special but not exclusive attention to relationship between the governments of these early states and the kinship associations within their nations.


The Zande represent an ethnic community that existed in the region of south eastern Sudan that today overlaps with the Democratic Republic of the Congo (Zaire). From the end of the 16th into the 18th century waves of immigrants began to move into this increasingly socially circumscribed territory. The last wave included a people known as the Mbomu who were led by chiefs of the Vongara clan. By the 18th century they were conquering other ethnic groups in the region and by the middle of the 18th century the Mbomu under the Vongara clan had established the state formation we know as the Zande, the name of the most prominent ethnic category in the nation which the Vongara clan governed. This state formation persisted until the intrusion of the British into the area in the middle of the 19th century (Kandert 1978).

By the 13th century the Inca were one of several tribes that occupied the Cuzco Basin what today is in the highlands of south central Peru. By the 14th century the Inca had become a chiefdom, and in the early 15th century they were at war with other societies in the basin. Between 1400 and 1450 they began to expand by conquering other chiefdoms in the region. In 1450 the Inca established a state with a government situated in Cuzco from which it dominated by 1530 an empire of about 8.0 million people across a territory of almost 1.0 million square kilometers. It was the largest empire in world history to be united by a people which lacked some form of four-footed animal as a means of conveyance (Schaedel 1978). The empire was bounded on the north by what is today the Ecuador – Columbia border and on the south by the Maule River in southern Chile. On the east it extended irregularly along the Andes to where the mountains graded off into the tropical rain forest. The Pacific Ocean established its western boundary (Rowe 1947; Katz 1974; Schaedel 1978). In 1533 with some help from Indians who opposed Inca domination, Spaniards conquered and overthrew the government of the Inca state and empire and formed a new government to rule the region.

After the empire of the Toltecs of Tula fell in the 12th century the Mexica, or Aztecs as they came to be known, were one of many societies that migrated into the Valley of Mexico. By 1325 the Aztecs had built Tenochtitlan, the capital and city-state, in the marshy lands of Lake Texcoco. In 1428 they established their hegemony over other city-states in the Valley of Mexico and then went on a binge of conquest that resulted in a loosely-knit empire5 that extended from the borders of the northern desert just above the valley of Mexico to the Gulf coast on the east. From there it stretched south before turning west above the Isthmus of Tehuantepec to the Pacific Ocean. It extended north along the coast to the southern border of the current state of Michoacan. At that point the empire's boundary moved east to the western edge on the central highlands where it went north and connected again to the southern border of the desert above the valley of Mexico. In 1521 Spaniards and large numbers of Indians opposed to Aztec domination overthrew the Aztec government. Thereafter the Spanish crown administered the first imperial colony in the New World (Soustelle 1961; Berdan 1982).

State, Government, Nation: case studies


Around 1850 Zande was a loosely knit nation of about thirty five relatively independent government units of varying size. These units were divided into 7 provinces of different sizes, each of which was governed by chiefs of one of the 7 ‘ruling houses’ of the Vongara clan. One of these clans provided the king of the Zande nation who ruled until his death. At that time another Vongara clan became hegemonic and provided the next king of Zande. Regardless of which Vongara clan provided the successor to the head-of-state, the structures of government and state remained about the same (Baxter and Butt 1953; Evans-Pritchard 1971; Kandert 1978).

The state was comprised of a structure of offices that included a head of state, royal oracles (two or three), provincial governors (chiefs?), sub-chiefs, deputies, military leaders, and some offices with special duties, such as commanders of the guard and deputy leaders of military companies. With minor variations due in part to the size of the province, each of the other 6 provinces was represented by a structure of offices that largely replicated that of the central state to whose government they were subordinate and gave allegiance, albeit, often grudgingly (Baxter and Butt 1953; Evans-Pritchard 1971).

The difference between these offices was reflected in the social status of who occupied them. The head-of-state and the governors of the provinces all were from the Vongara clan. Oracles that advised the king apparently were recruited because of their supernatural powers and did not seem to come from any particular class or clan. Sub-chiefs were appointed to their offices by the Vongara governors. Some were members of the Vongara; nepotism was common. Others were appointed from the Mbomu peoples, as were all other offices holders.

The office of head of state was imbued with powers that established the ruler as the nation's supreme judicial figure, commander of the nation's armed forces, manager of the government's tribute and taxes, and delegator of authority to those incumbents who occupied the offices of sub-chiefs and deputies of the state bureaucracy (Kandert 1978). The major duties of the offices were to defend the nation's borders, maintain order, collect taxes, organize corvée labor, and administer the assimilation of foreigners into the Zande nation. The functions of each of the 6 Vongara governors and their subordinates were largely a microcosm at the provincial level of the powers of the Vongara clan whose members comprised the state's central government.

In practice, however, the governments of the state and provinces were unstable. They were in continuous conflict with peoples on their borders. Provinces also were marked by a fission and fusion as governors, chiefs, and sub-chiefs fought for power and status and frequently realigned their territories. A national unity prevailed primarily when the central government or its lower echelons faced a threat from outside the nation. Conflicts were especially extreme when a head of state died; Zande governments had not established formal rules of succession. The interregnum was marked by conflicts that approximated rituals of rebellion (Gluckman 1963) and persisted until one of the contesting clans was successful in occupying the office of head-of-state. At this time the location of the central state was identified with the province of the winning contender and government continued much as before.


Around 1533 the Spanish conquered Cuzco, the capital of the empire and its 100,000 to 200,000 people, and took control over the Inca Empire. The Inca nation per se was restricted to the Cuzco basin where the Inca-by-blood originated and lived. However, Inca culture, material and ideational, spread so widely throughout the empire and Inca politics had penetrated so deeply into the polities it governed that it is not unreasonable to consider the empire the de facto nation of the Inca government and state.

At the apex of the Inca state structure was a head of state, or emperor, and a royal council of 4 to 12 offices. The council represented the 4 major divisions of the empire and these offices were located in Cuzco, as was the large federal bureaucracy. Each major division, or quarter, of the empire had a capital city which housed the offices of the governors of the province and some of the 80-odd bureaucracies throughout the provinces that included offices to administer affairs of the provinces. Lower provincial offices were classified according to the number of tax payers for whom each office holder, or chief, was responsible: 10,000, 5,000, 1000, 500, 100. Below these chiefs were 2 other offices (foremen) which were responsible for 50 and 10 taxpayers respectively (Rowe 1947).

The government consisted of the emperor and the members the royal council and the central bureaucracy. This council was a supra-provincial consultative body which advised the emperor and may have received his response through a secretary who served as an intermediary between the ruler and the council. Some of these members also may have served as ‘viceroys’ and in jural capacities in administrative affairs. Other of its members may have served as the ‘general staff’ of the Inca military and were charged with drafting men for military campaigns (Rowe 1947; Katz 1974; Schaedel 1978).

Many of the functions performed by the incumbents of the provincial bureaucracies overlapped with those of the central bureaucracy which also tended to ratify, oversee, and coordinate functions of the provincial bureaucracies. As well as involving himself personally in many matters, such as the production of luxury goods, the emperor delegated authority and power to the central bureaucracy to oversee economic production, administer higher education, central government cadastral functions (record keeping), supervision of bridge and road maintenance, messengers, ‘supreme court’ sanctions, and the operation of the royal court. Other offices in this bureaucracy were vested with authority to appoint provincial governors who in turn appointed local officials to offices such as judges and ‘inspectors’ which probably were vested with juridical duties. Still other offices vested their incumbents with authority over treasury function, such as maintaining the quipus related to government storehouses, provisions, and revenues. Finally, the central bureaucracy had an indeterminate number of offices whose incumbents were dedicated to supervising boundary markers of the nation's territories and communities. There was an overlap and probable coordination between the central bureaucracy and the authority vested in local government offices regarding the collection of taxes and tribute, management and records of the treasury and the storehouses for surplus goods, coordination of religious and juridical functions, and enforcement of the military draft as it was ordered by the royal council.

The central bureaucracy was staffed almost exclusively of Incas ‘by-blood’ and Incas ‘by-privilege’ that came from ethnic groups, largely Quechua speakers in the Cuzco area. The higher levels of the provincial bureaucracy were staffed by the ruling lineages of the conquered polities. However, the middle and lower echelons of each bureaucracy was staffed by ‘permanently settled bondsmen’ (Schaedel 1978), that is individuals who were obligated to service presumably without wages. With few exceptions, in addition to its political functions, the federal bureaucracy was distinguished from the provincial bureaucracy by ethnic consideration (Inca/non-Inca), and economic factors; sources of income for the central government came through taxation and tribute levied across the empire; local governments relied on income from taxes collected from their own constituents. Sometimes these levies were interdicted by the central government and diverted to Cuzco as a way of undercutting the authority of local governors (Katz 1974).


In 1519, just prior to Cortes' incursion into Central Mexico, the Aztec state consisted of a typical structure of offices. At the apex was the office of the head-of-state, or king-emperor. This office was vested with the usual authorities that connote a state structure: execution of the laws (the ultimate source of appeal), management of revenues from taxes and tributes, and command of the army. The office of head of state was complemented by a special office known as the cihuacoatl. The individual who occupied that office represented a kind of ‘Vice-Emperor’, or perhaps Viceroy. He assumed the status of head-of-state when the Emperor was away and on these occasions lived in the royal palace. Four other offices, which may have included that of the cihuacoatl, comprised the chief councilors to the king. Beneath those offices were 13 to 20 other offices that accommodated the supreme council of Tenochtitlan. The offices of head-of-state, council of 4, and supreme council were largely staffed by nobility by-birth (Soustelle 1961; Berdan 1982).

Intermediate offices for ambassadors and royal palace officials were numerous and a grade below the central offices of government. These offices were not necessarily filled by Aztecs of noble birth. Beneath these intermediate offices was an extensive bureaucracy composed of minor offices that accommodated judges, tax and tribute collectors, heads of the city's 4 wards, heads of the city's calpullis, or barrios of each ward, market personnel, scribes, police, bailiffs, executioners, street sweepers, and the like. Lower offices were occupied by commoners, although commoners could become nobility by appointment for exemplary service to the king, for example in warfare.

The incumbents of this office structure comprised the government of the Aztec state. Their duties and obligations were determined by the authorities and powers delegated to and vested in those offices, largely by the emperor. In practice the emperor apparently either appointed, confirmed, or sanctioned the incumbents of these offices, including those very low in the state's bureaucracy, such as street sweepers (Soustelle 1961; Berdan 1982).

The emperor relied on advice and information from his councils, especially the cihuacoatl, to render his decisions. The supreme council of 13 to 20 persons informed the council of 4 who then, through the cihuacoatl, conveyed to the emperor their opinions on political, economic, and other affairs that might affect the nation or empire. But unlike the Inca government, the Aztec government did not micromanage its empire to the extent the Inca did.

The Aztec empire was considerably more diffuse and loosely knit than that of the Inca. It consisted of about 38 tributary provinces associated with a dominant city-state and other small city-states (Soustelle 1961). The government of each city-state, including Tenochtitlan, more or less replicated each other in structure and function. The strategy by which the Aztec government extended its empire consisted of an offer by the emperor to another city-state to join the empire, the major stipulation being that if the offer was accepted the government of that city-state agreed thereafter to pay tribute to the Aztec government. Given the reputation the Aztecs had attained as unforgiving warriors, this offer was not to be taken lightly. The governments of city-states that agreed to be incorporated into the Aztec empire retained their rulers and government. The only Aztec officials they were likely to see would be the tribute collectors, and the only inconvenience would be the required visit to the Aztec emperor by the tributary ruler on certain scheduled times to pay him fealty. If they refused, the Aztec army would wage an internecine war against the city-state and, once vanquished, impose an Aztec military garrison and governor over it. The tribute then exacted by the Aztec emperor was considerably higher.

Unlike the Inca whose culture became widespread throughout its empire and, arguably, provided a rationale for equating the Inca empire with the nation, nothing similar existed in Mexico. A culture that might correspond to an Aztec nation was restricted largely to Tenochtitlan, and perhaps to some extent to the city-states around Lake Texcoco in which considerable numbers of Aztecs resided. Tenochtitlan was so densely populated with about 200,000 citizens that it could not hold all the Aztecs in the region. Aztec dominance in the Valley of Mexico may have given an Aztec ‘flavor’ to the culture of the cities around the lake. Nonetheless, the idea of an Aztec nation is perhaps best limited in scope to the population of the city of Tenochtitlan.

The vertical entrenchment of government authority (praxis)

Praxis refers to that unity of data and theory which provides strategies to explain the human condition. The praxis I employ here engages data concerned with the firmly social boundary networks of relations in the Zande, Inca, and Aztec nations. It uses these data to develop working hypotheses and explain how the vertical entrenchment of government authority subverts the loyalty peoples within their nations give to local firmly bounded social networks.


As noted, the goal of government entrenchment of authority is to shift the allegiance and loyalty individuals give to local institutions to the state's government and to the symbol the state represents. This hypothesis addresses the dialectic of central government control and local level autonomy (Giddens 1979) in early state formations and, by extension, the implication of this dialectic for state formations in general. One way to explore this problem is to relate government practices to those socially bounded networks within nations that governments perceive to be threats.

Cohen (1969b) analyzed the implication of socially bounded networks for social and political processes in two widely-spaced geographic and temporal social entities: a Puritan town in 17th century New England and the post World War II Tokyo ward of Mamachi. He determined that a firmly bounded social network manifests three characteristics. First, the roles individuals play within the network are transposable with respect to the activities of that network; that is individuals within the network can substitute for others in meeting the goals of the group that constitutes the network. Second, there is a consensus and lack of dissent among the members of the network regarding the goals of the network and the values and ideologies held by its members. Third, the network is differentiated and perhaps specialized in its activities from other socially bounded networks, but it is undifferentiated internally with reference to the role individuals play in the network. Conversely, loosely bounded networks will stand in an inverse relationship to these criteria.

Cohen's analysis explains reasonably well the significance of these criteria for determining the dynamic of social boundary relations in the Puritan town and Mamachi. They also account reasonably well for the degree to which kinship associations, lineages, clans, moieties, and the like, may be firmly or loosely bounded. They may also apply to nations. But the social boundary networks represented by nations are usually too differentiated and composed of too many differentially bounded networks to be distinguished neatly by these criteria. Nations also will vary in their social complexity depending on their level of technological adaptation. For example, in industrial nations socially bounded networks, such as kinship associations, witches covens, or organizations of sorcerers are not likely to represent the threat that they might to governments of early state formations. Thus when a nation is conceived as a maximal boundary network other criteria have to be introduced to account for it. Cohen (1969b) argues that a nation will be firmly bounded to the degree that persons and influences that threaten its government are excluded from it and the degree to which the recruitments of individuals as citizens of the nation is governed by a rite of passage. Cohen also suggests that the idea of a firmly bounded nation may, in most instances, represent a metaphorical way of thinking about nations and social boundaries instead of a body of hard facts ascertained from empirical data6.

The recruitment of individuals into firmly bounded networks, such as lineage in pre-industrial state formations, is likely to depend upon complicated and often protracted rites of passage (Van Gennep 1960; Cohen 1964). However, the rituals that may serve to recruit individuals into the boundary network represented by a nation may be no more complex than establishing one's citizenship as a result of one's birth within the nation (Cohen 1969b). Beyond that, governments strive to initiate policies to control recruitment of citizens and acquire their loyalty. For example, governments may regulate immigration and movements of people and/or require attendance in government managed schools. The first policy regulates outside influences that may affect the nation's boundaries. The second represents a common strategy by which governments strive to create good citizens by inculcating government values and ideologies into the nation's children.

A nation, however it is conceptualized, is constituted of internal networks that may, depending on a variety of circumstances, dilute the firmness of the nation's boundaries. In general, as noted, a nation will be either firmly or loosely bounded in inverse relationship to the firmness or looseness of the socially bounded networks within it. For example, the government of a nation that is constituted of many firmly bounded social networks, such as kinship associations, secret societies, castes, and the like is apt to be weaker and the nation more loosely bounded and open to outside influences and potential conflict internal and external to the state's formation. The government of a nation that has reduced or subverted the integrity of these firmly bounded networks and attained the allegiance and loyalty of their citizens is likely to be politically stronger, legitimate, and either less porous or better able to control the threat posed by bounded networks within the nation. The nations of all state formations are to some extent composed of such socially bounded networks. The viability of
a nation's government relies largely on how well that government copes with these networks.

For example, the governments of industrial nations have succeeded in establishing monogamous nuclear families as their nation's fundamental kinship structure. From the perspective of these governments, the monogamous nuclear family represents the ideal social structure for a nation to attain. A firmly bounded monogamous nuclear family – the more firmly bounded the better – poses the least organized threat to the integrity of a government. Such a family or household organization is least likely to develop into a large extended kinship association, such as a lineage, that might present such a threat. However, other firmly bounded networks may exist that do present real or perceived threats to these governments.

These include an array of associations that deny or refuse to accept the authority of their governments: politically oriented neo-Nazis in Europe and secessionist survivalists in the United States; religious cults, such as Fulan Gong in China and Doukhbors in 18th century Russia and 20th century Canada and the United States; political-religious dissidents, such as the Islamic Jihad or Al Queada. These and other such associations – anarchists, Moonies, hippies – have in the recent and distant past challenged and suffered responses from governments of industrial nations.

The ‘new states’ that emerged from the collapse of European colonial empires following World War II were closer in their social structures to early state formations and the firmly bounded social networks with which their governments had to contend. Many problems related to these networks in the new states were induced by purposeful practices of colonial governments, such as drawing national boundaries that included historically antagonistic populations. Internecine conflicts that erupted in post-colonial nations often appeared to outsiders and media to be irrational, explained by racist innuendo regarding the presumed incompetence of government in these nations. On the other hand, what appeared to outsiders without an anthropological predisposition, the media for example, to be irrational behaviors often was the dialectical drama between central government control and the autonomy of local firmly bounded social networks at large in these nations. One of the major threats to these governments, real or perceived, emanated from the corporate kinship associations, in particular lineages that claimed the allegiances and loyalties of their members (Gluckman 1955; Fallers 1965; Geertz 1963; Fried 1967).

It is no accident that lineages became a central focus and concern of anthropologists. Anthropologists had to unravel lineage organizations to understand the politics in stateless societies. When early state formations evolved lineages common to stateless societies persisted as basic sociopolitical structures of these nations, although now they often were dichotomized between lineages of the rulers and those of the ruled. Because of the firmness of their boundaries governments of early states had to contend with lineages as potential threats to their viability. If there is meaning to the typology of early states formation as inchoate, typical and transitional it should be reflected in the extent to which the governments of early states are successful in coping with the threats posed by lineage and other kinship associations.

A hypothesis from which to explore this phenomenon suggests that kinship associations as firmly bounded socially networks will be most viable in inchoate state formations and least-well developed in those that are transitional. This research will test this hypothesis as well as the general hypothesis related to the dialectic of government control and local autonomy in inchoate, typical, and transitional early state formations. Data from the Zande (inchoate), Inca (typical), and Aztec (transitional) state formations will provide the basis to explain the entrenchment of government authority in early state formations.


Perhaps the only criterion that was shared by the governments of each of these nations just prior to their conquest by Europeans was the existence of a ruling class comprised of members of firmly bounded royal lineages or clans. Members of these lineages provided the heads-of-state; others served as government incumbents and were responsible for policies and practices that impacted kinship networks at the local level7. In addition to this shared criterion, the nations that were ruled by the Zande, Inca and Aztec governments also contained other networks of firmly bounded social relations.

The Zande nation harbored secret societies and castes of iron-smiths (Evans-Pritchard 1971). The Inca nation included age graded organizations (Rowe 1947), classes of soothsayers and sorcerers (Brundage 1967), priesthoods and nobility by birth and appointment (Rowe 1947; Katz 1974), and ‘sisterhoods’ associated with the cults of deceased emperors (Brundage 1967). The Aztec nation contained professional traders, priesthoods and associations of temple virgins, elite warrior societies, associations of specialized craftsmen, sorcerers, nobility-by-birth and appointment, endogamous quasi-kinship associations, and slaves (Soustelle 1961; Berdan 1982). The extent to which some of these networks represented threat to their governments varied historically. For example, secret societies seemed to be increasing in significance in Zandeland, as were elite warrior societies among the Aztecs. The firmness of socially bounded kinship networks varied in each nation, but not always as the hypothesis regarding the entrenchment of government authority might predict.


The Zande nation was comprised of patrilineal clans. The sharpest class distinction was between the endogamous Vongara ruling clans and the local exogamous lineages and clans of the Mbomu and other ethnic categories that comprised the bulk of Zande citizenry. The ruling Vongara clans were firmly bounded and identified with the chiefs and seats of government of each territory. The integrity of local lineage, however, was badly compromised due largely to the wars by which the Vongara established their political hegemony over the nation and conflicts that persisted thereafter.

Zandeland was not a firmly bounded nation. Provincial chiefs were obligated to protect the nation's boundaries against outsiders and wars and conflict were constant on the nation's peripheries; Zande raided neighbors and were raided in turn. But Vongara chiefs also fought each other over territorial and other disputes. The king of Zande intervened to help organize defenses against outsiders. Local Vongara chiefs were largely left alone to resolve their differences, unless they posed threat to the ruling Vongara clan of the Zande nation.

The integrity of local lineages and clans was a casualty of these conflicts. The theory of government entrenchment suggests that governments attempt to subvert local level kinship associations. In Zandeland this subversion was accomplished by the international and national conflicts in which its rulers engaged. Because of these conflicts populations and lineages were dispersed; people moved frequently to avail themselves of better conditions, and the integrity of local kinship associations suffered accordingly. Individuals were aware of broader kin ties, but the only significant ones were those close at hand, and these often were destabilized as households moved as independent units to seek better conditions. Provincial chiefs and their administration frowned upon these movements. But they either did or could do little about them. Citizens were aware of this and had little to fear unless they returned to a province they had abandoned previously where chiefs might punish them (Evans-Pritchard 1971).

There is no direct evidence that secret societies, or ‘closed associations’ as Evans-Pritchard (1971, 1976) refers to them, filled the vacuum left by the breakdown of kinship associations. These societies took root in the southern borders of Zandeland where conflict was most intense and existed prior to the British invasion. Their expansion following British rule may be explained by the turmoil created by the British occupation. But descriptions of these associations also suggest that citizens who were dispersed by conflicts on the borders found meaningful affiliations in them.

The closed associations used magical powers to help individuals, men and women, cope with problems caused by war, dislocation, disease, bad luck, and the like. These associations crosscut ties of kinship, friendship and even enmities that might be assuaged by membership. Recruitment to the closed societies involved rites of passage and the boundaries were firm and in some instances enduring. But some closed associations also prove to be ephemeral. The Vongara chiefs and king were concerned about the spread of these societies, as were the British, and in some instances they were successful in subverting them. Still, a few did endure and established hierarchies of membership and bonds of trust and friendship that apparently, for many, helped them withstand conflicts and displacements of peoples in Zandeland.


Compared to the Zande, the Inca government had to cope with a large empire which I equated previously with the Inca nation. The central government of this nation resided in Cuzco, a city comprised largely of bureaucrats represented by the offices of the emperor, his council, and the central bureaucracy. In addition to regulating travel in and out of Cuzco, the government correlated the political, economic, religious and social affairs generally of the nation through some 80 provincial bureaucracies over which the central bureaucracy exercised authority. On advice from the central bureaucracy the emperor appointed local chiefs and conferred them in office, decided the allocation of tribute from conquered provinces, supervised craftsmen and traders who worked for the government, and oversaw the mobilization of troops for war (Rowe 1947; Katz 1972; Schaedel 1978).

Endogamous moieties were widespread in the central Andes and composed of ayllus, probably lineages, which were the fundamental kinship associations of the nation. They royal ayllu of the Inca ruling class was endogamous and polygynous. The ayllus of the commoners were exogamous within endogamous ayllus and moieties and largely monogamous (Rowe 1947; Schaedel 1978). Alliances between ayllus within a moiety and perhaps beyond were established through various patterns of cross-cousin marriage and sister-exchange (Harris 1986; Lounsbury 1986). In some instances the dynamics of these lineages may have replicated a segmentary lineage pattern that expanded and contracted as internal and external conditions demanded (Platt 1986). Compared to the dispersed Zande kinship associations, those of the Inca were stable and thriving systems marked by networks of relationships so firmly bounded that they probably paid tribute to the central government as corporate bodies (Katz 1974).

Not only do the governments of early state formation worry about firmly bounded kinship association, at least among the Inca – and even modern state formations – the intentions of young unmarried men are also suspect. Young men without family responsibilities are not ideal citizens. With too much time on their hands they are not especially productive and they may be recruited for actions inimical to a government's interests. Under the pretext of confirming adulthood by marriage, the central government required young men and women to marry, and should a young woman be indecisive about which swain to wed, the government decided for her (Rowe 1947; Katz 1974). This practice suggests a trade-off by the government between having tightly knit kin associations on the one hand and, on the other, lots of unmarried young men which might be induced to subversive acts. Still, practices and policies by the central government suggest that it perceived firmly bounded kin associations the greater threat.

The central bureaucracy of the Inca established different administrative policies which, except for the Spanish conquest, might have subverted the firmness of these associations. The central government did not attack these associations head on. That policy would have been too disruptive to the Inca political economy and the extensive redistribution system that helped weld the empire together. Instead the government engaged strategies aimed at integrating, perhaps eventually assimilating, them into the nation (Schaedel 1978).

The Inca government initially expanded its control over the empire through conquest. But force was not always desirable and could be disruptive. In one strategy intended to preclude conflict, the central government moved administrators and garrisons near the frontiers of those they wished to incorporate. Instead of invading and acting as an occupying power the government used these garrisons to introduce Inca culture to the governments and people they wished to incorporate. Gradually Inca administrators began to vertically entrench their culture and government and alter the local government to its satisfaction (Salomon 1986).

This strategy was not entirely benign. If the government deemed it necessary it might use force to accomplish this goal. But administrators aimed their policies largely at the governments of these areas. Initially local ayllus were not disturbed, although warfare did, as among the Zande, disperse some kin groups. But in general, Inca government policies strove to co-opt local governments without conflict. If this was successful ayllus, already accustomed to domination by central authorities, were largely passive (Katz 1974; Salomon 1986).

However, had the Spanish not intervened, entrenching policies by the central bureaucracy would have effected changes in local ayllus. These strategies were imposed either by Inca governors or local rulers subject to the Inca government. For example the Inca military did not rely on conscription. Instead the government required soldiers of conquered provinces to serve in the Inca army. This strategy aimed to shift the loyalty of those serving in the Inca military to the central government. As with many of their entrenchment strategies, the Inca government established these policies slowly so not to provoke resistance. Another strategy required citizens from the highlands to move into territories newly annexed to the empire and mix their culture with that of the local populations. Since the structure of ayllus was similar throughout much of the empire, intermarriages and cultural exchanges, backed by Inca provincial governors and the military, were expected to result in an ‘Incazation’ of these territories. Finally, in a more aggressive policy the central government might deport recently conquered population to other regions. This served to break up local primordial sentiments and reduce the firmness of boundary networks at the local level (Rowe 1947; Katz 1974; Schaedel 1978; Salomon 1986). The Inca government may have been aware of the potential threat posed by local ayllus. But at the time of the Spanish conquest much of its entrenchment was aimed at integrating the empire with the least conflict.


The Aztecs that entered the valley of Mexico in the 14th century were a chiefdom comprised of perhaps 20 lineages which may have ranked hierarchically. By 1519 Aztec social structure was similar to that which prevailed throughout central Mexico and, as it was with the Zande and Inca, the social structure of the nation was bifurcated between a ruling nobility and everyone else. Aztec nobility maintained a firmly bounded system of royal lineages that were endogamous, polygynous, and allied through patterns of cross-cousin marriage. Aztec citizens continued to be organized into between 14 and 20 calpullis. But the firmness of their boundaries was much diminished and kinship was replaced by social and economic classes. This was reflected in the organization of many calpullis as craft or some other specialized economic units (Soustelle 1961; Berdan 1982).

At the local level Aztec citizens traced descent bilaterally. This pattern correlates to an open, much less firmly bounded network of kinship relations. In general, marriage was agamous and post-marital residence was neolocal, although joint or extended households were patrilocal and established around brothers, the eldest of which tended be the head of the household. In short, in about a century, approximately the same length of time during which the Inca government established its empire, the Aztec nation had undergone a social metamorphosis that was different from that of the Zande and Inca.

In comparison with the Inca, Katz (1974: 310–323) attributes these differences to a variety of materialist factors that are internal and external to each state formations. The Aztec nation was more urban and densely populated over a much smaller area. Economic integration in central Mexico was established by trade and commerce in complex market systems, not the redistributive economy of the Inca. To maintain the integrity of the empire, the policies and practices of the Inca government were more rigid and deeply entrenched in the affairs of its empire's citizens. The Aztec government was content to collect tribute from conquered city-states which, so long as they didn't challenge Aztec dominance, were pretty much left alone. An extensive system of roads and communications allowed the Inca government better control of the empire, and deeper entrenchment of its policies into the affairs and social structures of its communities. This kind of control was exercised by the Aztec government only over the valley of Mexico which was heavily urbanized and densely populated. The Aztec government maintained a looser system of political controls over the farther reaches of its empire.

These conditions cannot be discounted as explanations for the differences between the Inca and Aztec state formations. But to consider them only would be overlook those strategies by which governments attempt to entrench their authority vertically into the local affairs of their nations. As we saw, the Inca government was accomplishing some of this by the deportations of conquered peoples and colonization that mixed Inca citizens and conquered peoples. But for what ever reason, perhaps the difference in scale of the Inca and Aztec empires, over the century of its existence the Aztec government established policies and engaged in practices that did, intentionally or otherwise, restructure the local level of the Aztec nation (Soustelle 1961; Berdan 1982).

Some strategies of the Inca and Aztec governments complemented each other. Both resettled citizens as hedges against threats from conquered peoples. The Aztec government moved people to the empire's western border as a buffer against the threat posed by the Tarascan government. The requirement for citizens to serve in the military is a nearly universal strategy in state formations and works to shift local allegiances to the state government. The Aztecs did not incorporate troops from conquered peoples into their military apparatus. Nor did they have a standing army per se. But all Aztec citizens were required to be on stand-by for military service, which was frequent, and the government established elite military units (Soustelle 1961; Berdan 1982).

In other ways the entrenchment policies of the Aztec government approximated practices of modern governments. Primary education universally is dedicated to creating good citizens; higher education is dedicated more to inculcating skills and knowledge that individuals require to manage state formations and empires. The Inca and Aztec governments established schools for children of the nobility which, in effect, trained them in skills necessary to hold high offices-of-state. However, the Aztec government also established schools for children (boys and girls) of commoners in the various calpullis and their education ‘produced ordinary citizens’ (Soustelle 1961: 169). For boys the education emphasized military training to create citizens willing to die for their governments and the symbol the state represents. Finally, taxes levied against social categories, such as ayllus or calpullis, for which all members are corporately responsible does not diminish the firmness of the boundaries of the category. Most of the tax or tribute paid in services, such as corvée labor by citizens of the Zande, Inca and Aztec nations were levied on social categories, such a kin associations and communities. Taxes in kind also, except for the Aztecs, were levied on social categories. But Aztec citizens in their calpullis also paid taxes severally, suggesting that the firm boundaries of the calpulli had been much diminished. The result of these policies supports the designation of the Aztec state formation as one ‘transitional’ between early and more modern state formations.

Conclusions and discussion

I have heard (or read) the comment that ‘states tend to be paranoid’. This cannot be since states lack the protoplasm that confers life. Nonetheless, social scientists continue to anthropomorphize the state as a political agent.

In this paper I have attempted to identify that component of state formations that has good reason to be paranoid: governments. Governments are comprised of incumbents – protoplasmic human beings – of state offices, and they know that in the wings off the stages upon which they act out their political dramas there are always other potential incumbents who believe that they can do the job better. Governments worry about these off-stage interlopers. But they also worry about threats to their existence that emanate from within the nations they govern.

I identified the threats internal to state formation as firmly bounded social networks that claim the allegiance and loyalty of their members at the expense of the allegiance and loyalty they might give to their governments. Governments keep a watchful eye on those would-be politicians in the wings and occasionally react to them forcefully. But governments are always concerned about networks in their nations that threaten them, and they develop policies and engage in practices to subvert them. The ideal and rarely attained goal of these practices and policies is to transfer the loyalty of the members of those networks to the government itself.

All state formations confront these problems. Here I focused on three early state formations: the Zande, Inca, and Aztecs. In these and other formations a major real or perceived threat to governments emanate from firmly bounded kinship association, especially the lineage. Using data from inchoate (Zande), typical (Inca), and transitional (Aztecs) state formations I explored the hypothesis that governments attempt to entrench their authority vertically into their nations to subvert local firmly bounded networks and transfer the loyalty of the members of those networks to the government and the state, as symbol, it represents. The Inca were exploring ways of accomplishing this when the Spaniards intervened. The Aztecs had through various policies successfully replaced most of these kin association with social classes at the time they confronted the Spaniards.

The Zande were an anomaly. The lineages and clans of the Zande nation were being subverted. But the Zande government had no apparent policies aimed at doing this. Instead, probably because the Zande state formation – its government, state, and nation – was so inchoate, marked by internal and external wars and conflict, that the integrity of its lineages was subverted by the sheer inchoateness of the formation of which they were a part.

I think I demonstrated reasonably well the idea that established the government as the political agent of a state formation, although not as specifically and detailed as I might have wished. I believe that conceptualizing social structures as networks that may be either firmly or loosely bounded has analytic value and potential. It provides a way long recognized but rarely demonstrated to account for the dialectic of central government authority and local autonomy; that dialectic shows how internal and external pressures can shift these relations, for example from a nation that is loosely bounded due to the firm boundaries of the networks that comprise it to a more firmly bounded nation due to the reduction of the firmness of the boundaries of its internal networks.

Readers will have to determine for themselves whether the methodology I used here to explain the dialectic of government, state and nation is worthwhile.


1 I concede that I have no special expertise regarding these societies, not even the Aztecs with whom I once had more than a passing familiarity. But I suggest that my control of these data is sufficient to support the explanations to which this research is dedicated.

2 In the social sciences there are literally hundreds of definitions of the state (see Titus 1931; Easton 1953; Abrams 1988; Kurtz 1993, 2006).

3 For an analysis of the evolution of the ‘political office’ see Kurtz (2004).

4 The idea of the state as a political entity came into frequent use in the 16th and 17th centuries and referred usually to the offices of the government, or the government itself. The idea of the state was formulated fully in its myriad, more contemporary usage only in the 19th century (Easton 1953).

5 The Aztecs of Tenochtitlan were member of the Triple Alliance that included the Acolhuacans of Texcoco and the Tepanecs of Tlacopan. The military forces of these city-states sometimes fought together and shared tribute from conquered city-states. The Aztecs of Tenochtitlan were the dominant member in the alliance. Here I refer to the Aztecs without consideration of their relationship to their allies, even though in some instances reference to the other members of the alliance might be justified.

6 I suggest as a hypothesis that in nations characterized by a ‘closed door’ policy, such as Tokugawa Japan, 1610–1840, it would be possible to ascertain data that would justify the example of a firmly bounded nation.

7 Earlier I mentioned that a distinction between rulers and ruled was an important criterion for the vertical entrenchment of state authority. I have considered the process by which this was established elsewhere (Kurtz 1978, 1984).


Abrams, P.

1988. Note on the Difficulty of Studying the State. Journal of Historical Sociology 1(11): 58–89.

Baxter, P. T. W., and Butt, A.

1953. The Azande and Related Peoples of the Anglo-Egyptian Sudan and Belgian Congo. London: International African Institute.

Berdan, F. F.

1982. The Aztecs of Central Mexico: An Imperial Society. New York: Holt, Rinehart, and Winston.

Brundage, B. C.

1967. Lords of Cuzco: A History and Description of the Inca People in their Final Days. Norman: University of Oklahoma Press.

Carneiro, R. L.

1970. A Theory of the Origin of the State. Science 169: 733–738.

Claessen, H. J. M.

1978. The Early State: A Structural Approach. In Claessen, H. J. M., and Skalník, P. (eds.), The Early State (pp. 533–596). The Hague: Mouton.

1988. Changing Legitimacy. In Cohen, R., and Toland, J. D. (eds.), State Formation and Political Legitimacy (pp. 23–44). New Brunswick: Transaction Books.

Cohen, Y. A.

1964. The Establishment of Identity in a Social Nexus: The Special Case of Initiation Ceremonies and Their Relation to Value and Legal System. American Anthropologist 66: 529–552.

1969a. Ends and Means in Political Control: State Organization and the Punishments of Adultery, Incest and Violation of Celibacy. American Anthropologist 71: 658–687.

1969b. Social Boundary Systems. Current Anthropology 10(1): 103–126.

Easton, D.

1953. The Political System: An Inquiry into the State of Political Science. New York: Alfred A. Knopf.

Engels, F.

1942. The Origin of the Family, Private Property and the State. New York: International Publishers. [original ed. 1884.]

Evans-Pritchard, E. E.

1971. The Azande: History and Political Institutions. Oxford: Clarendon Press.

1976. Witchcraft, Oracles and Magic among the Azande. Oxford: Clarendon Press.

Fallers, L. A.

1965. Bantu Bureaucracy: A Century of Political Evolution among the Basoga of Uganda. Chicago: University of Chicago Press (Phoenix Books).

Fried, M.

1960. On the Evolution of Social Stratification and the State. In Diamond, S. (ed.), Culture and History: Essays in Honor of Paul Radin (pp. 713–731). New York: University of Columbia Press.

1967. The Evolution of Political Society: An Essay in Political Anthropology. New York: Random House.

Geertz, C. (ed.)

1963. Old Societies and New States: The Quest for Modernity in Asia and Africa. New York: The Free Press.

Giddens, A.

1979. Central Problems in Social Theory: Action, Structure and Contradiction in Social Analysis. Berkeley: University of California Press.

Gluckman, M.

1955. Custom and Conflict in Africa. Glencoe: The Free Press.

1963. Order and Rebellion in Tribal Africa. London: Cohen and West.

Harris, O.

1986. From Asymmetry to Triangle: Symbolic Transformations in Northern Potosí. In Murra, J. V., Wachtel, N., and Revel, J. (eds.), Anthropological History of Andean Politics (pp. 260–279). Cambridge: Cambridge University Press.

Kandert, J.

1978. Zande. In Claessen, H. J. M., and Skalník, P. (eds.), The Early State (pp. 511–529). The Hague: Mouton.

Katz, F.

1974. The Ancient American Civilizations. New York: Praeger Publishers.

Kurtz, D. V.

1978. The Legitimation of the Aztec State. In Claessen, H. J. M., and Skalník, P. (eds.), The Early State (pp. 169–190). The Hague: Mouton.

1981. The Legitimation of Early Inchoate States. In Claessen, H. J. M., and Skalník, P. (eds.), The Study of the State (pp. 177–200). The Hague: Mouton.

1984. Strategies of Legitimation and the Aztec State. Ethnology 23: 301–314.

1993. A Reconceptualization of the Anthropomorphized State and the Centrality of Political Agency in State Formations. Political and Legal Anthropology Review 16(1): 16–30.

2001. Political Anthropology: Paradigms and Power. Boulder, CO: Westview Press.

2004. The Evolution of Politics and the Transition from Political Status to Political Office. Social Evolution & History 3(2): 150–175.

2006. Political Power and Government: Negating the Anthropomorphized State. Social Evolution & History 5(2): 91–111.

Lounsbury, F. G.

1986. Some Aspects of the Inka Kinship System. In Murra, J. V., Wachtel, N., and Revel, J. (eds.), Anthropological History of Andean Politics (pp. 121–136). Cambridge: Cambridge University Press.

Machiavelli, N.

1952. The Prince. New York: Mentor Books (published by the New American Library).

Murdock, G. P.

1949. Social Structure. New York: Macmillan.

Platt, T.

1986. Mirrors and Maize: The Concept of Yanantin among the Macha of Bolivia. In Murra, J. V., Wachtel, N., and Revel, J. (eds.), Anthropological History of Andean Politics (pp. 228–259). Cambridge: University of Cambridge Press.

Rowe, J. H.

1947. Inca Culture at the Time of the Spanish Conquest. In Stew-
ard, J. H. (ed.), Handbook of South American Indians. Vol. 2. The Andean Civilizations (pp. 183–330).Washington, D.C.: Smithsonian Institution Bureau of American Ethnology, Bulletin 143.

Salomon, F.

1986. Vertical Politics on the Inka Frontier. In Murra, J. V., Wachtel, N., and Revel, J. (eds.), Anthropological History of Andean Politics (pp. 89–117). Cambridge: Cambridge University Press.

Schaedel, R. P.

1978. Early State of the Inca. In Claessen, H. J. M., and Skalník, P. (eds.), The Early State (pp. 269–288). The Hague: Mouton.

Smith, M. G.

1960. Government in Zazzau: 1890–1850. London: Oxford University Press.

Soustelle, J.

1961. Daily Life of the Aztecs on the Eve of the Spanish Conquest. Stanford: Stanford University Press.

Titus, C. H.

1931. A Nomenclature for Political Science. American Political Science Review 25: 45–60.

Van Gennep, A.

1960. The Rites of Passage. Chicago: University of Chicago Press. [original ed. 1908.]