Supralocal communities organized around institutionalized social
inequalities emerged repeatedly and independently around the
world between about 1,000 and 7,000 years ago. The earliest
such societies, often broadly labeled chiefdoms, frequently,
but not always, came into existence after the establishment
of sedentary agricultural living. This fundamental transformation
in human social organization was speculated about by mid-19th-century
cultural evolutionists on the basis of comparative ethnography,
and it has been more directly documented by archaeological research
since the mid-20th century. Comparative study has drawn heavily
on earlier cultural evolutionary work, sharing its focus on
similarities and universals. Hierarchical relationships, of
course, have deep roots in dominance behavior in the human (and
other) species, and the institutionalization of such relationships
provides effective organization, especially at large social
scales. That such institutions have appeared repeatedly in human
history is no surprise, but we do not fully understand just
how they came to emerge when and where they did, and not in
other times and places. We are also increasingly aware that
early chiefdom communities did not all take the same form and
did not all emerge in the same way. There are indications that
this variation is not just idiosyncratic detail, and that comparative
study of its patterning can provide insight into the developmental
dynamics of institutionalized social hierarchy. Ironically,
the very richness of information that has brought about this
realization is itself an impediment to effective comparative
study, which finally will need to draw on the social trajectories
of dozens of regions. When a recent comparison of seven early
civilizations runs over 750 pages, one despairs of ever seeing
the forest for the trees. Here, we look at three regions, focusing
on changing activities, relationships, and interactions in human
communities where inequalities emerged and became institutionalized.
 |
Interdependence Within Chiefly Communities in Oaxaca, Mexico, During the First Millennium B.C. |
Anthropologists have often studied villages—small local
communities of people in daily face-to-face interaction—and
archaeologists are accustomed to thinking of early sedentary
agricultural living as organized into just such villages. The
Rosario phase (700–500 B.C.) occupation of the Valley
of Oaxaca, Mexico, for example, consisted of about 65 such village
communities in an area of 2,125 km
2 that were surveyed systematically
(
1–
4). Most of these communities had fewer than 50 inhabitants,
but one, San José Mogote, had a population of more than
500. Small villages had been the norm in Oaxaca since the beginning
of sedentary life about 1500 B.C. They consisted of around 10
nuclear families whose small thatch-roofed wattle-and-daub houses
were scattered across 1 or 2 ha at intervals of 20–50
m. Households are easy to identify in the archaeological record
from the remains of structures, hearths, middens, and pits,
and their activities are easy to reconstruct from associated
artifacts and ecofacts. These activities include subsistence
and household maintenance tasks carried out by virtually all
families as well as a number of possibly specialized crafts
(
5,
6). A multidimensional scaling of 10 households at Fábrica
San José, a Late Guadalupe phase (800–700 B.C.)
village (
7), shows five grouped tightly together (
Fig. 1). These
households have evidence of only basic domestic activities.
Four others are set off from this group by the presence of distinctly
different sets of tools related to different crafts. Such specialization
means economic interdependence and a level of daily interaction
that would have strengthened the forces that pulled these 10
households together into a compact community in the first place.
Fábrica San José seems typical of small villages
in Oaxaca, not only during the Guadalupe phase, but also during
the preceding San José phase (1200–900 B.C.).
View larger version (14K):
[in this window]
[in a new window]
|
Fig. 1. Scaling of artifact assemblages from 10 Late Guadalupe phase households at Fábrica San José (Oaxaca), based on quantities of bowls, jars, ceramic decoration, figurines, animal bone (especially deer), flaked stone (especially obsidian), ornaments (shell, mica, polished stone), and tools and debris from craft production. Evidence of salt production, found in several households, was especially abundant in LG4.
|
|
In addition to economic interdependence, social inequality is
a second major axis of variability in
Fig. 1. One household
is set off sharply from all of the rest in this regard; its
members had many decorated serving vessels, ate considerably
more deer meat, and owned more elaborate ornamental possessions
than their neighbors. A 60-year-old woman of this household
was buried with four ceramic vessels and 55 polished stone beads—much
more elaborate treatment than any other burial excavated at
the site. Three of the four craft specialist households also
had somewhat elevated ranks on this social scale. The locations
of higher status households at Fábrica San José
show enough consistency through time to suggest a degree of
institutionalization or at least heritability of status. In
some villages, higher-status residential compounds were more
elaborate, including features such as the San José phase
1-m-high stone and adobe platform at Tomaltepec (
8).
In addition to its much larger population, the community at San José Mogote differed from these small villages in several ways (6, 9). Household economic specialization was more extensive, including a larger variety of crafts, although initially higher status residences were no more impressive than those of small villages. San José Mogote also had non-residential structures from the time of its founding. By San José and Guadalupe phase times these took the form of plastered platforms with wattle-and-daub structures for public or ceremonial activities. True temples of much more monumental proportions appeared in the Rosario phase.
Rosario phase San José Mogote thus fits the archaeological description of a central place—the nexus of a larger web of social interaction. The application of distance-interaction principles to a smoothed plot of the distribution of Rosario phase occupation on the Oaxaca landscape (Fig. 2) permits the delineation of that web in the form of a district that incorporates 28 villages within about 12 km of San José Mogote (1, 2). None of these villages has produced any evidence of the kind of monumental-scale public buildings or residences known for Rosario phase San José Mogote. Participation in ceremonial activities at San José Mogote’s public buildings was clearly one form of social interaction that produced this district, and this activity was closely associated with the presence at San José Mogote of the region’s most elite families. It was in this context of stable intercommunity interaction that households at the small village of Fábrica San José began to specialize in the production of salt, presumably for distribution to other local communities, since this resource was not widely available. Small communities may also have acquired some goods produced by San José Mogote’s greater diversity of craft specialists. With unique ceremonial activities and the district’s highest ranking families at San José Mogote, its relationships with the small villages were clearly asymmetrical, forming a hierarchically structured regional-scale community composed of village building blocks. In similar fashion, the internal organization of the central place relied on residential wards not unlike a set of villages pulled very tightly together by more intense interaction. Like the outlying villages in the district, these wards interacted through their hierarchical social relations with San José Mogote elites and participation in ceremonial activities at the center’s public buildings, but it is at this local scale that the economic interaction from complementary economic specializations seems particularly important.
View larger version (45K):
[in this window]
[in a new window]
|
Fig. 2. Smoothed plot of Rosario occupation in the Valley of Oaxaca. The flanks of the San José Mogote peak "capture" outlying smaller villages to form a regional-scale district (red line). Supralocal clustering elsewhere is minimal.
|
|
The San José Mogote chiefdom was constituted in these
activities, relationships, and interactions, and this community
represents the largest scale of such interaction at the time
in Oaxaca. Other supralocal communities were much smaller; the
San José Mogote chiefdom contained some two-thirds of
the entire regional population. The economic interactions of
households in local communities and the complementary specializations
of wards at San José Mogote are detectable as early as
1200 B.C. in the San José phase. Social ranking began
by the same time and grew stronger up through the Rosario phase.
During the Guadalupe phase, there were public buildings at San
José Mogote as well as at potentially rival centers,
but the activities for which they were constructed became highly
centralized at San José Mogote in the Rosario phase.
In Oaxaca, the regional-scale chiefly community depended heavily
on social and ceremonial interaction for integration. The economic
component of community interaction was particularly well developed
at the local scale, both at small villages and at the large
local community of San José Mogote.
|
Ceremonial Chiefly Centers in the Alto Magdalena, Colombia, During the First Millennium A.D. |
Regional-scale chiefly communities are also visible on the landscape
in the Alto Magdalena of the Colombian Andes. A smoothed density
plot representing Regional Classic (A.D. 1–900) period
occupation in a completely surveyed area of 317 km
2 in the Valle
de la Plata (
10,
11) shows four districts (
Fig. 3). The two
chiefdoms that are relatively complete in
Fig. 3 are roughly
half the size of the San José Mogote district, although,
at perhaps as many as 7,000 inhabitants, their populations were
three or four times that of the San José Mogote district.
This survey area is only part of a much larger zone containing
probably 20 or more such districts of roughly similar size,
focused around complexes of funerary monuments. In each complex,
the central elements are the burials of single individuals in
stone slab tombs, covered by earthen mounds up to about 4 m
high (
Fig. 4). The overall impact of the Alto Magdalena’s
monuments is to focus attention strongly on these specific individuals
(
12), who may well be represented in the large sculptures incorporated
into the barrows and the plazas adjoining them. The sculptures
combine human and animal characteristics in clearly supernatural
references, as in human faces with fangs or two-headed reptiles
on the backs of human figures (
Fig. 4). These themes recall
ethnographically known shamanic beliefs in supernatural power
involving human–animal transformations (
13). The monument
complexes range from small (a single mound and a few statues)
to large (several large mounds and dozens of statues on adjacent
hilltops) and are widely distributed across the Alto Magdalena.
The ceremonial activities carried out at the monuments and the
hierarchical social relationships they express so conspicuously
seem the central elements in regional-scale social integration.
The multiple regional-scale chiefly communities of the Alto
Magdalena are, in this respect, very much like the lone San
José Mogote district in Oaxaca.
View larger version (52K):
[in this window]
[in a new window]
|
Fig. 3. Smoothed plot of Regional Classic occupation in the Alto Magdalena. The region is easily divided into several districts of similar size (red lines), each centered on a monumental burial complex.
|
|
View larger version (73K):
[in this window]
[in a new window]
|
Fig. 4. Monumental tomb in the Alto Magdalena (Left) and statue of naked man with fangs and two-headed reptile on his back (Right).
|
|
There are, nonetheless, important differences between the two
regions in the nature of ceremonial activities and the way in
which social hierarchy is expressed. It is in ritual that social
inequalities are most strongly expressed in the Alto Magdalena,
since principal ceremonies apparently consisted of rituals performed
at funerary monuments, continually reemphasizing the importance
of specific individuals. Oaxaca public buildings also created
ceremonial spaces, but they were not, in their very essence,
monuments to particular people, even though elite residences
were close at hand. When, late in the Oaxaca sequence, monumental
sculpture appeared, it depicted not elites or supernatural themes,
but rather the unfortunate victim of political violence (
9).
A naked and bleeding sacrificial victim carved in relief on
a stone slab formed a threshold to be trod upon by those approaching
San José Mogote’s public buildings (
Fig. 5).
A stronger contrast is the relative weakness in the Alto Magdalena
of the economic component to interaction that loomed so large
locally in Oaxaca. Households of different ranks and specializations
can be identified in the Alto Magdalena by their artifact assemblages,
but these differences are quite small by comparison to Oaxaca.
Artifacts related to such differentiation are much more abundant
in Oaxaca. In a sample of San José phase households at
San José Mogote, 299 status-related artifacts were recovered
for every 100,000 sherds (
6). In the Guadalupe and Rosario phase
households at the peripheral village of Fábrica San José,
there were 224 (
7). Assemblages from communities around four
of the Alto Magdalena’s most impressive monumental centers,
however, yielded only four status-related artifacts for every
100,000 sherds, reflecting only the most modest differences
in standard of living between households. Higher-ranking families
in the Alto Magdalena did not live in much larger or more elaborate
houses, as they did in Oaxaca. Craft specialization was present
in the Alto Magdalena, but there were only 17 craft-related
artifacts for every 100,000 sherds, compared with 338 at Fábrica
San José and 972 at San José Mogote. The greater
heterogeneity of Oaxaca households, particularly with regard
to craft activities, would have involved substantially more
intense local interaction—the interaction that formed
Oaxaca’s compact early villages. At the local scale, the
Alto Magdalena presents a very different picture. It has been
argued (
1,
2) that Regional Classic occupation is not comprised
of separate compact villages at all, but rather of widely scattered
individual households, probably living directly adjacent to
the lands they cultivated. The weaker development of specialization
in the Alto Magdalena would have meant less intense local interaction,
and this, in turn, could have produced this pattern of very
dispersed residence without village formation.
|
Economic and Symbolic Hierarchies in Northeast China During the Fourth Millennium B.C. |
The local community pattern for the Hongshan culture area in
Northeast China is much more like that of the Valley of Oaxaca
than the Alto Magdalena. For a systematically surveyed zone
of 783 km
2 in the Chifeng region (
14,
15), it is easy to delineate
a village pattern similar to that of Oaxaca in Hongshan occupation
(4500–3000 B.C.) (
1,
2). Hongshan villages, like those
of Oaxaca, ranged from only a few households up to over 500
inhabitants. Unlike the lone large village at San José
Mogote, however, the Chifeng region contained at least a half-dozen
with populations over 200. No one stands out as dramatically
larger than all of the rest. Chifeng villages of all sizes,
though, were considerably less compact than their Oaxaca counterparts.
The Chifeng Hongshan village of Fushanzhuang, for example, had
some 35–40 households in an area of about 50 ha—a
clearly defined village community, but one spread across 10
times the area that a Oaxaca village of comparable population
might cover. Subtle differences in the assemblages of lithic
tools associated with different households at Fushanzhuang indicate
different kinds and degrees of specialization (
Fig. 6), but
the distinctions between households are less marked than those
in Oaxaca. Fushanzhuang yielded a total of 131 craft-related
artifacts for every 100,000 sherds recovered, a value nearly
eight times that for the Alto Magdalena but still well below
those calculated for Oaxaca communities. This evidence is entirely
consistent with the notion that Hongshan villages, like Oaxaca
villages, were products of specialized local economies. It also
appears that the more intensive interaction patterns that result
from a greater degree of economic interdependence between households
play a role in the much greater compactness of Oaxaca villages.
View larger version (18K):
[in this window]
[in a new window]
|
Fig. 6. Scaling of lithic assemblages from 19 Hongshan households at Fushanzhuang (Chifeng), based on proportions of formal lithic tools. Ordinary households carried out agricultural and craft activities; others focused especially heavily on farming, on initial stone tool production, or on tool finishing.
|
|
At Fushanzhuang, some households have higher proportions than
others do of ornamental luxury items, of fine serving vessels,
and of incised, painted, or burnished pottery. There is considerable
variation along a scale from low to high, as in Oaxaca villages.
Higher-status households are differentiated from lower-status
ones to a similar degree in both regions, but a much higher
proportion of Fushanzhuang households fall near the bottom of
the social scale. As a consequence, the overall abundance of
status-related artifacts at Fushanzhuang (at 39 per 100,000
sherds) is much less than that calculated for Oaxaca communities.
Most excavated Hongshan residences are less than 20 m
2; at least
one was over 100 m
2, but all are single-room structures and
none was raised on a platform. Hongshan households, then, display
differences in the material conditions of daily life, although
such disparities were greater in Oaxaca communities, and a larger
proportion of Oaxaca households enjoyed elevated standards of
living. By contrast, evidence for the expression in daily life
of differences in wealth or status is extremely scarce in the
Alto Magdalena.
Both economic specialization and social ranking, then, are easily recognized in the archeological evidence of village life at Fushanzhuang, although both are more conspicuous in Oaxaca communities. There is little sign at Fushanzhuang of the connection between economic specialization and social status seen in Oaxaca. Thus, of the three cases compared here, it is in the early chiefdoms of Oaxaca where both craft specialization and the economic aspect of social ranking are not only most highly developed but also most strongly linked to each other, and where local communities are most clearly defined and nucleated. Hongshan local communities, although easy to delineate, are less nucleated; both craft activity and differences in standards of living figure less prominently and seem unconnected to each other. In the Alto Magdalena, with its very dispersed residence pattern, craft specialization is still less developed, and the economic component of social differentiation is weakest.
A smoothed density plot of Hongshan occupation in Chifeng (Fig. 7) clearly shows a number of larger-scale communities. Most of the large villages like Fushanzhuang form peaks whose lower slopes incorporate other nearby villages into districts, along the lines of Oaxaca. The dozen or so districts have populations generally numbering in the hundreds in an area no more than 3 km across, making these chiefdoms the smallest, both demographically and spatially, of any of the three regions. There is no indication, however, of any chiefdom substantially larger than its potential rivals or of any sociopolitical integration at a scale larger than these districts. Ceremonial activity may have, once again, been an important element in district-scale social interaction. At Fushanzhuang, for example, there are remains of at least four ceremonial constructions. No such structures have been excavated in the Chifeng region, but they have been investigated in detail in other parts of the Hongshan culture area, as at Niuheliang (16, 17). They consist of square or circular earthen stone-faced platforms of varying dimensions, usually 1–2 m high and up to about 25 m across (Fig. 8), although larger examples are known. Hundreds of 1-m-tall painted pottery cylinders might be incorporated into the platform construction. These open-ended cylinders were not containers and only occur in platform contexts. Unbaked clay figurines, usually female, and sometimes pregnant, are occasionally found in platform fill. Ceremonial, or at least public, activities performed atop these platforms would have been visible to sizeable crowds assembled around them. Occasional examples are alone on the landscape, but they more often occur as complexes of several platforms.
View larger version (50K):
[in this window]
[in a new window]
|
Fig. 7. Smoothed plot of Hongshan occupation in Chifeng. Tall peaks represent larger communities, some "capturing" outlying small villages to form districts (red lines).
|
|
View larger version (90K):
[in this window]
[in a new window]
|
Fig. 8. Hongshan platform at Niuheliang (Left) and central burial with carved jades from another platform (Right). [Reprinted with permission from ref. 17 (Copyright 2004, Liaoning Sheng Wenwu Kaogu Yanjiusuo).]
|
|
Most platforms have a large stone slab tomb in the center, and
other tombs (up to a half-dozen) may be arranged around it.
These burials contain elaborate jade carvings (
Fig. 8) of birds,
turtles, insects, and other animals, as well as supernatural
themes such as pig-dragons (
18). Most Hongshan villagers were
buried without offerings in village cemeteries, but graves not
in platforms were sometimes lined with stone slabs or had artifacts.
At Nantaizi, for example, one burial contained jade earrings,
a shell ornament, and a stone chisel; and another, three bone
tools, a ceramic bowl, two incised ceramic jars, a grinding
stone, and a shovel, an axe, and a knife of stone (
19). Large
jade carvings, on the other hand, are almost exclusive to platform
burials, where they are typically the only kind of offering
included. These two different sets of burials (platform and
non-platform) both reflect hierarchical elements in Hongshan
society, but they trade in different currencies. Non-platform
burials, with their varying quantities of tools and ornaments,
seem connected to the variation in material standard of living
seen in Hongshan residential architecture and artifact assemblages.
Platform graves and their jades, on the other hand, are not
obviously related to the material conditions of life. Their
currency is not economic but symbolic (
20). The occupants of
these tombs were a special group of people, set off from the
rest of the population by their connection to the supernatural.
They were not just well-to-do villagers, but shamans or some
sort of leader with great spiritual power—like chiefs
are often thought to have. The distinction between two different
(although not necessarily unrelated) social hierarchies is clear.
|
Economic Hierarchy, Religion, and Politics |
Hongshan monuments share a number of characteristics with those
of the Alto Magdalena. Both were, in their very essence, monuments
dedicated to specific individuals, who were buried in the central
tombs, frequently surrounded by less central burials of other
important individuals. The sculptural elaboration of the Alto
Magdalena monumental burials, like the Hongshan jades, emphasizes
the supernatural. Offerings suggesting personal wealth are scarce
in Alto Magdalena tombs (
21), as in Hongshan platform burials.
Highly visible monuments in both regions, then, signal hierarchies
with a strongly similar symbolic basis. In each case, the presumably
ritual activities for which these monuments were constructed
were critical to the integration of supralocal chiefly communities.
Both Hongshan and Alto Magdalena burial monuments were the central
elements of public spaces suitable for the assembly of large
numbers of people. Architectural layouts were open, and access
to theses spaces was completely unrestricted. [The lone Hongshan
enclosed ritual space (
22) seems the exception that proves the
rule.] The entire population of a Hongshan or Alto Magdalena
chiefdom could easily have assembled in the level or sloping
space around the monumental tombs of important people. In addition,
wherever information about habitational remains is available,
these public spaces were fully integrated into the central residential
zones of their respective chiefdoms. These ritual activities
are the most readily identifiable special functions of central
places.
As we have already seen, in the Alto Magdalena this ritual hierarchy is not complemented by a distinct one of more secular character, as it is in the Hongshan case. Precisely such a secular hierarchy in Oaxaca, though, provides a very strong parallel to Hongshan organization. The modest amounts of utilitarian and ornamental objects that distinguish the burials of higher ranking people in Oaxaca recall not the Alto Magdalena burial mounds or the Hongshan platform tombs, but rather the non-platform burials of Hongshan villagers. It is in these burials, along with household evidence, that social hierarchy in Oaxaca finds its expression. Each of the two distinct hierarchies present in Hongshan society has a parallel in the other regions we discuss—the ritual burial hierarchy in the Alto Magdalena, and the more secular hierarchy in Oaxaca—but only Hongshan society contains both as distinct social phenomena.
While the secular character of Oaxaca’s social hierarchy is most prominent (in residential architecture, household activities, and burial practices), it is by no means disconnected from religious belief. Evidence of religious activity in Oaxaca comes in various forms. Small ceramic figurines are widely dispersed through household remains, suggesting an important role for household-level rituals. Communal ritual probably also played an important role in the formation and integration of social groups from the time of the earliest settled villages. In contrast to the principal Hongshan ritual spaces, or those of the Alto Magdalena, Oaxaca village ritual spaces were small and of restricted access (23). Initially there were single-room structures no more than 4 x 6 m whose architectural form made them conspicuously different from ordinary houses but which could not have accommodated more than a very small proportion of any village’s population. Later public buildings, sometimes raised on platforms 1 or 2 m high, were also small and enclosed. Such buildings sometimes occurred in groups, on the four sides of unroofed rectangular patios. Although these patios were larger than the spaces inside the buildings themselves, they were still much smaller than the plaza areas adjacent to Hongshan or Alto Magdalena burial monuments, and could by no means incorporate entire district (or even local) populations. Admittance to space was highly controlled, since even the unroofed patios could be entered only through restricted openings at their corners. In all three regions, then, public or ritual spaces were created to bring groups of people together, but only in Oaxaca do these spaces seem explicitly designed to set apart these assembled groups from the rest of the population by excluding nonparticipants.
These simultaneously integrative and exclusionary ritual structures were spatially associated with elite residential areas at San José Mogote, and the participation of high-ranking people in this sort of ritual activity surely contributed to the process of social differentiation in Oaxaca. This association culminates in the late Rosario phase conversion of a temple structure at San José Mogote into an unprecedented stone and adobe multiroom elite residence on a platform 22 x 28 m (Fig. 9). Neither Hongshan nor Alto Magdalena elites ever enjoyed a standard of living so far above that of their lower-ranking contemporaries. The strongly economic character of social differentiation in Oaxaca is as clear as its reinforcement in ritual is unmistakable. Hierarchy in Oaxaca has its center of gravity, and perhaps its origins, in the economics of local communities, but it is clearly also strengthened through ritual, especially in its supralocal projection. The relief carving of a naked bleeding body (Fig. 5), walked on by those entering the patio adjacent to the residence of the San José Mogote district’s presumed ruler at the end of the Rosario phase, appears to be a high-ranking person from another community captured and killed by San José Mogote (9, 24, 25). Burned structures at San José Mogote provide additional indications of intervillage warfare and raiding. Supralocal chiefdom organization in Oaxaca, then, may have been a politico-ceremonial structure of relationships between elites at San José Mogote and preexisting communities—relationships that often turned violent as San José Mogote sought to dominate them. This regional-scale elite structure was built on a foundation of economic interdependence and differentiation within local communities. Thus, the unitary and fundamentally economic hierarchy of Oaxaca chiefdoms was laced tightly together with religious belief and ritual, especially insofar as supralocal politics were concerned.
View larger version (110K):
[in this window]
[in a new window]
|
Fig. 9. Stone masonry platform at San José Mogote built to support a temple; after the temple was burned, its place was taken by an elaborate multiroom residence.
|
|
|
Supralocal Organization and Conflict |
The twin hierarchies of Hongshan society (symbolic and economic)
represent a rather different dynamic in the relationship between
ritual, economy, and politics, and Hongshan districts did not
form in the same way the San José Mogote district did.
The comparatively rapid development of craft specialization
and social differentiation in Oaxaca during 300 or 400 years
produced a growing number of largely redundant independent village
communities, usually a few kilometers apart, each with strong
economically based internal integration and social hierarchy.
Valley-wide politics consisted of relationships, eventually
competitive, between the elites of these clearly defined village
units. San José Mogote’s early demographic edge
finally proved insuperable, as its elite progressively dominated
nearby villages until its domain included well over half the
valley’s population. Other chiefdoms were so much smaller
that, although raiding from them could have been a thorn in
San José Mogote’s side, by the Rosario phase there
was no real possibility of their dominating San José
Mogote militarily. This entire process of amalgamation of most
of the other local communities in the valley into the San José
Mogote chiefdom occurred during some 700 years of demographic
stability; it cannot have been driven by regional population
growth.
The Hongshan districts of the Chifeng region grew by a very different process. Craft specialization and social differentiation emerged only very slowly, after 1,500 years or more of sedentary living before the Hongshan period. Even then, village-level economic differentiation and integration were less strong than in Oaxaca. As chiefly districts formed in Chifeng, there was no amalgamation of preexisting villages. Instead, as villages grew, their relatively low level of internal interaction facilitated fissioning, producing multiple local communities within a very few kilometers of each other. It was these sets of mother and daughter communities that comprised Chifeng’s spatially compact chiefly districts. Growing population, then, was at the heart of the formation of Hongshan chiefdoms. It is difficult to tell whether most of the demographic change occurred rapidly in short episodes, because the periods that provide chronological control in Chifeng are so long, but it is clear that it took the cumulative demographic growth of some 3,000 years after the establishment of sedentary living to arrive at the full development of Hongshan chiefly districts. No one of these districts became very much larger than any other, nor is there any indication that any Hongshan district ever subjugated another one. (And this degree of political fragmentation continued to characterize the Chifeng region for another 2,500 years after the end of the Hongshan period as well.)
The resolutely small scale and conservatism of supralocal politics in Chifeng might be attributed to the separation between symbolic and economic hierarchies. Some of the pathways to consolidation of power and its supralocal projection were blocked to either of the Hongshan elites because of the existence of the other, potentially rival, hierarchy. The Hongshan archeological record has thus far not produced iconography of warfare or violence, or evidence of widespread burning, of weapons, or of clear-cut fortifications. The ditches often found at Chinese Neolithic village sites have sometimes been seen as fortifications, but they occur frequently in the absence of evidence for social hierarchy, so the two known instances for Hongshan period sites are difficult to connect to supralocal elite hostilities. Among these less cohesively led communities with less-developed economic integration, there was less potential for the emergence of the kind of conflict between the elites of different communities that provided the milieu in which supralocal political integration was effected in Oaxaca. By the end of the Rosario phase, San José Mogote’s economic elites had consolidated a degree of secular political power supported by religious ritual that enabled them to dominate a much larger territory than that of any Hongshan district in Chifeng. (And they were just on the threshold of 500 years of remarkable demographic, social, political, and economic change that culminated in the city of Monte Albán with tens of thousands of inhabitants—the capital of a conquest empire of well over 10,000 km2.)
In contrast to the Hongshan social ranking system, there was no alternate hierarchy as a potential rival to the supernatural authority of Alto Magdalena elites. As in Oaxaca, there is evidence that Alto Magdalena elites used violence in the projection of their position. Sculpted figures accompanying elite burials sometimes carry weapons or even apparent trophy heads. Evidence of violent conflict, however, is not abundant in the remains of daily life from the Alto Magdalena. Flaked stone points, for example, like those on spears depicted in statues are well preserved in the archeological record, but they are extraordinarily rare in Alto Magdalena lithic assemblages (less than 1 per 100,000 sherds). Alto Magdalena occupation does not favor defensible locations; its dispersal does not lend itself to easy defense; and no fortifications are known. Violent conflict in both Oaxaca and the Alto Magdalena, then, seems political, even though the elites engaged in it owed their positions largely to the economics of daily life in Oaxaca and to the supernatural in the Alto Magdalena. Although conflict in chiefdom development has often been attributed to population pressure and scarcity of resources, the two regions with the most evidence of violence have the highest and the lowest population densities (20–40 persons per km2 for the Regional Classic in the Alto Magdalena, and not much more than 1 person per km2 for the Rosario phase in the Valley of Oaxaca). Chifeng, with little evidence of violent conflict, had an intermediate population density (5–10 persons per km2). Not even the high population density in the Alto Magdalena created serious pressure on resources, and resource scarcity cannot have driven the forcible domination of neighboring communities in Oaxaca, given its extremely low population density (Fig. 10).
------分隔线----------------------------
