Barreto C.-Social - Complexity - in - Ancient - Amerindian
Barreto C.-Social - Complexity - in - Ancient - Amerindian
Barreto C.-Social - Complexity - in - Ancient - Amerindian
Cristiana Barreto
Models of social evolution, which rely on typologies for characterizing the degree
and structure of inequality and social complexity of past societies, together with
its corresponding trait lists of typical features and archaeological signatures, have
been largely used in South America to both describe and explain the emergence and
decadence of a variety of past social formations, ranging from the Inca empire, to
smaller polities which have been classified as tribes, chiefdoms, and states.
Study of the prehistoric South American lowlands has been fatally marked by the
longtime assumed absence of complex societies and the eternally restated contrast
with Mesoamerican and the Andean complex civilizations, powerful states, and vast
empires. Indeed, the view that the lowlands is one part of the world where evolution
has maintained societies in a relative state of social egalitarianism has been perpet-
uated in many ways and has served different philosophical ideals, from eighteenth
century colonial romanticism to 1960’s Latin American Marxism.
During the last decades, results of archaeological, ethnohistorical, and ethno-
graphic research have gradually changed the long-held view of native SouthAmerican
lowland societies as just simple and egalitarian. The greater time-depth perspective
provided by both ethnohistory and archaeology reveals now that such views were
strongly inspired by the study of present living native populations highly affected
by centuries of contact. Although it is now clear that the present cannot be taken as
examples of past social developments, we still know little in terms of how different
these past social developments were, especially in terms of social complexity. Fur-
thermore, the little we know seems to challenge previous models based on intensive
agriculture and population growth leading to complex forms of social organization.
C. Barreto ()
Museu de Arqueologia e Etnologia,
Universidade de São Paulo, São Paulo, Brazil
e-mail: [email protected]
On the other hand, the past decades have witnessed a dramatic rethinking of the
nature of the so-called egalitarian societies (Flanagan 1989; Paynter and McGuire
1991). Although ethnographers of Brazilian native populations have long recognized
that a narrow focus on tribal social organization has masked different kinds of so-
cial stratification, hierarchy, and inequality (Basso 1973; Carneiro 1993; Crocker
1979; Gregor 1990; Ireland 1986, 1996; Kracke 1978, 1993; Turner 1992; Werner
1981, 1982), we also know a lot more about the mechanisms which prevent lowland
Amerindian societies to turn into stratified states (Clastres 1979; Sztutman 2012;
Viveiros de Castro, 2008). Ideas about just what are “complex” and “hierarchical”
societies as opposed to “simple” and “egalitarian” have also moved away from tra-
ditional key concepts of techno-economic control and social prestige towards other,
perhaps subtler, forms of power and control, as well as less visible ways to insti-
tutionalize and perpetuate ideological systems (Heckenberger 2005). Even though
no strong states or vast empires have been revealed by new data, it is now widely
accepted that, at least in some parts of the lowlands, polities of considerable dura-
tion, formalized leadership, and hierarchical political structures, sometimes referred
to as chiefdoms, have indeed emerged along the lengthy history of South American
lowlands (Carneiro 1995; Heckenberger et al. 1999; Neves 1998, 2008a; Porro 1994;
Roosevelt 1999; Viveiros de Castro 1996; Whitehead 1996; Schaan 2004).
Some have argued that these chiefly societies were particularly common world-
wide in 1492 (Mann 1986) and that in the South American lowlands, as elsewhere,
such powers were based on notions rooted in “divine” ancestry values and hierarchical
sociality according to which persons were ranked relatively to one another (Heck-
enberger 2005). However, we are still lacking substantial archaeological evidence to
understand the dynamics of power and politics of such chiefdoms.
We do know that when the Portuguese arrived on the coast of Brazil in 1500
they encountered most of the territory occupied by Tupian groups organized into
chiefdom-like polities. Ethnohistorians and archaeologists have argued that from
all precolonial populations of lowlands Brazil, these, the Tupians from the coast,
were probably the only ones in a process of state formation, a process that was
brutally interrupted by the arrival of Europeans. Yet, besides living in large, settled,
agricultural villages, they apparently lacked the types of rigidly hierarchical social
structures commonly attributed to chiefdoms in many other areas (Clastres 1979;
Fausto 1992). In this case, we shall never know if state formation was to occur in fact.
Nor shall we know whether it was just a matter of time or if social development would
have taken another course. We can know, however, about other social developments in
the lowlands that were not affected by the conquest, which sequences can be studied
as “whole,” and about which we can ask ourselves whether and why formalized
political and social inequality, hierarchy and complexity ever happened.
From the start, it has to be said that from a large variety of reconstructed archaeo-
logical sequences in the Brazilian lowlands, covering a period of at least 12,000 years
of cultural developments, in quite diverse natural and historical contexts, the great
majority of such developments have not reached high levels of social complexity, in
the sense of having many vertical levels of political representation, leadership, and
authority. Rather, even among societies which have an extremely hierarchical ethos
Social Complexity in Ancient Amerindian Societies 3
Throughout the twentieth century, models of social evolution advanced for the South
American lowlands were largely based on environmental factors as either limit-
ing or promoting the local development of more complex societies. Regardless of
whether models explain the absence or the presence of societies more hierarchical
than the ones presently known in the region, debates have been consistently or-
ganized around the twofold argument of whether environmental conditions allowed
increasing sedentism and population growth leading to more complex forms of social
organization.
4 C. Barreto
Since Steward tagged Amazonian populations under his “tropical forest culture
type” (Steward 1946) not only has cultural development in this region been explained
mainly on the basis of ecological conditions, but also its implications have been
generalized to all of the South American lowlands, including nontropical forested
regions.
The “tropical forest culture type” model—now also called the “standard model”
(Neves 1998; Viveiros de Castro 1996)—assumed both environmental uniformity
across the region and a limited agriculture and protein capture potential to argue
that, in Amazonia, environmental conditions led populations to organize into small,
autonomous, semisedentary settlements across the entire region. Supposedly, a uni-
form environment, with an even distribution of resources discouraged any form of
competition, economic specialization, and stratification, whereas the low carrying
capacity inhibited the more formalized systems of social and political control that
usually emerge out of dense population aggregates.
Since the 1950s, this tropical forest culture type model led the few archaeolo-
gists working in Amazonia to interpret any indication of more complex societies
in the basin as an intrusion from the highlands (Meggers 1954, 1971, 1992, 1995;
Meggers and Evans 1957). In order to support this interpretation, attention focused
in both environmental studies and cultural diffusion with little concern to improve
understanding about the nature of such societies.
While data gathering has been consistently directed to explain larger and denser
sites as settlement superposition and to correlate archaeological sequences to en-
vironmental episodes, little has been advanced about the scale and organization of
settlements believed to belong to more complex societies (such as those of Marajó
and Santarém culture complexes, to cite only those admittedly to be more complex
than the natural course of local social evolution would allow). Moreover, if both
settlement and ceramic attributes indicated some degree of social stratification, other
topics, such as labor mobilization, craft specialization, and differential participation
in large trading networks, were not subject of archaeological research.
Over the past two decades the “standard model” has been going through a slow
and steady decline due to developments in ethnohistory, historical ecology, and
archaeology. The contrasting position that has gradually taken shape suggests that,
at least in some areas, large, fully sedentary villages were not only once present
in Amazonia but also were densely clustered along some riverine areas (Beckerman
1979, 1991; Carneiro 1986, 1995; DeBoer 1981; Denevan 1976, 1996; Heckenberger
et al. 1999, 2001; Lathrap 1970, 1977; Myers 1973, 1992; Porro 1994; Roosevelt
1980, 1981, 1991, 1993, 1994, 1999; Whitehead 1994, 1996; Wüst and Barreto
1999; Petersen et al. 2001; Neves et al. 2003; Neves and Petersen 2006; Schaan
2008). Most scholars have again focused on ecological determinants, only this time
environmental abundance and high levels of productivity (instead of scarcity) become
the key to explain large and dense settlements, and environmental diversity (instead
of uniformity) is the key to explain competition leading to population pressure and
stratification. While, again, a great deal of research has attempted to pin down the
subsistence basis of such dense populations, and to invert the diffusion arrows now
indicating the indigenous development of social complexity, little attention was
focused on the nature and scale of such complexity.
Social Complexity in Ancient Amerindian Societies 5
Lathrap (1970) was certainly the first to question the standard model suggesting
that while swidden agriculture did constitute an obstacle to population growth, the
floodplain areas of the Middle and Lower Amazon provided not only conditions for
a more “efficient” agriculture but also fishing resources that could support relatively
dense populations for long periods of time. Arguing for an indigenous development of
complex societies in the Middle Amazon, Lathrap inverted the direction of diffusion
arrows.
Significant ethnographic data were gathered by Carneiro (1985) among the
Kuikuro in the Upper Xingu, an inland interfluvial border of the Amazon basin,
suggesting that not only a fishing and manioc subsistence could support larger pop-
ulations but also that if more complex societies had not developed in this area it was
only because no social circumscription and population pressure had yet occurred.
Roosevelt has first argued for the complexity of Amazonian chiefdoms based
on the possibility of developing intensive agriculture, and specifically maize, in at
least some parts of the Amazon valley, only to revise her model after excavations in
Marajó island led her to believe that it was the particularly rich and diverse environ-
mental resources of the regions that allowed for complex chiefdoms to arise out of an
economy based on a mixed system of cultivation and foraging, including long-term
garden plantings and intensive fishing (Roosevelt 1999, p. 22). Schaan was able to
confirm that, indeed, Marajó chiefdoms grew out of intensive fishing, arguing that
once local kin groups obtained control over the surplus produced by built fisheries,
both population growth and competition fostered the process of increasing hierarchy
and social complexity (Schaan 2004, pp. 41–44).
More recently, a third approach to understand the dynamics of political organi-
zation, competition and conflict over resources, and its effects on different social
organization modes has emerged out of advances in both the historical ecology
of tropical forest, focusing on resource management by contemporary indigenous
groups, and from archaeological research in sites with “terra preta de indio” (indige-
nous black earth) and other earth features such as canals, mounds, and pathways.
The historical ecology perspective (Balée 1994) shows that indigenous populations
not only act on the environment so as to change the local conditions but also oc-
cupy and develop adaptive strategies to live on previously transformed places by
human activity. These strategies can be seen among very mobile hunter gatherers
(Politis 1996; Rival 2002) and sedentary agriculturalists (Balée 1989, 1995; Balée
and Moore 1994; Posey 1986).
The archaeological research proposes that much of the diverse sociopolitical dy-
namics behind dense and continuous occupations of particular sites along theAmazon
floodplain were not so much based on either food production such as in intensive
agricultural systems, nor on the control of, or access to, specific resources such as
fish, but strongly centered on the selection and management of resource diversity
and built landscape (Neves 2008a, b; Moraes and Neves 2012).
In sum, in the last decades we have witnessed debates about just which part of
Amazonia could have supported larger and denser populations, floodplains or hinter-
lands; about the feasibility of large populations being supported by shifting manioc
systems (Brochado 1980; Carneiro 1983, 1985; Heckenberger 1998b); whether more
intense agriculture of more protein-balanced products (i.e., maize) was possible
6 C. Barreto
(Roosevelt 1980, 1981); whether fishing could have been a sufficient source of pro-
tein in an otherwise protein-poor environment (Beckerman 1979, 1994; Carneiro
1986; Gross 1975; Schaan 2004, 2008); and that diversification of resources and
forest management caused cycles of population growth and political competition,
conflict and control over certain territories.
In one way or another, all of these arguments rely on a presumed causal relation-
ship between environmental conditions, be they natural or anthropogenic, population
growth, and increasing complexity. While the relationship between the environmen-
tal carrying capacity and the presence of large and dense settlements evolving in the
Amazonia have been greatly stressed (although without definite results—except for
Marajó, we still do not know what these dense settlements lived on), the second part
of the twofold argument has been left unexplored. Just how did population growth,
or population pressure led to the emergence of more complex forms of social orga-
nization? This question will require more than just theories based on old ecological
models.
Criticisms on the more general perspectives and causal theories relying either
on resource abundance and population density or on population pressure/resource
stress to explain the emergence of institutionalized inequality (such as those found in
Drennan 1987; Feinman 1995; Upham 1990) seem to have been ignored as debates
on Amazonian developments became more and more entangled on competing argu-
ments. Also ignored were criticisms about this kind of environmental determinism
being extrapolated to other areas of the lowlands (Moran 1995; Wüst 1994).
Again, other factors, usually considered in the examination of social development
and inequality elsewhere, such as social organization, resource control, labor mo-
bilization, craft specialization, trade, and networks (as approached in the volumes
edited by Erenreich et al. 1995; and Bacus and Lucero 1999) continue to be largely
unexplored in the lowlands. Furthermore, the specific environmental conditions tied
up to these revisionist models are narrowed to those of the Amazonian lowlands as
for its potential to support large communities, also overlooking conditions of other
environmental provinces in the Brazilian lowlands.
The cases presented below are an attempt to add new data to discussions about
social development in the lowlands in order to seek alternative models to the me-
chanical use of the twofold ecological determinants/population growth argument
and to improve knowledge about nature and scale of intermediate societies that have
evolved in the Brazilian lowlands.
In choosing to present data on cultural developments from three environmentally
different regions of Brazil, this chapter proposes to move discussion beyond the
strong environmental determinism that has shaped Amazonian archaeology and has
extrapolated to the South American lowlands in general, and turn archaeologists
attention to factors other than environment to explain different degrees of social
complexity reached by native Brazilian societies.
The three cases (Fig. 1) herein presented are (1) the shell mound culture complex
of southeastern Brazilian coast, as a strong example of nonagrarian societies that
developed out of extremely rich and stable natural resource conditions, and whose
Social Complexity in Ancient Amerindian Societies 7
Fig. 1 Map of Brazilian lowlands showing location of the three areas focused in the text: the
southeastern coast, where most sambaqui sites are located; the cerrado area, where most central
Brazilian ring villages are found; and the lower Rio Negro along the Amazonian varzea
Although shell mounds occur along almost the entire coast of Brazil, the southeastern
shore presents a particular situation in that a number of very dense concentrations of
this type of site occupy the rich estuarine areas that punctuate an otherwise narrow
strip of coast. Protected bays and lagoons edged by vast mangroves provide a high
nutrient influx and one of the highest productivity rates among worldwide marine
ecosystems. Especially rich in shellfish and fish, these areas have attracted the settling
of human groups specialized in shellfish gathering and fishing since at least 6,000
BP (it is possible that older sites have been destroyed by sea level fluctuations).
Probably due to expansion of horticulturalist groups from the mainland, these large
mounds stopped being built around 2,000 BP and much smaller deposits, including
shell mounds, took their place (Fig. 2).
Shell mounds (or sambaquis, the Tupian word for it) are the results of intentional
accumulation of faunal remains, notably shells, and also of a variety of stone and
bone artifacts. These shell mounds vary in size and structure, from small 2-m-high
heaps of shell-covered sandy dunes to 400-m-long and 30-m-high impressive mounds
containing artifacts, hearths, and burials disposed in a quite complex sequence of
layers. In fact, sites larger than 2,000 m3 are known in the southern shores, where
their monumental structures stand out in the open coastal landscape (De Blasis et al.
1999; Gaspar 1998).
For decades, sambaquis were believed to be the result of the gradual accumulation
of food remains by small, nomadic shellfish-gathering groups and many archaeo-
logical indicators of possibly more complex sort of social organization have been
overlooked. Although site size, and specially the volume of shells moved to form
those mounds were quite impressive and had indeed inspired early interpretations
of sambaquis as built monuments requiring organized labor mobilization (Duarte
1968), no data had been produced on the rate and nature of mound formation. The
same can be said for other indicators of social stratification and inequality such as
differentiated burials as seen in burial practices and associated burial paraphernalia,
differential access to prestige goods as attested by the distribution of exotic materi-
als; craft specialization as seen in the production and distribution of stone and bone
sculptures; and site hierarchies as site spatial distribution and size variability sug-
gest. Instead, a stress on ecological determinants had led to interpret faunal remains
exclusively as food remains and to a chronological ordering of sites into an evolu-
tionary scale in which the gathering of mollusks predominated first and later more
intensive fishing substituted for it just before sambaquis stopped being built (Dias
1972; Mendonça de Souza 1981; Perota 1974; Lima 1991). This ordering of sites
inspired archaeologists to think that after a 4,000-year-long period of shell-mound
building in estuarine areas, population growth caused depletion of mollusk banks
and led, first, to a diversification of gathered species (explaining the remains of less
nutritious shellfish species in mounds), and second, to an increase and specialization
in fishing, explaining why shell mounds became less imposing in size and finally
stopped being built (see Figuti 1993 for a review).
Social Complexity in Ancient Amerindian Societies 9
cemeteries and offering some clues for estimating population density. One large
sambaqui (Jabuticabeira II) is estimated to have ca. 43,000 buried individuals over
an estimated period of 400 years, while another (Cabeçuda) is estimated to have
more than 50,000 individuals. Preliminary estimates of number of buried individuals
per 25-year period generations are surprisingly high and indicate communities with
an unexpected high population density (Fish et al. 2001; De Blasis et al. 1999).
Other smaller mounds seem to have served as both residential and funerary areas,
with individuals buried under their houses, which led scholars to hypothesize some
sort of privileged status for those ritually buried in the large cemeteries (De Blasis
et al. 1999). The burials seem to provide the strongest evidence of social inequality
among sambaqui people: while some individual burials are simpler, others display
rich and diversified paraphernalia such as the elaborated and rare stone and bone
sculptures.
The degree of social differentiation and inequality that sambaqui builders have
reached and how this evolved along their developmental sequence have yet to be
determined and much has to be done to elucidate the forms of labor mobilization
involved in their construction, and what the observed elaboration of burial structures
and intensification of rituals may mean in terms of formalizing social inequality and
complexity. In other words, whether sambaqui culture will enter the hall of the few
nonagricultural societies to display a relatively high, formalized degree of social
inequality, only further research will tell.
The point we want to make with this case is that the stress on environmental vari-
ables, as either transposing the limiting-carrying capacity model from Amazonia or
viewing resource abundance as a stability maintenance factor, tended to mask impor-
tant indicators of the nature of sambaqui social organization. Alternative approaches
focusing on the nature of such sites (how they were built, for which purposes, etc.)
and their scale (how big they are, how complex are internal structures, how labor
Social Complexity in Ancient Amerindian Societies 11
demanding they are, etc.) will probably be more enlightening to understand sam-
baqui social evolution than ecological deterministic approaches limited to explain
its development in terms of subsistence practices, population growth, and carrying
capacities.
The ring villages of Central Brazil are relatively large settlements that have rapidly
spread across this immense region of the Brazilian Plateau around 800 AD. These
villages, in their circular, oval, or horseshoe arrangements of houses around a central
patio, correspond to a pattern of village organization still present in Central Brazil
among most Ge and Bororo speaking groups.
Ethnographic studies of these groups (in a context already highly affected by direct
contact with white colonizers) first led scholars to classify them as hunter/gatherer
marginal populations confined to the poorer and drier lands of plateaus that separate
the Amazon basin from other southern drainages. Indeed, the distribution of these
archaeological villages correspond to the savanna like “cerrado” and “caatinga” eco-
logical provinces of Brazil where agriculture conditions are limited by both poor
nutrient soils and long dry seasons. This marginal status drew the attention of ar-
chaeologists away from Central Brazil, especially those interested in studying more
complex societies.
However, surviving Central Brazilian groups did attract the attention of early
ethnographers for both the large size of their settlements (compared to the small
tropical forest villages) and the unusual complexity of village organization as wit-
nessed by a high number of social institutions, residential rules, corporate groups,
and rituals (Nimuendajú 1939, 1942, 1946; Levi-Strauss 1952; Lowie 1949).
But, if both size and complexity were considered anomalous, their nonstratified,
egalitarian status was not questioned. On the contrary, their typical village layout
was also taken to express the dichotomy between a very complex social system and
the egalitarian tribal structure. Houses (the households of extended families) are laid
out along the circle as to express many of the social partitions to which residents
belong (such as moieties, clans, and lineages), but at the same time all houses have
equal access to the central plaza where most public, ceremonial activities take place
and where most decisions about the village are taken (Maybury-Lewis 1979; Costa
and Malhano 1986; Barreto 2011).
Recent ethnography on social and political organization of Central Brazilian
groups now argues that inequality is latent to the typical form of village organi-
zation of Central Brazilian groups (even seeing the central plaza as an arena of
political control), and that social hierarchies are at the essence of the Central Brazil-
ian ethos (Crocker 1979; Heckenberger 1998a, 2005; Ireland 1996; Turner 1992;
Werner 1981, 1982). However, most studies touching on the dialectics of control and
the nature of chiefly power tend to agree that the critical transition from informal and
circumstantial forms of control into institutionalized, hereditary systems of power
12 C. Barreto
Fig. 4 Plan of site GO-RV-66, one of the largest archaeological villages in Central Brazil with a
total of ninety houses. (From Wüst and Barreto 1999)
was not reached among these groups for no other reason than the low population
densities and relatively high degree of mobility they display today (Spencer 1994;
Werner 1982).
Archaeological research, with a different time depth perspective, now provides
a quite different scenario for explaining both social complexity and hierarchical
structures among Central Brazilian groups. First, it is now clear that from 800 AD
onwards, throughout Central Brazil, population was organized into numerous, large,
fully sedentary, agricultural villages despite poor conditions for agriculture (Fig. 4).
Archaeological ceramics suggest varied use of processed and stored manioc and
maize products that could have supported dense settlements year round. Population
estimates based on number of houses per village and ethnographic data reveal that
communities as large as 2,000 people organized in villages formed by several rings
of houses existed in the past (Barreto 1996; Wüst 1990; Wüst and Barreto 1999;
Wüst and Carvalho 1996). Thus, models based on present village sizes and mobility
do not hold for the past.
Social Complexity in Ancient Amerindian Societies 13
Fig. 5 Graph of distribution of number of villages (in %) by village area (in hectares) from a total
of 65 surveyed ethnographic and archaeological ring villages (Uru and Aratu traditions) known in
Central Brazil. (From Wüst and Barreto 1999)
If both shell mounds and ring villages seem to represent borderline cases in which
only further research will tell whether and to which degree social/political hierarchies
and economic inequality were institutionalized at some point in the past, there has
been increasing agreement that, at least in some areas of riverine Amazonia, large and
dense settlements organized into well-established long-lived chiefdoms did exist.
Much of the data available has come from the Central Amazon Project coordinated
by Eduardo Neves since 1995, which originally focused in the area of confluence of
the Negro and Solimões rivers. First, regional surveys have produced a first scenario
for settlement variability based on black earth (terra preta) extensions (Neves et al.
2003; Neves and Petersen 2006). The distribution of small- and medium-sized sites
(ranging from 2 to 10 ha) and a few large sites (some with over 30 ha) in the Manaus
area give us now an idea of the magnitude of settlement size ranking. What this
settlement variability means in terms of settlement hierarchies and centralization of
Social Complexity in Ancient Amerindian Societies 15
power is still to be explored. However, it confirms the existence of large and dense
settlements that could be part of a chiefdom system. Comparisons with other areas
such as Central Brazil does seem to suggest that here not only more stable paramount
villages were in order, but also that these places were successively occupied by
different groups along the history of the area. Second, closer examination of one of
these large villages, reveals a well-structured plan displaying earthworks (ditches,
terraces, and mounds) organized around a large central plaza (of approximately
450 × 100 m). Radiocarbon dates also indicate that such structural elaboration of
the site was gradually built over a long period of time (more than 1,000 years) and
that fixed occupation and constant additions started at least as early as 150 AD.
The significance of these structural features is not yet fully understood, but ini-
tial interpretations based on distribution of archaeological materials such as prestige
items and more elaborated goods have led researchers to define special activities
linked to communal ritual and public affairs concentrated near the village cen-
ter and plaza, and to interpret the central plaza as a “politico-religious center of
community life” surrounded by special residential areas for “limited segments of
society”(Heckenberger et al.1999, p. 21). It is unclear whether such village layout
could be generalized as a pattern of paramount villages. Further data analysis on set-
tlement pattern in this area will be crucial to start defining the nature of concentration
of power, social structures, and scale of settlement hierarchies in Amazonia.
More interesting, however, is the variability found in the area concerning the
size, composition, and formation rate of sites. Other large sites, such as the mul-
ticomponent site of Lago Grande (Neves and Petersen 2006), show different rates
of formation corresponding to different phases of occupation, and perhaps different
groups. From around 400 AD till 900 AD, villages with central plazas and mounds
display thick layers of terra preta, correlated with a dramatic increase in site size and
density, signaling population increase and possibly multiethnic occupations (Neves
2008a; Moraes and Neves 2012). These thick and dense sites, related to Manacapuru
and Paradão ceramic complexes, seem to have been formed in only a few decades,
indicating a peak of occupation lasting about 500 years, only to be followed by a
rapid abandonment of the site, apparently related to warfare and defense activities,
and in some cases reoccupation, by groups related to the Amazonian polychrome
tradition. This subsequent occupation is organized around much smaller areas during
shorter periods of time.
These data bring back the question of reoccupation of sites as formerly proposed
by Meggers to explain large-sized, multicomponential settlements. Only here, it is
clear that reoccupation occurs within another timing and scale, i.e., as processes of
rapid and intensive reallocation of very dense populations.
Other than the formation of thick layers of terra preta, subtler forms of landscape
transformation include the constructions of funerary mounds, such as at the Hatahara
site, another large multicomponential site excavated in the area, reaching an extension
of 16 ha. Here, a great number of mounds built over a relatively short period of time
seems to indicate that mound building occurred into few but intensive episodes,
showing again multiple and short periods of dense occupation and transformation of
the landscape (Neves and Petersen 2006; Machado 2003).
16 C. Barreto
The sudden processes of site formation and site abandonment have led archaeol-
ogists to suggest “that chiefdoms in the Central Amazon were cyclical or centrifugal
social formations, characterized by alternate processes of centralization and decen-
tralization. Centralization, in the archaeological record of the CentralAmazon, would
be seen through the occupation of very large sites, some of them several hectares wide.
Decentralization, on the other hand, would be verified in the sudden abandonment
of some of those large sites” (Neves and Petersen 2006, p. 301).
While site abandonment in the Amazon has been formerly explained as the result
of adaptive problems (Meggers 1996), the new data seem to point to far more com-
plex processes due to competition and political conflict, causing frequent settlement
fractioning. “Such conflicts would emerge from a constant tension between, at one
hand, centralizing centripetal hierarchical ideologies—verified in the archaeological
record in, for instance, labor mobilization in mound building activities—and, at the
other hand, centrifugal pulverized and uncontrollable household-based productive
units” (Neves and Petersen 2006, p. 302).
In conclusion, these data seem to contradict models of Amazonian cultural de-
velopment as shaped by ecological constraints and which tend to explain large
settlements in the region as a consequence of repeated occupations of favored loca-
tions (Meggers 1990; Miller 1992). Although this view had already been questioned,
neither settlement pattern nor site structure data were yet available, limiting scrutiny
to reviews of ceramic seriations which inspired previous interpretations on settlement
sizes and mobility (DeBoer et al. 1996; Heckenberger et al. 2001).
Because it specifies settlement variability and organization, these different kind of
data can also provide the important link to the second part of the twofold argument that
relates conditions for development of dense and large settlements, and the existence
of such settlements to higher degrees of social complexity and inequality.
Within the present context of models of social evolution in Amazonia, whether
increased population density can be seen as a causal variable in changing complexity
in Amazonia (be it as in Carneiro’s model where competition for land leads to more
centralized organization, or be it in Roosevelt’s model where the need for more
intense use of land brings more complex social organization), the data from lower
Negro river can throw new light on the role of population in organizational changes.
Furthermore, looking at variability in scale and nature of such settlements in
Amazonia will probably also open up interpretation for less monocausal models to
include other factors such as specialized production, labor mobilization, warfare and
regional networks in the analysis of societal hierarchies and centralized polities as
described in the ethnohistorical literature.
Finally, the Central Amazon data also show that even though places were, in fact,
being reoccupied by successive and distinct human groups attracted by the managed
landscape favoring diversity of resources, appropriation of space and resources were
done in culturally distinct ways.
Social Complexity in Ancient Amerindian Societies 17
Conclusion
Studies of the three cases presented above, southern coastal shell mounds, Central
Brazil ring villages, and Central Amazon settlements, have yet to be refined in terms
of definite developmental sequences from which one can infer either the exact direc-
tion of changes towards social hierarchy and inequality or the degree to which these
were formalized in the past. However, in the three cases, it is the more direct focus on
the reconstruction of settlement size, organization, and their specific developmental
sequences which allow an initial evaluation of the longtime assumed relationship
between large, dense population aggregates and social complexity and inequality.
They show that determining the ecological basis from which such societies evolved
is certainly crucial, but not sufficient to understand the development of social differ-
entiation and inequality in the South American lowlands. Looking at how economies
and labor were organized in diverse environmental conditions can help to discern pre-
cisely the validity of and specify the particular circumstances in which the longtime
assumed relationship between ecological conditions, sedentism/population growth,
and societal change apply within the broad region. In particular, they point to the
need to define the role of population (and the relevance of different demographic
variables) to explain changes towards more complex societies.
The data presented above also point to a few factors whose role in the direction
of increasing social complexity and formalizing inequality should be considered
in further research, such as labor mobilization as strategies of emerging political
control, the intensification of rituals as shaping social hierarchies and leadership, and
the formation of supralocal spheres of interaction (through trade, marriage, warfare
alliances) as the basis for control over prestige goods and wealth.
Finally, the three cases presented above illustrate distinct pathways to perhaps
equally distinct degrees of social complexity and centralization of power and that
not one generalizing model for the broad region can be sustained. It is the comparison
of specific sequences within the region, and the resulting broad parallels and contrasts
that will deepen understanding of how social complexity and inequality has emerged
in the region.
References
Bacus, E. A., & Lucero L. J. (Eds.). (1999). Complex polities in the ancient tropical world. Arling-
ton: Archeological Papers of the American Anthropological Association Number 9. American
Anthropological Association.
Balée, W. (1989). The culture of Amazonian forests. Advances in Economic Botany, 7, 1–21.
Balée, W. (1994). Footprints of the forest. Ka’apor ethnobotany. The historical ecology of plant
utilization by an Amazonian people. New York: Columbia University Press.
Balée, W. (1995). Historical ecology of Amazonia. In L. Sponsel (Ed.), Indigenous peoples and the
future of Amazonia: An ecological anthropology of an endangered world (pp. 97–110). Tucson:
University of Arizona Press.
18 C. Barreto
Balée, W., & Moore, D. (1994). Language, culture, and environment: Tupí-Guaraní plant names over
time. In A. Roosevelt (Ed.), Amazonian Indians from prehistory to the present: Anhropological
perspectives (pp. 363–380). Tucson: University of Arizona Press.
Barreto, C. (1996). The emergence of Central Brazilian native villages: Investigating the transition
to sedentism and agriculture in a traditional Bororo territory. Paper presented at the 61st Annual
Meeting of the Society for American Archaeology, New Orleans.
Barreto, C. (2011). A construção social do espaço: de volta às aldeias circulares do Brasil Central.
Habitus, 9(1), 61–80.
Basso, H. (1973). The Kalapalo Indians of Central Brazil. New York: Holt, Rinehart and Winston.
Beckerman, S. (1979). The abundance of protein in Amazonia: A reply to Gross. American
Anthropologist, 81, 533–560.
Beckerman, S. (1991). A Amazônia estava repleta de gente. In W. Neves (Ed.), Adaptações e
diversidade biológica do homem nativo da Amazônia (pp. 143–159). Belém: Museu Paraense
Emílio Goeldi.
Beckerman, S. (1994). Hunting and fishing in Amazonia: Hold the answers, what are the questions?
In A. Roosevelt (Ed.), Amazonian Indians from prehistory to the present: Anthropological
perspectives (pp. 177–200). Tucson: University of Arizona Press.
Brochado, J. P. (1980). The social ecology of the Marajoara chiefdom. Unpublished Master Thesis,
University of Illinois.
Carneiro, R. (1983). The cultivation of manioc among the Kuikuro Indians of the Upper Xingu. In
R. B. Hames & W. T. Vickers (Eds.), Adaptive responses of the native Amazonians (pp.65–111).
New York: Academic Press.
Carneiro, R. (1985). Slash-and-burn cultivation among the Kuikuro and its implications for cultural
development in the Amazon basin. In P. Lyon (Ed.), Native South Americans: Ethnology of the
least known continent (pp. 73–91). Waveland Press, Prospect Heights.
Carneiro, R. (1986). The ecological basis of Amazonian chiefdoms. Unpublished manuscript,
Department of Anthropology, American Museum of Natural History, New York.
Carneiro, R. (1993). Factors favoring the development of political leadership in Amazonia. In
W. Kracke (Ed.), Leadership in lowland South America (pp. 4–8). Washington: American
Anthropological Association.
Carneiro, R. (1995). The history of ecological interpretations of Amazonia: Does Roosevelt have
it right? In L. Sponsel (Ed.), Indigenous peoples and the future of Amazonia: An ecological
anthropology of an endangered world (pp. 45–46). Tucson: University of Arizona Press.
Clastres, P. (1979). La société contre l’État. Recherches d’ anthropologie politique. Les Éditions de
Minuit, Paris.
Costa, M. H. F., & Malhano H. B. (1986). Habitação Indígena Brasileira. In B. Ribeiro (Ed.), Suma
Etnológica Brasileira 2. Tecnologia Indígena, (pp. 27–49). Petrópolis: Ed. Vozes/Finep.
Crocker, J. (1979). Selves and alters among the eastern Bororo. In D. Maybury-Lewis (Ed.), Di-
alectical societies: The Gê and Bororo of Central Brazil (pp. 130ted). Cambridge: Harvard
University Press.
De Blasis, P., Fish, S., Gaspar, M. D., & Fish, P. (1999). Some references for the discussion of
complexity among the sambaqui moundbuilders from the southern shores of Brazil. Revista de
Arqueología Americana, 15, 75–105.
DeBoer, W. (1981). Buffer zones in the cultural ecology of aboriginal Amazonia: An ethnohistorical
approach. American Antiquity, 46(2), 364–377.
DeBoer, W., Kintigh, K., & Rostoker, A. (1996). Ceramic seriation and settlement reoccupation in
lowland South America. Latin American Antiquity, 7(3), 263–278.
Denevan, W. (1976). The aboriginal population of Amazonia. In W. Denevan (Ed.), The native
population of the Americas in 1492 (pp. 205–234). Madison: University of Wisconsin Press.
Denevan, W. (1996). A bluff model of riverine settlement in prehistoric Amazonia. Annals of the
Association of American Geographers, 82(3), 369–385.
Dias, O. (1972). Sintese da pré-história do Rio de Janeiro, uma tentativa de periodização. Revista
de História, 1(2), 75–83.
Social Complexity in Ancient Amerindian Societies 19
Ireland, E. (1986). Our chiefs do not spill their anger: Secrecy and covert leadership in witchcraft
executions. Paper presented at the 85th Annual Meeting of the American Anthropological
Association, Philadelphia.
Ireland, E. (1996). Chefia e dinâ micas políticas entre os Waurá. Paper presented at the 20th Annual
Meeting of the Associação de Antropologia do Brasil, Salvador.
Kracke W. (Ed.). (1978). Force and persuasion: Leadership in Amazonian Society. Chicago:
University of Chicago Press.
Kracke W. (Ed.). (1993). Leadership in Lowland South America. South American Indian Studies,
No.1, Bennington College.
Lathrap, D. (1970). The upper Amazon. New York: Praeger.
Lathrap, D. (1977). Our father the cayman, our mother the gourd: Spinden revisited, or a unitary
model for the emergence of agriculture in the New World. In R. Reed (Ed.), Origins of agriculture
(pp. 713–751). The Hague: Mouton.
Lima, T. A. (1991). Dos mariscos aos peixes: um estudo zooarqueológico da mudança de subsistê
ncia na pré-história do Rio de Janeiro. Doctoral dissertation, Faculdade de Filosofia, Letras e
Ciê ncias Humanas, Universidade de São Paulo.
Lévi-Strauss, C. (1973). Social structures of Central and Eastern Brazil. In D. Gross (Ed.), Peoples
and cultures of native South America (pp. 263–274). NewYork: Doubleday/The Natural History
Press [1952].
Lowie, R. (1949). Social and political organization of the tropical forest and marginal tribes. In
J. Steward (Ed.), Handbook of South American Indians. Vol. 5: The comparative ethnology of
South American Indians (pp. 303–409). Washington: Smithsonian.
Machado, J. (2003). Complexidade social naAmazônia Central: um estudo dos montículos artificiais
do sítio Hatahara. Paper presented at the XII Meeting of the Sociedade de Arqueologia Brasileira,
São Paulo.
Mann, M. (1986). The sources of power. Cambridge: Cambridge University Press.
Maybury-Lewis D. (Ed.). (1979). Dialectical societies. The Gê and Bororo of Central Brazil.
Cambridge: Harvard University Press.
Meggers, B. (1954). Environmental limitations on the development of culture. American
Anthropologist, 56, 801–824.
Meggers, B. (1971). Amazonia: Man and nature on a counterfeit paradise. Chicago: Aldine.
Meggers, B. (1990). Reconstrução do comportamento locacional pré-histórico na Amazônia.
Boletim do Museu Paraense Emílio Goeldi 6(2), 183–203.
Meggers, B. (1992). Amazonia: Real or counterfeit paradise? Review of Archaeology 13(2), 25–39.
Meggers, B. (1995). Amazonia on the eve of European contact: Ethnohistorical, ecological, and
anthropological perspectives. Revista de Arqueologia Americana, 8, 91–115.
Meggers, B. (1996). Amazonia: Man and culture in a counterfeit paradise (revised edition).
Washington: Smithsonian.
Meggers, B., & Evans, C. (1957). Archaeological investigations at the mouth of the Amazon.
Washington: Smithsonian.
Mendonça de Souza, S. (1981). Pré-história fluminense. Instituto Estadual de Patrimônio Estadual,
Rio de Janeiro.
Miller, E. (1987). Pesquisas arqueológicas Paleoindígenas no Brasil Ocidental. In Investigaciones
paleoindias al sur de la linea equatorial. Estudios Atacameos, 8, 37–61.
Miller, E. (1992). Archaeology in the hydroelectric projects of Eletronorte: Preliminary results.
Eletronorte, Brasília.
Moraes, C., & Neves, E. (2012). O ano 1000: adensamento populacional, interação e conflito na
Amazônia Central. Amazônica, 4, 122–148.
Moran, E. (1995). Disaggregating Amazonia: A strategy for understanding biological and cultural
diversity. In L. Sponsel (Ed.), Indigenous peoples and the future of Amazonia: An ecological
anthropology of an endangered world (pp. 72–95). Tucson: University of Arizona Press.
Social Complexity in Ancient Amerindian Societies 21
Myers, T. (1973). Toward a reconstruction of prehistoric community patterns in the Amazon basin.
In D. Lathrap & J. Douglas (Eds.), Variation in anthropology. Essays in honor of John McGregor
(pp. 233–252). Urbana: Illinois Archaeological Survey.
Myers, T. (1992). The expansion and collapse of the Omagua. Journal of the Steward Anthropolog-
ical Society, 20, 129–152.
Neves, E. (1998). Twenty years of Amazonian archaeology in Brazil (1977–1997). Antiquity, 277,
625–632.
Neves, E. (2008a). Ecology, ceramic chronology and distribution, long-term history and political
change in the Amazonian floodplain. In H. Silverman & B. Isbell (Eds.), Handbook of South
American Archaeology (pp. 359–379). New York: Springer.
Neves, E. (2008b). Warfare in pre-colonial Amazonia: When Carneiro meets Clastres. In A. Nielsen
& W. Walker (Eds.), Warfare in cultural context: Practice theory and the archaeology of violence
(pp. 139–164). Tucson: University of Arizona Press.
Neves, E., & Petersen, J. (2006). The political economy of pre-Columbian Amerindians: Landscape
transformations in Central Amazonia. In W. Balée & C. Erickson (Eds.), Time and complexity
in historical ecology: Studies in the Neotropical Lowlands (pp. 279–310). New York: Columbia
University Press.
Neves, E., Petersen, J., Bartone, R., & Silva, C. (2003). Historical and social-cultural origins
of Amazonian dark earths. In J. Lehman (Ed.), Amazonian dark earths: Origin, properties,
management (pp. 20–50). Dordecht: Kluwer.
Nimuendajú, C. (1939). The Apinayé. Washington: Catholic University Press.
Nimuendajú, C. (1942). The Sherente. Washington: Frederick Webb Hodge Anniversary Publication
Fund.
Nimuendajú, C. (1946). The Eastern Timbira. Berkeley: University of California Press.
Oliver, J. (2001). The archaeology of forest foraging and agriculture in Amazonia. In C. Barreto &
E. Neves Unknown Amazon (pp. 50–84). London: The British Museum Press.
Paynter, R., & McGuire, R. (1991). The archaeology of equality and inequality. Annual Review of
Anthropology, 18, 369–399.
Perota, C. (1974). Resultados preliminares da região central do Estado do Espírito Santo.
Publicações Avulsas do Museu Parense Emílio Goeldi, 15, 140–162.
Petersen, J. B., Neves, E. G., Heckenberger, M. J. (2001). Gift from the past: Terra preta and
prehistoric Amerindian occupation in Amazonia. In C. McEwan, C. Barreto & E. Neves (Eds.),
Unknown Amazon: culture in nature in ancient Brazil (pp. 86–105). London: The British
Museum Press.
Politis, G. (1996). Nukak. Instituto Amazónico de Investigaciones Científicas, Bogotá.
Porro, A. (1994). Social organization and political power in the Amazon floodplain. The eth-
nohistorical sources. In A. Roosevelt (Ed.), Amazonian Indians from prehistory to present.
Anthropological perspectives (pp. 79–94). Tucson: University of Arizona Press.
Posey, D. (1986). Manejo de floresta secundária, capoeiras, campos e cerrados (Kayapó). In B.
Ribeiro (Ed.), Suma etnológica brasileira. Vol. I: Etnobiologia (pp. 173–185). Petrópolis:
Vozes/FINEP.
Rival, L. (2002). Trekking through history: The Huaorani of Amazonian Ecuador. New York:
Columbia University Press.
Roosevelt, A. (1980). Parmana: Prehistoric maize and manioc subsistence along the Amazon and
the Orinoco. New York: Academic Press.
Roosevelt, A. (1981). Resource management in Amazonia before the conquest: Beyond ethno-
graphic projection. In D. Posey & W. Balée (Eds.), Resource management in Amazonia:
Indigenous and folk strategies (pp. 30–62). New York: New York Botanical Garden.
Roosevelt, A. (1991). Moundbuilders of the Amazon. Geophysical archaeology on Marajó Island.
New York: Academic Press.
Roosevelt, A. (1993). The rise and fall of Amazonian chiefdoms. L’Homme 126/128, 33(232),
2556/12.
22 C. Barreto
Wüst, I. (1990). Continuidade e mudança: para uma interpretação dos grupos pré- coloniais na
bacia do rio Vermelho, Mato Grosso. Doctoral dissertation, Departamento de Antropologia,
Universidade de São Paulo, São Paulo.
Wüst, I. (1994). The Eastern Bororo from an archaeological perspective. A. Roosevelt (Ed.), Amazo-
nian Indians. From prehistory to present: Anthropological perspectives (pp. 315–342). Tucson:
University of Arizona Press.
Wüst, I., & Barreto, C. (1999). The ring villages of Central Brazil: A challenge for Amazonian
archaeology. Latin American Antiquity, 10(1), 1–21.
Wüst, I., & Carvalho, H. (1996). Novas perspectivas para o estudo dos ceramistas pré-coloniais
do centro-oeste brasileiro: a análise espacial do sítio Guará 1 (GO- NI-100), Goiás. Revista do
Museu de Arqueologia e Etnologia, 6, 47–81.