CONCLUSION

Although not common, there are examples of chiefly societies that evolved on the basis of a foraging economy, especially through the exploitation of abundant aquatic resources (Arnold 2001; Erickson 2000; Goggin and Sturtevant 1964, quoted by Carneiro 1981: 49; Moseley 1975, but cf. Haas and Creamer 2004 and Pozorski and Pozorski in Chapter 31 of this volume; Widmer 1988). In such societies rulers may control the facilities that allow intensification, requiring the population to work in building and maintaining the facilities (such as fish weirs, boats, and drying and storage places) in exchange for food and protection (Johnson and Earle 2000: 262-263). With the ability to produce surplus to sustain the ruling elite, the chiefs also guarantee their followers the security of having food supplies year round due to storage facilities and technology. This arrangement may be especially attractive in places where wild resources are seasonally abundant requiring both management and storage capabilities.

As Widmer (1988: 280-281) has pointed out, the intensive exploitation of aquatic resources, usually seen as a foraging economy, actually follows the logic of agricultural systems. Aquatic resources are not only renewable, they can be managed in order to assure reproduction and availability. Indeed, in areas where aquatic resources are highly productive, agricultural intensification is dispensable (Carneiro ms.). Therefore, investment of time and resources in order to intensify fishing may not be a matter of lack of choice, but rather of cost-benefit reasoning.

At the Camutins site, the intensification of fishing enabled populations to rely on a stable protein resource, promote population growth, and develop complex sociopolitical institutions. It is likely that traditional kin groups claimed rights over the administration of the aquaculture systems, which probably first started as cooperative units, promoting intensification with little labor investment (see Stanish 2004). This dominance was justified by means of a religious system in which ancestor worship and access to long-distance exchange networks played important roles. It is predicted that the existence of similar ecological conditions at the headwaters of several rivers that drain the savanna grasslands enabled the development of similar social formations across the island, providing a model that awaits further testing. It is possible that other mound groups, not yet studied, will reveal several competitive chiefdoms or social formations with different levels of complexity. Preliminary data indicate the absence of a supra-regional political center, which suggests a system of alliances between chiefdoms that is compatible with peer-polity interaction models (Renfrew 1986; Roosevelt 1999; Schaan 1997).

Practices for the maximization of food production, such as river damming and fishing in temporary lakes and creeks that fill with aquatic life at the end of the flood season have often been reported among Amazonian peoples. The abundance of aquatic resources described in early chronicles and in colonial documents, both along the Amazonian floodplains and in the estuary (Marajo Island included), have encouraged scholars to consider

That the protein obtained from aquatic fauna was critical in promoting sedentary life and cultural developments in the region (Carneiro 1995, ms.). However, a bias that considers foraging as an unreliable economic enterprise has prevented scholars from adequately interpreting the archaeological record.

The enduring “agricultural chiefdoms” paradigm is ironic considering indigenous cosmologies that “privilege social and symbolic interactions with the animal world,” as Viveiros de Castro (1996:194) has pointed out. That paradigm just does not fit the archaeological evidence in Amazonia. Earthworks have been identified in several locations across Amazonia such as the central Amazon (Neves and Petersen in press) and the Upper Xingu (Heckenberger 2005) and, despite the fact that many of them are located next to lakes and streams, scholars have not correlated them to intensified fishing economies that would provide the necessary surplus for the emergence of complex sociopolitical systems. Mass fish harvesting is common in several locations in Amazonia today, especially in lakes and rivers affected by tides and seasonal water levels. Many of these locations were formerly inhabited by populations known for their elaborate pottery industries. For example, in the lakes that border the Trombetas River (where there are sites related to the Incised-Punctated Tradition; see Hilbert and Hilbert 1980) net fishing at the end of the rainy season is highly productive, suggesting that intensive fishing was also important there in pre-colonial times.

Increased attention to earthworks related to aquatic resource management would benefit our understanding of pre-Columbian complex societies in Amazonia. Practices that are presently common among indigenous and caboclo (mixed Brazilian Indian and European or African ancestry) populations most certainly originated in pre-Columbian times and may be a key to understanding the particular sociopolitical features that differentiate Amazonian complex societies from everything else we have seen on the continent.

ACKNOWLEDGMENTS This chapter summarizes the results of my doctoral research, so I am grateful to those individuals and institutions that helped me along the way. The research was possible thanks to funds provided by the Heinz Endowment Latin American Archaeology Fellowship from the University of Pittsburgh, the National Science Foundation (Grant No.

0233788), and CNPq - Conselho Nacional de Desenvolvimento Cientifico e Tecnologico of the Brazilian Government. I thank the MCT/CNPq-Museu Paraense Emilio Goeldi, which has always supported field and laboratory work. to Morgan Schmidt for helping me with the editing.