An in-depth study in the area took place in 1999, when other
six sites were also excavated (PA-JO-50: Rio Branco, PA-JO-51:
Saparará, PA-JO-52: Casinha, PA-JO-53: Vista Alegre, PA-JO-54:
São Benedito, and PA-JO-55: Leal). Together, they represented
nearly 1,300 years of occupation (from A.D. 300 to A.D. 1,600),
encompassing Marajoara Phase. More interestingly, those sites
were located on the periphery of the known Marajoara domain.
Parallel
to the EIS project, Schaan started the Anajas Project, whose aim
was to broaden the area of study, encompassing all of the prehistoric
communities located on the Middle and Upper Anajas River area.
Most of those sites belonged to the Marajoara culture, related
to a society that was organized socially and politically at a
regional level. For this reason, the study was meant to take into
account the political, economic and cultural interactions among
the populations who inhabited the various sites. This was done
in order to address questions related to the process of cultural
change that affected the entire region.
Within the Anajas Project design, it was important to understand
the relations between the ceremonial centers located at the Anajas
River headwaters and those settlements located further down the
river. Initially, a group of mounds located along the Camutins
stream (a tributary of the Upper Anajas River) was visited. It
was the first step in order to review and extend previous surveys
carried out by Meggers and Evans (in 1949) and Hilbert (in 1951).
In March 1999, mound number 16 (M-16, Hilbert count) and the Cuieiras
mound, where located, and were found to be almost completed destroyed,
due to looting. At the Camutins main mound, a 2-meter profile
was excavated, and charcoal samples were collected and dated.
At that time the project obtained the first absolute radiocarbon
dates for that mound: A.D. 750 (level 2 m) and A.D. 850 (level
1.15 m), showing an accumulation of one meter of sediment on the
top of the mound during a period of nearly one hundred years.
In
July 2000, with the help and support of Earthwatch volunteers,
the Anajas Project (entitled then " Lost Civilizations of
the Amazon") carried out the excavation of the PA-JO-52:
Casinha site, located in the middle Anajas River area. The research
aimed to investigate activity areas and the distribution of artifact
types, in order to compare spatial organization there with other
sites in the region. Despite the fact that the Casinha site was
partially excavated during the EIS project the year before, it
was not intensively studied because it was not located in the
area of direct impact of the Hidrovia do Marajó Project.
However, the site was considered important for its location: in
a forested area, peripheral to the area of floodplains taken as
characteristic of Marajoara settlement patterns. In that sense,
this study improved the understanding of the settlement hierarchy
and the expansion of Marajoara culture to other areas. At Casinha,
the project identified and excavated both household and burial
areas. Radiocarbon analysis dated the occupation between A.D.600
and A.D.1,200.
During July 2001, with the support of the Department of Anthropology
and the Center for Latin American Studies (University of Pittsburgh),
Schaan undertook another survey along the Camutins River. The
fieldwork consisted of identifying and mapping the mounds located
along the river margins. At that time, 28 of the 37 mounds situated
along the Camutins river floodplain were visited, and mapped,
updating and expanding the survey conducted by Meggers and Evans
in the late 1940´s, when the technology did not provide
the accuracy that the Geographical Positioning System (GPS) does
today.
The data collected made it possible to conclude that the mounds
were located on three distinct strategic sections of the river.
At the lower Camutins stream there were two large ceremonial mounds
(Camutins and Belém), flanked by two smaller mounds (respectively
Camutinzinho and Arraial). In the middle course of the river there
were ten habitation mounds, and at the upper Camutins stream there
were both small ceremonial and habitation mounds.
In the second half of July, 2001, again with Earthwatch volunteers
and their financial support, eight profiles were excavated in
Mound-1 (the Camutins main mound). Four radiocarbon dates, ranging
from A.D. 645 to A.D. 1,030, were obtained, relating to six meters
of stratified deposits. The data reflects that at least part of
the mound may have reached a height of four meters around A.D.
660, and seven meters around A.D. 1,000. During the excavations,
the project identified baked clay floors, layers of charcoal,
and a burial urn. All but one of the excavations were located
next to the western limits of the mound, where the deposits were
still intact. The mound had been partially destroyed, due to looting.
A thick layer of black soil full of sherds covered part of the
mound, as a result of looting, since they moved huge amounts of
soil and discarded broken pottery, not good for sale.
From September to October 2002, with the support of the NSF -
National Science Foundation, the mounds already mapped were again
visited in order to collect pottery samples, and the Belém
mound (the second in the hierarchy) was excavated. There, several
features related to both household and cemetery areas were excavated.
An extensive excavation of the burial area was completed later
in November 2002 with the financial support of Earthwatch Institute
and five volunteers. In a 20-meter square area, 24 different burials
were excavated and studied. The data should provide a better understanding
of household activities and gender at the local level, as well
as also contributing to the understanding of the sociopolitical
organization at the regional level.
The preliminary data from the Anajás Project was published
in Antiquity (2000) and papers were presented at the 2001 meeting
of Society of American Archaeology and at the 2001 meeting of
the Society for Brazilian Archaeology. Denise Schaan is currently
analyzing the artifacts collected during the 2002 field season,
and the study should be completed by May 2004.
