Marajoara Culture

← Encyclopedia

At the mouth of the Amazon the river ceases to be a river and becomes an inland sea, and in its center sits Marajó, an island the size of Switzerland, half savanna and half flooded forest. For roughly half the year the rains and the tides press the floodwaters up across the campos until the grasslands disappear beneath a sheet of water four meters deep; for the other half the rivers shrink and crack and the same ground bakes hard. A society that meant to stay on Marajó, rather than retreat before each flood, had to manufacture dry land. The builders of the Marajoara culture did exactly that. Across the eastern savannas of the island they raised tesos — also called aterros — great embanked mounds of packed earth, some standing as high as ten or twelve meters and spreading over several hectares, on which they set their longhouses, buried their dead, and rode out the annual inundation on islands of their own making. The mounds are the first fact of the culture, and the largest. They are why the dead and the living shared the same elevated ground, and why a people without stone left a built landscape legible two thousand years later.

Setting and chronology

The Marajoara phase is conventionally dated from about 400 to 1350 CE, the framework Anna Roosevelt fixed by radiocarbon in her 1991 study and that subsequent work has adjusted at the edges rather than overturned. Its classic florescence — the moment of the most elaborate ceramics — is usually placed between roughly 700 and 1100 CE. Mound-building on the lower Amazon is older still; some basal levels in the broader tradition return dates well into the first millennium BCE, so the Marajoara represent an intensification of a long regional habit rather than a sudden arrival. After about 1200 the pattern frays. Site after site falls out of use, the funerary ceramics coarsen, and by the fourteenth or early fifteenth century the great mound complexes stand largely abandoned — before, not after, the Iberians reached the estuary. When the Portuguese did arrive they met other peoples on the island, the Aruã and the loosely grouped Nheengaíba, whose presence on Marajó belongs to a later horizon than the urn-builders and should not be folded into them.

What ended the culture is genuinely unknown. The candidate pressures are the island’s own: the violence of the wet–dry oscillation, a run of drought or excessive flood severe enough to break the food base, and with it the surplus that sustained potters, ritual specialists, and whatever authority commissioned the mounds. The collapse of craft quality through the late phase is read by some as the visible signature of that unraveling. No single cause is demonstrated, and the abandonment is one of the open problems of the field.

The mounds

The tesos are not uniform heaps. Geological work by Dilce Rossetti and colleagues showed that the largest of them were not raised wholly from flat ground: the Marajoara recognized pre-existing natural rises — old fluvial bars and levees standing slightly proud of the floodplain — and built upward and outward from them, so that a great teso is part inherited landform and part anthropogenic embankment. That refinement does not diminish the labor. The earth of the mounds is shot through to its base with potsherds, ash, refuse, and the dead, the dense cultural fill of generations living on the same spot, and the scale of the work implies coordinated effort over long periods.

The mounds cluster. The best-studied cluster is the group along the Camutins River in the island’s interior — Os Camutins — some thirty-four mounds strung along the watercourse, the largest known Marajoara complex and the type-case for arguments about how the society was organized. Excavation at one of them, Teso dos Bichos, conducted by Roosevelt in 1983 and 1985, exposed lines of hearths arranged in a way that suggests multifamily longhouses set in rows atop the mound, the elevated equivalent of a village. By Lake Arari to the east stands the most famous single mound, the Pacoval teso, oval, roughly 150 meters long and 70 wide, rising about five meters above the water so that in the rainy season it stands clear as the Island of Pacoval. Its earth too is filled with pottery to the base. It was the Pacoval mounds, plundered for their urns in the nineteenth century, that first carried the name Marajó into the museums of the world.

Around and between the tesos the Marajoara worked the water itself. Roosevelt and Schaan both documented river-linked lakes, earthen dams, and excavated ponds associated with the platform mounds, some dating to around 500 CE — a managed hydraulic landscape of reservoirs and fish-holding basins integrated with the dry islands of the mounds. The mound and the pond are two halves of one technology for living on a floodplain: raise the ground for the people, shape the water for the food.

The funerary ceramics

If the mounds are the culture’s frame, its ceramics are its signature, and among the most accomplished pottery of the pre-Columbian Americas. Marajoara ware belongs to the Polychrome Horizon of the Amazon — the broad lowland tradition of painted ceramics that links it across the basin to the downriver Tapajó of the Santarém-Tapajós culture, the nearest and most often confused comparator, though the two are distinct ceramic worlds and distinct polities. The Marajoara potters commanded a deep repertoire: painting in red and black over a white slip; fine incision; and, most distinctively, excision, the cutting away of the clay background so that the design stands in relief, a labor-intensive technique that gives the finest pieces a carved, architectural density. Modeling adds three-dimensional faces and limbs. Two named styles anchor the sequence — Pacoval Incised, with its dense engraved fields, and Joanes Painted, the polychrome mode of the classic phase.

The most charged objects are funerary. The Marajoara practiced burial in large ceramic urns, many over eighty centimeters tall and some approaching a meter, in which the dead were placed — frequently as secondary burials, the bones gathered and reinterred after the flesh was gone — and set within the mounds. The urns are not plain containers. They carry anthropomorphic and zoomorphic imagery: stylized human faces and bodies, the eyes and serpentine forms that recur across the iconography, reptiles and birds rendered as effigy. More than twenty vessel forms are documented, the burial urn only the largest among bowls, plates, stools, and effigy jars. The ware is tempered in the Amazonian manner, with crushed sherd and the silica-rich ash of bark and freshwater sponge, and fired to carry color.

A second class of object turns the ceramic art toward the body of the living. The Marajoara made tangas — small triangular pottery pubic covers, modeled to the wearer’s anatomy and pierced at the upper corners for a cord at the waist, painted red and black on white like the urns. They were made for women and girls, and they are recovered alongside ceramic stools, labrets, spindle whorls, and seated figurines. The art is conspicuously concerned with the female: women appear across the imagery as ancestral and generative figures, founders of lines, and the pattern has led many scholars to read Marajoara society as organized through maternal descent. The tanga and the urn together describe a world in which the same hands that clothed a girl for life shaped the vessel that would hold her bones — a single ceramic tradition spanning a person from initiation to the grave.

The population-and-complexity debate

For most of the twentieth century the Marajoara were read through a thesis about what the Amazon could not do. Betty Meggers, with Clifford Evans her collaborator on 1950s Marajó fieldwork, argued in Amazonia: Man and Culture in a Counterfeit Paradise (1971) that the basin’s leached soils and apparent abundance set a hard ceiling on population and social complexity. On that reading the genuinely complex Marajoara — their monuments, their elite ceramics — could not have grown in place; they had to be the relic of an Andean-derived society that drifted down and then devolved under tropical-forest conditions toward the simple riverine band. The “counterfeit paradise” was a green forest that starved its inhabitants of the surplus a civilization needs.

Anna Curtenius Roosevelt dismantled that frame. Her Moundbuilders of the Amazon (1991), the product of geophysical survey and excavation through the 1980s, presented the Marajoara as an indigenous, in-situ complex chiefdom — densely populated, organized into a hierarchy of mound centers, fed by intensified cultivation of floodplain crops (maize or a local seed crop), and capable of monumental public works. She called it one of the outstanding indigenous cultural achievements of the Americas, and she made the argument polemically, against Meggers and Evans by name. Roosevelt’s reading carried the island’s population at its peak into the tens of thousands; figures up to roughly one hundred thousand for the whole of Marajó in the Marajoara phase have followed from this line of work. The vastly larger numbers that circulate for the Amazon as a whole belong to basin-wide estimates and to other sites; for Marajó itself the million-scale claim is rejected.

Denise Pahl Schaan then revised Roosevelt without retreating to Meggers. Her work on the Camutins group — a regional settlement survey of the thirty-four-mound complex, the kind of systematic landscape study Roosevelt’s site-by-site comparison had not done — recast the subsistence base. The Marajoara, on Schaan’s reading, were not principally grain farmers but managers of water and fish: the ponds, dams, weirs, and river-linked lakes around the mounds constitute an aquaculture, a deliberate engineering of aquatic resources, and the ceremonial mounds sit precisely where they command the most productive waters. The complexity is real — a chiefdom, or a set of them — but it rests on fish and floodplain management rather than on maize, and Schaan is correspondingly cautious about the height of any political hierarchy, with the question of ranked chiefdom versus a flatter, heterarchical order left genuinely contested. What the Roosevelt–Schaan exchange settles, against Meggers, is that Marajó grew its own complexity from its own ground; what it leaves moving is the shape and scale of that complexity.

The wider stakes are continental. The Marajoara are one of the cornerstone cases for the now-dominant view that the pre-Columbian Amazon was densely settled and deliberately shaped — a case built here not from airborne survey but from a century of urns, mounds, and pond systems read on the ground. They take their place beside the savanna urbanism of the Casarabe culture in the Bolivian Llanos de Mojos, the ditched geometric earthworks of the Acre geoglyphs, and the garden-city network of Upper Xingu Kuhikugu — each a separate earthwork tradition, each evidence that the basin held engineered landscapes long before the forest closed over them.

Scholarship and the funerary corpus

Marajó was among the first places anthropological archaeology was practiced in Brazil. The mounds and their pottery were described from the 1870s and 1880s by the geologist Charles Frederick Hartt, by Orville Derby, by Domingos Soares Ferreira Penna, and by João Barbosa Rodrigues and Ladislau Netto, and the spectacular urns dredged from the Pacoval mounds filled the National Museum in Rio de Janeiro and the Museu Paraense Emílio Goeldi in Belém while drawing attention abroad. Helen Palmatary’s mid-century study of the pottery and then the Meggers–Evans excavations set the stage Roosevelt later contested. The modern foundation of the field is Roosevelt’s Moundbuilders of the Amazon (1991) and Schaan’s Camutins study (2004), with the geological reassessment of the mounds themselves carried by Rossetti and colleagues.

The ceramic corpus that anchors all of this is scattered. The urns and tangas are dispersed across the Goeldi and the National Museum, the American Museum of Natural History, and collections in Europe — the legacy of nineteenth-century excavation and the museum trade, and the source of a continuing concern with provenance, dispersal, and the heritage rights of Marajó’s communities, in line with the cautions that govern any handling of Indigenous American material. The visual language of the urns had a long afterlife of its own: the excised geometry of Marajoara ware was taken up by Brazilian designers and artists in the early twentieth century as a national decorative idiom, the “estilo marajoara,” lifting the pattern off the funerary vessel and into the architecture of the living.

What the Marajoara made of death is the most exact thing they left. They did not bury in the earth so much as on it, raising the dead with the living above the flood, and they sealed the bones in vessels worked with the same red-and-black hands, the same cut-away geometry, the same ancestral faces that ran through every object a person touched. The urn was the last and largest piece of a continuous craft — the form a Marajoara life finally took, set into a mound built to outlast the water.

→ Related: Santarem Tapajos Culture · Casarabe Culture · Acre Geoglyphs · Upper Xingu Kuhikugu · Tiwanaku · Moche · Chavin · Olmec · Nazca · Shipibo Conibo · Tukano Vaupes Cosmology · Land Without Evil · Soul · Eschatology