# The river was full of people ## An Amazonian field note A starving Spanish expedition descended the Amazon and came back with a report of towns, chiefs, roads and war. For centuries, the report sounded impossible. Then the canopy began to lift. ## The detour In late 1541, Gonzalo Pizarro's expedition was coming apart in the eastern foothills of the Andes. The men had crossed into a country they did not understand while searching for cinnamon and wealth. Rain ruined supplies. The ground swallowed horses. Hunger turned every green wall into a question. Francisco de Orellana took a small party down the Napo in a rough brigantine called the *San Pedro*. The plan was to find food, load the boat, and fight the current back to Pizarro. Then the river widened. The current became the strongest person in the expedition. They had gone too far to return quickly, and the farther they traveled, the more impossible the return became. The boat was no longer a supply run. It was a one-way door. They repaired hulls, made nails, traded for provisions, and learned the river by surviving it. At one settlement they heard drums before they saw the people. At another, local leaders arrived by canoe. The crew had been looking for a pantry. Instead, it had entered a political world. ## The record Gaspar de Carvajal, the Dominican friar who traveled with Orellana, wrote down what the crew saw. His account is a witness statement, not a census. > "Five leagues of houses, with no space between." Carvajal's most startling passages describe long runs of settlement along the water. He writes of food stores, canoes, chiefs, defensive displays, and houses continuing until the bank seemed to become one inhabited place. He also records roads leading away from the river. A settlement is not only a row of houses visible from a boat. It is a system of fields, paths, people, and obligations extending inland. The report still has sharp edges. It was written by a member of a conquest expedition, for an audience that wanted reasons to fund more conquest. It translates unfamiliar political structures into European words. It includes the famous story of warrior women, a claim that cannot be treated as settled fact. The account can show that a witness saw dense settlement. It cannot give us a clean population count. It does not prove one Amazonian empire. The river connected many peoples, and the surviving text does not establish a single ruler or capital. ## Put the layers beside each other It is tempting to draw a straight line from Orellana's report to every recent discovery. That would be a better story and worse history. The written route is a schematic reading aid for the 1541–42 descent from the Andes through the Amazon basin to the Atlantic. The archaeological points are independent windows into other regions of Amazonia. They are not stops on Orellana's route. The shared landscape does not make a shared identity. The relationship is a question for evidence, not a shortcut for the story. ## How a living world becomes a blank The later silence was real. The idea that it proves the first account was fiction is the part that does not hold. ### 1541–42: The first report Carvajal records stretches of settlement, leaders arriving by canoe, stored food, and roads running inland. His account is extraordinary, interested, and impossible to ignore. ### After contact: The river changes European intrusion brought disease, enslavement, warfare, mission pressure, and forced movement. The populations that later observers met were not living under the same conditions as the populations the first expedition encountered. ### The 18th to 20th centuries: The forest closes Wood, thatch, gardens, paths, and raised earth do not announce themselves like a stone wall. Once people leave, vegetation returns. A road becomes a low line. A plaza becomes a level patch. A town becomes a handful of mounds under trees. ### Now: The ground answers back Archaeologists use dark earth, ceramics, ditches, mounds, roads, pollen, charcoal, and laser mapping to read the landscape as a record of work. The forest is still there. It is no longer being mistaken for the whole story. ## The canopy lifts Recent archaeology does not give us a key that opens every sentence in Carvajal. It does something more useful. It makes the old idea of an empty Amazon impossible to keep intact. ### Upano Valley, Ecuador LiDAR revealed more than 6,000 human-made structures across an approximately 300 km² survey area, including mounds, plazas, roads, terraces, drainage works, and fields. The network began around 500 BC and lasted for centuries. It is a different valley, not a missing pin from Orellana's route. ### Upper Xingu, Brazil Archaeology documents large settlements, ditches, causeways, and planned relationships between communities. Researchers use "low-density urbanism" for a landscape organized around settlements and gardens rather than a single dense stone core. ### Llanos de Moxos, Bolivia Airborne laser mapping exposed elevated roads, canals, reservoirs, and monumental mounds built by the Casarabe culture. Some earthen pyramids rise about 22 metres above the surrounding plain. The structures show that a tropical wetland could hold planned, connected centres. They do not establish one basin-wide political system. ### Aquiry region, southwest Amazonia A 2026 Nature study combined satellite imagery, fieldwork, and LiDAR to estimate 24,000–30,000 earthworks across the Aquiry region. The same paper models 1.25–3 million people around AD 100–300. Both figures are extrapolations, not a census. ## Was Carvajal telling the truth? The honest answer has more texture than a yes or a no. - The expedition reached the Atlantic: **attested** - Some river stretches held dense settlements: **supported** - Amazonian societies engineered land at large scales: **supported** - One empire ruled the whole river: **unproven** - Every reported settlement has been found: **unknown** - The forest was untouched before contact: **wrong frame** The report is strongest when it describes a river lined with organized Indigenous societies. It becomes fragile when later readers turn a moving, plural landscape into one lost capital. ## The lost city was the wrong question There may never be one spectacular ruin waiting under the leaves to close the case. No single ruin is needed. Across different Amazonian regions, Indigenous peoples built towns, roads, canals, fields, ceremonial spaces, and political worlds that left a lighter signature than stone. Carvajal's account is not a passport to every dramatic claim. It is a damaged window. Through it, the river looks less like a wilderness that briefly surprised a few Spaniards and more like a populated world that colonial violence made difficult to see. ## Sources - [Gaspar de Carvajal, *The Discovery of the Amazon*, American Geographical Society, 1934](https://archive.org/details/discoveryofamazo0000carv) - [Heckenberger et al., "Pre-Columbian urbanism, anthropogenic landscapes, and the future of the Amazon," Science, 2008](https://pubmed.ncbi.nlm.nih.gov/18755979/) - [de Souza et al., "Pre-Columbian earth-builders settled along the entire southern rim of the Amazon," Nature Communications, 2018](https://doi.org/10.1038/s41467-018-03510-7) - [Rostain et al., "Two thousand years of garden urbanism in the Upper Amazon," Science, 2024](https://doi.org/10.1126/science.adi6317) - [Pärssinen et al., "Over 20,000 precolonial earthworks in the Southwest Amazonia," Nature, 2026](https://doi.org/10.1038/s41586-026-10835-7) - [Massimo Livi-Bacci, "The Depopulation of Upper Amazonia in Colonial Times," Revista de Indias, 2016](https://doi.org/10.3989/revindias.2016.013)