Online chapter

Las sociedades originarias
(The Indigenous societies)

Director
Teresa Rojas Rabiela (Mexico)

Codirector
John V. Murra (United States of America)

Chapter 2: The Original Peopling of Latin America
Alan L. Bryan

People probably first entered what is now Latin America along the coast of Baja California. During the early Last Glacial the western cordilleras of North America were glaciated to the southern end of the Sierra Nevadas in southern California; so, except for the Columbia River gap, people would have been confined west of the Cascade/Sierra mountain chain. Some adventurous people could have expanded up the Columbia into the Columbia Basin and onto the Snake River Plain anytime during the Last Glacial, and some could have crossed south of the Sierras into the Mojave Desert, but most people probably remained along the littoral coast and smaller river valleys within the traditional productive ecosystems their ancestors had long occupied. Unfortunately, most coastal sites dating to the Last Glacial would now be submerged on the continental shelf. Early sites not submerged would be buried in river terrace deposits or other geological contexts. Such sites have been reported in San Diego just north of the Mexican border and in the Mojave Desert, but as they contain only simple unifacial cores and flakes, most professional archaeologists have concluded that there must be something wrong with the reported evidence. Although not dated to the Pleistocene because of the rising sea level, tectonic uplift on the channel islands of southern California, has preserved shellmiddens with deposits dated as early as 10,000 years (Meighan l989). These early shell midden deposits also contain simple cores, flakes, generally without retouch or other modification, and only a few knives flaked on both sides. If it were not for the fact that they had been excavated from shell middens containing tons of food refuse, hearths, and human skeletal remains, sceptical archaeologists would question their status as man-made tools. Meighan (l989) points out that early Californians primarily dependent on ocean resources, whose descendants later developed complex maritime adaptations, gathered shellfish and fished by wading in tidal pools or shallow estuaries using nets, fish spears, or weirs. Boats could have been simple dugouts or tule balsas used to dive from or fish with lines.

Shellmounds have been reported from raised beaches in Baja California, but none have yielded early dates. So far, the earliest site reported in Mexico is El Cedral south of Monterrey, where a possible hearth which dated 33,000 B. P. was surrounded by proboscidean toe bones. Two unifacial flake scrapers and some possible bone artefacts were also excavated (Lorenzo and Mirambell l986b). Completely permineralized artefacts worked and carved of extinct animal bone have been dredged from the bottom of Lake Chapala and collected from the bed of Pleistocene Lake Zacoalco near Guadalajara (Solarzano l990). No occupation sites have been located, and the bones have not been dated because of insufficient collagen. Several localities with extinct fauna were excavated in the Valsequillo Reservoir south of Puebla (Irwin-Williams l967). One locality yielded a flake scraper associated with shell which dated 20,000 B. P. Hueyatlaco, the main locality, yielded unifacial flake tools in addition to bifacial willow-leaf-shaped points, which seem to be too early if the 20,000 B. P. date can be extrapolated from the only radiocarbon dated locality.

Most of the evidence for early occupation has been recovered in the Basin of Mexico near Mexico City. At Tlapacoya, several blade-like flakes were found near a hearth dated 20,000 (Lorenzo and Mirambell l986c). Several mammoths have been excavated from ancient lake shores (Lorenzo and Mirambell l986a). Most of these sites have yielded only unifacial flake tools, but bifacial projectile points were found at two localities near Iztapan. The points are lanceolate and stemmed but not fluted. A date of 9,000 B. P. at Iztapan I seems too late for mammoths, although an associated shouldered point is considered to be like the Scottsbluff form, which dates about that time on the Great Plains. Fluted points have been found scattered in northern Mexico, including one at Zacoalco, but the only one excavated in a stratigraphic dated context (9460 B.P.) was in the Cueva de Los Grifos rockshelter near Ocozocoautla in the southern state of Chiapas (Garcia-Barcena l979). The point is a "fishtail" form, several isolated examples of which have been found farther south in Central America, but which has come from dated contexts only in South America. As the oldest dated (ll,000 B. P.) context is at Fell's Cave near the Straits of Magellan, and they appear soon thereafter in Buenos Aires Province on the Argentine pampas, it seems most reasonable to conclude that this distinctive form was developed in southern Patagonia as part of a local adaptation to hunting horses, and subsequently diffused northward after the horses became extinct. Two of the Fell's Cave fishtail points and the two points from nearby sites on the pampas have been described as fluted; however, technological analyses are required to determine if they are truly fluted like several from farther north, including the collection excavated from El Inga, near Quito, Ecuador, in a context dated no earlier than 9,000 B. P.

Clovis-like fluted points have also been found in Central America as far south as Panama; however, only one fluted base has been excavated from a context dated l0,700 B. P. at Los Tapiales, located on the continental divide at 3,000 m in Guatemala (Gruhn, Bryan, and Nance l977). Blades, burins, and simple bifaces were associated with the base and a flute flake. Bones were not preserved at this site, so we do not know whether these people were hunting extinct animals. Complete Clovis-like as well as fishtail fluted points have been collected from areas within a few kilometres of Los Tapiales, and mammoth bones have also been reported, so the potential exists for finding the association of man and mammoth in Guatemala. However, the association need not be with fluted points. Interestingly, a shouldered point quite similar in shape to one from Iztapan I was collected from the Quetzaltenango (Xelajú) Basin, which may have held a Pleistocene lake, like those in the Basin of Mexico (Bryan, in press).

At Turrialba in highland Costa Rica several Clovis-like points, blades, and burins have been collected from a sugar cane field. A fishtail point came from a lower terrace, suggesting that it was used at a later time (Snarskis l979). Several fishtail points, some fluted, have been surface collected from Madden Lake, an artificial pond built to hold water for injection into the Panama canal during dry periods (Bird and Cooke l978). The southernmost Clovis-like points have been reported from La Mula West, near the Gulf of Panama (Ranere and Cooke l989). Fishtail points, some of them fluted, have been found in Colombia and on the Paraguaná Peninsula in north-western Venezuela, but Clovis-like forms have not yet been reported from South America.

The distribution of Clovis-like and fishtail fluted points in Central and South America is most parsimoniously explained by the hypothesis that Clovis fluted points, which are dated on the Great Plains and in south-eastern Arizona between l0,900 and 11,200 were diffusing southward from their North American centre at the same time the Magellanic fishtail form, which is dated in the southern cone between 11,000 and 10,500 B. P., was diffusing northward. The two technological traditions met and merged in Ecuador and Central America about 9,000 years ago. Furthermore, the localized distribution of early projectile points suggests that local people developed special techniques for hunting large game animals, including the idea of bifacial projectile points, which were experimented with and adopted if they proved to be effective. Hunting around lakeshores and in open grassland ecosystems such as the high parámos would explain the sparsity of early points in lowland forested areas of Central America. The popular alternative explanation is that the earliest Central Americans all emphasized big game hunting with fluted points, and these people quickly moved through Central America and on to the Andean cordillera where there was more abundant game.

The model that people already well-established in their ecosystems developed special techniques for hunting land mammals that lived in grassland refugia as the forests were encroaching at the end of the Pleistocene means that early sites should be found that lack bifacial projectile points. Bifacial flaking disappeared in Panama about 7,000 B. P. because by then dense forests were everywhere and herbivores could no longer run in herds. However, in Guatemala, southern Mexico, and probably in Costa Rica, projectile points continued to be used in open areas of the highlands where some game survived. Because bifacial flaking disappeared early in densely forested areas, archaeologists identify early sites by the presence of bifacial flaking, and assume that all sites lacking bifacial flaking are late. This conceptual problem can be circumvented only by dating all sites lacking bifacial flaking, or by finding stratified sites without bifacial flaking in lower layers.

The Cueva de Espirito Santo in El Salvador yielded such a stratified sequence.

Beneath a ceramic occupation and an intervening sterile layer, the German archaeologist Wolfgang Haberland (personal communication l989, l99l) encountered an abundant lithic assemblage which lacks bifacial flaking. End and side scrapers, perforators, and choppers were recovered in addition to hundreds of flakes of exotic materials. The early component remains undated because of lack of charcoal and bone. The possibility of a Pleistocene context is suggested by what appears to be a faded painting of a proboscidean on the rear wall of the rock shelter. Not far south in north-western Nicaragua, a few pieces of flaked chert were recovered among abundant bones of extinct animals, including mastodons and ground sloths. As the chert occurs only downstream from the El Bosque site, the context suggests that people were scavenging the bones of dead animals (Espinosa l976 , Gruhn l978).

The only other site in Central America that was once thought to be early is located in the city of Managua. El Caucé or Acahualinca is one of those rare sites where human footprints are perfectly preserved because people walked across saturated volcanic ash that was soon covered by more ash. As the footprints are buried 8 m below the present surface, they were thought to be very old. However, a humic acid date of 5945 B. P. was collected in a nearby drainage canal that was projected to extend beneath the layer containing the footprints (Bryan l973a).

Last update 13/10/00