Cerro Mangote

Folder Type:
Archaeology Site
Primary Title:
Cerro Mangote
Summary Description:
Cerro Mangote (AG-1) is a 35 x 65 m accumulation of pre-Columbian cultural debris containing abundant marine shells. It is located in Coclé province on the central Pacific coast of Panama on the summit of a low (48 m) and elongated hill at the landward edge of high tidal flats, or “albinas”, which have recently been turned into cultivated shrimp tanks. The hill itself is about 1.2 km long and lies 0.2 km north of the northern bank of the present (2016) course of the River Santa María. The site’s eastern edge is now 8 km from the active marine shoreline of Parita Bay, a mangrove-fringed inlet of Panama Bay, but when first occupied about 7800 years ago, it was likely to have been 7 km further inland. The site’s coordinates are 8º8’11.73”-N80º33’45.4”E. When Cerro Mangote was first reported in the professional literature in 1955 by Charles R. McGimsey, it stood out as the only Preceramic coastal site yet identified in the New World humid tropics. Pottery was sporadic, superficial and scattered. Its lithic assemblage was quite unlike that of any other site in Panama except ceramic-using Monagrillo (5500-3200 cal BP), which lies 20 km to the south near the current outlet of the River Parita on the eastern side of the Azuero Peninsula. An important feature of the stone tool assemblage--shared by Cerro Mangote and Monagrillo--were “edge ground cobbles,” eg, hand-sized river stones, which were used for grinding on their edges, rather than on the flatter areas of the tool. These diagnostic instruments, which were later shown to be widespread at preceramic and early ceramic sites in Panama and northern Colombia, were assumed by McGimsey and Willey to have been used for preparing plants, inferred to be wild at Cerro Mangote, and perhaps incipiently domesticated at Monagrillo. Much more recently, micro botanicals analyses of phytoliths and starch grains embedded in the grinding surface of stone grinders and bases at this and otherLate Preceramic sites in Pacific Panama, eg the Aguadulce Shelter, vouched for agricultural practices during this time period, currently inferred to span the period 8000-5500 calibrated radiocarbon years ago. When the publication of the first (1955) season at Cerro Mangote appeared, there seemed little doubt that the site was largely occupied in preceramic times: very few artifacts other than stone ones were found, and the lithic assemblage was strikingly unlike that of any other site in Panama except Monagrillo. It consisted of choppers, chopper-grinders, one-hand manos, edge ground cobbles, boulder metates, a stone disc, nutting stones, a fragment of cobblestone bowl, river pebble pounders, and flakes that lacked secondary chipping. In the 1955 excavations, which lasted only one week, 12 human burials were found. None was unequivocally associated with an artifact. The following year (1956), however, when the excavations were significantly enlarged by McGimsey over a six week period, several more burials were encountered. Immediately beneath one of these (31D [see below]), a carved shell pendant was found. McGimsey believed that this artifact fell “unmistakably into the tradition of the so-called curly-tailed monkey.” This “curly tailed” animal motif (of course, not necessarily a monkey, but any mammal that raised its tail!) was attributed to many gold and stone ornaments found at the Sitio Conte cemetery, further to the east, before the Second World War. The depiction of a curly-tailed mammal on the shell pendant at Cerro Mangote, in addition to twenty-five more pottery sherds and a fragment of polished celt, cast aspersions the exclusively preceramic nature of the deposit whose antiquity was confirmed in 1957 by a radiometric radiocarbon date of 4853±100 uncalibrated years BP, obtained in Stratum C, and opened the possibility that some of the burials were not contemporaneous with the deposition of the midden. The presence of different methods of treating the deceased also inferred that Cerro Mangote may have been utilized for funerary rites by more than one social group or at different times. Although the Late Preceramic age of the midden is clear, doubts persist about the real age of individual burials, which were placed within the midden. McGimsey originally reported that five pottery sherds had appeared in the excavations. Four of these were found on the surface, and one was considered by McGimsey to have been knocked into a position 1.35-1.5 m below the surface by workers climbing in and out of the pit. In a later article, he raised the number of sherds to 30, scattered over an area of 134 m2 and mostly concentrated within an area of 5 sqm. m. He inferred that they may have represented as few as two vessels, which he identified as the Escotá type of the Aristide ceramic group, which had been described by Willey and Stoddard in a 1954 article about the cultural stratigraphy of the Cerro Girón village site. This site lies 300 m to the west on and around a small hill, and whose debris extended to the bank of the River Santa María. (The Aristide group pottery dates from about 300 bc to ad 500). McGimsey described the Cerro Mangote midden as “hourglass-shaped” with a maximum depth of 2.2 m. His western trench measured 6x1 m with extensions added for recovering burials. A stratum of loose gray-brown soil with lenses of crab, oyster and other shells, nearly a meter in depth, lay over a reddish-brown clayey sediment with many shells. It bottomed out between 0.9 and 1.8 m depth. The eastern trench measured 8.5x1 m and comprised six internal divisions. It reached the basal reddish clay between 1.1 and 1.3 m below surface. Again, the upper soils consisted of loose gray-brown soil with frequent shell--mostly crab claws, oyster, mangrove clams (Anadara tuberculosa) and sand-gravel clams (Protothaca spp.) (Ranere later called this sediment, the “brown zone”). The lower deposit was a reddish clay with shells (Ranere’s term was the “red zone”). In addition to the trenches, McGimsey dug four test pits, one placed between the trenches and one each at the western and eastern end of the midden. The basal "red zone" of laterized clay has fewer cultural materials per sediment unit than the overlying "brown zone,” which is an organically rich refuse deposit that probably accumulated rapidly. McGimsey was conscious of the fact that Cerro Mangote was much nearer the active marine shore around 5000 uncalibrated radiocarbon years BP than today. In 1979, Temple University geologists led by John Adams returned to the area in order to improve knowledge about the timing of Holocene marine transgression in Panama and Parita bays, followed about 8000 cal BP by coastal progradation as the River Santa María’s sediments built up. John Barber’s 1981 Master’s Thesis (Temple University, 1981) reconstructed Cerro Mangote's relation to continental and marine sediments on the basis of two 2.6 m "Vibracore" samples. By correlating sedimentology, 14C dates and published sea-level curves for Panama, Barber proposed a facies change model, which related archaeological sites to coastline mechanics. He inferred that Cerro Mangote's initial occupation (ca. 5600 cal BC) coincided with the closest approach of a marine setting (1.2 km to the east) and that its abandonment (ca. 3600 cal B.C.) occurred at the end of a period of rapid delta progradation. An inferred location close to meso- and oligohaline waters, mangroves and mangrove channels, and tidal pools was confirmed by the taxonomic and size composition of the abundant fish remains in the Cerro Mangote middens. These emphasize estuary-tolerant marine catfish (Ariidae), snook (Centropomus spp), corvinas (Cynoscion spp.), toadfish (Batrachoides spp.), and small Pacific sleepers (Dormitator latifrons). Outer estuarine marine fish such as lockdowns, Selene peruviana and thread-herrings (Opisthonema)—abundant at much later ceramic sites along the River Santa María—were scarce or absent. The reptile, bird and mammal taxa in the middens are in harmony with a terrestrial-marine interface consisting of mud flats, mangroves, riverine woods and open areas like the remnants of wooded savannas that persist today among cane fields and cattle pastures. The commonest birds in Cerro Mangote’s midden are waders (Scolopacidae) and ibis (Eudocimus); black and green iguanas were taken frequently, and the commonest mammals by far were raccoons (probably all Procyon lotor) and white-tailed deer (Odocoileus virginianus). McGimsey’s work team did not use sieves. They gathered large vertebrate bones as they came across them while digging. In the 1970s, McGimsey asked an Arkansas archaeologist, Ralph Medlock, to analyze the abundant white-tailed deer while Richard Cooke studied remains from other vertebrates. A most interesting find were two distal ends of manatee ribs (Trichechus), which were identified by Smithsonian Institution cetacean specialist James Mead. Manatee does not exist today on the Pacific coast of tropical America. It is inferred that these specimens were brought to the site from the Caribbean coast. Another important zoological find of McGimsey’s field seasons was a canid humerus, almost certainly from a collie-sized domestic dog (Canis lupus familiaris). This is the oldest record for a domestic dog in Central America. Richard Cooke and Máximo Jiménez analyzed the vertebrate faunal data recovered in Anthony Ranere’s excavations in 1979 and have published several articles on this archaeofauna. They paid special attention to the fish, bird and reptile remains. The entire sample is currently (2016) undergoing a complete revision since the Smithsonian Tropical Research Institute’s collection of modern reference collections has been enlarged considerably since the 1980s when the bulk of the faunal analysis was undertaken. On encountering several disorderly holes dug by clandestine looters, Ranere’s team emptied them out searching for undisturbed cultural deposits and then using the stratigraphic information observed in the walls to recover undisturbed sediments. They also dug three new 1x2 m blocks. The third of these blocks was placed alongside one of McGimsey’s 1956 trenches in order to use the profile to guide the excavations. A total of 6 m2 of undisturbed sediments was excavated y Ranere’s team. Ranere’s reading of the natural stratigraphy was similar to McGimsey’s: 5 (bottom) – andesite boulders, 4 - Red clay between 25 cm and >1m thick (the base of this deposit is culturally sterile although the upper portion contains the earliest cultural materials and much animal bone), 3 - Light red silty clay with more cultural and biological material than stratum 4, 2 - The densest occupational refuse consisting of horizontally bedded lenses of marine shell and crab remains-- often badly crushed—and animal bones, 1 - a layer of silt between 10 and 15 cm thick representing the post-occupational deposit. McGimsey’s team recovered 67 human individuals in the 1955 and 1956 excavations, and Ranere’s team 12 individuals during the 1979 tests. The McGimsey sample is currently stored at the University of Arkansas, but will soon be repatriated to Panama. Ranere’s burials are housed in the archaeology laboratory of the Smithsonian Tropical Research Institute in Panama City, Panama. McGimsey and his collaborators identified four salient characteristics of their mortuary sample: 1) the variety of burial forms, 2) a high incidence of osteopathology, 3) the presence of cut marks on bones suggesting de-fleshing, and 4) physical variability. Burial was accomplished by placing one or more individuals in a simple pit in the Preceramic midden. McGimsey wrote that it proved very difficult to identify burial pit limits because of the softness of the ‘brown zone’ midden above the red clay zone, and the inclusion therein of abundant invertebrate remains. McGimsey’s team inferred that 18 of the 67 individuals recovered had been disarticulated. Fourteen of these were bundle or packet burials. Twenty-five tightly flexed apparently primary burials were identified, sometimes in conjunction with bundle burials in the same feature. This mortuary pattern has since been identified at other Middle Ceramic sites in central Pacific Panama, especially La Cañaza in Los Santos province and Cerro Juan Díaz on the lower La Villa River near the towns of Chitré and Los Santos. Individuals in the mortuary samples ranged in age from infant to mature adult. No artifacts were associated with any of the skeletons in Ranere’s excavation. McGimsey’s burial sample contained few mortuary objects, and all were made of shell. McGimsey’s burial #19C--an adolescent buried face-down in a tightly flexed position--owned a necklace of 53 shell beads. Shell beads likewise accompanied two individuals buried as members of a single group burial (McGimsey’s #31). The most interesting find in this feature (31D’) was the carved “curly-tailed monkey” pendant mentioned above (3.5x2.5x0.9 cm). It is very similar in size, material and form to thorny oyster (Spondylus princeps) ornaments found in Feature 16 of Operation 3 at the site of Cerro Juan Díaz, dated between 1900 and1500 calibrated years BP. The human remains from Cerro Mangote, which are well preserved, are important for ancient disease research because several individuals show signs of infection by the bacterium Treponema pallidum. McGimsey’s team cautiously inferred that five adult individuals exhibited periosteal lesions that corresponded closely to congenital syphilis of the bone, but warned that other explanations are possible. In order to re-evaluate this and other hypotheses concerning thee health status of the Cerro Mangote population, Aimée Huard re-evaluated the Cerro Mangote mortuary population for her Ph.D. dissertation from Binghampton University, New York. She inferred that the biological profiles from 93 individuals among McGimsey’s and Ranere’s samples pointed to low stress and low carbohydrate diet. Dental health was quite good with low values for dental defects, periostitis, pathologies, and calculus. The pathologies were mostly non-specific indicators of health with minimal cribra orbitalia and porotic hyperostosis the sample. A particularly important inference was that, although some individuals exhibited well-re-modelled and expanded long bone shafts--potentially consistent with a treponemal disease--the skeletal elements present were insufficient for a differential diagnosis. Huard hypothesized that unusually for tropical regions, some Cerro Mangote individuals showed evidence for scurvy, a disease caused by lack of vitamin C in the diet. This, in turn, suggests that some individuals in this population had limited access to fresh plant foods especially fruits. In view of the fact that fruits such as nance (Byrsonima crassifolia) and hog-plum (Spondias spp.) are characteristic species on the southern Coclé landscape, with high vitamin C values, scurvy seems theoretically unlikely if the individuals in question were of local origin. The final point we address in this summary is that of chronometric dating and the Preceramic nature of Cerro Mangote. This is still a controversial issue among archaeologists. A total of 17 radio-carbon determinations have been run on organic materials from Cerro Mangote since the first sample was analyzed by Yale in 1956 (Y-458d). Using radiometric techniques at the dawn of the technique, a date of 6810±100 BP was inferred for Yale charcoal sample, obtained in McGimsey’s Stratum C at a depth of 130-145 cm above the Red Zone. Using Intel04 and an intuitive 13δC value of -25, this age calibrates at 2σ to 7845-7465 cal BP (cal BC 5895-5515). A second charcoal sample from the basal red clay zone gave an Intel04-calibrated date (intuitive 13δC value -25) of 6670±215 BP (7940-6635 cal BP [cal BC 5990-5240]). A third radiometric sample run on charcoal by Temple University in 1979 inferred a date of 3555±100 BP, which, again using Intel04, calibrates with an intuitive 13δC value of -25 to 2200-1620 cal BC. (This date is not consistent with its stratigraphic position in the red zone at the base of the site; rather, it overlaps with dates from the Monagrillo Early Ceramic A site, and should be ignored). This errant charcoal sample was recovered in the same stratigraphic context as estuary clam (Protothaca) and estuary oyster (Crassostrea) The Protothaca shells were radiometrically dated by Temple University in 1979 to, 6710±170 BP (13δC 1.75, Intel04 at 2σ: 7580-6850 cal BP [5605-4900 cal BC]) and Crassostrea shells to 5055±155 BP (13δC 2.22, Intel04 at 2σ: 6180-5625 cal BP [4230-3675 cal BC]). Thus the radiometric dating of the Preceramic middens at Cerro Mnagote is by no means perfect especially since the reservoir effect for marine shell needs to be calculated empirically near the site. But a maximum age for these middens of 7845-5625 cal BP [5895—3675 cal BC] is likely to be quite accurate and is consistent with the stone and bone tool assemblage, and a Late Preceramic age. The radiocarbon dates for the human remains, however, are sufficiently disharmonious with the midden dates that the possibility that at least some of the human burials at Cerro Mangote are considerably later in time than the kitchen refuse, and were buried within it without the intrusive pits being adequately identified in the field, remains a very strong possibility. McGimsey himself was circumspect about the Preceramic age of a very large group of skeletons which he named Burial 31 since it contained shell ornaments, one of which was the Spondylus curly-tailed animal figure, mentioned at the beginning of this summary. There is no evidence from other sites that small personal ornaments in Spondylus and Pinctada mazatlanica are earlier than the La Mula polychrome, which have been dated by over 30 radiocarbon determinations to between 250 cal BC and 250 cal CE. Besides, the modes of treating the dead and their deposition as primary flexed individuals and secondary bundles or packets (often rectangular in form) are strikingly comparable at Cerro Mangote, Cerro Juan Díaz and La Cañaza between about 250 BCE and 700 CE. During this time period, the sequential polychrome styles in the Great Coclé semiotic tradition – La Mula, Tonosí and Cubitá – were in production. Many funerary features do not contain any pottery vessels although they often include polished stone beads and larger ornaments, shell jewelry, and perforated animal teeth. From a cross-cultural point of view, then, linking the Cerro Mangote mortuary record with that of communities that practiced Middle Ceramic burial traditions around Parita Bay and the Gulf of Montijo, remains a defensible hypothesis. The radiocarbon dating of human bone is known by archaeologists to be fraught with difficulties. In a site where abundant marine shell was deposited over many centuries, the possibilities exist of carbon exchange between the archaeological shells placed in strata of different ages and the human bones within them. But Cerro Mangote is not itself right on the coast, exposed to recent injections of marine carbonates from the sea. It is on top of a ~50 m hill whether hunters and fishers carried up quite sizeable or bulky prey to be prepared and consumed. As Amy Huard recently emphasized in her Ph.D. thesis, and as archaeozoologists have been at pains to point out for a long time, this site was occupied by people who spent considerable amounts of time there even acquiring manatee ribs—probably for carving--from the Caribbean. Cerro Mangote was not a shell collection or fishing station. Four intercostal bones and one tibia+fibula sample from the Cerro Mangote burials were radiocarbon-dated by the AMS method at the University of Arizona in 1988 (letter by Austin Long to Anthony Ranere, 22nd August). The results were: AA-4942, McGimsey’s Burial 31E: 2250±50 BP, δ13C: -14.1, N14/15: 6.5, cal BC 400-190 (2σ); AA-4944, McGimsey’s Burial 26: 1850±45 BP, δ13C: -13.8, N14/15: 7.8, cal AD 65-255 (2σ); AA-4945, Ranere’s Burial 69: 2200±45, δ13C: -12.8, N14/15: 7.4, cal BC 390-170 (2σ); AA-4946, McGimsey’s Burial 23A: 1970±60, δ13C: -13.7, N14/15: 7.3; cal BC 1100-cal AD 140 (2σ AA-4947, McGimsey’s Burial 20A: 2015±50 BP, δ13C: -13.5, N14/15: 7.7, cal BC 160-cal AD 80 (2σ). Summarizing, the five bone dates run by the University of Arizona’s Laboratory of Isotope Geochemistry in 1988 inferred the antiquity of five human individuals at Cerro Mangote to be between cal BC 400 and cal AD 255 (maximum 2σ range). This time, period represents the beginnings of the painted pottery tradition of the Greater Coclé culture area.
PIDTypeTitleMetadataURL
si_248CSV iconCerro Mangote UNIT 1 RZ.xlsDownload
si_249CSV iconCerro Mangote UNIT 3 RZ.xlsDownload
si_2754912Cerro Mangote 79 excavations for pid si_2754912Cerro Mangote 79 excavationsDownload
si_2754913Cerro Mangote clearing for pid si_2754913Cerro Mangote clearingDownload
si_2754914Cerro Mangote SE view for pid si_2754914Cerro Mangote SE viewDownload
si_2754915Cerro Mangote trench for pid si_2754915Cerro Mangote trenchDownload
si_2754916Cerro Mangote hill's base for pid si_2754916Cerro Mangote hill's baseDownload
si_2754917Cerro Tigre for pid si_2754917Cerro TigreDownload
si_2754918Cerro Mangote air view for pid si_2754918Cerro Mangote air viewDownload
si_2754919Cerro Mangote excavations for pid si_2754919Cerro Mangote excavationsDownload
si_2754920Cerro Mangote 1979 for pid si_2754920Cerro Mangote 1979Download
si_2754921Cerro Mangote Block 2A for pid si_2754921Cerro Mangote Block 2ADownload
si_2754922Cerro Mangote's excavations for pid si_2754922Cerro Mangote's excavationsDownload
si_2754923Cerro Mangote salt flats for pid si_2754923Cerro Mangote salt flatsDownload
si_2754924Cerro Mangote salt flats for pid si_2754924Cerro Mangote salt flatsDownload
si_2754925Cerro Mangote excavations for pid si_2754925Cerro Mangote excavationsDownload
si_2754926Cerro Mangote albinas for pid si_2754926Cerro Mangote albinasDownload
si_2754927Cerro Mangote high tidal flats for pid si_2754927Cerro Mangote high tidal flatsDownload
si_2754928Cerro Mangote burial 4B for pid si_2754928Cerro Mangote burial 4BDownload
si_2754929Mangote air for pid si_2754929Mangote airDownload
si_2754930Cerro Mangote burials for pid si_2754930Cerro Mangote burialsDownload
si_2754968Ranere's excavations for pid si_2754968Ranere's excavationsDownload
si_2795360PDF iconCerro Mangote: A preceramic site in PanamaDownload
si_2795362PDF iconCerro Mangote: Interpretations of space based on mortuary analysis Download
si_2795364PDF iconFurther Data and a Date from Cerro Mangote, PanamaDownload