The Archaeology of Costa Rica

By Michael J. Snarskis, from Arqueo Costa Rica

In spite of its small size, Costa Rica can be divided into three general zones whose cultures produced artifacts of distinctly different styles, especially after c. 500 A.D. Natural boundaries, like the Cordillera Central and the Talamanca mountain range, with the contrasting climatic regimes that they create, were of considerable importance in the development of this cultural diversity. You may click on the names of these zones (listed to the left in the table above) to read about the particular characteristics of each zone. Click here for a short discussion of alternative names for these zones.

The time periods listed are based on a scheme for Central America agreed upon by a School of American Research Seminar in 1980, whose divisions correspond approximately to cultural thresholds important throughout Nuclear America. There are very few data for the first three periods (Period I - approx. 12,000 B.C. to 8000 B.C.; Period II - 8000 B.C. to 4000 B.C.; and Period III - 4000 to 1000 B.C.) of this chronology in Costa Rica. The last three periods (Period IV - 1000 B.C. to 500 A.D.; Period V - 500 A.D. to 1000 A.D.; and Period VI - 1000 A.D. to 1550 A.D.) are discussed in detail.

For a summary of findings, see Conclusions at the bottom of this page. To learn more about the archaeologists who have contributed to Costa Rican studies, as well as the tomb-robbers and collectors who have made such studies more difficult, click here.

The Guanacaste-Nicoya Zone
The northwest quarter of Costa Rica, on the Pacific side, forms a prehistoric cultural zone that will be referred to here as Guanacaste-Nicoya. Guanacaste is the name of the modern political province that includes long stretches of gently sloping inland plains that are today mostly divided into large cattle ranches, with some farming. The more mountainous, squared-off peninsula to the west is Nicoya. Both are part of what archaeologists call the Greater Nicoya Subarea, a Precolumbian cultural designation including part of western Nicaragua. Guanacaste-Nicoya is set apart from the rest of Costa Rica by relative aridity and marked seasonality. Its original cover of tropical dry forest is today mostly destroyed by agriculture and conversion to pasture, an alteration that began in Precolumbian times. Only .5-1 meter of rain falls each year, usually between May and December; some years not a drop of rain falls for four or even five months. Many small streams dry up, some trees lose their leaves, and the man-made grasslands turn brown and sere. Strong, dust-raising winds that buffer the countryside are responsible for this Verano, or dry season, as they rush the moist air from the Caribbean side over the plain and out into the Pacific before rain clouds have a chance to form. The broken Pacific coastline, with its numerous, small embayments and rocky headlands, also played a role in shaping Precolumbian cultures. Although only two bays (Santa Elena and Culebra) are large enough to shelter ocean-going craft from the gusty winds, they and the much larger Gulf of Nicoya provide myriad marine-estuary biotopes, environmental niches produced by reef formations, sand splits, and swampy, mangrove-filled backwaters, which are (or were, before overexploitation and pollution) hosts for an array of fresh and salt-water fauna, especially shellfish.

The long and partially navigable Tempisque River has formed, with its tributaries, a large and fertile alluvial plain between the Nicoya Peninsula and the line of volcanoes that marks the beginning of the Pacific watershed in Guanacaste. This region, heavily settled in Precolumbian times, together with receptive sections of the coast, formed the primary backdrop for indigenous cultural development in Guanacaste-Nicoya.

Periods I - IV (12,000 B.C. - 500 A.D.)
Periods I, II, and III (12,000 B.C. - 1000 B.C.)
When the Americas were first populated by Homo sapiens, who came across the Bering Strait between 15,000 and 35,000 years ago, the Central American isthmus had to be crossed. C-14 dates from sites in Patagonia tell us that this occurred around 10,000 B.C. The only evidence of the presence of these migratory bands of hunters and gatherers in Guanacaste-Nicoya is a single fluted spear point, obtained by Hartman in the 1890s and not identified until 50 years later.

Period IV (1000 B.C. - 500 A.D.)
Although the first pottery in northwest Costa Rica probably appeared in Period III, our present knowledge of the ceramic sequence picks up between 1000 and 500 B.C. Through the controlled stratigraphic excavation of only a handful of potsherds from depths of nearly six meters at some coastal sites, archaeologists have been able to identify a few whole vessels in museum collections as products of early Period IV. Decoration usually consists of wide, round-bottomed, incised lines, sometimes traced around thickened vessel rims, or in parallel bands separating areas of red slip; two colors predominate, red and the natural buff or brown of the fired clay itself. "Zoned Bichrome" is the name given Period IV in the regional periodization. This pottery shows a general stylistic relationship to other contemporary ceramic traditions throughout Nuclear America, especially those of the Middle Preclassic in southern Mesoamerica.

Sedentary settlements are suggested by a circular oven at the Vidor site, which produced a C-14 date of c. 800 B.C. Early Period IV is known as the Loma B phase in the Bay of Culebra archaeological sequence, the longest sequence for Guanacaste-Nicoya.

A trend toward increasing size and complexity of sites began during middle to late Period IV (300 B.C.-300 A.D.). M. Coe and Baudez first described the archaeology of this time; their stratigraphic excavations also provided supporting radiocarbon dates. Like Lange and his students some years later, they noted extensive cemeteries with differential mortuary goods, implying rankordered social structure. Village size and layout and house forms are as yet unknown, but, so far, more sites of this period have been found inland, on the foothills of the central volcanic range, than near the coast. Interestingly, the occupants of Zoned Bichrome coastal sites did not utilize marine mollusks, a major resource in later periods. It is likely that hunting, gathering of wild fruits and nuts, and agriculture were all practiced.

High-ranking burials frequently contain metates (grinding tables) along with jade pendants. Some archaeologists, including Lange, feel that the metates were "thrones" for high-status personages and not grinding tables. Their owners may sometimes have sat on these highly valued objects, but I do not believe this to have been their primary function, for several reasons: (1) the majority show considerable wear by grinding, even in sections of the upper surface that are decorated with low-relief carving; (2) the MNCR excavated two decorated examples of the period 300-500 A.D. at the Nacascolo site in 1980, each with heavy wear and a long "overhang" type of mano, or muller, in association; (3) the noticeable increases in population and social stratification during this period were almost certainly linked with more productive agricultural techniques, and it is not surprising to find among the paraphernalia of the ruling elite articles symbolic of food preparation and its ceremonial redistribution, major sources of political power in chiefdom societies.

I would also link the other well-known high-status object of this period, the jade "axe-god" pendant, to the spread of full-scale-probably maize-agriculture. Stone axes, or celts, were the standard forest-clearing tools in Nuclear America. Used for girdling and splitting tree trunks from late Preceramic times onward, they were especially numerous in agricultural societies. The axe-gods incorporate the celt form, usually with an avian effigy. These symbolic celts, the associated zoomorphic effigies, and the sometimes elaborately decorated ceremonial metates may have formed part of a politico-religious complex associated with the control of agricultural lands and the processing and redistribution of foodstuffs.

Behind most developed prehistoric subsistence systems was a mythological framework, the knowledge and perpetuation of which explained the origins of the system, usually in an allegorical fashion, and defined the rituals that formalized seasonal and other cyclical, necessary procedures. Priestly and administrative classes arose to handle the organization and sanctions required. Unfortunately, iconographic research in Costa Rican archaeology is scanty. Mark Graham has taken the first steps toward interpreting the various effigies that appear on stone sculpture.

There are indications that Mesoamerican influence was important in northern Costa Rica during several centuries before and after Christ. Elizabeth Easby has noted the importance of the celt form in both Olmec and Costa Rican lapidary work and has suggested that important gem-quality jade sources were located in Costa Rica, stimulating trade with Olmec centers like La Venta to the north. Recent physical analyses dispute this hypothesis, indicating that the highest quality jadeite may have been traded into Costa Rica from the north. In any case, there is increasingly better evidence for well-established trade routes between northern Costa Rica and southern Mexico from late Olmec times (800-400 B.C.) on. Since, as Easby notes, the stylistic links between the Olmec and Costa Rican Zoned Bichrome cultures are tantalizing yet incomplete, we cannot describe with certainty the nature of Olmec influence in Costa Rica. The few Olmec jades found here appear to have been heirlooms, associated with local objects hundreds of years younger. This suggests a "down-the-line" trading system (in which objects from point A reached point D through intermediaries at points B and C) rather than a direct Olmec-Costa Rica contact.

Ceramics from Guanacaste-Nicoya during the second half of Period IV (300 B.C.-500 A.D.) often display strong sculptural qualities and an elegance of line that strike sympathetic chords in modern aesthetic sensibilities. The human and zoomorphic effigy forms of the ceramic type archaeologists call Rosales Zoned Engraved are probably the most outstanding ceramic objects of this time. Some have been found in the Central Highlands-Atlantic Watershed zone, in contexts suggesting trade between elite groups. Many pieces, especially spouted forms, are reminiscent of formative styles in both Mesoamerica and the Andes. Like most prehistoric pottery from Costa Rica, these red, black, and brown or cream-colored vessels were made by a combination of molding and coiling, and were fired at relatively low temperatures in open hearths or rudimentary kilns. The potter's wheel was not used.

Mortuary evidence suggests that, in the last centuries of Period IV (200/300-500 A.D.), population and social stratification continued to increase. Many cemeteries of this time have been excavated in part, beginning with Hartman's Las Huacas excavations. There, in what was obviously a high-status burial ground, decorated metates, ceremonial mace heads, and jade pendants were' found in many tombs, with relatively little pottery. In the nearby Bolson cemetery, Baudez found both secondary and primary burials with more modest grave goods, mostly ceramics. This pattern was also observed in the MNCR excavations of contemporary cemeteries at Mojica, near Bagaces, Guanacaste, and at the coastal site of Nacascolo on the Nicoya Peninsula, where culinary pottery, a few modeled and incised ceramic effigy vessels, and mostly undecorated metates were recovered from burials containing both articulated, flexed individuals and secondary interments. Archaeologists deduce from these differential mortuary patterns well-developed class or rank divisions, although enduring evidence. such as architecture, is absent.

The repertoire of ceramic decoration begins to enlarge importantly in the two or three centuries before 500 A.D. The early bichromes are now often trichromes, the black motifs usually outlined with white. Incised and applique techniques are more frequent, and motifs become increasingly angular and geometric. Linear, rather than sculptural, qualities are emphasized. Modeled and/or effigy vessels of Guinea Incised ware include many striking representations, but they lack the subtle quality of line characteristic of earlier ceramic sculptures. Some Guinea vessels incorporate the image of one or more people reclining in a hammock, an object associated with southern tropical-forest peoples. Bat motifs begin to appear with greater frequency; the large, modeled, and painted ollas of Tola Trichrome are often decorated with bat forms. Figurines, ocarinas, stamps, and other specialized ceramic objects were particularly abundant during this time; they were perhaps indicative of a preoccupation with ritual required to bind together a society of increasing complexity.

The wider distribution of portable status objects like jades and ceremonial mace heads suggests that geographically dispersed upper classes may have possessed strong social or even hereditary ties, and that their status was produced and maintained, at least in part, by their access to foreign trade articles or peoples. Regionally important chiefdoms had probably developed in many parts of Costa Rica before 500 A.D

Period V (500 - 1000 A.D.)
The famous Guanacaste-Nicoya polychrome-pottery tradition and the beginning of a significant stylistic divergence between the matrtial culture of this zone and that of the Central Highlands-Atlantic Watershed mark this period. Although the causes are far from understood, the rupture of old northern trade routes, combined with new political, commercial, and perhaps spiritual allegiances to the south, may have been of considerable importance. A trend toward greater marine exploitation (at least of mollusks) began c.400-500 A.D., and it is possible that other ocean products like salt and purple dye from Murex Mollusca became increasingly valuable trade items.

Domestic features have been found at coastal sites like Vidor or Nacascolo: large, circular, rock-filled ovens or kilns; small, ellipsoidal, fired adobe hearths; fired adobe chunks with cane impressions, testifying to wattle-and-daub house constructions; floors; and postholes. The same sites had apparently differentiated burial grounds: at Vidor, more than 20 burials of females, children and infants (the last often in urns) were excavated in a vicinity of domestic activity and refuse, while Nacascolo yielded tombs of the same period, constructed entirely of natural basalt columns, overlain by a cap of fieldstone, and positioned along the contours of a steep natural slope. Grave goods were much more elaborate in the latter tomb type.

Carrillo Polychrome, with angular geometric patterns of red and black lines (often carelessly executed) on a buff or brown ground, is one of the most representative ceramic types of the period 500-800 A.D. It begins at the end of Period IV as a variant of Tola Trichrome, and eventually grades into the striking Galo type. Most forms are bowls, jars, and ollas; motifs thought to be stylized alligators or bats occur. Chavez White on Red, characterized by zoomorphic effigies as well as by functional bowls, also has roots in earlier styles. This type probably spanned the Period IV-V transition. Some archaeologists find in it stylistic affiliations to certain Ecuadorian ceramics. Pelicans or similar water birds are often represented.

Perhaps the finest ceramics from the period c. 500-800 A.D. are Galo Polychrome. Their mirror-bright burnished surfaces are technically unsurpassed by any Precolumbian pottery, and the yellows, reds, oranges, creams, maroons, and blacks of their polychrome decoration are impressively vivid. This ware is closely related to the Ulua Polychromes found in western Honduras and El Salvador. Galo, with cylindrical vessel forms and slab tripod feet, often looks very Mayoid. Guilloche and woven-mat patterns, seen on contemporary Maya pottery, where it is indicative of high rank, are also conspicuous. Galo has been found by looters in Nicoya with a fresco-painted vessel reminiscent of Teotihuacan pottery, and marble (alabaster) vessels from the Sula plain in Honduras. Galo was even found in the earlier Las Huacas cemetery and in Nicoya sites where Classic Maya glyph-incised jades have surfaced. The largest and most impressive ceramic figurines of prehistoric Costa Rica fall into the Galo category. They may be full human figures or effigy heads; the former usually have a pubic cover, and are painted with elaborate, busy representations of tattoos or body paint. Human faces are often exceptionally expressive. A striking and ornately modeled ceramic type of this same time is Potosi Applique or Modelled Alligator pottery. These vessels, which perhaps served as incense burners, often have a large, hemispherical base and a ventilated lid, elaborately decorated with a multicrested animal effigy, usually an alligator or crocodile. Variants of this pottery type persist into Period Vl.

The brilliant polychrome tradition that began in Guanacaste-Nicoya with Carrillo and Galo may represent an important new social dimension; when the northern trade network that brought jade, slate-backed pyrite mirrors, foreign ceramics, and other luxury goods broke down (c. 500 A.D., it seems), the Nicoyans responded by producing their own special-purpose pottery. Inspired by northern models, it also incorporated local and southern elements, forming a vigorous hybrid style that would be traded around Central America and southern Mesoamerica in the centuries to come.

Striking changes took place in almost all aspects of prehistoric life in Guanacaste-Nicoya from c. 800 A.D. on. Sites increased dramatically in both number and size. There seems to have been a marked shift of population concentration toward the coast, although inland zones some distance from the mountains apparently remained occupied. Lange has noted the presence of a thin volcanic-ash layer in coastal sites near the middle of Period V; volcanic activity may have driven people from the central piedmont toward the Pacific. While agriculture must have remained important, a much greater exploitation of marine protein sources began; most of the large and numerous Nicoyan shell middens began to accumulate after 800 A.D.

External factors also functioned as agents of change in northwestern Costa Rica during the eighth and ninth centuries. The central Mexican empire of Teotihuacan had long since dissolved, and now the Classic Maya lowland centers had also suffered catastrophic collapses. When these basically theocratic empires were fragmented by more militaristic nations. the ceremonial context in which elite-associated artifacts played important roles was drastically altered. In Costa Rica, the symbolic importance of carved jades, ceremonial mace heads, and elaborate metates suddenly declined midway through Period V.

Today, archaeologists tend to play down the importance of actual migrations of peoples as causes of cultural change, preferring to see cultural history as a result of the interplay of social and environmental elements. Nevertheless, there is evidence for the arrival in Guanacaste-Nicoya during this period of Mesoamerican peoples of the Oto-Mangue language group, probably as a result of social disruptions in the north. No doubt these influxes changed the stylistic tenor or local material culture.

The time of greatest diversity and production of polychrome ceramics (Middle Polychrome) was 800- 1200 A.D. The earliest varieties of Mora Polychrome, like the older Galo type, incorporate elements common in certain Late Classic Maya pottery -- -the seated figure with headdress, the mat pattern, and the Kan cross. Typical Mora painted decoration is mostly geometric, executed in red, black, and maroon on a buff-orange ground; vessel forms are usually simple hemispherical bowls. The other hallmark of the period is Papagayo Polychrome with a cream-to-white slip and brilliant orange-red paint, usually paired with black, and sometimes gray, in later varieties. Motifs on Papagayo pottery range from simple bands to complex figural scenes, including humans, jaguars, and a version of the Mesoamerican plumed serpent. Papagayo's great range of forms includes bowls, jars, zoomorphic effigies, and effigy-head tripods; some forms resemble those of Mesoamerican Early Postclassic horizon types like Tohil Plumbate and X Fine Orange.

Papagayo brought renown to the "Nicoya Polychromes," but it is just one of several related white-slipped wares that began to be manufactured in Pacific Central America during this time. Baudez noticed that polychromes like Mora and Birmania are found with great frequency in the southern half of Greater Nicoya, while Papagayo tended to augment toward the north. There is increasing evidence for a system of local centers of production for each major polychrome type; Abel-Vidor has postulated that almost all white-slipped pottery, beginning with Papagayo, was made at least as far north as Rivas, Nicaragua, and then traded to Guanacaste-Nicoya. Papagayo-like ceramics have been found as far north as the Toltec capital of Tula, in central Mexico.

There was a trend away from sculptural qualities toward painted ceramics in the second half of Period V; new effigy styles appeared, however, notably those of Guabal and early Birmania Polychromes. The former emphasize broad-legged seated human figures, most with intricate painted clothing, tattooing, or body painting; an invariably flattened headdress and face suggest possible cranial deformation, and ear flares are prominent. Birmania, with technically inferior painting, generally takes zoomorphic effigy forms that incorporate a small bowl; felines, birds, and even sea turtles appear. Ocarinas, especially turtle and avian forms, also are seen in these types.

Potosi "alligator incensarios" persist through Period V, becoming less elaborate. Brown-slipped "Chocolate Ware" begins in, and lasts through, this period and most of the next. Early types in the tradition, like Huerta Incised, are more skillfully executed, displaying motifs also seen in Galo Polychrome; later, the geometric painted designs of other polychrome types, especially Mora, are translated into incised decorative bands on mostly open bowls and dishes of the Belen Incised type.

Period VI (1000 - 1550 A.D.)
The trends in settlement patterns that begin in mid-Period V generally continue throughout Period Vl. Centers of population appear nearer the Pacific coast of the Nicoya Peninsula. The utilization of marine resources becomes increasingly important, while manos, metates, and similar ground-stone tools decrease dramatically. Archaeologists do not know if this indicates a drastic shift in agricultural systems and food-processing technology, if the associated tools began to be made mostly of wood (not preserved), or if they have simply not appeared in the small samples excavated so far

Data on house forms and sizes are still scanty, but part of one house, dating 1000 - 1300 A.D., was recently excavated by the MNCR at La Guinea on the Tempisque River. This structure was apparently ellipsoidal or rectangular, and 30-50 square meters in area. An unusual feature was the use of fired adobe blocks at intervals along the perimeter of the house, apparently placed to chock wooden poles in much the same way that river cobbles or field stones were used in other parts of Precolumbian Costa Rica; natural stone is scarce in the Tempisque drainage around La Guinea. Along one edge of the house, a large fragment of cane-impressed fired adobe was found, showing that the house walls were made of upright canes, c. 2.5 centimeters in diameter, lashed together with vines or ropes, and covered with adobe to a height of at least 50 centimeters. In other parts of the site, compressed sandy clay floors, with post holes, were found. The stratigraphy in the trench walls illustrated many prehistoric flooding episodes and, unfortunately, our excavation was destroyed by a modern flood before it was completed

During Period Vl, simple primary interments and multiple secondary burials were practiced. At Nacascolo, La Guinea, and other sites, dental mutilation, also observed in southern Mesoamerica, has been noted. Tomb structures may consist of a single, naturally columnar stone slab (a vertical marker) or a group of stones placed above (or single stones within) a burial, or may be unmarked pits. Ceramic grave goods of variable quality are usual. Long, triangular celts of chipped or polished stone have also been found in tombs; the chipped forms are probably tree-felling tools (almost identical versions come from the Atlantic Watershed), but the elegant polished examples suggest a ceremonial role. Many elaborate burials of this period have been looted; only one has been excavated by archaeologists. It contained the primary extended skeleton of a middle-aged male, surrounded by high-status polychrome pottery, a copper bell at his wrist, red ochre on the pelvis, and the skulls of six people of varying ages placed on his chest; their long bones rested nearby.

Early Period Vl corresponds to the last half of Middle Polychrome. There is an increasing emphasis on white-slipped polychrome pottery; new types like Vallejo, using blue-gray paint, and Mombacho, with underslip incising, incorporate Mexican-looking design elements. While the nature of this northern "influence" is not clear, there is reason to believe that the Postclassic Mixteca-Puebla expansion was instrumental in the dissemination of certain deity concepts and motifs. These and other white-slipped types concentrated in northern Greater Nicoya inspired somewhat inferior copies to the south, for example, Jicote Polychrome apparently manufactured along the lower reaches of the Tempisque River. Buff-orange-slipped polychrome types persist through at least the first half of Period Vl.

After c. 1150-1250 A.D., Papagayo Polychrome grades into the striking black and-red-on-white pottery called Pataky. This obviously elite-associated ceramic may have been manufactured as mortuary furniture. Its intricate, lacy black-on-white panels repeat unusual stylized jaguar motifs; the best-known vessels are modeled jaguar effigies, incorporating a pear-shaped container (cat. nos. 107, 108). The jaguar replaces the alligator (or crocodile) and the bat as the key animal figure in what are almost certainly mythologically symbolic contexts.

Two unusual ceramic types make their appearance rather late in Period VI. Murillo Applique is a glossy black or red pottery that features only modeled decoration. It has few antecedents in the region, and has been thought to indicate a late, undefined South American (or, at least, Atlantic Watershed Tropical Forest) influence. Unfortunately, it cannot be associated with antecedent pottery traditions either to the south or to the east. An even greater enigma is posed by Luna Polychrome. Its varieties of painted decoration, from "minimalist" patterns-with large, open zones of cream slip-to busy, honeycomb designs, resemble the Late Polychrome pottery made on Marajo Island, at the mouth of the Amazon River in Brazil. Luna is seen more often in southern Nicaragua, and has been found along with Spanish iron artifacts in burials. Was there a trans-Caribbean trading network that extended along the navigable San Juan River that divides Costa Rica and Nicaragua? Such a concept is not to be discarded out of hand, for Columbus described 40-man trading boats of the coastal Yucatan Maya plying the Caribbean coast of Central America in the early 16th century.

Prismatic blades of obsidian, while not numerous, are found in many Period VI sites. These are almost certainly trade articles from at least as far north as Nicaragua or Guatemala, for Costa Rican obsidian deposits are unknown. A more complex pattern of trade and technological diffusion is provided by metallurgy, which first reached Costa Rica from the south c. 500 A.D. Cast gold or tumbaga (a gold-copper alloy) artifacts are rarely found in Guanacaste-Nicoya, and it was thought that they were trade articles from Diquis or the Atlantic Watershed. Recently, however, Lange surface-collected a small, gold frog pendant, as well as a clay mold for a virtually identical, but different, piece at two different sites around Culebra Bay. Perhaps at least some gold work was produced in Nicoya. While copper artifacts were produced in the southern (Colombian-Panamanian) tradition, certain types of copper bells, occasionally found in northern and Central Costa Rica, are thought by some archaeologists to be products of the Mesoamerican trade network. One such bell (type IA3 in David Pendergast's classification) was found in the important Period VI burial at Nacascolo.

Crude, basin-shaped, stone metates have been found in a few sites of this period, but the ornate varieties seem to have declined or disappeared. Much of the columnar stone sculpture known from around Lake Nicaragua apparently can be placed in this time. Supposedly, similar sculptures were removed from the site of Nacascolo many years ago. Extremely crude versions of such statuary are still to be found there, and were also recovered by Baudez at the nearby site of Papagayo.

The first Spanish visitors to Greater Nicoya found large villages ordered around a kind of central plaza, which itself was bordered by residences and tombs of the ruling household. They recognized subsistence patterns (maize and beans), fragments of language, deity concepts, and even certain ritual activities (voladores, men suspended "flying" from a pole, and patolli, in which grains of maize were moved around a board according to throws of a die), similar to those previously observed to the north, in Mexico.

The Central Highlands / Atlantic Watershed Zone
By far the largest and most disparate of Costa Rica's archaeological zones, the Central Highlands-Atlantic Watershed is a composite of four or perhaps five geographic subzones, grouped here because the stylistic similarities of their artifacts suggest that they shared more or less common cultural traditions.

The largest and most characteristic part of this zone is made up of small to medium-sized valleys, with clear, rushing rivers, and the extensive, fertile lowland plains that make up the Atlantic Watershed of Costa Rica north and west of modern Port Limon. Below that city, the plains are abbreviated because the Talamanca range approaches the Caribbean. Here the Precolumbian cultural pattern seems to resemble more closely that of the Bocas del Toro region of Panama. The extreme northwest corner of the Atlantic Watershed seems to tend to Greater Nicoya affiliations. Neither subzone has been investigated systematically to date. Throughout most parts of the Atlantic Watershed, rainfall is heavy, generally from two to five meters per year, with no distinct dry season; less rain usually falls in March and April. 'The steeper eastern face of the Cordillera Central causes the moisture-laden easterlies to rise, cool, and release most of the resulting rain on the Caribbean side. The original vegetation in the zone was tropical rain forest. and it still remains in a few parts, although most of it has been cut down to make pasture lands and banana plantations, and to stock the strong market for tropical hardwoods. In spite of extreme weathering, soils are, for the most part, rich, dark, and of good drainage, although typical rain-forest laterites are sporadically present. Most rivers run with sparklingly clean white water in the- upper reaches, where they are filled with rounded volcanic cobbles, and are partially navigable in the more sedate meandering stages near the Caribbean. Cyclical flooding is the rule, not the exception.

The Central Highlands can also be divided into two subzones, the temperate valley where the modern capital of San Jose and most of the country's population are located, and the central Pacific drainage, composed of parallel ranks of rugged mountains and steep valleys which terminate in a limited strip of coastal plains. Although a part of the Pacific drainage and subject to its sharp seasonality, the Central Valley is closely related to the Atlantic Watershed throughout the known prehistoric cultural sequence. The central Pacific drainage appears to follow the same pattern, but this subzone is, archaeologically, one of the least known in Costa Rica; its limits on the Pacific coast fall roughly between the modern towns of Quepos and Puntarenas. Like the Central Valley, it has a basically Pacific climatic regime.

Periods I - IV (12,000 B.C.-500 A.D.)
Period I (12,000-8000 B.C.)
A workshop site from this period was recently discovered on the eastern slopes of the Cordillera Central. Called Turrialba, it shares with Madden Lake, another Paleo-lndian site near the Panama Canal, the distinction of having yielded two distinct types of chipped-stone spear points used by hunters of the Pleistocene megafauna 10,000- 12,000 years ago. One is a variation of the Clovis-point type, known throughout North America, while the other, the so-called fishtail, or Magellan point, was typical of paleo-hunters in South America. Costa Rica and Panama seem to for m the border between the spatial distributions of these two well-known classes of Paleo-lndian points. Even at that early date, this part of Central America functioned as a buffer or transitional zone between important cultural traditions on the two American continents.

Periods II (8000 - 4000 B.C.) and III (4000 - 1000 B.C.)
No sites can yet be placed in these periods with certainty, although one site has chipped-stone tools and debris much like that found in certain Pacific Panamanian sites of the Tropical Forest Archaic (Period II).

Period IV (1000 B.C. - 500 A.D.)
The earliest securely radiocarbon-dated pottery known in Costa Rica is the La Montana complex, from a site of the same name in the Turrialba valley. Five C-14 dates from the site range from 1500 to 300 B.C., clustering around 500 B.C. La Montana-phase pottery, almost entirely monochrome, is well made and has a fairly wide range of forms; it was almost certainly preceded by earlier ceramics. In general, it resembles pottery dating to 2000-1000 B.C. from northern South American sites like Barlovento, Colombia. This impression is strengthened by the presence of flat, rimmed griddles (budares), which are associated with the processing of bitter manioc or cassava in Colombian, Venezuelan, and Brazilian archaeological sites. Since they, and an unusual beveled type of mano, do not appear again in the Atlantic Watershed archaeological sequence, it is possible that La Montana peoples were the last to rely on root and tree crops as food staples (a carbonized avocado seed was found with La Montana deposits).

In 1977, the year La Montana was discovered, a very different-looking pottery came to light in the San Carlos region of the northern lowland plains. Called the Chaparron complex, it is a distinctly zoned bichrome pottery, characterized by a hard, glossy, red slip separated from the polished-brown or buff clay surface by wide incised lines. The Chaparron complex is most like the Conchas-phase ceramics (Middle Preclassic) from the Pacific coast of Guatemala both in form - tecomates, or in-curving, restricted-mouth bowls, predominate - and in decoration. Chaparron form and decoration suggest that it is nearly contemporaneous with La Montana. If so, Chaparron might represent a northern, Mesoamerican influence in Costa Rica, while La Montana most closely resembles southern styles.

The period 500-100 B.C. is still poorly known, although pottery probably dating to this time has been found at several sites. From c. 100 B.C. to 200 A.D., there was a veritable explosion of sites (i.e., population) and a trend toward social stratification, evidenced by a new series of high-status artifacts-elaborate metates, ceremonial mace heads, carved jade or similar stone, flutes, rattles, and, undoubtedly, a wide range of objects in perishable wood (staffs, drums, etc.), cloth, and bone. Sites of the El Bosque (middle Atlantic Watershed) and Pavas (Central Highlands) phases, dating from c. 100 B.C.-500 A.D., are numerous and large.

Contact with more developed Mesoamerican cultures c. 600-200 B.C., probably through elite-oriented trade, may have resulted in the gradual propagation in northern Costa Rica of a new mythic complex, or politico-religious "world view," in which different deities, a reverence for jade amulets, and possibly intensive maize agriculture were important components. The popularity of zoned red-on-buff pottery, common in Mesoamerica, but rare in northern South America, can perhaps be traced to this interaction. The population boom may have been produced by successful, intensive maize farming, producing increasing competition for prime agricultural lands, and a need to ritualize cyclical agriculture procedures. Warrior, priest, and administrative classes probably evolved to handle related duties, creating a market for luxury articles that were badges of office.

It should not be assumed that a Mesoamerican-type mythology obviated other belief systems. The predominance of toad, lizard, and especially cayman effigies is important, if one accepts Donald Lathrap's association of cayman symbols with manioc farming in South America. A patina of Mesoamerican symbolism may have combined with earlier tropical-forest animist beliefs, resulting in the menagerie of zoomorphic adornos, or ornaments, typical of Central Highlands Atlantic Watershed pottery.

Recent archaeological excavations by the MNCR have provided considerable information on house forms, tomb constructions, and associated artifacts. At Severo Ledesma, near Guacimo, in the eastern lowlands, three El Bosque phase houses were found. The two smaller ones were 3.5 x 12-meter rectangles, delimited by river cobbles stood on end; a perishable structure of wood, cane, and thatch was probably erected on this foundation, but no trace of it remains. Each house had two cobbles with cup-shaped depressions placed along one wall; these may have been mortars or receptacles. Numerous fragments of metates and other stone tools surrounded the houses, and several simple burials, excavated in the subsoil, were also associated. In the rainy Caribbean climate, no bones are preserved; burials are recognized by the tomb edifice and/or grave goods, mostly pottery.

The third El Bosque house was also rectangular, but much larger, 15 x 2.5 meters; because it was divided into rectangular segments by interior stone foundations, it may have been two adjoining structures. In any case, it housed people of a higher rank. The interior beneath the floor was honeycombed with burials and caches; one of the former contained an individual with a necklace of jade disk beads, surrounded by 27 pieces of grave furniture that included the plate of a "flying-panel" metate, fancy ceramic tripods decorated with animal and human modeled effigies, pottery ocarinas and rattles, and ground-stone celts.

Carbonized palm nuts of the species Elaeis oleifera HBK, an American oil palm related to the commercially important African oil palm, were found within this house, and a maize cob was found in another part of the site. No houses of the contemporary Pavas phase in the Central Highlands have yet been recognized, but there are considerable data on subsistence. During salvage excavation of a later architectural site at Barrial de Heredia, a deep trench bisected a large conical pit (two meters in diameter at the base and two meters deep) containing broken Pavas culinary pottery, stone tools, and a carpet of carbonized floral remains. Although botanical analysis is still in progress, it is known that the feature contained thousands of maize kernels; five maize-cob fragments, similar to Swasey 1 and 2 types from Cuello, Belize; several pieces of unidentified nuts or hard-shelled fruits; unidentified erect rhizomes; and several desiccated, pitted "cherry-like" fruits. Also present were one - perhaps two - varieties of Phaseolus vulgaris (common bean) that are "closer to Mesoamerican than Andean types," as well as seeds of the Convolvulaceae family, which includes both sweet-potato and morning-glory species. The final identification of these seeds will be revelatory; if the former, a new cultigen will be added to the Costa Rican prehistoric subsistence complex, while the presence of the latter would be hard evidence for psychotropic drugs, a tradition well documented in other Precolumbian cultures and indirectly indicated by El Bosque-phase clay double-tubed nasal snuffers.

When he first identified conical features at the type site of Pavas, a San Jose suburb, Aguilar referred to them as "bottle-shaped tombs" because their contents included human skeletal remains and whole ceramic artifacts. Nevertheless, a case can be made for classifying them as "bell-shaped" storage pits, associated with the domestic-activity zone surrounding a dwelling, as in the Formative Mesoamerican pattern. Marcus Winter notes their occurrence from the Valley of Mexico to Guatemala City, and emphasizes their almost universal use as maize-storage pits, which, upon abandonment, were often "filled with household debris including burnt daub, ashes, carbonized corn cobs and fruit seeds, animal bones, cooking pots and discarded manos and metates; some also had burials." Other Pavas-phase burials were unmarked except for associated grave goods, so mortuary patterns varied.

El Bosque-phase tombs may be one- or two-meter rectangles of cobbles; ellipses; corridors up to 12 meters long; or simply a scooped-out oval area in the subsoil, with no tomb edifice. Tombs in separate cemetery zones always have walls of cobbles and are usually long rectangles, often ordered neatly in ranks and files; these tombs repeat, on a smaller scale, the shape and proportions of El Bosque houses. Groups of 15 to 30 or more tombs, separated from other groups by empty corridors, may have corresponded to different lineages or clans. Although no bones or even teeth have been recovered, it seems that most burials were of the primary extended type; jade pendants, found face up in the bottoms of some tombs, were probably suspended around the neck of the deceased. Some El Bosque cemeteries cover several acres and contain hundreds of tons of volcanic-stone cobbles brought from river beds anywhere from 50 meters to several kilometers distant.

Chipped-stone artifacts are rare in El Bosque and Pavas deposits, but daggers of slate or fine basalt were produced by this technique. Pecked and ground-stone tools abound; mostly andesite, they are usually related to food processing or agriculture. Petaloid and trapezoidal celts, most with signs of hafting, are found frequently. Other ground-stone artifacts include bark-beaters, pestles, mortars, edge-battered cobbles, crude mace heads (probably weapons), loaf- and stirrup-shaped manos, and several kinds of metates. Loaf-shaped manos and basin- or trough-shaped metates, the typical maize-processing tools in prehistoric Mesoamerica, are found most frequently. Flat, tripod metates with raised rims and stirrup-shaped mullers are found less often and in contexts that sometimes suggest nondomestic roles - the preparation of special, ceremonial foodstuffs or drugs. Since the edges of these metates are carved in the shape of small heads, the taking and shrinking of trophy heads by warriors in battle seems to have been connected with them.

Metates are also employed as a funeral bier in the most prestigious tombs; the body is laid out on two or three of them placed side by side. Many of the extraordinary "flying-panel" examples may have been manufactured especially for high-rank burial. An exceptional burial of this kind was found in a salvage excavation at Tibas. Besides artifacts of jade, stone, and pottery from both the Central Highlands and Guanacaste-Nicoya, it contained a remarkable large Olmec jade clam shell (33 centimeters), with low relief on the interior showing a human hand holding a mythological composite animal, half feline, half moth. It is the only Olmec jade recovered in controlled excavations in Costa Rica.

Ceramics of the El Bosque complex in the Central Atlantic Watershed are most often red on buff with polished dark-red lips, interiors, and bases, and a collar of naturally buff-colored clay, smoothed, but left exposed around the vessel shoulder and neck. This area may be blank or decorated by a series of tool-impressed techniques, applique motifs, or painted linear patterns. Shell and reed stamping, combing, scarifying, fluting, pattern burnishing, and applique pellets and adornos are some of the decorative techniques found in El Bosque pottery. There are also red- and orange-slipped vessels with maroon paint. El Bosque pottery may be baroque, with piled-on applique, or exquisitely elegant and simple. Technically, the pottery is very well made, showing a dominance of the ceramic craft that disappears in later periods.

Pavas-phase pottery of the Central Highlands is modally similar to El Bosque, but orange slip and maroon paint predominate. Vessels are generally larger. Whereas the Ticaban tripods of the El Bosque phase are rather massive solid-leg vessels with zoomorphic adornos on each suppor, Pavas-phase ceramics of the Molino Channeled type are usually more graceful. Small, rotund figurines in a variety of domestic and ceremonial poses are frequent El Bosque-phase finds. Men with feathered capes, large headdresses, and zoomorphic masks, sometimes holding trophy heads, give a glimpse into complex ceremonial life. Women are portrayed holding children or carrying burdens on a tumpline. Animals are shown in naturalistic postures; what look like small dogs are often bicephalic. Most of these hollow figurines (the so-called Santa Clara type) double as rattles and may be finished with black, white, or yellow fugitive paint; red-slipped varieties also occur.

Other special-purpose ceramic artifacts include maracas; small rattles made with rings, to be worn on the fingers; ocarinas, flutes, and whistles, of various forms; flat and rolled stamps or seals (probably for body painting or cloth imprinting); and single- or double-tubed pipes - the last were probably used for inhaling drugs. Five C-14 dates for the El Bosque phase range from 50 B.C. to 425 A.D.

Period V (500 - 1000 A.D.)
After maintaining close contact with Mesoamerican culture for several centuries, the Central Highlands-Atlantic Watershed, during the middle part of Period V, underwent striking changes, which can be observed in house forms, ceramics, and high-status artifacts. Most evidence suggests that an undefined "southern influence" produced these changes. It was perhaps fortuitous that the fall of Teotihuacan in the sixth century, with the consequent disruption in central lowland Maya centers and the Pacific trade route to the south, coincided approximately with the introduction to Costa Rica of metallurgical techniques from Colombia and Panama, but it may turn out that there was a causal relationship, with elite-oriented gold objects and their associated mythology being brought in to fill the vacuum produced by the sundering of ties with Mesoamerican elite groups. Guanacaste-Nicoya did not react to these influences in the same way; after c. 500 A.D., the ceramic traditions of the two zones began a marked stylistic divergence, the former emphasizing polychrome painting and the latter, plastic decorative techniques.

In the first part of Period V, settlements seem to have followed the El Bosque pattern of dispersed villages of several houses, usually located on alluvial terraces; one incomplete excavation of a La Selva A-phase house in the Turrialba valley suggested a rectangular form. Long "corridor" tombs are also typical of La Selva A. Examples at the La Montana site near Turrialba were defined by rows of cobbles in two or three courses, measuring 2-2.5 x 7 meters. No domestic zones of corresponding Curridabat A-phase sites in the Central Highlands have been reported, but Aguilar excavated tombs of this date at Tatiscu, near Cartago, that were shaped like large, shallow basins and contained multiple interments. In one, he recorded a large fragment of a tumbaga figurine in the "Cocle," or Sitio Conte, style - one of the earliest metal objects found in Costa Rica.

The same kinds of ground-stone artifacts described for the El Bosque phase continue in La Selva A and Curridabat A. There is a decline in the technical skill of lapidary work, and frequently lesser stones are used as raw material. At La Montana, we recovered a necklace composed of tiny disk beads made from tiza, a light-green, chalky stone; at intervals among them were strung miniature beads of jade and tree resin, with a somewhat larger jade pendant placed to hang centrally on the chest. Ceramics sometimes reflect the zoned, red-on-buff El Bosque tradition, but the simple elegance that sometimes characterized that style is lost. Shiny maroon-on-orange painted decoration appears on the interior of open bowls, and purple or maroon paint on orange-brown slip, accompanied by a variety of incised, stamped, or applique motifs, becomes more common. This is what Hartman called Curridabat ware. The massive Ticaban tripods of El Bosque are gradually replaced by the hollow-legged Africa type, whose modeled adornos, perched above the supports, display a variety of ritual and domestic poses.

In one tripod we found a carbonized maize cob. Burnt maize in a mortuary offering might be symbolic of the funeral chicha, a thick, fermented brew made from maize or palm-fruit. In another part of the same cemetery, a mass of 15 to 20 of the same kind of tripods was found, smashed, near the surface at one end of a corridor tomb. This discovery recalls the two- or three-day funeral chichadas (rowdy, drunken feasts) described for historical times by Maria Eugenia Bozzoli and others. Another possibility is that the tripods were incense or offertory burners, since many are smudged on the exterior.

Red-slipped Gutierrez lncised/Engraved and the earliest brown-slipped incised types of Guanacaste-Nicoya have analogues in Zoila Red and La Selva Brown of the Atlantic Watershed. Incised triangles with simple hatching, probably symbolizing alligator scutes, are a common motif, and zoomorphic effigy vessels sometimes occur. Negative, or resist, painting - in which pottery is painted with a design in wax or impermanent clay and then smudged or repainted, after which the "resist" substance is removed to reveal a design the color of the original surface (a technique similar to the batik process on cloth) - increases greatly toward the middle of Period V. It is usually applied in curvilinear patterns reminiscent of Colombian and Panamanian motifs.

Around 700-800 A.D., the preferred house form became circular, and tombs became what have been called stone cists: oval or rectangular boxes of cobbles or flagstones, usually with both a floor and a lid of stone. When floors and lids are missing today, they were probably made of wood, now decayed. Wooden cist tombs - used because of taboos against the funerary bundle touching earth - are recorded historically. Some of the better-made prehistoric stone cists were sealed so well with natural volcanic flagstones that today they are only half-filled with fine dirt that has sifted in over a thousand years. At about this time, a few Early Polychrome trade ceramics (usually Galo or Carrillo) from Guanacaste-Nicoya appear in Central and Atlantic sites, beginning a substantial trade in polychromes between the two zones.

The earliest known site with circular house forms is La Fabrica, near Grecia in the Central Highlands, where the MNCR did archaeological salvage. Thirteen circular foundations of field stone and river cobbles were mapped; at least as many more may be hidden in nearby sugar-cane fields. Most foundations varied from 10 to 20 meters in diameter, and the largest had two rectangular entry ramps opposite one another. A cobble-paved causeway, nine meters wide, enters the site from the north and runs toward the principal structure. La Fabrica shows evidence of occupation from the first few centuries after Christ until c. 1100-1200 A.D.; associated pottery of the Curridabat B phase suggests that the circular structures date to 700 - 900 A.D.

La Fabrica has no stone cist tombs. Instead, the mortuary patterns recall those of Guanacaste-Nicoya: tombs were marked with natural stone columns and/or accumulations of field stone. Fired adobe floors appeared 50 to 200 centimeters below the present surface in cemetery zones and beneath houses, where burials were also located. Most La Fabrica burials were of the primary extended type. One cemetery burial, laid on three decoratively sculpted metates, was accompanied by artifacts known to be indicative of high status in most of Costa Rica before c. 800 A.D.: jades, ceremonial mace heads, teardrop-shaped polished black celts, and a stirrup mano with zoomorphic motifs. A collared jade tube found in this burial is identical to examples recovered from the Sacred Cenote at Chichen Itza, dating to the Maya Late Classic (c. 800 A.D.). While this burial is similar to the Tibas burial that contained an heirloom Olmec jade, artifacts from the La Fabrica burial place it somewhat later.

At La Fabrica, carbonized remains of maize, beans, and palm nuts were found, which, together with the large quantities of quotidian manos, metates, and chipped-basalt tools, indicate a nucleated agricultural village. Differences in architecture and grave goods (a copper or tumbaga bell and deer antlers were found in the principal house) reveal a rank-structured society.

Barrial de Heredia was another Central Highlands site excavated as part of an MNCR salvage program. Its architectural remains date to the transition between Periods V and Vl (900-1100 A.D.). Although later than La Fabrica, Barrial did not have round houses. Of eight structures excavated and mapped, three were ellipsoidal and five were square or slightly rectangular. The two shapes apparently correspond to functional differences, for the ellipsoidal cobble foundations contained much more domestic refuse (burned food remains, broken culinary pottery, and stone tools) and large ovens and/or hearths, and lacked burials beneath the floor, while the quadrangular examples showed less evidence of domestic activity but had burials with imported polychrome ceramics beneath the floor. The largest quadrangular and ellipsoidal structures directly adjoined, suggesting that occupants of the latter (wives?) were involved in the domestic maintenance of those domiciled in the former. Two kinds of tombs were noted at Barrial: a variant of the long, corridor variety, using only one line of standing cobbles and found only beneath the largest quadrangular houses, and a simple rectangular trench capped with flagstones. The latter type was found under houses and in a small cemetery zone, 100 meters away.

Basin-shaped metates and simple cobble mortars appear at Barrial; the most commonly observed stone tool was a small boxlike mano-machacador (combination grinder-pounder) showing use-polish on its flat facets and battering at the extremities. Carbonized maize and beans were found in and around the structures. Although Curridabat B-phase pottery was found beneath many parts of the site, ceramics associated with the architectural features were more closely related to the crudely executed applique styles of Period VI. Other frequent finds were small, open dishes with tripod zoomorphic-effigy head supports, a brown slip, and geometric incised panels on the exterior (Tayutic Incised), and a large number of caches of tiny ollas, often placed well away from burials and al no great depth.

Of the hundreds of Guanacaste-Nicoya polychrome sherds found at Barrial, a considerable percentage showed crack-lacing holes, a mending technique whereby perforations are drilled on either side of a fracture, and thongs are employed to bind the weakened part together. This bespeaks esteem for the foreign polychrome pottery, since almost no locally made vessels were thus mended. The repeated presence of this polychrome in higher-status tombs at Barrial indicates a flourishing, elite-oriented trade network between these two archaeological zones. Since Central Highlands Atlantic Watershed ceramics are not found in Guanacaste-Nicoya sites, we know that something else was being traded in return, possibly perishable commodities (carved wooden objects, feathers, poisons, drugs, cacao, or slaves). Two C-14 dates from Barrial fall between 800 and 1000 A.D.

Period VI (1000 - 1550 A.D.)
A recent catchment analysis - in which a series of natural resource zones and environmental variables are related to the location of human settlements - of a small sample of Atlantic Watershed archaeological sites revealed an interesting trend. Through Periods IV and V (some 2,000 years) there was an increasing tendency to locate settlements on reasonably flat alluvial plains suitable for farming. In Period Vl, the distribution of sites in several environmental zones becomes random, suggesting that other factors besides good farmland became important in choosing a site; these factors were probably socio-political frontiers and defense.

Period VI habitation sites in the Central and Atlantic zones are usually easily recognized because of their rudimentary but distinctive architecture: round, earth-filled mounds or simple foundations with retaining walls of stone cobbles; calzadas, or cobble-paved causeways; small ridged enclosures or plazas; and even aqueducts and giant flagstone bridges in the larger sites. Petroglyphs (rock carvings) are frequent. Just as noticeable to the archaeologist is the degree of agglomeration (houses and other features packed together) which gives the sense of definite site boundaries. There seem to have been networks of tightly organized sites, small and large, often surprisingly close together.

The largest, most complex site known for this period is Guayabo de Turrialba. Other major sites in the Atlantic lowlands are Las Mercedes, now destroyed; Anita Grande; Costa Rica Farm; and, on a smaller scale, La Cabana. Such sites probably number well over 100 in this archaeological zone, although only Guayabo and La Cabana have been excavated horizontally, in part. MNCR excavations in La Cabana in 1976-77 exposed two circular mounds, 20 meters across and less than two meters high; one non-mound house circle, 12 meters in diameter; and a square, ridged enclosure surrounding an empty "plaza," with a cobble-paved causeway leading into it. Horizontal stripping of the two main mounds showed that Mound 1, the higher, had only a central hearth, while Mound 2 (from which a curved porchlike area projects near the entrance) had along one part of the interior a series of four or five boulder metates, sometimes surrounded by stone seats; one had a mano nearby. In addition to the central hearth, several smaller ones were noted. The presence of food processing on one mound and not the other would seem to indicate a different function for each. The taller Mound 1 might be construed as the residence of the ruling individual or group, with Mound 2 housing wives or others having to do with the domestic maintenance of Mound I inhabitants. Fray Agustin de Zevallos, writing in 1610, describes several customs of the peoples then living in eastern Costa Rica, "who live in palenques, which are forts built in native fashion.... The chiefs have the women that they desire all in the same house, and the common people generally have one....".

At La Cabana stairways off the two major mounds, as well as one major and three minor paved causeways, enter the empty plaza area. Within the walled, square enclosure surrounding the plaza were found small, stone cist tombs or caches containing prestigious ceramic artifacts. A plaza area with enclosure has been noted in all major Period VI sites, invariably near the principal mounds. At Anita Grande, two very large quadrangular plazas are connected by a causeway almost 500 meters long. It is tempting to interpret this plaza configuration as the formal place of contact between the ruling class and the lesser population of the site, perhaps for ritual redistribution of goods.

Stone cist burials predominate in Period VI, under or around houses and in special cemetery zones. Much more use was made of flagstones as floors or lids of cobble-built tombs, or as the sole element in tomb construction. Several stone cist cemeteries excavated around Cartago by the MNCR contained extended primary and secondary burials, some tombs containing both kinds. Cups made from human skulls were found above many tombs. Exhumation and reburial seem to have been practiced, for many tombs were divided, added to, or built in as many as three vertical layers. Stone cists in the Atlantic lowlands are more often made of carefully chosen cobbles, chocked with smaller stones; no mortar was used. Grave goods in stone cist tombs are usually fewer and of poorer quality than those of the preceding periods. Many tombs contain no offerings, although perishable articles may have been included. Tiny quartz crystals, used as gravers, were found in several stone cists in the Cartago valley (Hacienda Molino and El Cristo sites). One tomb yielded a copper bell with a quartz crystal inside it.

In Period VI, cobbles were used as quotidian metates. Slightly used mortars and hammerstones seem to have been frequent. Ornate ceremonial metates did not disappear, however, although styles changed. Four-legged jaguar-effigy metates appear, and round versions with pedestal or atlantean bases become more common. Free-standing stone sculpture increases sharply, tending to repeat "standard" or "archetypal" poses; idiosyncratic poses (women braiding their hair, a man urinating) are also seen. Some sculptures seem to be portraits of specific individuals, even depicting deformities or facial tics. While earlier stone sculpture almost always portrayed zoomorphic effigies or, at best, a human figure with a zoomorphic mask, Period VI sculpture primarily records human subjects. This must represent a fundamental philosophical shift, wherein military or political power was augmenting at the expense of the traditional "religious" power base, probably as a result of population pressure on certain resources and/or new modes of conflict resolution. The hubris of emergent warrior-chiefs may have caused them to erect stone images of themselves as large as, or larger than, those of their zoomorphically symbolized deities.

With a few exceptions, Period VI pottery is of poor quality compared with that of earlier cultures, and there seems to have been less of it. Crude little ollas and tripod dishes of a coarse, poorly knit paste are the most frequent finds. Better preserved examples are overloaded with decoration (incision, tool stamping, applique pellets, fillets impressed to look like chains, and crudely modeled heads or animal figures). Lothrop called this Stone Cist Ware. Brown incised/engraved types from the end of Period V continue throughout Period Vl, increasingly carelessly made. Resist decoration usually appears only on high-status pieces. Dishes or slightly flaring cylindrical jars with animal-head tripod supports are found, as they are in Guanacaste-Nicoya, but the long-legged tripods of the two preceding periods virtually disappear. Seen more frequently is an unusual skillet-like form, known as a frying-pan censer. apparently used for burning incense. Cartago Red Line begins in this period, first as simple or tripod dishes, often with a stylized feline head and tail added on; red finger-painted lines on orange slip decorate the earlier varieties. Later, the red paint becomes more vivid, a cream slip replaces the orange, and more animated design possibly attempts to copy the brilliant Papagayo Polychrome traded in from Guanacaste-Nicoya.

In other Period VI ceramic types, a Diquis or Chiriqui stylistic influence can be felt. Irazu Yellow Line shows geometric designs in thick yellow paint on two-tone orange and brick-red slip. while Cot Black Line has similar motifs in weak black and red paints on orange-brown slip. Especially important are open dishes or bowls, with effigy-head tripod supports. The geometric motifs recall those of Chiriqui Polychrome, a late Diquis type that may have been inspired by Guanacaste-Nicoya polychromes. Another Atlantic type, Turrialba Bichrome, exhibits very thin vessel walls and elegant, simple olla forms in the best examples, recalling the delicacy of the exceptional Tarrago Biscuit type of Diquis. Trade sherds of Tarrago Biscuit found at Guayabo de Turrialba, provide more concrete evidence for ties to the south.

Metallurgy was the most important material-culture introduction from the south, reaching its apogee in the Central Highlands-Atlantic Watershed during Period Vl. The disappearance of jade working by 1000 A.D. has previously been attributed to the exhaustion of local jade sources, but we may now consider the possibility that Atlantic and Central cultures at this time were simply not concerned with jade as mythologically significant material, preferring gold amulets or other articles important in a southern-oriented ritual complex.

Figurines, ocarinas, rattles, and other small special-purpose ceramic articles (with the exception of a well-made brown incised type that spans the transition between Periods V and Vl) are very rare in this period. Stone and wood carving may have taken precedence over the ceramic medium for figurines. Five C-14 dates from La Cabana-phase sites range from 1000 to 1400 A.D.

The Diquis Zone
The Diquis archaeological zone is also composed of topographically and climatically diverse subzones. While most of the Pacific coastal strip below Quepos and the broad upland valleys to the southwest of the Talamanca mountains have a severe, clearly demarcated dry season, a localized Pacific wind pattern causes parts of the Osa Peninsula and localities around the Golfo Dulce to experience the intense rainfall typical of much of the Atlantic Watershed, up to five meters a year, with no detectable dry season. Although they support a plentiful and varied natural flora, large upland valleys like the General have mostly red, relatively infertile soils compared with those of the Central Valley and the Atlantic Watershed. Fundamentally, this is because the nonvolcanic Talamanca range has failed to provide the soil nutrients so generously bestowed on other parts of Costa Rica by the volcanoes running along the center of the country.

Of the three archaeological zones, Diquis, with the smallest number of controlled excavations, is the least understood. No radiocarbon (C-14) dates are yet available, and the regional sequence of archaeological phases must be considered tentative. Culturally, Diquis is considered to form part of the larger subarea known as Greater Chiriqui, which includes the Panamanian province of Chiriqui, to the southeast.

Periods I - IV (12,000 B.C. - 500 A.D. )
Nothing is yet known about periods I - III in Diquis. We pick up the prehistoric thread well into the second half of Period IV with the Concepcion and Aguas Buenas archaeological complexes. For specific data on chronology, settlement patterns, and subsistence, it is necessary to extrapolate trom published work carried out in Panama, where highland valleys or upland ridges seem to have been preferred locations for villages, a pattern also observed in Diquis. With the exception of a possibly oval or rectangular house (suggested by postholes) at the Panamanian highland site of Pitti - Gonzalez no house forms are known. Larger, probably ceremonial sites like Barriles, in Panama, and Bolas, in Diquis, include stone-faced earthen mounds and terraces or platforms.

Considerable uncertainty exists as to the spatial and temporal relationship of the Concepcion ("Scarified Ware") and Aguas Buenas ceramic complexes, which predominated in this zone from several centuries before Christ until c. 500-700 A.D.. Concepcion, known primarily from Chiriqui, is characterized by multiple-line incising or rough brushing on the naturally buff or brown clay exterior, giving the appearance of rough-edged, fine corrugation, hence "Scarified." Chimney-shaped and conical vessels, some with tripod feet, are typical.Clay roller stamps are known. Although featured prominently in the earliest archaeological monographs, very little of this pottery has since come to light.

Aguas Buenas pottery may be red, brown, or red on buff; small recurved-rim dishes and ollas are typical, and zoomorphic modeled adornos are a frequent decoration. Certain very large urns were apparently used to house secondary burials. The unusual Carbonera figurines, reputedly from the tip of the Osa Peninsula, are suggestive of a style at once older and foreign. The Osa may have offered a convenient point of arrival for seagoing peoples, in this case from South America.

Cobble metates, with and without tripod legs, open at one end and raised along the other three sides, are known from Aguas Buenas sites, as are many waisted, double-bitted axes of chipped stone. Some Barriles-like artifiacts have been recovered; carved stone barrels of andesite and granite, with low reliefs carved at either flat end, have been found near the Costa Rican town of San Vito. It is said that 15 were found around a huge petroglyph-bearing boulder called Piedra Pintada, where they may have functioned as seats. Also found in that region were fragments of stone figures, some with conical headgear like that known from the famous Barriles statuary. Shaft-and-chamber tombs at Barriles contained huge ellipsoidal metates, adorned along the edge with carved human heads and supported by four legs sculpted into detailed human figures. Missing in this region is intensive lapidary work in jade; although a very small amount has come to light, it seems that other materials available locally, like agate, were used.

Some archaeologists believe that the giant stone spheres of Diquis began to be made at this time. Some spheres are over two meters in diameter, yet vary from the perfectly spherical by a mere centimeter or two; they are made of granite, andesite, and even sedimentary stone, and weigh up to 16 tons. Certain balls found on the Diquis Delta were apparently rafted and hauled to their present location from places many kilometers distant. The spheres have been found in alignments on the surface. some mounted on cobble platforms. Although no burials have been found beneath them. the spheres are often grouped in the vicinity of a cemetery zone.

Since maize, beans, palm nuts, and avocados have been recovered from late Period IV sites in Chiriqui, we may infer a similar subsistence for Diquis populations at that time. Root crops were almost certainly important as well, but they leave few macroscopically identifiable remains.

Some Concepcion-phase tombs in Chiriqui, stone-lined and incorporating metates in the walls, suggest extended primary burials. Other Aguas Buenas tombs were simple oval pits, while Barriles had well-made shaft-and-chamber tombs, a form known early from Colombia and the Andean area. Urn burials are also reported. Burials are usually found beneath or around dwellings, but appear in a separate cemetary zone at Boquete, Panama.

Periods V and VI (500-1550 A.D.)
As in the Central Highlands-Atlantic Watershed zone, important cultural changes occurred in Diquis between 500 and 800 A.D., seeming to indicate an influx of South American peoples and/or cultural traditions, possibly the arrival and eventual hegemony of Chibcha-speaking peoples from Colombia.

In their survey of the Terraba (Diquis) drainage, Drolet and Markens found that Chiriqui-phase sites were most often located on broad terraces jjust above major water courses, suggesting a more intensive utilization of the major rivers in later times. While increased travel and commerce along these mostly navigable rivers were probably a factor, the extrodinary amounts of river cobbles used in the construction of Chiriqui-pase sites must also have played a part in the shifting pattern of settlements.

Several kinds of sites are known for Period Vl. Special burial grounds often incorporating cobble platforms and walls, as in the case of Sabana de Caracol, have been found on the summits of rather large hills. Cemeteries, rich in cast gold pendants and other metal artifacts, were discovered along high, sharp ridges, today overgrown with rain forest ( e.g., the Coquito cemetery). A second class of sites seemingly combined ceremonial and perhaps funerary activities with habitations. Finca Remolino incorporates large, low, stone-faced platforms of many shapes, including quadrangular ones, interspersed with what seem to be house mounds. On, or alongside, the platforms were embedded large natural monoliths (of collumnar basalt) up to four meters in length, some pecked to a pointed end for easier insertion. A third kind of site seems to been primarily habitational, with circular house foundations of cobbles, terraces with retaining walls, and cobble-paved walkways. This kind of site is most frequently found along major rivers; almost all have associated cemetetaries on terraces above the habitation zones .

Murcielago, a very large site (almost 4 square kilometers) of the third kind, has circular house foundations, 12 to more than 20 meters in diameter, surrounded by a gently sloping pavement of river cobbles. Although many manos and metates were found outside the houses, most chipped and pottery refuse was apparently deposited in unusual features composed of yellow and red oxidized subsoil rocks, which surround the houses. Whether these features were activity areas or dumps is unclear, but they were placed systematically throughout the site. A complex system of pavements and ramps connected living areas in any one part of the site. These new, important settlement data for Diquis, along with the previously known stone cist tombs, make up an "architectural" complex much like that known from the Central Highlands-Atlantic Watershed during Periods V and VI, and we may assume that both zones experienced the same pervasive "southern" influence during that time. In addition to stone cist tombs. we know that shaft-and-chamber tombs continued to be constructed into the Chiriqui phase.

There is a fairly large range of pottery types known from Diquis during the last six or seven hundred years before the Spanish arrival. Tall, hollow-legged tripods, slipped in reddish brown and usually decorated with applique fillets, pellets, adornos, or white-painted lines, recall somewhat similar (but earlier) vessels of the Atlantic Watershed In Diquis, the tripod legs are often in the form of crocodiles or fish. Organic black resist decoration, often combined with positive red paint, adorns white or orange-slipped small ollas, in patterns recalling Panamanian, and especially Columbian, ceramics. It is significant that small cylindrical, capped phials, of the kind used among Colombian cultures to carry lime for chewing with coca leaves, appear in Diquis in this ceramic type, since coca-chewing is a South American trait.

Chiriqui Polychromes are usually executed in black and red on a cream slip, with simple geometric motifs. While Panamanian motifs seem to dominate, there was some stylistic influence from northwestern Costa Rica, and a few Nicoya polychrome trade sherds have been found in Diquis. As in the rest of Costa Rica during Period VI, tripod supports shaped like animal (mostly feline) heads are diagnostic for this time in Diquis.

Perhaps the pinnacle of the ceramic craft in Period VI Diquis was the Tarrago Biscuit. This fine, buff-colored pottery may have walls less than two millimeters thick, and displays supremely simple yet elegantly proportioned shapes. Tiny modeled adornos frequently emphasize the voluptuous forms. Zoomorphic effigies include what is almost certainly an American camelid (llama, guanaco), an animal whose natural habitat extends only as far north as the Colombian Andes

Diquis stone sculpture of Period VI is radically different from that of the other two zones of Costa Rica, and, at the same time, reminiscent of Colombian forms. A link with the Atlantic Watershed is found in the tetrapod jaguar metates and circular "Atlantean" varieties that appear in both zones; ceramic models of the latter kind probably served as seats. Large stone spheres have been found around Chiriqui phase cemeteries.

Periods V and VI witnessed a florescence of the metallurgical craft, which produced quantities of pendants, bracelets, plaques, headbands, and other articles of gold or tumbaga. The Diquis gold-working style is basically the same as that of Chiriqui.

Conclusions
In a very general way, we can sum up prehistoric cultural development in Costa Rica as follows: starting c. 1000 B C. a few small and sedentary communities, in which pottery and perhaps a northern South American subsistence pattern, i e, mostly root cropping, were followed bv a rapid increase in population and social complexity, perhaps stimulated bv developing maize agriculture, complemented by polycropping (in regions of fertile, alluvial soils and abundant rain) and hunting. A culmination occurred around the time of Christ in sedentary, fairly large nodes of population, which were characterized by stratified society with complex ritual connections to Mesoamerican trade networks, and probably a redistributive hierarchy. In Guanacaste-Nicoya, a gradual shift to coastal foci occurred, accentuated markedly after c. 800 A.D.; it continued, with variable links (spiritual and commercial) to Mesoamerica, revealed by the transition from "Mayoid" to "Mexicanoid" iconography and art styles. Eventually, Greater Nicoya was defined as a buffer zone between Mesoamerica and tropical-forest cultures of southern origin. In the Central Highlands-Atlantic Watershed and (probably) the Diquis zones, the first five or six centuries A.D. saw sporadic intergroup resource competition and warfare, with head hunting and sacrifice of captives possibly indicating population pressure; the apparent intromission, c.500-700 A.D., of foreign (probably southern) peoples and tradition; changes in house and tomb forms; and the gradual degradation of ceramics, although not of other prestige items (gold replaces jade). The "balkanization" of these zones took place in the late period; they broke into relatively small, agglomerated, rudimentary-architecture settlements, for political control and defensive strategy, with occasional strong leaders who organized several centers into a site hierarchy or alliance for brief periods.

Why did the cultural evolutionary process in Costa Rica (and the Intermediate Area in general), similar in its early stages to that observed in Mesoamerica and Peru, sputter and stall? Why did no "urban" centers with large pyramid complexes appear? Robert Carneiro believes that historical evidence shows that no autonomous socio-political unit, large or small, will voluntarily relinquish sovereignty in the name of cooperation or the "greater social good." Only through forceful domination (war) are states and empires forged. Betty J. Meggers postulates that endemic warfare in an "open" environment like Amazonia. overtly waged for reasons like revenge, supernatural mandates, and the taking of exogamous marriage partners, is, in reality, a regulatory device for human population in an area with a precarious ecological balance. Warfare in Costa Rica may have functioned in this fashion, and may have been even more intense, given greater population densities. Why did this conflict not result in the amalgamation of larger, more complex socioâ€"political structures, as it apparently did in parts of Mesoamerica and Peru? The answer is that oppressed populations could successfully flee the threatened domination, emigrating to other, similar localities instead of being incorporated by force into the larger or more powerful conquering group. William Sanders and Barbara Price, in their essay on the development of "civilization" in Mesoamerica, note that it is not the lack of productive potential in tropical-forest areas like Amazonia that prevented the development of a complex society, but, rather, the presence of huge amounts of at least nominally agricultural land acting as an incentive to successful emigration. The juxtaposition of very different environments in Mesoamerica produced a cycle of competition and cooperation between "symbiotic regions," with growth and expensions trends in all participating "environmental niches", culminating in a socio-political whole bigger than the sum of its parts. Ironically, the abundance -- not the lack -- of viable ecozones may have stifled the cultural evolutionary development of much of the area between Mesoamerica and the Andes. In no way, however, did this detract from the graphic expression of a complex mythological "world view", which, in Costa Rica, combined elements of pervasive cultural traditions from both Middle and South America.

A Word on Terminology
In certain publications and textbooks, especially within Costa Rica, Guanacaste-Nicoya is referred to as the "Chorotega" cultural area, the Central Highlands and the Atlantic Watershed as "Huetar," and Diquis as "Brunka." The use of these names stems from a misunderstanding of the Spanish chroniclers. The names were those of individual chieftains or of larger socio-cultural groups that dominated part (but not all) of the three respective archaeological zones of Costa Rica around the time of the Spanish arrival, or, in the case of the Chorotega, some centuries before. To use the names as descriptive of the whole prehistoric cultural tradition of a zone is as misleading as referring to all the occupants of Manhattan Island during the last 10,000 years as New Yorkers. Although the actual names of many different peoples occupying Costa Rica when the Spanish came are known, we do not know, and probably never will know, what more ancient cultures called themselves. Costa Rica, unlike parts of Mesoamerica, has produced no evidence of a written, or hieroglyphic, record from Precolumbian times.

The Europeans who arrived in Costa Rica at the beginning of the 16th century observed indigenous cultures which in most cases have since been characterized by anthropologists as "chiefdoms," organized around a centralized, hereditary-status hierarchy with a theocratic orientation, but lacking the rigid social stratification and institutionalized means of forceful repression that are the products of civil law in a formal political state. The monumental architecture, writing systems, and calendrics that often characterize the state, or "civilization," are usually absent in the chiefdom. Instead, we see a succession of richly diverse styles in pottery, stone carving, lapidary work, and metallurgy, the preferred media changing through time. Craft traditions and religious symbolism are almost always highly developed, a result of the status-reinforcing needs of a "warrior-priest" chief and his coterie.

*****************
Michael J. Snarskis is one of Costa Rica's foremost archaeologists. A graduate of Yale University, Dr. Snarskis received his Ph.D. from Columbia University and directed the Archaeological Research Program at the National Museum of Costa Rica for 10 years and was a professor of archaeology at the University of Costa Rica for 14 years. He guides tours to Guayabo National Monument and the archaeological museums of San José. For further information, please contact Dr. Snarskis.