AM ERI CA N AN TIQUITY VOL. XXI OCTOBER,1955 No. 2 THE ARCHAIC TRADITION IN PUERTO RICO RICARDOALEGRiA, H. B. NICHOLSON, AND GORDON R. WILLEY BOTH HISTORICAL and archaeological gladiolitos, and pieces of red ocher are diagnos- sources on the West Indies provide evidence tic features of the preceramic assemblage on that these islands were once occupied by this island. Harrington also found burials in peoples with a culture markedly different from cave floors; the skulls were undeformed and that of the agricultural,pottery-making Arawak. the bones often stained with red ocher. The The archaeological remains of this Archaic antiquity of these Cuban Archaic finds is mode of life are found on several of the islands demonstrated by their association with the in contexts which indicate their separateness bones of Megalocnus, an extinct ground sloth. from the pottery horizons and imply their The Archaic of Cuba has been divided into greater antiquity. The ethnohistorical sources 2 distinct phases, perhaps chronologically se- also attest to the survival of this Archaic tradi- quent, called by Rouse (1951a: 253) Guayabo tion into early historic times (Martyr 1912 Vol. Blanco and Cayo Redondo. The former con- 1: 100, 380; Las Casas 1867: 35; 1951 Vol. 2: tains no evidence of stone grinding and is 240; Velasquez 1869: 424-5; Oviedo 1852 Vol. characterized by what Rouse refers to as the 1: 90). "battering stone" and deep vessels made of the The characteristics of this West Indian conch. Cayo Redondo, on the other hand, Archaic tradition, as revealed by archaeology lacks the deep conch shell vessels, and has and supplemented by the ethnohistoric ac- a ground-stone inventory consisting of various counts, are: absence of agriculture and pottery; grinding implements, stone balls, and the seminomadic living in small bands; frequent curious peg-shaped stones, the gladiolitos. use of caves for shelter and burial; crude arti- On the island of Hispaniola the existence of facts made of conch shells, flint, and other Archaic cultures in prehistoric times has been classes of stone; use of hematite or red ocher; demonstrated by the work of Gabb (1881), and the absence of cranial deformation as Krieger (1929), and Rouse (1939; 1941). The practiced by the agricultural peoples of the early historical accounts also mention the Antilles. existence of primitive pre-agricultural groups Cultural tradition is here defined, following who lived in the Peninsula of Guacayarima Phillips and Willey (1953), as "a major large- up to the time of the conquest, and who may scale space-time-cultural continuity, defined have been marginal surviving remnants of the with reference to persistent configurations in old tradition (Martyr 1912 Vol. 1: 380; Oviedo single technologies or total (archaeological) 1852 Vol. 1: 90). culture, occupying a relatively long interval of Rouse subdivides the Archaic of Haiti into time and a quantitatively variable but environ- 3 phases: Couri Cabaret mentally significant space." (northern); (central); and Bay of Conch In Cuba the Archaic tradition is represented (southern) (Rouse 1951a: by the preceramic finds of Harrington (1921), 253). The chief diagnostic of Couri is an as- labeled by him with the controversial ethnic semblage of crude stone tools "made by finely term, "Ciboney," and by those of Rouse (1942: rechipping the edges of large lamellar flakes" 131), Osgood (1942), Ortiz (1943), Coscul- (p. 253). Ground stone is also well represented luela (1922), and others. Rock-shelter and in this phase and includes single- and double- coastal shell-midden sites are typical of the bitted axes, balls, dishes, beads, gladiolitos, Cuban Archaic. Shell gouges. shell celts, vessels mortars and pestles, and milling and polishing and dishes made of conch shells, shell or stone stones. The Bay of Conch inventory is similar pendants of simple design, pitted hammer- to Couri, with the addition of pointed stone stones, hammer-grinders, stone balls, slightly tools and single-bitted eared stone axes. Caba- retouched flint flakes. ne,-s-hnned st-nne nr ret is much the simplest of the 3 phases. Only 113 114 AMERICANANTIQUITY [ XXI, 2, 1955 chipped stone work is found, comparable to scribed. No Archaic remains have as yet been Cuba rather than Couri or Bay of Conch. reported from the Lesser Antilles, but much In the Dominican Republic both Gabb and of this area is an archaeological terra incognita. Krieger dug into cave stratifications in the It is not until Trinidad (Rouse 1953) and vicinity of Samana Bay in which upper level -Venezuela (Rouse 1951b) are reached that evi- deposits held pottery while the lower strata dence occurs of comparable nonceramic phases. contained only implements of shell, bone, and The putative preceramic or Archaic mani- stone. Krieger (1929: 68) holds that these arti- festation in Puerto Rico, which Rouse has facts show great similarity to those found by labeled the Coroso, is the subject of this paper. Harrington in his preceramic sites in Cuba. Recent excavations have brought to light new For Puerto Rico there are no ethnohistorical evidence which proves its existence. We also references to nonagricultural groups similar to wish to make some comparisons which have those mentioned by the chroniclers in Cuba not been made before. and Hispaniola; however, their prehistoric ex- istence has been postulated by Rouse (1952: TH E LOIZACAVE EXCAVATIONS 355). During his survey of Puerto Rico and One of the most obvious desiderata con- Vieques, Rouse located several sites with small nected with the problem of a possible pre- shell heaps which appeared to contain no pot- ceramic occupation of Puerto Rico was the tery. Excavation at 5 of these sites verified this locating of an actual occupation site over and absence of pottery, but the material recovered above the scanty shellheaps investigated by was so meager and of such dubious artifactual Rouse. In 1948, Ricardo Alegria, as part of the character that Rouse hesitated to claim with program of the newly organized Centro de certainty that a genuine preceramic occupation Investigaciones Arqueologicas of the University of these islands had been demonstrated. Al- of Puerto Rico, initiated excavations at a rich though he felt that it was unlikely that these ceramic site near Loiza Aldea, close to the nonceramic shell heaps were "the by-product northeast coast. A few hundred yards south- of a later Indian or Spanish occupation" (Rouse west of this site is a large limestone cave, 1952: 382), he was convinced that further known locally as Cueva Maria de la Cruz. excavations were necessary before the problem Portions of its rocky floor had long been utilized could be resolved. All the sites were located as a stone quarry, but near the western entrance close to the shore in regions where shellfish was a fairly extensive area of hard-packed were, or had been, abundant. In all cases earth mixed with detritus fallen from the ceil- "there is a bay in the immediate vicinity, ing which appeared to be relatively undis- backed by a mud flat or a swamp which could turbed. Supplementary to the excavation of have been used as a base for hunting and fish- the ceramic site, 2 test pits, totaling 5 sq. m. ing" (p. 335). In these respects the sites re- in area, were dug into a portion of this floor semble the nonceramic shell heaps of Cuba and near the western wall of the cave. The stone artifacts de- Hispaniola. possible The pits were dug by 1-foot levels. The 1st non- scribed by Rouse from these Puerto Rican level yielded a number of Igneri phase sherds, "several bat- ceramic sites include: pebbles including 1 large rim fragment of the diagnostic tered on their another with several ends, grind- Cuevas white-on-red ware, 7 stone artifacts number of of ing facets, a sharp-edged pieces quite unlike those found in the nearby ceramic or other and several flat stone slabs" flint stone, site (all artifacts will be described in the next (p. 335). section), large numbers of shells, animal and To no evidence of the Archaic has date, fish bones, nondescript fragments of rock, and has a appeared in Jamaica. Krieger reported a few scattered human bones. The 2nd level possible nonceramic assemblage from the Baha- yielded only 5 sherds, plus 5 artifacts of stone mas (Krieger 1937: 98), but it has not been and an even thicker of shells and fully described or illustrated. In the Virgin deposit 3rd and levels were Islands Hatt (1924) reported nonceramic refuse animal bones. The 4th deposits in shell heaps at Krum Bay, St. completely pottery-free. The heavy concentra- Thomas. Hammerstone grinders, red ocher, tion of animal bones (many of them fossilized) and a "peculiar long and narrow type of stone and shells continued. Most importantly, many axe" are listed as associated artifacts, but the of the animal bones showed evidence of burn- full assemblage has never been thoroughly de- ing, and thick deposits of gray ash were en- ALEGRiA AND OTHERS1 THE ARCHAIC TRADITION IN PUERTO RICO 115 FIG.37. Tools from the preceramic or Archaic levels of the Loiza Cav-e, Puerto Rico. a, circular pitted ham merstone; b, iectangulate pitted hammerstone with one edge (right side) smoothed by grinding; c, scraper made from conch shell; d, pebble grinder showing grinding surface at left; e, pebble chopper showing chopping edge at left. The pebble grinder and chopper ate the most distinctive types of this Loiza assemblage. (Dimensions: a, 8.5 by 4 cm.; b, 11 by 8 by 3.5 cm.; c, 10 by 6.5 cm.; d, 12.5 by 6 by 4 cm.; e, 10 by 7.5 by 4.5 cm.) 116 AMERICAN ANTIOUITY I xxi.