Traces Behind the Esmeraldas Shore

4. Vagrant Vestiges: Valdivia and the Mafa Phase

4 VAGRANT VESTIGES Valdivia and the Mafa Phase For many archaeologists, the earlier the better. While the most elaborate pottery dating to a mere A.D. 500 may be described in a technical monograph or regional journal, even the scrappiest of sherds, if dated to several millennia B.C., warrant a press release. Participants in the Santiago-Cayapas Project are not immune to this common archaeological ailment of the "earlies"; we always hoped that truly ancient remains would be encountered. Alas, preceramic and early ceramic occupations in Esmeraldas continue to be elusive. Only one flaked projectile point has been reported for the province, but this specimen is probably of ceramic age (Valdez 1987: fig. 45a). Although lithic scatters have been recorded in our survey, the associated chips, flakes, and chunks of bashed stone are indistinguishable from those known from ceramic contexts. Such nonceramic sites, therefore, probably represent special activity loci of ceramic age. At present, the beginning of our sequence is fully ceramic and not all that early Valdivia, Where Are You? One candidate for early ceramics would be, of course, Valdivia-that precocious pottery that continues to spawn a large and often contentious literature (Meggers et al. 1965; Lyon 1972-74; Lathrap et al. 1975; Damp 1988; Marcos 1988). In its heartland in southwestern Ecuador, Valdivia spans the entire third millennium and continues through the first half of the second millennium B.C. Its areal distribution encompasses El Oro province to the south and the Jama valley in Manabi to the north Oadan 1986). In the latter locale, the Valdivia occupation is assigned to the terminal phase VIII ofthe refined chronology developed by Hill (1972-74), a dating that conforms to a time-transgressive distribution northward along the Ecuadorian coast. If there is a Valdivia presence in the SantiagoCayapas area, therefore, one might expect it to be late within the Valdivia sequence . This expectation is in accord with certain ceramic features recorded in the area of the Esmeraldas valley that are said to be Valdivia related but, if such, are more likely to be Valdivia derived and of a later age (Guinea Bueno 1986). To the north, in the Tumaco region, the earliest discovered ceramics date to the latter half of the first millennium B.C. and do not evince any Valdivia vestiges (Bouchard 1985). The relevant evidence from the Santiago-Cayapas is exiguous at best, consisting as it does of two sherds from a mixed surface depOSit. But the sherds would appear to be vintage Valdivia. The sherds in question come from R53, a ca 0.2 ha surface deposit located about 100 m west of the Estero Maria (fig. 4.1). A 4 m high hillock is located at VAGRANT VESTIGES 69 ICM I ~ . . o Fig. 4.1 Distribution ojMaJa sites (solid circles). R53 is the siteyielding possible Valdivia materials (open circle). 70 VAGRANT VESTIGES Me . . eM Fig. 4.2 Valdivia sherds of the type Broad-line Incised. Top, R53. Center: Real Alto (Marcos 1988, II: fig. 458). Bottom, Meggers et al. (1965: plate 38j). the center of the site, but a shovel probe atop this elevation suggests that it is a natural formation. The surface collection from R53 is a mixture of Selva Alegre, Guadual, and other materials including the piece shown in fig. 4.2. This latter specimen has a dense, coffee-colored paste and well-smoothed surfaces. The exterior design consists ofnested stepped figures executed in broad-line incision. A second sherd, probably from the same vessel, has an identical color and paste but lacks incision. As seen in fig. 4.2, the similarity of the R53 specimen to the Valdivia type Broad-line Incised (Meggers et al. 1965: 47-51) is sufficiently specific to argue for this identification. Thus, however fugitive it might be, the R53 evidence tentatively supports a Valdivia presence in the Santiago-Cayapas, a presence almost missed in the survey and excavations carried out to date. Although the 3580 B.P. date from nearby R36 (table 3.12) is perfectly concordant with a late Valdivia time frame, the necessary association between sherds and dated carbon is lacking . Documentation ofsuch an association should be a target of future fieldwork. The Mafa Phase The Mafa phase is named after a left-bank estero of the upper Cayapas where the distinctive ceramics of the phase were initially discovered by Paul Tolstoy in his reconnaissance of December 1985. The Mafa actually consists...