Machalilla Phase Salango

The Ecuadorian Middle Formative Machalilla phase (Lippi Reference Lippi1983; Meggers et al. Reference Meggers, Evans and Estrada1965; Zeidler Reference Zeidler, Silverman and Isbell2008) is dated to 1450–800 cal BC (Zeidler Reference Zeidler, Scott Raymond and Burger2003). Settlements extended 200 km along the central coast from the Río Chone down to the Gulf of Guayaquil, with occasional outlier sites in both directions. Much of this land was hilly and covered in tropical dry forest, with cloud forest at the peaks and scrub and mangrove in the lower reaches and around the river estuaries.

Although Machalilla was initially distinguished from Early Formative Valdivia by changes in pottery vessels and figurines (Lippi Reference Lippi1983; Meggers et al. Reference Meggers, Evans and Estrada1965), there were also shifts in settlement pattern and, it was inferred, in social structure (Raymond Reference Raymond, Scott Raymond and Burger2003; Schwartz and Raymond Reference Schwarz and Scott Raymond1996; Zeidler Reference Zeidler, Silverman and Isbell2008). No mainland ceremonial centers have yet been identified, but the offshore Isla de La Plata was visited as an oceanic sanctuary (McEwan and Lunniss Reference McEwan, Lunniss, Beekman and McEwan2022). Subsistence was based on sea fishing, shellfish collection, and mixed agriculture with an intensification of maize production; however, hunting and collecting forest fruits and other products such as honey were also practiced (Béarez et al. Reference Béarez, Gay and Lunniss2012; Lippi et al. Reference Lippi, Bird and Stemper1984; Pearsall Reference Pearsall, Scott Raymond and Burger2003; Stahl Reference Stahl, Scott Raymond and Burger2003; van der Merwe et al. Reference van der Merwe, Lee-Thorp, Scott Raymond, Lambert and Grupe1993). Interaction with the Ecuadorian Andes is suggested by the similarity of Machalilla ceramics with those of Cotocollao and Catamayo (Guffroy Reference Guffroy2004; Villalba Reference Villalba1988; Zeidler Reference Zeidler, Silverman and Isbell2008). Thus, Machalilla was an expression of novel cultural traits that were shared across the larger part of the central coast but also reached to the central and southern highlands 300 km or more away. At Salango, meanwhile, more local aspects of identity arose through specific relationships with place.

Salango lies at the midpoint of the Machalilla coastline. Its history as a sea-fishing site begins around 4000 cal BC with Late Archaic and then Valdivia occupations (Lunniss et al. Reference Lunniss, Zeidler, Aguilú and Jadán2021; Norton et al. Reference Norton, Lunniss and Nayling1983). Archaeological remains are concentrated at the south end of a sheltered sandy bay that harbored abundant and diverse fish species (Béarez Reference Béarez1996; Béarez et al. Reference Béarez, Gay and Lunniss2012). Three sea-turtle species nest at Salango and nearby (Miranda Reference Miranda2019), and there are many invertebrate species (Ministerio del Ambiente del Ecuador 2007), including the bivalves Pinctada mazatlanica, used for making fishhooks (Béarez et al. Reference Béarez, Gay and Lunniss2012), and Spondylus crassisquama, a symbol of marine sacred essence then of growing importance in the Andean world (Carter Reference Carter, Infantidis and Nikolaidou2011; Lunniss Reference Lunniss, Beekman and McEwan2022; Moore Reference Moore, Barber and Joyce2017).

The Machalilla phase villages at nearby Puerto López and the eponymous Machalilla site occupied raised marine terraces overlooking the beach and sea (Meggers et al. Reference Meggers, Evans and Estrada1965; Prümers et al. Reference Prümers, Ugalde and Dyrdahl2022). At Salango, however, the area chosen was at the foot of the headland, close to and just above the high-tide line. Artifact-rich deposits extend across 1 ha, with further stratified remains over another 2 ha to the north and south.

Evidence for the Machalilla village came in 1982 from excavation of a 6 × 6 m cut at OMJPLP–141A (Figure 2). At 2–3 m below ground level, 75 cm of black sandy layers with large quantities of fish bone, charcoal, pottery, shellfish hooks, and other artifacts overlay 25 cm of lighter-colored, less organic deposits (Béarez et al. Reference Béarez, Gay and Lunniss2012; Norton et al. Reference Norton, Lunniss and Nayling1983). Thin layers of yellow clay and postholes suggest the presence of houses or more ephemeral structures, and other features include rubbish pits and fire pits. The main set of 30 Machalilla individuals was found in this sector, along with complete, large chicha (maize beer) production or storage jars that were buried nearby. Kitchen jars and decorated fine ware bowls, jars, and stirrup spout bottles were also represented, along with occasional fragments of solid pottery figurines and larger hollow figurines (Figure 3). In the same year a 2 × 1 m cut at OMJPLP–141B, 20 m to the west (Norton et al. Reference Norton, Lunniss and Nayling1983), exposed another Machalilla burial, and in 1983 similar black sandy deposits and possible house floors were found at OMJPLP–141C, 50 m to the north of 141A and 141B, along with two possible Machalilla phase burials (Norton et al. Reference Norton, Lunniss and Nayling1983). Finally, in a 2014 watching brief, cultural deposits with fishbone and Machalilla pottery and figurine fragments were found along Calle Larga from a point 150 m south of 141A to Calle 22, which is 150 m north of 141A and close to the old river estuary that marked the northern edge of the settlement (Lunniss Reference Lunniss2016).

Figure 2. Map of the Machalilla site at Salango (illustration by Luke Dalla Bona).

Figure 3. Face fragment of a large hollow Machalilla figurine from Salango Sector 141A.

That Salango was a fishing village with a focus on pelagic tuna is indicated by abundant fish remains and fishing equipment (Béarez et al. Reference Béarez, Gay and Lunniss2012), stable isotope analysis of the human skeletal material (van der Merwe et al. Reference van der Merwe, Lee-Thorp, Scott Raymond, Lambert and Grupe1993:77), and fishing net designs on pottery vessels (as discussed later). Other lines of evidence illuminate the social and ritual context.

First, the figurine fragments reference two spheres of ritual practice across the Machalilla landscape. The simpler solid and sometimes red-painted figures (Lathrap et al. Reference Lathrap, Collier and Chandra1975; Lippi Reference Lippi1983; Meggers et al. Reference Meggers, Evans and Estrada1965) are comparable with solid Valdivia figurines perhaps used as instruments of life-cycle and curing rituals (Stahl Reference Stahl1986; Stothert Reference Stothert, Scott Raymond and Burger2003; Zeidler Reference Zeidler, Canuto and Yaeger2000). The hollow painted figurines—in fact, anthropomorphized bottles—share details with the solid figures such as ear piercing and ornamentation with bands of red paint; however, they present a new form of imagery that indicates both a new perception of human identity and a new type of ritual performance (Lathrap et al. Reference Lathrap, Collier and Chandra1975; Lippi Reference Lippi1983; Stothert Reference Stothert, Klein and Cevallos2007). There is yet no direct evidence to suggest why, when, or where rituals took place involving either type of figurine, although one ear from a hollow figure was found buried on top of a large chicha jar. It can be suggested, however, that the simpler and smaller solid figures were used for personal or household ceremonies and the larger, more elaborate hollow figures for more public or community-oriented performance.

Second, stable isotope analysis indicated that the predominantly marine diet also included maize, likely obtained via reciprocal relations with agriculturalists in exchange for the fish and shellfish harvested at Salango (van der Merwe et al. Reference van der Merwe, Lee-Thorp, Scott Raymond, Lambert and Grupe1993:77). The large jars buried near the human interments suggest that this maize was at least in part consumed as chicha, that chicha was drunk in commensal events, and that chicha and chicha consumption were associated with celebration of the dead (Stothert Reference Stothert, Scott Raymond and Burger2003). That the jars were found complete and generally upright suggests that they may have contained chicha when buried.

The jars are of two classes. First are vessels with undecorated bodies that measure from 260 mm tall and 330 mm wide to 380 mm tall and 456 mm wide; their volumes range from 14 L to 30 L. These are associated with the end of the burial sequence and were found singly, paired, and once in a group of four (Figure 4). They correspond to globular jar forms as suggested for Cabuya Plain by Meggers and colleagues (Reference Meggers, Evans and Estrada1965:121, Figure 75.5). A larger form with an embellished neck and the entire exterior decorated with a banded design in red paint on a buff background stands 600 mm tall and 580 mm wide, with a capacity of around 100 L (Figure 5). Perhaps used from an early stage of midden development, these jars match Meggers and coworkers’ Machalilla Red Banded type: although the form is previously undescribed, the treatment of the neck is a variant of the “step-like projection” below the neck of Form 12 for that type (Meggers et al. Reference Meggers, Evans and Estrada1965:134–136, Plates 149–151). The reticulated design of the large chicha jar shown represents a fishing net, and even though the jar expresses the mutually necessary values of maize agriculture and sea fishing, the design itself prioritizes the latter.

Figure 4. Four large chicha jars buried intact adjacent to area of the Machalilla burials at Salango Sector 141A.

Figure 5. Massive Machalilla chicha jar decorated with fishing net design from Salango Sector 141A.

Both classes are larger than Valdivia jars (DeBoer Reference DeBoer, Raymond and Burger2003) and indicate significant changes in ceramic technology and the socioeconomic aspects of chicha brewing and consumption. At the contemporary central and southern highland sites of Cotocollao and Catamayo there were storage vessels of similar size to the smaller large jars at Salango (DeBoer Reference DeBoer, Raymond and Burger2003). Salango thus reflects a more widespread change in pottery technology and feasting customs. But the larger jars are of a different order of magnitude and, with their careful design and decoration, were surely designed for chicha consumption that was both large in scale and formalized in character. Notably, they match in volume the chomo ani—ceremonial chicha jars—of historic Shipibo of the Peruvian Upper Amazon (DeBoer Reference DeBoer, Raymond and Burger2003), whose chomo anitama, quotidian storage jars, match in volume the undecorated smaller large jars at Salango.

Third, there were nine exotic pottery sherds of a polished black or dark-brown fine ware, with broad, deep incised, and punctate designs on the exterior alone or on both the interior and exterior, sometimes infilled with white pigment (Figure 6; Norton et al. Reference Norton, Lunniss and Nayling1983:49, Figure 12). Scattered throughout the burial sequence but not found in any grave, these vessel sherds, as well as part of a roller stamp, are unlike anything yet described for pottery complexes of the Ecuadorian coast, highlands, or eastern lowlands. They are, however, matched by two sherds each from Machalilla and Cabuya that were explained as deriving from contact with a “Kotosh-like complex” of the Peruvian highlands (Meggers et al. Reference Meggers, Evans and Estrada1965:141, 173–174, Plates 158 g–i, 191). Obsidian flakes, also found at La Ponga (Lippi Reference Lippi1983:179–182), are another inbound element of exchange reaching Salango and derived from highland sources near Quito, which is 300 km away (Burger et al. Reference Burger, Asaro, Michel, Stross and Salazar1994).

Figure 6. Fragment of exotic incised ware from Salango Sector 141A.

In brief, life at Machalilla phase Salango extended beyond utilitarian concerns and fishing. At feasts chicha was served from decorated jars of unprecedented magnificence and drunk from bowls of equally novel elegance. Household and public ceremonies centering on anthropomorphic figurines linked the community to the wider world of Machalilla settlement and ritual. And Salango, looking across the ocean to an island sanctuary that marked the northwest horizon, was itself the destiny of rare and precious items brought from distant communities in probable exchange for Spondylus and other products of the sea.

The Machalilla Burials at Salango

OMJPLP–141A was divided into six units of 2 × 3 m each; three of these units on the east side were later extended, as was the southwest corner, to allow recovery of burials reaching beyond the original excavation area (Figure 7). Twenty-seven burials were registered, including three double interments, all set in oval or circular pit graves distinguishable for their darker fills. The burials present as two sets, Group A (14 burials / 16 individuals) and Group B (13 burials / 14 individuals), separated by an unused area a meter wide that ran north–south between them. The skeletons were often poorly preserved, but with one certain and one possible exception, all individuals appear to have been buried as fully articulated corpses. There was also a single burial at OMJPLP–141B, 20 m to the west. In total, then, 31 Machalilla individuals were identified during excavation at 141A and 141B. Determinations for age and biological sex (for adults) were made by Ubelaker in Quito in 1984 following procedures described by Gilbert and McKern (Reference Gilbert and McKern1973), McKern and Stewart (Reference McKern and Stewart1957), Phenice (Reference Phenice1969), Steele (Reference Steele1976), Stewart (Reference Stewart1958), and Ubelaker (Reference Ubelaker1978). Methods and theories used were limited to those available at the time of analysis. These determinations indicated 12 infants up to 4 years old, 12 adult females of 15–50 years, and 7 adult males of 30–50 years. Full details are presented in Table 1.

Figure 7. Plan of Salango Sector 141A showing distribution of Machalilla human burials among suggested Groups A and B. The circled numbers indicate the center points of the graves (illustration by Luke Dalla Bona).

Table 1. Machalilla Phase Human Burials at Salango, Sectors OMJPLP–141A and 141B.

^a Inds = Number of buried individuals.

^b Or = Orientation.

Relative and Absolute Chronology

At 141B, Burial 30 lay within the midden matrix. At 141A, Burials 1–24 were associated with the later stage of midden formation. Burial 25 may belong to the early stage of the midden and Burial 26 may predate the midden, but both match the configurations of the others. Thus, although there appears to have been increased interment in the later stage, the burials can be treated as elements of a single continuous tradition.

Further evidence for relative sequencing is provided by the direct superpositioning of graves, indirect stratigraphy, and measured levels below the ground surface (Figure 8). The longest sequence so far identifiable includes five individual burials or sets of burials running from Burials 26 to 6 via Burial 25, Burials 15 and 16, and then Burials 8a, 8b, 9, 10, 11, and 12. The remaining burials all fall within other sequences of four successive individual burials or sets of burials, where in each case again the earliest two burials are Burials 26 and 25. These strings were, however, doubtlessly integrated within a longer actual sequence.

Figure 8. Sequence diagram of the human burials and radiocarbon dates from Salango Sector 141A (illustration by Luke Dalla Bona).

This ordering is supported by 13 dates obtained from bone and charcoal samples assayed in 1983 by the University of Cambridge Sub-Department of Quaternary Research and calibrated using the OxCal v4.4 program (Bronk Ramsey Reference Bronk Ramsey2009) and the SHCal20 southern hemisphere calibration curve developed by Hogg and others (Reference Hogg, Heaton, Hua, Palmer, Turney, Southon and Bayliss2020; see Table 2). Seven samples were taken from seven tombs, all associated with the upper 50 cm of the midden. At 2σ, they suggest a range of 1496–922 cal BC. Another sample taken from the lowermost level of the midden, probably predating Burial 25, gave a date range of 1506–1266 cal BC. Two samples from levels beneath the midden but roughly contemporary with Burial 26 gave ranges of 1534–1281 cal BC and 1496–1226 cal BC. Combined, these results date the Machalilla occupation and burials to 1534–922 cal BC. Three samples with an overall span of 1743–1286 cal BC, taken from layers beneath those associated with the graves, suggest occupation reaching back into the Late Valdivia (Zeidler Reference Zeidler, Scott Raymond and Burger2003).

Table 2. Radiocarbon Dates for Salango, Sector OMJPLP141A.

Note: Official laboratory numbers were never given, and the numbers used here are those provisionally assigned when the samples were submitted.

Burial Distribution, Population Composition, and Burial Configurations

Although the total number and distribution of the Machalilla burial population at Salango are unknown, 141A and 141B show that the original assemblage extended at least 26 m in an east–west direction, and 141C showed that it possibly reached 50 m to the north. The distribution pattern at 141A, with the graves forming two groups, then suggests that there may have been several separate groups, with at least one other represented by the single grave at 141B and yet another perhaps at 141C.

How the burials were situated with respect to the overall settlement design is also uncertain, but present evidence suggests that they were clustered at and around the central area of the site currently occupied by the modern fish factory. In early Machalilla times, this was a residential area with perhaps an occasional human burial conducted there, such as Burials 26 and 25, but later this area was chosen more expressly for the placement of the dead.

The distribution pattern at 141A has further implications. Of the two groups of graves identified there, Group A included five infants, six adult females, and five adult males, and Group B included seven infants, six adult females, and one adult male (Supplemental Table 1). Though neither group has been completely recovered, the joint presence of adult females, adult males, and children; the stratigraphy; and the radiocarbon dates suggest these separate groups may represent family or kinship units spread over several generations. Alternatively, the two main burial groups may have referenced spatial settlement division or moieties such as suggested for the Valdivia community at Real Alto (Zeidler Reference Zeidler1998; Zeidler and McEwan Reference Zeidler, McEwan, Hoopes and McEwan2021), though at Salango there is no other suggestion of a dualistic structure.

Two infants were buried with adults, one in front of an adult male (Burial 25) and the other on the back of an adult female (Burial 6). In one instance, two infants were buried side by side (Burial 3). Other burials seem also to have been intentionally sited next to each other, whether back to back and headed in opposite directions (Burials 10 and 11, 20 and 21) or in some other relation of proximity (Burials 13 and 14, 11 and 12, 8a and 8b): these burials suggest another level of conscious relational organization within each group.

Burials were of complete articulated bodies, except for Burial 13 (discussed later) and perhaps Burial 22, where bodies were placed in a variety of positions (Table 3). Among burials with complete bodies, adults could be flexed and seated (n = 7), flexed on the left or right side (n = 6), or fully flexed on their backs (n = 5). The one exclusive tendency that is suggested relates to adults who were laid on their side: all females so positioned (n = 4) were on their right sides, and the two males were on their left sides (Béarez et al. Reference Béarez, Gay and Lunniss2012: Figure 5). Except for the two infants associated in double burials with individual adults, infants were found on their backs or semi-flexed on one or the other side. There is no evidence for any preferred orientation, either for adults collectively, men, women, or children.

Table 3. Positions according to Age and Sex of Machalilla Individuals Buried at Salango, Sectors OMJPLP–141A and 141B.

In addition to the offerings of Burial 13, certain objects found close to buried individuals or in their tombs would also have been grave goods. Pincatada mazatlanica pearl shells were set by the arm of Burial 15, by the feet of Burial 7, and by the hip of Burial 25, with a Spondylus valve and three sherds by the feet of Burial 7. An exotic sherd with Burial 13; shellfish hooks with Burials 5, 11, 23, and 30; and stones with Burials 24 and 30 may also have been intentionally placed with the bodies. One of the infants of Burial 3 and the infant of Burial 25 were both covered in some way by pottery. Thus, although pottery sherds and stones, as well as shellfish hooks, were present, an association with complete shells, also present in Burial 13, is especially indicated.

Extraneous human bones were associated with four burials. Infant Burial 19, 40 cm from the nearest neighboring tomb, was accompanied by the remains of one other infant and four adults. These included a complete male cranial vault with occipital flattening, and although the modification is not as extensive as that of the skulls at La Cabuya and San Pablo (Munizaga Reference Munizaga, Meggers, Evans and Estrada1965:228–229), it adds to the evidence for Machalilla phase cranial modification. The adult male Burial 30, also isolated, was accompanied by an adult right mandible. Accidental contamination can be ruled out because neither of these two graves touched any other previous burial. Thus, the bones were derived from graves situated elsewhere, though why or how the bones were selected and removed from their original locations is unclear. These two cases suggest that the addition of adult bones to Burials 10 and 11 was also intentional.

Burial 13: Female Authority and the Spirit World

In Group B, Burial 13 was an adult female of 35–40 years who, before interment, was dismembered and beheaded. The torso, with legs and arms beneath and to either side, was then wrapped in a bundle and set in the grave with the pelvis to the east and the neck to the west. The head was found over the center of the south edge of the main packet. Evidence for dismemberment, rather than secondary burial, is that the torso, legs, and arms were each found with their respective elements still articulated, indicating that the flesh was present at the time of burial. At the west (neck) end of the bundle, there was a large ceramic bowl set on one of its long sides, leaning over and so containing or protecting the bones (Figure 9). There was also a round stone (10 cm diameter) with two valves of P. mazatlanica above and below, each facing upward—with the upper valve covering the head while the stone and second valve lay a little farther to the south. A third P. mazatlanica valve, also facing upward, lay over the neck end of the bundle, a second stone was reported, and there was an exotic sherd in the fill.

Figure 9. View of the pottery vessel protecting Burial 13.

Treatment of the corpse and the overall burial configuration differ from those of all other Machalilla burials at Salango and identify the individual as having been of considerable importance. Such importance has antecedents in the high social, ritual, or spiritual status indicated by the treatment and siting of females found buried in Middle and Late Valdivia contexts (Marcos Reference Marcos1988:163–165; Staller Reference Staller, Staller and Currie2001; Zeidler Reference Zeidler, Canuto and Yaeger2000; Zeidler and McEwan Reference Zeidler, McEwan, Hoopes and McEwan2021). Disarticulation of the corpse was also an earlier feature of Middle Valdivia ritual at Real Alto, but that involved young adult males whose interments were elements of an extended ritual associated with the primary burial of a high-status female (Marcos Reference Marcos1988:163–165; Zeidler Reference Zeidler, Canuto and Yaeger2000). Burial 13 thus echoes an earlier tradition of female authority but repurposes dismemberment, indicating that it was significant here in terms of the authoritative identity and function of the individual so treated. In this context, disarticulation of the corpse and the protecting presence of the pottery vessel, both unique and both requiring explanation, provide intersecting lines of evidence to suggest that this individual was a shaman.

First, dismemberment is central to ideas about the ritualized death and rebirth of shamans (Eliade Reference Eliade and Trask1964:33–66; Sullivan Reference Sullivan1988:400; Wallis Reference Wallis2016:741). We suggest that the Burial 13 individual, having been a shaman in life, was dismembered at death before being repackaged for rebirth as a shaman in the afterlife. Contemporary Mapuche reconstruct the memory of deceased Mapuche machi following social disremembering, rather than repackaging of the corpse after physical dismembering; via this unique process, the identity of the dead machi is reconstituted so as to incorporate the ancestral qualities of earlier deceased machi (Bacigalupo Reference Bacigalupo2016:198–225). Shamanic practice and community relations with shamans undergo constant change and reinvention, often leading to specific situations that have no other direct parallel (Vitebsky Reference Vitebsky and Harvey2000), and Burial 13 is perhaps another such case.

The ceramic bowl adds to the argument. Unprecedented among Machalilla pottery collections, it was found complete and with no signs of use or predepositional damage: the bowl was of ordinary brown ware, lightly wiped but undecorated. It had an oval shape measuring 43 × 38 cm and was 16 cm deep (Figure 10). What would normally be described as the base was rounded, and there was a shallow rounded groove around the exterior that separated the slightly flaring rim from the main body. The ends were shallower with respect to the “base” than the sides, and although the vessel displayed general irregularities, the change of curvature of the rim at the two ends seems more probably intentional than accidental.

Figure 10. Drawing of the pottery vessel placed as protective covering for Burial 13, showing side (top) and end (bottom) views (illustration by Luke Dalla Bona after Norton and colleagues Reference Norton, Lunniss and Nayling1983:Figure 11).

As a container of liquids or solids, it would have been awkward and unstable, and although such a normal vessel function is possible, the depositional context combined with morphology points in another direction. First, the slight disorder of Burial 13 suggests that the bowl may have been originally on top of the overall arrangement of bones, shells, and stones; it may have later slipped to one side and the skull, stones, and shells slipped to the other. In any event, the bowl appears to have been so placed as to contain and protect the body. Second, given its undamaged condition at the time of burial and its unusual form, the bowl is likely to have been made specifically to accompany the individual in the tomb. Third, the vessel's oval form matches that of the shells of sea turtles. Not only does the treatment of the ends mimic that of the elevated head and tail ends of turtle shells in general but the rim also echoes the narrow flange that runs round the edge of the shell of the green sea turtle (Chelonia mydas)—the most common of the species that still visit beaches at and near the Salango site. Given the presence and likely human interaction with these creatures at Salango, it can be suggested that the bowl was made and used as a representation of a green sea-turtle shell.

A Terminal Valdivia young adult female buried at the Capaperro site in the Jama Valley was accompanied by shamanic paraphernalia that included an ocelot (Felis pardalis) skull likely once attached to a cape made of an entire skin; it was concluded that this would have represented the feline aspect of the young shaman's being that was invoked for purposes of communication with the world of the spirits (Zeidler et al. Reference Zeidler, Stahl, Sutliff, Oyuela-Caycedo and Scott Raymond1998). Similarly, we suggest that given its imitation of the form of a sea-turtle shell, the pottery vessel of Burial 13 referenced a sea-turtle spirit that was the woman's alter ego and that this woman was a shaman who had embodied, on behalf of the community, a sense of shared identity between the Machalilla people of Salango and the sea turtles who came to that place. The burial of her reconfigured remains, under the vessel, would thus have confirmed her own identity as a sea-turtle spirit in the afterlife.

Human interaction with sea turtles on the Ecuadorian coast is evidenced by Cheloniidae remains from Valdivia onward (Stahl Reference Stahl, Scott Raymond and Burger2003). More significantly, each of two burials dating to 1000–600 BC at Playa Vicente Mena in Chile is reported to have been covered by a real sea-turtle carapace (Frazier Reference Frazier, Lutz, Musick and Wyneken2003), and a black ware bowl from Ancón in Peru, dating to the Initial period (2000–600 BC), carried a complex, dualistic, incised image of a supernatural sea turtle (Burger Reference Burger1992:Figure 47). These separate and roughly contemporary instances echo central aspects of the ceramic vessel at Salango: its function as protection for the dead individual and its reference to the spiritual realm that she embodied. They thus show that there was a widespread, if rarely documented, impulse to integrate human society with the spirits of these creatures of the sea, and they provide support for our interpretation.

The stones and shells are of another order. The set comprised by the one stone with shells above and below recalls two Valdivia offerings in this same area of Salango, where P. mazatlanica valves enclosed Strombus peruvianus and galeatus conches (Norton et al. Reference Norton, Lunniss and Nayling1983). Now, it was a stone, perhaps a hand-grinding stone, rather than a conch, that was associated with the pearl shells, thus making possible reference to the maize that was consumed by these fishing folks.

Comparisons with Valdivia

Very young children represent 38.71% of the Machalilla phase assemblage at Salango (Supplemental Table 1), a proportion almost double that registered for Valdivia times at Real Alto, where children aged up to 4.9 years comprised just 21.00% of the sum assemblage (Ubelaker Reference Ubelaker, Scott Raymond and Burger2003). The 5.0–14.9-year age range, however, which represented 16.00% of the Real Alto assemblage, was absent at Salango. This perhaps reflects sample bias rather than some selection procedure or a zero-mortality rate for that age group. But if we look at Salango's mortality figures in two main age groups of 0–19.9-year-olds and 20.0–50.0-year-olds (Supplemental Table 2), at 45.16% and 54.84% respectively, they are similar to those for comparable age groups at Real Alto: 46.00% and 54.00%. In other words, despite the absence of the 5.0–14.9 age group, the Machalilla burials at Salango indicate a general population structure like that for Valdivia times at the one settlement for which there are substantial and systematic data. Likewise, although surviving offerings at Salango reflect the site's proximity to and dependence on the sea, their limited number compares with that of Valdivia burials at Real Alto (Zeidler et al. Reference Zeidler, Stahl, Sutliff, Oyuela-Caycedo and Scott Raymond1998).

Machalilla phase burial configurations at Salango, however, as well as their location and grouping, are notably different from those typically recorded for Valdivia. In the better-studied Valdivia Phase III at Real Alto, there was a close relation between human burials and residential structures: adults and subadults generally lay buried in a flexed position in shallow graves next to the wall trenches of houses, whereas infants were laid directly in the trenches and unborn babies were buried enclosed in upturned cooking jars set in the trenches (Marcos Reference Marcos1988:161–162; Zeidler et al. Reference Zeidler, Stahl, Sutliff, Oyuela-Caycedo and Scott Raymond1998). Moreover, the burials in trenches were foundation offerings made prior to the placement of the wall posts. At Machalilla phase Salango, in contrast, the flexed burial on the side, the format identified for the two Machalilla burials at La Cabuya (Meggers et al. Reference Meggers, Evans and Estrada1965:110), was just one of three main possibilities for adults: adults and infants were buried in similar formats and in adjacent sites or even together, and burials were apparently sited as discrete groups. Thus, in contrast with the general practice at Real Alto, the Machalilla burials at Salango not only seem to emphasize the complete extended family as a social construct but also ignore any discrimination by age. At the same time, there does not appear to have been any concern with associating the interments directly with actual residential structures.

When we compare Burial 13 with the burials of the high-status female at Real Alto, differences emerge in the expression of social relations at the community level. The female at Salango was among a group of individuals, undistinguished by any evident architectural setting, that included seven infants, one adult male, and five other adult females all treated in standard ways without discrimination. In addition, this was just one of at least two similar groups likely composed of kin relations. The Middle Valdivia female at Real Alto, however, although accompanied by a total of 19 individuals potentially representative of a kin group, was interred in a stone-lined grave at the entrance to a structure on top of a specially constructed mortuary facility that projected into the center of the main village plaza (Marcos Reference Marcos1988:163–165; Zeidler Reference Zeidler, Canuto and Yaeger2000; Zeidler and McEwan Reference Zeidler, McEwan, Hoopes and McEwan2021). The other individuals—8 males and 11 juveniles—were all differentially treated and distributed with respect to the principal burial and the structure itself, according to age and in ways that diverge notably from standard practice elsewhere at the site.

Thus, whereas at Salango the role and identity of the female are indicated simply by her own postmortem treatment and grave accompaniments, at Real Alto the female's elevated status was emphasized both by placement in a special crypt at the entrance to a highly visible and formalized mortuary facility and by the unusual treatment of the dead who accompanied her—pointing to an elevated rank, whether achieved and ascribed, for the family as a whole (Zeidler and McEwan Reference Zeidler, McEwan, Hoopes and McEwan2021). By comparison, inclusion of the Salango shaman within the burial area of a possible family or lineage group suggests a community structure that avoided overt expression of social ranking.

Machalilla Phase Salango and Cotocollao

In view of their differences from Valdivia practice, it is notable that the Machalilla burials at Salango are more directly comparable for siting, distribution, and position with those excavated at the highland Cotocollao site (Ubelaker Reference Ubelaker and Villaba O.1988; Villalba Reference Villalba1988:75–109, Appendix I:Tables 1 and 2). In both the early (1500–1000 BC) and especially in the late (1000–500 BC) phase, the dead were buried together at Cotocollao in dedicated funerary zones with no discrimination as to age and in such a way as to suggest that families and descent groups used their own areas repeatedly over long periods of time. Furthermore, the bodies were generally flexed on the right or left side or else were flexed and seated, and four adult male crania of the late phase presented occipital flattening. All this evidence supports the idea that Cotocollao and Machalilla were linked by sociocultural ties and significant aspects of ideology.