Ornaments at Üçağızlı I Cave, Turkey

Ornaments are abundant in the Initial Upper Paleolithic and Ahmarian layers of Üçağızlı I cave, and they are also present in the small Epipaleolithic sample (Stiner 1999; Stiner et al. 2002). With the exception of one notched raptor talon (probably vulture, Gyps or Gypaetus) from layer B, all of the ornaments were made from small marine and freshwater mollusk shells. Most of these shells were collected from active shorelines and rivers of the area, but tusk (Dentalium) shells used mainly during the Epipaleolithic were obtained from fossil sources roughly 15 km from the cave. The presence and abundance of shell ornaments is unrelated to that of mollusks exploited as food (turbans & limpets) at this coastal cave site.


The “ornamental” shell artifacts are easily distinguished from species consumed as food and from land snails on the basis of an aggregate of damage characteristics. The ornamental species were often perforated with a pointed tool, and the placement of holes is very consistent, normally through the flange or aperture of gastropods, and through the umbo of small bivalves. Some of these shells were stained with red ochre, an iron mineral pigment. The human-made holes have irregular edges and were made with a small hard punching tool or direct pressure from a pointed object. A rarer perforation method was used only on some moon shells in the Ahmarian layers and involved sawing a slit in the lip margin. The irregular edges of the human-made holes differ conspicuously from the symmetrical beveled hole that naticid or muricid predatosr typically drill into the shells of live prey.


Shells with Red Ochre Shell with Beveled Holes


The wide recurrence of wave-induced abrasion and predator-drilled holes on Üçağızlı I ornaments tell us that humans relied mainly upon beach-cast raw material rather than capturing live specimens for the purpose of ornament-making. These shells were collected from nearby rocky shores, soft marine shores, inland freshwater lakes or rivers, and at least one paleontological source. The marine species, Nassarius gibbosula, dominates the IUP assemblages and is co-dominant with Columbella rustica in the Ahmarian series. Freshwater types (mainly Theodoxus and Melanopsis) are most common in layers C-E, whereas a more even mix of tusk shells, Gibbula, Columbella, and other types occurs in the Epipaleolithic assemblage.


Taxonomic diversity in the ornament assemblages—number of species utilized and the relative proportions of each species—increased steadily with time, rising more than threefold from the IUP to the early Epipaleolithic. This trend is not explained by changes in the distance between site and shoreline. Shellfish were used in moderation during the earliest IUP (layer I), the ornaments from which are overwhelmingly Nassarius gibbosula.


Whereas the terrestrial faunas of the region experienced few changes in composition during the Late Pleistocene, the hydrodynamics of the Orontes river mouth and the complexity of the marine shoreline could have varied greatly and unpredictably relative to general environmental indicators, including sea level. Local forces governing aquatic habitat complexity remain to be understood in the study area and could account for the trend in ornament species diversity. On the other hand, a dominance of Nassarius gibbosula or similar forms of Nassarius is reported for some of the earliest ornament assemblages in North and South Africa and is suggested to imply a narrow human preference for this type of shell in the earliest periods of ornament use.


Human selectivity is apparent in the ornament assemblages from the prevalence of ecologically uncommon taxa (e.g., Nassarius is a carnivorous scavenger) and with respect to shell shape, size, and color (Stiner 2003). Rounded, basket-shaped or pearl-shaped forms are especially common at Üçağızlı I and in UP sites across the Mediterranean rim more generally. In fact there is remarkable consistency in ornament shell sizes at this and other sites, irrespective of phase.


Unworked shells, along with specimens broken during manufacture, indicate that many of the ornaments were produced on site. Only a few were worn-out and abandoned after extended use. Other shells were broken during attempts to perforate them. Less than 5% display fine polish on the edges of the hole from prolonged contact with fiber.


The ornaments accumulated with other camp litter and were shed gradually by individuals in a variety of contexts. We have yet to find articulated sets of ornamental shells, as if still strung, but at least one bundled concentration was found in layer G, and certain other loose clusters in the IUP deposits may also represent caches of some kind.


Back to Top

Notched Raptor Talon





























Species Diversity
















Shell with Polish


Shell Concentration

and Rib Fragment

ges/thumbnails/subsistence_change.jpg" alt="" width="175" height="123" />