Antiquity of Early Holocene Small-Seed Consumption and Processing at Danger Cave

Article excerpt


The adoption of small seeds as a food staple is a crucial turning point in human dietary history. Instances of the so-called 'broad-spectrum revolution' of subsistence diversification (Flannery 1969, 1973) occur worldwide during the late Pleistocene/Holocene transition (see, for example, recent summaries in Cowan & Watson 1992; Harris & Hillman 1989; Price & Gebauer 1995). Dramatic and rapid climate-induced fluctuations in the abundance and distribution of food resources during this period forced foraging peoples to adopt new foodstuffs (including small mammals, aquatic resources and small-seeded plant foods) that were abundant but entailed high procurement and processing costs (Richerson et al. 2001; Stiner 2001 ; Watson 1995).

The archaeological record of the North American Great Basin shows a similar trend towards dietary reliance on small seeds, but the timing of small-seed adoption is not certain. Danger Cave, located on the western margin of the Great Salt Lake Desert, Utah (Figure 1), has long played a central role in the interpretation that intensive small-seed use began during the earliest Holocene in parts of the Great Basin (Aikens & Madsen 1986; Beck & Jones 1997; Fowler 1986; Grayson 1993; Jennings 1957, 1978; O'Connell et al. 1982; Willig & Aikens 1988). This is because Danger Cave contains seed-processing residue, human palaeofaecal samples and abundant groundstone artefacts (grinding stones and handstones) associated with small-seed processing all found in deposits dated as old as 10 300 b.p. These three lines of evidence have been widely thought to demonstrate a reliance on small seeds since earliest Holocene times (Fry 1976; Grayson 1988, 1993; Harper & Alder 1972; Jennings 1957).


In this paper, we re-examine each of these three fundamental sources of evidence of earliest Holocene small-seed consumption and processing at Danger Cave. We demonstrate that small-seed consumption and processing began at Danger Cave c. 8600 b.p. Although not as early as previously supposed, this is the best documented and one of the earliest dates for the onset of significant small-seed use in western North America.

Early Holocene occupations at Danger Cave

Jennings (1957) identified two major early Holocene stratigraphic units in Danger Cave, which he labelled DI and DII (Figure 2). The earliest occupation, in the DI level, takes the form of several small fire hearths accompanied by a sparse scatter of artefacts and ecofacts on a bed of beach sand ('Sand 1'), capped by a variably thick layer of wind-deposited sand containing abundant artiodactyl scat ('Sand 2'). Jennings's original radiocarbon dating of the hearths, together with recent excavations we have conducted, confirm that this earliest human occupation of Danger Cave took place at 10 300 b.p. (Table 1). This earliest occupation was likely coeval with the existence of a large shallow lake that coveted the Great Salt Lake Desert at the level of the Gilbert Shoreline, approximately 20m below the cave's portal (Oviatt et al. 1992, 2003).


The DI level contains a small number of palaeofaecal specimens and groundstone artefacts that have been commonly thought to indicate earliest Holocene use of small seeds. Five palaeofaecal samples reported from the DI level (Fry 1976) contain pickleweed seeds (Allenrolfea occidentalis). This halophytic playa-margin shrub produces tiny seeds that rank low in caloric efficiency (Simms 1987), but the seeds were evidently processed and eaten in substantial quantities during the middle and late Holocene occupations at Danger Cave (Fry 1976; Madsen & Rhode 1990; Rhode & Madsen 1998). Jennings (1957) reported three grinding stone fragments and three possible handstones from the DI level, but these he considered to be too inconclusive to demonstrate small-seed processing; one handstone and milling slab were covered with red pigment, suggesting grinding of ochre, not seeds. …