This paper presents an archaeological approach to the study of culture change and persistence in multi-ethnic communities through the study of daily practices and based on a crucial tenet of practice theory-that individuals will enact and construct their underlying organizational principles, worldviews, and social identities in the ordering of daily life. The study of habitual routines is undertaken in a broadly diachronic and comparative framework by examining daily practices from a multiscalar perspective. The approach is employed in a case study on the organization of daily life of interethnic households composed of Native Californian women and Native Alaskan men at the Russian colony of Fort Ross in northern California. Recognizing that different opportunities and choices existed for household members in this colonial setting, we explore how they constructed their own unique identities by examining the spatial layout of residential space, the ordering of domestic tasks, and the structure of trash disposal. We argue that trash deposits and middens in built environments, which often accumulate through routinized tasks, present great promise for examining the processes of culture change and persistence in archaeology.

A major focus of archaeological field investigations over the past four decades in eastern North America has been the excavation of rockshelters. Many of the Southern highland rockshelters investigated during this period yielded evidence of initial occupations by Dalton horizon (10,500 to 10,000 B.P.) hunter-gatherers. Data concerning the Dalton components from a sample of 45 of these shelters are reviewed and discussed in order to identify variability in site functions and to address the question, Why were Dalton peoples the first North American hunter-gatherers to systematically inhabit rockshelters? Factors such as shifts in hunting patterns and mobility strategies appear to have been central to this development in early Holocene landscape utilization.

Histories of Americanist archaeology regularly confuse frequency seriation with a technique for measuring the passage of time based on superposition—percentage stratigraphy—and fail to mention interdigitation as an important component of some percentage-stratigraphic studies. Frequency seriation involves the arrangement of collections so that each artifact type displays a unimodal frequency distribution, but the direction of time's flow must be determined from independent evidence. Percentage stratigraphy plots the fluctuating frequencies of types, but the order of collections is based on their superposition, which in turn illustrates the direction of time's flow. Interdigitation involves the integration of sets of percentage-stratigraphy data from different horizontal proveniences under the rules that (1) the order of superposed collections cannot be reversed and (2) each type must display a unimodal frequency distribution. Ceramic stratigraphy is similar to occurrence seriation, as both focus on the presence-absence of types with limited temporal distributions—index fossils—but the former uses the superposed positions of types to indicate the direction of time"s flow, whereas occurrence seriation does not.

Fossil pollen assemblages from Cliff Palace Pond, Kentucky, characterize changes in forest composition through the past 9,500 years of the Holocene. Early-Holocene spruce and northern white cedar stands were replaced by mixed mesophytic forests after 7300 B.P. Hemlock declined around 4800 B.P., and eastern red cedar became locally important. After 3000 B.P, mixed oak-chestnut and pine forests were dominant. The fossil charcoal record from Cliff Palace Pond demonstrates that Late Archaic and Woodland peoples cleared forest gaps to cultivate native plants in the Eastern Agricultural Complex and that anthropogenic fires served to increase populations of fire-tolerant oaks, chestnut, and pines in upland forests of the northern Cumberland Plateau.

Domesticated squash (Cucurbita pepo) was recovered from King Coulee, a multicomponent habitation site. Recent accelerator mass spectrometry (AMS) radiocarbon determinations on two seeds indicate that Cucurbita was used as early as 2530±60 B.P, during the Late Archaic. This marks the earliest occurrence of domesticated plant use in the upper Midwest. Another seed dated to the Late Woodland (1170±40 B.P.) is consistent with an inferred pattern of greater plant use throughout the area. The use of Cucurbita played an important role in the long transition from foraging to farming. These new data provide valuable insights into the economies of the people inhabiting the region.

Baker Village was a Fremont horticultural community occupied between about A.D. 1030 and 1250. The residents of Baker Village built at least five pithouses and seven aboveground storage structures arranged around a much larger, centrally located building. The spatial patterning of faunal remains indicates that more ungulate and leporid bones were recovered from floors, hearths, and middens in the Central Structure than in the pithouses and surface storage structures. These data may support the interpretation that private ceremonies or feasts were held inside the Central Structure. Together with evidence of community planning, reduced mobility, and relatively low population size, the data also indicate that an influence-based political hierarchy probably existed within Baker Village.

The California School of Midden Analysis represents a long-standing tradition of using weight, rather than minimum number of individuals (MNI), to analyze shell recovered from archaeological sites in California. This method originated at the University of California, Berkeley, in the early twentieth century and continues to the present, in spite of the advent of counting measures such as MNI and NISP (number of identified specimens) in faunal studies. We argue that MNI estimates are more reliable than weight as a measure of taxonomic abundance for most research issues being addressed with California shell data. Examples using both weight and MNI measures for shell from California coastal sites produced divergent results. This disparity shows that weight measures produce potentially misleading interpretations regarding the importance of marine habitats exploited and the diet of the site’s occupants.

McGuire and Saitta (1996) give voice to widespread dissatisfaction with artificial dichotomies that lead to the classification of historic and late Prehispanic puebloan societies as “egalitarian” or “hierarchical” in organization. They suggest a solution, a dialectical approach, that rejects processual archaeology in general, although not in its entirety. Another alternative approach, proposed here, relies on the concept of heterarchy (e.g., Crumley 1994), which, surprisingly, has not yet been used in southwestern archaeology. The proposed use of this concept does not involve rejection of a processual framework or represent a comprehensive critique of McGuire and Saitta's proposed dialectical approach.

Rautman's critique of our article “Although They Have Petty Captains They Obey Them Badly: The Dialectics of Prehispanic Western Pueblo Social Organization” (McGuire and Saitta 1996) provides us with an opportunity to clarify some points about our theoretical perspective. Rautman shares our dissatisfaction with attempts to characterize Prehispanic western pueblo social organization as either egalitarian or hierarchical. She, however, questions our dismissal of processual theory and our advocacy of a dialectical approach to the problem. She proposes instead an alternative approach that relies on the concept of heterarchy. We have little problem with the use of heterarchy as a descriptive label for late Prehispanic pueblo social organization, but we desire a more dynamic understanding of that organization than the concept of heterarchy allows. We find that understanding in a dialectical approach.