Temporal scales and archaeological landscapes from the eastern desert of Australia and intermontane North America

Time gets much less attention than space in discussions of archaeological scale. This may seem strange in a primarily historical discipline for which the demonstration of human antiquity is something of a defining moment (Grayson, 1983). Part of the reason may lie in the nature of time. Time unfolds along a continuum, and the way observers perceive time depends on their location and the scales they adopt. Compare the contemporary Western experience of earth time, for example, with time at the scale of the universe. A person traveling at the speed of light would experience a different time (Hawking, 1998; Ramenofsky, 1998) than the person caught up in the linear progression of our planet-bound life. Of course, archaeologists rarely deal with quantum time, but the example serves to remind us that time is not an absolute dimension. Archaeologists create their own conceptual units for measuring time. They project these units at different scales and choose their own observation points, dividing the continuum of time into arbitrary packages that relate in some way to specific research goals (Ramenofsky, 1998). Few archaeologists have grappled explicitly with scale issues. Crumley (1979) and Marquardt (1992; see also Crumley and Marquardt, 1987) emphasize that social and economic processes may each resolve best at different spatial scales. Stein (1993) attempts to reconcile the vastly different temporal scales of geology and archaeology. Most recently, Dobres (2000; see also Lock and Molyneaux, this volume) differentiates between the phenomenological scales at which events contributing to archaeological deposits unfold (i.e., activities, behaviors, and practices) and the interpretative scales of archaeological reasoning (i.e., generalized, theoretically informed). The former are comprehensible in what Binford (1981) refers to as ''ethnographic time'' and what Stein (1993) calls ''human time''; the latter are timeless or time-free. Dobres contrasts phenomenological and interpretative scales with the analytic scales that researchers use. The choice depends on their research interests (see also Crumley, 1979; Marquardt, 1992) and on the nature of archaeological deposits. As her focus is primarily on agency at individual and collective levels, she emphasizes the phenomenological scale, but she insists that phenomenological, interpretative, and analytical scales have no necessary relationship. Thus, when pursuing such interpretative goals as agency, archaeologists are not limited to one particular scale of phenomena. Nor, according to a close reading of Dobres, are they limited by the nature of the archaeological record, as the scale at which they view the material record is not related to any particular phenomenological scale of agency. In this chapter, we explicitly focus on archaeological temporal scale, by which we mean both the temporal structure of the phenomenon we study, i.e., the archaeological deposit, and the scales of measurement and interpretation we bring to that phenomenon. Temporal structure refers to (1) the grain, resolution, or microstratigraphic acuity and (2) the extent or scope of phenomena represented in archaeological deposits, observations and interpretations. Grain (Binford, 1980; O'Neill and King, 1998:7), resolution (Behrensmeyer, et al., 2000; Ramenofsky and Steffen, 1998:4 -5; Stein, 1993:2) and microstratigraphic acuity (Schindel, 1982) refer to the smallest resolvable temporal interval in an observation set. Extent (O'Neill and King, 1998:7) and scope (Schindel, 1982) refer to the total expanse of time represented in an observation set (see also inclusiveness - Ramenofsky and Steffen, 1998:4 -5). To these, Schindel (1982) adds (3) temporal sequence completeness, as many deposits are records of depositional gaps as well as accumulations. Ecologists O'Neill and King (1998:7) offer important observations on how scale of observation and measurement affect the effective grain and extent of deposits. They note, for instance, that the sampling frequency in time influences the grain of observation, a relationship described elsewhere as the Nyquist principle. Similarly, the time span of a particular measurement necessarily influences grain. The practice of calculating means for some span of time necessarily coarsens the grain while subsampling a sequence reduces its extent. In archaeology, both behavior and geological processes contribute to grain (resolution), extent (scope), and completeness of sequences. Measurement practices further affect these aspects of temporal scale. While Dobres argues that there is no necessary relationship between phenomenological, interpretative, and analytical scales, we follow geoarchaeologists, geomorphologists, paleontologists, and ecologists in emphasizing that the nature of archaeological deposits very much determines the analytical scale - and therefore the range of interpretative scales (see Murray, 2003; Stein, 1993:5; Stern, 1993, 1994; Stern, et al., 2002). Thus while the archaeological record may potentially be viewed at a variety of different scales from a range of different view points, issues of compatibility between data, analysis and interpretation cannot be ignored. We begin with this point, using it as the basis for a critique of the recent and current hunter-gatherer literature and drawing on our current work from western New South Wales, Australia, and southwest Wyoming, USA. We argue that neither of the current interpretative approaches to the hunter-gatherer archaeological record, ethnoarchaeological models or insights derived from behavioral ecology, deal adequately with the temporality of the record. Integrating the temporality of data and interpretation suggests to us a third way, whereby we can use explanations developed by viewing the archaeological record at a variety of scales to create a rich historical tapestry of past human behavioral variability.