The Classical lands of the Mediterranean present the thoughtful observer with the paradox of the homelands of great early civilizations in landscapes now characterized by limited and discontinuous arable soils, bare rocky hillsides, and silted harbors. Already in late Classical times writers speculated about the destruction of formerly richer landscapes by abusive land-use practices. Early environmentalists used the Mediterranean case as a moral lesson, threatening similar impoverishment to heedless peoples elsewhere (e.g., Marsh 1965). This view of things is necessarily based on the assumption that the damage had been done during classical times and that later populations simply endured the burden of their poor inheritance, which doomed them to economic marginality in the modern world.

By the decade of the 1960s, informed observers had noticed that the massive alluvial deposits in circum-Mediterranean valleys contained Roman and younger sherds, and that in some instances they buried Classical and Byzantine sites (e.g., Judson 1963). These observations particularly impressed Claudio Vita-Finzi, who inspected valley fills around the Mediterranean and published in 1969 a monograph on his investigations.

Vita-Finzi observed two major episodes of Mediterranean valley fills, which he called the “Older” and “Younger” Fills. The older and more massive was very rocky in places, was typically a deep red color, and had been deeply incised by stream-cutting before the deposition of the Younger Fill that was “nested” within it.

The truly dynamic components of the biosphere are members of the Animal Kingdom, most of whom are capable of motion and intentional behavior. Behavior (e.g., feeding, competition, migration, cooperation) immensely complicates environmental modeling by reducing predictability in system states. Here, the emphasis is on describing and interpreting the faunal components of past ecosystems, including humans, as a basis for understanding paleoecology.

Paleoenvironmental reconstruction has been important in paleontology for a very long time, beginning two centuries ago in the fossiliferous Paleolithic caves of Europe. However, interest in paleoenvironmental reconstruction from archaeological faunas did not travel intact across the Atlantic (Grayson 1981). The best American work in the genre has been done by paleontologists, whether or not working for archaeologists (e.g., Graham, Guilday, Guthrie, Klippel, Parmalee). The problem in America seems to derive from the fact that few paleontologists are interested in Holocene faunas, while neoecologistswork in a timeless dimension that assumes the validity for ancient times of actualistic study in present conditions. Consequently, the development of critical theory for paleoenvironmental reconstruction from archaeological faunas in the Americas has been delayed, to the detriment of research designs and excavation strategies. Archaeozoology with an environmental emphasis is more at home in the rest of the world than in the western hemisphere, where autecological studies are in short supply and zooarchaeological emphases on human behavior dominate the archaeological literature.

Admittedly, there are stringent limitations to the applicability of archaeological data for paleoenvironmental work with fauna, as indicated below.

Decades ago, b.p. (before 1950), dating Bronze Age sites in the eastern Aegean seemed direct and reliable, given artifactual cross-ties with the astronomically calibrated Egyptian king lists that served as calendars. For instance, the association of Late Minoan IA ceramics in a Greek tomb with Egyptian scarabs provided a terminus ante quem age estimate for the ceramic style at ca. 1500 b.c., a nice round, memorable date. Ceramics of LMIA style were buried when the Thera volcano erupted on Santorini island, north of Crete, and 1500 b.c. was proposed as the date of the event (Marinatos 1939). The scale of the eruption, devastating to the town of Akrotiri, led archaeologists to try to relate it to catastrophic fires and building destruction elsewhere in the Aegean. When radiocarbon dating became available, scholars wanted to use that method to refine the dating of the eruption and test its synchroneity with destructive events nearby. The result of those efforts, and applications of additional dating methods, has been a vast, expanding, contentious literature that remains inconclusive. Why?

In the decade of the 1970s, charred organic samples from the Akrotiri excavations were sent to the radiocarbon laboratory at the University of Pennsylvania, a respected research facility. The immediate results supported the traditional age, but tree-ring calibration produced dates implying an age greater than 1600 b.c. (Fishman et al. 1977). Efforts to explain the results focused at first on contamination by ancient carbon in volcanic gases venting nearby (Weinstein and Michael 1978).

Human beings perceive environmental change mainly as change in the state (qualitative character or structure) or condition (quantitative composition or amount) of nearby communities of living organisms, or of the weather. For example, a change of state for living communities might be gains or losses in the diversity of plants or animals represented; a change in condition might be an increase or decrease in the numbers of plants and animals. For the weather, a switch fromwinter rains to predominantly summer rains in mid-latitudes would constitute a change in state, whereas a marked decrease in precipitation over a month or more would be a change in condition. We notice such changes, because they violate our expectations that things vary little from year to year. Our experience and observations of the environment are at the local scale and are mediated by language and opportunity, so that each of us has a slightly different idea of things. Because of thewaysweperceive environmental change, our “commonsense” leads us to seek the causes of change wherewe perceive it – among living communities and in the weather systems. Recent research in geophysics and climatology has demonstrated that this approach is oversimplified and misleading.

Every surface on which humans lay foot or artifact is a potential archaeological site, requiring only that subsequent processes not dislodge and transport the surficial deposits. Of course, disturbance of surficial sediments of every kind is the normal case. This vulnerability ensures that archaeological sites are neither ubiquitous nor permanent.

The focus of this chapter is on sediments and soils as matrices of archaeological sites, at local and micro-scales. We occasionally lift our eyes to regional-scale phenomena, as in considering the information potentials of widespread deposits of loess or volcanic ash, but we pay no attention here to the mega- and macro-scales of phenomena or to regional-scale interpretations.

MESSAGES IN THE MATRIX

Sedimentological analyses are undertaken to learn about the sources, transportation agents, depositional and transformational history of the materials comprising deposits (Chapter 11). Although archaeologists typically treat that information as background, environmental archaeology must begin with the environments in which materials, whether cultural or natural sedimentary particles, were brought to a site, deposited, and affected by postdepositional processes including pedogenesis and diagenesis. The enclosing matrix is the fundamental source of information about all the processes essential to understanding the context of human behavior at a site. Not all evidence is visible, and not all is extractable by techniques currently known. However, for sites lacking written evidence the matrix is the only source of non-artifactual information; for sites with written histories, the matrix will variously confirm, expand, or contradict elements of that record.

The conflict of land and water creates the dynamism characteristic of shorelines, whether of rivers, oceans, or lakes. Humans are drawn to water because it is essential to the maintenance of organic life, and is therefore the location of basic resources. Archaeological sites on the shores of lakes and oceans present special opportunities and challenges for paleoenvironmental studies. Sites near water typically exhibit preservation conditions conducive to the survival of a range of organic materials. The sediments in and near them are likely to be organically enriched as well and hence excellent sources of climatic proxies and remains of plants and animals. Landforms shaped by waves and currents are typically informative about past climatic and geotectonic states and conditions. The dynamism of sedimentary regimes typically creates stratified sites, which are nevertheless subject to frequent erosion.

COASTAL GEOMORPHIC CONCEPTS AND PROCESSES

Landforms at the edge of water, like those on land, are shaped primarily by processes in the atmosphere, geosphere, hydrosphere, and cryosphere. The biosphere's influence is expressed mainly at small, local scales. The large-scale processes most responsible for changing the elevational relationships between land and water, and thus initiating erosion and landform evolution, are tectonism and climate change, and their combined product – eustasy.

The study of the human past requires knowledge of the solar system as well as of the home planet and its geophysical and biological systems, of which we are inextricably a part. The eternal fascination of archaeological research is that it challenges all our creativity, discipline, and enthusiasms; scarcely any knowledge is irrelevant to it. That is especially true of environmental archaeology – the study of paleoenvironments as human habitats. Habitats pose problems and opportunities for resident organisms of whatever size and complexity; humans are not excepted from this imposition. If we are to understand the behaviors of human beings in their unique cultural contexts, we must be able to define and examine crucial aspects of their habitats. Humanenvironments, originally restricted to sub-Saharan Africa, now include the entire world and parts of space – so, the study of human ecology, which is at the core of environmental archaeology (Butzer 1982), is necessarily comprehensive and resolutely dynamic. Not surprisingly, it is still very immature and experimental.

The means for defining and interpreting elements of human environments, both past and present, are expanding. Archaeologists, especially environmental archaeologists, employ techniques and concepts developed in the disciplines of anthropology, biology, ecology, zoology, botany, geology, oceanography, climatology, and pedology (soils), among others. Of course, no one can be expert in all these subjects; both compromise and consultation are required.

In the first sentence of Chapter 1, we quoted Hardesty (1977: 290) defining ecology as “that branch of science concerned with [the study of] the relationships between organisms and their environment.” In Part VI, we introduced “neoecology” as a special aspect of ecology devoted to the study of living species in their environments or in laboratory situations, and “paleoecology” as the application of principles from neoecology along with disciplined inference to the study of organisms in environments no longer directly observable. On the basis of these definitions, paleoecology is the best that archaeologists can do in researching societies of the past. Furthermore, because archaeologists are particularly devoted to the ecology of human beings, our study is necessarily anthropocentric. This volume is an extended argument for the centrality to archaeology of an anthropocentric paleoecology; the concept imposes on research and interpretation broader contexts and goals than are typically included in the currently controversial term “environmental archaeology”.

The argument that humanbeings can exist and act independently of their environments is a fallacy deeply rooted in Western culture (Glacken 1967). The false dichotomization of nature and culture is supported by religious and technological ideologies, and embraced by development economics. The environmental crises of today are direct results of such dismissal of interactive mutualities between organisms and their physical and social environments.

Landform reconstruction in archaeology is most challenging with truly ancient sites, as in the study of human origins. The site of Laetoli in the southern part of the Serengeti Plain in East Africa is rich in hominid and other fossils and, uniquely, in footprints on a buried land surface about 3.6 million years old. Among the prints are those of upright, bipedal primates walking with a “shambling” gait (Leakey and Hay 1979) on thin layers of volcanic ash. What we can know of the context of those prints and of the landscape in which those strolls were taken – the habitat of a remote ancestor – we must learn from landform analysis, sedimentology, and paleontology. Diligently applying skill and imagination to these complementary sets of data, Richard L. Hay (1981) achieved a remarkable reconstruction of a Pliocene landscape at the beginning of human time.

The current landscape at Laetoli is a product of plate tectonics; continental plates are pulling apart and new crust is forming by volcanic action in the East African Rift Valley. Large-scale faulting has shaped a landscape of abrupt changes in relief and elevation, where dry uplands loom over lakes in the basins. South of Olduvai Gorge in northern Tanzania, the Eyasi Plateau lies on the uplifted northwest side of a major fault. Near the fault on the south edge of the plateau, the Laetoli area rises to an elevation of 1800 meters; at the foot of the fault lies Lake Eyasi.

Vegetation, with bacteria, is the foundation of the biosphere, the base of the food chain, the mediator of atmospheric composition, the organizer of the water cycle, and the pulverizer of the geosphere. Human biological and cultural adaptations today and in the past grow out of relationships with plant communities at all scales, and must be understood in those contexts. This chapter presents some basic concepts of ecology and paleoecology (the application of principles from the ecology of living systems to the study of organisms in environments no longer directly observable), and indicates some methods for achieving knowledge of aspects of past vegetation states and conditions and the mutual relationships between those and human societies.

ASSEMBLING THE DATABASE

Paleoenvironmental reconstruction begins with defining, assembling, and describing the data available, and moves on to interpretation. Again, description and interpretation must be separate and sequential, although ideally there are reflexive loops in each process.

The diversity of data sources for paleovegetation is advantageous since the entire set is rarely if ever available at once. As discussed in Chapter 13, plant and animal remains, soils classes and distributions, paleotopography and paleohydrology, paleoclimate data, and ecological theory all potentially contribute to reconstructions. Data can be assembled from archaeological sites (on-site data) or from the locale and region (off-site data). The more diverse the data and the sources, the more reliable the results. The best information comes from research projects clearly defined and structured so that sampling has been broad, careful, and suitable to the goals.

In an ideal homeostatic world, there would be little change. However, in a world such as Earth, dominated by living things, there can be no stasis, no equilibrium. For any organism, the successful continuation of life requires the ability to adjust to changed conditions. The paleoenvironments that were the contexts of past human actions must be known if we are to understand human history and evolution. Social environments, the crucial contexts of human planning and decision-making, are the subjects of anthropological and archaeological social theory, with their own vast literatures. In this chapter attention focuses on human strategic responses to changes in physical and biological environments, distinguished as much as possible from social environments.

Environmental change, loosely defined, is a departure from the “mean” or perceived normal state or condition of any aspect of the environment. Humans respond only to those changes that they perceive, and then only to those that affect conditions or resources that are important to them. To illustrate with a simplified example: a competitive replacement of one species of mouse by another on a mountainside should evoke no response whatever from a community of farmers in valleys nearby. The feeding habits of the new mice, however, could initiate changes in the ratios of grasses available to grazing animals, which might be crucial to pastoralists using the highlands. The pastoralists would observe the vegetative change, and might try to mitigate it by firing the grasses; they might not recognize the subtle role of the mice in the changed conditions.

Sediments are composed variously of particles of disaggregated rock, dust from whatever source, bits of dead animals and plants, and chemical precipitates. Their deposition on the surface of the Earth or the bottom of lakes and seas creates threedimensional sedimentary bodies (deposits) which are subsequently modified in characteristic ways by the five spheres of the climate system. In company with bedrock, sediments underlie the landforms on which life processes occur. For archaeologists, sediments are the enclosing medium and the environment for the physical and chemical remains that comprise archaeological sites.

INTRODUCTORY CONCEPTS

In contrast to the readiness with which archaeologists borrow geomorphological techniques for identification and description of landforms, developments in petrographic techniques seem to be adopted slowly and reluctantly by them. Methods for the technical description and interpretation of sediments and soils, particularly, need further development and more intensive application in archaeology. As with fish that cannot be expected to be aware of water, archaeologists often take for granted the materials within which their sites occur, rather than seeing them as problems and interpretive opportunities. Skilled geoarchaeological work remains, regrettably, a specialist domain instead of being incorporated as a matter of course in all field work.

Minerals are inorganic chemical compounds in crystalline form; rocks are composed primarily of minerals, sometimes accompanied by organic detritus and chemical precipitates. Mineral matter of the regolith recirculates through cycles of exposure, erosion, deposition, and burial at the surface of the Earth.

How do we recognize climate change in the cacophony of proxy signals in the paleoenvironmental record? A prime criterion has long been recognition of the same or similar change over regional or larger spatial scales. The idea is that only climatic forcing can elicit parallel and nearly synchronous response over significant distances. Is this criterion sound, and is it adequate? The history of the concept of the European “elm decline” is the story of efforts to explain a phenomenon that has exercised the ingenuity of scientists for over fifty years, yet its environmental significance remains unclear at best.

Early in the development of the northwestern European pollen studies, investigators noticed a sharp mid-Holocene decline in the abundance of elm pollen, a loss of 50% or so in a century. Peat-bog stratigraphy indicated that the decline occurred very close to the transition between the Atlantic and Sub-Boreal phases of the Blytt–Sernander scheme (Fig. 8.2). That transition had been earlier interpreted as the result of climatic change from a wetter “Atlantic” to a drier “Sub-Boreal” phase, both falling within the postglacial peak of warmth. Efforts to interpret the elm decline in causal terms emphasized (1) the wide geographic extent of the decline throughout northwestern Europe, (2) its apparent synchrony, and (3) its coincidence with the Atlantic/Sub-Boreal transition.

The Star Carr site represents a turning point in prehistoric archaeology, especially in the paleoenvironmental archaeology of wet sites. Grahame Clark's monograph of 1954 established new standards for excavation, recovery, and interpretation. It has since inspired several reinterpretations. The site was reopened for additional field work in the 1980s and 1990s (Cloutman and Smith 1988; Mellars and Dark 1998). Such ongoing interest is a tribute to the quality of the original work and to the continuing importance of its role in archaeology.

Star Carr, a wetland site near the north edge of the Vale of Pickering in eastern Yorkshire, England, comprises a deposit of early Mesolithic artifacts of stone, bone, antler, bark, and wood, preserved with other organic remains in peat. It is culturally affiliated with the Danish Maglemosian, Mesolithic woodland hunter-gatherers who lived partly on terrain now inundated by the North Sea, during the early Holocene Preboreal period about 9600 radiocarbon years ago (see Day and Mellars 1994). Artifacts and organic debris were deposited on the shore and shallow margin of a lake occupying the Vale of Pickering.

The outstanding feature of the site was a “platform” of birch trunks and branches exposed by careful excavation of the overlying peat and mud. The platform lay on and in a reed swamp bordering the lake, between dry ground and open water where the latter was closest to land.

Not long ago the biosphere encompassed by definition two taxonomic kingdoms: Animals and Plants. Fungi and algae were included in the plant kingdom, and bacteria were somewhere between. The currently dominant taxonomy (systematic classifi- cation), based in large measure on organic form, recognizes five kingdoms, with bacteria and some algae included in Monera, other algae and simple “eukaryotic” organisms (having discrete nuclei in their cells) in Protista, and Fungi in a kingdom of their own (Margulis and Schwartz 1982). The kingdom of Plantae includes most of what we recognize as plants (vegetation) – trees, shrubs, flowers, mosses, ferns, and so forth. Recent research in microbiology has forced reconsideration of the organization and history of life, and initiated a period of classificatory revisions. An emerging taxonomy based on molecular criteria proposes three “domains” at the foundation of life: Archaea (microbes unlike bacteria), Bacteria (with blue-green algae), and Eukarya (all organismswith distinct cell nuclei). Members of the newly recognized group, Archaea, are separate from but differentially related to the two other groups, and with bacteria comprise the prokaryotes (organisms lacking cell nuclei). The domain of Eukarya includes plants, animals, fungi, and protists (single-celled organisms with cellular nuclei). As the new criteria are tested against extant classifications, the results are likely to rearrange classes even further and modify phylogenetic trees (Pace 1997). Archaeologists can wait for these developments but should be aware of the new uncertainties and alert to the need for clear analytical language.

For as long as the Earth has had an atmosphere, its surface has been continually shaped by air, water, and ice disaggregating, transporting, and depositing mineral matter. Human lives are lived on surfaces, but archaeological surfaces are not necessarily those on which the archaeologist walks. Depending on their age and situation, ancient surfaces have been buried or lost to erosion.

Conceptual reconstruction of past landforms and surfaces is an essential aspect of modern archaeology because the spatial context of a site is crucial to its interpretation, and to understanding its relationships to other sites. Space and landscapes define the resources available to any human group, and landform changes through time are related variously to changes in other elements of the environment – climate, hydrography, and biota (mainly vegetative). The mechanisms by which landforms, as elements of the geosphere, are shaped by processes originating in the hydrosphere, cryosphere, atmosphere, and biosphere are imperfectly known although believed to be determinable. Because of these interdependencies, landform reconstruction informs about the past states of variables in all five spheres. However, as with all complex dynamical systems, our ability to predict future states or understand past states and conditions is limited by the element of chance influencing combinations of mechanisms in several scales of space and time.

The “reconstruction” of ancient landforms is not done with earth-moving equipment; it is, rather, a conceptual exercise based on the study of remnants available for observation in the present.