Introduction: Goal of this paper 2
 

1. Conceptions of Space and Time

1.1 Conceptions of Space 2

1.2 Conceptions of Time 4

1.3 Space and Time as a framework for Anthropology 5

1.4 Back to GIS 6
 

2.1 The field of Geographic Information Research (Geomatics) 7

2.2 Integrated Geographic Information Systems 8

2.3 Spatial Analysis 10
 

3.1 History of Spatial Analysis in Anthropology 11

3.2 History of Temporal Analysis in Anthropology 13

3.3 Current Applications of GIS within Anthropology 13
 

Literature cited 21

Anthropology is commonly defined as the study of humankind, in all times and places (Haviland, 1997). This definition of anthropology centrally positions its object of study within the nexus of two highly philosophical concepts: space and time. The importance of space and time to anthropology comes forth out of the simple fact that all entities and objects in the world exits in time as well as in space.

Aristotle's Physics written in 400 BC contains what might be the first known systematic theory of space and time. Two thousand years ahead of Newtonian physics it already tried to reach beyond the mechanistic model of the world implicit in that view. Aristotle's Physics was centrally concerned with spatio-temporal reasoning, which concerns the appearance, change, and disappearance of things in space over time (Couclelis, 1998). This form of reasoning is fundamental to the explanation of complex phenomena in both the natural and human world, and as such important to the development of anthropology as the science of humankind in all spaces and all times. For this reason it is perhaps surprising to note that anthropologists have given only relatively little priority to the investigation of these fundamental conceptualizations (Adam, 1998).

This purpose of this paper is to further investigate the consequences of this lack of critical perspectives on time and space within anthropology. Central in this discussion will be a recently developed, highly powerful new set of technologies: Geographic Information Systems, or "GISs. GISs have come to play an important role in anthropology, especially in its subfields ecological anthropology and archeology, as well as in many other disciplines. In the past two decades, GISs have revolutionized the way in which we conceive of space, and time, with applications ranging from being an academic metaphor for human memory (Hirtle, 1998), a tool applicable to "any postmodernist problem in which space is of relevance" (Aldenderfer, 1996), to a measure of family planning accessibility using satellite data (Entwisle et al., 1997). As such, this paper aims to explore some of the consequences of this impact on anthropology and academic life in general by paying attention to the philosophical assumptions underlying these technologies. The paper consists out of three sections. In the first section, I will try to place the discussion within a general theoretical framework. In the second section an overview will be given of the concepts of GISs, the "spatial perspective," and the use of GISs in anthropology. Finally, the implications of these developments for anthropology will be discussed in more detail.
 

1.1 Conceptions of Space

Space may not be what it seems. The way we normally conceive of it today is only one of many perspectives which exists and have existed in the human world. The conception we, westerners, hold of space today is generally called absolute space, and based on the universal language of Newtonian physics. This conception holds a physical view of space as a three dimensional extension of the world, the intervals, separations and distances among people, between people and things, and among things (Couclelis, 1998). Newtonian space has its roots in seventeenth century philosophers such as Descartes and Locke, and follows the Enlightenment ideal of a universal language which gets rid of the particular (Curry, 1998). In this rule based representation of space, the relationships between place and object are of absolute necessity and not contingent or accidental. Thus, the assumption is that "it is almost as though one could hold the map up to the world--from space--and everything would perfectly match. This is fundamentally a structuralist notion of space, following Levi-Strauss and Chomsky.

Another important conception of space was developed by the philosopher Leibniz, based on the relativistic language of indirect space (Werlen, 1993; Curry, 1998). According to Leibniz, all elements are intrinsically connected to each other. Therefore it makes no sense to talk about space with nothing in it; space came into being with the objects that are filling it. Thus, instead of the rigid relationship between space and objects proposed by Newtonians, Leibniz--backed by Kant--proposes a relationship between persons and space as the fundamental principle. This view argues that rules cannot be totally spelled out, and culture does not follow a mathematical practice. Leibniz argues that we can never really "see" space directly. The best we can do is inferring space from perceivable objects.

Underlying the development of these conceptions is the older known conception of space as articulated in Aristotle's Physics (Couclelis, 1998). Prevalent in medieval times, this space is couched in a language of significance and must be seen as symbolic space. Although similar to Newton in its reliance on the metaphor of a "container-like" space, this conceptualization differs from the strictly mechanistic Newtonian model by its power to draw things in its proper place. In Medieval pictures there is no grid like space, consequently striking the viewer often as peculiarly flat. People and things vary in size, and paintings have within it various scenes which may involve the same people at different times. The significance of social and religious scenes depicted is based on the placement of more important figures in the foreground or in larger proportions. The meaning of these kinds of spatial representations is to provide interpretations of reality through the narrative of the landscape. As such, medieval space is a mirror of an ordered world in which everything has its place.

Werlen (1993) argues that the debate on the above mentioned conceptualizations of space has led to the conclusion that space in its Newtonian sense can not be maintained. Based on a general theory of action, Werlen argues that implied in Newtonian space is the erroneous idea that space is an object by itself, and as such an appropriate research object as well. However, if this were the case, he maintains, we should be able to locate the place of space in the physical world, and this is impossible. Based on this argument, a clear fundament for the discovery of positivist "spatial laws" as striven for by mainstream geography is lacking in validity. Thus, the definition of "spatial science" should be broadened to include other perspectives, such as ecology, ethology, biology, planning, architecture, environmental sciences, sociology, psychology, and the various disciplines of anthropology, including archeology. Indeed, Werlen's emphasis on action theory seems congruent with current trends in anthropological theory which emphasizes practice (Bourdieu, 1977; Ortner, 1984), situatedness (Haraway, 1991), connectivity (Latour, 1988), and methodological individualism (Smith & Winterhalder, 1992).

For anthropologists, understanding space clearly involves more than the study of an abstract, geometrical map based on a Newtonian space. Accepting the interpretation of Werlen's action theory involves an acceptance that space cannot cause or determine anything by itself, but only in relationship to action. Giddens (1993) suggests that seen from this perspective location is only socially relevant when filtered through the frames of reference that orient individual's conduct. Thus, these frames of reference themselves become what is constituted as "space." As such, many classifications of space exists and have been classified (Rapoport, 1998). Space can be human or non-human. It can be designed or non-designed. Space can be symbolic (Geertz, 1976), although such space may be indistinguishable from other forms of space, such as economic space (Malinowski, 1922), or appear not to be organized at all. Behavioral space can be defined as the setting or system of settings in which behavior of groups or individuals take place and influences psychological or cognized space, which determines what is known (informational space) and perceived (perceived space). Perceived space could be further broken down in experiential and sensory space. Cognized space consists out of mental schema or maps which are inherently dynamic and distorted relative to geometric space (Laszlo et al., 1993). Cognized space can also be seen as cultural space when related to culturally characterized predispositions of groups or individuals. All these behavioral spaces can be very different from the kinds of spaces shown on a geometric map. Clearly then, anthropologists’ space resembles a heterogeneous, culturally defined pattern, similar to the concepts put forth in human geography, and differing from the strict "objective" geographical grids of the maps used in our normal lives today.
 

1.2 Conceptions of Time

Although not the central topic of this paper, an excursion into the concept of space automatically links to the concept of time. Reason for this is that in human perception these concepts are bounded together in the form of spatio-temporal reasoning, which, according to Couclelis (1998) deals with the three related dimensions of being and becoming, objective and subjective, and space and time. Time, as some philosophers have asserted, does not exist: the "temporal" is no more than a way in which events are ordered (Giddens, 1993). This somewhat extreme and perhaps postmodern notion of time could however be used to suggest the possibility of alternatives better capable of dealing with cultural and even physical reality. Time as formulated by Newtonian physics is related to an absolute, true and mathematical time, which of itself, and from its own nature, flows equably without regard to anything external. Within this absolute notion of time two alternative social conceptions have been identified by anthropologists and historians (Couclelis, 1998). In the first case, task or event-orientated time, the sense of time is shaped by the onset, duration, and completion of daily, weekly or seasonal tasks. As Evans-Pritchard (1940) puts it:

1.3 Space and Time as a framework for Anthropology

The relationship between space and time is helpful in the interpretation and positioning of certain schools of thought. For example, Harvey (1990) suggests that on the one hand social theory can be seen to treat time over space: space is pre-existently ordered within which temporal processes operate. Progress is the annihilation of space by time, emphasizing the process of becoming rather than being (Harvey, 1990). Jagtenberg and McKie (1997) argue that postmodernism is precisely a reaction on this dominance of time over space. On the other hand, aesthetic theory treats space over time through its search for the rules of eternal truths and timeless beauty in the midst of flux. Here the emphasis is on being instead of becoming. Couclelis (1998) suggests that although treated separately, Aristotle's time and space are intimately connected through motion. Aristotle defines physical locomotion as "a motion with respect to place". Furthermore, because Aristotle's organic time does not flow "without regard to anything external," as is in the case with Newtonian time, it is fundamentally connected with locomotion (or change), though distinct from it.

This suggested connection between space and time helps anthropologists interested in understanding the dynamics of change--either physical, cognitive, cultural or ecological--to focus on a combined diachronic and synchronic perspective. Thus, "spatial ecologies" of identity, behavior, settings, settlements, trading routes, territories, processes, ecosystems, etc. are put in a historical and evolutionary framework. Such multi-temporal and multi-spatial relativity has become increasingly accepted in cultural anthropology (Haraway, 1989; Harvey, 1990; Anderson, 1991; Appadurai, 1996). However, these forms of analysis prove to be quite complex, especially when set against the pervasive positivist notion of Newtonian ,objective meanings for both space and time. At the same time the powerful language embedded in such forms of analysis is illustrated by for example Harvey's (1990) analysis of the postmodern condition. Couched entirely in terms of an analysis of changing space-time relationships, Harvey argues for the recognition of a multiplicity of objective qualities which space and time can express, and the role of human practices in their construction. Following a Leibnizian conception of space--in which space can not exist independent of the objects it contains--Harvey proposes that it is only through the investigation of material processes, which serve the reproduction of social life, that one can understand space and time. Given the socially, historically and geographically situatedness of these material practices, different bundles of social time and social space exists at different places based on different materialist production.

In ecological and evolutionary anthropology, one would expect to find an explicit attention to conceptualizations of space and time foremost in archeology. The recent incorporation of practice theory in this field indeed opened some doors to meaning (Hodder, 1985), but explicit attention to changing space-time conceptualizations have remained difficult (Hastorf & Johannessen, 1991; Jochim, 1991; Kowaleski, 1995) by large due to the challenge of reconstructing ancient space-time conceptualizations on a valid basis. However, the increasing attention given to historical processes in ecological studies motivates the general move to an integrated spatial-temporal framework. Within human ecology, it is especially the approach has been taken by historical ecology which by giving attention to cognitive aspects seems to captures the goals inherent in a critical analysis of space-time relationships. Crumley (1994) defines historical ecology as the "study of the ongoing dialectical relations between human acts and acts of nature as made manifest in the physical landscape." The premise is made that landscapes retain the physical evidence of practices, decisions, and ideas of past and present cultures, while providing practitioners of many disciplines--from ecology to geography to architecture to philosophy--with a common and useful concept.

Historical ecology seems quite up to date given its explicit attention to the landscape in. In recent globalization theory, Appadurai (1996) has sketched a framework for understanding global, cultural flows by emphasizing various "scapes," which serve as a metaphor for the fluid, irregular, deeply perspectival, situatedness of landscapes. Appadurai talks about ethnoscapes, mediascapes, technoscapes, financescapes, and ideoscapes to create a spatially grounded model focusing on disjunctures. Crumley's interpretation of landscapes through Historical Ecology and Appadurai's globalization model thus seem to have much in common. Both investigate the fractal nature of complexity (rejecting boundaries), the incorporation of chaos theory, a focus on dynamics, and an interest in an Aristotelian order and self-organizatizing systems (analog to the Medieval idea of "force"). Thus, in both globalization studies and historical ecology, theories are being developed aimed to capture the dynamics of space and time conceptualizations. As anthropologists who attempt to study humankind in all spaces during all times, these moves seem to be highly relevant to their goal.
 

1.4 Back to GIS

So, where is the relevance to GISs in all this? As any human evolution textbook could easily prove (try Lewin, 1998), technologies have always been an important factor in human biological and cultural evolution. Going back to recent history, Anderson (1991) points at the overwhelming influence which print capitalism had on the development of nationalism through the emergence of new time and space conceptualizations. Appadurai (1993) and Harvey (1990) have both followed Anderson's lead by pointing at the important influence of capital, technology, and imagination in the postmodern crisis. According to Harvey, the postmodern condition must be seen as an extraordinary situation of space-time compression, in which a fierce annihilation of space through time has made all spatial barriers collapse. Appadurai suggests that psychological space now embraces the entire planet—through the world wide flow of commodities and images—while showing stressful signs caused by deterritorialization. Both authors point to the explosion of communication and transportation technologies as major factors affecting this profound change.

As the crown of modern spatial technology and rooted in a history of carthographic colonialization (Anderson, 1991), GISs are deeply implicated in the rise of "postmodern stress." As such, it seems important for anthropologists to be somewhat cautious when treating the advances made by GISs as tools for the improvement of the world's ecological condition---which is not disputed here—as separated from their centrality in a complot aiming to destroy our social-psychological health, however unintentional this may be. To rebuttel this accusation, it is often argued that GISs are a "value free" technology, as is the case with any technology, and therefore not itself responsible for any of the consequences to which its presence has motivated the corrupt human soul. However, taking technology out of political debate much resembles controversial statements made by the American National Rifle Organization, which claims that it is not the guns who do the shooting, but the people. Such reasoning ignores two factors: first of all, the very presence of guns motivates behavior in a certain direction (the behavior of objects stand in direct relationship to the behavior of subjects, see Latour,1988), and secondly, guns are simply made to kill. It is this analogy I aim to explore in this paper on behave of the community of anthropologists fascinated by the indeed overwhelming application of GIS in solving real world problems.
 

2.1 The field of Geographic Information Research (Geomatics)

The application of Geographic Information Systems (GISs) is often classified into three levels (Arnold & Applebaum, 1992):

Obviously then, GISs have come to stand for more than just a little techno-toys in the hands of mapmakers from the Geography Department. The simultaneous occurrence of increasing ethical concerns and rapidly developing application illustrates the new situation in which the concept of "space" is embedded today. Goodchild (1997) argues that one of the more interesting questions in this area concerns "the nature of the geographies that will emerge in the information era, as telecommunication and the internet bring us closer to a world in which there is no more 'there'; everything will be 'here'."
 

2.2 Integrated Geographic Information Systems

When narrowly defined, GISs can be seen as a software packages for the storage, analysis and presentation of geographical information. In recent years GISs have collected several impressive technological partners to its repertoire, such as satellite driven remote sensing systems and global positioning systems, computer assisted carthographic systems, computer assisted drawing systems, and of course the global internet infrastructure. Together, these technologies are used to create a database which can be called an Integrated GIS, or IGIS (David et al. , 1991; Cowen et al., 1995; McGregor, 1998). An integrated GIS combines data from a very diverse set of sources: digital data (satellite data, elevation data, and thematic data such as land use data), legacy data (existing data such as census data), in-situ data (obtained from surveys), GPS data (exact point locations obtained by hand held receivers), and ancillary data (other data such as photo's, reports, websites, film, audio, etc.). The integration of these data based on a spatial location is key characteristic of an integrated GIS.

The basic concept of GISs is that one creates a computerized "layer cake" (see Fig. 1) of themes of spatial information, where each layer represents a different spatial attribute such as a transportation layer, a hydrological layer, an elevation layer, a land use layer, etc.
 
 

These individual data layers (from whatever sources) are all georeferenced, which means that they are all entered and stored in the GIS based on the same coordinate system, such as the Universal Transverse Mercator (UTM) projection system. In cartography a projection refers to a system of intersecting lines (of longtitude and lattitude) by which part or all of the three dimensional globe is represented on a flat surface. Today there are more than hundred different projection systems in use. The most common one is UTM, which was developed by Gerhardus Mercator in his 1569 navigational map (Haviland, 1997). Projections do not provide absolute locations. In fact, coordinate systems such as UTM are local systems, and based on different "datum’s". A datum is a mathematical model of the Earth, and because the earth is quite irregular, these models differ regionally. Thus, depending on the datum, one point can have a different set of coordinates. For example, Mercator's projection grossly exaggerates the size of land masses in northern hemisphere (coincidentally enlarging Europe and North America!)

GISs are represented by two fundamentally different modeling approaches. These are the vector (object) and raster (field) approaches. The vector approach represents the world as populated by simple objects: lines, points areas, typically roads, buildings, etc. This approach is good for describing data as gathered by traditional ground methods. The raster approach represents the world as a grid, with fields of data. Examples of such fields are elevation, concentrations of pollutants in the soil. This approach is good for describing data acquired with Remote Sensing. Briefly, while vector describes well defined boundaries and objects, raster approaches are associated with regions. The latter perspective suggests a sampling heuristic to test belongingness to a class (Goodchild, 1992). The difference between these two approaches is more fundamental than it appears, but beyond the scope of this paper.
 

2.3 Spatial Analysis

The eventual goal of any GIS is its application onto the real word. This application is couched in the particularities of spatial analysis. With regard to this form of analysis, Goodchild (1998) remarks that the idea that maps are part of (in)formal process of data analysis is still novel in many areas of the social sciences. Here, the orthodoxy still supports a non-spatial way of thinking, which searches for theories that are universal within human societies. In these theories, the sole role of maps is as statistical samples, each equally representative of a hypothetical population. Usually, a ranked list of, for example, nation states (see Table 1) is seen as informative enough in these sorts of analyses, because of their clear values representing each state.

In contrast to ordered non-spatial analysis, spatial analysis at first seems to lump data into a jungle of classes in many colors and shades. However, spatial analysis is a set of techniques which are developed to make sense of the patterns observed, depending on the location of the objects of analysis (Goodchild, 1998). Spatial analysis might be mathematically sophisticated, computationally intensive, or simply descriptive and intuitive. Several general themes can be discerned as unique to spatial analysis (Goodchild, 1998). These themes are patterns which reoccur due to the nature of spatial analysis itself. A first theme deals with spatial anomalies. The eye is remarkably efficient at scanning otherwise random patterns and detecting anomalies such as clusters or high density, which often occur in the distribution of diseases like cancer. Spatial analysis invites the eye to find such patterns. But it also corrects the eye by providing access to objective tests based on statistical principles. A second theme is spatial coincidence. Insight comes not so much from observation of anomalies, as recognition that it coincides with some other factor and the possible implication that this is somehow involved in causing the factor. A third theme is spatial proximity. Here, linkage of the causal factor over space is central. For example, a disease pattern might be linked to a specific cluster of neighborhoods. The less intense the disease, the further the distance to the source. A fourth theme is spatial dependence, which is based on the assumption that things that are close together in space are more alike than things further away in space. Some phenomena display greater spatial dependency than others. The distance over which dependency is observed varies as well. Because of spatial dependency, the number of truly independent observations in a given sample taken over geographical space is normally lower than one might expect—

spatial dependency lowers the effective numbers of degrees of freedom in a statistical test. This questions the traditional assumption behind the "random" statistical sample. A final theme is spatial heterogeneity. A landscape is heterogeneous and locally variable. Spatial heterogeneity invalidates the assumption of statistical tests that all cases are drawn randomly from the same population. The influence of a certain gradient causes a sample to be unequal in pattern relative to the entire population. For example, if an air photo had been positioned further upslope, the mix of settlements in the sample would have been different and the conclusion would have been affected.

GISs are build upon the principles of spatial analysis and combines these characteristics with rigorous statistical and mathematical analysis on an limitless database of spatial and other kinds of information (like census information). These qualities allow for comprehensive modeling and description of landscape structures (spatial organization), functions (interactions between the elements), and changes. Fox et al. (1995) suggest several quantitative methods which provide insight in landscape structures, including the simple analyses of patchiness (the number of different patches), dominance (measures homo- or heterogeneity of land cover types), contagion (measures the aggregation of land cover types, and fractal dimensions (measure the complexity of shapes). Walsh et al. (1996) explore the application of spatial metrics in GIS modeling in relation to human village settlements. Spatial metrics is an offspring of landscape ecology theory, and has been used to quantify landscape structures through composition and pattern metrics. Temporal change within GIS has traditionally been assessed using transitional matrices, which overlay grid-cell data of a given period with grid-cell data for various other periods. Recently, more research has been focused on the issue of change (Egenhofer & Golledge, 1998), including explorations of qualitative time (Couclelis, 1998).
 

The issues of conceptualizations of space and time in relationship to anthropology have already been discussed in Chapter 1. Here, however, the topic is not the importance of space and time to anthropology or alternative conceptions of space relevant to anthropology. Instead, the history of spatial and temporal analyses in anthropology will be briefly visited in an attempt to position the field relative to the practice of GISs today.
 

3.1 History of Spatial Analysis in Anthropology

Spatial thinking has long been a part of anthropology. In cultural anthropology at the beginning of the 20th century, attention was focused on the culture area (Kroeber, 1939; Barth, 1956) and diffusion concept. Diffusion is the borrowing of cultural elements from one society by members of another (Haviland, 1997). This notion identified spatial proximity and coincidence to explain similarities and differences between cultures. The rate of diffusion was held to be relatively unconstrained by the reality of physical space. The culture area concept originated in the late 19th century, and was further developed in the USA (a parallel existed in Europe) by Kroeber in the 1930s. In its most elementary form it is the spatial delineation of entire social formations, or associations of linked cultural particularities (Ellen, 1982). According to Ellen, culture areas were often selected spatially in too gross of a manner and lumped together indiscriminately. Furthermore, the arbitrariness of the factors employed to designate culture areas often resulted in conflicting typologies. However, Aldenderfer (1996) points out that one resembling trait of modern GISs with the culture area concept is precisely this scaling in size depending on the problem of interest. For example, Kroeber's Cultural and Natural Areas of Native North America (1939) listed continental scale areas such as California, Eastern Woodlands, and the Great Plains. At the same time, in California the regional "Pomo Culture Area" was defined by Gifford & Kroeber (1937) as to study variation in material culture of the Native American tribes. Kroeber used maps, map overlays, and tables, as well as so called "culture element distribution lists" to classify his areas. During this same period, archeology emphasized both important single sites as well as the diffusion cultural traits in prehistory. Gordon Childe’s influential books, The Dawn of European Civilization (1925) and The Danube Prehistory (1929). told the story of the spread of civilization from the Near East core area. Distribution of material culture—artifact styles—were called traditions, foci or phases. These notions were similar to culture areas, but with a difference in their explicit concern with time (often from a limited temporal frame) (Aldenderfer, 1996). The regional distribution of artifact types at sites was identified, establishing spatial boundaries for certain phases.

In the 1930s - 1940s anthropologists began to reject diffusion theories. New schools developed, few of which integrated space and spatial thinking in a meaningful way into the research process (Aldenderfer, 1996). These schools turned inward to demonstrate the role of history, place, and locality as primary means to understand cultural diversity with an increasing emphasis on evolutionary explanations of innovation. Thus, space became more passive and sterile as an analytical concept in this shift to the small scale village and participant observation.

In the 1950s - 1960s space became again reintegrated in anthropology. Four different sources motivated this development: aerial photography, the ecosystem concept, settlement approach and the new archeology (Aldenderfer, 1996). The use of aerial photography helped to maintain practical interest in larger scale phenomena. Already used in 1907 for taking photos of Stonehedge, aerial photography was first used by Archeologists after WWI. Perhaps the first ethnographic applications where done in 1937 by Marcel Griaule in Africa (Conant, 1993). Ethnographers were, however, reluctant to adopt aerial photography because of the use of a theoretical perspective which did not value spatial data. The combined influence of cultural geography and rural sociology and especially the emerging field of ecological anthropology led to John Rowe's (1953) use of aerial photography in ethnography to get regional scale data on settlement patterns, land tenure, house types, cultivation cycles, etc. (Aldenderfer, 1996). Ecological anthropology developed an emphasis on the integration of small scale ethnographic data and data obtained from research in regional-scale phenomena, such as deforestation. However, despite this renewed interest, maps remained visual aids or iconic devices rather than data themselves.

More directly interested in the analysis of the spatial layout of sites, archeology did, however, retain the spatial perspective from the 1930s on. The emergence settlement pattern approach in the 1950s boosted spatial analysis. Early use of this approach was focused on regional scale variability in the kinds of sites made by historic people and the degree these distributions reflected social norms and social interactions. Butzer (1993), however, argues that the empiricists of the 1950s did not actually use geographic techniques in a very sophisticated way, often leading to simplistic treatments of environmental variables within archeology. For archeology, the introduction of the ecosystem concept (Rappaport, 1967) connected with the settlement approach and the new archeology. The interest in explanation, quantification, and a scientific perspective motivated looking at other perspectives, notably geography. Many archeology students studied human geography. However, Butzer (1993) notes, again, that the concepts taken from these discipline remained only peripheral variables.

In 1972 the introduction of the new technology of remote sensing (LANDSAT 1 satellite) boosted regional scale studies. For the first time, very large areas could be viewed with modest levels of detail, while images could be updated every 18 days introducing an important temporal sequence into analysis. Thus, temporal and spatial variability in resource availability could be studied simultaneously and not solely dependent on being in the field. According to Aldenderfer (1996), the emergence of GISs in 1980 was quickly seized first by archeologists and followed by ecological anthropologists.
 

3.2 History of Temporal Analysis in Anthropology

Traditionally, anthropologists have assumed a classical dichotomy between "modern" and "traditional" societies with regard to the perception of time. In traditional societies cyclical was dominant rather than the linear time, qualitative change rather than quantitative change, stability rather than change, etc. In recent decades, this conception has become criticized based on, among others, the writings of Levi-Strauss (1962). Additionally, the tendency to look at the "ethnographic moment" in the frozen present of Anthropological discourse (Fabian, 1983), the rising awareness of the constitutive nature of knowledge (Marcus & Fisher, 1986) and the need for reflexivity (Ortner, 1984) have become central problems to the practice of anthropology with respect to time. However, despite these criticisms, Adam (1998) argues that the positivist, Newtonian belief in an uncontaminated, objective time is so deeply engrained in human and anthropological practice that many researchers still tend to view their own assumptions as immaterial and irrelevant. Ironically then, the omnipresent quality of time compels its ignorance.
 

3.3 Current Applications of GIS within Anthropology

Today, GISs are in use in mostly ecological and developmental anthropology, archeology, and cultural resource management. Five broad themes have been distinguished to give an overview of themes in which these systems are applied in these disciplines (Kvamme, 1989; Aldenderfer, 1996).

First, GISs have been introduced in regional data management. Within the context of federal legislation mandating historical inventories, archeologists have fruitfully applied spatial analysis in historic preservation. The usefulness of geographic databases in this effort may be obvious when considering that California alone has over 160,000 historic properties. In ecological anthropology regional data management has been incorporated to combine small scale informant data with regional analysis of remotely sensed images and to integrate the ecosystem concept on practical level. GISs are also used in development anthropology, where the goal is to find ways to reduce the impact of rapid cultural change. Here, the spatial scale is often considerably larger than in traditional anthropology.

A second area of application is in the management of remotely sensed data. Global environmental change is best observed at the regional and local levels. Anthropologist use remotely sensed regional data to cross check and add to local-level studies (Loker, 1996). As mentioned earlier, IGIS includes satellite data which is analysed by means of GISs. Furthermore, GISs are used to test formal models of processes of deforestation by indigenous groups by using remotely sensed data as a larger regional check.

A third area of application is in regional environmental analysis in the determination of the set of physical features which have the greatest influence on the character of human settlement in region. This type of research has a long tradition in archeology. Masschner (1996) describes a problem which has risen in this field with the limitations of predictive power of purely environmental models in archeology. The suspicion has risen that such environmental models impose a perceptual representation on the past that is not isomorphic with what may have existed in prehistory. Combinations of economic variables (arable land, soil type) may not be entirely sufficient because humans are culture bearers. To solve this problem, Masschner explored a GIS technique called viewshed analysis to analyze the importance of perception of the environment together with evolutionary psychology.

A fourth area of application is in simulation, or "what if..?" problems, where models are used to explore consequences in a dynamic matter. Relatively few studies like these are done, and this is surely not unrelated to the less developed issue of time in GISs. An example of this approach can be found in West and Kohler (1996), who model prehistoric agricultural activity to explore responses on climatic variability.

Finally, GISs can be used in locational modeling. The most obvious application here is the prediction of the location of some aspect of human behavior on the landscape. This is a much used technique in archeology in the prediction of the probability of sites based on physical, social, and ecological constraints. Relatively little has been done in ecological anthropology, although there is great desire to find ways in which to implement it.
 

4.1 Who's space and time are we talking about here?

With the sweeping history and success stories of GIS technology in anthropology, Aldenderfer (1996) concludes that it has a very bright future as a tool in anthropological and archeological research. Although this is undoubtedly true, the tone of Aldenderfer and Maschner's book (1996) seems to ignore some of the larger implications of the GISs endeavor. According to them, the only requirement for GISs to work is that the problem at hand has a spatial dimension, and this could even include postmodern perspectives! Would postmodernist agree with such a claim? Indeed, in Eco-Impacts and the Greening of Postmodernity Jagtenberg and McKie (1997) make ample use of the GIS discourse, proposing that mapping as a methodology to understand contextual fields is especially useful for the interdisciplinarians who think about the creation of new discursive spaces: (p. 36)

Remarkably, and at the same time not surprising, Aldenderfer goes on to make the following statement in his conclusion:

4.2 The hegemonic character of GIS

Following Latour (1988) the advocated alliance of GIS as the "model of the future" with the quest to put "space" back on the anthropological table must be taken with a grain of political salt. And even though my personal feelings are that Aldenderfer does not mean it that way—one would assume an anthropologist would be somewhat keen on alternative conceptions of space—the consequences of an alliance between Newton and GIS is of importance to anthropological research.

Is GIS value free? Following Curry (1998), the philosophers Russel and Whitehead claim that computers are complex logic machines that work in terms of a two-valued logic: true and false. As a result, systems created in the realm of science and computer technology carry the ideal of equations and modeling, and as such the ideal of the perfect, rule bound language. As a result the ideal language used in GIS is based on the artificial language of computer science, rather than natural languages used by humankind. Thus, using GIS to model ecological relationships in the real world frames these relationships in the restrictive language of ones and zeros. Thus, categorization of pixels which represent "real" spaces is based on either "in" or "out." A place cannot escape the compelling force of this in and out categorization by trying to be both at the same time. Being both in and out is absurd, even though this is exactly the kind of problem encountered when one accepts the possibility of multiple conceptualizations of space (and time) based on culture, practice, perspective, cognition, etc. Acknowledging this problem, a fuzzy logic approach has been proposed to remedy this problem. In fuzzy logic, places are allowed to be "inbetween." But inbetween of what? Of in and out! The problem remains that such conceptualization is itself highly dualistic, and not necessarily representative of many other cultures living in the global space which GIS aims to portray. Furthermore, being inbetween is still locked up in a discrete, pixelized notion of space which is used to categorizate entities. Such categorization is obviously not value free, since it is based on culturally relevant as well as politically charged interpretations of the nature of things. Many anthropologists have pointed out the problems inherent to such categorizations (Said, 1979; Marcus & Fisher, 1986; Abu-Lughod, 1991; Lutz, 1995; Gupta & Ferguson, 1997).

From a realist philosophy of science, then, the question becomes if an anthropologist can ethically support such hegemonic discourse of western space-time conceptualizations in the practice of his or her work. And unfortunately, reversing this homogenizing trend has become a lot harder than one would expect. Since a few years, GIS language is mandated in a Spatial Data Transfer Standard which describes how language ought to be conceptualized within GISs, and how language can and ought to operate within these systems. Furthermore, the evolution of GISs is guided by people who often believe in an universalist idea of time and space. From their perspective GISs are a value free technology, which ought to be applied everywhere, on all places, and in all times.
 

4.3 A cost/benefit analysis as seen from the normal world

Because I do not intend to skew this discussion toward a postmodern critique on the nature of objectivity, I would like to step back and take an objective (western, white, middle class, male) look at some of the benefits and problems given by GISs—as Newtonian as they might be—to the goal of anthropology. This I would like to do because, after all, GISs do seem to help the world become a better place. Or do they?
 

4.3.1 Some of the benefits of using GIS

First of all, it might be obvious that all the applications mentioned in section 2.5 provide an overview of the benefits of using GISs in applied anthropological settings. If this were not the case, few explanations could account for its widespread use. Here I intent to not repeat these applications and their real world benefits, but rather focus on some general benefits inherent to the structure of GISs.

GISs provide an excellent platform from which the integration of small with regional and global scale data can take place (Aldenderfer & Maschner, 1996). Loker (1996) gives an example of a study which applies GIS in analysis of land degradation in the Peruvian Amazon. When based on existing data, he sees GISs as a relatively low cost methodology for the analysis of an entire region. Such regional analysis using GISs can show where agricultural induced degeneration is taking place (high risk areas) and provide preliminary info on what sort of activities promote degradation and how. This analysis must be seen as a first phase in a multi-phase research program. In a second phase, fieldwork is needed to resolve ambiguities and provide crucial additional cultural information. Such information would not have come available using conventional methods due to the insurmountable costs underlying surveying and research. Policy makers often need regional scale information, and GISs facilitate the interaction of anthropologists with regional scale models. A further benefit of GISs lie in the bridging quality it provides for interdisciplinary cooperation. As data from many different sources become available, GISs integrate all this information, familiarizing the analysts and specialists with each others perspectives. Another related advantage lies in the multi-scalar nature of GISs which allow investigators to manipulate the scale and dimensions of the image being analyzed to discover the best format. Thus, multi-dimensional analysis is allowed through the relative ease with which the system can switch from one "view" to another. This feature of GISs make it especially relevant to anthropology’s study of multi-scalar ecological and social interactions. GISs also greatly facilitate the analysis and dissemination of information. GISs facilitates rapid preparation of maps for conveying related sets of information and open up new analytical possibilities because it allows for rapid numerical transformations. This allows the investigator to explore multiple questions which would be too cumbersome or time consuming to explore by hand. Most importantly perhaps, GISs greatly improve informed decision making. The combination of spatial and statistical analyses with a large amount of integrate data leads to more integrated perspectives on the problems at hand. GISs should lead to a greater dialogue with other social sciences on the opportunities opened by these decision making possibilities.
 

4.3.2. Some of the problems of using GIS

A first problem related to the issue of representation of space (Goodchild, 1996; Jagtenberg & McKie, 1997; Curry, 1998; Couclelis, 1998; ). As main theme of this paper, I believe this issue has been amply covered. All left here is to mention that the argument goes very similar when made with respect with the issue of time. Thus, a second problem deals with time. Spatial analysis is primarily analysis of form, whereas understanding requires analysis of process. GISs represent a "snapshot" of a Newtonian reality, ignoring the strongly temporal nature of mechanisms of cause and effect. It is suggested that animation is a way out of this problem. Recent research is looking into the possibility of multidimensional GISs (Egenhofer & Golledge, 1996). However, animation still might not capture the significance of narrative, because there is a difference between chronology--clocktime--and narrative--meaningful time. The large story of time can be told in many different ways and the narrative is always told by a narrator. Essentially, conceptualizations of time need much exploration in the terrain of GIS research. A third problem has already been mentioned (see 2.5) and deals with developmental differences in human landscape perception when viewed over evolutionary time (Aldendorfer, 1996). Obviously this problem is highly problematic and important to archeologist trying to deal with stylistic issues of related to artifacts of prehistoric cultures. The issue of variability in conceptualizations of space and time lies at the heart of this problem.

A fourth problem deals with the problem of transparency. GISs are products of many forces and does not have available a way to establish a relationship of trust with the author of its narrative, or to develop means of feeling competent to judge reliability (Curry, 1998). There are many reasons for this:

• hardware is product of multiple creators
• hardware organization is not completely transparent to the viewer
• software is difficult and complex to read
• software algorithms are kept secret by corporations (see below)
• belief in adequacy of data can be merely act of faith, despite testing.
• standards of data are set remotely (see below)
• sense of understanding of the "whole thing" is never revealed

A seventh problem relates to the hegemony of geography on geotechnology. Especially in the Unites States, disciplinary boundaries prevent integration (Curry, 1998). "Geomatics" or Geographic Information Research is an interdisciplinary field (Goodchild, 1997). An eight problem is that the political development of global standards for GISs are based on the needs of companies exploiting the earth and governments seeking to control these resources. As a result, standards are manipulated and set in the political context of these interests. Key issues of concern in this context are: privacy, copyright, access, and ultimately power in the information society (Arnold & Applebaum, 1996; Dow, 1998; Curry 1998). Standards have promoted a united view of geographic data through the National Spatial Data Infrastructure. The 1992 Special Data Transfer Standard (SDTS), has been developed in collaboration between government, academics, and industry and created a standardized formal system for marking out the face of the earth. These standards categorize elements and attributes according to a universal applicable vocabulary: "a river here will be a river there." Although this certainly has its benefits when exchanging data, it also normalizes the perception of space and time (Foucault, 1978). A ninth problem deals with the increasing intrusion of capitalism in the research world. Because GIS software and hardware is very complex, difficult to learn, and expensive, it becomes difficult to routinely use more than one system. This creates increasing dependence on commercial manufacturers. Commercial research funding further shifts power to corporate interests. A wide range authors suggests that models of the computer, computer reason, and of scientific reason are closely, even intrinsically tied to capitalism, industrialism, and masculine culture (see Curry, 1998). A tenth problem comes with the underrepresentation of less powerful in existing spatial data (Loker, 1996; Dow, 1998). Especially with regard to the developmental world data is often absent or incomplete. Data derived from secondary sources is often suspect. This results in the systematical underrepresentation of transients, homeless, undocumented workers, gypsies, etc. in GIS data systems. An eleventh problem is related to the issue of geodemographics and impact on society. The recent fusion of US Census Data with geographic ZIP codes have jeopardized privacy (Curry, 1998) and boosted direct marketing technologies, as well as governmental and business control. People are increasingly seen as aggregates based on categorizations over neighborhoods ("people who are alike tend to cluster") in an ongoing reconceptualization of the objects that make up the world. Ownership of the data resigns with the corporations who determine information about people. The geographical question has thus shifted from "where am I" to "where am I on their map"

A twelfth problem relates to the perception that information is increasingly seen as representing the real world. Technologies of making out the face of the world and tracking locations of people motivate the idea that everything can be characterized in terms of information (Curry, 1998). A thirteenth problem is the lack of sense of place in GISs. In GISs, places are seen as points, and not as loci of human actions. This denial of practice is further destructive because of the underrepresentation of time. Places are historical social communities and created through time. Without time, there can be no place. A fourteenth problem is the incapability of GISs to represent normatives of cultures. "What is good or bad, what is right and wrong." There are no normative and self-correcting mechanisms built into computer tools. As such, the tendency of these complex, non-transparent tools to assist in important decision making processes might jeopardize ethical norms.
 

As I have hoped to show in this paper, GISs are powerful players not only in the future shaping of these two rapidly evolving spaces, but also in the subjugation of other conceptualizations of space and time to the Newtonian perspective. This includes culturally colored spaces which are highly relevant to anthropological investigations. Aiming to preserve cultural diversity, anthropologists need to become aware of this normalizing character of GIS technologies (Foucault, 1978). However, at the same time, GISs offer incredible benefits to the solution of complex social problems such as population growth, planning, environmental degradation and cultural preservation. Unfortunately, detailed anthropological studies on the evaluation of the language of mathematics underlying GIS analyses and application are far and between. Egenhofer & Colledge (1998) provide some encouraging examples of such studies from the disciplines of engineering, psychology, geography and philosophy. The participation of anthropology in this discussion seems very relevant and perhaps even highly necessary given the political implications of different conceptualizations of space and time. With a tense relationship to discourses of colonization and hegemony, attention to the fast expanding world of GIS supported decision making is of great importance. Because of its unique disciplinary history, it would perhaps be misguided for anthropologists to assume that other disciplines would and could bring a similar perspective to the discussions on the relationship between GIS and society as anthropology could (see for example Butzer, 1993). Furthermore, epistemological issues related to the anthropology itself are being played out dramatically in the philosophical issues underlying the design of improved GISs. The irreversibility of this development demands the attention of a discipline involved in history and evolution. Perhaps a critical, historical ecological perspective could take on this very challenge, although I would not want to imply that this is the only perspective relevant to anthropological information research. The congruence of historical ecology’s focus on cognitive and cultural perception of the physical landscape from a multi-spatial and multi-temporal perspective with the very spatial nature of GISs modeling tools almost speaks for itself.

However, I believe some caution is here in place. The very reliance of historical ecology on finding its traces in the physical landscape might in the end jeopardizes the extent to which such a critical reflection on conceptualizations of space and time will manage to grasp the entire scope of GISs. The dominant spaces of the postmodern world—chaotic and unstable imaginary space and the quickly emerging and powerful reality of electronic space (cyberspace)—are examples of anarchistic powers trying to avoid the grasps of a physically grounded historical ecology. One could argue that this occurs despite the fact that these spaces directly influence spatial organization, and thus the physical landscape. Reason for this is that the fusion of imaginary and electronic space function to create a new dimension of reality which is entirely embedded in electronic fantasy. The tendency of disconnectedness of this imaginary electronic space to the natural world—so important to the survival of the global ecosystem—is all too well known by concerned environmentalists trying to convince multinational corporations that community is of importance to the world. Perhaps historical ecology could solve this problem by including in its definition the historical and evolutionary analysis of the electronic landscape as well. Interestingly, such an electronic landscape has become increasingly hardwired by the standardized and globalization of the GIS network.

Concluding, I believe anthropology is much behind in exploring the historical situatedness of the language of spatio-temporal reasoning, which ironically lies at the fundament of the definition of the discipline itself. Anthropologists applying and interested in using GISs therefore have to be aware of this neglect in order to use, criticize, and improve its application in the spaces of being and becoming.
 
 

• Abu-Lughod (1991) Writing against Culture. In: Recapturing Anthropology. Pp 137-162.
• Adam. B. (1998). Perceptions of time. In: The Encyclopedia of Anthropology, Ed. T. Ingold. Routledge.
• Aldenderfer, M. (1996). Introduction. In: Anthropology, Space, and Geographic Information Systems. Eds. Aldenderfer & Maschner, Oxford University Press.
• Anderson, B. (1991). Imagined Communities. Verso.
• Appadurai, A. (1996). Modernity at Large: Cultural Dimensions of Globalization. University of Minnesota Press.
• Arnold, C.G. & Appelbaum, R.P. (1996). The Use of GIS to Measure Spatial Patterns of Ethnic Firms in the Los Angeles Garment Industry. In: Anthropology, Space, and Geographic Information Systems. Eds. Aldenderfer & Maschner, Oxford University Press.
• Barth, F. (1956). Ecological Relationships of Ethnic Groups in Swat, North Pakistan. American Anthropologist, Vol. 58.
• Bourdieu, P. (1977). Outline of a Theory of Practice. Cambridge University Press.
• Childe, V. (1925). The Dawn of the European Civilization. London.
• Childe, V. (1929). The Danube in Prehistory. London.
• Conant, F. (1993). 1990 and Beyond: Satellite Remote Sensing and Ecological Anthropology. In: The Ecosystem Approach to Anthropology. Ed. E. F. Moran, University of Michigan at Ann Arbor.
• Couclelis, H. (1998). Aristotelian Spatial Dynamics in the Age of Geographic Information Systems. In: Spatial and Temporal Reasoning in Geographic Information Systems. Eds. Egenhofer and Golledge. Oxford University Press.
• Cowen, Jensen, Bresnahan,. Ehler, Graves, Huang, Wiesner and Mackey (1995). The design and implementation of an integrated geographic information system for environmental applications. Photogrammetric Engineering and Remote Sensing. Vol. 61, No.11, Nov. 95, pp. 1393-1404.
• Crumley, C.L. (1994). Historical Ecology, A Multidimensional Orientation. In: Historical Ecology, Cultural Knowledge and Changing Landscapes. Ed. C.L. Crumley. School of American Research Press. Santa Fe.
• Curry, M. (1998). Digital Places. Living with geographic information technologies. Routledge, London.
• David, Quattrochi, Ridd, Lam, Walsh, Michaelson, Franklin, Stow, Johanssen, and Johnson (1991). Environmental anaysis using integrated GIS and remotely sensed data. Some research needs and priorities. Photogrammetric Engineering and Remote Sensing, (57)6:698-97.
• Dow, J.W. (1994). The mapping of cultural traits from field data. Social Sciences Review 12(4):479-492.
• Egenhofer and Golledge (1998) Spatial and Temporal Reasoning in Geographic Information Systems. Oxford University Press.
• Ellen, R. (1982) Environmental determinism and causal correlation. In: Environment, Subsistence and System. The ecology of small scale social formations. Cambridge.
• Entwisle, B, Rindfuss, R.R., Walsh, S.J., Evands, T.P., & Curran, S.R. (1997). Geographic Information Systems, Spatial Network Analysis, and Contraceptive Choice. Demography, Vol. 34(2):171-187.
• Fabian, J. (1983). Time and the Other. Columbia University Press, NY.
• Foucault, M. (1978). History of Sexuality. Vintage.
• Fox, J., Krummel, J., Yarnasarn, S., Ekasingh, M., Podger, N. (1995). Land Use and Landscape Dynamics in Northern Thailand: Assessing Change in Three Upland Watersheds. Ambio Vol 24 No.6 Sept. 1995.
• Geertz, C. (1976) From the native's point of view: on the nature of Anthropological understanding. In: Meaning and Anthropology. Eds.: K. Basso & H. Shelby. Santa Fe: SAR Press.
• Giddens, A. (1993). Preface. In: Society, Action, and Space. An Alternative to Human Geography. Werlen. Routledge.
• Gifford, E. & Kroeber, A. (1937). Culture element distributions: EV, Pomo. University of California Publications in American Archeology and Ethnology, 37:117-254.
• Goodchild M.F. (1992). Geographic data modeling. Computers and Geosciences. 18:401-408.
• Goodchild, M.F. (1996). Geographic Information Systems and Spatial Analysis in the Social Sciences. In: Anthropology, Space, and Geographic Information Systems. Eds. Aldenderfer & Maschner, Oxford University Press.
• Goodchild, M.F. (1997). Postscript: new directions for research. In: Geographic Information Research, Eds. Craglia & Couclelis. Taylor & Francis
• Gupta, A. & Ferguson, J. (1997). Anthropological Locations: Boundaries and Grounds of a Field Science. University of California Press.
• Haraway, D. (1989) Introduction: the persistence of vision. In: Primate Visions: Gender, Race, and Nature in the World of Modern Science. Routledge.
• Haraway, D. (1991). Situated knowledges. In: Simians, Cyborgs, and Women: The Reinvention of Nature. Routledge.
• Harvey, D. (1990). The Condition of Postmodernity. Blackwell, Oxford.
• Hastorf, C. & Johannessen, S. (1991). Understanding Changing People/Plant Relationships in the Prehispanic Andes. In Processual and Postprocessual Archaeologies: Multiple Ways of Knowing the Past, ed. Preucel. Occasional paper No. 10, Center for Archeological Investigations, Southern Illinois University.
• Haviland, W.A. (1997) Putting the world in perspective. In: Anthropology. Harcourt Brace College Publishers.
• Heuvelink, G.B.M. (1998). Error Propagation in Environmental Modeling. with GIS. Taylor and Francis, Bristol, PA.
• Hirtle, S. C. (1998). The Cognitive Atlas: Using GIS as a Metaphor for Memory. In: Spatial and Temporal Reasoning in Geographic Information Systems. Eds. Egenhofer and Golledge. Oxford University Press.
• Hodder, I. (1985). Postprocessual Archeology. Advances in Archeological Method and Theory. 8:1-26.
• Jagtenberg, T. and Mckie, D. (1997) Eco-Impacts and the Greening of Postmodernity SAGE Publications, CA.
• Jochim, M. A. (1991) Archeology as long term ethnography. American Anthropologist 93(2):308-321.
• Keat, R. & Urry. J (1982). Social theory as science. Routledge & Kenan Paul.
• Kowaleski, S. A. (1995). Large-scale ecology in aboriginal Eastern North America. In: Native Interactions: Multiscalar Analyses and Interpretations in the Eastern Woodlands, ed. M Nassaney & K. Sassaman. University of Tennessee Press, Knoxville.
• Kroeber, A. (1939) Cultural and Natural Areas of North America. University of California Publications in Anthropology, Vol. 38.
• Kvamme, K.L. (1989). Geographic information systems in regional archeological research and data management. In: Archeological Method and Theory, vol. 1. Ed. Schiffer. University of Arizona Press, Tuscon.
• Latour, B. (1988). The pasteurization of France. Harvard University Press.
• Laszlo, E., Masulli, I., Artigiani, R., Csanyi, V. (1993). Introduction. In: The evolution of cognitive maps. Gordon and Breach.
• Levi-Strauss, C. (1962) The Savage Mind. University of Chicago Press.
• Lewin, R. (1999) Human Evolution; an illustrated introduction. Blackwell Science.
• Loker, W. M. (1996). Land degradation in the Peruvian Amazon: Applying GIS in Human Ecology Research. In: Anthropology, Space, and Geographic Information Systems. Eds. Aldenderfer & Maschner, Oxford University Press.
• Lutz, K. (1995). The Gender of Theory. In: Women writing culture. R. Behar & D.A. Gordon. University of California Press.
• Malinowski, B. (1922). Argonauts of the Western Pacific. Waveland Press, Illinois.
• Marcus, G. & Fisher, M. (1986). Anthropology as a Cultural Critique. Chicago University Press.
• Maschner, H.D.G. (1996). The politics of settlement choice on the Northwest coast: cognition, GIS, and coastal landscape. In: Anthropology, Space, and Geographic Information Systems. Eds. Aldenderfer & Maschner, Oxford University Press.
• McGregor, S. (1998). An integrated geographic information system approach for modeling the suitability of conifer habitat in an alpine environment. Geomorphology, 2:265-280
• Ortner, S. (1984). Theory in Anthropology Since the Sixties. CSSH 126-165.
• Rappaport, R.A. (1968) Pigs for the Ancestors: Ritual in the Ecology of a New Guinea People. New Haven: Yale University Press
• Rapoport, A. (1998) Spatial organization and the built environment. In: The Encyclopedia of Anthropology, Ed. T. Ingold. Routledge.
• Said, E. (1979). Orientalism. Vintage.
• Walsh, S.J., Evans, T.P., Welsh, W.F., Rindfuss R.R., Entwisle, B. (1996). Population and environmental characteristics associated with village boundaries and land use/land cover patterns in Nang Rong district, Thailand. In press, proceedings 1996 Pecore Symposium on "Human interactions with the Environment: Perspectives from Space."
• Werlen, B. (1993). Society, Action, and Space. An Alternative to Human Geography. Routledge.
• Winterhalder, B. P. (1994). Concepts in Historical Ecology: The View from Evolutionary Theory. In: Historical Ecology, Cultural Knowledge and Changing Landscapes. Ed. C.L. Crumley. School of American Research Press. Santa Fe.
• Winterhalder, B. & Smith, E.A. (1992). Evolutionary Ecology and the Social Sciences. In: Evolutionary Ecology and Human Behavior. Eds. Smith & Winterhalder. Aldine de Gruyter, New York.