Our motivation for promoting synthesis in human ecology is premised on the recognition that information flow is as central to understanding human systems as matter and energy flows. Our premise is that a truly human ecology should be responsible for the integration of both sociocultural ecology (traditionally the domain of the humanities) and biophysical ecology (traditionally dominated by the so-called `natural sciences'). The history of human ecology's disciplinary development has led to increasing division of labor and fragmentation of knowledge, often delimiting potential know-how and restricting the scope of skill acquisition.
This has sometimes resulted in such specialized worldviews that practitioners can not share common vocabularies or conceptualizations (for example, the anthropology department at Stanford University dividing into two departments: cultural and social anthropology vs. anthropological `sciences'). On the other hand, the focus resulting from specialization within human ecology has resulted in considerable refinement of knowledge within each restricted domain. The discipline as a whole, then, appears in need of synthetic integration.
There have been several calls for just such an effort in human ecology. Redman, Grove and Kuby (n.d.) state that "[a]lthough it is not novel to recognize the interconnectedness of humans and the[ir] environment[s] (cf. Marsh 1864; Thomas 1956), constructing a new approach emphasizing an integrative framework equipped with comprehensive models, reinforcing methods, and complementary data is a growing and urgent priority." Scoones (1999) reviews the potential cross-fertilization of social science and ecological thinking, identifying several research topics in social science that actively use the principles of dynamic equilibrium, spatial and temporal variation, complexity and uncertainty, including historical ecology, `structuration' or contextual/interactionist perspectives, and complexity in socio-ecological systems. Our own view, that a truly holistic approach to understanding human ecosystems must recognize and investigate the central role of information, was articulated almost thirty years ago by Flannery (1972:400):
Up until now, it has mainly been the humanists who have studied the informational aspects of complex societies-- art, religion, ritual, writing systems, and so on. The `ecologists' have largely contented themselves with studying exchanges of matter and energy...humanists must cease thinking that ecology `dehumanizes' history, and ecologists must cease to regard art, religion, and ideology as mere `epiphenomena' without causal significance. In an ecosystem approach to the analysis of human societies, everything which transmits information is within the province of ecology.
Go back to the Introduction to this essay and look at Figures 1a, b, c, d and e again. Keep our goal of integrating biophysical and sociocultural ecology in mind as you read the following proposed framework for a theory of human ecosystems.
Radical Synthesis
A partially nested hierarchy of paradigms relevant to human ecology can be seen in Box 4. What would be the nature of potential integration across these contrasting paradigms?
For example, population and ecosystem paradigms historically represent two apparently opposed approaches within bioecology (Box 5). These paradigms can also be applied to ecological anthropology (Figure 10). A recognition that populations function as parts of ecosystems is a potential point of integration. The ecosystem is the most inclusive ecological conceptualization at any given level in the ecohierarchy. But because it is holistic, it is also difficult to achieve conceptually. Conceptual stumbling blocks include the prevalence of diffuse causality, indeterminism and the difficulties of bounding variables and systems.
In addition to being part of natural ecosystems (sensu lato), humans also define themselves by their worldviews (± praxis) as part of constructed world systems. This results in a hierarchy of human ecological levels something like that depicted in Table 1. Human ecosystems may be recognized at each of these levels, a potential starting place for conceptual integration across subdisciplinary paradigms. The central role of information in human ecology makes such integration more plausible than it would be in biological ecology, where the integrative potential may have to come from focusing on questions that lie at the intersection of the subdisciplines, questions that cannot be addressed by either of the two paradigms already described acting alone.
In the development of an integrated theory of human ecosystems, it is important to remember that no single objective of theory takes precedence over all the others. Every theoretical objective contributes to the dialog of understanding. This pluralistic view should help alleviate unproductive and damaging debate about what single method is best. Moreover, a theory of human ecosystems could at once be evolutionary/historical, dynamic/dialectical, abstract/empirical and general/specific (Table 2). In restricted domains, critical theory, for example, might be self-contradictory and still contribute to understanding.
Fundamental Questions for an Integrated Human Ecology
In order to radically integrate sociocultural ecology and biophysical ecology, it is necessary to identify fundamental questions that might motivate or `drive' this effort. Examples of such fundamental integrating questions might include the following.
To what extent are the causes of modern human ills inherent to the nature of our socio-cultural systems (Robbins 1999:ix)? Related to this, why do humans persist in destroying the life-support systems of the planet (Shepard 1982:1)? At best, will the next millenium see the conversion of a biologically diverse planet into a completely human-dominated noösphere (cf. Wyndham 2000:87)? Questions such as these may be addressed a number of ways at a number of levels. Perhaps new theoretical components are required to successfully address these questions, components that would not be recognized as relevant otherwise. Ultimate (or historically distant) causalities, as opposed to proximate causalities, may be involved, separated from considerations of system function and the ontogeny of reproducing patterns of daily life. Addressing these questions systematically may force us to challenge the completeness, applicability or coherence of current theory, leading us to encompass components or relationships well outside currently accepted domains of explanation and understanding.
Methodological Tools for Investigating the Complexity of Human Ecosystems
Methodological tools are designed to effectively relate conceptual constructs to observable phenomena (Figure 4). Some of the methodological tools (m-tools) listed in Figure 4 that are most relevant to the construction of a theory of human ecosystems include the following.
Expanding Contextualization
While used here for theoretical purposes, the method has applied beginnings in the East Kalimantan project of UNESCO's Man and the Biosphere program (Vayda 1983). For this project, units of analysis were sought by analyzing the social or spatial system by which resources were managed (Vayda 1983:267). To do this, initially a system of a small size was bounded, then made wider and wider or denser and denser (Vayda 1983:265). The strengths of expanding contextualization include: avoidance of assumptions that unnecessarily reify unit or system stability; allowance for flexibility in required time/money/effort; demonstration of practicality of results to lay people (policy makers); and better understanding of dynamism as well as stasis and persistence.
Marxian Method
In the Marxian method, priority is given to the `How' (Box 6). Historical materialism creates "...an understanding of social and structural conditions based on internal relations of entailment rather than external ones of cause" (McKinlay and Taylor 1998), such that cause and effect become interchangeable (Harvey 1996:54). The role of theory is not to predict, but to explain through a set of generative and transformative principles that reveal possibilities. "The purpose of materialist inquiry is not to test...but to show...forms... domains...effects and transformative possibilities...The problem is to explore the forms and domains of operation" (Harvey 1996:67). Harvey's dialectical method closely parallels our system analysis, with emphasis on understanding processes, flows, fluxes and relations rather than on elements, content, things, fixed structures and static organizations. The dialectic (see Figure 4, methodological tools) prioritizes a search for fundamentally contradictory forces. For human systems, dialectical fields can be informational with associated material/energy sinks, as well as the creative dialectics identified in Marxian (and Hegelian) methodology. In the former case, increasing amounts of energy may be dumped into points of structural contradiction in a human system, and these may become runaway sinks. In a `bubble' economy, for example, governments will often keep throwing money at fear to boost `consumer confidence' even though it is explicitly recognized that the market `value' is drastically inflated relative to real production.
Postmodern Method
Postmodern and poststructural theorists are decidedly against grand theory, integration and universalism, yet there are several valuable methodological insights that a developing theory of human ecosystems can borrow from postmodern thought (Box 7). The postmodern position advocates critical social theory that analyzes and deconstructs the historical basis for our discourse/practice, our notions of truth and modes of intellectual domination, thereby hopefully undermining their appearance as the natural, i.e., inevitable, outcome of sociocultural evolution (cf. Poster 1984:159). This point of view provides valuable shifts in methodological perspective that forces us to continually recognize the constructed nature of theory itself. Also, the production and reproduction of `power,' a central theme in postmodern and poststructural analyses (cf. Russell 1938; Foucault 1972; Jameson 1991), is clearly relevant to the evolution of human ecosystems.
Causality
The targets and some of the aspects of causal explanation are shown in Box 8. Causalities may be conceptualized in a number of ways, including direct, indirect, ultimate and proximate.
The Aristotelian conceptualization of causality consists of four main components: material, efficient (mechanical), formal and final (Table 4). Three of the four causes work at different levels of scale, and can be considered to be hierarchical in nature (see Figure 11). The efficient cause will act on a small field of influence. The formal cause operates at the `focal' level of observation. Events will transpire under conditions constrained and permitted by the final agents of causality (Ulanowicz 1997). Newton's Principia led to the dominance of the mechanical form of efficient cause in the physical sciences (Ulanowicz 1997). In the sciences influenced by Darwinian biology, however, final cause (based on purpose or design) dominates in the development of theoretical explanation (Gleick 1987:201). This is true of functionalist and materialist anthropology too, as seen in the search for final causes that explain cannibalism, religious practices, etc. and in the recent enthusiasm for evolutionary psychology's explanations of modern social behavior. Care must be taken to avoid preconceiving the kind of causality relevant to systems of interest; in most cases it is likely that at least both efficient and final causality play important roles.
Causation may also be either direct or indirect. Patten (n.d.) discusses both direct and indirect causation. Transactions are the observable transfers of resources between organisms. Relations are the direct and indirect consequences of transactions. Indirectness is a separation between the organisms, a lack of direct connections between them. Indirect effects may influence a system through transactions that are not directly linked. Indirect causality is more important than direct causality in complex ecological systems (Patten n.d.).
Ultimate forms of causality refer to inherited structures or enabling and constraining conditions whose origins can be traced deep into the past. In order to understand present structures/systems (cultural, social, biological, physical) one must identify not only their genesis and context, but also their evolutionary histories. Figure 12 depicts some important features of historical transformation. Initial conditions set the stage; boundary conditions restrict potential outcomes; and transformations rupture systems to effect significant change and mark periods of `internal logic.' Internally, priority effects and echoes of the past provide historical continuity.
Hypothesis Testing
Theories that attempt universal explanations of human ecosystem phenomena are subject to a wide range of scale limitations that affect our ability to derive testable hypotheses (Table 5). Currently it is not possible for the individual researcher to judge the success of theory that forecasts or predicts over a very broad time scale or world system spatial framework. Creative attempts to transcend time/space compression and limitations of the individual observer include, for example, historical fiction and predictions based on thought experiments that incorporate current data into science fiction. Depending on the framework in question we find various shortcuts used in testing hypotheses.
The dominant mode of testing is the comparative method (Box 9), in which complex, multicausal models are subject to varieties of evidence. The conclusions drawn are usually probabilistic and contingent (Box 10). Because of the immature development of practitioner and theory in human ecology, predictions often seem riskier (Box 11) than conventional forecasting and prediction in other (non-human) fields of study.
Multiple Working Hypotheses
Chamberlin (1890, 1965) offers an important methodological discussion that can be partly summarized as follows. Hasty explanation quickly leads to the development of tentative theory. With unconscious bias, a favored tentative theory rapidly passes into an adopted theory, and then on to a ruling theory. The defects of this common chain of events have led historically to both the condemnation of theorizing in general, and to the method of the working hypothesis. Unfortunately, a working hypothesis easily becomes a controlling idea. To guard against this, Chamberlin urged the method of multiple working hypotheses. He points out that one of the chief merits of this method is that it encourages the development of complex explanations, with multiple causalities. The method also promotes thoroughness and a habit of parallel or complex thought wherein the practitioner appears capable of simultaneous vision from different standpoints. Phenomena appear to be viewed analytically and synthetically at once. He further notes that this type of complex thought cannot be expressed verbally in words, and that words and thoughts lose the close association that they usually maintain with those whose silent thoughts, as well as spoken thoughts, run in linear verbal courses. He confesses that one drawback of the method is that it introduces that difficulty in expression, and concludes that therefore there is a certain predisposition on the part of the practitioner to taciturnity. We should note that in the late 1800s, when Chamberlin wrote, graphical models had not yet been developed to the point where they could be used to express the kind of complex scientific thought that he knew so well from his own personal experience.
Concepts of Evidence
Concepts of evidence (Box 12) are the ways-and-means that allow for confirmation and rejection of the empirical content of theory, i.e. facts and well-established generalizations. The focal mode of thinking is white hat, the `checking and re-checking mode' (see Appendix), in which it is important to specify limits to knowledge and common sources of error in judgement. Consilience with the D.I.F. criteria (Diversity, Independence, Fit, see Box 12) is the ultimate standard.
