Introduction

The notion of industrial metabolism draws attention to a materialistic view of the economy as a physical system, driven by energy flows. Such a conception is less trivial than it seems, since money functions as a unifying principle of economy to such an extent that it is difficult to raise awareness and understanding of physical (non-monetary) concepts. The physical dimensions of the economy usually are discussed only as tools for the development of monetarization, not as autonomous concepts. Similarly, the social sciences, as well as economics, tend to view social reproduction as a system of communication, and not in physical terms.

If you conceive of the economy as a physical system, drawing physical inputs from its natural environment, processing them internally, and generating physical outputs to this environment, you have to define a boundary between the "system" and its "natural environment": you have to be able to tell what is "inside" and what is "outside." This boundary is both omnipresent and fugitive. It certainly cannot be a "physical" or topographical boundary: The same physical elements will be both part of the economic system and part of its natural environment, depending on the point of view. One can only construct a functional boundary, and this has to be done with care. Two approaches may be chosen.

First, an a priori theoretical approach would discuss the possible functional labels of physical entities and processes that should define them as inside or outside the system. This might be their function as goods and services in markets (a narrow approach that would leave aside the so-called "free goods," and could not easily be applied to elements of subsistence economies); it might be their function for "humans" in terms of a biological species (which would be a very broad approach, difficult to connect to a specific concept of economy). We feel the most promising approach would be the functional link to property: property is specifically human and it constitutes a functional connection between physical entities and economic "subjects." But we will not pursue this question here, though we would like to encourage such a discussion. As long as this question is not resolved satisfactorily on a theoretical level, we prefer to speak of the "socio-economic system" rather than of "industrial economy." Sometimes we also use the term "economy," implying a wider historical range of modes of production.

Another, second, approach is a strictly constructivist one. It presumes that society "constructs" the boundary of its natural environment by the environmental information system it uses. The environmental information system itself defines what is to be considered part of the system and what is to be considered as an element of its (relevant) "natural environment." Practically speaking, this is the approach we chose in this paper; it leads to an implicit definition of the boundary between the socio-economic system and its natural environment.

This leaves us with the need for a mode of selection of physical processes that are relevant within an environmental information system that is supposed to describe socio-economic metabolism. This mode of selection should be selfreferential to the socio-economic system, in the sense that it selects, for the possible present or future, harm that feedbacks from the natural environment to the system may cause. In view of our limited knowledge of interdependencies, it should also take into account the self-regulating qualities of natural systems for their own sake.

The second section of this chapter, therefore, attempts to outline what we think are the basic paradigms for conceiving of such interdependencies. The third section is devoted to the overall structure of an information system that might qualify for the standards set. We have proposed this information system to the Austrian government, with some chances of success. In the last two sections we provide empirical illustrations of how the Austrian economy would perform within such an information system.