Industrial ecology : Reflections on a colloquium ( environmental technologies / dematerialization / decarbonization )

Industrial ecology is the network of all industrial processes as they may interact with each other and live off each other, not only in the economic sense, but also in the sense of direct use of each other's material and energy wastes and products. This paper, which reflects upon the papers and discussions at the National Academy ofSciences Colloquium on Industrial Ecology on May 20-21, 1991, is structured around 10 questions. Do sociotechnical systems have long-range environmental goals? How is the concept of industrial ecology useful and timely? What are environmental technologies? Is there a systematic way to choose among alternatives for improving the ecology of technologies? What are ways to measure performance with respect to industrial ecology? What are the sources and rates of innovation in environmental technologies? How is the market economy performing with respect to industrial ecology? What will be the effect of the ecological modernization of the developed nations of the North on the developing countries of the South? How can creative interaction on environmental issues be fostered among diverse social groups? How must research and education change? Frosch (1) has defined industrial ecology as the network of all industrial processes as they may interact with each other and live off each other, not only in the economic sense but also in the sense of direct use of each other's material and energy wastes. As the field has rapidly taken form over the last few years (2, 3), we can ask what are the provocative and fundamental questions that should frame its progress over the next few. Such questions should reflect the full span of relevant thought and practice, ranging from philosophy of nature and history of technology through science and engineering to economics and management. Drawing on this Colloquium on Industrial Ecology, I believe 10 questions come to the fore. 1. Do sociotechnical systems have long-range environmental goals? 2. How is the concept of industrial ecology useful and timely? 3. What are environmental technologies? 4. Is there a systematic way to choose among alternatives for improving the ecology of technologies? 5. What are ways to measure performance with respect to industrial ecology? 6. What are the sources and rates of innovation in environmental technologies? 7. How is the market economy performing with respect to industrial ecology? 8. What will be the effect of the ecological modernization of the developed nations of the North on the developing countries of the South? 9. How can creative interaction on environmental issues be fostered among diverse social groups? 10. How must research and education change? 1. Do Sociotechnical Systems Have Long-Range Environmental Goals? Societies assert broad goals such as reduction of poverty, universal education and health care, population stabilization, and enhancement of environmental quality. Concrete longterm technical projects, such as the building of highway and water supply systems, are periodically conceived and carried out. But what are the general long-range goals of such systems as agriculture, transport, energy, and production? There was never a goal to become reliant on fossil fuels; this reliance was reached through dynamic optimization of the energy system with respect to transport and storage ofenergy and other factors. Society has directions in which it is driven, but is it driven by intent? It is unclear to what extent sociotechnical systems are directed toward an end or shaped by purposes (4). There may be purpose at the micro level, but the evolutionary track is the summation ofongoing processes whose interactions are not well understood. For example, the transportation system appears to be coded to seek low-cost speed to enable individuals to maximize range. It seems societal development is fundamentally evolutionary and thus to a large degree without purpose though with strict rules of choice at every stage. In some cases these rules of choice have been environmentally favorable. The teleological question has several aspects for industrial ecology. One is the capacity for coordinated, creative design in the economy. On the one hand, there is the recognition of a need for some kind of larger-scale optimization of industry that takes better account of environment. On the other hand, there is the appreciation of myopia in economic systems. Blindness to consequences provides freedom to explore and experiment, and heterogeneity of preferences and expectations is required for evolution. Moreover, how far into the future can social radar look? How far into the future can societies effectively and sensibly plan? The question "To what end?" also forces reflection on the basic question about products and services. What products and services do people need according to various criteria? Would it be preferable to have a particular system or get along without it? This question is hard for both private and public enterprises to ask. Most organizations want to sell more ofany particular product they make and have more products. The issue of long-term public good is rarely asked in fundamental ways in technological or environmental impact assessment. The issue may be interpreted as the traditional one in living systems of the difference between growth and development. Growth implies an increase in size that is a quantitative phenomenon, while development implies additionally a real879 The publication costs of this article were defrayed in part by page charge payment. This article must therefore be hereby marked "advertisement" in accordance with 18 U.S.C. §1734 solely to indicate this fact. Proc. Natl. Acad. Sci. USA 89 (1992) ization and enhancement of potential. It is a qualitative phenomenon. There is a widespread sense that the ad libitum feeding ofthe industrial economies cannot be sustained except at great environmental cost (5). The beneficiaries ofeconomic development wonder whether they are becoming obese, complacent, unhealthy, and vulnerable like a bunch of fat laboratory rats. The rise of industrial ecology may indicate the maturity of industrial society to ask questions seriously about not only growth but also development and its consequences. Industrial ecology also implies concern with equity, not only over time but also as the current network of production and consumption distributes goods and bads. Societies are accustomed to discussing and struggling over distribution of material wealth, wealth that industrialization has been astonishingly successful in generating over the last 200 years. The transformations of the economy have also generated new possibilities of loss and injury. They have reduced many hazards and dangers. We now meet mad hatters and wheezing chimney sweeps only in literature. But clearing away old risks allows latent ones to surface, and new ones also arise (6). How is the industrial web now affecting and distributing the risks people face and where may risk reduction be most effective? Who is going to shift what burden to whom? In the end, industrial ecology suggests both some broad goals for sociotechnical systems, such as waste reduction and dematerialization, and revised rules of selection for the technologies, products, and enterprises that should survive. The goal in general terms for technologists should be to offer the possibility of superior efficiency and productivity in serving human needs such as food, clothing, shelter, health, transportation, and communication with reduced environmental impact, reduced consumption of raw materials, and substitution of morefor less-abundant raw materials, as well as inclusion of wastes and products at the end of their lives in the industrial food web as both material and energy (3). 2. How Is the Concept of Industrial Ecology Useful and Timely? Traditionally concerned with landscapes and animals, ecology is the branch of science that considers how organisms are embedded in their environment and how they interact with it. Ecosystems are defined from the inner surface of their environment, or their ecological niche. The key is that the parts are conceived with respect to the whole (itself often poorly understood) and not the other way around. Ecology recognizes connectedness as the condition of existence. At the same time existence itself is constantly evolving. Nature is very much unfinished, quite provisional (7). Of course, ecology is not the only discipline to make claims for a sound model of flows of energy, resources, and information. Its nemesis, economics, does as well. The field of chemical engineering also stresses dynamical systems. The value of industrial ecology will depend on the extent to which it can provide the grounds for synthesis and interrelation of the variables normally incorporated or ignored in each of these and other relevant fields. It is about deepening appreciation of technology and extending what is valued in economics, broadening the domain of what must be considered in engineering design and practice, and ending the isolation of ecology from the man-made world. The concept appears now not only because of the accumulation of problems that already exist but also because of prospective multiplication of needs. The current anxiety can be illustrated with numerous examples. One powerful fact is that in 1991 the United States had about 185 million motor vehicles, whereas in 1970, when the first Clean Air Act was passed, there were only about 100 million. Simply keeping pace requires getting better. Projecting needs into the future, individuals will give their own preferred numbers. A conventional guess is that there will be a doubling of the world population in the next century. To meet the needs of a world of 10 billion people will likely entail at least a 4-fold increase in agriculture, energy use, and industrial production if the majority of people are to have better housing, diet, transport, and other services than today. This multiplication of needs means that we need a science and industry for a small planet. Industrial ecology can contribute both understanding and solutions. If the current ratios of emissions, pollution, and waste creation to production and consumption are maintained, the environmental problems are certainly going to become worse. Fortunately, there is reason for confidence that we can increase and probably double efficiency in many systems over 30 years or so (8). The question is whether the introduction and diffusion of technologies, including social technologies, will proceed in directions and at rates such that there is net improvement or deterioration. The question of the scale of application and use of technologies, products, and services is fundamental to the emergence of anxiety. Many of the most serious impacts relate to scale. Can societies accurately anticipate the scale of application and use of processes and products? Few people, certainly not inventor Thomas Midgely in 1931, would have guessed the extent of the markets for chlorofluorocarbons that would exist 60 years later (9). Even in optimistic moments about the car market Henry Ford probably would have failed to predict the quantity of auto emissions. The same is probably true of fertilizers and chemical pesticides. Alfred Lotka, who understood exponential growth, nevertheless wrote in 1924 that it would take 500 years for the atmospheric carbon dioxide concentration to reach the level it is likely to attain in another 50 (10). Lotka drastically underestimated the rate of expansion of fossil fuel use. Humans continue to be a rapidly growing species, and we are attempting to be aware of our impacts on the other organisms with which we share the planet. New techniques and their interactions will nevertheless have undesirable and unforeseen impacts, even if our "radar," technology, and technology assessment improve. Of course, not everything that is unforeseen in the web of industrial connections will be undesirable. There are numerous wonderful examples from the history of technology where clusters of inventions turn out advantageously. The late Lynn White, Jr., illustrated industrial ecology vividly in an essay on technology assessment from the point of view of a medieval historian (11). White sketched a sequence that started with textiles and ramified unexpectedly. The arrival of the spinning wheel in Europe in the 13th century sped yarn production, lowered the price of cloth, and increased its consumption. Because it was unpatterned, seldom dyed, and bleached only by sunlight, linen was particularly affected. By the 14th century there was an immense increase in linen shirts, underwear, bedding, towels, and headwear. Increased use meant more and cheaper rags, and linen rags were the best material for making paper. The burgeoning paper industry could expand production, lower prices, and seek new markets. Formerly it had taken the skins of between 200 and 300 sheep or calves to produce a Bible. With the advent of cheap paper, the wages of the scribe became by far the greatest cost of manufacturing a book. Thus, it was the wastes created by the spinning wheel that created the conditions for success for Gutenberg's venture in mechanical writing. We live inevitably in such a web of connections. As described by Frosch (1), we live off each other and sometimes off one another's wastes. We must now try more consciously to evolve to make use of waste products. 3. What Are Envirommental Technologies? There appear to be several definitions or categories of environmental technologies. Some remedy; some conserve; 880 Colloquium Paper: Ausubel Proc. Natl. Acad. Sci. USA 89 (1992) 881 some prevent pollution. Some people speak of the gardening or improvement of the planet. Some define environmentally sound technologies by their quality or performance relative to present practices. Examples are as diverse as compact, efficient lighting devices (12) and completely biodegradable plastic materials produced by bacteria from agricultural raw materials rather than petrochemicals (13). There are environmentally friendly technologies that relate to manufacturing and operations, such as just-in-time-inventory and on-demand generation of toxic chemicals to obviate mass storage and transport (14). Stein (15) points out the potential importance of design-for-disassembly, which would facilitate recycling of basic constituents of a product. The Volkswagen Passat can be disassembled for recycling in 20 min. There has yet to be careful thinking about categories or criteria for environmental technologies. A successful taxonomy of environmental technologies ought to clarify opportunities for fast, generic progress. For example, large streams of contaminated water from various processes are a problem that arises repeatedly, as do problems associated with oxidation in regular air. Chemical engineering and other professions ought to be able to make rapid advances in a number of such areas. Design principles could change quickly with regard to use of pure oxygen for oxidation and processes involving excessive formation of salts or use of water. 4. Is There a Systematic Way to Choose Among Alternatives for Improving the Ecology of Technologies? Patel (16) suggests there are six strategic elements in industrial ecology: selection of materials with desired properties at the outset; use of just-in-time materials philosophy; substitution of processes to eliminate toxic feedstocks; modification ofprocesses to contain, remove, and treat toxics in waste streams; engineering of robust and reliable processes; and consideration of end-of-life recyclability. Are there systematic ways to decompose existing designs so that alternative processes to eliminate existing pollution sources can be identified? Boyhan (17) offers a case study of eliminating chlorofluorocarbon use in manufacturing by substitute cleaning agents, by processes that require no cleaning because precise amounts of materials are used, and by other alternatives, such as conductive epoxies, which would obviate the entire soldering process. There is also the prospective question. Are there systematic ways to foresee pollution problems early and simply during development of techniques (18)? The question and responses should be central in the curriculum of industrial ecology. The need is to note all the decisions required to complete a design and their consequences in a context that encourages imaginative generation of alternatives processes. This is more than a requirement of good design software, although many key queries and outcomes could be captured in programs. Ultimately, as Duchin recognizes (19), it would be desirable to have coherent frameworks for examining potential long-term advantages of each web of industrial changes and identifying short-term bottlenecks that may emerge. 5. What Are Ways to Measure Performance with Respect to Industrial Ecology? There are hundreds of familiar indicators of environmental quality and of economic performance (20). There are specific, promising reports of change. For example, according to one inventory of releases of toxic pollutants, major U.S. manufacturers may have reduced their emissions about 20% between 1988 and 1989 (21). However, few such indicators represent effectively the networks of industrial processes and how they are changing (19). All such measures require goals and rules to be meaningful. William Clark suggested at this colloquium that one way to measure performance might be to identify major transitions that would take place in industrial ecology and assess standing in relation to these transitions. Such transitions are familiar reference points in other fields. For example, there is the demographic transition, at which fertility rates begin to fall, and labor force transitions, signaled by declines in agricultural workers and increased participation ofwomen in the labor force. What would be the transitions expected or sought in industrial ecology? The transition from materialization to dematerialization could be one (22). "Dematerialization" is the decline over time in weight of materials or "embedded energy" in industrial products (23). Dematerialization could translate into less waste from both production and consumption. Although statements about dematerialization ofindustrialized societies have been made casually, only a few short series of data are available as evidence. Time series extending back 30 years and more need to be assembled and kept current for a broad sampling of system levels (firms, industries, individuals, municipalities) in different countries. The shift from increasing reliance on carbon fuels to "decarbonization" of the energy system might be another key transition in industrial ecology (Fig. 1). Carbon matters because it is the main element used to spin the industrial web and is also associated with greenhouse warming, smog, oil spills, and deforestation. Decarbonization might be defined in several ways. It could indicate the evolving mix of fossil fuels used. Coal, the most environmentally damaging fossil fuel, is heavy in carbon and would weight the measure; natural gas, which is mostly hydrogen, would lighten it. It could refer to the ratio of carbon used to total energy consumed or to economic activity. Between 1973 and 1986 Canada, the United States, Sweden, and France, in absolute terms large polluters, nonetheless moved on what might be labeled a green trajectory toward high energy efficiency and low carbon intensity. Meanwhile, Mexico and India moved in the reverse direction. Data for this measure for the former U.S.S.R. and China suggest these areas continue to function with a Victorian industrial ecology (Fig. 1A). An alternative formulation of decarbonization includes biomass (fuelwood and hay) in addition to fossil fuels. By this measure all nations are on favorable trajectories, though proceeding at different rates and by different routes (Fig. 1B). Viewed as one system, the globe is decarbonizing steadily (Fig. 1C). Analysts might enjoy fun and profit over the next few years improving these measures and developing others, providing aggregate and disaggregate measures for performance with regard to industrial ecology. The discussion of transitions may be generalized. At what point after materializing or carbonizing does an economy bend around and move back down? Is this an "eco-transition" as discussed by Ayres (26)? Can the arrival of transitions be hastened? It is important to include moral and aesthetic criteria in the evaluation of system performance. We each have a diffuse but deep sense that there exists a world of artifacts, of clay pots and plastic water bottles, and some of these artifacts correspond to what might be called peace with nature, while others seem to violate peace with nature (27). We know how to distinguish between gardens and garbage dumps, although bacteria might like both. Our measures must ultimately relate to concepts of what is right and beautiful, difficult and contentious though this may be. 6. What Are the Sources and Rates of Innovation in Environmental Technologies? In contrast to sectors such as health and national security, there have been few studies of patterns of innovation and barriers to diffusion in the environmental area. Many quesColloquium Paper: Ausubel Proc. Natl. Acad. Sci. USA 89 (1992)