Engineering Within Ecological Constraints (1996)

John Cairns, Jr.

The significant problems we face cannot be solved at the same level of thinking we were at when we created them.

Albert Einstein

Every country can be said to have three forms of wealth: material, cultural and biological. The first two we understand very well, because they are the substance of our everyday lives. Biological wealth is taken much less seriously. This is a serious strategic error, one that will be increasingly regretted as time passes.

E. O. Wilson

Society has become so accustomed to technological services that they are obvious only in their absence. This happened recently in the eastern United States during the ice storms and blizzards of 1994, in the Mississippi River drainage during the severe floods of 1993, in Florida following Hurricane Andrew in 1992, and in Georgia during the flooding of the Flint River in 1994. Technological services generally involve the substitution of fuels for human effort. They include delivering diverse foods independent of season or local climate; delivering potable water directly into homes; heating, cooling, and humidifying or dehumidifying the climate in buildings; delivering communications such as telephone calls and television and radio signals; and removing noxious wastes such as sewage and trash from homes. Civilization, especially in wealthy countries, has developed an extensive infrastructure for the delivery of technological services, including electrical transmission lines, roads, airports, telephone lines, satellites, and sewer systems.

Society also depends on ecosystem services that have existed for much longer, in fact, probably since the appearance of life on Earth. Ecosystem services are those functions of ecosystems that society deems beneficial, including the maintenance of atmospheric gas balance, flood control, carbon storage, cap-

ture of solar energy and subsequent production of food and fiber, maintenance of water quality, and maintenance of a genetic library that provides the raw materials for improved foods, materials, and drugs. While these services have long been taken for granted, society is beginning to realize that many functions of natural systems are not immutable and can be affected by human actions.

The definition of an ecosystem service is a matter of societal perception because it hinges on valuation. Of all the processes or functions carried on by ecosystems, only those contributing to the well-being of human society are considered services. On those rare occasions when the societal value of minimally managed functions of ecosystems is evaluated, different people reach different conclusions. The debate is even more heated when management actions to protect ecosystem services are proposed. What scientific evidence is necessary to facilitate the societal debate on value and help establish a reasonable level of management? Does the delivery of necessary ecosystem services depend on ecosystem health? Not only is additional, well-conceived research needed to clarify these relationships, but it is also crucial to be able to communicate the results of such research and its uncertainties to the wider society that is properly involved in the debate on values. These issues provide recurring themes in the following discussion of balance.

All ecosystem functions could possibly be viewed as ecosystem services and any distinction between the two as a reflection of the limits of human knowledge rather than an actual difference. In addition to the term ecosystem services, the term sustainable use is often used to describe human benefits from ecosystems. When an ecosystem service is being provided at a rate that meets society's demands without compromising future use, there is sustainable use of the ecosystem. However, as Costanza notes (in this volume): "The problem is that one knows one has a sustainable system only after the fact . Thus, what usually passes for definitions of sustainability are actually predictions of what set of conditions will actually lead to a sustainable system." The alternative is an unsustainable use that not only may fail to meet the needs of society but also may damage the ecosystem and impair the rate at which the desirable service is provided.

The least controversial examples of ecosystem services are those for which an economic value is easily derived. These economic values can then be incorporated into existing decision making tools. The rapidly developing field of ecological economics (e.g., Costanza, 1989, 1991, and in this volume) has identified several useful approaches. There are problems with these approaches, but in their absence ecosystem services are too often completely ignored as externalities.

Some cases of free market respect for ecosystem services rather than technological ones can be cited. Natural systems are replacing chemical technology for waste treatment (Hammer, 1989). Natural systems complement energy-using

technologies in local climate modification; for example, planting trees may save 200 billion kilowatt-hours annually in the United States by reducing the need for air conditioning (Committee on Science, Engineering, and Public Policy, 1992). Avise (1994) calculated the cost of Biosphere 2, which partially regulated the life-support systems for eight humans over a two-year period with an electricity subsidy from outside the sphere. The cost was $150 million (U.S.), or $9 million per person per year. The complications of calculating per capita ecosystem services at vast global or continental scales were definitely not present in this mesocosm experiment. Nevertheless, there is little doubt that practices tolerated in Biosphere 1 (the earth), such as human population growth, overexploitation of ecological capital, and massive destruction of habitats and species, could not be tolerated in Biosphere 2 even for a relatively short time (Avise, 1994). Clearly, some essential functions of natural ecosystems would be difficult or expensive to replace through technological systems.

Ecosystem health is a complex concept, but a consensus definition has been derived (Haskell et al., 1992; see also Karr, in this volume):

It seems reasonable that a close correlation should exist between health and performance at any level of biological organization. Healthy plants capture more solar energy. Growing forests store more carbon. Indeed, one definition of stress demands that captured energy be diverted from growth into coping mechanisms (Calow, 1991). However, the relationship between ecosystem health and ecosystem services is not as well defined.

A number of interrelated hypotheses on the relationship between ecosystem health and ecosystem services deserve serious attention (Cairns and Pratt, 1995).

1. A close correspondence exists between ecosystem health and the production of ecosystem services.
2. Deterioration of ecosystem health does not affect the various ecosystem services uniformly.
3. Ecosystem services decline monotonically with declines in ecosystem health.
4. Decline of ecosystem services important to human society can be predicted accurately from measures of ecosystem health.
5. Ecosystem services of importance to human society may vary markedly even for ecosystems in robust health or condition.
6. For ecosystems in robust health with highly variable delivery of ecosys-

6. tem services, these changes can be predicted from climatic, life cycle, or other similar information.
7. Some ecosystem services are important globally and, therefore, must be maintained at a global level, while some ecosystem services are local or regional, so management at those levels could ensure appropriate delivery of services.

It has been suggested that some of these hypotheses are too broad to test, while others are truisms. However, because these ecosystem services play a role in contributing to the life-support system of human society, they are matters of vital importance. To test these broad hypotheses, it may be necessary to approach them, like other big and important issues, obliquely or piecemeal.

Bradshaw (1983) has said, ''The acid test of our understanding is not whether we can take ecosystems to bits on pieces of paper, however scientifically, but whether we can put them together in practice and make them work." Bradshaw's statement is as valid today as it was more than a decade ago when it was written. There is, however, a second test for ecology; namely, whether ecologists can document the services ecosystems provide in sufficiently explicit terms that society will not only protect and preserve those ecosystems still delivering such services but repair to whatever degree possible those ecosystems capable of delivering services at a level far beyond their present capacity.

Quantifying ecosystem services, while a formidable task, may not be as esoteric as it appears. John Harte, of the University of California, Berkeley, has carried out studies at Rocky Mountain Biological Laboratory in Colorado on the effects of global warming. During this research, he found that the soils at the laboratory itself acted as a methane sink. He estimated that, if one extrapolates from that small patch to the entire county, the amount of methane assimilated by the soils approximately equals the amount of methane produced by the cattle in the county. Not only is this information ecologically useful, but it can be easily communicated to the ranchers and other inhabitants of the area. Put in a local context, ecosystem services are no longer regarded as an esoteric issue but as one of considerable interest to area residents.

Some more specific questions related to the documentation of ecosystem services follow.

1. What is the relationship between species richness and delivery of ecosystem services?

Biotic impoverishment, or loss of species, is fairly well documented (e.g., Wilson, 1988), but the relationship between delivery of ecosystem services and the number of species present is almost certainly not linear. There is some evidence of redundancy in function, which means that, if 10 or 12 species were carrying out roughly the same function simultaneously, the loss of one or two would not cause a serious decline in delivery of services because the deficiencies would be made up by the remaining species expanding their numbers.

The decrease in species diversity, and presumably services provided by eco-

systems subject to sustained pressure by humans, appears widespread. Regier and Hartman (1973) provide an interesting account in which valued sport and commercial fish were exterminated in Lake Erie as a result of tremendous fishing pressure. These species were replaced by less valuable ecological equivalents. The substitute species still provide "services," though not of the same commercial value as those potentially contributed by the exterminated native species.

Other evidence indicates that what appears to be redundancy is not redundancy. For example, although two different species of zooplankton may feed on phytoplankton in lakes, one may be much more selective than the other, thus producing different ecological consequences. Selective feeders remove individuals of only certain species from a community. This gives the remaining species, those not chosen, a considerable advantage since their competitors are removed and the resources used by the competitors will almost certainly become available to the remaining species. This produces a marked alteration in community structure, as often occurs in rangeland cattle grazing when the most succulent plant species are removed and the less succulent species (e.g., thorny shrubs) become dominant or at least much more abundant. Nonselective feeding, on the other hand, is more equitable and is more likely to produce similar reductions in the various prey species. Thus, selective feeding is likely to alter the relative abundance of species present, whereas nonselective feeding is likely to have much less effect on the relative abundance of species.

2. How close is a crucial break point, or threshold, in the delivery of services by natural ecosystems?

There is good documentation of the increase in human population as well as the marked increase in technological and industrial activity, particularly energy consumption from fossil fuels. Since human population numbers, levels of affluence, and use of technology have increased dramatically in the past 10,000 years, and ecosystem services still appear adequate to support life (although regionally impaired here and there), one may ask whether delivery of ecosystem services is really an important problem?

3. To what degree are the ecosystem services of natural systems replaced by agroecosystems, managed forests, and even such things as vegetation on golf courses? As Harte (1993) notes:

4. To what degree do natural systems provide global, regional, or local services?

For the maintenance of atmospheric gas balance, one could make a fairly strong case for hemispheric management. For protecting water quality, a regional management for services might be more effective and, for aesthetic purposes, highly local management is called for (see also Norton, in this volume).

5. Can one extrapolate from the services provided by one ecosystem to the services likely to be provided by another?

Of course, one would not expect to extrapolate ecosystem services from the Kalahari Desert to the California redwood forests with any substantive degree of correspondence. However, one might reasonably expect to be able to extrapolate from one East Coast temperate zone wetland to another. The degree of uniqueness of each ecosystem and the degree to which extrapolations can be made with confidence will require a robust database, which can probably be gathered relatively efficiently.

6. What is the relationship between ecological resilience and the delivery of ecosystem services (see Holling, in this volume)?

Ecosystems have, throughout their existence, been exposed to natural perturbations, such as drought, hurricane damage, fires, and floods. Resilience, as defined by Holling (1973), is the ability to regain normative or characteristic structural or functional attributes following a perturbation. Some ecosystems are thought to be perturbation-dependent and others perturbation-independent (e.g., Vogl, 1980). Will full ecosystem services resume more quickly in the former than in the latter?

7. To what extent do exotic invading species disrupt or impair ecosystem services?

As a result of human transportation, animals and plants now have means of dispersal that were not available previously. The new colonists can wipe out or displace native species, and are frequently dispersed alone, without the control measures (predators, parasites) that keep them in check in their original habitat. The rate of human transport of exotics to new habitats will increase markedly as the global marketplace develops. How will these assisted invasions affect delivery of ecosystem services?

Any attempt to balance technological and ecosystem services must be supported by the general public. To be included in the debate, it is essential that the general public have access to the information that ecologists take for granted. It will be necessary to communicate convincingly to a majority of a country's inhabitants that they themselves are dependent on services provided by natural systems. If this dependence is not understood, then the drive to protect wild areas

and ecosystem services will be characterized as an extravagance driven by religious beliefs rather than sensible self-interest. That is why scientific investigations of the linkages between environmental services and human quality of life are essential.

At present, environmental literacy is rudimentary for most members of human society, including college graduates in many disciplines (e.g., Wallace et al., 1993). There is a widespread belief that there are technological solutions to every problem and that stabilizing the size of human population will lead to economic stagnation.

Skinner (1983) argued persuasively that human behavior is selected or determined by its consequences and that substantial numbers of people cannot be expected to change their behavior as the result of information or advice alone, especially when the information is about a distant future. He further stated that people might follow advice when the information from the advice giver has led to beneficial consequences in the past; however, this situation requires that people have experienced reinforcing consequences of prior compliance with similar advice givers or similar rules. Such operant learning is difficult or impossible when reinforcement lies in the future or punishing consequences are unclear, uncertain, or remote. Lack of clarity, uncertainty, and remoteness are all common characteristics in the models scientists must use to make large-scale predictions about ecosystems. A major ecological disaster would provide persuasive evidence of the links between human and environmental health, and failures of some technological solutions would radically change human society's relationship to natural systems.

Rather than wait for severe consequences, what events are most likely to change human behavior with a modest increase in environmental literacy? Professionals in various disciplines must make the added effort to combine their contributions to facilitate policy and decision making. The obvious example of this is ecological economics. Professionals must also make their information and conclusions accessible to the public. Science reporting for the mass media is improving and has room to continue to do so. Another area of concern in environmental literacy is the way in which ecologists, managers, and engineers communicate the uncertainties inherent in their projects.

Although human society is well into the information age in world trade, economics, military strategy, marketing, investment, and a host of other activities, the use of information feedback loops to modify society's relationship to ecosystems has not progressed nearly as far as it has in other areas. A major reason for this curious discrepancy is that the consequences of failure in military, economic, and other sectors of society are more apparent to policymakers than are comparable failures in ecological systems. In contrast, the consequences of

ill-coordinated responses to changes in ecosystems may not be intrusive until much time has elapsed.

In ecological systems, one form of information feedback is coevolution, a term first used by Ehrlich and Raven (1964) in examining the evolution of plants and the insects that feed on them. They defined coevolution as a pair-wise process in which the appearance of a trait in one species elicits a response in another species. For example, increased speed in a carnivorous mammal may result in increased speed in its prey, or a variety of other adaptations to offset the predator's speed, such as improved camouflage. Futuyama and Slotkin (1983) indicated that development of a particular trait in one or more species may result in a suite of traits in several other species. Ghersa et al. (1994) described the coevolution of agroecosystems and weed management. They postulate that weed-management practices have become closely linked to social and economic, rather than biological, factors. As Harlan and deWet (1965) have noted, weeds and weed problems are anthropocentric terms applied to populations of plants when they are considered undesirable.

Before the agricultural revolution, the needs of Homo sapiens were met in ways not dramatically different from the ways in which needs of other species were met. Human populations were comparatively small and kept from the explosive growth of the past century by disease, starvation, and even predation. These are, in fact, much the same population control factors affecting other species. Throughout history, ecosystems provided all the services necessary for the continuation of the human species: breathable air, potable water, food, and a consistent climate over the short-term. Now human society has a codependence on both a technological life-support system and an ecological life-support system. However, the maintenance needs of technological services get much more attention than those of ecosystem services. In a very real sense, natural systems and human systems are coevolving since only those opportunistic and communal species tolerant of the anthropogenic alterations of natural systems are likely to thrive. These may be opportunistic species resistant to pesticides and habitat fragmentation and tolerant of a wide range of ecological conditions—in short, pests.

Those species that are tolerant of anthropogenically changed conditions might provide some of the services that more complex natural ecosystems provided previously. However, it seems unlikely that they will perform in precisely the same ways. Kauffman (1993) proposes a bold hypothesis: complex, adaptive systems operate on the edge of chaos. He feels that not only organisms, but economic entities and nations, do not simply evolve, but rather coevolve, and that coevolving complex systems mutually operate at the edge of chaos. If the adapting system is itself in the ordered (rather than the chaotic or boundary) regime, Kauffman (1991, 1993) believes the system itself adapts on a smooth landscape (see Holling, in this volume). In the chaotic regime, the system adapts on a very rugged landscape, and, of course, in the boundary regime, it is intermediate.

Kauffman uses the word landscape in association with the word fitness,¹ a term commonly used in ecology to mean the degree of adaptation to the habitat or ecological niche of an individual species. I suspect it can also describe the goodness of fit of two complex, multivariate systems to each other. Kauffman uses chaos in a technical sense, whereas in this discussion it is used in a more colloquial sense. However, it seemed important to begin this discussion with Kauffman's publications.

With Kauffman's analysis in mind, this author draws three conclusions about the balance between technological and ecosystem services: (1) abrupt changes are not prudent in the coevolution of complex, multivariate systems, (2) in two coevolving systems with substantive interdependence, chaos in one is likely to result in chaos in the other, and (3) coevolution can drain energy and deplete resources of the coevolving partners, as in the case of predator and prey. This is also true for nations coevolving in a climate of hostility, as was the case for the USSR and the United States during the Cold War. Finally, coevolution is likely to proceed most smoothly if at least one of the parties recognizes the dynamics of the situation.

In the coevolution of a predator species and its prey (the prey evolving toward more elusiveness and the predator increasing its foraging efficiency), one might make the case that the relationship is beneficial to both species because the capabilities of each species are thereby improved. However, less benign forms of coevolution are also evident. Application of pesticides leads to pesticide resistance in the target organisms, thereby requiring more and more pesticides to achieve the same result and ultimately increasing the risks to nontarget species, including humans. Thus, the application of pesticides, if not done skillfully, can pose a serious threat to human health, a situation well documented in the literature. As a consequence, one might reasonably ask if the coevolution of human society and natural systems is mutually beneficial or mutually destructive. Since the beginning of the agricultural revolution, society has attempted to alter natural systems so that more and more of the energy captured by photosynthesis is converted to foodstuffs and other products of interest to human society (Vitousek et al., 1986). Not only has there been a substantial loss of space devoted to unmanaged production of diverse ecosystem services as a result of agricultural activities, but relatively natural ecosystems, particularly those adjacent to agroecosystems, have often been affected by runoff and airborne contaminants such as pesticides and dust, fragmentation, and, finally, changes in microclimate. It seems possible that this could represent hostile coevolution. Erwin (1991) states, "Within a few hundred years this planet will have little more than lineages of domestic weeds, flies, cockroaches, and starlings, evolving to fill a converted and mostly decertified environment left in the wake of nonenvironmentally adaptive human cultural evolution." The pesticide tolerance and increased pesticide use described earlier are one such example. In hostile coevolution, the deleterious effects of human society on natural ecosystems will select for those organisms

and communities of organisms most resistant to this stress, with an equal selection for pioneering species capable of taking advantage of chaos. More than 100 years ago, Kew (1893) provided an eerie insight into the kinds of species that would coevolve with human society if hostile coevolution continues. He described the durability and colonizing potential of Dreissena (the zebra mussel), which relatively recently became a major problem in North America (Ludyanskiy et al., 1993), as follows:

Thus, even if ecological losses or problems can be predicted, it seems as though society is often willing to trade these problems for jobs, particularly in uncertain economic times (i.e., during recessions). Pratt and O'Connor (1994) describe a situation at Gettysburg Historic Park, where certain conditions that existed at the time of Pickett's charge across a cornfield were to be maintained for historical reasons. The cornfield was located between two lines of trees. During the Civil War, only a few deer inhabited the area. Deer were controlled by hunters, who could harvest a substantial percentage of the population. At that time, deer were not protected either by legislation or by being adjacent to areas where hunting would be highly objectionable. For many years, farmers were willing to pay a modest fee to grow corn in this historic area because they could harvest enough to make a profit. Eventually, the deer herd expanded to a size that made harvesting corn no longer profitable. In addition, harvesting the deer at the necessary rate was objectionable to components of society for a variety of reasons. Replacing the cornfield with astroturf or some other nonhistoric condition was also unacceptable. Under the best circumstances, a substantial expenditure would be necessary to preserve the historic condition of the area by excluding the deer. Consequently, the coevolution of the deer herd and human society became more and more expensive with no socially acceptable alternative in view. In this example, the natural control measures regulating the deer population were removed and resulted in densities unlikely to have been achieved previously in natural systems. All this occurred because humans provided an extraordinary food base and freedom from predation for the deer.

Chaos at the interface can be manifested in a variety of ways. A contrasting illustration for an aquatic system is the invasion of North America by the Asiatic clam, the zebra mussel, and the quagga mussel (Russian mollusc, D. bugensis). These invaders from different parts of the world might well have taken hold in pristine systems in North America had they been able to get there. However, without the help of human society's transportation system, this would have been

unlikely. Not only can these molluscs invade areas occupied by North American species, but they can also occupy habitats created by an industrial society, such as power plant cooling systems, irrigation canals, waste treatment ponds, and, in the case of the zebra mussel, surfaces such as ship hulls (e.g., Cherry et al., 1986; Garey et al., 1980; Hayward et al., 1982; Nalepa and Schloesser, 1993).

In addition to the problems caused in the technological portion of human society, there is also an increased risk to natural systems resulting from the marginally effective control measures designed to minimize the impacts of these species on the industrial system, including agro-industry. These invaders of North America have created chaos in both the technological and the ecological components of human society's life-support systems.

Thus, while there is ample cause for concern about biotic impoverishment (e.g., Wilson, 1988), there seems to be little concern about the ultimate consequences of this coevolutionary process. If the assumption is that humans are incapable of driving all species to extinction, those left will be highly adapted to exploit the new environmental conditions resulting from overemphasis on the maintenance of technological services. Pest species will be difficult to eradicate—those capable of invading habitats unsuitable for most other species; those selected for resistance to pesticides and other control measures; and, in many instances, those so intimately associated with human society (such as the Norway rat, the housefly, and the cockroach) that control measures may also be a risk to human health. However, failure to exercise the control measures may pose an equal danger to human health as a result of spread of disease.

Nonsuch Island provides excellent examples of the ability of exotic species to colonize, as do the Hawaiian Islands and many other island ecosystems. On Nonsuch, Wingate (1990) reported the necessity of a major and continuous effort to eliminate invading species. Plant invaders are brought to Nonsuch from other Bermuda islands by birds that feed there and then defecate the seeds and other propogules while roosting at night. Even reestablishing the species that existed on the islet before heavy colonization by Europeans would not enable that community to resist invasion by exotics if the sources of colonizing species were in either the conterminous or contiguous ecological landscape in which Nonsuch exists.

Although experimental evidence is not robust, one might still reasonably conclude that, once human society has created chaos in a natural system by destabilizing a complex system, the new system, whatever ecologists may think of it, is now a coevolutionary partner with human society. This plausible analysis provides grounds for persuading human society to consider the coevolutionary consequences of the impact of its technological life-support system on its natural life-support system.

The situation described as hostile coevolution is a form of "social trap." The social trap in this case is the feeling that any major change in human behavior will result in such chaos in the economic system that the natural systems will just have

to take it. Besides, the reasoning goes, there is no "scientific proof" of any failure in ecosystem services, and most predictions of the consequences of loss of biodiversity, global warming, ozone holes in the atmosphere, and the like seem much less threatening than job loss, reduction in gross national product, or loss of present amenities resulting from high per capita energy consumption and lowered product costs because environmental externalities are not included in economic analyses. Substantial literature exists on social traps, but some illustrative materials are Brockner and Rubin (1985), Cross and Guyer (1980), Platt (1973), Teger (1980), and, particularly relevant to this discussion, Costanza (1987, and in this volume) and Geller (1994).

In developing a guiding model for achieving a balance between technological and ecosystem services, it is helpful to see what humans did when the technology was primitive and human populations comparatively small. At least some preindustrial tribes revered their local environment, but Diamond (1992, 1994) has suggested that the relationships between primitive peoples and their natural environments were not always sustainable. Probably the best known, much earlier exponent of this outlook was Rousseau (1754), whose discourse on the origin of inequality traced humanity's degeneration from the golden age to the misery that now exists in all too much of the world.

Where peoples living in harmony with nature persist, they have been driven to areas that no one else wants. This is true of the Bushmen in the Kalahari Desert and some of the few remaining indigenous societies in the Amazon basin and parts of Australia. Since the agricultural revolution some 10,000 years ago, the agrarians have always outnumbered the hunters and gatherers and, even if they were less skillful in warfare, could overwhelm them by sheer numbers. Some of the literature that is particularly interesting in this regard is Hughes (1975), Kirch (1984), Martin (1973), Martin and Klein (1984), Mosimann and Martin (1975), Samuels and Betancourt (1982), and Yoffee and Cowgill (1988). Despite uncertainties and controversy about the literature from which these selections are taken, it is clear that even the primitive technologies of human society when used with teamwork and intelligence could cause considerable environmental havoc.

Ornstein and Ehrlich (1989) speculate that there could be no selective pressures likely to result in a genetic trait prohibiting excessive depletion of resources or conditioning an individual toward long-term sustainable use of a resource base. The rewards for exploitative behaviors are too immediate; the consequences too delayed. Even if the level of technology used by the hunters and gatherers was acceptable to human society (which is no longer the case), population densities alone preclude using such a model, at least for many generations. Of course, many species are left, so, as Diamond notes, the golden age may not have been as golden as it was thought to be, but neither was it all black. Human

population growth cannot continue at present rates indefinitely without temporary and possibly irreparable damage to ecosystem services.

However, as a caveat, human society is probably unlikely to eliminate all the species on the planet, especially since some of them are capable of living in thermal vents on ocean floors and in other nearly equally inhospitable environments. It is also important to remember that much ecological damage has been done and is being done by people who are suddenly thrust into a new ecosystem and transpose methods suitable for an entirely different ecosystem to the new one. In this regard, much may be learned from examining society's response both to technological advances, such as the agricultural revolution, and to the displacement of large groups of people, such as pioneers, into new ecological systems and habitats.

Agriculture appears to have had its origins approximately 10,000 years ago in the Near East (Quinn, 1992). Probably the most positive feature of this revolution is that many times more people could be supported per unit area under an agricultural than a hunter/gatherer system. This concentration of people accelerated the spread of diseases but also made possible the division of labor, which could lead to either art or war. The agricultural revolution made it possible for a relatively few people to feed much larger numbers, and this in turn made the industrial revolution possible.

Multistory buildings, elevators, fossil fuel transportation, and the like made it possible for enormous numbers of people to live in a relatively small area— people who, for the most part, have infrequent interactions with natural systems and therefore are unfamiliar with how they work. This is particularly true of policymakers, elected officials, captains of industry, and others who may have high technological literacy but relatively low environmental literacy. Even people in agribusiness, mining, and ranching, though far removed from cities, may have only the most superficial interactions with systems that have substantial biological integrity. The people most familiar with natural systems are not usually in positions of power and are probably not frequently consulted by those who are. Balancing technological and natural life-support systems will be exceedingly difficult if the operational prerequisites of the natural systems are not well understood and the prerequisites of the technological system are.

Unless the global population stabilizes, it seems unlikely that a balance between technological and ecosystem services can be achieved. The United Nations report Children and the Environment quoted in the San Francisco Chronicle July 20, 1990, noted that fewer than 1.5 billion people, less than half the number alive in 1968, have yet achieved the standard of living that most Americans (or citizens of other rich nations) would find acceptable. In short, the present technological and natural life-support systems of the planet are not func-

tioning in a way that most inhabitants of developed countries would consider ''necessary.'' Yet, the planet has a net gain of 95 million people annually. It is difficult to be optimistic about either stabilizing or reducing the human population when both men and women in many countries want a number of children far beyond the replacement rate. For example, in Niger, women still want 8.5 children, although men want 12.6 on the average (e.g., Sachs, 1994).

If the human population is stabilized or approaching stabilization, it would be worthwhile to attempt to achieve concomitantly a no-net-loss of ecosystem services locally, regionally, and globally. This would mean a balance between destruction and repair of ecosystems with the proviso that destruction is usually accomplished much more rapidly than repair, although in some cases both can be incremental. A stable human population coupled with a no-net-loss of environmental services would mean that, if there were equitable distribution of both, the ecosystem services per capita would not diminish from the time that stabilization and no-net-loss were simultaneously achieved and maintained. Ehrlich and Ehrlich (1991) have stated that the present population density and level of affluence are possible only because society is expending ecological capital, such as topsoil, old-growth forests, fossil water, and ocean fisheries, at rates that are not sustainable (i.e., far greater than the rates of replenishment) for many more years, let alone hundreds of years or millennia. Achieving a long-term, sustainable-use balance between technological and natural life-support systems will require replenishing ecological capital at the rate it is being used (Daly, 1990). One problem with this approach is that the rates of replenishment of soils and aquifers are difficult to calculate on a short-term basis, and, for some things such as old-growth forests, it is not clear whether even millennia will be adequate in some locations. Nevertheless, achieving a balance between replenishment and depletion of ecological capital will be an important consideration in achieving the balance discussed here.

The transition from expenditure of ecological capital to living entirely on ecological "interest" (i.e., services) might be improved by restoring a substantial number of damaged ecosystems, particularly those providing highly desirable services. It is worth emphasizing that these services can include recreational values or aesthetic values. The National Research Council (1992) provides illustrative targets for restoring aquatic ecosystems. This debate should focus on the types of ecosystems, the area or size of the restoration effort, the time frame within which the effort should be made (conditioned, of course, by the fact that biological systems may be capable of reaching a new equilibrium only after substantial time periods), and what should be done nationally, regionally, and locally. These are but a few of the many topics that should be discussed. Initially, restoration need not affect prime agricultural or urban sites. There are numerous abandoned or derelict damaged ecosystems, such as mined sites or floodplain farmlands (150 million acres in the United States), that could easily be restored without interfering with the technological life-support system, and these

should, of course, be restored first. However, inevitably, the time will come when society must consider which should have the highest priority—the technological or the ecological life-support system. The technological life-support system is getting almost exclusive attention, and the ecological life-support system practically none for reasons already discussed. However, assuming that the damage to the ecological life-support system cannot continue indefinitely, a time will come when trade-offs must be made between the two systems.

Until there is more robust information on the services provided by the ecological component of society's life-support system, several steps can be taken to keep options open.

1. A diverse array of habitats should be maintained in as nearly pristine form as possible so that their ecosystem services can be measured. These will provide models for reconstructing damaged ecosystems and also furnish information on the impairment of ecosystem services, if any, when these habitats are used for a multiplicity of purposes as opposed to leaving them in their wild state. These pristine habitats also serve educational roles.
2. The biotic impoverishment, or loss of species, that is occurring globally at a disturbing rate should be substantially decreased as a matter of prudence until more is known about the relationship between biodiversity and ecosystem services. Of particular concern are migratory species, such as birds, which may provide a variety of ecosystem services, many of which are not yet recognized. The crucial issue for migratory species is the fact that a loss of habitat anywhere in their migratory cycle could result in their extinction or cause a dramatic reduction in population size and, thus, affect ecosystem services at points distant from the area of lost habitat.
3. If reaching the maximum possible number of humans on the planet is a societal goal, there should be at least some discussion of whether this goal is most likely to be achieved over a long period through sustainable use or through the depletion of ecological capital as is now being done. If achieving the maximum possible number of humans is not a goal, some discussion of the desirable population size and the human condition permitted by that size is necessary.
4. Enlightened discussion of the issues raised here will require a level of both environmental and technological literacy far beyond that now acquired by most graduates of educational institutions. Wholesale changes in beliefs may also be required to break the "jobs first, then environmental protection" mindset. These changes are likely to occur only following enhanced education.
5. As society moves toward a global economy, it is essential to move toward global consensus on society's relationship to the ecological portion of the life-support system.
6. Financial and other incentives should be devised to ensure that the appropriate data are gathered and that a group of professionals competent to make these measurements and judgments is produced by the educational system.

I am indebted to Bruce Wallace, B. R. Niederlehner, and Eric Smith for commenting on a rough first draft of this manuscript and to John Harte and William Calder for useful comments and information for the third draft. I am equally indebted to Teresa Moody for transcribing the dictation of the manuscript and for making the many adjustments necessary in subsequent drafts, and to Darla Donald for editorial work so necessary in ensuring that the manuscript meet the requirements for publication.

Arise, J.C. 1994. The real message from Biosphere 2. Conservation Biology 8(2):327-329.

Bradshaw, A. D. 1983. The reconstruction of ecosystems. Journal of Applied Ecology 20:1-17.

Brockner, J., and J. Z. Rubin. 1985. Entrapment in Escalating Conflicts: A Social Psychological Analysis. New York: Springer-Verlag.

Cairns, J., Jr., and J. R. Pratt. 1995. The relationship between ecosystem health and delivery of ecosystem services. In Evaluating and Monitoring the Health of Large-Scale Ecosystems, D. J. Rapport, ed. Germany: Springer-Verlag.

Calow, P. 1991. Physiological costs of combating chemical toxicants: Ecological implications. Comparative Biochemistry and Physiology 100C:3-6.

Cherry, D. S., R. L. Roy, R. A. Lechleitner, P. A. Dunhardt, G. T. Peters, and J. Cairns, Jr. 1986. Corbicula fouling and control measures at the Celco Plant, Virginia. Pp. 69-81 in Proceedings of the 2nd International Corbicula Symposium, Special Edition No. 2 of the American Malacological Bulletin.

Committee on Science, Engineering, and Public Policy. 1992. Policy Implications of Greenhouse Warming. Washington, D.C.: National Academy Press.

Costanza, R. 1987. Social traps and environmental policy. BioScience 37:407-412.

Costanza, R. 1989. What is ecological economics? Ecological Economics 1:1-7.

Costanza, R. 1991. Ecological Economics: The Science and Management of Sustainability. New York: Columbia University Press.

Cross, J. G., and M. J. Guyer. 1980. Social Traps. Ann Arbor, Mich.: University of Michigan Press.

Daly, H. E. 1990. Toward some operational principles of sustainable development. Ecological Economics 2:1-6.

Diamond, J. 1992. The Third Chimpanzee. New York: Harper Collins.

Diamond, J. 1994. Ecological collapses of ancient civilizations: The golden age that never was. Bulletin of the American Academy of Arts and Sciences 47(95):37-59.

Ehrlich, P. R., and A. H. Ehrlich. 1991. Healing the Planet. Menlo Park, Calif: Addison-Wesley.

Ehrlich, P. R., and P. H. Raven. 1964. Butterflies and plants: A study in coevolution. Evolution 18:596-608.

Erwin, T. L. 1991. An evolutionary basis for conservation strategies. Science 253:750-752.

Futuyama, D. J., and M. Slotkin, eds. 1983. Coevolution. Sunderland, Mass.: Sinauer Publishers.

Garey, J. F., R. M. Jordan, A. H. Aitken, D. T. Burton, and R. H. Gray, eds. 1980. Condenser Biofouling Control. Ann Arbor, Mich.: Ann Arbor Science Publishers.

Geller, E. S. 1994, The human element in integrated environmental management. Pp. 5-26 in Implementing Integrated Environmental Management, J. Cairns, Jr., T. V. Crawford, and H. Salwassser, eds. Blacksburg, Va.: Virginia Polytechnic Institute and State University.

Ghersa, C. M., M. L. Roush, S. R. Radosevich, and S. M. Cordray. 1994. Coevolution of agroecosystems and weed management. BioScience 44(2):85-94.

Hammer, D. A., ed. 1989. Constructed Wetlands for Wastewater Treatment: Municipal, Industrial, and Agricultural. Chelsea, Mich.: Lewis Publishers.

Harlan, J. R., and J. M. J. deWet. 1965. Some thoughts about weeds. Economic Botany 19:16-24.

Harte, J. 1993. The Green Fuse. Berkeley, Calif.: University of California Press.

Haskell, B. D., B. G. Norton, and R. Costanza. 1992. What is ecosystem health and why should we worry about it? Pp. 3-19 in Ecosystem Health, New Goals for Environmental Management, R. Costanza and B. G. Norton, and B. D. Haskell, eds. Washington, D.C.: Island Press.

Hayward, S. J., V. H. Gomez, and W. Sterrer. 1982. Bermuda's Delicate Balance. Hamilton: Bermuda National Trust.

Holling, C. S. 1973. Resilience and stability of ecological systems. Annual Review of Ecology and Systematics 4:1-24.

Hughes, J. 1975. Ecology in Ancient Civilizations. Albuquerque: University of New Mexico Press.

Kauffman, S. A. 1991. Origins of Order: Self Organization and Selection in Evolution. New York: Oxford University Press.

Kauffman, S. A. 1993. The sciences of complexity and "origins of order." Annals of Earth 11(3):19-26.

Kew, H. W. 1893. The Dispersal of Shells. An Inquiry into the Means of Dispersal Possessed by Fresh-Water and Land Mollusca. London: Kegan Paul, Trench, Trubner and Co., Ltd.

Kirch, P. V. 1984. The Evolution of Polynesian Chiefdoms. Cambridge, England: Cambridge University Press.

Ludyanskiy, M. L., D. McDonald, and D. MacNeill. 1993. Impact of the zebra mussel, a bivalve invader. BioScience 43:533-544.

Martin, P. 1973. The discovery of America. Science 179:969-974.

Martin, P., and R. Klein. 1984. Quaternary Extinctions. Tucson: University of Arizona Press.

Mosimann, J. E., and P. Martin. 1975. Simulating overkill by Paleoindians. American Scientist 63:304-313.

Nalepa, T. F., and D. W. Schloesser. 1993. Zebra Mussels: Biology, Impacts, and Control. Ann Arbor, Mich.: Lewis Publishers.

National Research Council. 1992. Restoring Aquatic Ecosystems: Science, Technology, and Public Policy. Washington, D.C.: National Academy Press.

Ornstein, R., and P. Ehrlich. 1989. New World New Mind. New York: Simon and Schuster.

Platt, J. 1973. Social traps. American Psychology 28:642-651.

Pratt, J. R., and R. E. O'Connor. 1994. Public Involvement in Scoping for Environmental Impact Assessment: A Case Study at Gettysburg National Military Park and Eisenhower National Historic Site. Technical Report NPS/MARO/NRTR-94 for the National Park Service. Environmental Resources Research Institute, Pennsylvania State University, University Park, PA.

Quinn, D. 1992. Ishmael: A Novel. New York: Bantam/Turner Books.

Regier, H., and W. Hartman. 1973. Lake Erie's fish community: 150 years of cultural stresses. Science 180:1248-1255.

Rousseau, J.-J. 1754. Discours sur l'origine et les fondements de l'inegalite parmi les homme. 1983. Discourse upon the Origin and Foundation of the Inequality Among Mankind. Translated and edited by D. A. Cross. Indianapolis, Ind.: Hackett Publishing Co.

Sachs, A. 1994. Men, sex, and parenthood in an overpopulating world. World Watch 7(2):12-19.

Samuels, M. L., and J L. Betancourt. 1982. Modeling the long-term effects of fuelwood harvests on pinyon-juniper woodlands. Environmental Management 6:505-515.

Skinner, B. F. 1983. A Matter of Consequences, Part Three of an Autobiography. New York: Knopf.

Teger, A. I. 1980. Too Much Invested to Quit. New York: Pergamon Press.

Vitousek, P. M., P. R. Ehrlich, A. H. Ehrlich, and P. A. Matson. 1986. Human appropriation of the products of photosynthesis. BioScience 36:368-373.

Vogl, R. J. 1980. The ecological factors that produce perturbation-dependent ecosystems. Pp. 63-94 in The Recovery Process in Damaged Ecosystems, J. Cairns, Jr., ed. Ann Arbor, Mich.: Ann Arbor Science Publishers, Inc.

Wallace, B., J. Cairns, Jr., and P. A. Distler. 1993. Environmental Literacy and Beyond, President's Symposium Volume V. Blacksburg: Virginia Polytechnic Institute and State University.

Wilson, E. O. 1988. Biodiversity. Washington, D.C.: National Academy Press.

Wingate, D. 1990. The restoration of Nonsuch Island as a living museum of Bermuda's precolonial terrestrial biome. Pp. 133-150 in The Earth in Transition, G. M. Woodwell, ed. New York: Cambridge University Press.

Yoffee, N., and G. L. Cowgill, eds. 1988. The Collapse of Ancient States and Civilizations. Tucson: University of Arizona Press.