Using the Columbia River Basin in the Pacific Northwest as a case study, Kai Lee describes the concept and practice of "adaptive management," as he examines the successes and failures of past and present management experiences. Throughout the book, the author delves deeply into the theoretical framework behind the real-world experience, exploring how theories of science, politics, and cognitive psychology can be integrated into environmental management plans to increase their effectiveness.

Taking Measures

... I never ceased to ponder [in the early 1950s] ... the obvious deterioration in the quality both of American life itself and of the natural environment.... Allowed to proceed unchecked, they spelled—it was plain—only failure and disaster. But what of the conceivable correctives? ... Would they not involve hardships and sacrifices most unlikely to be acceptable to any democratic electorate? Would they not come into the sharpest sort of conflict with commercial interests? Would their implementation not require governmental power which, as of the middle of the twentieth century, simply did not exist, and which no one as yet—least of all either of the great political parties—had the faintest intention of creating?

—George F. Kennan, Memoirs

We've been in bad shape ever since Columbus landed.... But that's O.K. You can't go back. We must live in this modern world and do what we can to keep it livable.

—Billy Frank, Jr., chairman, Northwest Indian Fisheries

Human activity disrupts environmental stability on a planetary scale. Spectacular instances—the erosion of the ozone layer, the creation of long-lived toxic and radioactive wastes—receive increasing public attention. But it is the mundane momentum that is impressive in the long run. Humans already appropriate 40 percent of net primary productivity on land; two of every five beams of sunlight captured by living things are already in the service of our species. Our demand for energy, mostly from nonrenewable fossil sources, equals 2 tons of coal per person per year; each human accounts for more than 300 pounds of steel annually. The world population is expected to double over the next century, with most of the increase occurring in the developing countries. In 1989 a United Nations panel on sustainable development estimated that economic output must rise between five and ten times to keep up with minimal aspirations for betterment. What remains unanswered is how we might double our numbers and quintuple our economic activity without impairing the longterm ability of the natural environment to feed, clothe, house, and inspire our species.

Social Learning

Today, humans do not know how to achieve an environmentally sustainable economy. If we are to learn how, we shall need two complementary sorts of education. First, we need to understand far better the relationship between humans and nature. The strategy I discuss in this book is adaptive management—treating economic uses of nature as experiments, so that we may learn efficiently from experience. Second, we need to grasp far more wisely the relationships among people. One name for such a learning process is politics; another is conflict. We need institutions that can sustain civilization now and in the future. Building them requires conflict, because the fundamental interests of industrial society are under challenge. But conflict must be limited because unbounded strife will destroy the material foundations of those interests, leaving all in poverty. Bounded conflict is politics.

This combination of adaptive management and political change is social learning. Social learning explores the human niche in the natural world as rapidly as knowledge can be gained, on terms that are governable though not always orderly. It expands our awareness of effects across scales of space, time, and function. For example, we pump crude oil from deep within the earth and ship it across oceans; we burn in a minute gasoline that took millennia to form; with petroleum and its end products we foul water, soil, and air, overloading their biological capacity. Human action affects the natural world in ways we do not sense, expect, or control. Learning how to do all three lies at the center of a sustainable economy.

The Compass

Adaptive management is an approach to natural resource policy that embodies a simple imperative: policies are experiments; learn from them. In order to live we use the resources of the world, but we do not understand nature well enough to know how to live harmoniously within environmental limits. Adaptive management takes that uncertainty seriously, treating human interventions in natural systems as experimental probes. Its practitioners take special care with information. First, they are explicit about what they expect, so that they can design methods and apparatus to make measurements. Second, they collect and analyze information so that expectations can be compared with actuality. Finally, they transform comparison into learning—they correct errors, improve their imperfect understanding, and change action and plans. Linking science and human purpose, adaptive management serves as a compass for us to use in searching for a sustainable future.

To see how adaptive management differs from the trial and error by which humans now learn, consider what happens when a tract in the rain forest is logged. Cutting and removing trees tests beliefs about soil erosion, what plants will grow in the cleared space, pollution of the streams that drain the land, and other aspects of that ecosystem's response to logging. If those beliefs are correct, lumber or cleared land can be obtained without permanent damage to the ecosystem's ability to support life, and understanding is affirmed. Unforeseen results, however, usually bring only loss, because people are seldom prepared to infer lessons that are both clear and capable of being checked against others' experience. In contrast, adaptive managers make measurements so that action yields knowledge—even when what occurs is different from what was predicted. Properly employed, this experimental approach produces reliable knowledge from experience instead of the slow, random cumulation gleaned from unexamined error. When reliable learning prevails, a wide range of outcomes is valuable, and unexpected results produce understanding as well as surprise.

Adaptive management plans for unanticipated outcomes by collecting information. Usually, the greater the surprise, the more valuable the information gained. But the costs of information often seem too high to those who do not foresee such surprises. Framing an appropriate balance between predictable cost and uncertain value is a principal task of the chapters ahead.

The Gyroscope

The environment is necessarily shared by us all; but as every littered park reminds us, what is shared by many is typically abused. Reconciling control with the diversity and freedom essential to a democratic society is the task of bounded conflict.

The conflicts that already pervade environmental policy are likely to increase as nations encounter the domestic and international stresses of moving toward sustainable resource use. Conflict is necessary to detect error and to force corrections. But unbounded conflict destroys the long-term cooperation that is essential to sustainability. Finding a workable degree of bounded conflict is possible only in societies open enough to have political competition. In the United States, environmental politics draws its energy from citizen groups, grass-roots nongovernmental organizations (NGOs) that have skillfully engaged governments, businesses, and individuals. American NGOs have in turn influenced environmental activism all over the world, while Europe's parliamentary democracies have been a center of innovation for "green" political parties, a complementary means of articulating, organizing, and empowering environmental concerns.

Political competition is a messy process. Winston Churchill called democracy the worst form of government save for all others. Yet the existence of more or less open competition in political systems is, paradoxically, what bounds conflict in them. Political competition can persist only where there are rules, both unwritten and written. Chief among them is a shared commitment to address important issues through continual debate. In tyrannies, losers are not just down but out, excluded from further decision making by winners who need respect no limits. Like a spinning gyroscope, competition is motion that can stabilize.

People seek individual freedom (and the political competition it fosters) as a fundamental and universal human right. If their aspirations are fulfilled everywhere, the capacity of bounded conflict to correct error may turn unexpectedly into the salvation of our species. Like a gyroscope deep within a vessel, pointing a true course because it is independent of crosswinds, conflict bounded by legitimate restraint may yet provide a lodestar for all.

At the global level the search for an environmentally sustainable economy must involve many nations that still lack constitutional democracy, at least for now. As a result, negotiation must often involve limited objectives, without aiming to resolve all the differences that divide nations. Finding ways to bound conflict in these circumstances is another major task of the chapters ahead.

Both adaptive management and bounded conflict are essential for social learning to occur. Adaptive management—the compass—is an idealistic application of science to policy that can produce reliable knowledge from unavoidable errors. Bounded conflict—the gyroscope—is a pragmatic application of politics that protects the adaptive process by disciplining the discord of unavoided error. Together they can bring about learning over the decades-long times needed to move from the current condition of unsustainability toward a durable social order.

Large Ecosystems

Social learning is most urgently needed in large ecosystems: territories with a measure of ecological integrity that are divided among two or more governing jurisdictions. Large ecosystems present some of the most difficult problems of environmental science and policy. They are complex, often badly damaged, riven by deep-rooted rivalries among several jurisdictions, and essential to the well-being of large populations. Nothing larger than an aquarium or smaller than the planet forms a closed biological system, although some units, such as river basins, are more nearly integral than others. Large ecosystems provide opportunities for learning from and about the real world. Their governance presents challenges of science, management, and politics, often entangled in ways that resist simple approaches. But without some degree of simplification there can be no learning and no transfer from one case to others. Rather than pursuing conceptual neatness, we should study how human institutions deal with the interdependence created when human boundaries cut across ecological continuities. In the case of large ecosystems, pragmatism is a prime virtue: to learn what we can, and to recognize its limits.

What makes an ecosystem "large" is not acreage but interdependent use; the large ecosystem is socially constructed. Rivers nurture fish and plants, water fields and cities, provide transport for trade and sometimes hydroelectricity for industry. Multiple use of a river or other large ecosystem requires trading off qualities that are hard to compare, controlled by or benefiting different people. Social constructs can be difficult to alter, and the boundaries between competing claimants to a natural resource have often produced stalemate rather than problem solving. But an adaptive approach can loosen deadlock with surprising outcomes. The social dynamism of learning can undermine socially constructed stalemate. Although this book focuses on ecosystem management, the subversive flexibility of social learning has wider implications for environmental and public policy.

Large ecosystems are arenas of interdependence. Humans recognize interdependence only when it is more inconvenient not to do so. The institutions of property, privilege, and authority that predate the recognition of interdependence are durable and likely to remain so. Yet large ecosystems are also laboratories of institutional invention. Out of the fractious chronicle of ecosystems ruled by divergent human interests comes most of our small fund of ideas for managing the planet, the largest ecosystem, the one least likely to come under a single government. The cases of the Columbia River, examined in detail in Chapter 2, and of others discussed in Chapter 5 provide important lessons about what sort of governance can be built and how it might be operated. Focusing only on the global case affords a sample of one; by looking at large ecosystems we expand the possibilities for learning.

Large ecosystems offer the possibility of observing cumulative and other large-scale effects. The problems of suburbs cannot be easily projected from the difficulties faced by rural communities. Biological populations can thrive even while many of their individual members perish. Many social and ecological problems become apparent only in settings of sufficient size and complexity. Such places are usually highly imperfect laboratories. No large ecosystem is perfectly matched to any other; what works in one place, in one case, may not transfer. Yet the fact that knowledge must be incomplete is not a reason to disparage what understanding we can glean from these laboratories.

A First World Bias

This is a book mainly about the American experience. But five centuries after Columbus's discovery of a new world, it is obvious that environmental quality is a global issue in which the developing countries must play a central role: they are three-fourths of humanity now, and likely to become a larger fraction still; they are the stewards of the biological treasures of the tropics; and their aspirations for a better life are likely to become the principal engine of human impacts upon the environment.

Sustainable development is a more vexed issue in the Third World than in industrialized countries, first because of the insistent need for progress, and second because of the continuing call for equity. The poor nations as a group have lost ground over the past generation as the social and political tensions of decolonization, changes in international commodity markets, and continued population growth have produced upheaval and corruption more often than investment and economic growth. The pressures of international debts and disrupted domestic economies leave little room to consider longterm environmental goals—and bitter impatience with suggestions from the rich to forgo economic gain.

There are three reasons why I do not put the role of the developing world at the center of analysis here. The first reason is conceptual. In the search for a future in which living standards characteristic of industrial economies can be maintained, it is essential to consider the institutional limitations and possibilities of advanced industrial societies. The second reason is practical. Because industrial economies have made the biggest mistakes and have had the wealth to learn from some of them, their experience is most of what we have to go on. The third reason is speculative. Nations that have changed the economic role and status of women and the life-chances of the old through social insurance have tempered their rate of population growth. We may have reason to expect that sustainable development can also benefit from conditions in nations able to experiment with different modes of living. That is a hopeful reason to learn.

"Sustainable Development": An Elusive Idea

International organizations such as the United Nations now predictably call for "development that meets the needs of the present without compromising the ability of future generations to meet their own needs." Like national "security" or "affirmative" action, sustainable development seems to have become a slogan to elude hard questions: how might the world achieve sustainability, and how would we recognize that we had done so? I offer a partial answer here: social learning enables us to search for sustainability. The crucial test will be learning over time scales of biological significance—learning that lasts long enough to sense the long-run response of the ecosystems we use. Such long-term learning is the acid test of whether a policy is in fact headed toward sustainability. But a process is not a result, nor is the existence of a process the same as the will to use it.