Replacing the administrator at the Environmental Protection Agency is a step forward. But William D. Ruckelshaus has a task more difficult than ending the shoddy practices. Unfortunately, the real difficulties are not political. The fundamental problems are scientific uncertainty, public attitudes and the efficacy of government regulation.

An industrial economy inevitably generates air and water pollution, toxic wastes, occupational accidents and disease, and some unsafe consumer products. A condition of employment at the health, safety and environmental agencies since their inception has been persistent criticism for being unresponsive, uninformed and for promising too much and delivering too little. No one should be deluded into thinking that changing personnel at EPA will solve these problems.

The main difficulty is that while danger lurks everywhere, it is nowhere very great. Pollution in the air, the water and in the environment generally increases disease risks, including cancer. Yet rarely if ever are the risks high enough to constitute an emergency. For example, even at Love Canal, exposure to this stew of hazardous substances would be estimated to increase lifetime cancer incidence by perhaps 1 percent. I do not mean to disparage these concerns, but rather to provide a rough estimate of the problems faced. Life styles, including smoking habits and diet, have a much greater effect on disease rates, including cancer, than current environmental exposures.

Yet most regulation takes place in a crisis atmosphere where decisions must be made immediately, without adequate scientific investigation. These problems will not disappear; the industrial society of our grandchildren will present the same sorts of risks and give rise to the same concerns.

The basic elements of a solution can be outlined. None is easily accomplished, but no satisfactory solution can emerge without addressing them.

The single most important step is a clarification of goals. It is impossible to make the environment pristine or to reduce environmental risks to zero. Stating unrealistic goals, such as eliminating all discharges into the nation's waterways by 1985, is a major source of current problems. It encourages the crisis orientation, short-term fixes to long-term problems and confrontation, since environmentalists are dissatisfied with half a loaf and industrialists see no point in taking steps toward an unreachable goal.

Just what does society desire? How much are we willing to give up to enhance environmental quality? Survey data consistently show that the vast majority of the public supports environmental goals, but not, presumably, at the cost of a large reduction in living standards. The public needs to be better informed about what would be gained by each proposal and what it would cost.

A second step is to move away from litigation and confrontation more generally. Crises require immediate action, while permanent problems require planning and analysis. EPA must develop an agenda and pursue it vigorously. Litigation to change priorities generally doesn't work.

A third step is to improve the scientific basis of agency decisions. A conference that my colleague, Robert Crandall, and I organized in 1979 came to the conclusion that there is almost no scientific basis for current health and safety regulation. Regulators do not usually understand the nature of scientific data (they yearn for one-handed scientists). They have such short deadlines that studies cannot be mounted to answer critical questions and they do not know how to use data that are less than conclusive. For their part, scientists don't understand the regulatory process and shun its rough-and-tumble atmosphere. The resulting regulations are often arbitrary and uninformed, inefficient and unenforceable: they are costly and don't accomplish their goals.

Recent innovations allow quantification of risks, albeit with uncertainty. The uncertainties themselves can be analyzed and decisions made with a view to when critical uncertainties will be resolved. The scientific basis of EPA decisions can be improved without appreciable increases in resources or delay.

The fourth step is improved implementation. The problem is so large and regulatory procedures so cumbersome that the EPA cannot possibly protect the whole environment. About 1,000 new chemicals are introduced each year, adding to the 60,000 in common use, yet the EPA cannot regulate more than two or three dozen new chemicals a year. Nonregulatory mechanisms are needed for targeting chemicals, setting standards and controlling exposures. Before the establishment of EPA, companies relied on voluntary standards set by industries and professional associations. The corporate conscience--plus the threats of adverse publicity and lawsuits--promoted widespread compliance with these "voluntary" standards.

EPA enforcement of its standards is hampered by the small number of its inspectors and by burdensome court procedures. Substituting economic incentives for judicial enforcement, where catastrophe is not threatened, could improve enforcement.

There are EPA success stories. For example, chlorobenzilate was regulated by EPA in a proceeding that was ultimately satisfactory to all parties. This pesticide poses significant risks to workers, but is important to citrus growers. Rather than some bold move like banning the substance, EPA carefully defined the uses where chlorobenzilate has no substitutes, limited use even there, and specified how workers were to be protected.

The lesson is that facing up to a problem, carefully reviewing the scientific evidence, and involving all parties can produce good solutions, even to difficult problems.