To a remarkable degree, decisions about who should be funded to do science have been made on an essentially nonpolitical basis, even though government has been the main patron. There has never been the slightest doubt that Congress has had the power and the right to decide on the disposition of every dollar of federal research appropriations. Nor has there been any doubt that university people have the "right" guaranteed them by the Constitution to petition Congress to meet their needs. The fact that both have chosen, with relatively marginal exceptions, not to do so has enabled American science to avoid both automatic funding formulas and decisions based on political favoritism.
I do not include among the exceptions national facilities like the great accelerator centers. They are not marginal, and neither are they exceptions. While decisions about the siting of those centers have had a strong political component, the political competition followed rigorous review of competing proposals by leading physicists. The bias in funding decisions has been toward scientific quality and the promise of quality, judged by people with the professional competence to make such judgments.
All human systems need to be watched and changed as the need for change becomes apparent. It would be hard to find a governmental system that has met that test better than the system of merit review. There is a large analytical literature attempting to assess the operation and effects of the system. Over the years the system has been the object of numerous congressional hearings and … agency studies, as a consequence of which there have been many changes and improvements. Certainly, the systems now in use in the various agencies are different and stronger than those first put in place.
At the moment, we know of several relevant developments. The National Science Foundation has just completed a new study of its merit review system and has announced changes in its operation. The Government Accounting Office and the Department of Defense are conducting studies at the request of Congress, and it may well be that further changes will result from them.
Some have argued that the change represented by the increase in congressional earmarking—a change away from decisions in which informed judgments about quality are central—represents just that kind of needed correction. I think not. My own view, and it remains the view of the organized academic and scientific community, is that we will do ourselves great harm if we move informed judgments about quality away from the core of scientific policy, and that the advantages of doing so are short-term and accrue exclusively to individual beneficiaries, while the costs are long-term and must be borne by all.
We stand now at what may be a historic point. The practice of earmarking has grown in recent years, but until now its visibility has been greater than its substantive importance. However, the pressures that have produced it are real and will not go away. There is a real and painful capital deficit, and for nearly two decades the federal government has essentially abdicated responsibility for meeting its share. The pressures on colleges and universities to contribute to the economic development of their areas are real, too, and if they cannot be met in an orderly manner, they will find expression in less desirable ways. In short, the longer the practice of earmarking continues, the larger and more pervasive it is likely to become, and the greater will be its threat to America's scientific and technological future. Therein lies the challenge to our beliefs and to our ingenuity. The question is how to meet it.
Let me make four suggestions.
1. Congress should authorize programs and appropriate funds for the construction and renovation of research facilities. Those programs should provide a regular and systematic way for institutions to win support for facilities that show promise of producing quality science and technology.
2. All programs should preserve competition and informed judgments of merit as necessary conditions for funding. The elements that make up the final judgment may differ according to the purpose and the nature of the program or project. Clearly, different considerations will govern the disposition of the Superconducting Supercollider than are appropriate for an investigator-initiated grant from NIH or NSF, and those, in turn, may be different from the sum total of the elements that must be taken into account in deciding among competitors for a cancer center or a supercomputer center.
The point is that whatever components constitute merit for the purposes of a particular action should be the subject of informed judgment, and wherever possible, should be based on competitive proposals.
3. Project research funding should be based solely on informed judgments about the intellectual merits of the work and of those who propose to do it. In practice this means at least protecting the programs of NSF and NIH and the basic research programs of DOD, the Department of Energy, and NASA from political influence.
4. The academic community should work with Congress to ensure that its funding decisions are made in an orderly way and include provision for informed judgment about the merits of proposed projects.
I want to emphasize that I am speaking to the academic community—faculty, department heads, deans, and presidents, all of those who may see the opportunity for gain. In constitutional terms, Congress and the President are responsible for shaping national science policy; but in practical terms, they always have been responsive to cues from the scientific and university world. Several of the propositions I have put forward are cast in terms of actions for Congress to take, but they are in fact propositions for us to urge on them. It is our collective conception of wise science policy that will prevail, and it is necessary now to say what we believe that to be.