By Marc Kirschner
If there is one principle of scientific inquiry we may cite, it is Francis Bacon’s:
“Those who have taken upon them to lay down the law of nature as a thing already searched out and understood… have therein done philosophy and the sciences great injury. For as they have been successful in inducing belief, so they have been effective in quenching and stopping inquiry… ”
Francis Bacon Novum Organum (translation from the Latin in Harvard Classics)
In considering reform of the US biomedical research enterprise, we are naturally driven to trying to patch the systems we have, systems that seemed to have worked well enough in the past. Yet dealing one by one with existing flaws may not assure the degree of improvement that one would hope for. Scientific creativity and accomplishment are the products of complex systems: education, mentoring, distribution of limited resources, recognition etc. Merely correcting individual programs without considering the logic of the whole structure risks unintended consequences of our actions. It seemed to me that it would be useful if we should first try to agree upon the basic principles of effective scientific conduct. I think we can rely on historical precedent for that, as it seems that although the experimental and theoretical methods used by scientists have changed a great deal, the fundamental motivation and activity of scientists have changed very little. Here are some principles for effective science, which I believe are self-evident. All our proposed solutions should be measured against them.
1. Science best proceeds in an environment of free inquiry.
The individual scientist should be free to raise new questions and propose original experiments, regardless of how heretical or unfeasible it may seem to the majority and in particular how independent of direct or immediate application they are. History has taught that the cost of not trying is much greater to society than the cost of failure. This poses obvious problems when science is not merely a statement of philosophy but requires expensive laboratories and materials. It means that within the context of some responsible structure of scientific funding and publication every effort should be made to nourish the iconoclast, including the researcher who is straying outside his or her own discipline. The last 150 years of biology is a tribute to the success of that philosophy: Mendel, Banting and Best, Delbruck, Ham Smith and Werner Arber, Nusslein-Vollhardt and Wieschaus, Mello and Fire.
2. Science progresses most rapidly when scientists can focus the majority of their energies on science.
In the distant past many scientists were from wealthy families and several, like Darwin, had no need for work. Yet these scientists put tremendous efforts into their science and produced amazing work. More recently, much science was done in academic institutions where the researchers’ entire salary was paid by the institution, or fully funded industry research institutes such as Bell Labs. Today, the vast majority of scientists have unwillingly traded this situation for the insecure life of a freelance entrepreneur, with many mouths to feed including their own. Senior scientists, and many junior scientists, spend the majority of their time on matters other than science: grant writing, compliance with federal regulations, animal protocols, human subjects protocols, rewriting rejected papers. In such an environment is it any surprise that these people are apt to be reading the winds for the newest funding opportunity rather than following their own idiosyncratic ideas? Research, teaching and discussions with colleagues should be the major focus of a scientists’ life. No matter how much money goes into science, progress will be undermined by the diversion of the scientist’s concentration on peripheral matters no matter what their justification.
3. Science proceeds best in the context of a lively, self-organizing community.
There have been many magical groupings of scientists: the Cold Spring Harbor Phage Group in the 1950s , the NIH in the 1960s, the Laboratory of Molecular Biology in the UK in the 1960s and 70s. These groupings were characterized by local autonomy and the availability of funding for common resources, and they nurtured outstanding postdocs and students. These environments socialized young scientists in the ethic of hard work, rigorous analysis, high achievement. Crucially, community-controlled resources made it possible for these assemblies to make their own judgments about which scientific areas were exciting, which risky, long-term projects to attempt, and which scientists to support. We need more such communities.
4. Effective communities are built from many types of contributors
Currently we value highly the individuals who are the first to be recognized as making a great discovery. Of course these individuals are worth celebrating, but also valuable are those who take on a new and difficult problem and lay the groundwork for others to make great discoveries, or those who take the time to listen to students and colleagues and offer constructive criticism or those who teach or take on administrative responsibilities enabling the work of others. How to maintain such individuals in our communities, in circumstances where funding is driven by shallow highly individualized metrics, is a major challenge.
5. Communication is the life-blood of science.
Publishing one’s results, once a straightforward process, has become an incredibly painful one. As journals and reviewers have demanded “complete works” extraordinary amount of time has been wasted filling the pages of supplemental information and this has prolonged graduate and postdoctoral careers, blunted the dynamics of scientific discovery, and interfered with the collaborative spirit of scientific across subspecialties. I see two roots to this problem: the largely justified belief among young scientists that their very careers depend on publishing in a very small set of journals, and a complete lack of penalty for reviewers and editors who deliver rude, incorrect and unconstructive criticism. The major attempts the community has made to change this situation have focused on creating new competitors to the “general” journals, and have been mostly ineffective so far. In the past, specialist journals were recognized as publishing the majority of the most important papers. The barriers to publication were high with regard to professional quality, but much less emphasis was placed on trying to imagine the significance of these results in the future. We need to find a way to re-affirm the importance of major results within a field; we need to train young scientists in the art of constructive criticism; and we need to find a way to restore civility and collegiality to the reviewing process. Above all we need to change the balance of power between anonymous reviewers, editors and authors.
6. Science is unpredictable.
Much to the chagrin of funding agencies and disease groups, science — whose goal is often prediction (such as the orbits of the comets) — cannot predict its own future trajectory. Science operates at the boundaries, at an endless and ever-expanding frontier. The more we know the more there is to know, the more questions to be asked, the more dissatisfaction with current answers. The war on cancer gave us ways to combat HIV, just as the penicillium mold that landed on Fleming’s agar plate ruined one experiment but saved millions of lives. Funding mechanisms that focus on a detailed process of investigation rather than on the investigator’s abilities, the clarity of ideas or simply the determination to describe unknown phenomena, act as a strong constraints on progress. Idiosyncratic attempts to reward “high-risk, high-gain” science have typically not been effective, as they have typically chosen novelty of approach over novelty of the question.
7. Science is not easily divided into basic and applied subdivisions.
Great science has always often inspired application, such as Faraday who turned electricity from an idle amusement into the electric motor. A recent term, “translational science”, has been used to give distinction to research that is considered to be of immediate application. It is not at all clear that this effort will be successful even on its own terms. Julius Comroe studied retrospectively the scientific discoveries that went in to life saving advances in cardiology and concluded that they were 2 to3 times more likely to come from basic research than from applied research (Comroe and Dipps, 1976). Although this may not have been the original intent, today the words “translational science” have been misappropriated to mean “fundable science” or “publishable science.” In particular an overly strong emphasis on immediate applied outcomes has been used to belittle the study of model organisms or biochemical approaches to questions. The criteria we use to determine relevance to human health need to be revisited and revised in the light of history and not wishful thinking.