The history of science shows that curiosity, imaginative tinkering, dead-ends, randomness, and serendipity all play an outsized role in gaining insight into the natural world. The same can be said for the social sciences. Real insight into the human condition or our current predicament often comes in mysterious ways, and may involve as much rediscovery as discovery.
Robbert Dijkgraaf, director of the Institute for Advanced Study in Princeton, has written a companion essay for Princeton University Press’s republication of IAS founder Abraham Flexner’s famous article “The Usefulness of Useless Knowledge.” Following in Flexner’s footsteps, Dijkgraaf makes an important case for the social and scientific value of “unobstructed curiosity” and the need to resist the growing pressure to prioritize “short-term goals” and direct all effort to “more immediate problems.”
While demands for accountability, usefulness, and specific “deliverables” of economic or social relevance differ in the natural and human sciences, they foster the same tendencies. From my experience in the social sciences, three such tendencies stand out, and each now has a life of its own.
One tendency is to move away from broad, inductive, and holistic research—such as whole-organism physiology in medical science or broad cultural transformations in social science—and toward reductionist explanation and narrowly circumscribed empirical investigations. Providing predictable outcomes and problem-solving outputs typically requires a building block model, where each piece of research is some small step toward a larger goal, another brick in an emerging edifice—the science of happiness, for instance. The edifice is the promise that justifies the investment in the little pieces, which in themselves typically don’t amount to much. This approach enjoys an intuitive, deductive logic and some important results, but most path-breaking ideas do not develop in this way.
This narrowing move leads to the next tendency: prioritizing conventional, low risk ideas, which can easily be the next building blocks in an already familiar model. Setting, predicting, or promising specific outputs is not possible with real trial-and-error work. Such research is iterative rather than deductive. The findings cannot be stated in advance; they will only emerge in the process of investigation and interpretation. And there might not be any novel findings, which absence, though always of scientific value, is unlikely to be regarded as a suitable deliverable.
Yet another tendency—and by itself a crucial reason why innovation has slowed—is the overvaluing of individual achievement. The dominant entrepreneurial model of science today, natural and social, prizes not the search for truth or a community of joint inquiry but secrecy and the competitive development of ways to transform ideas into intellectual property (such as patents, papers, grants, and awards).
The context of innovation, by contrast, is typically one of collaboration. It often takes place in what the microbiologist Ludwik Fleck called in 1935 a “thought collective,” a community of persons engaged in a mutual exchange of ideas. Over time, members adopt certain ways of perceiving and thinking, and in their exchanges, continually transform their understanding and think new thoughts. Think Bell Labs in the immediate postwar decades, or, in a very different way, the Frankfurt School.
In his important essay, Dijkgraaf calls for a changed funding strategy, the directing of more resources to research untethered to immediate outputs. That is all to the good, but the underlying reasons for the problems in science funding go deeper. The benefits of basic research are unlikely to be realized without deliberate efforts to counteract the accompanying tendencies to reductionism, deductivism, and individualism.
Joseph E. Davis is publisher of The Hedgehog Review.