It is a familiar story, if one with great power to break the heart. The company has entered a period of extreme financial stress. Layoffs have occurred at the periphery, and more substantial ones are clearly to follow at the core. Strange memoranda have begun to appear in which the company testifies to its steadfast commitment to research, even as managers emerge from weekly meetings with long faces, unable to look their friends in the eye. Things look bleak. Just at this critical moment, rumors begin circulating of a series of astonishing technical breakthroughs. These seem miraculous, so everyone is curious about how they were achieved, but the authors are understandably playing their cards close to their chests. The company feels warmly toward the accomplishments and comes to encourage frequent and aggressive interaction with the press. The lawyers busy themselves with patents and licensing agreements. The crisis mitigates, but has not disappeared, for the claims are false, the patents are empty, and the scientists have stepped across the line from seekers of truth to seekers of advantage, never to return.
Scientific misconduct is not unique to industry, for the professional pressures in universities, nonprofit organizations, and federal laboratories are similar and in many respects greater. The 1974 painted mouse episode at Sloan-Kettering Institute 1 and the recent report of data fabrication at Lawrence Berkeley National Laboratory (Physics Today, Physics Today 0031-9228 55 9 2002 15 https://doi.org/10.1063/1.1522199 September 2002, page 15 ) are cases in point. However, the economic inducements to misconduct are especially easy to understand in the case of private-sector research. Anyone who has worked in industry long enough to have experienced a business cycle knows how unbearable the job pressure can get when a company is in trouble and how this pressure can turn otherwise excellent and honest scientists into willing deceivers. It is neither uncommon nor hard to understand. Threaten a resourceful person with loss of home and endangerment of family and it is scarcely surprising that the person “innovates.” The recent findings of fraud at Bell Labs, Lucent Technologies (Physics Today, Physics Today 0031-9228 55 11 2002 15 https://doi.org/10.1063/1.1534995 November 2002, page 15 )—the ostensibly false claims of a new technology to suppress avalanche breakdown in field-effect transistors—are noteworthy only because of Bell’s special stature in American science and its reputation, both partly attributable to Bell’s having been shielded from such pressures by the old AT&T monopoly.
Research linked to property
The recent events are nonetheless extremely important because they force us to confront a fundamental flaw in modern beliefs about science: Research linked to property has a built-in conflict of interest toward the truth. For a research investment to be justified, it must produce value equal to or greater than that of the investment. When that value takes the form of intellectual property—knowledge that one can sell—as it commonly does, it must be kept secret, since no one will buy knowledge that is available for free. The core content of useful industrial research can rarely, if ever, be submitted to public scrutiny. This secrecy increases the opportunity for impropriety and thus makes the knowledge inherently less reliable than comparable knowledge produced in the open.
Sadly, the problem is greater than the mere opportunity to deceive. Although outright fabrication of data by scientists is rare, scientific deception is commonplace. The academic who refuses to exaggerate in proposals, for example, will not get grants. The industrial worker who explains the core of his technical niche to someone else will jeopardize his job. Even at Bell Labs in its heyday it was common for the scientists working in the public domain to be ignorant of matters deeply important to the company—even while being exhorted to be “relevant”—because the knowledgeable technical people would not reveal the problems to them. The mandate to generate property forces us to deceive. Members of Congress and managers in NSF and other federal agencies would do well to reflect on this effect and understand that some fraction of the industrial-style research portfolio of which they are so proud is simply lies.
It is often argued that industrial deception is a uniquely self-rectifying problem, in that an engineering firm with rotten patents will get a reputation of shoddiness and unreliability and will eventually suffer in the marketplace. But this rottenness can take an extremely long time to reveal itself, while secrecy, accumulated mistakes, and economically motivated disinformation slowly clog the pipeline of innovation. In this sense, ownership—more accurately the secrecy it necessitates—is not the engine of progress, but its enemy. One cannot both expose knowledge to scrutiny and keep it for oneself to sell. It has to be one or the other.
This process is why making over universities in the image of business is such a terrible idea. The great power of university research is its openness and the inherent truthfulness—stemming from this openness—of the knowledge it generates. Bell Labs, IBM, Xerox, and others have been traditionally held up as shining examples of how industry could match, and even surpass, university truth-seeking. But recent events tell a much sadder story. In a truly competitive environment an industrial laboratory cannot do academic-style research—studies on the cutting edge of knowledge with no obvious immediate applicability. Doing such research with private money makes no sense except as a charity or part of an advertising budget. Similarly, the use of public money in public institutions to generate private property is inherently dicey, no matter how cleverly its proponents argue otherwise. Although making universities over into businesses may generate more patents, it also corrupts scientific traditions and causes people of goodwill to become hamstrung by the need to generate property, sliding into mediocrity and dishonesty as a result. I think it is time to realize that the entire model is wrong, and that the product of science in general, and university science in particular, that counts is not intellectual property at all but young people with fiery independence and the courage to take risks.
An obsession with truth
We have entered a time in which correctly understanding and promoting the values of physics are essential for its survival. The tradition of sending government money down to universities without some value coming back has ended, and we are increasingly seen in the government and depicted in the press as an irrelevant science of the past. It thus behooves us to think carefully about our traditions and seriously consider the possibility that our key asset may not be our ability to build instruments at all but our obsession with fundamentals and truth. This is especially true vis-à-vis the life scientists, who have more money, less oversight, and much more tolerance for imprecision than physicists. Rather than allow ourselves to be defined by the property we generate, I suggest we take the high ground and turn ourselves into the gold standard of truth. This is the way to make physics relevant and important in this “age of biology.”
Everything we do in physics is built upon a scaffolding left behind by our predecessors. This scaffolding is reliable and strong because it is lovingly maintained in public by the community of science, not because it is in anyone’s self-interest to do so. Thus we are reminded that scientific traditions are a remarkable invention of civilization that, like banking, representative government, and law, are difficult to improve. The traditions of science rest fundamentally on a foundation of teaching and learning, and are dedicated to the task of creating new knowledge that is true, not of creating industrial secrets that may or may not be true. This dedication is their unique value to society—the thing science brings to the table that the natural working of the economy cannot. Human weakness and outright criminality pervade our world, and it is simply naive to assume that technical deceptions, some quite sophisticated, will never occur. But so long as the details are made public and dutifully subjected to the harsh light of scientific scrutiny, those deceptions will be hunted down and destroyed, just as they have always been. Only in the shadows of proprietary knowledge can they survive to work their mischief.
The career of every scientist is a struggle between the animal needs of economic survival and the higher obligations of citizenship. None of us can repeal the laws of economics, nor would it be desirable even if we could, for economic spinoff is the important tangible benefit society gets from its investment in science. But economics is not fundamentally what science is about. Science is instead about personal sacrifice and uncompromising dedication to clarity, meaningfulness, and truth. For each of us aspiring to a technical career, there comes a moment when we must choose between creating knowledge and creating property. Both choices are legitimate and important, but only one is science.
Over the past several months we have witnessed events disturbingly inconsistent with the belief that proximity to industrial needs is a valid substitute for traditional scientific openness. Let us accept the evidence, think carefully about what it means, and modify our theories as appropriate.
REFERENCES
Robert Laughlin is the Anne T. and Robert M. Bass Professor in the School of Humanities and Sciences and a professor in the departments of physics and applied physics at Stanford University in Stanford, California.