There's an extensive body of knowledge devoted to the management of people who think for a living—so-called "knowledge workers"—and to the knowledge-based projects they engage in. While this body of knowledge arose mainly from the software industry, where complex projects can be tough to manage and the stakes high—case in point: Healthcare.gov—it is applicable to other domains, including science. 

In the traditional management theory—the so-called agency model, which developed to support industrial work—the relationship between the employee ("agent") and the employer ("principal") is oppositional, and workers' motivations are external. As management theorist Robert Austin wrote in Science Careers in 2006, "The objectives of principal and agent are thus diametrically opposed: The agent wants to be paid as much as possible for doing as little work as possible, whereas the principal wants to get as much work as possible from the agent while paying as little as possible." (Full disclosure: He's my brother.)

Any resemblance between such an arrangement and your postdoc appointment is regrettable, because the best motivations of knowledge workers—and scientists above all—are entirely different from those of factory workers. Knowledge workers are motivated by the work itself and the pleasure of doing it, by an internal drive to find answers or to make things. As most readers of this essay surely know from experience, anything that undermines that motivation—pressure to produce, meddling by management, fear of sanctions, anxiety, resentment, even gratuitous performance bonuses—worsens work performance. The best approach to managing knowledge workers, then, is to clarify the objectives, provide the tools and support they need, facilitate collaboration, and get out of the way.

About a decade ago, when I first started thinking hard about these issues, I had lunch with Tom DeMarco, an engineer and expert in managing knowledge-based projects. DeMarco started his career—it was his first real full-time job—at Bell Laboratories (Bell Labs) in 1963. Later, he founded the Atlantic Systems Guild, a consulting firm specializing in "methods and management in software development." Today he is among the leading thinkers on the management of software projects.

Among the 12 books DeMarco has published in his 50-year career (including a short story collection and a novel about project management) is one called Slack: Getting Past Burnout, Busywork, and the Myth of Total Efficiency. When we met for lunch the book had just been published and he gave me an inscribed paperback copy. I reread it recently and found DeMarco’s vision at least as relevant today as it was when it was published.

The book's premise is that in a knowledge-work context, too much pressure—too much emphasis on reducing wasted time or resources—can cripple knowledge-based projects, in large part by corrupting the environment in which knowledge workers thrive.

When knowledge workers are pressed too hard, the intimate connection between workers and their work is compromised and many things go wrong. Stress and anxiety harm mental health and, hence, performance. Competition rises and team cohesion—a major source of productivity gains—declines. Animosity may develop among staff, or between staff (who feel exploited) and management. Overworked workers take on extra tasks and pay a task-switching productivity penalty that DeMarco estimates at 15%, minimum. And it isn't just average productivity that declines; it is also peak productivity, those rare moments of transcendence when important breakthroughs are made.

When there's not enough slack, knowledge workers play it safe and stop taking risks. Organizations, too, do less to facilitate innovation: DeMarco points to the "bankruptcy of inventiveness" that results from "a failure to set aside the resources necessary to let invention happen."

When the pressure gets too high, people cut corners. They make mistakes. Under intense pressure and faced with the difficult challenge of meeting core goals, many workers aspire instead to proxy goals used to evaluate productivity: Scientists, for example, may pursue publications instead of real, important insights. Worse, faced with intense productivity pressure and dubious productivity metrics, some may cheat and lie to create the appearance of success.

Even in 2002, this insight was not new. When DeMarco joined Bell Labs, the workday there was 7.5 hours long. That length was chosen because the leadership considered it optimal: If the day were any shorter it would leave productivity untapped, they surmised, but if it were any longer it would reduce the creativity and effectiveness of the labs' scientists and engineers.

The results were impressive: The Number One Electronic Switching System, which DeMarco worked on; the Telstar 1, the first orbiting communication satellite; the laser; the metal-oxide semiconductor field-effect transistor; and the touchtone telephone. A couple of years later, Arno Penzias and Robert Wilson would discover the cosmic microwave background, which would yield the Nobel Prize in physics in 1978. The balance of the '60s brought molecular beam epitaxy and the charge-coupled device, among other insights and inventions.

The history of science includes many stories, some real and some apocryphal, of "aha" moments during mountain hikes and periods of quiet contemplation, but today the idea of giving scientists time to think seems quaint. Even before science's current fiscal squeeze, many graduate students and postdocs were working 60 to 80-hour weeks under heavy pressure to produce meaningful results. A shrinking job market exacerbated this pressure.

Post-sequester, and following a decade or so of declining budgets and several decades of overproduction of scientific labor, graduate students and postdocs are under more pressure than ever to produce important results, just to stay in the game. Probationary faculty members know they have little time to lose in getting their labs up and running and independently funded—against historically low success rates—and produce important work before their tenure decisions. Home pressures, too, are higher. The rise of two-career families—a welcome development—means the home workload must be shared.

For scientists in the trenches, more than just careers are at stake. For foreign scientists, failure can mean deportation. For soft-money researchers, it can mean unemployment. When a grant isn't renewed, tenured professors may be forced to lay off valued staff. As one scientist—one of a growing number of soft-money principal investigators working in the United States—wrote to me in a e-mail recently, "NIH is bombarded by grant-proposals not to advance the cause of science but to keep families housed and fed." Such pressures affect the way scientists work, the decisions they make day to day.

Even under such intense stress, today's science remains highly productive, but indications of strain are easy to see. While it's difficult to prove—there can be no experimental controls after all—science and society are likely paying a very high price for the intense pressure that so many scientists are under.

Jim Austin is the editor of Science Careers. @SciCareerEditor on Twitter

10.1126/science.caredit.a1300266