Recently, I posted a message to the Postdoc Network Listserv asking postdocs to share stories about the worst abuses they had observed and experienced. If you spend much time around postdocs, you hear lots of these stories: stories of favoritism, deception, and various forms of abuse.

Even if it's rare--and I think it is rare--mistreating postdocs, or any other group of scientists, is a serious problem. Young scientists, not to mention society, need to feel good about the work scientists do and the way they do it. As one respondent to my query wrote, "I'm being asked to mentor little girls and encourage them to be scientists when they grow up. On the other hand, I am scraping to survive as a postdoc day-to-day: professionally, emotionally, physically, and financially. We really need to educate the public, and ourselves, about the realities of science. It's the only way we're going to be able to improve anything."

Postdocs are a discontented lot, and justifiably so. For years, Postdoc Network writers have documented their lousy pay, inadequate benefits, long, inflexible hours, inadequate vacations, dearth of good long-term job opportunities, and so on. Economist Richard Freeman has called postdocs "one of the greatest bargains in the U.S. economy," and he's right. Here's another way of saying the same thing: Postdocs are among the least adequately compensated employees in the United States. Given this assessment, with which few serious people disagree, a bit of disgruntlement is to be expected.

But not all the stories I've heard can be written off to disgruntlement. I'll spare the details for now. It suffices to say that some principal investigators don't treat their people very well.

I know a lot of PIs. Many of my friends are PIs. I used to be one myself. We aren't all the most pleasant people in the world--many of us lack a bit of social polish--but generally we're decent people. Are these decent people the same ones who are treating their postdocs dishonestly and, in some cases, abusively? Why? Are these all isolated events, or is there some pattern to them?

Based on the stories I've seen and heard so far, I think that there is a pattern. This is a highly unscientific study--indeed, even calling it a "study" rather misses the mark--but I'm ready to make what I think is a reasonably well-supported generalization: Within my small sample, many of the stories of postdoc abuse have been about brand-new assistant professors.

It makes sense when you think about it. New assistant professors tend to share three traits: stress, inexperience, and limited competence as managers of people.

A certain amount of stress is inevitable, and only, um, experience can cure inexperience. But people-management skills are relatively easy to acquire for smart, determined people, once they've realized they don't have them and that they need them. They just have to know where to look.

So why is learning how to manage people so important? All that matters is doing good science, right? Well, no. Management skills are especially important for young faculty members because in times of stress people's instincts, their innate understanding of and compassion for their fellow person, can abandon them. I've seen it happen. To me.

I was a sprinter when I was in college--not much of a sprinter, but I showed up to practice most days. About once a week, my fellow sprinters and I would do a certain, dreaded drill. We'd sprint a kilometer or so up a steep hill. We'd do this over and over again, maybe a dozen times, until we were ready to die and wanted to. But I can hear my track coach saying to me, with his Trinidadian accent, "Watch your form, Austin! Your form will get you up the hill!"

Good form will get you up the hill. In times of stress you need good form, and good form comes from repetition (i.e., experience), training, and careful attention. In the context of an organization like a research laboratory, good form means establishing and following clear policies and guidelines. It means systematically, in a disciplined manner, honoring your obligations and treating people with respect. These things are easy to forget when you're faced with tenure pressures, grant-proposal rejections, and the responsibility for meeting payroll. But forgetting them is unpleasant for everyone, not to mention counterproductive. You need good form. You need management skills.

In case it isn't clear, what I'm suggesting is that some of the problems postdocs encounter can be attributed to the combination of unmanageable PI stress and a lack of PI management skills. It's an untested hypothesis, and even if it's true it's not the whole problem, but I'm comfortable asserting that if PIs, in general, had better management skills, then postdocs would, in general, be a more contented lot.

And there is, I think, reason to hope that by giving today's postdocs a taste of what it takes to be a good manager, by helping them learn good management form, we can help make tomorrow's PIs better managers and tomorrow's postdocs a little less disgruntled. We'll have to continue to address the other issues, as well--pay, benefits, long-term job prospects, and so on. But helping PIs acquire management skills is, I think, an important piece in a complex puzzle.

It is, of course, necessary to find and deliver the right information on management; if you pick up the wrong management book and read it too uncritically, you may end up worse off than you were when you started. There are different schools of management, and not every management theory applies to every context.

A science laboratory isn't a factory. Scientific knowledge is not a commodity. Running a research laboratory is not like making socks. Scientists are knowledge workers, technical creatives, and there's plenty of scholarship out there on how to manage knowledge workers, if you know where to look. I've distilled a few tips from those places that I find make best sense. But pretty much all of it makes a single point: Micromanaging knowledge workers--let alone whacking them over the head, pinching them on the *ss, and making them feel like dirt--is a bad idea.

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