Gregory Petsko - Brandeis/weill cornell medical center - The post - Doctoral situation

One of the ways you know you've become a senior scientist is that people start asking you to be on committees. I've done a pretty good job in my life of avoiding that. But there are some committees that you don't want to avoid. And a couple of years ago, I was asked by the president of the National Academy of Sciences to chair a committee about the postdoctoral experience in science in the United States. This is something that interested me enormously for a variety of reasons. One of them being, I never was a postdoc in the United States. I did a post-doc in France, but that was for less than a year. I've trained more than 50 postdocs in my career and I sort of thought I knew what kind of experience they had, but never having had one myself, I thought this was an ideal opportunity to learn something about a subject that's really important for the conduct of modern life sciences, but about which I didn't know nearly enough. Now, the committee hasn't issued its report yet, so there are no conclusions from the committee itself that I can talk about, but what I can do is give you some of my personal conclusions based on the data that the committee was able to gather. The first thing I discovered about the postdoctoral experience in the life sciences is that postdocs are the invisible people. It's extraordinary. We asked institutions to tell us how many postdocs they had. Almost without exception, they couldn't do it. In fact, in many cases they couldn't even come within an order of magnitude of how many postdocs they had. There are a variety of reasons for this. One reason is that postdocs, of course, come in a variety of sizes, shapes and flavors, including postdoctoral research associates, postdoctoral fellows, glorified technicians, graduate students who have finished their PhD and are staying on for a year or so to write papers, and so forth. And also, they come in a variety of support. Some are on fellowships, some are on training grants, some are on grants, some are on a variety of instruments, including support from private companies or, in some cases, foreign governments. But the main reason that postdocs are invisible is because of what they are and the way we think about them. I've spent my life believing that the way we talk about something influences the way we think about it. And nothing could be more true than in the case of the postdoc. Think about how we talk about other things in science. Undergraduates are undergraduate students. Graduate students are graduate students. Faculty members are professors, or assistant professors. They all have titles that reflect to some extent the duties that we expect of them and the obligations that the institution has toward them. If they're an undergraduate, they're an undergraduate student. We're supposed to teach them. Postdocs have a name that reflects nothing except the time frame of their career. It doesn't say what they are, it doesn't say what they're supposed to be doing, and it doesn't say what we are supposed to do with respect to them. I think that alone is one of the biggest reasons why the institutions allow postdocs to fall through the cracks. It immediately also suggests one of the things that I think we must do to try to make the postdoctoral experience better for everybody. And that is, every institution needs to have an office, or at the very least, an individual, who is responsible for postdocs on that campus. And that individual needs to figure out how many there are, where they come from, how they're supported, and what careers they go on to. And those data need to be accumulated, made publicly available, on websites, for the institution, and gathered together by organizations, like, say, the National Science Foundation, where they can be looked at on aggregate. Right now, most institutions don't have a postdoctoral officer. Some expect that the person responsible for graduate students, most institutions have a dean of graduate studies, or a graduate program officer, will have that responsibility. But I think it would be better if there was a specific person whose sole responsibility was to worry about postdocs because, as you'll learn in a moment, I discovered that a lot of people aren't worrying about postdocs, except, of course, the postdocs themselves. So, that was the first thing I found out. Second thing I found out was that nearly all postdocs go on to careers other than in academia. If you ask someone how many postdocs go on to academic careers, most people come up with a figure like 30, 40 or 50 percent. And they're probably deliberately trying to guess low. In actual fact, the figure is much below 20 percent. We're fond of saying that we should prepare people for alternative careers without realizing that we're the alternative career. Only a handful of postdocs are going to become like me. That's probably just as well, I don't need the competition. But the fact of the matter is, if we believe, as scientists, that the people we're training in our labs are being trained for academic careers, we're fooling ourselves and we're doing them a disservice. Most of them will not do that during their life. And we need to worry about whether we are giving them adequate preparation for careers that are not like the careers that we have. For example, there are very exciting careers in the sciences in science journalism, science policy, scientific funding, venture capital, patent law, and a variety of other things that I could list. Nearly all of which are dying for people who are trained in the kind of analytical thinking, critical analysis of data, and creative approaches to problem solving that we like to think we're training our postdocs in. But we almost never give the postdocs exposure to careers like that. We don't have people who are doing those things coming and talking with them, we don't have internships available for them to go and experience a month or two months or a summer, in an office in a patent law firm, or in a policy making position. Either at the local level or at the national level. I think we've got to do that. We need to be responsible for making it clear to postdocs, and actually, I think we probably ought to be doing it earlier than that, we probably ought to do it when they're graduate students, that realistically, most of them are not going to become academics, and their training should reflect that and give them some of the skills and exposure that will allow them to identify the careers they actually do want, and be well prepared for it. So, that's a crucial step that I think very few institutions are taking in a sensible way. Now, what was the third thing that I learned. The third thing that I learned is it's amazing how many postdocs seem to believe that this was an inevitable part of their scientific education. You became a graduate student, and once you were done being a graduate student, of course, the next step was to be a postdoc. The postdoc has become the default. And I think that's a huge mistake. First of all, as I said, postdoctoral experience doesn't prepare you for every career. For example, it's pretty lousy preparation for, say, high school science teacher. Moreover, I think you could argue that a lot of the career preparation that people need for non-academic careers could be done as early as graduate school. Maybe even, in some cases, a little earlier than that. So, being a postdoc shouldn't be a default. It shouldn't be as easy to be a postdoc as it is, and people shouldn't think it's automatic, it's something that they should do. No, there needs to be a set of career options presented to people, ideally as early as possible, that will allow them to make good decisions and those good decisions might well be, I don't want to be a postdoc. I don't need to be a postdoc, it's not for me. Now, those are some of the main things that we found out when we were doing the investigative work that we needed to do for this committee. But it got me thinking about a broader question. And that's the question of whether, as scientists, we're training too many people. There's tremendous pressure on scientists, especially young scientists today, to produce papers in boutique journals to get lots of grant support, and much of this now, because the bar keeps getting raised, and whether that's sensible or not would be a whole other talk, the bar keeps getting raised, and so, studies that used to come out as five papers now come out as one paper, that the length of a PhD thesis, or involve a PhD thesis worth of work, and in order to produce at that kind of level, labs have gotten bigger, people are training more people, they're taking more graduate students, they're especially taking more postdocs. And a system that was created when science was a mom and pop activity in the 1960's and 70's, has, I would argue, a structure that may no longer be suited for the kind of science that we're trying to do now. And it may serve both science and the young people in it, ill. Now, there's a lot we could say about the hegemony of certain journals and why that's a bad thing, and the raising of the bar for publishable work, and why that might be a bad thing. And the whole grant reviewing process, and why it's too conservative, and why people like myself are doing a bad job of reviewing creative proposals. But regardless of how we talk about that, the realistic point is that labs have gotten bigger, and we're training a lot more people. And the job market has not, especially in academics, expanded to keep pace with the increased number of people we're training. Now, if we train them properly for careers outside of academia, that might be ok. It might be that we can accommodate the number of graduate students and postdocs that we have in the community as a whole without them feeling like they wasted their time, were lied to, or in general, had a disappointing and frustrating future. That would be fine. But, it's also possible that we actually need to cut back. I'm not sure that's true, but I think we ought to have, as a community, a creative discussion about it. I think we should ask ourselves realistically, are we training too many people, and if we are, what's the best way to go about reducing the size of the enterprise. I don't think that a good way to do that is for institutions to legislate how large labs should be. I don't like top-down management of science. Anyway, I don't think that works very well. But let's suppose for the sake of argument, that we decide, and this is just hypothetical, that we decide we are training twice as many people as we should. How do we shrink the enterprise by a factor of two? There's an obvious way of doing that. We double the stipend. We're basically, in the life sciences, the NIH minimum salary sets the floor for people supported by almost every type of support. Whether it's industrial, or other types of fellowships or whatever, the ballpark figure is set by the NIH minimum stipend. And so, if we decide, for example, that we're training twice as many people as we should be, then what we ought to do is over a four year period, let's say, it could be five, I don't care, we should double the stipend. Phase in a doubling of the stipend. This will have two consequences. It will force labs, like mine, to get smaller. And in so doing, I'll have to make cold-blooded, carefully reasoned decisions about who I really want to keep and who I think really has a future. And who I should be training, and who I shouldn't be training. And that's something we're not doing very well right now, so being forced to do it for economic reasons might be a good thing. And the second consequence, of course, is we would finally be paying these wonderful young people, our graduate students and postdocs, because we'd probably double both stipends, we would finally be paying them something approaching what they were worth. So, regardless of what the figure is, right, whether we're training 25 percent too many people 100 percent too many people, I think if we decide as a community and we must, I believe, have this conversation now, if we decide that this is what we're doing, I think the best solution to the problem is a purely economical one. Because it has mostly beneficial consequences, not negative ones, and if it forces some of the huge labs we have to get smaller, and stop cranking out quite so many papers, you know, I think science will be just fine that way. Realistically, I probably would have more fun if my lab were smaller. I know for a fact the people in it would probably get a better training experience. And I'd be surprised if anybody else could say any differently. So, that's sort of the way I've been thinking about the postdoctoral situation. You can see that it spills over into the way I think we have to think about the graduate student situation, and the whole training of scientists in general. We've allowed the present system to develop by circumstance over a period of ten to twenty years. Things have changed not because we wanted them to change, but because we've allowed them to change. And we drifted into the situation that we have. Part of it's laziness. I mean, one of the reasons that we worry about stupid things like impact factors in journals and where you publish is because it's become a proxy for actually reading the papers and trying to figure out if somebody's done good work or not. If we weren't so lazy about that, these problems wouldn't have developed the way they have. So, we've drifted into the present situation and that includes the present situation where postdocs are the invisible people of science. I think you can't drift your way out. I think if you want to do something about it, you need to sit down as a community, we need to talk about it, we need to think about how to do it right. And we as scientists then need to tell the science administrators at our institutions, in Washington, that this is what we want done. We shouldn't let them decide for us and we shouldn't do nothing. We should control of the enterprise or take it back, because there was a time, actually, I'm old enough to remember this, when we sort of had control of the enterprise, and now the enterprise is sort of controlling us. If we don't do this, we risk having an entire generation of young scientists who are frustrated, who are disappointed, whose lives aren't turning out the way they wanted to, or expected to. And I think that this problem is so important that it requires the attention of the whole scientific community now, because the postdoctoral experience is supposed to be the best time of your scientific life. It's supposed to be the time when you have finally the freedom to pursue the things that you're interested in, without some of the educational responsibilities associated with taking courses and passing exams. It should not be a time of anxiety. It should not be a time of frustration and disappointment. It should not be something that you drift into because there's no alternative. It should be something worthy of the wonderful young people that we have who are trying to be scientists. And if we give them the kind of training they deserve, the kind they should be getting, it will enrich their lives, not just as scientists, but also as human beings.