Why postdoc ?

A few of the blogs that I read regularly are written by postdoctoral researchers in different fields of science. A post by one of them (alas no longer available) has got me thinking about the whole postdoc issue.
It is a frequent subject of conversation for me, these days. Not only did it repeatedly come up at the March meeting; two of my PhD students are about to graduate, trying to decide what to do with their lives. A former student of mine is also thinking of doing a second postdoc. Questions that I am asked most often are: “What exactly is the point of doing a postdoc ?”, “What am I supposed to get out of it ?”, “What makes a postdoc good, as opposed to a waste of time and an aggravation ?”.
Having done two postdocs myself, and having been a faculty for over ten years, you would think that, at least to this type of question, I would have an answer ready. And well, I do, except that it keeps changing…
So, this is what I think now (read quickly):

First of all, I don’t think that there is any objective, compelling reason for making the postdoc a necessary stage in the career path of aspiring science academics. I know, things are the way they are, and are not likely to change anytime soon, but aside from the usual this-is-my-blog-dammit argument, I can easily imagine things functioning just as well without postdocs. One need only look outside the sciences, where postdoctoral appointments are virtually non-existent. English, education, philosophy departments hire their faculty straight out of graduate school, and most of these individuals perform quite adequately (hey, don’t get me wrong, I know that in the sciences we do everything better — especially in physics, right ?).
I am not aware of any research on this subject, but I do not expect the teaching performance of a tenure-track assistant professor to be significantly enhanced by postdoctoral experience. It will generally make him/her more mature and effective a researcher, but this additional maturity comes at a non-trivial cost [0]. Abolishing postdocs in the sciences would simply mean that the bar for tenure-track faculty would have to be re-adjusted, expectations should be more reasonable, and some additional research mentoring should be provided, for example by faculty colleagues, but also by former PhD advisors (scary, eh ?).
Frankly, I doubt if the overall effect would be even noticeable.

Anyway, this is all academic (no pun intended). From a very practical standpoint, at least one postdoc (and more often two or even more) is a necessary (by no means sufficient) condition to seek an entry-level faculty position at a research intensive university, in most fields of science and engineering [1]. It has been so for at least three decades in North America, and it is like that now pretty much all over the world. So, I suppose an easy answer to the question “Why do a postdoc ?” is just this: if being a professor is the ultimate goal, there is simply no escaping the postdoctoral route.
This is not a satisfactory answer though. It does not work for many, perhaps for most. Fact is, being a postdoc is not easy for many a personal and professional reason. The uncertainty about the future, the short duration, the feeling that one’s life is put on hold, are only a few of them. If “because I have to“, or even worse “because I can’t find anything better“, are the only answers that come to mind when asking yourself “OK, why am I doing this, again ?”, those two (three, four…) years will be no fun.
It is important to have a sense of purpose, if a postdoc is to be fruitful and rewarding an experience, a credible pathway not only to professional, but to intellectual and personal growth as well. This is especially true for someone not wanting to pursue an academic career in the long run (yes, it can make sense to do a postdoc in this case too).

So, here is my most compelling reason for doing a postdoc: to put it simply, it is the time to show ’em what you got. It is an opportunity for a talented young scientist to establish him/herself as an independent thinker and researcher. The moment you start out on a postdoc, you are basically unproven. There is no reliable way for an external reviewer to assess to what extent what you have accomplished up to that point, is really yours. It could be mostly the work of your PhD advisor, of a postdoc supervising you, or even of a fellow graduate student. There is no reason to assume, at that point, that you are able, on your own, a) to come up with an interesting and original research project, and b) to carry it to fruition. As a postdoc, the chance that you will be able to work in the shadow of someone else (if that is what you have done until then) is slim, as your postdoctoral advisor (PA) has his/her own graduate students to think about, and your fellow postdocs are (to a point) your competitors — they will not do the work for you.
It’s a challenge, of course, but for the most part an opportunity. And it is also a necessity, at this particular juncture. In these times, with academic communities basing their judgment to a large extent on containers rather than on contents, abolishing the postdoctoral position may well take away from anyone without a “pedigree” the only chance to present him/herself to the community as a credible alternative to the superstars of the year, when the time comes to look for a faculty job.

Based on the above rationale, these are my thoughts on postdoc strategies, in no particular order:
0) A well laid out research plan, with one or two concrete goals to achieve, is essential. This is not to say that one’s research interests or priorities may not shift in due course; that’s how science works, and there is nothing wrong with that. But one should have a plan at the outset.
1) Make sure that, whichever place you go, you are given the freedom to pursue your own research, at least half of the time. Beware of any potential PA who says things like “I need someone who can do xyz“, xyz being something very specific and technical. You do not want to make your postdoc about doing xyz all the time, nor do you want to be branded as “someone who does xyz ” (read: just a technician. Departments typically seek to hire scientists, not technicians).
2) Beware of any potential PA who talks and acts like a micro-manager. You do not want your postdoc to be “graduate school — part deux”. Your PA is a senior colleague. When (s)he proposes to you possible research projects, make sure they are broad, original (not a mere continuation of what that group has been doing for years) and potentially significant [2]. More generally: Take risks, go for big things — publishing a bunch of irrelevant papers will take you nowhere.
3) Teaching is an important part of the academic profession, but postdoc is not the time to do it. Postdoc is for building your research portfolio and career [3]. Having taught a large physics course will almost never impress search committees at research universities [4]. Even if you wish to teach at a 4-year college, having taught one or two courses will not give you much extra mileage over other applicants. Making a good teacher is difficult, and experience plays an important role.
4) Do not pass on the opportunity of supervising graduate students. As a faculty you will be expected to do that, so anything that can make the case that you can be a PhD or MS advisor will help, especially if your PA will state this clearly in his/her letter of recommendation.
5) Some of the criteria that I would adopt to choose a graduate school apply to the choice of a postdoctoral position as well. The department should be highly ranked, the PA influential, well-known and with a good track record of placing advisees… the usual stuff.

I would give the above advice to anyone looking for a postdoc, including those who saw themselves in industry, or outside research or academia in the long run. A profitable postdoc is always an asset and a valuable undertaking, even to someone who ends up teaching in high school (if more high school physics teachers had postdoctoral experience… well, let us not go there today). If you disagree with me, no worries — chances are by the time you read this line I have changed my mind.

Notes

[0]. An obvious problem with extending the average time that it takes for a science PhD graduate to land a tenure-track job, is that it may discourage a (possibly significant) fraction of potentially talented young people from pursuing academic science as a career path. On the one hand, it may be beneficial to weed out those who lack passion and tenacity. On the other hand, even passion and tenacity cannot always overcome difficult life circumstances.
[1]. In fact, it is not uncommon these days for science departments at 4-year colleges, and other institutions whose primary mission is teaching rather than research, to require some postdoctoral experience of candidates for faculty appointments.
[2]. Assessing the potential significance of a project is obviously very difficult. Help from senior mentors is usually very valuable. This is one of those cases where being in good terms with your former PhD supervisor may prove very useful, as (s)he can be a tremendous resource.
[3]. Obviously if the grant that pays your salary falls through, and you need to teach to bridge the funding gap, there is no alternative (not that I am thinking of any one to whom this may have happened — not at all).
[4]. Indeed, in a search committee meeting the issue may be raised of why, given the short time the person had at his/her disposal to build solid research credentials, (s)he opted to spend part of that time teaching instead of focusing on research.

Tags: Academia, Graduate Studies, Physics, Research, Science