Some recent tweets about fellowship requirements got me to thinking. As an astrophysicist with an interest in planet formation, the analogies between growing a planet and growing an academic have been yammering in my brain all day, so I might as well splurge them here.
I should start by explaining how to grow a planet. In the most commonly accepted model, planets start out as interstellar dust grains inside a circumstellar disc. These grains collide and stick by electrostatic attraction (just as dust does behind the sofa at home). These blobs of dust continue to grow, and become less fluffy with each collision. As the grains grow past a critical size (about a few metres, i.e. boulder size) the growth becomes more efficient. Eventually, the boulders grow into protoplanetary cores, which can either a) collect gas from the circumstellar disc (if they grow quickly enough) and become giant planets like Jupiter, b) collect some gas and become icy giant planets like Neptune, or c) not collect much gas at all, and simply become terrestrial planets like Earth.
There are two serious obstacles to grains wanting to become a planet. The first is that getting to the critical size is hard. As the grains get to a few centimetres in size, collisions become erosive and shattering, breaking up the grains, sending them back to square one.
The second is that the disc is not an extra in this play: it is an important actor. The gas in the disc exerts drag forces on the dust, extracting orbital energy and causing the dust to spiral in towards the central star. Near the critical grain size, this drag force is significant, pulling grains out of play within a very short period of time, as the above graph shows (a few thousand years, which is very short given that you need at least a million years to make a planet this way). Getting past the barrier raised by these two obstacles is very difficult, perhaps impossible without some extra assistance (like spiral waves or vortices in the disc preventing the dust from moving inwards).
So what does this have to do with postdocs? The key word is bottleneck.
Just like the bottleneck that grains face towards becoming planets, postdocs face a similar bottleneck (at least in science, as I see it). When starting out as a PhD student, things are relatively easy – in astronomy, STFC have been protecting numbers of studentships for some time, and will do for the foreseeable future. Once your PhD is finished, getting the first postdoc is difficult, but reasonably doable compared to the subsequent career steps. With every postdoc you do, you move away from positions with protected numbers, and closer to a no-man’s land where a great deal of talented people end up leaving for various reasons, which can vary from inability to travel abroad for work, to simply tiring of the cut-throat competition for jobs that are well-paid, but often less well-paid than jobs available to people with science PhDs. If you can get past this hurdle, e.g. if you get an advanced fellowship of some type, then the career path becomes significantly better-defined. These fellowships typically come with enough money to pay yourself for a few years, and get some students into the bargain. The chances of a permanent academic position begin to rise – departments commonly take on an Advanced Fellow as a permanent staff member, as they are effectively free for the first few years. In essence, you’re well past the bottleneck.
If you’re feeling pretty dismal about the careers situation, which I admit I do from time to time, you might adapt the above graph to this:
This is admittedly a very gloomy picture, but it does have a grain of truth to it. Advanced fellowships typically require a certain amount of time in research since completing your PhD, usually about 5 years, and as much as 8 years in some cases. As this is the least securely funded portion of your career, you have to work very hard to get through it, and you may have to make pretty big sacrifices too.
The saddest part is that prospective PhD students are not told this, and have little means to find this out for themselves. Perhaps we should be teaching postdoc formation as well as planet formation at undergraduate level.