Back to the drawing board - for a new PhD system

Imagine: the concept of a ‘PhD trajectory’ doesn't exist yet. There's no policy on how Dutch PhD candidates should be promoted. We could start from scratch. How would we design that PhD system? And would we end up with the system we have now?

We should start by asking ourselves what a PhD is for. Before I returned to Utrecht, I worked at Imperial College London for a few years. As a junior lecturer, I was immediately allowed to supervise a PhD candidate – no hassle with Ius Promovendi there. I had to do only one thing: attend a mandatory course on "the why of PhD candidates."

Because why are there PhD candidates? I still remember the course well. There was a very lively discussion about the purpose of PhD candidates. Are they there to carry out research? Are they there as "laboratory hands," indispensable for conducting experiments or delving into the archives? Are they there to select the next generation of academics from? Are they there to support our education? Are they there to eventually enter the professional labour market as highly specialized knowledge workers? Are they there to enhance the socio-academic status of supervisors? Or are they there so universities can secure their ‘promotiepremies’?

As a novice supervisor, I found that discussion incredibly valuable. But I've never had such a conversation in the Netherlands. If we had, I’d expect we won't reach an agreement. That the answer will be something like "a bit of everything."

The PhD trajectory is still based on a master-apprentice relationship

There are aspects of the Dutch PhD system that we, as universities, can be proud of. For example that our PhD candidates are full-fledged employees, with the corresponding employment conditions. In few other countries do PhD candidates accrue a pension and have the right to parental leave.

But while the institutional integration of PhD candidates is very progressive, the supervision is rather outdated. Not to say medieval. Because that's where it rubs most: the fact that the PhD trajectory is still based on a master-apprentice relationship. The PhD candidate under the tutelage of the all-knowing supervisor – the master of the guild. Why do we cling to this so much? Surely this medieval structure doesn't fit the way we want to train young academics? Wouldn't we organize supervision this way if we were starting from scratch?

There's a lot of discussion – rightly so – about the social safety of PhD candidates. Universities build additional safeguards and safety systems into the PhD process. But how can a system based on a master-apprentice relationship ever become socially safe? If ultimately, one person is the gatekeeper and determines when a dissertation is "finished"?

I've been a huge supporter of ‘Everyone Ius’

Because of Brexit, I returned from London to Utrecht in 2017 – with a personal ERC grant. This allowed me to recruit a PhD candidate, with whom I worked very well for three years. Until, at one point, that candidate asked me: “do I have enough content now to submit my thesis?” The only thing I could say was that I thought so, but that it wasn't my decision. That candidate had to consult the promotor – after all, I didn't have Ius yet. And that promotor had never really gotten involved. He was a co-author on all the papers and duly attended the annual interviews, but otherwise kept a distance.

I felt powerless that at one of the most crucial moments in that candidate's PhD trajectory, I didn’t have the authority to answer whether the work was finished. Even though it was our project. Since then, I've been a huge supporter of ‘Everyone Ius’. Just like in London and Sydney, where I worked before. Just like almost everywhere, really.

As critique on ‘Everyone Ius’, I often hear: “yes, but shouldn't we train promotors first?” Of course; training is very important. But why shouldn't a trainee promotor also take on the tasks and responsibilities of a promotor? That's how we do it with lecturers too: lecturers in a BKO program already teach. They are coached in this by a more experienced lecturer. Similarly, trainee promotors could be coached by experienced promotors until they obtain their ‘basic qualification promotor’.

Without PhD candidates, research would grind to a halt

But this whole discussion about Ius Promovendus is only one part of the bigger question: what exactly is a PhD program? Or what ought it to be? Let's peel that back a bit.

Many beginning PhD candidates think a PhD program is about becoming an academic. But for most PhD candidates, especially in the natural and life sciences, that's nonsense. There are simply far too many PhD candidates for all of them to become academics. If the goal is to cultivate the next generation of academics, then we've set up a rather inefficient process for it.

I once spoke to a colleague in the US who worked at a government lab. His goal was to supervise two PhD candidates throughout his career: one to succeed him and one as a backup. That colleague wouldn't last very long in the Dutch system, where there's constant pressure to supervise more PhD candidates.

Another goal could be the PhD candidate as a researcher. The PhD candidate who has the time and space immerse themselves in research – because, after all, they have hardly any administrative work or teaching duties. Without PhD candidates, research would grind to a halt. This is the PhD candidate as lab hands: as glorified pipette-taker, mouse-keeper, archivist, or interviewer.

Although many, including myself, will cringe at this, it's at least clear. A PhD is then just a job. That aligns with the good working conditions for our PhD candidates, by international standards. But why do we then expect a dissertation at the end? And why is every PhD candidate part of a graduate school?

Students who want to do impactful research don't go to university

No, most will still consider a PhD program a training program. The doctorate at the end is the biggest advantage. At least, that's what many people think.

When I told my parents I wanted to pursue a PhD, my father advised against it. He worked in IT and told me that PhD candidates were at a disadvantage when applying for jobs at his employer. Twenty years ago, PhD candidates there were stigmatized as loners who worked far too slowly and in too much detail. The opposite of what they were looking for at that company.

My father is retired, so he doesn't know if the reputation of PhD candidates there has improved. But I do understand where the ideas come from in such a company. After all, PhD research is individualistic, slow, and detailed. They probably could use some of those kinds of people at ASML or in the pharmaceutical industry, but they're not exactly the skills of the future.

And we don't just have this somewhat sluggish reputation with employers. Many talented students also seem to be unwilling to pursue a PhD. When I returned to Utrecht in 2017 with my ERC grant to investigate plastic pollution in the ocean, I immediately started looking for PhD candidates to recruit. I was invited to an event organized by The Ocean Cleanup, Boyan Slat's organization. I expected to meet some good interns there. And indeed, I did – but none of them wanted to do their PhD. No matter who I asked, I always got the response that they’d crazy to trade the trendy startup for the rigid university.

And that is a major problem. Talent doesn't just flow naturally to universities anymore. It has been a problem in computer science for a while that the best research facilities are not at universities but in the private sector. This lack of state-of-the-art facilities is going to affect many more disciplines. Students who want to do impactful research don't go to university.

Back to the drawing board

And that's why we must prioritize education in the PhD program. That's what we, as universities, are strongest at. That's our edge.

But let's then fundamentally change four things. Back to the drawing board.

One: get rid of the master-apprentice relationship, and thus improve social safety. No longer be dependent on just one promotor. Why do we consider it perfectly normal for an MSc program to have a full teaching staff, but a PhD program only has one or two examiners? Why not a different set of supervisors every year, for a different part of the learning process?

And then we also need to stop with the dreaded phrase "my PhD students." As a researcher, you simply can't own your PhD candidates!

Two: get rid of the dissertation. Turn it into a portfolio. And not just with academic chapters, but also offer opportunities for educational chapters and impact chapters such as white papers or policy documents. 

Three: all PhD candidates should have a mandatory internship outside academia. A PhD candidate from my team once spent three months at the Nationale DenkTank. In those three months, they learned more about collaboration, project management, and responsibility than in their four years on my team.

Four: get rid of the promotiepremie. It's a perverse incentive that perpetuates the enormous influx of PhD candidates. I'm sure I'm not the only one who has been asked by someone at the Research Support Office to convert a postdoc position on a project proposal to a PhD position, even though a shorter postdoc made much more sense in the project proposal.

I'm very happy to be here, and I had to earn a PhD for that. But to be honest: if I had known in 2004 what a PhD program really was, I don't know if I would have started it. Too medieval. The PhD system must and can be improved!

The views expressed above belong to the authors and do not necessarily represent the views held by DUB.