Default!
Academic Engzell problems, exploding cans and bucks and what to do about them
Per Engzell posts (bisks? skeets? bloots?) a lot over on Bluesky. Which also means he posts a fair amount of highly insightful observations when he’s not trying to outdo Hadas Weiss as the platform’s snarkiest academic.1 Among other things, he has an excellent and very simple theory on how to make it in European academia (more on that another time) and he is the only person I know who has publicly pointed out a glaring problem in the scientific community which will be coming back to bite us before long:
We must be grateful for everyone in academia who is a) clear-sighted enough to see such problems and b) sane and honest enough to openly name them a problem. I was aware of the frequency of bad research projects but I didn’t think about them as systemic oproblems to the same extent before I read Engzell’s rant.
So in his honour, I want to call these sorts of social constellations “Engzell problems” (a dubious honour, surely) and say a bit more about why they are so highy corrosive for academia (as well as other industries). They are not limited to research projects but a general syndrome in universities, scientific publishing and adjacent activities. Hence why I want to give a short definition and a few examples before I explain how they come about and how they can potentially be solved.
Academic demand and supply
What does Engzell mean when he speaks of “a project that was dead on arrival” and “sorting out the mess”? Although many academics lament a lack of funding, there is an astonishing amount of third-party money from the European Research Council (ERC), national public funding bodies and private foundations to do all kinds of stuff in all sorts of fields and disciplines, so at any point of the year there is a flood of more or less specific calls for applications and rolling deadlines.
And yet the lamenting academics are not totally wrong because these calls rarely address their expertise directly and ask for a slightly or totally different background and approach. For example, there is currently much, much more funding for research on artificial intelligence or climate change in Europe than there are competent researchers in those fields.
Conversely, even the handful of people who study Celtic languages or pseudoscorpion taxonomy often have a hard time finding even one call at any time that comes close to their expertise. This mismatch between demand and supply is mirrored on all levels of the academic system: from theme issues of scientific journals to job offers to conference panels.
The result is that academics will try to dress up to get the chance to find a job, receive research funding or get their work published. If you don’t have the necessary ingredients for a substantial meal, just be generous with sauce!
This is how we get papers and projects like “Pseudoscorpion taxonomy in the age of AI: Employing LLMs when we should instead be collecting in the field” or “Old Irish in shifting tides: Why there is barely a trace of changing CO2 levels in consonant shifts” that promise a lot and deliver very little. In that sense, the problem also exists in any other industry that is subject to fluctuating political programmes and the subsidies attached to them, such as NGOs, journalism or the arts.
It would be nice if the problem were to end here, if it simply existed due to external incentives and internal restraints and if it only consisted in time wasted and tax money misspent. Alas, it is much bigger than that.
Exploding bucks and cans down the road
In an Engzell problem, a mismatch of supply and demand produced by the academic reward system meets with a particular power structure of academic organisation. Universities, research labs, chairs, scientific journals and basically every other organisational structure and process in academia are built in a way that makes it very easy to pass down responsibility for one’s actions—usually to people who are worse-equipped to handle that responsility.
In the case of research projects, this usually means that even the most ridiculously cobbled-together project with no merits on paper has a chance of getting funded; simply because the mismatch of demand and supply goes both ways.
If there aren’t enough good projects for a funding call, the agency will still be inclined to spend all the money rather than hand it back to the taxpayer.2 (It would be nice if more people would keep that in mind when submitting dressed-up applications; perhaps it would prevent them from bending over backwards several times over.) In effect, the agency is not living up to its resposibility of only funding sound projects.
Once those projects are granted, this responsibility is concentrated in the principal investigators (PIs), the people nominally in charge of the project. They are only nominally in charge because the very proposition of the project application already means that they are out of their depth: they cannot possibly deliver on what they promised for the money.
To make things worse, the hierarchical structure of the academic system means that aspiring researchers often need the money and reputation from reeling in a big research grant to be eligible for a permanent position at a university. This hierarchical structure also means that they will likely not be the ones actually responsible for delivering on their promise: They will instead hire postdoctoral researchers and PhD students.
Postdocs and PhDs often face the same problem that PIs faced earlier on: The positions they are applying for in these projects often don’t match their skills and knowledge. This is unsurprising: If professors haven’t caught up with the research fads of the day, how should junior researchers? In any case, they need a job and are constantly told that they don’t have any alternative, so they should apply anyway. In the job interview, like in the funding call before it, both sides try to make themselves believe that the other side is a perfect match for them, out of a (perceived) lack of alternatives.3
While the PI is busy with other duties, usually related to their new role as a tenured senior academic, the junior researchers face a proposition that is impossible to deliver. How should they? The PI who originally drafted the application might remember the difference between fact and fiction in the text he sent to funders or know which of the promises he actually intended to deliver on (he usually doesn’t, more below).
Yet an inexperienced junior scholar has way less resources to tell substance from sauce. She would need her superiors to help with sorting them out, relieving her from the duty of delivering on an impossible promise or at least finding a creative way of making the promise a little less impossible. But the PI has already relinquished responsibility, both for the junior scholar and the project, and there is no one else in sight to take both off her.
Let’s be very clear: This is academic abuse. It is abuse because it overwhelms someone who is largely defenseless and it is abuse because all the instances above the junior scholar gain disproportionally from this arrangement.
At this point we have reached the end of a long line of what Dan Davies (2024) so aptly calls “accountability sinks”: there is no drain through which to pass the buck to someone else further down. Indeed this one-way street of accountability is what makes this relationship between PI and junior researcher a power relation, much more so than between the PI, the funding agency and other actors like reviewers or publishers, where responsibility and the role of a sink is passed back and forth several times over. The junior scholar is the final sink; a shaky multi-million Euro project stands and falls with her performance.
Next to the card game reference of “passing the buck”, which also exists in other languages, English has the nice metaphor of “kicking the can down the road”, which mostly just means deferring an individual problem. Applying it to Engzell problems might seem stretched as those imply handing one’s problem to someone else to deal with this. Yet often that someone else is you yourself, only six months or four years in the future. Indeed that is a way out for the junior scholar: just dump it on your future self, she’ll find a way to figure out how to solve a problem too big for you.
But that’s also what the PI and the funding agency do, because at some point the project will be evaluated—after all, research grants aren’t gifts, they are loans, just ones that have to be repaid in a different coin. At this point the latest, the PI will be reminded that this project is indeed his and that there might be repercussions for him for not publishing enough papers or conducting enough research.
Similarly, the funding agency might ask (at least internally) why this problem wasn’t properly vetted beforehand, when many of the problems could already have been anticipated. The easiest way to answer that question, however, is to increase pressure on PIs to deliver even impossible results, just like the easiest solution for PIs for making the problem go away is to increase pressure on the junior researchers to lean in even harder.
What characterises the trajectory of the buck and the can is that at any point of its travel it is likely to end up with someone less equipped to deal with it than at its current point. It’s constantly increasing in relative size until it’s about to explode: There are countless projects like the ones described by Engzell that stand out by a long period of frustration only apparent to the postdocs and PhDs working on them before this frustration turns into stress and panic when the PI and the funding agency realise the extent of the mess they made, only to blame the next person in line and trying to force them to make the problem go away.
You can easily imagine what working in such a job feels like during those bouts of stress, anxiety, yelling and passing the blame. Here the implicit abuse becomes explicit. And perhaps you can picture what that means during a time when many other things important for junior professionals—future jobs, relationships, children—are both dangling and dependent on their job success.
No wonder depression, anxiety and substance abuse are rife in academia. To see PhD students joking about their life on social meadia is painful to watch. They are trying to take their hardship and abuse with humour. Yet in the process they are often just degrading themselves further by taking the blame when it’s other people’s fault.
Engzell problems or Engzell syndrome?
Engzell’s posts focus on third-party funded projects. But it would be wrong to think that Engzell problems are limited to these cases or indeed academia. They emerge whenever and whereever people who think they have few alternative options see a way—no matter how temporary—out of their problems by overpromising, cashing in and then passing on their liabilities to someone else.
Academia is rife with such opportunities, and we might well call it a hotbed for Engzell problems. I will give you another example that is particularly interesting: that of theme or special issues in scientific journals. These special issues are often outsourced by the actual editors of the journal to one or more guest editors, who will be (co-)responsible for recruiting contributors, editingsubmissions and commissioning peer-reviewers.4
The assumption behind this is that these issues are held together by a common theme, often timely and incredibly relevant, for which there is both a need in the scientific community and a large number of ongoing work that the theme speaks to. The reality is that the theme often sounds intriguing but is not thought through; that guest editors hope the journal editors and contributors will help them figure out what exactly it means, and vice versa.
The situation is reminiscent of that of calls for project applications, only that there is a “buck triangle” in which each party variously serves as giver and recipient of bucks to the others. The decision to do a theme issue in the first place relieves the journal editors of some of their editorial work. A full issue! Timely! Internally coherent! And, above all, almost for free: someone else will handle the buld of the work. The buck is passed to the guest editors.
But not so fast: At the same time, the guest editors have passed their buck to the journal editors: A scientific publication! Peer-reviewed! Fast-tracked! Handled by professionals! At the low, low price of a daring extended abstract and perhaps a conference panel. Surely the journal editors will make sure that this project lives up to the highest standards, all there is to do is a few e-mails to write.
Since this obviously wouldn’t work out, another party is needed: People who are in need of directions and under pressure to publish, i.e., people working in projects as the ones described above. You have a timely topic? And some data you can’t quite figure out what to make of? Don’t worry, this special issue will get them both published and polished. Surely the guest editors will give you substantial feedback and helpful advice for how to develop your manuscript? Certainly the journal will accept it because someone has told them it’s essential to the theme issue?
These mutually agreed misunderstandings would immediately collapse if it weren’t for what Pierre Bourdieu (2008) has described as “the work of time”: the ability of people and collectives to shift around temporal horizons, allowing them to selectively take into account certain actions and consequences while blissfully ignoring others. Here the work of time consists in the certitude, or better hope, that all these contradictions will clear up with time: I don’t quite know how to deliver what I sketch out in the short description of my article, but surely it will be easier once I have to expand it to fifteen pages of text. (Reader, it won’t.)
On the way to impending disaster, other people will be drafted into this particular Engzell problem: peer reviewers. Their role is slightly different, because they are passed a can and will usually kick it straight back to where it came from, but the can will still expand in the process. An unfinished but dressed-up manuscript, often handed in a month past the original deadline, is sent to peer reviewers by the triangle of authors, guest editors and journal editors in hope that they will finally tell them what to make of this and turn it into a proper article. Of course they won’t, for the same reason that junior scholars can’t turn a PI’s sauced-up grant application into a working research project: They are usually less equipped to judge what is good and bad in such a text, to tell sauce from substance.
The best the peer reviewers can do is to send back the manuscript with an educated guess about what in that text is of merit and warrants publication. In the second-worst case, that would mean a rejection, since a text no one can tell what it is supposed to say will likely not take shape before the publication deadline. In the worst case, the reviewers will read some substance into the sauce of the text and encourage authors to expand on and develop it, in which case the authors suddenly find themselves in the role of a junior scholar: it is the peer reviewers who ultimately get to decide if their work will make it into publication. They thus have a large incentive to follow the peer reviewer’s recommendations but usually lack the means to do so: there was, after all, not substance behind all that sauce.
But the journal and the guest editors aren’t out of the picture yet: The process invariably leads to a wild mix of papers, takes and approaches that are more divergent and contradictory than coherent. In a certain regard, it may make the individual paers better but the whole issue worse. Who will be responsible for giving back coherence to the collection? You guessed it, the guest editors.
The editorial piece, in which the guest editors introduce all the pieces and tell you what they have in common, is the space in which all the loose ends will have to be tied together. These texts only work because they are a) usually short and b) often unrelated to the papers in the theme issue. For the same reason that it’s easier to make a movie exciting in a trailer than in 90 min, it is easier to entertain a bold idea on five than on fifteen or 300 pages.5
And since the editorial is usually the last piece drafted for a special issue, it can speak to the individual contributions but they do not get to (or have to) talk back. That is why an editorial can sell you a parking lot of used cars hastily fixed as an unheard-of opportunity for investment and mobility—a proposition that often falls apart the moment you take one of them for a test drive.
I don’t mean to insult or ridicule anyone who has partaken in a special issue in one role or another. In ten years, I have guest-edited four such theme issues as a guest editor and contributed to three as an author and another two as peer reviewer. Plus five edited volumes, which are are rarely any better, only that they get to skip the peer review process. You may take a guess in which of these roles I have followed the above model: It was all of them, and it still doesn’t fully prevent me from signing up to further special issues. The problem isn’t one of a lack of personal standards; it’s a systemic mishandling of responsibility.
There are many more examples I could name and describe: grading students, university management, course design, personal abuse by superiors, tenured positions that are undersupplied and thus a perfect environment for summoning additional Engzell problems. It’s not far-fetched to say that academia is in many ways suffering from Engzell syndrome and that the underlying causes are a) unclear, intransparent and fluctuating relations of responsibility and power and b) plenty of ways for passing around buckets and kicking cans down the road because academic work operates on long time frames and both produces and relies on a large amount of externalities.
I’m not naive enough to believe this is limited to academia, only naive enough to only have ever really worked in academia. To the extent that other industries and fields are similarly patient (just think venture capital and start-ups) you will find exactly the same patterns of escalating bullshit, sacrificing more vulnerable people to save your own arse, lying into one’s own pocket, wishful thinking and an essentially flat learning curve.
The question therefore is: How can we at least address the last point so that the future will bring less Engzell problems, not more?
Distributed responsibility
The key argument that I hope you will be taking away from this post is that academia (and other work contexts) have a very bad way of distributing responsibility. As long as we keep using them, we should take seriously the labels of “senior” and “junior” scholar: A senior scholar is a grownup, a junior scholar can be likened to a child or adolescent.
As a grownup, a senior scholar gets to enjoy privileges that a PhD student doesn’t, but in turn she also has more responsibilities. Likewise, a junior scholar is not the one who should be handling a flawed research project; this is the very reason most funding agencies do not give him the opportunity to submit grants in the first place.
The simplest definition of a grownup is a person who can take responsibility for themselves and for others. A grownup can catch and dismantle a rolling can about to explode instead of kicking it down to someone else. What we are seeing in academia, fuelled by the current funding system and only half a hierarchy (one of privileges but not of parallel duties) is the opposite: supposed grownups who cannot even take responsibility for their own actions and instead roam the academy to find someone else to do it for them.
They often end up with people even less equipped to handle their problem, if only because these people are also badly equipped to say no.6 If a good academic system produces responsibility, the current one tends to circulate liability and encourage irresponsibility.
Per Engzell rightly points to the funders and their complicity in this system. Much of what happens is made worse by the way project formats are designed, calls are issues and money is handed out by funding agencies, in Europe and everywhere else. ERC Grants, perhaps the most prestigeous funding line in the world with a volume of €1.5 to €2.5, do a lot of damage, because across Europe they serve as a golden ticket to tenure on the one hand and require that you employ at least two two three other people to spend that money. They encourage you to overpromise and make it very likely for you—and much more importantly, your staff—to underdeliver if you are not careful about it from the very start.
Other formats are even worse. The German Research Foundation (DFG) has a number of grant schemes that explicitly require you to hire a disproportionate number of PhD students, by definition the weakest links in this system and the people most likely to be crushed by their superiors’ rejected responsibilities. The same goes for many other funding bodies, whose implicit rules or grant sizes preclude hiring a postdoc.
But why even hire someone else in the first place? Engzell is right that there is a warped understanding of “team” at the heart of this, namely a collective of people at the whims of one person and their responsible behaviour. Smaller projects that only fund one researcher but distribute responsibility for supervision, support and controlling to a collective should be the logical inversion:7
But should we really wait for the funders to change their philosophy? Would this get rid of all the other Engzell problems in academia, such as special issues? Making Engzell problems someone else’s problem would be a convenient way of passing the buck once more. We also need to ask what is behind PIs’, guest editors’, referees’, authors’ and other academics’ inclination not to live up to their responsibility.
That PIs write their project proposals “in a weekend” gives us some hints as to where the problem lies: If people had a year to design a solid project with reasonable hypotheses and a workable research programme, their applications would be less likely to blow up further down the road. The longer you plan something, the more feasible it becomes when you start to implement it. The same goes for the time a reviewer invests into giving substantial feedback on a draft.
PI time (just as that of authors, reviewers and editors) is a serious bottleneck that has to be addressed. Proposals written on weekends spell disaster for everyone involved, only disproportionally so. Hence why people need to be given proper time to figure out what they want to do (together with others) before they do it.
Yet we are still very high up the road for trying to stop the can from being kicked down. It is likely that in giving more time to PIs and professors we will make Engzell problems proliferate because they will simply take on more responsibilities, simply to pass them down onto others who will be crushed in the process. This is why we should perhaps best start at the end of the buck line: with the junior scholars who end up with oversized responsibilities and undersized capacities.
The task is very easy to formulate at this point: We can safely increase their capacities because they are not at risk of passing their responsibilities down the academic hierarchy anymore. One way of giving capacities to PhD students and early postdocs is to provide them with material resources: longer contracts, higher wages, more freedoms for how to define and conduct their work in someone else’s project.
But the most important change we could make to their situation is giving them the ability to default on their debt: to rightly say “this problem is too big for me and it was not mine to begin with”. There are good rules in financial regulation about the appropriate size of liabilities and the role of responsibility for both creditor and debtor. The moment you default on a bad loan, it returns to being your bank’s problem. It is not far-fetched to think about something similar for transferring duties and responsibilities onto others.
Of course that brings us back to the system and waiting for someone else to change the rules for defaulting. But what people need to realise on all steps of the buck ladder is that defaulting is always possible, even now. The reason they shy away from it is that they fear the current costs of defaulting, of being punished for rejecting oversized responsibilities dumped onto them.
This is why funding agencies do not say no to bad applications, why PIs don’t say no to ill-suited applicants, why journals don’t refuse half-baked papers and theme issues: It would imply a loss of money, publication, reputation, time. But as much sympathy as I have for any of these actors, as little patience do I have for someone who keeps dragging others into their misery. If you are old enough to write an article, a project application, a funding call or a call for papers, you should also be old enough to think for 5 min what this implies for others.8
Maybe bailing out funding agencies, professors or journal editors is indeed too much to ask. But I don’t think bailing out struggling PhD students and postdocs is: At the very least we can remind them that quitting a project that should not exist (and certainly does not exist for their sake) is neither a crime nor a failure. Nor is it the end of the world, even it may be the end of an academic career.
Whether we can offer them material alternatives or not: We need to encourage them to pursue their interests in the same breath that we should encourage them to refuse to solve the problems that other have dumped on them.9 Let the grownups sort out their own mess.
References
Bourdieu, P. (2008). The Work of Time. In The Logic of Practice (pp. 98–111). Stanford University Press.
Davies, D. (2024). The Unaccountability Machine: Why Big Systems Make Terrible Decisions – and How the World Lost Its Mind. Profile Books.
Perhaps this is indicative of quantitative sociologists like Engzell being much better than qualitative ones like me at thinking reflexively about the academic system, despite their reputation to the contrary. But I admit the sample size is indeed to small to support or reject that hypothesis.
The agency has good reasons for this: First of all, there are no good routine ways for handing back money to the taxpayer that is already allocated somewhere in the public system. Second, it would imply that the agency receives less money in the following round, meaning it could fund less projects even if supply of topics and experts had caught up with demand in the mean time.
People who have been in the hiring position will object that this lack of alternatives is anything but “perceived”. This is true because there are fixed deadlines from the agency’s side for spending grants: There simply are no perfect candidates on the job market and if you’re not hiring this person, you’re likely not hiring anyone else. It is also wrong, however, because it implies that not hiring anyone, scouting the market before you submit a project or renegotiating with the funding agency are not decisions you could make. They absolutely are.
There is a lot of talk currently about special issues as a predatory scheme by dubious publishers. I hope I can contribute a little to the understanding of its non-monetary economy here.
The way I describe it here there are no ultimate sinks but of course there are (co-authors, colleagues, family members, rjected submissions). They are often just less localised than junior researchers in third-party projects.
Once you think about Davies’s accountability sink, this shouldn’t come as a surprise: bucks are passed on and cans are kicked down the road until they reach a dead end, simply because the last recipients are unable to force them onto someone else. This is usually where they do most damage, because the people who have the least power to force others to comply with their needs also have the least power to handle other people’s problems.
Ironically the European funding landscape already knows such projects! DFG has two lines that work exactly like this (Heisenberg Programme and Individual Research Grant) as does the Spanish Ministry of Research with its Ramon y Cajal fellowship. I have yet to understand what allows these programmes to coexist with much more problematic team/PI-oriented ones. What is more, the ERC isn’t forcing you to hire PhDs and postdocs foryour project! You can as well just buy your tenured colleagues sabbaticals.
This also applies to me—there are things I wished I had done differently in the past and things I wish I were doing differently now. I certainly appreciate other people’s (unsolicited) help with them but I cannot demand that someone else be parachuted in to solve them when I am old and smart enough to do it myself.
What this means for me personally is that if I learn you are abusing your dependents in whatever way I will work my arse off to offer them a better alternative instead of being stuck with you. I can only hope that others will do onto me like I would do onto them.