Small is beautiful – and other thoughts about university governance

If one had to choose one reason for why things are not going well in academic life, the managerial, top-down style of governance that reigns in many universities would be a top candidate (with budget cuts as a close competitor). But what is a better way of running universities? For me, this is a question in which theoretical and practical-professional interests intersect.* I’ve long been a defender of workplace democracy, and since 2023, I’m on the board of a small faculty – so the question became: What does it mean for a faculty to be a democratic workplace? Especially if the official rules do not allow for, say, an election of the faculty board by the faculty members…

But the internal structures of small units are only one dimension of the problem. Another is how a university as a whole are governed. In my various jobs, and in conversations with many colleagues, I’ve seen and heard of many bad examples – but I’m looking for good ones! So, I’ll share some thoughts about university governance, to invite a discussion about what works and what doesn’t! Here is a list of ideas, loosely building on each other.

• At their core, universities should be self-governing bodies. This is how they have historically been run, and how some universities still function today. Of course, historically these self-governing bodies had most of the time been exclusionary along the usual lines of gender, class, nationality, religious affiliation, etc. But that need not be the case, and the principle of self-governance should not be thrown overboard but rather be made inclusive.

• The most important resource of a university are its people. Their intrinsic motivation to do good work, and to invest into their university (as opposed to investing in all kinds of other activities or outside networks), is one of the most important things a governance structure needs to preserve. Investment in a university (or a unit within it) has a public good structure: everyone is happy if others do it, but one might not want to do so much oneself, because it’s often more exciting to plan one’s next conference trip than to sit on a committee that looks through student evaluations. The best way to ensure such investment is probably that people feel they get a fair deal overall, and that their colleagues also do their bit, so it becomes a matter of reciprocity and “taking one for the team.” This, by the way, holds for administrative staff as much as for academics. And it requires that people can see how much others do (or understand why someone is doing less for a certain time, e.g. because of illness in the family). That has an impact on seize, to which I come back below.
• Decision-making structures about research, teaching, and impact should be kept as close as possible to those doing the actual work. You want representatives of all relevant groups in the relevant decision-making bodies. For example, the Dutch system of “student assessors” in faculty and university boards is an excellent way of keeping short lines with the student body. The argument for this principle comes from epistemic democratic theory (see e.g. Hélène Landemore’s account): those governing over a unit need to be close to those doing the actual work (or themselves being among them), so that they gain a deep understanding of the dynamics and constraints of the work on the ground. Having part-time, rotating management roles probably works best for keeping decision-making close to the actual practices. It allows organizational units to build their own research culture and to adapt formal processes, e.g. promotion criteria, to them. In this way, you can minimize (though never completely eradicate) the tensions between the “practices” (which realize value in research, teaching, and impact) and the “institutions” (the external measurements, policies, and incentives) (to use a distinction from MacIntyre**).
• There are different kinds of efficiency that can inform governance issues (this thought is inspired by Heath’s paper on different kinds of efficiency). There is the efficiency of “economies of scales” that pushes for processes to be done on large scale. But there is also what might be called the efficiency of “economies of trust”: short lines and trustful collaboration. The former pushes towards larger structures, the latter towards small, close-knit communities in which people know each other well. The art is to find the right balance between those. As a general principle, I find it plausible to aim at economies of scale for support structures, such as IT, and economies of trust for the primary processes of the university, i.e. research, teaching, and outreach, the actual work done day by day by the employees of the university.
• How big should such units be? Because I believe strongly in “economies of trust”, I’m a fan of “small is beautiful” (a slogan stolen from F.E. Schumacher). People should be able to know each other, to build trust, which requires personal connections. On the other hand, if units become too small, this creates other kinds of risks: too few people for multi-perspectival deliberation, or concentrations of power with insufficient counterpower. My experience in Groningen, with a faculty of around 60 staff members plus circa 30 PhD students, is very positive – this seems a good seize for a unit.***
• Lots of universities want interdisciplinarity and transdisciplinarity, and there seems to be an idée fixe that to create it, you best throw different disciplines into one organizational unit. I think this idea is ill-founded. The best kid of inter- and transdisciplinary research is rooted in strong disciplines. You may need to adjust other things, for example evaluation criteria, and you might want to create spaces, e.g. additional interdisciplinary unit or cross-cutting networks, to encourage cross-disciplinary collaborations. But such work will remain difficult and challenging, so you help it neither by making it a facile buzzword, nor by trying to conjure it up through forced marriages between units.
• If a university board claims that it cannot deal with a large number of small units (fair enough), the solution can be federations and rotating representation: several units, together, can be represented in certain central bodies. I don’t know what the ideal size and composition here is, but considerations from epistemic democracy suggest, again, that diversity of voices and perspectives is crucial. By the way, I strongly suspect that different units can also do many things traditionally done in hierarchical lines between themselves instead, on a horizontal level. For example, the heads of units could have their annual performance review in a peer system with other heads of units. Only in case of conflicts or problems need a case be escalated to a higher level or an external party, e.g. an ombudsperson.
• For specific tasks – whether substantive research projects, or things like developing a new partnership with an outside organization – committees can be put together with members from the different units. Here, the principle of lottocracy (as defended, for example, by Alexander Guerrero) might be useful: to draw a random but representative group from the whole population in question. I am not aware of whether this has been tried anywhere, but a lottocratic assembly at a university seems a very interesting experiment!
• Last but not least, however good or bad a governance structure may be, there is also the level of culture: the informal social norms that determine how processes actually go, and how people treat each other. Good governance structures can support, but never guarantee, a collaborative, trusting culture that is oriented towards the values of the academic community. Ill-conceived attempts at reforming the governance structures, however, can do incredible harm to such a culture. So, if you have units with a strong culture, with economies of trust and good records in teaching, research and impact, don’t undermine them – you can lose in months what has been built in years, and you will not get it back easily.

Those are some of my thoughts, and I’m interest in yours. What form of university governance is best for good academic practice? How much does governance matter compared to other factors (am I overestimating its importance compared to the idiosyncrasies of individual characters, maybe)? And what are successful democratic practices or good institutional devices that universities could learn from one another?

*) I should be fully open that my university has announced that it wants to restructure its governance for the social sciences and humanities sector, so this topic is closer to home than I would like it to be…

**) I am not convinced of all elements of MacIntyre’s thought, but I find the distinction between practices and institutions very useful. It is from After Virtue. A Study in Moral Theory (2nd ed.) (Notre Dame: Notre Dame University Press, 1984). To add a bit more background: MacIntyre argues that practices aim at a specific good internal to it, and as such offer a field in which the relevant virtues can be exhibited (pp. 148, 187). In practices, individuals are not motivated by external goods, e.g. money or prizes, but want to get better at the practices themselves (pp. 188-9). Institutions, in contrast, are the external shells within which practices are hosted, but in which people also strive for other goods, such as promotions or higher incomes, which are “characteristically objects of competition in which there must be losers as well as winner” (p. 190). Internal goods, for example excellence in an artistic practice or in the care for patients, are not competitive. Rather “their achievement is a good for the whole community who participate in the practice” (pp. 190-1).

***) There might be an exception to that rule, though: if you have narcissists in a unit, smallness can increase toxicity. Maybe the best way of dealing with such a situation is to dilute the narcissism in larger units, I don’t know. There is a broader question about how to keep narcissism out of academia, but that’s for another occasion.