I first wrote these reflections in February 2021, almost a year after the beginning of the Covid-19 pandemic. At the time, I had started to notice with some frequency a particularly aggressive tone in academic discussions: papers in which criticism came dangerously close to personal attack, unnecessarily harsh peer-review reports, and discussions in which disagreement seemed almost to require some form of hostility.

It is possible that the particular conditions of the pandemic made these attitudes more visible, or perhaps simply more difficult to tolerate. Yet the problem certainly predated the pandemic, and it led me to reflect on a concept that is widely discussed in organisational studies but much less present in our conversations about the way academia works: psychological safety.

Amy Edmondson defines psychological safety as the shared belief that a group represents an environment in which it is possible to take interpersonal risks. In simpler terms, it means being able to ask questions, acknowledge a mistake, express a doubt, or voice disagreement without fearing humiliation or inappropriate professional consequences.

Applying this concept to academia, however, requires a certain degree of caution, because research lives precisely through criticism and confrontation. An argument may be wrong, an interpretation unconvincing, the evidence adduced insufficient, and a method inadequate. All of this must be stated clearly, because abandoning criticism would mean abandoning one of the fundamental conditions of scholarly work. The question concerns, rather, the way in which criticism is formulated and, above all, the environment in which it takes place.

This problem is particularly relevant in academia because doing research means, at least to some extent, proposing interpretations that are not yet accepted. We are expected to question established positions, reconsider existing interpretations, and sometimes advance hypotheses that may initially appear implausible. Part of scholarly work consists precisely in taking this risk. The situation becomes even more delicate when we consider the early stages of an academic career. On the one hand, we ask younger researchers to be original and to produce something new; on the other hand, their work is constantly evaluated by people who generally possess far greater academic and institutional authority than they do. Therefore, a healthy academic environment should allow disagreement without turning it into a personal or professional risk.

From this perspective, I think that two aspects are particularly important. The first concerns constructive criticism. When we criticise an argument, we should try to explain clearly where the problem lies and why that problem matters, offering, whenever possible, some indication of how the argument might be reconsidered. In teaching, this principle seems rather obvious to us: simply telling a student that something is wrong is rarely very useful, because what matters is making the reason for the error intelligible and allowing the student to reformulate their position more effectively.

I do not see why this principle should cease to apply when we move from our relations with students to relations among scholars. Peer review, book reviews, conference discussions, supervision, and even ordinary conversations among colleagues all participate, in different ways, in the same intellectual process. Their purpose should be to improve arguments and, through this exchange, to understand more clearly what we are studying. Aggressiveness adds very little to this process and often ends up simply increasing the personal cost of being criticised.

The second aspect is related to divergent thinking. The expression is associated with J. P. Guilford’s work on creativity and, in very general terms, refers to the capacity to explore different solutions without immediately converging on an already established answer. Something analogous is essential in research, because we need to remain willing to take seriously interpretations that differ from our own, including those that may create difficulties for positions we have defended for many years. And this is probably where the issue becomes more complicated.

Imagine that scholar A develops, early in their career, a particular interpretation of x. The interpretation is well received, and A begins to use it in order to interpret y and z as well. Over the years, A publishes articles and books based on the same general framework, to the point at which the original interpretation of x becomes one of the elements supporting a considerable part of their work.

At a certain point, scholar B proposes a profoundly different interpretation of z. If B is right, the consequences do not necessarily concern z alone, because A’s interpretation of y may also need to be reconsidered and, moving back through the structure of the argument, even the original interpretation of x may eventually become problematic.

Suppose, moreover, that B’s interpretation is in fact better, because it explains the sources more effectively, resolves some of the problems left open by A, and offers a more coherent reconstruction overall.

Now imagine that A finds themselves evaluating B’s work as the referee of an article, the reviewer of a book, the respondent after a conference paper, or a member of a committee deciding on B’s career.

At this point, the problem becomes evident. In order to evaluate B’s proposal properly, A may have to recognise that part of an intellectual construction developed over decades needs to be revised. Disagreement thus acquires a dimension that is no longer exclusively intellectual, because it inevitably also involves biographical and professional elements. Naturally, in circumstances of this kind we rely on the scholarly integrity of researchers. Yet I think there is an even more fundamental issue here, concerning the way in which we conceive of our own work.

Our interpretations always remain provisional. They may be influential, widely accepted, and capable of sustaining years of research, but none of this prevents them from turning out to be wrong or from requiring substantial revision in the light of new evidence or better arguments. In this sense, psychological safety is closely connected with epistemic humility. B should be able to propose an alternative interpretation without having to calculate in advance whether disagreeing with A may have consequences for their career. At the same time, A should be able to modify their own position without this being perceived as a personal or professional defeat. Indeed, changing one’s mind because another interpretation explains the sources better should be regarded as one of the most ordinary outcomes of academic work.

Therefore, constructive criticism and openness to alternative interpretations concern something deeper than simple courtesy in academic relations, because they directly affect the conditions under which knowledge is produced. When a researcher fears ridicule, possible retaliation, forms of exclusion, or consequences for their career, they inevitably become more cautious. Gradually, they learn which questions can be asked without too much risk, which positions it may be preferable not to challenge, and which disagreements may carry an excessive cost. Research continues under these conditions, but the range of positions that can genuinely be explored tends to become narrower.

Psychological safety, naturally, cannot by itself solve problems that also depend on the profound asymmetries of power that characterise academic life. It can, however, contribute to creating an environment in which it is possible to disagree, acknowledge an error, and revise an interpretation without all of this immediately becoming a personal matter.

And this is precisely the point that seems most important to me. Academia cannot function without criticism, but criticism makes sense insofar as it remains part of a shared attempt to understand something better. When disagreement instead becomes intimidation or turns into a professional risk, one of the conditions that make research itself possible begins to disappear.

Reference

Edmondson, Amy C. 1999. “Psychological Safety and Learning Behavior in Work Teams.” Administrative Science Quarterly 44 (2): 350–383.

Edmondson, Amy C. 2018. The Fearless Organization: Creating Psychological Safety in the Workplace for Learning, Innovation, and Growth. Hoboken, NJ: Wiley.

Guilford, J. P. 1950. “Creativity.” American Psychologist 5: 444–454.

©️Nicola Polloni
Latest update: September 2026