Summary

A Nature Comment by Leiden University rector Sarah de Rijcke proposes six principles for protecting academic freedom while recognising legitimate concerns such as research security and public accountability. It argues that independent enquiry is a practical condition for producing reliable, challengeable knowledge.

Universities should protect academic freedom not only as an abstract right, but as a working condition for producing knowledge that can be checked, challenged and corrected, argues Sarah de Rijcke in a Nature Comment published on 21 September 2026.

De Rijcke, rector magnificus of Leiden University, sets out six principles for governments, research funders and university leaders. Her central test is whether a decision affecting higher education strengthens independent scrutiny and the production of reliable knowledge, or shifts control towards politically acceptable conclusions.

Contents

Academic freedom as a quality-control system

The Comment places a university’s responsibility to produce reliable knowledge before the question of individual rights. Independent peer review, openness about methods and data, and the ability to correct earlier work all depend on researchers being able to pursue questions without instructions that determine the acceptable answer in advance.

Political or commercial pressure can influence the questions researchers ask, the methods they select and the outcomes they present. De Rijcke argues that measures described as protecting quality or integrity should therefore be judged by how they operate in practice. Measures that make methods and results easier to inspect can strengthen knowledge. Grant calls that favour predetermined answers, screening researchers for their views, or removing uncomfortable data reduce the scrutiny through which errors are found and corrected.

The same principle applies to teaching. When political instructions replace the testing of arguments against disciplinary standards, students are trained to adapt their thinking to approved answers rather than evaluate evidence and reasoning.

Why open-ended research needs protection

A second part of the framework concerns research whose value cannot be demonstrated at the start. De Rijcke points to the path from research on how bacteria defend themselves against viruses to CRISPR gene editing, and from decades of work on messenger RNA and lipid-nanoparticle delivery to COVID-19 vaccines. Linguistics and digital humanities also contributed to translation and search systems, while long-running epidemiological cohort studies supported public-health guidance on smoking, diet, workplace safety and environmental exposure.

These examples illustrate why narrowly defined relevance criteria can be restrictive. If researchers are required to specify immediate societal value, funding systems may favour safer and more incremental projects while excluding questions whose eventual importance is not yet visible.

The Comment links this issue to public policy. Governments seeking robust evidence need research systems capable of generating and testing knowledge independently. Reducing investment in fundamental research may lower spending in the short term, but can also reduce the range of expertise and options available when future health, environmental or technological problems arise.

Autonomy, democracy and research security

De Rijcke also treats universities as part of democratic infrastructure. They train professionals who can question official decisions, produce analysis that challenges government or corporate claims, and bring independent expertise into courts, public debate and the media. Institutional autonomy, in this view, means accountability to the public for the quality and integrity of knowledge rather than to a government for whether conclusions are politically acceptable.

The Comment uses developments in Hungary as an example of how changes to research funding, higher-education law and other institutions can reduce university independence. It also describes concerns raised in the Netherlands about budget cuts, commissioned research, proposed knowledge-security screening and intimidation of researchers in public debates.

Research security is presented as a legitimate responsibility rather than an alternative to academic freedom. Universities must address espionage, intellectual-property loss and the harmful use of sensitive knowledge. At the same time, security rules can narrow enquiry if they give excessive control to those managing access to data, infrastructure and international collaboration. The proposed approach is to design security measures and openness policies together, preserving necessary safeguards without making political or administrative control the centre of research decisions.

The result is a decision framework rather than a new regulatory standard. Its practical question is whether a proposed funding rule, accountability requirement, collaboration restriction or security measure improves independent scrutiny while addressing the public concern that prompted it.

Sources