PSR: You discuss coping with problematic articles published within academia. How do you define a “problematic article” and how is that related to academic freedom?
Kristian Skrede Gleditsch: The article mainly discusses articles that people generate controversy, where people flag specific issues as problematic. This could range from the topic itself, the data used, the analysis, or the inferences drawn from data or analysis. There can obviously be main problems in articles or research that are ignored, but my focus here was on how to best deal with controversies, inspired by the debate on the article on “The Case for Colonialism”. Academic freedom is normally defined as the freedom of researchers to pursue research without interference. Research that generates controversy can face additional barriers to publication or open discussion.
What are major myths or misperceptions about common ways of managing controversial publications you’d highlight? How can one avoid political biases during the evaluation process?
Science is always to some degree uncertain, incremental and gradually revised, and should be open to debate. However, in practice, it is often difficult to publish comments on articles, as many journals are reluctant to consider comments on published work. Dialogues are often more informative than monologues, but journals are skewed towards the latter. There is a strong status quo bias, where published work is often left uncontested, and important questions or qualifications often become sidelined. There are many biases that could affect the evaluation process, and I am not convinced that if political biases are more problematic than other biases, even if they are likely to generate more heat. Ultimately, we can only call for all of us as reviewers and editors to try to consider not just whether they agree with or are convinced by something, but whether it is a debate that is worth making public.
You argue that “calling for retraction for articles that one disagrees with is clearly problematic on grounds of academic freedom, commonly understood as the right of researchers to have freedom in conducting their research and seeking to publish the results”. What are better approaches to such questions?
Retraction may be appropriate for clear cases of misconduct or fabrication, but it is not an appropriate response to resolving disagreements and risk suppression of research. Allowing for more debate post-publication would help clarify the sources of disagreement and allow for others to make up their minds about the merits of a contending argument. Ultimately post-publication debate can also allow for better science and advancing knowledge.
Are traditional control mechanisms such as peer-review and editorial judgment a sufficient safeguard for academic publishing?
Peer review is a valuable way to evaluate scientific research, but it is not an infallible guide to “truth” or scientific insights. Peer review can both fail to detect important problems, and it may recommend against important contributions that deviate from conventional approaches. For example, Akerlof’s article on the “Markets for Lemons” on the problems in markets with incomplete information was rejected three times before it was published, and Ioannidis in an important article argues that most published findings in medicine are more likely to be wrong. I think it is a mistake to focus too much on keeping out material that we may later find to be incorrect. Sometimes we are unlikely to find out unless something was published, and learning from the past is a key part of scientific progress.
You say, that it is not possible to divorce one’s own political views when conducting research. What is your view on a more complex level of this issue: scientists’ engagement in public/political life?
I think it is entirely legitimate for scientists to engage in public and political life, and scientists have much to contribute to policy debates. But scientists should be careful to claim privileges because of their position. The policy is ultimately a question of objectives, and determining what our objectives ought to often lie outside scientific knowledge per se. For example, science may tell us passive investment on average will provide a higher return than active management, but we may wish for public investment to support other objectives than just maximizing return (i.e., “ethical investments”). Scientists should separate value judgements from claims about means-end relationships.
You emphasise the significance of a post-publication debate. The other pole of a post-publication article’s trajectory can be a post-publication lawsuit. As, for instance, in this case, where two leading Holocaust historians were accused of defamation. Should courts be places for validating scientific evidence?
I do not think courts are a suitable place for making decisions on scientific debates. It is hard to think of examples where such lawsuits have been helpful, and they appear to most often been dismissed or withdrawn. In 2017, for example, an academic sued the US National Academy of Sciences over a study the criticized his claims over green energy, demanding more than $10 million in damages and retraction, but he ultimately withdrew the lawsuit and may be liable to pay legal costs.
Article: Skrede Gleditsch K. (2021), Houston, We Have a Problem: Enhancing Academic Freedom and Transparency in Publishing Through Post-Publication Debate, Political Studies Review 2021, Vol. 19(3) 428–434
Questions and production
Dr Eliza Kania, PSR/Brunel University London