Last week was the most memorable week of my scientific career. Accompanied by family, friends and colleagues, I was honoured with the award of a Nobel Prize in an unforgettable ceremony and banquet. That same week, I also chose to express highly critical views about deficiencies I perceive in the system scientists use for publishing and rewarding scientific research, for which I was both attacked and praised.
My remarks focused on the power of certain journals, which I refer to as luxury journals, that have distorted how science and scientists operate.
I was not surprised by the range of opinions my comments provoked, but I have been impressed by their quantity. The evidence that the scientific community wants and needs this discussion could not be stronger. I write this to respond to some of the criticisms, to expand on some points I made, and to suggest some next steps.
It is understandable that some see hypocrisy in my criticism of a system that has served my own career well. I have published extensively in Nature, Cell and Science. I have now, of course, won the Nobel Prize. It is therefore easy, some have said, for me to voice my concerns. But that, in some ways, is exactly the point. I am saying what many others believe but feel they cannot say, because they fear their careers might be damaged.
Yet others have spoken out. I recognise that I am not the only person to criticise luxury journals and an academic reward system that relies too much on them. I applaud those who reached this view long before me. I accept that I could have spoken out earlier in my career, but the Nobel Prize has afforded me a platform from which to speak loudly. The charge of hypocrisy would be fair were I still submitting my own research work to luxury journals. I see none in speaking out, while doing as I say.
It has also been pointed out that I have a conflict of interest. I have edited a major subscription journal (Proceedings of the National Academy of Sciences, PNAS), and now edit an open access one (eLife), both of which compete with the luxury journals in different ways. But I have long held a negative view of the role of impact factors, an imperfect measure of the importance of a journal and its content, and shared my views with the staff and editorial board members who served with me during my term as editor-in-chief of PNAS. The problems with the scientific rewards system extend beyond the competition among the journals.
I have also been clear as to the extent of this conflict. As was declared in the Guardian article, I am leading a challenge to the luxury journals as editor-in-chief of eLife. I am doing this work because I believe that journals need to be radically improved and we have the means to achieve this. Though I draw an employee’s salary, I have no wider financial stake in eLife’s success, and I have always been entirely open about my role. I believe my argument would be weaker if I were not also attempting to change the system in some ways.
I understand, too, concerns that my stance will have career implications for junior colleagues in my lab. I shared these concerns, which is why I discussed the issue with them more than two years ago, when I took on the editorship of eLife. My colleagues agreed then, as they do now, that we should be challenging the big journals, and that papers we would once have submitted to Science, Cell and Nature should go elsewhere.
I am deeply committed to developing the careers of younger scientists I work with – that, indeed, is a major motivation for my argument. I do not want them to have to play a system where the artificial scarcity of prestige publications makes recognition and advancement such a lottery. It is gratifying that several of my lab colleagues have publicly supported me.
My purpose in avoiding luxury journals, other than being seen to walk the walk, is not necessarily to prompt others to do the same. Rather it is to prompt reflection among researchers, institutions and funders, who are in a position to limit the poor incentives that the reliance on luxury journals has created. I want scientists and administrators, especially those involved in funding, promotion, recruitment and tenure, to think hard about the influence that publishing decisions and research assessments have. That is the way we will drive change.
One of the most important changes we need is for journals to exploit the advantages of publishing online rather than in print. Too many journals remain wedded to print, artificially limiting the number of papers they accept. This made sense when journals were constrained by page counts, but makes much less in a digital world. It makes journals more selective than they need to be, driving extreme competition for space that is good for subscription businesses but bad for science.
Intense competition for space in key journals means that the editorial process often involves multiple rounds of revision, review and resubmission, causing long delays in publication. Additional experimental data and information are often demanded by reviewers who might later, as authors, be competing for space in the same journals. Much of this data is then relegated to supplementary appendices. The experience can be highly dispiriting for researchers.
I see a solution in open-access journals. They generally cover their costs upfront, for example using a business model whereby a fee is levied for publication. This model is more suited to the digital medium: all the work that meets the editorial criteria for the journal can be published, and it can be made freely available to everyone. As high quality science grows, so can the number of articles published. This, more than anything, is what makes eLife not like the luxury journals: it is selective, but will publish everything that meets the editors’ standards. There is no picking and choosing to meet a quota. It also tries to address some of the other issues listed above, for example using a much more efficient editorial process. And when eLife receives an impact factor, it will not be promoted.
Journals, however, are only one half of this equation. We also need to address the demand for luxury journals, from researchers themselves and from the institutions that use them to judge scientific quality. We need to discuss what researchers, universities and funders can do to remove the incentives that make it rational to publish under the biggest brands. I would like to suggest four places to start.
1) Academics who serve a role in research assessment could shun all use of journal names and impact factors as a surrogate measure of quality. New practices and processes must be devised and shared so that we can rapidly move forward. My Berkeley colleague Michael Eisen has added an important point: we must speak up in appointment and funding committees when we hear others use journal names this way. Here we need peer pressure as much as we need peer review.
2) Researchers applying for positions, funding and tenure should avoid any mention of impact factors in their applications or CVs. Article metrics might have a role to play, but narrative explanations of research significance and accomplishments would be more helpful.
3) Funders, universities and other institutions should make it clear to their review committees that journal brand cannot be used as a proxy for scientific quality. If reviewers object, they should find different reviewers.
4) Many of us serve as editors or editorial board members of journals – and we could insist that the publishers of these journals stop promoting impact factors. Instead, the journals could emphasise the other valuable services they provide to authors and readers to promote their worth to the community.
No doubt others will come up with bigger and better ideas to move us away from the problems that we currently face. If I have helped to spark a discussion, I’m delighted. Now we have to turn our attention to action.