PhD students contribute a vast proportion of the work that constitutes scientific research, be it in data collection, data analysis, development and refinement of methods, or the production of the main vehicle of research communication, peer-reviewed publications. Peer-reviewed publications are often the primary measure used to assess PhD students’ achievements and employability. Yet although PhD students make up the scientific work force of the future, there is not much in the way of open debate among all those with a stake in PhD training as to whether the current system to train and assess PhD students is fit for purpose.

In this issue, we feature a collection (https://www.nature.com/collections/phd-publishing) of World Views that debate the merits and shortcomings of focusing on publications as the primary output of PhD training and on the wider context in which this training takes place. These contributions were selected among numerous excellent suggestions we received in response to an open call, encouraging submissions not only from PhD students themselves, but also postdoctoral researchers, tenured faculty and funders across the globe. Many remarkable contributions are further hosted in the form of blog posts on the Behavioural and Social Science community page (https://socialsciences.nature.com/channels/2140-is-it-publish-or-perish).

The collection brings together a wide range of diverse opinions. Unsurprisingly, it is not a simple consensus view that emerges, but a multifaceted picture that brings into relief the fact that what value publications have is highly dependent on the context. For instance, the preoccupations of PhD students in Germany or the UK—where it’s mostly publish or perish—are not the same as those of students in Brazil or Japan, where it may be publish but perish regardless. Irrespective of geography, however, common themes surface across the contributions: science is not a level playing field and publication pressures may do more harm than good, both to students and the scientific enterprise.

Peer-reviewed publications present a form of hard currency that is widely understood and applicable to many disciplines. As a hard currency, evaluated by independent experts, publications can contribute to meritocracy and equality, especially in nepotistic academic systems. As a point in case, one author explains how the introduction of publication requirements for PhD students in Vietnam now allows female PhD students to demonstrate that their competency equals that of their male peers, leading to greater recognition of their achievements (https://doi.org/10.1038/s41562-019-0667-6). There are other positive aspects, too: the practice of writing and disseminating scientific results and novel ideas presents a learning opportunity that allows development of transferable skills, and each publication is a landmark that provides structure for the many years of PhD candidacy.

However, several of our authors point out that in many academic systems, publications fall short of promoting meritocracy.

The external conditions under which trainees pursue their PhDs, such as a nations’ research spending and funds for student support, a field’s standards in what is publishable, an institution’s working environment and facilities, differ widely. But science is an international endeavour, and placing the same measure on candidates from vastly different backgrounds neglects this issue. Within the same context, a candidate’s individual circumstances, such as immigration status and minority background, create systemic differences that render even the most narrowly defined playing field, such as the same research group, uneven.

The World Views and blogposts go beyond discussing the issue of fairness and question whether this way of assessing PhD students is indeed fit for purpose. Does it prepare PhD students for their future careers? Only if that career is within academia and only in countries with significant economic investment in science. The majority of PhD students, however, leave academia, and their publication-focused PhD training does little to prepare them for anything other than the elusive professorship.

Many of our contributors say that the frenzied race to the longest publication list threatens the quality of the work they or their trainees do. Not only the quality of science but its potential to benefit humanity suffers if narrow and ill-chosen criteria are applied in the selection and promotion process. Powerful contributions in this collection stress the detrimental real-world consequences of current practices, particularly for the Global South. As long as promotion is contingent on a specific type of publication in a select range of journals, this creates knock-on effects by which important research is never conducted or is not communicated to affected communities, as outreach to the public is deprioritized.

The conditions of PhD support, training and assessment differ. Nevertheless, the majority of PhD students who contributed to this collection either have cause to believe, or know with certainty, that they are assessed predominantly on a single dimension: the number of publications on their CVs. Using a hard currency to determine merit is preferable to a system relying on favouritism and personal connections; nevertheless, the use of publications as a single metric fails PhD students and can be detrimental to science. Across the selection of World Views and blogposts, there are many proposals for fundamental first steps to bring about the system-wide changes, targeting not only assessment, but also support and training practices that are needed to achieve a meritocracy.