Narcissus by Caravaggio depicts Narcissus gazing at his own reflection.

The motto for the American Association for the Advancement of Science (AAAS) is simple, elegant, and symmetrical: “Advancing science. Serving society.”

What I love about these four words is that they convey a duel role of the association. The AAAS aspires to improve our understanding of the world, but the ultimate goal of knowledge is to benefit society. Science is not just an end in of itself, but a means to a higher end — the improvement of human welfare. Most scientific societies have similar mission statements.

But what if that ultimate goal of science was being chipped away, eroded, such that science begins to turn inward and become an end in itself. According to a new study of the perceptions and behaviors of scientists, this is taking place. And the cause of scientific narcissism is the “publish or perish” culture that universities have shaped for their academics.

Hendrik van Dalen and Kène Henkens, researchers at Tilburg University in the Netherlands, report their findings in Intended and unintended consequences of a publish-or-perish culture: A worldwide survey which appeared online in the Journal of the American Society for Information Science and Technology on April 20th.

Surveying demographers — a multi-disciplinary group of researchers in fields covering sociology, epidemiology, geography, anthropology, economics, and history, among others — Van Dalen and Henkens report what many of us have heard anecdotally for years, that publication is valued above all other accomplishments and is rewarded at the expense of other academic responsibilities. They write:

Population scientists find publishing in international refereed journals and being cited by other scholars the most rewarding element in their job. Writing referee reports and making insights visible by writing articles for newspapers rank among the least appreciated elements of their work.

Scientists, however, are not fixed to their first place of employment, nor do they conform to a single behavioral model. Those who do not enjoy (or succeed) in the high-pressure publish-or-perish environments of research universities gravitate to institutions with less pressure to publish and the reverse is true as well. Concentrating ambitious researchers in elite research institutions, however, has its costs as competition requires researchers to publish more in order to keep up with one’s peers.

What is refreshing about this article is that van Dalen and Henkins provide historical and sociological context to the normative behavior of scientists. The publish-or-perish model and our reliance on publication metrics, they write, is not altogether bad–as it helps to identify and reward scientists based on merit and not on favoritism and nepotism–only that it comes with consequences for the role of science in society. They write:

The results presented in this article show that the publish-or-perish culture can have both beneficial and detrimental effects. . . . A consensus can be detected on the benefits of publications, as they improve the upward mobility of scientists. However, the detrimental effects revealed are the widening gap between science and policy, and especially for those scholars working outside the United States the incentive to publish in peer-reviewed journals is perceived to discourage the production of local knowledge.

As economist Paula Stephan described in her interview with Kent Anderson, public policy does not have to be radical to change the behavior of scientists. Shifting the emphasis on rewarding scientists for quality rather than quantity of publication, for example, by limiting grant applicants to their top five publications, would help stem “salami-style” publication behavior. And Stephan does not advocate against abandoning publication and citation metrics, only limiting their importance.

When we wish to change a complicated system, we should avoid limiting our conversation to grandiose policy shifts, as they often arrive with unintended consequences. Sometimes just a little nudge will do the trick.

Phil Davis

Phil Davis

Phil Davis is a publishing consultant specializing in the statistical analysis of citation, readership, publication and survey data. He has a Ph.D. in science communication from Cornell University (2010), extensive experience as a science librarian (1995-2006) and was trained as a life scientist.


10 Thoughts on "Publish-or-Perish Culture Promotes Scientific Narcissism"

It is nice to know that the way the scientific world is, is “not altogether bad.” More seriosly, there seems to be a misconception of how scientific knowledge serves society. It does so by informing applied scientists, typically via journals, not by appearing in newspapers or informing local knowledge or policymakers, whatever that means. Applied scientists, including enginerrs and doctors, then develop practices and technologies that render knowledge beneficial. Educators develop teaching materials to spread the most basic new knowledge. Nomadic postdocs also spread new knowledge, as do conferences, etc. It is a complex, slow diffusion process, with few shortcuts. The basic researchers role is to do the research and publish their findings, which is why the system rewards them for publishing important results. The alternative world suggested by this study is wrong headed.

It does so by informing applied scientists, typically via journals, not by appearing in newspapers or informing local knowledge or policymakers, whatever that means.

David, aren’t you setting up a false dichotomy here? It seems to me that the reality is nowhere near so binary: scientific knowledge does indeed serve society by informing scientists, but surely scientific knowledge also serves society by being made available to the general public for application in daily life. The germ theory of disease; the carcinogenic properties of cigarette smoke; the transmission vectors of malaria, parasitic diseases, and HIV — all of these are examples of scientific knowledge that has been transmitted both from scientist to scientist and from scientist (or science organization) to the general public and to policymakers, to the general public’s tremendous benefit. In light of such examples (and hundreds or thousands of others), the idea that scientific knowledge doesn’t serve society “by appearing in newspapers or informing local knowledge of policy makers” seems absurd to me.

I agree, scientific information does need to be communicated to the public in an easy to digest manner. The problem I see is that newspapers do tend to sensationalize results, leading people to believe that the next big treatment is just around the corner, whereas in reality the breakthrough may lead to something happening a number of years later. A lot of scientists do try and communicate their work, either by blogging or via organisations such as EuroStemCell ( in a way that is understandable to non-specialists yet tells it how it is!

I do think there needs to be more communication between science policy makers and actual scientists. This seems to be really lacking at least in the UK.

True enough — the communication of research results to the public does pose difficulties, among them the problem of getting accurate information through the filter of mainstream journalism intact. But this doesn’t change either the necessity of communicating scientific knowledge to the public, nor the reality that such knowledge is transmitted to the public all the time — sometimes more perfectly, sometimes less so — and has historically done so to the public’s great benefit. What I’m taking exception to is the patently absurd notion that scientific knowledge benefits the public only by being transmitted between scientists.

Rick, you seem to have missed the issue. The theme of this study seems to be that scientists are spending far too much time reporting results in journals, as opposed to spending their time informing the public and policy makers. I disagree. The issue is binary only in the important sense that there is a strict budget on people’s time. If you want more public communication it means less research or less scientific communication.

The system of science is very large and complex, involving several million people. The system of policy making is likewise large and complex, perhaps roughly the same size, but I have not actually looked at that. It has many active channels to the scientific system.(This is my research area.) The public communication system is probably smaller, because it is less important. Blogs may be changing that.

People and policymakers do not have to follow the article by article scientific research. While science has implications for policy and people, individual studies are almost never significant at this scale. Science only changes policy, or education, or public behavior, at the aggregate assessment level. Blogs may be changing that, leading to whip sawing.

In short, this study reflects a common, yet wildly oversimplified, misunderstanding of the systems involved. The idea that scientists should shift from reporting their results in journals, to sticking it straight into society, is wrong. It is, however,popular.

David, I’m responding to the specific assertion you made in your comment, which is that scientific knowledge does not serve society “by appearing in newspapers or informing local knowledge or policymakers.” I haven’t “missed” the more general issue you discuss in your rejoinder above; I’m simply addressing a specific problem with that specific assertion.

Well Rick, you have missed my point, which oversimplified is that publishing research results in the newspaper instead of a journal would not be better.

I agree in that the academic world at present seems to be highly focussed on publications, both in terms of quantity and journal quality. However, I do not think this a symptom of how scientists want to communicate their work and I personally do not find it to be the most rewarding aspect of being a researcher. Getting published for the first time is a fantastic feeling and it is nice for your work to be recognised as good quality, but I do not think it serves to promote narcissism as such, as there is always someone better out there. I think the primary reason for the focus on publications is to do with attracting funding- first author papers are key to getting funding at the post doctoral level. The whole research grant thing is another issue. In the life science field at least, it seems to be becoming increasingly difficult to attract funding without a ‘translational’ aspect i.e. you need to aim for some form of clinical application to the research. This is going to impact ‘basic’ research, which I think is necessary as you need information to translate. Anyway, I shall shut up now as I don’t believe that this is the right forum for this comment!

Comments are closed.