The Hidden Power of Citations – Reflections on scientific citation practices

This famous metaphor, attributed to Bernard of Chartres in the 7th century, vividly illustrates how science works. Each advance, each step further, is only possible thanks to the height provided by an already constructed body of knowledge. To add their own contribution to this structure and ensure its solidity, scientists must be able to support each of their claims with one or more references, carefully chosen to substantiate their point.

“La cigale ayant chanté tout l’été (Williams & Simons, 1995), se trouva fort dépourvue quand la bise fut venue (Hoppendsteadt and Keller, 1976).” could have been written by a Dr. La Fontaine working on hemipterans (for our fellow english readers: this is a famous french poem, which can be translated as “The cicada, having sung all summer, found itself quite destitute when the north wind came“).

Scientific writing is shaped by these citation choices. In theory, these choices are straightforward: the scientist selects the most relevant source to support their argument. In practice, however, other factors sometimes come into play—often unconsciously—such as an article’s visibility, the reputation of the journal in which it is published, or its impact factor.

Scientific articles are published in peer-reviewed journals, now primarily online, and managed by publishers. They can be broadly classified into three categories based on their economic model: for-profit journals, non-profit journals, and “academic-friendly” journals that reinvest their revenues into research (for example by funding and/or organizing conferences).

Today, an increasing share of publications is controlled by large commercial publishers. Over just a few decades, their influence has grown considerably: the share of articles published by these journals has risen from around 20% in the 1970s to more than 50% in 2013. The more visible a journal is, the more submissions it attracts; the more articles it publishes, the more it is cited. This feedback loop helps consolidate the dominance of the most prestigious journals, often for-profit.

In response to this observation, postdoctoral researchers and data scientists at Cesab propose the idea of a “virtuous circle.” Without ever compromising scientific rigor, the aim would be to cite more articles—whenever possible—from “academic-friendly” journals, in order to increase their visibility.

Here we find the familiar logic mentioned earlier: that of the consumer choosing free-range eggs. After ethical consumption, why not consider a form of ethical citation?

This good intention raises a practical challenge: who has the time to examine the editorial policies of every journal listed in a bibliography? Very few people—and the authors of the study are well aware of this. That is why they created: fairpub.

This tool allows researchers to automatically analyze a list of references and assess their distribution according to the economic model of the journals. It thus makes potential biases in citation practices visible.

*Non-contractual illustration

The tool notably relies on the Dafnee database, which classifies journals according to their operating model and reinvestment policies—an initiative of the Isem (Institute of Evolutionary Sciences of Montpellier). For now limited to certain fields such as ecology and evolution, this initiative could eventually be extended to other disciplines.

En savoir plus