What are paper mills, how they act and their consequences to science: an introduction to the tip of the iceberg

In this post, I will briefly introduce paper mills and their consequences to science.

 

I decided to do so because (as long as I know), not too many scientists are aware of the problem, but there is a lot of effort from sleuths (people who work for decontaminating science, for reference please read The Rise of the Science Sleuths) dedicated to investigate it further.

 

Let’s start from the very beginning. Unfortunately, scientists have wrong incentives to share their research findings in the form of publications: the publish or perish pressure. PhD and master students are pressured to publish in order to have their thesis and dissertation defended; academics in general are pressured to publish for getting promotions, higher salary, grants, etc. However, publishing is not so easy, as a lot of parameters are involved in this process. To state, experimental sciences (as chemistry and biology) are highly affected by the replication crisis, where researchers can struggle to reproduce experiments or achieve successful results. In this scenario, there is a general refusal to publish bad (or negative) data, and as long as your article does not contain innovative or fascinating results, it will be very hard to get it published. In this vicious cycle, we still have deadlines to meet, such as grant or promotion applications.

On the other hand, there are fraudulent organisations selling articles’ authorship on social media, as Telegram, WhatsApp and Facebook, the so-called paper mills. Their activities are basically to fabricate fake manuscripts (for instance, by automatic text generators), submit these manuscripts to journals and manipulate the peer-review process. As an example, some journals request the authors to suggest reviewers during the submission steps; then the fraudsters suggest themselves (with fake emails) and in the reviewer report they recommend the article to be accepted. After the acceptance, the advertisements on social media start selling authorship for an already accepted article in a given journal (usually with higher prices for 1st and 2nd authors). After selling authorship, it is just a matter of asking the editors to change the authors list and that is it! The article is published without any effort (apart from paying) from the authors.

 

Below one example, from this post on X by @author_for_sale: 

 

“There’s this entire economy, ecosystem of Facebook groups, Whatsapp groups, Telegram channels selling authorship for papers, selling citations, selling book chapters, selling authorship of patents.” – Nick Wise for Retraction Watch

 

The damages of paper mills to science are huge: a lot of fake articles contaminating the literature. However, some signs of paper mill products can be flagged, and a detailed list (as well as more technical discussion about the subject) can be found in the article Paper mill challenges: past, present, and future by Parker et. al., 2024. Here I want to provide at least one example of these signs: tortured phrases, a “dubious writing style” that replaces established terms. The Table below (from Tortured phrases: A dubious writing style emerging in science, by Cabanac, Labbé and Magazinov, 2021) shows some examples of tortured phrases detected by the Problematic Paper Screener.

 

Reproduced from Cabanac, G., Labbé, C., & Magazinov, A. (2021). Tortured phrases: A dubious writing style emerging in science. Evidence of critical issues affecting established journals. arXiv preprint arXiv:2107.06751.

 

As a community, what can we do to avoid the effects of paper mills? From the recommendations by Parker et al., 2024 two caught my attention: change incentives in research (I would say that a better acceptance of the scientific community for publishing “bad” or negative data); and share knowledge (the main objective of this personal blog, to make more people aware of the problems and how to mitigate them).

 

I would like to highlight the key findings by Parker et. al., 2024:

Key findings

• Paper mills are fraudulent organizations that make money by writing fake manuscripts and offering authorship slots for sale to academic customers.

• Paper mill outputs are large scale, and many thousands of fake manuscripts have been successfully published in peer-reviewed journals.

• Paper mill activity is still relatively unknown outside of the publishing industry, meaning that many researchers and readers do not know the signs.

Parker, Lisa, Stephanie Boughton, Lisa Bero, and Jennifer A. Byrne. ‘Paper Mill Challenges: Past, Present and Future’. Journal of Clinical Epidemiology, 9 October 2024, 111549.

 

For concluding, paper mills promote fake research, and usually the same “results” provided in an article will be reproduced in other works; this practice has a lot of negative impact in science, as the pollution of the scientific literature and the corruption of knowledge. On the other hand, some bad activities related to authorship do not necessarily affect scientific knowledge, but still compromise research evaluation. To state, a valid (not fabricated) article can be authored by people who pay (or have other kinds of agreements with co-authors) for having their names added to the manuscript. In such a case, the wrong person receives credit for a work on which she/he has not contributed, but this addition to the CV can be a differential for getting job positions, be distinguished from other candidates, receive awards, grants, promotions, and so on.

 

To get more insights about paper mills, I recommend you Find an Issue in the six cases presented by Anna Abalkina.

 

Written by Nathanne Rost, with special thanks to Jennifer Byrne and Willem Halffman, for contributing for this text before its publication.