The scientific world must not only produce new knowledge but also safeguard its reliability. All actors in the academic world must take responsibility for eliminating zombie articles and preventing their creation.
Scientific publishing has long existed as a self-correcting system. As was customary, erroneous results were criticized over time, corrected with new studies, and scientific knowledge gradually became more robust. Recent findings show that this ideal scenario is being seriously challenged. In particular, the rapid increase in the number of retracted scientific articles has raised concerns about trust and quality in the scientific community.
According to Retraction Watch (https://retractionwatch.com), an independent platform that systematically monitors scientific publications, the number of retracted articles has increased dramatically in recent years. Research shows that more than ten thousand scientific articles were retracted in 2023 (https://www.nature.com/articles/d41586-023-03974-8). This number is a record in the history of scientific publishing. Moreover, the increase in the number of retracted articles cannot be explained solely by the increase in the number of publications; the retraction rate itself is also rising (https://pmc.ncbi.nlm.nih.gov/articles/PMC10485848/).
These developments highlight a phenomenon that some researchers now call a “retraction crisis.” Retraction of a scientific article means that the article is no longer reliable and should not be used as a reference in the scientific literature. This decision is usually made in the following cases:
- Data fabrication or manipulation
- Plagiarism
- Ethical violations
- Fraudulent peer review
- Serious methodological errors
Unfortunately, retracted articles are not limited to studies published in small or low-impact journals. From time to time, even the world’s most prestigious scientific journals face such cases. For example, a case report recently published in The Lancet, which has been the subject of much debate for a long time, was re-examined. According to Retraction Watch, serious doubts have been raised about the reliability of a study concerning an infant who suffered opioid poisoning through breast milk (https://retractionwatch.com/2026/02/04/lancet-flags-long-scrutinized-report-of-infant-poisoned-by-opioids-in-breast-milk/).
One of the best-known cases in the history of science is Andrew Wakefield’s 1998 paper, which claimed that the measles-mumps-rubella vaccine was linked to autism. Later retracted due to ethical violations and data manipulation, this study continued to be used for years as a key reference point for anti-vaccine movements.
Retractions are not limited to young researchers or small laboratories. Retraction Watch data reveal that studies by some Nobel Prize-winning scientists have also been retracted. This suggests that problems in the scientific system cannot be explained solely by individual ethical violations. Structural factors such as academic competition, publication pressure, and research funding profoundly affect scientific production processes.
One of the most interesting problems is that the impact of retracted articles doesn’t easily disappear. Even after being retracted, an article can continue to receive citations in the scientific literature (https://journals.plos.org/plosone/article?id=10.1371%2Fjournal.pone.0277814). Some retracted studies have been cited hundreds or thousands of times even after the retraction decision. In other words, these continue to live on as “zombie articles.”
It is estimated that the number of scientific articles published worldwide today has reached approximately fifty million. Tens of thousands of these have been retracted (https://www.nature.com/articles/d41586-025-00455-y). The percentage may seem small at first glance. However, given the cumulative nature of scientific knowledge, this number can have a significant impact. When an article is retracted, it doesn’t simply mean that a single study was flawed. Research, meta-analyses, and clinical applications based on that article may also be indirectly affected.
When considering the root of the problem, the first thing that comes to mind is publication pressure (publish or perish). The fact that academic careers are largely dependent on the number of publications leads some researchers to produce a large number of articles quickly. The emergence of so-called “paper mills” in recent years has made this problem even more visible. These companies can produce fake or low-quality articles for a fee and submit them to scientific journals. In some cases, fake peer-review processes or organized citation networks are also involved (https://www.theguardian.com/science/2024/feb/03/the-situation-has-become-appalling-fake-scientific-papers-push-research-credibility-to-crisis-point). On the other hand, the focus of some commercial publishers on increasing publication volume rather than quality also contributes to this problem.
Retracted articles are not just an academic issue. Incorrect scientific studies directly affect health policies, clinical practices, and public trust in science. Therefore, scientific trust is a common concern not only for academic circles but for the entire society.
A rethinking of research evaluation systems is necessary for a healthier academic environment. Making research data and analysis processes more transparent, expanding open data practices, and more effectively monitoring scientific publications are important steps in this direction. Furthermore, retracted articles should be clearly marked in databases, and researchers should be informed about this.
The scientific world today faces not only the task of generating new knowledge but also the responsibility of protecting the reliability of that knowledge. Eliminating “zombie articles” and preventing their creation places significant responsibilities on all actors in the academic world.
