Study: ~180 online-only, open-access science, social sciences, and humanities journals have disappeared in the last 20 years and their content wasn't preserved
Context & Ripple Effects
This study quantifies a failure mode that other coverage has documented piecemeal: the Pew finding that 38% of webpages from 2013 are already gone shows link rot erasing the general web, and this research shows scholarly publishing is not exempt — roughly 180 online-only open-access journals across science, social sciences, and humanities vanished over two decades with no preserved archive.
The timing matters because the open-access movement spent the 2010s removing paywalls — search engines like Google Scholar and Semantic Scholar helped surface more papers — while the preservation layer never kept pace. Even the project to stabilize Library Genesis, built around pirated copies, is in part a response to how fragile legitimate access to the record has proven.
First-order effects
- Researchers who cited or need work from those ~180 journals now face dead links and no authoritative copy, since the content was never deposited with a preservation service.
- The journals' publishers — often small scholarly operations — are gone, leaving no rights-holder to authorize rehosting, which complicates any rescue effort.
Second-order effects
- Libraries and national archives inherit the preservation burden by default, pushing the cost of keeping the scholarly record onto institutions that had no role in publishing it.
- Shadow archives like the LibGen ecosystem gain a de facto preservation role that legitimate infrastructure failed to provide — an uncomfortable outcome for publishers arguing piracy has no upsides.
Third-order effects
- If vanishing journals are a recurring pattern rather than a one-off, open-access mandates will likely need to pair free publication with deposit-at-publication requirements, making preservation a condition of the OA label rather than an afterthought.
- The credibility of online-only publishing — for both OA and subscription titles — becomes a selection factor for authors and funders, favoring journals backed by archives or large publishers over standalone operations.
The trend: The shift of scholarly publishing online is outrunning its preservation infrastructure, turning the academic record into another casualty of digital decay unless archiving becomes a built-in condition of publication.