Sage retracts dozens more papers from journals it inherited. Some authors say they were never shown the evidence.

Sage has retracted dozens more papers from two journals it inherited when it bought IOS Press in 2023, and in each case the stated reason was manipulation of, or concerns about, peer review. What makes this round worth a closer look is not the count. It is that several of the affected authors are pushing back publicly, and their objection is about process rather than guilt.

What happened

The retractions came in two batches, both reported by Retraction Watch on 17 August 2026.

• On 15 July, Sage retracted 43 papers from the Journal of Intelligent and Fuzzy Systems (JIFS), several tied to a special issue linked to a 2020 conference held in Türkiye. The notice said an internal investigation found the submissions carried "indicators of third-party involvement and attempts to manipulate the peer-review process."
• On 5 August, a second notice covered 34 papers published between 2017 and 2025 in the International Journal of Knowledge-Based and Intelligent Engineering Systems. It cited suspected third-party involvement, peer-review problems, citation manipulation and "tortured phrases," the mangled paraphrases that often signal automated text laundering.

Sage defines third-party involvement as paper-mill activity, peer-review manipulation, or "any other evidence suggesting the wholesale generation of research submissions."

The context that matters

This is not Sage's first cleanup of the portfolio it acquired in November 2023. JIFS alone retracted 1,561 papers across earlier batches through April 2025, a single-journal record, after Clarivate put the title's indexing on hold over quality concerns. That flag still shows as "on hold." Sage has taken similar action at other former IOS Press titles. The publisher says the July and August retractions came from a separate investigation, not the 2024 to 2025 one.

The comparison writes itself: Wiley bought Hindawi in 2021, retracted thousands of papers, closed journals overrun by paper mills, and eventually dropped the brand. Acquiring a compromised portfolio means inheriting the job of cleaning it, and that job is unfinished years later.

Where the authors disagree

Some of the retracted authors are named academics who say the process was unfair to them specifically.

Cengiz Kahraman, an engineering professor at Istanbul Technical University and an author on six of the retracted JIFS papers, called the publisher's actions "legally and editorially unacceptable." He told Retraction Watch that "retraction decisions are complex and lengthy processes that should be concluded by taking into account the editorial and peer-review system under which the articles were originally published."

Selcuk Cebi, an engineer at Yildiz Technical University, said he has "never been presented with any article-specific evidence indicating unauthorized third-party involvement or manipulation of the peer-review process in relation to my paper." He added that he had already sent Sage his original manuscript, revised versions, reviewer responses and the final accepted version in an earlier appeal, without a substantive reply.

Sage, for its part, acknowledged a real limit. Public affairs manager Chris Burnage said the publisher has "limited information about the original publication process," the process that ran under the previous owner. A colleague, Laura West, said Sage has "applied the same rigorous publication ethics and research integrity standards that guide all of our journals publishing."

Analysis: the due-process problem in inherited mass retractions

The following is Bizarus analysis, not reporting.

The evidence that a special issue was compromised at scale, and the evidence that any one paper in it is fraudulent, are not the same thing. A publisher that inherits a special issue with manipulated peer review has a strong basis to distrust the batch, and a weak basis, sometimes no independent basis at all, for adjudicating each paper inside it. That is precisely the gap the authors are pointing at. When the original submission and review records sit with a previous owner, a new publisher may be retracting on the pattern rather than on the paper.

None of this establishes that the contested authors are innocent. Batch-level manipulation is real, tortured phrases are a genuine signal, and mass retraction is often the only proportionate response to a portfolio a paper mill has worked through. But a retraction is a permanent mark on a named person's record, and "we have limited information about the original publication process" is an uncomfortable thing to sit next to that mark. The reproducibility literature Bizarus follows keeps returning to the same lesson: transparency of process is what lets an outside party check a claim. Here the process itself is what nobody can fully reconstruct.

The honest reading is that both things are true at once. The portfolio needed cleaning, and the mechanism for cleaning it cannot cleanly separate the guilty paper from the merely adjacent one. That tension will not go away as more acquired paper-mill portfolios come up for the same treatment.

What to watch

Whether Sage publishes article-level evidence in response to the appeals, or whether inherited mass retractions settle into a norm where batch-level suspicion is treated as sufficient for individual retraction. The answer sets a precedent for every publisher that has bought a troubled journal and found the receipts missing.