The Death of the Science Journal: A Good Thing Too!
The traditional science journal is dying, and not only from the ideological capture that has hollowed out its credibility. It is also being overtaken by simpler, faster, and more open forms of communication that no longer require its expensive gatekeeping apparatus. The prestige titles still command attention, but the model that sustained them, slow, costly, subscription-walled, and editorially controlled, is increasingly an anachronism.
For most of the twentieth century the journal performed a clear function. It provided a mechanism for quality control through peer review, a permanent record of priority, and a signal of professional standing. That function depended on scarcity. Printing and distribution were expensive. Access was limited. The journal's brand therefore carried weight. Scientists submitted work, waited months or years, and accepted the verdict of anonymous reviewers and editors because there were few alternatives for reaching a serious audience.
Those conditions no longer hold. Preprint servers, open repositories, and rapid online commentary have collapsed the time between discovery and dissemination from years to days. A competent researcher can place a paper in front of the relevant community almost immediately, receive criticism in public, and revise accordingly. The formal journal publication that follows, if it follows at all, often adds little beyond a typesetting fee and a paywall. The gatekeeping that once protected standards now frequently protects narratives. When major journals adopt ethics guidelines that empower editors to reject or retract work judged to undermine "socially constructed or socially relevant human groupings," the claim to neutral arbitration of evidence becomes difficult to sustain. When the same journals fill their pages with woke social-justice framing and political endorsements, the signal of quality is further degraded.
Cost compounds the obsolescence. Institutional subscriptions to the major commercial publishers run into the millions. Individual articles are locked behind fees that place them beyond the reach of independent researchers, journalists, and the public that ultimately funds much of the work. The result is a system in which knowledge is produced with public money, refined through the unpaid labour of peer reviewers, and then sold back to the institutions that paid for it in the first place. Open-access mandates have partially addressed the problem by shifting costs to authors, but the underlying extraction remains. The public still does not enjoy frictionless access to the literature that shapes policy and medicine.
Online publishing does not automatically solve problems of quality or bias. Preprints can be wrong, commentary can be tribal, and reputation systems can be gamed. Yet the structural advantages are obvious. Distribution is nearly free. Access can be universal. Versioning is transparent. Criticism is visible rather than hidden behind confidential review. The old journals retain residual prestige and the ability to confer career rewards, but those advantages erode as more scientists and readers conclude that the formal publication process adds delay and ideology without commensurate gains in reliability.
The death of the science journal is therefore not a single event but a gradual displacement. The institutions that once organised scientific communication around scarcity and authority are being left behind by systems organised around speed, openness, and lower cost. Ideological capture has accelerated the loss of trust, but the deeper cause is technological and economic. When the marginal cost of distributing a paper approaches zero, and when readers can bypass the paywall and the editorial filter entirely, the traditional journal becomes a luxury good of diminishing necessity. What replaces it will have its own failures. It will not, however, be the closed, expensive, and increasingly politicised model that is currently losing its grip.
https://cliscep.com/2026/08/03/the-death-of-the-science-journal/
