pre-publication peer review
The system where a manuscript is evaluated by a few editor-selected reviewers before it is allowed to appear in a journal.
As defined in 95 | Liam Kofi Bright on Knowledge, Truth, and Science
Where it comes up · 1
Most important claims · 4 of 4Search all claims for “pre-publication peer review” →
- An illustrative experiment in which classic, highly-cited papers were anonymized with non-essential identifying details lightly changed and resubmitted for review found about half were rejected on grounds of serious technical flaws; relatedly, super highly-cited papers tend to be more daring, and boring-but-true papers are easier to publish than interesting-and-daring ones, indicating a conservative bias in review.95 | Liam Kofi Bright on Knowledge, Truth, and Science
- The most defensible function of peer review is not filtering out bad work (people put bad work on the internet anyway) but providing a signal—being in a journal acts as a stamp of approval that helps non-experts and scientists in other fields know where to look for decent ideas; but since peer review's signal value depends on its selecting for good ideas (which Bright disputes), it would be worth investing effort in finding a better way to provide that orientation signal.95 | Liam Kofi Bright on Knowledge, Truth, and Science
- Bright proposes replacing pre-publication peer review with directly posting manuscripts to a publicly accessible archive (as already done in physics, math, and increasingly economics), going straight to community-wide assessment—which amounts to wanting more peer review by more scientists (open peer review), since the labor a few reviewers expend deciding what is 'worth sharing' has no firm evidence of doing anything good.95 | Liam Kofi Bright on Knowledge, Truth, and Science
- Bright and Remco Heesen argue that most of the things pre-publication peer review is supposed to do—e.g., sorting good ideas from bad or true claims from false—have either no evidence that it succeeds or some evidence that it fails; it is labor-intensive, time-consuming, and lucrative for publishers, yet the claim that a couple of reviewers' judgments reliably select for quality is not borne out, with a small-c conservative bias against daring work.95 | Liam Kofi Bright on Knowledge, Truth, and Science