Peer review is the mechanism scholarship leans on most heavily and scrutinises least. For most of its history it has happened in the dark: reviewers anonymous, reports unpublished, the reasoning behind a decision invisible to everyone but the editor. Open peer review is the family of practices that opens parts of that process up — and because review quality and review credit both bear directly on who gets heard in scholarship, it sits squarely in the knowledge-equity domain, with strong ties to the research-integrity domain. This article sets out the main models, what they achieve, and the questions that remain genuinely open.
“Open peer review” is not one thing
The single biggest source of confusion is that “open peer review” is an umbrella term covering several distinct, separable practices. A journal can adopt one without the others. It helps to name them precisely.
- Open identities (signed review). The reviewer’s identity is disclosed to the author, and sometimes published alongside the article. This is the dimension most people mean by “open review”, but it is only one.
- Open reports (transparent review). The review reports themselves — and often the authors’ responses and the editor’s decision letters — are published with the article. Reviewers may still choose to remain anonymous; what is opened is the content of the review, not necessarily the identity.
- Open participation. The pool of reviewers is widened beyond editor-selected referees, allowing the broader community to contribute comments.
- Open pre-review / post-publication. Review happens in the open on a preprint or a published article, rather than behind a closed submission system before publication.
Because these dimensions are independent, two journals can both call themselves “open” while doing very different things. Reading an open-review policy carefully — which dimensions are open, and whether they are mandatory or opt-in — matters more than the label.
Models in practice
Several established models illustrate the spectrum.
The publish-then-review model, pioneered by F1000Research and adopted by a number of funder-backed platforms, inverts the usual order: an article is published rapidly after a basic editorial check, then openly peer-reviewed, with the reports, the reviewers’ names, and their approval status displayed on the article itself. The article’s status is transparent to readers, and the review history is part of the public record.
eLife moved to a model in which papers sent for review are published as Reviewed Preprints accompanied by public reviews and an editorial assessment, with the journal stepping back from a simple binary accept/reject verdict in favour of openly published evaluation. It is a notable example of transparent reports combined with a rethink of what the “decision” even is.
Many traditional publishers now offer transparent peer review as a feature: the review reports and author responses are published alongside the version of record, while leaving the reviewer’s choice to sign or remain anonymous to the reviewer. This is the most incremental and widely adopted form — opening the reports without forcing open identities.
What opening review achieves
The arguments for open review are about accountability, credit, and quality.
- Accountability and quality. Reviews written in the knowledge they may be published, or signed, tend to be more constructive and better-reasoned. Publishing the reports also lets readers judge the scrutiny a paper actually received, rather than taking “peer-reviewed” as an opaque stamp.
- Credit for review labour. Peer review is substantial intellectual work that has traditionally been invisible. Open and recorded review makes it creditable — through signed reviews, published reports, and reviewer-recognition records that can be attached to a researcher’s profile. CRediT itself has no review role (it covers authorship contributions), which is one reason the surrounding vocabulary for reviewer recognition matters.
- A teaching resource. Published reviews show early-career researchers what good review looks like — something the closed system kept hidden.
- Integrity. Transparent reports make certain abuses harder to hide: coercive citation demands, superficial review, or reviews that do not match the published decision become visible.
The open questions
Open review is not a settled good with no costs, and honest advocacy acknowledges the unresolved tensions.
- Power and candour. Signed review can chill criticism: an early-career reviewer may hesitate to write a frank negative report on a senior figure’s work under their own name. Open identities can protect the powerful as easily as they hold them to account.
- Reviewer recruitment. Some evidence suggests inviting reviewers to be named, or to have their reports published, makes them more likely to decline — a real cost when reviewer fatigue is already acute.
- Equity effects cut both ways. Transparency can surface and deter bias, but it can also expose reviewers from marginalised positions to retaliation. The knowledge-equity case for open review is therefore genuinely contested rather than one-directional.
- What counts as the review of record. When review is open, continuous, and possibly post-publication, the tidy notion of a single pre-publication gatekeeping event gives way to something more fluid — which is arguably more honest, but harder to summarise as a simple “reviewed: yes/no”.
Where shared vocabulary fits
“Open peer review”, “signed review”, “transparent review”, and “post-publication review” are used interchangeably far more often than they should be, which muddies both policy and the credit researchers can claim for reviewing. A shared, federated vocabulary that distinguishes these models precisely — and connects them to reviewer-recognition records — is what lets an open-review policy in one venue be compared honestly with another. Supplying that definitional layer is the role the CASRAI dictionary is designed to play; the relevant terms sit in the knowledge-equity domain, with adjacent entries in the research-integrity domain.
