Double-Blind Review: A Commitment to Fair Editorial Practices

In recent months there have been several articles, panels, and publishing projects that critique or propose new approaches to the editorial practices involved in peer reviewing. The American Economic Association announced in 2011 that it would end double-blind manuscript reviews in its various journals, saying that “the practice will make it easier for reviewers to spot conflicts of interest, and will reduce the administrative costs of the review process.”¹ In keeping with the topic of a special issue on “Shakespeare and New Media,” Shakespeare Quarterly conducted an experiment in open peer review on the Internet from March to May 2010. This past January, the Modern Language Association convened a panel on “The Future of Peer Review,” which focused both on techniques and underlying philosophies.² Each of these initiatives has been accompanied by an active flurry of positive and negative commentary on various academic blogs, e-lists and newsletters.

Much of this re-examination is prompted by the possibilities of the Internet in scholarly communication. Open access publishing and open source software have led to an academic and creative culture demanding openness, reduced barriers, and shared authoring in all kinds of contexts, even while it is evident that there is not an agreed-upon single definition for either of these terms. The ease of doing a name or phrase search simultaneously across a staggering range of online texts and websites calls into question the assertion that there can be any truly “anonymized” manuscripts or blind reviewing.³ At the same time, substantial improvements in productivity, speed of publication, and in the ability to undertake large-scale editing projects have been demonstrated through such new forms of collaboration as crowd-sourcing and the solicitation of interactive critical commentary through posting preprints and drafts to blogs and subject repositories.

In the vigorous pro and con debate over these ideas and practices there is some unnecessary conflation of terms and polarization of options. One can have open access publishing, for example, and still retain double-blind peer review. Open review, and crowd-sourcing, can be set up so that only an editor sees all the identities, or so that final decisions are still made by a selective appointed editorial board. Newer forms of review and publishing can coexist with, or be a first phase in, a cycle of scholarly communication that includes later stages that are more formally vetted and packaged.

The argument over the value of peer review, the necessity of a double-blind process, and the potential for bias or error that may occur without it, is not in general a new debate, though the digital environment has added new layers of logistical and policy complexity. An excellent review article written in 2010 synthesizes the issues across a wide range of fields and cites over 120 articles dating back to 1973.⁴ There are significant differences across the disciplines in the attitudes of authors, editors, and the scholarly cohort in a given field. In science and medicine there seems to be greater skepticism that blind peer review can be achieved, or that it even has a provable benefit in prioritizing the most important research or preventing fundamental scientific errors from making their way into the literature.⁵ The influential journal Nature seems to be straddling the fence in a 2008 editorial casting doubt on double-blind peer review while still acknowledging that there needs to be protection from bias, and seeking commentary and consensus from the field.⁶ But the argument proferred by the American Economic Association, that potential bias is ferreted out more effectively if the refereeing process is not blind, runs counter to quite a cluster of articles that document the biases that exist when the authors and institutions of each paper are known.⁷ In highly specialized fields it is acknowledged that researchers know who is working on what topics and it is almost impossible to have a blind process. Scientific research involving large laboratory groups and a wide network of public agency grant reviewers operates in a very different landscape from humanistic research articles that are usually the product...