Quality assurance and open review

While many of the things that we envisaged in the DFG proposal (Müller & Haspelmath 2013) are up and running already, one important thing is still missing: open reviewing. As was already argued by Pullum (1984), open review increases the quality of publications because reviewers will have to do their job carefully, as it is their reputation that will suffer if their name is associated with a bad publication. In addition, Pullum pointed out that reviews may improve a publication quite substantially and in a closed reviewing scenario the reviewers contribution cannot be acknowledged as it should be.

As was noted by Osterloh & Kieser (2015), research metrics like the h-index and impact factors of journals are not well-suited to evaluate researchers. The authors argue for open reviewing systems as well. At Language Science Press, we have the ambition to introduce optional open review in linguistics. The following is a sketch of our current plans and ideas.

Language Science Press uses several ways of ensuring quality and transparency: First, all monograph submissions are reviewed by at least two external reviewers. The reviewers may decide to make their review publicly available or they may decide to reveal their identity without revealing the complete review. The reviewers thus stand in with their names for the quality of the book. This is known from backcovers of the books of some publishers (Prof. A: This is a great book which the field has waited for.), but here it comes with a little more commitment, since the community will know that our reviewer actually read the book carefully and is convinced that it is worth publishing it with Language Science Press.

The next level of transparency is reached by publishing the complete review. Of course this could be paired with the publication of the original submission. Having a version system that documents the initial submission ensures that ideas are documented in their earliest available form.

After the reviewing by Language Science Press, if the reviewing is positive, a book counts as officially accepted. In addition to having this phase open and transparent, there is an optional second phase of public open reviewing. The book can be published in a preliminary version. The scientific community is invited to comment on the book. After a certain time span (2 month) the author reacts to the comments either in an online discussion or by addressing them in the book and then the final release will be published as the official version. Public reviews will get a DOI and will be citeable documents.

These three levels of transparency will be combined with our gamification system (which will be implemented within the next year) and awarded accordingly: non-anonymous reviewing gets more credits than anonymous reviewing, open reviewing gets more credits than non-anonymous reviewing. Reviews can be voted up or down as it is known from stackexchange, ResearchGate, and reddit.

There are some proponents of open access who argue for a market-driven view (Osterloh & Kieser 2015). That is, everything that is sound gets published and then the post-publication reviews and/or citations by other researchers will indicate the quality of a publication. We did not adopt such a model for two reasons.

  1. We think that reviews by selected external reviewers almost always improve the quality of a book.
  2. Open Access is often associated with predatory journals that publish everything they get since they live from the publication fees.

We believe that quality assurance is one of the key points in the transition to Open Access and therefore insist on peer review. But we also believe that the scientific community can contribute to further improvements and hence allow for and encourage open reviewing by the community.

We already have a few books with open reviews. For instance, Eric Fuß wrote an open review of Roland Schäfer‘s book Einführung in die grammatische Beschreibung des Deutschen and my review of Timm Lichte‘s book Syntax und Valenz: Zur Modellierung kohärenter und elliptischer Strukturen mit Baumadjunktionsgrammatiken will also be public. The first book that will undergo the second phase of public reviewing, the community reviewing, is my book Gram­ma­ti­cal theo­ry: From trans­for­ma­tio­nal gram­mar to cons­traint-ba­sed ap­proa­ches. This book is a translation of a German book and the German book had some reviews in journals already. However, the book will be available for a much larger readership now and there may be experts on some of the topics covered or touched in the book that can add valuable insights, suggest further reading or point out inconsistencies (I hope there aren’t any …). The book is not finished yet but I hope to be able to finish it before the summer and I am curious to see whether open reviewing will work in this case.



Müller, Stefan. 2012. A personal note on open access in linguistics. Journal of Language Modelling 0(1), 9–39.

Müller, Stefan and Haspelmath, Martin. 2013. Language Science Press: A publication model for open-access books in linguistics. Grant Proposal to the DFG.

Osterloh, Margit  and Alfred Kieser. 2015. Double-blind peer review: How to slaughter a sacred cow. In Welpe, Isabell M., Jutta Wollersheim, Stefanie Ringelhan, and Margit Osterloh: Incentives and Performance, Springer, 307-321.

Pullum, Geoffrey K. 1984. Stalking the perfect journal. Natural Language and Linguistic Theory 2(2), 261–267.


Leave a Reply

Your email address will not be published. Required fields are marked *

Hinweis / Hint
Das Captcha kann Kleinbuchstaben, Ziffern und die Sonderzeichzeichen »?!#%&« enthalten.
The captcha could contain lower case, numeric characters and special characters as »!#%&«.