Open Access News

Jan Velterop, Reviewed reviews, The Parachute, January 18, 2008.  Excerpt:

"Book self-archiving cannot and should not be mandated, for the contrary of much the same reasons peer-reviewed journal articles can and should be."

Stevan Harnad
18 January 2008
contribution to liblicense-l

I agree with him.

I think.

Why I can't be entirely certain is because by peer-reviewed journal articles he may mean the same as the NIH in the description of the types of articles that fall under the mandate, which says:

"The Policy applies to all peer-reviewed journal articles, including research reports and reviews. The Policy does not apply to non-peer-reviewed materials such as correspondence, book chapters, and editorials."

That's a mistake, in my view. Review articles belong in the second sentence, with editorials and the like; not the first. More often than not, review articles are initiated by a publisher, inviting a distinguished author to write one. More often than not the author is offered some payment for writing it. Seldom if ever is a review article the result of a funded research project.

Review articles have a lot in common with books. And if self-archiving of books "cannot and should not be mandated", the same applies, grosso modo, to review articles.

Even OA publisher par excellence, BioMed Central, requires subscriptions to access review articles, for instance in the journal Breast Cancer Research. I think they are right to do that....

Comments.  Whether OA mandates should ever apply to books is a fascinating question.  Some thoughts:

1. First, see the whole discussion thread in which Stevan made his comment about books.
2. Jan is right that review articles have important similarities to books.  But if a funder like the NIH mandates OA for review articles, then clearly the mandate only applies to those review articles that result from its funding.  If Jan is also right that "seldom if ever is a review article the result of a funded research project", then the question is generally moot --and in those exceptional cases where funding does give rise to a review article, the funder policy could be justified.
3. Funders mandate OA because they want the results of their research to be disseminated or available as widely as possible.  They might not care whether the research results in journal articles, books, or some other genre.  But they might start to care (or one might argue that they ought to care) if they want to allow grantees who write books to earn royalties on them.  But even in that case, funders wouldn't have to give up on the idea of mandating OA to books.  They would have to evaluate the mounting evidence that for some kinds of books, including research monographs, OA is compatible with royalties and might even stimulate a net increase in sales. 
4. One critical variable is consent.  Even a funder "mandate" is conditional, or depends on grantee consent.  A funder book mandate could never say, "You must make your book OA" --or if it did, it would be unenforceable.  It could only say, "If you take our money, and use it to write a book, then you must provide OA to the text; if you don't like that, then don't take our money."  I say more about the priority of consent over genre in the Richard Poynder interview at p. 40.
5. Some researchers might not take grant money if it required OA to any books resulting from the grant.  But some would still take it.  I know this from my own case.  In the mid-1980s (I forget the year!) I received a grant from the National Endowment for the Humanities for research that resulted in a book.  The NEH didn't require OA, but I made an OA edition on my own as soon as I was legally allowed. 
6. I've participated in drafting several funder OA policies, and in every case I recommended an exemption for royalty-producing books.  For practical and strategic reasons (to get policies adopted, to get researcher support), it's important to focus OA mandates on royalty-free literature.  But points 3 and 4 above show that a funder could depart from this advice for its own reasons (to disseminate its research results more widely) and with grantee consent.
7. I can think of two policies that come interestingly close to OA mandates for books.  (1) The Scholarly Editions Program of the US National Endowment for the Humanities (NEH, not NIH) doesn't require OA for the books it funds, but since September 2006 it has given preference to proposals that promise OA for the resulting books.  (2) Ilmenau Technical University has required dual editions (OA and TA) for every book published by Ilmenau University Press since March 2007.  I applaud them both. 

Update.  Stevan Harnad has now blogged, and elaborated, his original forum comment on books.