The RCUK open-access policy now open for comment
SPARC Open Access Newsletter, issue #87
July 2, 2005
by Peter Suber
On June 28, the Research Councils UK (RCUK) published its long-awaited open-access policy for public comment.  The comment period ends on August 31, 2005.

RCUK page on the open access policy (collecting related links)

The RCUK Consultation, covering note

RCUK Position Statement on Access to Research Outputs
(PDF, 119 KB)

Summary of RCUK Position Statement on Access to Research Outputs
(PDF, 39 KB)

Consultation to date

RCUK press release, June 28, 2005

Comments on the draft policy should be sent to Dr. Astrid Wissenburg (before August 31, 2005)

Bottom line:  The policy is superb.  It has four primary strengths.  First, it mandates OA and does not merely request it.  Second, it applies to all publicly-funded research, not just biomedicine.  Third, it gives authors some flexibility about the OA archive in which to deposit their work.  Fourth, it offers to pay the fees at OA journals that charge fees.  My concerns about the current draft are minor compared to its strengths, but I'll say more about them below.

As far as I know, the Research Councils are the first public funding agencies in any country to mandate OA to the results of agency-funded research.  The first private funding agency to do so was the Wellcome Trust, also from the UK.  By chance, the two policies will take effect on the same date, October 1, 2005.  The two precedents together make the UK a leader in the worldwide campaign for open access.

In a posting yesterday to the AmSci OA Forum, Les Carr estimated that the Research Councils fund about half the peer-reviewed research in the UK.  So in addition to improving upon the NIH policy, and offering a better precedent for other funding agencies worldwide, the RCUK policy will directly bring about OA to a very significant body of research.

Here are the key provisions of the draft policy.  (I cite it by paragraph numbers in parentheses.)

* Grantees must deposit the results of their research in OA repositories.  This is a requirement, not just a request, encouragement, or exhortation (#13).

* The requirement applies to all published journal articles and conference presentations resulting from RC funding (#14.b).  It applies to all disciplines, including the social sciences and humanities (#1).  However, there are some exceptions noted below.

* The policy applies to grants awarded from October 1, 2005, and beyond (#14.b).  Recipients of earlier grants are encouraged but not required to follow the same policy (#14.e).

* The grantee may choose the repository in which to deposit the research.   While the RCUK prefers institutional repositories to disciplinary repositories, both are eligible for this program (#14.c).   All eligible repositories "should" be OAI compliant (#14.b). 

* The policy does not require any institution to create an OA repository if it doesn't already have one (#14.d).

* Grantees should deposit their articles "at the earliest opportunity, wherever possible at or around the time of publication" (#14.b).  The ultimate state of OA may be required, but the timing is merely recommended.  The policy does not distinguish the time of deposit from the time of free online accessibility (as the NIH policy does, for example) and it provides no firm deadline for deposit or for access.

* The repositories must clearly label published articles as postprints and unpublished articles as preprints.  They must also distinguish "between different preprint versions" (#20).

* Grantees who choose to publish in OA journals that charge author-side processing fees may pay the fees out of their grants "subject to cost-effectiveness being demonstrated" (#27).  But the viability of this funding model is sufficiently uncertain that the RCUK was not persuaded to set up a special fund (as recommended by the Gibson committee) to pay these fees (#26). 

Regrettably, the RCUK uses the term "author-pays" for journals that charge processing fees (##25, 26, 27, 28) even though it acknowledges (#25) that the term is misleading.

* There are two exceptions to the mandatory archiving policy.  First, it is subject to "copyright and licensing arrangements" (#14.b).   Second, it does "not apply to authors who do not have reasonable access either to an institutional or to a subject-based e-print repository" (#14.d).  I'll treat the two exceptions separately, below.

Note the background principle that "[o]nly in exceptional circumstances should public funding of research not lead to publicly-available outputs" (#3.a).

* On the first exception (copyright and licensing arrangements, #14.b):

The policy refers readers who wonder about this clause to paragraphs 16 and 17.  Paragraph 16 points out that "a large majority of publishers" allow postprint archiving.  While true, this does not mean that the same publishers will not impose embargoes on articles deposited as a result of this policy.  Even some green publishers are responding to the NIH policy with new embargoes, sometimes applying to all OA archiving and sometimes only to NIH archiving.  Paragraph 17 questions the custom of automatic transfer of copyright from authors to journals.  It points out that this is not the only way and not necessarily the best way to protect authors against plagiarism or to achieve the optimal "balance between the rights and responsibilities of authors, their employing institutions, funders and other stakeholders --including the general public-- particularly in a context where the Research Councils are funding research for the public good."

Neither #16 nor #17 dispels the vagueness of the phrase "in accordance with copyright and licensing arrangements" (#14.b).  However, we have helpful clarification of the RCUK intent from Astrid Wissenburg, an historian on the Economic and Social Research Council (one of the eight Research Councils) and the interim head of a committee that formulated the RCUK policy.  Aisha Labi quotes Wissenburg (and me) in the June 19 issue of the Chronicle of Higher Education:

"If [RCUK grantees] have signed an agreement with a publisher that either restricts them completely or gives a time restriction -- for example, if the publisher says they are only allowed to deposit their work in six months -- then they can wait six months. So the phrase 'at the earliest opportunity' means when someone is legally allowed to do so. We're not overruling any agreement publishers have in place with authors." ...Ms. Wissenburg conceded that the RCUK exception -- allowing for delay because of copyright and licensing restrictions -- might create an incentive for publishers to begin imposing such restrictions on authors, as a way of dictating when their work could be placed in open-access repositories.  "It will be interesting to see if they do," she said. "We'll be keeping an eye on that. I wouldn't be surprised if some publishers took that attitude and put in restrictions, but I think the majority will wait to see what the impact will be." ...Based on the NIH experience, Mr. Suber is certain of the outcome. "With the NIH policy, we've seen that publishers are requesting embargoes," he said. "They're saying, If you don't comply, we won't publish you. We'll see the same thing with the RCUK unless the language is tightened up before it's made final."

It's clear then that the RCUK is deliberately opening the door for publishers to demand embargoes on OA archiving.  At the same time, the RCUK is promising to respect any embargoes that publishers choose to impose.  Finally, the RCUK is predicting, or perhaps hoping, that most publishers will not impose embargoes unless their experience with unembargoed archiving forces them to do so.  There are two problems here.  First, the NIH experience suggests that the prediction or hope is groundless.  If publishers are allowed to impose embargoes, most will do so.  Second, the policy language could express the framers' intent more clearly and directly.  For example, it could add the clear statement that publishers may impose embargoes of any length and it could subtract the unclear implication that this deference to publishers has something to do with copyright.

Here's how I put the point in a comment to Stephen Pincock for his story in the June 23 issue of The Scientist:

"Researchers sign funding contracts with the Research Councils long before they sign copyright transfer agreements with publishers.  Funders have a right to dictate terms, such as mandated open access, precisely because they are upstream from publishers.  If one condition of the funding contract is that the grantee will deposit the peer-reviewed version of any resulting publication in an open-access repository, then publishers have no right to intervene.  That is, publishers have no valid objection based on copyright law or a licensing contract.  However, publishers do have the right to refuse to publish any article for any reason, and they may well tell authors that publication depends on agreeing not to provide open access to any version for a certain period of time.  But publishers have this power without any action or recognition from the RCUK.  If the [Research Councils] are expressing their own willingness to accommodate this power of publishers, then they muddy the waters by referring to copyrights and licenses, which have nothing to do with this power.  In short, if the clause is supposed to recognize the right of publishers to publish only what they wish, or only on their own terms, then it's superfluous --and harmful for opening other doors through vagueness."

* On the second exception (lack of a suitable repository, #14.d):

How big is this exception?  The RCUK itself counted only 26 institutional repositories in the UK, as of January 2005 (#12).  Tim Brody's Internet Archives Registry counts 55 today.

Brody's list also describes the archive type.  Of the 55 in the UK, 34 are institutional or departmental, 8 are cross-institutional, 1 is for e-theses only, 2 are databases, 6 are e-journals, 4 are other.  (Thanks to Stevan Harnad for these tallies.)

But there are 191 universities in the UK, leaving a majority of them without an OA repository. 

It's much harder to tell how many faculty work in disciplines without disciplinary repositories, because there isn't yet a good directory of disciplinary repositories. 

But it's clear that this exception is non-trivial.  Many UK faculty do not have institutional repositories, many do not have disciplinary repositories, and (what matters) many do not have either.  However, there are reasons to think that the number of UK faculty left in the cold will decrease with time.

One solution is for UK universities without repositories to ask the RCUK to help fund one.  Currently the RCUK mandates deposit in OA repositories but not submission to OA journals.  However, it is willing to pay for OA infrastructure on the journal side (through journal fees, #27) but not on the repository side.  In fairness, it should be willing to support its mandate by paying at least some of the infrastructure costs for repositories.  This compelling argument was first made by Michael Fraser the AmSci OA Forum.

Another solution is to ask JISC for repository funding.  The RCUK itself points out that JISC is supporting institutional repositories (#11), although it also suggests that JISC funding will not suffice (#24).  In fact, just two weeks ago JISC announced a 4m funding program for institutional repositories in the UK.  There is some ground for hope in the fact that #24 has not been revised since a draft circulated in April.  The JISC announcement two weeks ago may be a sign of new money or new willingness to meet the occasion.

Note that while the RCUK is very concerned about the long-term preservation of OA eprints (esp. ##29-31), the policy says that the host for preservation purposes needn't be the same as the host for OA purposes (#30).  Hence, repositories need not take on responsibility of long-term preservation in order to be eligible for this program, which should make them much easier to fund.

Finally, a third solution lies in the "universal repository" that Brewster Kahle and I are setting up at the Internet Archive.  This repository will accept deposits from any scholar in any discipline in any country.  I'd like it to be open for business by October 1, 2005, but it's still too early to say whether this will be possible.

* The policy has two underlying rationales:  "to maximise the impact of [the Research Councils'] investment in maintaining and improving the research base" and to "increas[e] the contribution it makes to the benefit of the UK's society and economy" (#3).

* The RCUK will review the policy before the end of 2008, especially the development of OA repositories and the requirement to deposit (#14.d).  It will also review its decision to let authors choose the destination repository (#14.c).

* On several points the RCUK did not create policy but promised to start inquiries or discussions for future policy.  For example, it will investigate the best ways to provide easy and rapid access to data (#8), but the current policy only applies to articles, not data, that arise from publicly-funded research (#8, #14.b).  It will start discussions with managers of OA repositories to develop "a common and recognisable standard to ensure that the distinction between pre-prints and post-prints is clear to all users and also to maintain document integrity and authenticity" (#20).  It will start discussions with the British Library, the Legal Deposit Libraries, and other stakeholders about the long-term preservation of the OA research literature (#31).  It will start discussions with learned societies about "ways in which they can adapt to and exploit new models of publication" (#32). 

Those are the main provisions.  Here are two aspects of the background.

First, this policy is not from the eight Research Councils alone.  It's a "joint position statement" formulated by the RCUK in "partnership" with other UK agencies "including the Funding Councils, the Joint Information Systems Committee (JISC), the new Research Libraries Network (RLN) and the British Library" and in "consultation" with unnamed others.  It also acknowledges significant reliance on the July 2004 report of the House of Commons Science and Technology Committee (#4).

Second, while the key recommendations of the RCUK were made last summer by the House of Commons, those recommendations were rejected by the government in November.  The RCUK has been able to propose the same measures now, and will be able to implement them in October, because the Research Councils are independent of the government.  (Technically, the Research Councils are "Non-Departmental Public Bodies" outside government, accountable to Parliament, using taxpayer money allocated by the Department of Trade and Industry and channeled through the Office of Science and Technology.  In practice, they are more independent than government agencies but probably less independent than the architects of their status intended.)  For the same reason, their degree of independence makes them less vulnerable to publisher lobbying than the government, a significant matter when most of the commercial publishers who oppose the policy are incorporated in the UK.  Note that the RCUK plans "to liase closely with Government, which is defining its own policy" (#4).

Scientific Publications:  Free for All?  Report of the House of Commons Science and Technology Committee (July 20, 2004)

Government response, rejecting the House committee recommendations (November 8, 2004)

A March 2005 recommendation from the House of Commons Science and Technology Committee to give the RCUK more independence was rejected by the government last month.

All friends of OA, especially those from the UK, should send comments to Astrid Wissenburg ( before August 31. 

* My own comment will focus on the two exceptions to the OA archiving requirement. 

(1) This key sentence needs revision:  "Deposit should take place at the earliest opportunity, wherever possible at or around the time of publication, in accordance with copyright and licensing arrangements" (#14.b).  The combination of flexible timing and vague exceptions for copyright will invite publishers to impose embargoes.  Dr. Wissenburg's comments to the Chronicle of Higher Education confirm that this loophole is open.  The experience of the NIH is that when this loophole is open, most publishers will take advantage of it, even publishers that already permit postprint archiving without embargoes.  However, most authors can bypass NIH-specific embargoes by self-archiving outside NIH in their institutional or disciplinary repositories.  By contrast, RCUK-specific embargoes will apply to the only institutional or disciplinary repositories where authors have deposit rights and leave them without archiving alternatives.  Hence, it's even more important to close this loophole in the RCUK policy than in the NIH policy (though I'd like to see it closed in both).

The RCUK is trying to reduce embargoes with an exhortation.  But this will only work if publishers heed the exhortation.  If they don't, the only way to reduce embargoes is with a firm deadline on deposit and access.  The only problem with a deadline --say, 6 months-- is that it creates a permissible embargo length up to the deadline.  The NIH tried both paths at once, exhorting grantees to deposit their work "as soon as possible" after publication but within 12 months.  Most publishers with public responses are imposing 12 month embargoes.

In my comment, I'll urge the RCUK to consider the track record of publishers in responding to the NIH policy, as documented in SOAN.
(Also see the update to this story in the current issue, below.)

Then I'll urge it to mandate OA archiving immediately upon publication.  Of course publishers will object that this will harm them.  The RCUK should be prepared with these replies.  (a) The mandate will only apply to the author's peer-reviewed manuscript, not the published edition.  (b) Most journals publish more than just RC-funded research and more than just research articles.  (c) Immediate OA has not harmed subscription-based journals in physics.  (d) The author and the funder both have an interest in the maximum impact of their research with the minimum delay.  (e) Journals that start to lose subscribers can always convert to full or hybrid OA.  If they really think the cause of their subscriber attrition is the large number of RC-funded authors, then they can charge fees on accepted papers to cover their costs and collect fees from every one of those RC-funded authors.  (f) The RCUK policy is based on four fundamental principles.  Here's how it elaborates on the second principle:  "Historically, subscription-based academic journals (priced or electronic) have provided [peer review].  But there is not reason in principle why other publishing models cannot accommodate similar mechanisms" (#3.b).

(2) Currently the policy does not apply to grantees who have no "reasonable access either to an institutional or to a subject-based e-print repository" (#14.d).  This leaves too many UK faculty outside the scope of the policy.  The RCUK could solve this problem by hosting its own central repository.  But a better solution, more compatible with the policy's commitment to distributed, interoperable repositories, is for the Research Councils to respond to university requests to fund new repositories.  The investments would not be large.  The software is open-source.  The maintenance of an institutional repository for OA purposes is much less expensive than maintenance for preservation purposes, and the RCUK has already agreed that these functions may be provided separately (#30). 

If you don't agree with me that the RCUK should close these loopholes or at least make them smaller, then send a comment supporting the policy as it stands.  All supportive comments will help the RCUK resist pressure from publishers to water it down.  If the policy is adopted as it stands, it will still be a breakthrough worth celebrating.

* Here's some of the early coverage of the RCUK policy.

The ALPSP issued a response to the policy.  The document is dated Apil 19, 2005, but the ALPSP home page dates it June 30, 2005.

Anon., British group takes step beyond NIH open access, Research Research, June 30, 2005.

Aisha Labi, British Research Group Calls for More-Liberal Open-Access Policy Than NIH Supports, Chronicle of Higher Education, June 29, 2005.

Donald MacLeod, Research councils back free online access, The Guardian, June 29, 2005.,3605,1517384,00.html

Richard Wray, Funding aid for open access, The Guardian, June 29, 2005.,9865,1517116,00.html

* Postscript.  The meeting that drafted the Bethesda Statement on Open Access two years ago was convened by funding agencies:  the Howard Hughes Medical Institute and the Wellcome Trust.  At the time, no funding agencies anywhere mandated OA to the research they funded even though there was consensus around the room that they wanted it and even needed it.  The most commonly heard caution was, "Let's use the carrot, not the stick."  Hence the primary recommendation was to pay processing fees at OA journals, not to mandate OA archiving.  Then last year the US House of Representatives asked the NIH to "require" OA to the results of NIH-funded research.  At the same time, the UK House of Commons Committee on Science and Technology (the Gibson committee) recommended mandated OA across the board.  These recommendations broke the ice even though the NIH eventually weakened the requirement to a request and the UK government rejected the Gibson committee recommendations.  Now the RCUK is reviving the Gibson committee recommendations and mandating OA across the board.  It has taken two years and three large steps to get here.  If funding agencies did not take this step earlier (inter alia) because it was unprecedented or unheard of, then finally that hurdle is behind us.  Funding agencies are now more free to be more effective.  In pursuing their mission to fund research in the public interest, they may now decide that if a research project is useful enough to fund, then it's useful enough to share.


Read this issue online

SOAN is published and sponsored by the Scholarly Publishing and Academic Resources Coalition (SPARC).

Additional support is provided by Data Conversion Laboratory (DCL), experts in converting research documents to XML.


This is the SPARC Open Access Newsletter (ISSN 1546-7821), written by Peter Suber and published by SPARC.  The views I express in this newsletter are my own and do not necessarily reflect those of SPARC or other sponsors.

To unsubscribe, send any message (from the subscribed address) to <>.

Please feel free to forward any issue of the newsletter to interested colleagues.  If you are reading a forwarded copy, see the instructions for subscribing at either of the next two sites below.

SPARC home page for the Open Access Newsletter and Open Access Forum

Peter Suber's page of related information, including the newsletter editorial position

Newsletter, archived back issues

Forum, archived postings

Conferences Related to the Open Access Movement

Timeline of the Open Access Movement

Open Access Overview

Open Access News blog

Peter Suber

SOAN is licensed under a Creative Commons Attribution 3.0 United States License.

Return to the Newsletter archive