Elsevier permits postprint archiving
SPARC Open Access Newsletter, issue #74
June 2, 2004
by Peter Suber
In a May 27 email to Stevan Harnad, Karen Hunter announced an important change of policy at Elsevier.  The world's largest publisher of scientific and scholarly journals now permits postprint archiving.   Elsevier authors may now provide open access to the final editions of their full-text articles by posting them to their personal web sites or their institutional repositories.  They may not deposit them to repositories elsewhere.  The archived or OA edition must be author-made, not Elsevier's PDF or HTML, and must include a link either to the journal's home page or the article's DOI.  Hunter is Elsevier's Senior Vice President for Strategy.

The new policy is not yet spelled out on the Elsevier web site or copyright transfer agreements, but the updates will follow shortly. 

See Stevan Harnad's listserv announcement, quoting Karen Hunter's email with comments of his own.

Yes, this is the breakthrough that it seems to be.  We may disagree about how well it matches public definitions of "open access", how it weighs against other Elsevier policies, or even how much it was foreshadowed by earlier Elsevier actions.  But there is no question that it marks significant progress for OA.  Permission for postprint archiving is all that authors need to provide OA to the final, peer-reviewed editions of their own work.  Elsevier deserves our thanks for adopting this most helpful policy.

How important are the limitations on the policy?  If authors cannot provide OA to Elsevier's own HTML or PDF, then they will not be able to provide OA to the "official" version of the article.  This hardly matters if authors can provide OA to the final edition of the text.  Scholars will not refuse to read the OA edition, but will they refuse to cite it?   From the standpoint of OA, it doesn't matter.  Will authors be prohibited from adding the final pagination to their own editions?  The policy doesn't say so, although many authors will not be bothered.

How important is it that Elsevier will not allow authors to deposit their postprints in OA repositories outside their own institutions?  This restriction rules out deposit in arXiv and PubMed Central, for example, though only for postprints.  On the one hand, as long as the final version of the full-text is OA through the author's institutional repository, and as long as the repository is interoperable with other repositories through the OAI protocol, then all OAI-compliant data services can find it and share it.  Moreover, an increasing number of non-0AI services, like Google and Yahoo, will also be able to find it and share it.  But does Elsevier's restriction mean that when users find a copy in the author's institutional repository, they are not free to make and redistribute their own copies?  If so, the restriction would be inconsistent with most public definitions of "open access", inconsistent with the usage policies of most institutional repositories, and unenforceable.

A policy-change that supports OA archiving does not entail policy-changes that ameliorate high prices, inflexible bundling terms, or harsh negotiating tactics.  And it is not a ground for overlooking any policies that hinder the access to published research.  Individual scholars and institutions will have to decide for themselves how these pluses and minuses net out.  However, if Elsevier authors want free online access to their work, this opportunity makes it possible, regardless of every other policy and practice at Elsevier.  By helping authors without helping libraries, Elsevier may challenge institutions that support OA only in order to help libraries.  But the breakthrough is that Elsevier is now helping authors by removing access barriers to their work.

Until now, Elsevier permitted preprint archiving but not postprint archiving.  That is, it did not to ask authors of published articles to remove their preprints from the web, at least when the preprints differed from the postprints. 

Its position on postprint archiving was complex and in some ways unclear.  Authors could put their postprints on non-public directories within their employer's web site.  Authors could not update their OA preprints to match the texts of their published postprints.  Authors who wanted to put their postprints on a public, OA server had to ask Elsevier case by case for permission.  This is the clear part of the older policy. 

See Elsevier's page on rights retained by authors (checked May 31, 2004)

Also see Elsevier's page on electronic preprints (checked May 31, 2004)

However, at a time when the Elsevier web site described the same policy against postprint archiving that it describes today, Derk Haank gave an interview to Richard Poynder in which he proudly described a much more liberal archiving policy (Information Today, April 2002).  At the time, Haank was the Elsevier CEO.

[Haank] "We consider open archiving to be in line with our policy of open linking, which we have always supported. As a founding father of CrossRef, we realize that other initiatives like open archiving could be another means to the same end...."

[Poynder] "You imply that open archiving is the same as CrossRef, but CrossRef assumes that linked articles are all behind a financial firewall.  Open archiving, by contrast, depends on researchers self-archiving their articles on the Web so that anyone can access them at no cost.  Supposing an academic wants to publish a paper in one of your journals, but to self-archive it on the Web as well.  Would that be acceptable to Elsevier?"

[Haank] "You can put your paper on your own Web site if you want. The only thing we insist on is that if we publish your article you don't publish it in a Springer or Wiley journal, too. In fact, I believe we have the most liberal copyright policy available."

This interview made it unclear whether Elsevier permitted postprint OA without case-by-case permission, whether it permitted postprint OA through archives or only personal web sites, and whether authors should follow the policy on the company web site or the policy articulated in public by the CEO.  When I wrote to Haank for clarification, he did not reply. 

Other highly-placed Elsevier officers used the conservative web-site edition of the policy in their public statements, not the liberal Haank version.  For example, Jeffrey Young cited the conservative version of the policy in a July 5, 2002, article for the Chronicle of Higher Education.  His source was Arie Jongejan, CEO of Elsevier's Science and Technology Division.  Young's paraphrase:  "Elsevier does allow its authors to publish their papers in institutional repositories or other noncommercial archives, provided that the authors ask permission first.  [Jongejan] says that fewer than 5 percent of authors ask."

Another sign that Elsevier's upper management was not uniformly friendly to OA archiving was Pieter Bolman's column ("Chairman's Corner") in the Fall 2003 issue of the Professional Scholarly Publishing Bulletin.  At the time Bolman was the Vice President and Director of STM Relations at Elsevier.  In the column Bolman questioned both the practicality and the legality of OA archiving.

I really don't know whether Haank meant to assert such a liberal policy in his conversation with Poynder.  Even if he did, I don't know whether he was the only one in the company at the time to support wider OA archiving or whether his support outlived his tenure at the company and affected the recent decision to liberalize the policy.

However, one reading of the events is that Elsevier always opposed OA journals, at least for itself, but did not always oppose OA archiving.  The new policy could be the blossoming of a seed planted in 2002 or before.

I don't want to put a cynical reading on the new policy permitting OA archiving.  On the contrary, I want to celebrate it.  But it's not cynical to wonder how the change could be in Elsevier's interest.  What might Elsevier have been thinking? 

Here's one possibility:  The new policy gives it an edge over most other publishers.  If Elsevier permits postprint archiving, then it is suddenly more advantageous to authors than before.  (Authors take note:  this is true.)  This will help Elsevier journals in the competition for high-quality submissions, at least against journals that do not permit postprint archiving.  If other journals follow suit, then (Elsevier may be hoping) either they will do it late, giving Elsevier a period of unanswered competitive advantage, or they will do it with fewer resources to survive a decline in subscriptions that might --or might not-- result.  If Elsevier's own subscriptions decline more than it can absorb, then it can always change its archiving policy.  If postprint archiving does not cause a decline in subscriptions, then Elsevier has found a costless way to attract authors and get a jump on most of its competitors.

Pre-authorized postprint archiving is a real, if little-noticed, area of common ground between OA advocates and subscription-based journal publishers.  As I put it in the February SOAN, 2004 may be the year that "journal self-interest will join author and reader self-interest" in fueling new progress toward OA.

Two recent reports support Elsevier in thinking along these lines --if indeed it was thinking along these lines.  The new CIBER report on what authors want shows that most senior researchers know very little about OA.  A February report from JISC and OSI shows that when authors do learn about OA, they support it in very large majorities.  Put the two together and they suggest that demand for OA will grow roughly in proportion to familiarity with the concept.  Since the viral growth in familiarity with OA cannot be stopped, it is prudent for any publisher to position itself to appeal to this growing constituency, the sooner the better. 

CIBER report on what authors want (dated March 18, 2004, but released in May 2004)

JISC/OSI authors survey, February 2004
(I discuss these two reports further in the essay on author primacy, below.)

Here's another possibility:  By lending its weight to OA archiving, and helping it spread, Elsevier might be hoping to reduce the demand for OA journals.  But will any amount of OA archiving reduce the demand for OA journals?  Nobody knows.  It might satisfy some of the stakeholders who have been working for OA.  But it would not negate any of the benefits or undercut any of the arguments for OA journals.  For example, it would not turn their business models from sustainable to unsustainable or reduce the integrity of their peer review.  My guess is that even rapidly spreading OA archiving will do little to reduce the drive for OA journals.

Let me conclude by making two appeals.

* Appeal to authors:  Don't let your other quarrels with Elsevier distort your understanding of what just happened.  By all means weigh this development against the other advantages and disadvantages of submitting your work to Elsevier journals.  But don't prejudge the outcome by failing to take account of changing circumstances.  The new policy lets you provide free online access to your work, even if it doesn't solve other problems.  It's an opportunity worth seizing, even if you are dedicated to solving some of the other problems the new policy does not address.

* Appeal to publishers:  Consider following suit.  If you allow postprint archiving, then you can support OA without converting your business model.  You can provide this kind of OA immediately and let other OA-related policy changes depend on the results of experiments now under way at OA and hybrid journals.  You can match and neutralize the incentive Elsevier just created for authors to submit their work to Elsevier journals.


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 <SPARC-OANews-off@arl.org>.

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