Balancing author and publisher rights
SPARC Open Access Newsletter, issue #110
June 2, 2007
by Peter Suber
In order for authors to provide OA to their own work, they don't need to retain full copyright, and in order for publishers to publish, they don't need to acquire full copyright.  This raises the hope that we might find a balance giving each side all it needs.  But even with good will on both sides, this win-win compromise may be out of reach; each side might give and receive significant concessions and still not have all it needs.

There were two developments in May that could affect the balance between author and publisher rights.  First, a group of universities adopted an "author addendum" to modify standard publisher copyright contracts and a pair of non-profits enhanced their own author addenda.  Second, a group of publishers adopted a position statement on where the balance lies.  Not surprisingly, the two groups strike the balance in different places.  I'll look at them in order.

(1) Author addenda

An "author addendum" is a lawyer-written document that authors sign and staple to a publisher's standard copyright transfer agreement.  It modifies the publisher's contract to allow authors to retain some rights that the default contract would have given to the publisher.  Because it's a proposed contract modification, the publisher may accept or reject it.

The idea is that most authors are unfamiliar with copyright law and intimidated by the prospect of negotiating contract terms with publishers.  More authors are willing to take the first step, and to more likely to make a prudent request, if they can start with a written proposal crafted by a lawyer to promote OA.  Another advantage is that if a specific addendum gains institutional backers, then authors who use it gain bargaining power.

One problem that author addenda won't solve is author fear that merely asking for different contract terms will cause publishers to reject an already-accepted paper.  But I haven't heard of a single case in which this has happened.  The fear is groundless; there's no harm in asking.

Author addenda may not be necessary for the approximately 70% of green subscription journals that already give blanket permission for postprint archiving.  But addenda (or their equivalent in individual requests or negotiations) are necessary for the ungreen 30%.  Addenda can be helpful even when not strictly necessary, for the green 70%, by assuring a continuing legal basis for OA in case the journal later changes its access policy.  They also help when greenish journals aren't green enough, for example, because they prohibit deposit in certain repositories, impose fees or embargoes on self-archiving, or limit re-use rights. 

There are many author addenda in circulation.  The major ones, in chronological order, are from SPARC (May 2005), MIT (January 2006), Science Commons (June 2006), OhioLINK (August 2006), SURF-JISC (October 2006), and the Committee on Institutional Cooperation (February 2007).

The Committee on Institutional Cooperation (CIC) is a OA-friendly consortium of 12 research universities:  Chicago, Illinois at Chicago, Illinois at Urbana-Champaign, Indiana, Iowa, Michigan, Michigan State, Minnesota, Northwestern, Ohio State, Pennsylvania State, and Purdue.  If you're not affiliated with a CIC institution, you may know it best as the group behind the first open letter from university provosts in support of FRPAA (Federal Research Public Access Act) in July 2006.

The CIC letter, eventually signed by 25 provosts, triggered a wave of other letters now totalling 132 signatures.

Early in 2007 the CIC provosts wrote a statement in support of OA and a draft author addendum, and began circulating them to member institutions.  It released the final version on May 17, 2007.

The CIC author addendum retains three rights for authors:  (1) a non-exclusive right to make and use derivative works, even for future publication, (2) a non-exclusive right to self-archive the published version six months after publication, in any repository, and (3) a non-exclusive right for the author's institution to use and copy the work for any activity at the institution.

In my view, the only significant omission is a non-exclusive right to provide *immediate* OA to the final version of the author's peer-reviewed manuscript (not the published edition).

Even before the CIC approved the final language of the addendum, CIC member institutions began adopting it:

The University of Illinois at Urbana-Champaign adopted it on May 3.

The University of Minnesota adopted it on May 3.

The University of Wisconsin at Madison it on May 7.  (At the same time, Wisconsin strengthened the CIC language by adding a section stating that the publisher agrees to the addendum by publishing the article.)

Other CIC member institutions may soon follow suit, and it's possible that some have already done so without fanfare. 

What's new here is that universities are endorsing an author addendum.  This is a powerful signal that the institutions support OA and want authors to self-archive.  I hope it's also a signal that the same institutions are ready to do even more to support OA archiving.

I don't know of any university that *requires* faculty to use an addendum.  Wisconsin, for example, will merely encourage it.  And it's not clear what form the encouragement will take.  Will it be limited to the abstract encouragement of passing a resolution in the Faculty Senate?  Or will there also be some case-by-case encouragement?  Either way, will the adopting universities recognize any exceptions?  Will they recommend use of the addendum even at the green 70% of non-OA journals that already permit postprint archiving?  In a standoff between a publisher and faculty member, what will universities do to support their faculty member? 

Here's the bigger question:  What else will these universities do to encourage OA archiving?  If they take the step of adopting an author addendum, they should also adopt a policy to require OA archiving.  If permission is not a problem (because publishers already give it or because an addendum worked), what will these institutions do to insure that faculty postprints are actually archived? 

About 30 universities, departments, or labs around the world mandate or strongly encourage OA archiving, and none of them needed the initial step of adopting an author addendum.  That suggests that the permission problem is not the real hurdle, even if it's a real problem in a minority of cases and a perceived problem in many more.  The permission problem is worth solving, but we have to remember that solving it is only a means to the end of OA.  Universities adopting an author addendum are moving in the right direction, but they must keep moving.  Permission for OA isn't yet OA itself.

We don't yet know how the adoptors of the CIC addendum will follow it up.  But so far the news is good.  If just one university adopted an addendum, then publishers might refuse the addendum's terms or even refuse to publish articles by that university's faculty.  But as more universities join the critical mass, more publishers will accommodate them.

Other universities should see the action by Illinois, Minnesota, and Wisconsin as an opportunity to create a critical mass.

But that isn't the end of the story, even for May 2007.  The same day that CIC released the final version of its addendum, SPARC and Science commons (SC) announced that they were consolidating their addenda, strengthening them, and releasing an online tool to produce customized versions of them.

Together SPARC and SC now offer four coordinated addenda, depending on the author's needs.  The online "Addendum Engine" lets authors select the addendum best for them and print a copy with article and publisher information already filled in.

One of the four is the pre-existing MIT addendum from January 2006.  The others are the three published by Science Commons in June 2006 with one of them modified to incorporate elements from the SPARC addendum of March 2005.  All three of the SPARC-SC addenda allow the author to retain the right to make, use, and distribute derivative works.  One uses a Creative Commons (Attribution, Non-Commercial) license to free up users as well as the author.  One allows immediate self-archiving of the published version of the article, and one allows immediate self-archiving of the peer-reviewed manuscript and only delayed self-archiving of the published version.

The Scholar's Copyright Addendum Engine

The new Addendum Engine should make it easier for authors to use an addendum, whether or not their universities stand behind their decision.  But at the same time, they make it easier for universities to endorse an addendum, knowing that online tools (which can be hosted locally) simplify the process for authors. 

I believe that each of these addenda gives publishers all they need, but not all publishers would agree.  The only formal publisher response I've see to any addendum is the joint ALPSP-STM statement in response to the MIT addendum (June 27, 2006), in which the publishers said that "Author posting (of any version of an article) immediately upon publication risks competing with the journal itself."  But even this position, which may take more ground than publishers really need, is compatible with the CIC addendum.

When universities adopt an addendum, they can frame the message for their own faculty in many different ways.  Here's how I'd do it:  "When you publish a research article, the prestige of your journal is not enough.  Its access policy matters at least as much.  We're going to be looking at both.  We still want you to get the imprimatur of a good journal.  Among other things, it shows that you're good.  But we also want your work to be accessible to those who need it and used by those who can use it.  To make this happen, you could publish in good open-access journals or in good conventional or non-OA journals that let you provide OA on your own, for example through our institutional repository.  The least important reason why we have a publish-or-perish requirement is for you to prove that you're good.  The most important reason is to share the knowledge produced at this institution with everyone who can benefit from it."

(2) Publishers describe how they'd balance author/publisher rights

Meantime on May 9, three publisher associations released a position paper titled, "Author and Publisher Rights For Academic Use: An Appropriate Balance."   The three groups were the ALPSP (Association of Learned and Professional Society Publishers), AAP/PSP (Association of American Publishers / Professional/Scholarly Publishing), and STM (International Association of Scientific, Technical & Medical Publishers). 

For convenience I'll call the authors of this document "the publishers".  But everyone should understand that not all publishers share the views set forth in this document, perhaps not even all publishers who belong to the ALPSP, AAP/PSP, or STM.

The heart of the document lies in two assertions, one on author rights and one on publisher rights:

[1] Academic research authors and their institutions should be able to use and post the content that such authors and institutions themselves provide...for internal institutional[,] noncommercial research and education purposes; and

[2] Publishers should be able to determine when and how the official publication record occurs, and to derive the revenue benefit from the publication and open posting of the official record (the final published article), and its further distribution and access in recognition of the value of the services they provide.

* The first statement, on author rights, seems to say that free online access should be limited to the author's own institution. 

But I'm not sure.  Does the adjective "internal" apply only the word immediately following it or to all the words following it to the semi-colon?  If the former, then the statement would allow OA postprints to be used for non-commercial research whether or not it was internal to the author's institution.  That's good; authors need at least that much, and OA archiving provides it.  If the latter, then the statement would restrict the use of archived articles to the author's institution.  That would mean the end of OA archiving, which by design makes content accessible to all users everywhere.  If the publishers meant to limit free access to the author's institutions, then their position is one-sided, insufficient, and a retreat from the permissions most publishers already give to post to an institutional repository. 

The sentence in the press release slightly supports the first reading by inserting a comma after "institutional", while the sentence in the body of the paper slightly supports the second reading by omitting the comma.

Several contributors on several discussion lists interpreted the statement in the second way and publishers on the same lists did not contradict them.  So it appears that publishers did mean to limit free access to the author's institution.

But what does it mean for publishers to have meant that?  The question arises because most of the same publishers permit OA archiving without delay, fee, special requests, or scope restrictions.  The publishers' position paper is at odds with their own copyright transfer agreements. 

In explaining this disparity, one possibility is that publishers permitted self-archiving in the first place without quite understanding what they were getting in for, and are now testing the waters for a retraction.  (That is, they underestimated the power of OAI interoperability and didn't anticipate crawling by Google, Yahoo, Microsoft, and Scirus, let alone archiving mandates from funding agencies and universities.)  They may see the position paper as the first step toward creating a "new normal" in which permission for self-archiving is limited to permission to make a work accessible to the author's own institution.  Or, they may have no plans for a retraction, but feel the need to push back against the rising talk of author rights, for example, represented by the proliferating author addenda.  They may eventually revise their copyright agreements to match the position paper, which would obstruct research, alienate authors, and deter submissions.  Or they may revise the position paper to match their copyright agreements, which they would probably cast as a concession to authors.  But as long as they do neither, it's hard to know their real position.

The publishers don't even acknowledge the disparity between the position paper and the widely-granted permission for unrestricted postprint archiving.  In one section of the position paper, alluding to the permission to self-archive, they write that, "[g]iven the scholar-friendly nature of most academic journal publishers' copyright policies, a further question may be raised as to whether anything more is needed from publishers in order to accommodate the needs of academics and academic institutions."  If publishers do grant permission for unrestricted self-archiving, then indeed authors may not need more.  But if publishers are cutting the heart out of that permission, then authors need a lot more.

What matters for authors is that about 93% of surveyed non-OA journals currently allow preprint archiving without institutional or geographic limits, and about 70% allow equally unrestricted postprint archiving.  Our job is to make sure that authors understand that self-archiving is easy, lawful, beneficial, and an opportunity that only they can seize.

Note that the position paper also limits free access to non-commercial use.  While I've often argued that researchers should permit commercial re-use of their work, publishers needn't permit commercial re-use of theirs.  On the other hand, publishers only need to restrict the commercial distribution of their publications, not the use of their publications by researchers with commercial plans or motivations.

* For the rest of the reason why the publisher position is unbalanced we have to look at the next statement, on publisher rights.  Unfortunately it's even more difficult to parse. 

"Publishers should be able to determine when and how the official publication record occurs...."  If the occurrence of a record is the publication of an article, then determining "how" it occurs is the whole question.  Giving this entirely to publishers is to give up the quest for balance. 

"Publishers should be able to...derive the revenue benefit from the publication...." Is the "revenue benefit" more than just "the revenue"?  Publishers do have the right to sell their publications, of course.  That's non-controversial.  But are publishers also claiming the right to all the revenue that anyone can make from it?  What if I'm offered an honorarium to speak at a conference on the strength of my publication?  What if Google indexes the repository copy and puts ads on the page of search returns?  What if I link to a copy, even the publisher's copy, from a page hosting ads?  What if a team of industrial scientists pays for access but uses the resulting knowledge to make a product which their company sells for a profit?  I'm not saying that the publishers are claiming all this revenue, merely that the statement needs clarification.

"...and open posting of the official record (the final published article)..."  What's the open posting of the official record?  What's the revenue benefit of the open posting of the official record?  If this is about open access to the publication, then what revenue are the publishers talking about?

"... and its further distribution and access in recognition of the value of the services they provide."  I understand that publishers want the revenue from the paper's "further distribution and access".  But is this "further distribution" more than the publication already mentioned?  If so, what does it include?  Self-archiving?  If so, again, what revenue are they talking about?

More importantly, are the publishers saying they deserve the revenue because of the value of the services they provide?  It seems so.  But there are three problems here.  First, authors, referees, and funders provider valuable services that enhance the same final product, competing with the publishers' claim to exclusive rights.  I'll say more about this one below.  Second, a significant fraction of publisher revenue doesn't come from the value they add but from price increases made possible by monopoly power and market dysfunction.  Reducing their prices to the value they add would be a nice change.  And third, in order to keep the revenue stream flowing, publishers take many steps that actually subtract value from the final product, such as password protection, packaging in locked PDFs, cutting good articles solely for length, turning processable data into unprocessable images, and turning gifts into commodities which may not be further shared.

Let's look more closely at the first of these.  The publishers are arguing that because they add value to the publication, they deserve exclusive rights in it, in effect letting them control access beyond the author's own institution.  This is neither balanced nor good for research.  Publishers do add value, primarily the organization of expert volunteers who provide peer review.  But no matter how many other forms of publisher-added value we recognize, and no matter how we estimate their overall benefits, there's no doubt that publishers add *less* value to the final product than authors, who do the research and writing, and funders, who pay for the original research.  When there are funders in the picture at all, their support is usually at a level greater than the cost of publication and sometimes at a level thousands of times greater.  But in current practice authors and funders don't get the right to control access to the final product, and in this document publishers want to perpetuate the arrangement in which the right to control access ends up in their hands, not those of contributors who add greater value. 

There are two main reasons why we find ourselves in the odd situation in which publishers get to control access even though they add less value than authors or funders.  The first is that publishers demand compensation for their services, while authors and funders do not.  The second is that publishers believe the only way to be compensated is to control access and charge for it.  This is their business model from the age of print, when it was physically impossible to make perfect copies for a worldwide audience at zero marginal cost.  Their business model depends on scarcity, which for digital texts in a networked world is always artificial scarcity.

Publishers are not appealing to the principle that adding value carries the right to control access.  If they were, then all contributors who added value would have to share control.  Nor are they appealing to the principle that the right to control access belongs to the contributor who adds the greatest value.  If they were, they'd have to make a serious argument that their contribution is more valuable than the author's or funder's.  They are demanding the right to control access because they need compensation for their services and choose a business model that depends on access barriers and artificial scarcity. 

Even if we don't think this situation is perverse and cries out for change, at least we should notice that their position is not about balance.  It's about what publishers need or want, regardless of what authors need or want. 

Am I saying that publishers should join authors and funders in working without direct monetary compensation?  Not at all.  Publishers deserve to be paid for the value they add.  But it doesn't follow that they deserve to control access or that they deserve a package of exclusive rights that bars author-initiated OA.  These extra demands don't arise from the value they add but from an access-limiting business model that is optional, obstructive, and obsolete.

Instead of starting from the proposition that publishers add value, and deserve whatever they think it takes to compensate them for it, including artificial limits on access to knowledge, I suggest that it's more balanced to start from the proposition that many contributors add value, that they all deserve suitable compensation, and that letting publishers limit access prevents this equitable division of rewards, harms the other contributors, and harms research.

The position paper makes several other, lesser assertions.

* "Exclusive rights also provide a legal basis for publishers to...enforce copyright claims with respect to plagiarism...." 

It's inaccurate and disingenuous to argue that publishers need exclusive rights to prosecute plagiarists.  First, the rights are rarely used this way.  Plagiarism is typically punished by the plagiarist's institution, not by courts --that is, by social norms, not by law.  Second, if it's ever desirable to pursue a plagiarist in court and authors don't give publishers the right to do so on their own, then authors retain that right to use as they see fit.  Third, many authors would rather have a larger audience and impact than give their publisher the seldom-used legal tools to prosecute plagiarists.  Authors should make this decision, not publishers.  Finally, if an author discovers a plagiarist and the publisher really wants to get involved, the author can always delegate the publisher to act as his/her agent.  For this purpose, publishers don't need rights from the time of publication, nor do they need exclusive rights, let alone a policy to limit access to the author's work.

Behind this argument there's a confusion of plagiarism and copyright infringement.  Someone can commit plagiarism without infringing copyright (by copying a fair-use excerpt and claiming it as one's own) and infringe copyright without committing plagiarism (by copying a larger excerpt but with attribution).  One can also commit both together (by copying a large excerpt and claiming it as one's own), but that doesn't collapse the distinction.  One can commit just about any two offenses together.

* Publishers "are concerned about the potential to waste monies with unnecessary duplicate systems" (when public funding agencies mandate OA archiving for publicly-funded research). 

It's also disingenuous for publishers to argue that OA mandates at public funding agencies will lead to wasteful duplication.  Some publishers do provide OA to some content when it's sufficiently old.  But this is a far cry from providing OA to virtually all publicly-funded research within six months of publication.  If ALPSP, AAP/PSP, and STM are saying that the voluntary efforts of their members will approach what FRPAA (for example) would mandate, then the duplication argument starts to make sense.  But in that case they have to stop arguing that OA to publicly-funded research would kill their revenues, kill their journals, and kill peer review.  They can't have it both ways.

* "[F]unding agencies, search engines, and other third parties who wish to use or distribute the publisher versions of journal articles should only do so upon consultation and under an agreement with the publisher...."

It's vast over-reaching to say "use or distribute" here.  Third parties like readers may lawfully "use" publisher versions in countless ways without consultation or permission. 

If we limit the statement to "distribute", it's fair enough but not really responsive to funder mandates for OA archiving.  These funder mandates do not apply to the published version of an article, but only to the final version of the author's peer-reviewed manuscript. 

* Finally, publishers argue that fair use is limited to print:  "there are exceptions and limitations to copyright laws that may in certain limited circumstances permit the copying of journal articles for certain purposes, but...these exceptions are thus far limited to traditional photocopying and do not permit the exploitation of such materials over the Internet."

This is simply untrue.  For example, the US District Court of Nevada ruled that it was fair use for Google to index and cache a copyrighted online news story by Blake Field (Field v. Google, 2 F. Supp. 2d 1106 (D. Nev. 2006)).  The Ninth Circuit Court of Appeals ruled that it was fair use for Arriba to display thumbnails of Kelly's copyrighted online photographs even if it might not be fair use to display full-sized copies (Kelly v. Arriba Soft, 280 F.3d 934 (CA9 2002)).

Even if publishers could subtract fair use from the online freedoms of researchers, they would simply have to add even more to the authors' pan of the scale if they really want to achieve a balance.

* I haven't answered the big question whether a win-win balance can be struck that gives both authors and publishers all that they need.  I'm not ready yet.  I know the publishers' position statement is not that balance, and they may say that no author addendum they've yet seen is that balance either.  Universities adopting author addenda are at least trying to approach such a balance by correcting the imbalance that currently favors publishers.  The publisher position statement is trying to tilt it further toward publishers.

If publishers really need the rights this position paper says they need, then no win-win balance is possible.  The closest we could come is a set of mutual concessions that at least one side will find insufficient.  But if some of the positions the publishers are taking here, like limiting free online access to the author's institution, are just trial balloons, then we can still hold out hope of a win-win balance.  Much constructive work remains to be done.  I expect it from individual publishers who grant authors more rights than this position paper would allow and then demonstrate that they can still survive and prosper, perhaps even increasing their submissions.

Committee on Institutional Cooperation (CIC)

The CIC Author Addendum (May 17, 2007)

University of Illinois at Urbana-Champaign announcement
--The draft resolution adopting the addendum (apparently the same as the adopted resolution)
--UIUC copy of the addendum

University of Minnesota
--The draft resolution (apparently the same as the resolution as adopted)

University of Wisconsin - Madison
--Resolution as adopted by the Faculty Senate, May 7, 2007
--University press release, May 7, 2007

Science Commons, SPARC announce new tools for scholarly publishing, a press release from Science Commons and SPARC, May 17, 2007.

The Scholar's Copyright Addendum Engine (SCAE)

Instructions for institutions wishing to host a copy of the Scholar's Copyright Addendum Engine

Science commons home page on the Addendum Engine

Science Commons FAQ on author addenda

Author and Publisher Rights For Academic Use: An Appropriate Balance, May 2007.  A joint position paper by ALPSP, AAP/PSP, and STM.
--The ALPSP announcement
--The ALPSP text
--The STM announcement
--The STM text

Les Carr, PSP and ALPSP: j'accuse!  American Scientist Open Access Forum, May 11, 2007.
--Discussion thread based on Carr's message.

Les Carr, Starting a community response to ALPSP/PSP?  American Scientist Open Access Forum, May 12, 2007.

Kevin Smith, Publisher position on author rights, Duke Scholarly Communications, May 20, 2007.


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