Skip to main content
SearchLoginLogin or Signup

7-6. A Bill to Overturn the NIH Policy

Published onMar 14, 2019
7-6. A Bill to Overturn the NIH Policy

From “A bill to overturn the NIH policy,” SPARC Open Access Newsletter, October 2, 2008.

Six months after the new, strengthened version of the NIH OA policy took effect, it faces a bill in Congress to overturn it.

Rep. John Conyers (D-MI) introduced the Fair Copyright in Research Works Act (H.R. 6845) on September 9 [2008]. Conyers is the chairman of the powerful House Judiciary Committee, which is the House committee most responsible for copyright legislation, especially through its Subcommittee on Courts, the Internet, and Intellectual Property. The subcommittee held a hearing on the Conyers bill on September 11. (For links to the bill and hearing, see the bibliography below.)

The gist of the bill is to prohibit federal funding agencies from requiring grantees to transfer any rights or licenses to the government as a condition of funding, for any works (1) even partially funded by a source other than a US federal agency and (2) even partially reflecting the “meaningful added value” of any other party.

Because peer review undoubtedly counts as “meaningful added value,” because it is organized by private-sector journals at some expense to themselves, and because the NIH policy applies to peer-reviewed manuscripts, the bill clearly covers the NIH policy. The bill would not only overturn the NIH policy, but block similar policies at every other federal agency.

In December 2007, when Congress directed the NIH to adopt an OA mandate, it did not amend the US copyright statute. It merely asked the NIH to change its funding contract with grantees. But to overturn the NIH mandate, the Conyers bill would actually amend the copyright statute.

I won't do a section-by-section analysis of the bill. This has been done well by Mike Carroll and the Alliance for Taxpayer Access.

Mike's analysis is particularly good at showing how crude and unsurgical the bill is, and how it would inadvertently repeal a good deal of federal procurement law. Federal agencies which procure copyrighted works routinely make their purchase or funding conditional upon a license to use the work. But here I'll focus on the intended damage rather than the collateral damage.

The bill isn't called the “Stop Open Access Act” or the “Put Publishers First Act.” It's called the “Fair Copyright in Research Works Act.” The name suggests that the purpose of the bill is to right some copyright wrong. Wrong.

Public statements from the publishing lobby are explicit in suggesting some kind of copyright problem, even if they are not specific about what it is:

A joint statement by the DC Principles Coalition (DCPC) and the Professional/Society Publishers division of the Association of American Publishers (AAP/PSP) asserted that “the mandate is being implemented in a manner that is inconsistent with U.S. copyright law. …”

In a separate press release the AAP/PSP asserted that the mandate “undermin[es] copyright protection,” “diminishes copyright protections,” and “compromises copyright protections” for research articles.

The American Association of University Presses (AAUP) likewise asserted that the NIH policy would “weaken” and “diminish … copyright protection.”

The Copyright Alliance asserted that after the embargo runs on articles deposited in PubMed Central, the US government treats them as “public domain work[s].”

The American Chemical Society asserted that research articles covered by the new bill “would be protected by copyright laws” as if they were not already protected.

In his statement introducing the bill, Rep. Conyers said that it would “preserve the intellectual property rights of our Nation's researchers” and “restore intellectual property protections for scientists, researchers and publishers. …” (Not online.)

In an earlier joint statement (April 2008), before the Conyers bill was introduced, the AAP/PSP and DCPC asserted that the NIH policy denied “authors and publishers the benefits of their copyrights. …”

The DCPC, the AAP/PSP, and the Copyright Alliance all say that the NIH policy “forces publishers to surrender their copyrighted scientific journal articles. …” The Copyright Alliance goes a step further and says that a public research grant is no reason to “commandeer the resulting research paper and treat it as a public domain work.” BTW, the word “surrender” was first introduced into this context by the press release announcing the launch of PRISM in August 2007.

Note that not one of these statements says that NIH policy infringes copyrights.

If the publishers could claim infringement, they would. They would have a remedy at law and would not need to amend the law to get it. But a specific charge like infringement is too easy to evaluate and dismiss. Hence, we face nebulous charges like “inconsistency with copyright law” or “diminished copyright protections.”

The NIH uses a simple and elegant method to avoid infringement. When researchers publish an article based on NIH-funded research, they must retain the right to grant PubMed Central a non-exclusive license to disseminate a copy of their peer-reviewed manuscript. They may transfer all the remaining rights to publishers, if they wish, and they usually do.

There many ways to describe the result. OA through PubMed Central is expressly authorized by the copyright holders. Publishers no longer acquire full copyright to articles by NIH-funded authors, at least when those authors comply with the policy. Publishers don't acquire the rights they would need to negate the NIH's non-exclusive license or claim infringement. When researchers sign their funding contracts with the NIH, they are not committing any publisher to anything, let alone taking any intellectual property from anyone; they are only committing themselves to demand certain terms when they later write up and try to publish articles based on their funded research. The NIH is taking advantage of two important facts: (1) that authors are the copyright holders until they decide to transfer one or more of their rights to someone else, and (2) that researchers sign funding contracts before they sign publishing contracts. When NIH-funded authors do sign publishing contracts, they may only sign them subject to the terms of their prior funding contract.

The NIH succinctly describes this aspect of the policy in its policy FAQ: “Authors … may assign … rights to journals (as is the current practice), subject to the limited right that must be retained by the funding recipient to post the works in accordance with the Policy. …”

When I refer to this later, I'll just call it “retaining the key right.” It's what gives PMC permission to host OA copies of the author manuscripts. It's what prevents publishers from acquiring the full bundle of rights. It's what prevents the NIH policy from infringing the rights that authors may decide to transfer to publishers.

Let me digress on this point for a moment. There are two basic ways for funder OA mandates to avoid copyright infringement or secure permissions for OA. First, they may make an exception for publishers who won't allow OA on the funder's terms. Second, they may close that loophole and require grantees to retain the key right and use it to authorize the funder's OA, even if grantees may transfer all other rights to publishers. The first depends on permission from the publisher and the second depends on permission from the author prior to transferring rights to a publisher.

When publishers don't want to allow OA on the funder's terms, the first approach allows them to opt out. The second approach requires the author to find another publisher. In the first case, publishers may choose between accommodating the policy and resisting it. In the second case, they must choose between accommodating the policy and refusing to publish work by the agency's grantees.

The Conyers bill would block the stronger second path for US federal agencies, and allow the weaker first one. The NIH would have to retreat from assured author permission to contingent publisher permission, which is where it stood with the earlier, non-mandatory version of its policy, under which fewer than 10% of manuscripts based on NIH-funded research ever made it to OA from PubMed Central.

The strong second approach is taken by at least these eight funding agencies: the Arthritis Research Campaign (ARC, UK), Cancer Research UK (CR-UK), Department of Health (UK), Howard Hughes Medical Institute (HHMI, US), Joint Information Systems Committee (JISC, UK), Wellcome Trust (WT, UK), Medical Research Council (MRC, UK), and—for now—by the National Institutes of Health (NIH, US).

The weaker, first approach is taken by at least these seven: the Arts & Humanities Research Council (AHRC, UK), Canadian Institutes of Health Research (CIHR, Canada), National Cancer Institute of Canada (NCIC, Canada), Consejo Superior de Investigaciones Científicas (CSIC, Spain), Economic & Social Research Council (ESRC, UK), Higher Education Authority (HEA, Ireland), and Istituto Superiore di Sanità (ISS, Italy).

If the NIH policy doesn't infringe copyrights, or violate the federal copyright statute, then perhaps it violates our international IP treaties. Publishers have complained in the past that it violated US obligations under Article 13 of TRIPS (Agreement on Trade Related Aspects of Intellectual Property Rights) and the Article 9 of the Berne Convention. For example, see the AAP's 19 page letter to the NIH in May 2008,

But these objections would only apply if Congress had amended the Copyright Act in order to implement the NIH policy, which it did not do, or if it modified US law on copyright exceptions and limitations, which it did not do. On the contrary, it left copyright law unmodified and only modified the contract between the NIH and its grantees.

For detail, see the SPARC legal memorandum from July 2008, NIH Public Access Policy Does Not Affect U.S. Copyright Law.

The issue is also addressed in detail in the letter from 46 law professors to the Judiciary Committee, in response to the Conyers bill: “[T]he NIH Policy governs the terms of contracts, not exceptions to copyright law. As such, the Policy in no way implicates Article 13 of TRIPS or Article 9 of the Berne Convention, which address permissible copyright exceptions. These treaty provisions are completely silent on the issue of the terms a licensee can require of a copyright owner in exchange for valuable consideration.”

In his opening statement, Rep. Conyers didn't say the NIH policy violates US treaties, but he did say that it “could severely impact important negotiations with our international trading partners. Already we are hearing reports, in conversations at the World Health Organization and in other international forums, [that] the new [NIH] policy to limit the exercise of copyright by authors and owners is being taken as a sign that the United States is shifting its position away from being a strong proponent of intellectual property rights and enforcement.” (Not online.)

This appeal to trade issues conceals several problems. (1) The NIH policy doesn't violate the US copyright statute or treaties, and is completely compatible with intellectual property rights and their enforcement. If any of our trading partners think otherwise, they misunderstand the policy. Perhaps they only know what the publishing lobby has told them, which seems to be the case with several members of the Judiciary Committee. (Details below.) (2) The NIH OA mandate reflects US policy. In December 2007, it was approved by both houses of Congress and signed by the President. (3) A growing number of our trading partners are countries, or located in countries, with similar or even stronger policies. Worldwide, 29 funding agencies in 14 countries now have OA mandates: Austria, Belgium, Canada, the EU, France, Germany, Ireland, Italy, the Netherlands, South Africa, Switzerland, the UK, Ukraine, and the US. Without exception, all the medical research funders limit the embargo to six months, while the NIH allows 12 months. (4) If the NIH policy advances medical research without violating copyright, should we sacrifice it merely to ingratiate trading partners? What if their objections could be erased with a little accurate information? If this turns out to be a hard question of national priorities, do we really want to settle it in a committee devoted to copyright law? (5) If our trade partners understand that the policy doesn't violate copyright law, but still see it as a worrisome sign of national trends, and if the Judiciary Committee intervenes to act on their worries, then the Judiciary Committee is no longer defending the law of copyright, with its balance of competing interests, but the ideology of copyright maximalism.

But if the NIH policy doesn't infringe copyrights or violate our IP treaties, then in what sense is it “inconsistent” with copyright law? I don't see any non-cynical answers to that question. Perhaps publishers are deliberately blowing smoke, hoping that journalists and policy-makers will think infringement even where there isn't infringement. Perhaps they are dressing up a business setback as a copyright problem in order to get the attention of the Judiciary Committee. Perhaps publishers actually believe that their business setback, even without infringement, violates the “spirit” of the Copyright Act, as if the spirit of the act were to favor publishers over all other stakeholders, even for decisions made by authors at a time when authors, not publishers, are the copyright holders. Some of these theories are supported by the evidence that the Judiciary Committee is willing to credit the bogus copyright arguments, whether from misreading the NIH policy or a predisposition to help publishers.

In their rhetoric, publishers speak as if they are the copyright holders for these articles, and as if the NIH is blocking their full exercise of these rights or even expropriating them. But that is uninformed or deceptive. Because the NIH requires grantees to retain a key right, NIH-funded authors now transfer less than the full bundle of rights to publishers. Publishers don't like that, and it may be a problem for them, but it's not a legal problem. Despite their pose, publishers are not the copyright holders in these articles, without qualification, even after authors sign copyright transfer agreements. The NIH method of avoiding infringement means that there are plural rightsholders and divided rights in these articles: the authors have retained at least one and publishers have the rest. Publishers don't acquire the key right which would allow them to deny permission for OA or claim infringement or expropriation. As for the rights publishers do acquire, the NIH policy does nothing to diminish publisher freedom to hold and exercise them.

Have publishers forgotten this central feature of the NIH policy? Have its legal consequences still not sunk in? I find that theory hard to believe. It would entail that they haven't read, haven't remembered, or haven't understood the policy on which they have focused so much animus and lawyer time. And it doesn't square with their justified reluctance to claim actual infringement. But if they do understand this aspect of the policy, then we're only left with another cynical theory: that publishers deliberately stretch the truth by speaking without qualification as if they were the copyright holders for these articles. But strong or weak, the theory would explain a lot. If publishers did receive full copyright from authors, or if they believed they did, or if they had some reason to say they did, then their public rhetoric would make start to make sense. In that world, it would make sense to say that OA through PMC, against their wishes, would violate, diminish, or nullify one of their rights.

The snag, of course, is that the rhetoric is false, no matter what explains it. NIH-funded authors retain the key right and don't transfer full copyright to publishers. This is what I meant when I said (in SOAN for February 2008) that “publishers cannot complain that [the NIH policy] infringes a right they possess, only that it would infringe a right they wished they possessed.”

In the DC Principles Coalition press release on the Conyers bill, Martin Frank puts the publisher complaint this way: “NIH has undermined our publishing activities by diminishing a basic principle under copyright—the right to control the distribution of the works we publish.” (DCPC 9/12/08.)

It's true that if a publisher owns full copyright in a work, then it should have the right to control its distribution. But publishers of NIH-funded work do not own full copyright to those works. That's the point the publisher rhetoric keeps missing. The principle Frank thinks is diminished is intact. Instead, authors are retaining a right that publishers covet, something that authors are entitled to do under another basic principle of copyright.

Publishers can complain that under the NIH policy they don't acquire all the rights they want to acquire or all the rights they formerly acquired. They can complain that this creates problems for them. But they can't complain that the rights they do acquire are infringed, diminished, nullified, or made unenforceable by the NIH policy. Nor can they complain that authors don't have the right to transfer less than the full bundle to publishers. If the “benefit of copyright” is the power to enforce the rights you actually hold, then publishers enjoy the full benefit of copyright.

One version of the publisher complaint is that the NIH hosts and distributes peer-reviewed manuscripts without compensating publishers for their expenses in supervising peer review. At least this argument has a basis in fact and doesn't depend on distorted appeals to copyright law. Before we look at its flaws, we should recognize the truths on which it is built. It's true that publishers organize peer review. It's true that that costs them money, even when referees work without pay. It's true that the NIH policy applies to peer-reviewed manuscripts, as opposed to unrefereed preprints. And it's true that the NIH does not routinely compensate publishers for the service of organizing peer review. To assess the argument in its strongest form, I'll ignore the $100 million/year the NIH pays to journals to defray the costs of publishing NIH-funded research.

Why isn't that a slam-dunk?

It forgets or suppresses a critical fact. The NIH policy does not apply retroactively to published articles on which publishers hold full copyright. It applies to future articles published in conformity with the policy. The policy tells grantees, “When you publish future articles based on your NIH-funded research, you must demand the following terms from a publisher.” When NIH-funded authors approach publishers, they ask publishers not only “will you publish my article?” but also “will you publish under these terms?” It's a business proposition, and publishers are free to take it or leave it. But either way, they have no complaint. If they accept the proposition, they have no complaint: They have agreed to accommodate the policy. They have agreed to allow PMC to provide OA to the peer-reviewed manuscript after some embargo period. If they reject the proposition, they have no complaint: The authors go elsewhere and the publisher never invests in peer review for that manuscript.

Publishers may accept the business proposition only with reluctance. They may want a return to the days when they put terms to authors (“will you transfer full copyright?”) and when authors’ bargaining power was not buttressed by their funders or employers. Publishers may want to sweeten the deal, but they don't have a right to sweeten the deal. By contrast, however, authors do have the right to transfer all, some, or none of their rights to a publisher.

The expropriation argument hides the timeline of events and the role of publisher consent. It's not the case that publishers invest in peer review and then learn after the fact, helplessly, that the NIH will host copies of the peer-reviewed manuscripts. NIH-funded authors ask publishers in advance whether they are willing to publish under the terms required by the NIH. The decision is up to the publisher.

The current version of the NIH policy does not directly depend on publisher consent. But whether an article subject to the NIH policy appears in a certain journal does depend on publisher consent. The absence of publisher consent from the first step is what makes the NIH policy a mandate. Publishers cannot opt out except by refusing to publish NIH-funded authors or by charging a fee to do so, as the American Psychological Association once tried to do. The presence of publisher consent in the second step is what undercuts the publisher complaint about taking, surrender, expropriation, or lack of compensation. The NIH never disseminates the fruits of a publisher's investment in peer review without that publisher's knowing consent and cooperation.

One member of the subcommittee at the September 11 hearing, Bob Goodlatte (R-VA), suggested that we could moot the compensation question if the NIH policy applied to unrefereed preprints rather than to peer-reviewed postprints. But everyone else around the table rejected the idea, even Martin Frank representing the publishers. Frank said it would be “disastrous” to provide OA to unrefereed preprints in medicine, even with a disclaimer, since 90% of them are rejected by journals. (Piper 9/12/08.) This may be the first time that the Ingelfinger rule removed, rather than erected, an obstacle to OA.

The unsound compensation argument becomes dishonest when publishers disguise it as a copyright argument and introduce a hint of entitlement. There is nothing in copyright law requiring authors to transfer all rather than some of their rights to publishers, and there is nothing in copyright law protecting publishers from the risks of their own business decisions. It's dishonest to suggest that the letter or the spirit of copyright law should assure publishers of undiminished revenue or save them from hard bargaining. Publishers can complain that authors and funding agencies are waking up to the bad contracts authors have been signing with publishers, and waking up to their interest in retaining the right to authorize OA. But that's life in a changing world. Publishers can complain about the end of easy pickings, but they can't pretend that the easy pickings were ever an entitlement or that the end of easy pickings is inconsistent with law, any law.

The dishonesty is doubled when publishers pretend that they are protecting authors rather than themselves—for example, when the AAP/PSP says that the NIH policy denies authors the benefits of their copyrights, or when Rep. Conyers says that his bill would restore copyright protections for scientists. (More on this in the next section.)

Publishers sometimes put the compensation argument this way: supporters of funder OA policies want something for nothing. But on the contrary, supporters of funder OA policies want something for something. They want OA in exchange for the taxpayer investment in research. The “something for nothing” argument not only forgets that taxpayers pay for the underlying research, but also that publishers pay nothing to receive the written results. If publishers and taxpayers both make a contribution to the value of peer-reviewed articles arising from publicly-funded research, then what's the best way to split this baby? The NIH solution is a reasonable compromise: a period of exclusivity for the publisher followed by free online access for the public. The compromise is tilted further toward publishers by providing OA only to the peer-reviewed but unedited manuscript, not the published edition. If publishers want to block OA mandates per se, as the Conyers bill does, rather than just negotiate the embargo period, then they are saying that they want no compromise, that the public should get nothing for its investment, and that publishers should control access to research conducted by others, written up by others, and funded by taxpayers. That would be getting something for nothing.

The NIH policy is a compromise of mutual compensation: in consideration for publishing articles whose unedited manuscripts will become OA 12 months after publication, publishers get high-quality submissions, free of charge, reporting the results of expensive, quality-controlled research. In exchange for funding the research, at hundreds or thousands of times the cost of publishing the resulting articles, taxpayers get OA to a certain version after a certain delay. If the precise balance needs to be tweaked, then let's tweak it. But OA has to be part of the balance in order to give taxpayers something for something. And before we decide that a tweak favoring publishers is necessary, consider the evidence that it isn't: some publishers may be complaining, but with no tweaks at all they are almost unanimous in accepting the business proposition from NIH-funded authors.

William Patry, former Copyright Counsel to the House Judiciary Committee, puts it this way, with the accent less on publisher consent than balanced accounting:

[W]hy do you disregard the money that NIH has sunk in? Why shouldn't you flip the analysis and require the publisher, as a condition of publishing the article and charging for the journal, to reimburse NIH for some of NIH's expenses? STM publishers seem quite exercised over articles they pay nothing for being made available to the public, but apparently have no qualms about making money off of research funded by the public. Their moral outrage and accounting seems curiously unidirectional.

Finally, I shouldn't leave the compensation argument without noting a deeper problem. The publishers who oppose the NIH policy have adopted a business model which depends on access barriers. Their method for generating revenue is to block access by default, or create artificial scarcity, and then open it up for paying customers. OA publishers need compensation too, but their business models do not depend on access barriers or artificial scarcity. Toll access (TA) publishers believe that the OA business models are inadequate. We can debate that, for example, in light of the evidence that more than 3,000 peer-reviewed OA journals are finding ways to pay their bills, the fact that several for-profit OA publishers are already showing profits, and the fact that most of the money to support OA journals is currently tied up supporting TA journals. But in the end it doesn't matter whether the TA publishers are right or wrong to believe that their compensation depends on access barriers. The deeper problem, as I argued in July 2007, is “making a public commitment to use public money to expand knowledge and then handing control over the results to businesses who believe, correctly or incorrectly, that their revenue and survival depend on limiting access to that knowledge.”

If the TA publishers are right that they must erect access barriers to reimburse themselves, then we'd be foolish to let them be the only outlets for publicly-funded research. If the TA publishers are wrong, then we'd be foolish to overturn the NIH policy, and erect access barriers to publicly-funded research, just to satisfy their wrong belief.

Another version of the publisher complaint appears to concede that authors have a right to transfer all, some, or none of their rights to a publisher, but points out that the NIH policy is a “mandate” which interferes with author freedom.

Of course publishers are not mounting an expensive lobbying campaign in order to enhance author freedom. They are acting for themselves. Far more publishers than authors complain about the supposed restrictions of NIH-style funding contracts. Alma Swan's empirical studies have shown that “the vast majority (81%)” of researchers would “willingly” comply with an OA mandate from their funder or employer, and that 95% would comply willingly or reluctantly.

But what about the “mandate” problem? Does this limit author freedom in a way that should alarm legislators or in a way that is “inconsistent” with copyright law?

Let's not forget the sense in which the NIH policy is mandatory. The NIH is putting an OA condition on a voluntary contract. It doesn't require OA unconditionally, and couldn't possibly do so. The NIH is saying, in effect, that in exchange for a grant of public money, we ask you to retain one of your rights as copyright holder and use it to authorize OA. This is a contract that researchers are free to decline. Researchers who don't like the terms (and I haven't heard of any), don't have to accept public money for their research. Just as publishers are free to accept or reject papers by NIH-funded authors, authors are free to seek or avoid seeking NIH grants.

As the 46 law professors put it:

[T]he Policy does not create an involuntary transfer, a compulsory license, or a taking of the publishers’ or investigators’ copyright. … [I]f the investigator chooses not to receive NIH funding, the investigator has no obligation to provide the article to PMC or a copyright license to NIH. But if the investigator elects to receive NIH funding, he or she accepts the terms of the grant agreement, which include the requirement to deposit the article with PMC so that the article can be made publicly accessible within one year after publication. Because the investigator has this basic choice, the policy does not constitute an involuntary transfer.

The relevant author freedom here isn't the freedom to deposit in PMC, which existed under the previous (non-mandatory) version of the NIH policy, but the freedom to accept a research grant which requires deposit in PMC. But all contracts contain terms that require something or another. The NIH funding contract requires authors to turn in a report at the end of the grant period, to spend the money only on approved research-related expenses, and many other things. Now it also requires them to demand certain terms when publishing articles based on the funded research. Publishers don't object to the other mandatory contract terms, and couldn't. But if they object to this one on the ground that it's mandatory, then they must object to them all. More, they must object to their own publishing contracts, which also require authors to transfer certain rights.

The publisher argument here skirts hypocrisy, and the only way to avoid it is with breathtaking one-sidedness: authors should be free to sign publishing contracts with mandatory terms but not funding contracts with mandatory terms. If publishers are serious, they are calling on Congress to tilt an unbalanced copyright system further toward themselves.

As I was going to press, I just read about H.R. 286, which would put another mandatory condition on federal funding contracts: the grantees must use the metric system. Will the publishing lobby object that this limits researcher freedom? Or because it doesn't affect their own business model, will they let it go?

Let's take the money complaint seriously for a minute. Perhaps the publishing lobby understands that NIH-funded authors retain a key right, that copyright law allows them to do so, and that publishers acquire less than the full bundle. Perhaps the complaint is that the missing right prevents publishers from making money from the remaining rights. Publishers may be able to exercise their rights in full but they can't make as much money as before.

That may be their complaint, and it may even be true. The answer is a variation of my earlier answers. First, nothing in copyright law says they must always make the same amount of money they previously made. It's not an entitlement. Second, NIH-funded authors are still putting forward a business proposition (“will you publish this article under these terms?”) and publishers are still free to take it or leave it.

Moreover, if that is the complaint, why aren't publishers more candid in stating it? Why disguise their grievance as a copyright problem? Why speak as if they are the full copyright holders, somehow denied the rights of copyright holders? Why hide the fact that NIH-funded authors retain a key right and transfer only what's left? Why call the NIH policy “inconsistent with copyright law”? Why say that the NIH policy forces them to “surrender” their articles? Perhaps those are just exaggerations to get a hearing before the Judiciary Committee. But serious question: Why should publishers exaggerate about copyright violations to a Congressional committee which specializes in copyright law? Serious question: Why should we entrust the management of peer review to serial exaggerators?

If publishers admit that they have full legal authority to exercise the rights they acquire from authors, and merely want to acquire more rights from authors in order to make more money, then they'd have to admit that they are not suffering a legal injury. They'd have to admit that they are only facing a more difficult business climate because one source of their raw material is starting to offer it on less attractive terms, though still free of charge. They'd also have to admit that their preferred remedy would itself diminish copyright protection for authors, who are the original copyright holders here and who should be protected in their freedom to bargain away one of their rights in exchange for a large research grant.

Even if the NIH policy does lower the value of these works as investment opportunities for publishers, because publishers are no longer their exclusive distributors, it multiplies their value to authors and readers. I call this a step toward balancing the competing interests. But even if you wouldn't, it remains the case that publishers know when a submission comes from a NIH-funded author, know the terms, and accept it anyway.

For the record, by the way, it's far from clear that publishers can't make as much money as before. For the argument and evidence, see Sections 1–10 of this article from September 2007,

In response to Elias Zerhouni's claim at the hearing that “there is no evidence that [the NIH policy] has been harmful” to publishers, Martin Frank told Science Magazine “that some journal editors believe the new policy is leading to ‘fewer eyeballs coming to their sites.’ ” (Kaiser 9/11/08.) It's well-known and not surprising that OA archiving can reduce downloads from publisher web sites. But there's no evidence that these reduced downloads are reflected in reduced subscriptions. If there were, then publishers would cite the reduced subscriptions instead of the reduced downloads, to show that the policy is harming them.

Any systematic study of the harm question would also have to compare the few publishers who have publicly endorsed the Conyers bill with the hundreds who are voluntarily going beyond the terms required by NIH-funded authors:

Journals participating in PMC (many with embargo periods shorter than 12 months)

Journals submitting the published editions of articles by NIH-funded authors to PMC (not just the unedited manuscripts)

Journals depositing their articles in PMC, under open licenses, whether or not their authors are NIH-funded

For more detail on various publisher positions, see the OAD list of publisher policies on NIH-funded authors

One more attempt at a non-cynical theory: Publishers know that they have a right to resist the NIH policy, but they are mistaking the legal basis for that right. They think it derives from copyright law when in fact it derives from their background right to refuse to publish any work for any reason.

I have more to say about this distinction in an article from October 2006.

The two rights are independent, and the second is not rooted in copyright law. Publishers have a right to refuse to publish any work, whether or not they possess any rights in the work. I support that right, strongly, and I hope we all do. It would be intolerable for a publisher to lose the right to reject a submission or for the government to dictate what a publisher must publish. My point here is not that it is in jeopardy but that it is unrelated to copyright.

When the OA policy at a funder like the AHRC or CIHR (see above) includes a loophole for publisher copyright policies, then publishers have a copyright-based right to escape the OA policy. But funders like the NIH and Wellcome Trust close that loophole, deny publishers the easy opt-out of changing their in-house rules, and leave only the hard opt-out of refusing to publish work based on research funded by those funders. It's time for publishers to choose the hard opt-out if they really dislike the NIH policy, or to admit that they have chosen not to.

At the very least, it's time to stop pretending that they are victims of a legal wrong with an entitlement to restore the previous arrangement. Authors and funders here are not violating copyright; they are making a canny and lawful use of copyright in order to advance their own interests. Publishers have the same right to resist that every business has in the face of hard bargaining: just say no.


Here are eight facts I'd like every member of Congress to understand, if they ever have to cast a vote on the Fair Copyright in Research Works Act. Or, here are eight corrections to eight false impressions created by the rhetoric of the publishing lobby.

(1) There is no copyright infringement here. OA through PMC under the NIH policy is expressly authorized by the copyright holders.

(2) There is no violation of our IP treaties here. The treaties would only come into play if Congress had amended the Copyright Act in order to enact the NIH policy, which it did not do. The NIH policy only changes the contract between the agency and its grantees. (However, the Conyers bill would amend the Copyright Act.)

(3) The NIH policy regulates grantees, not publishers. It doesn't require publishers to provide OA, to permit OA, to relinquish anything they possess or acquire, or to do anything they are unwilling to do. It merely requires grantees to ask certain terms from publishers.

(4) There is no surrender, commandeering, or expropriation of publisher property here. There's a business proposition which publishers can take or leave. The only taking here is that most publishers are taking it.

(5) NIH-funded authors who comply with the new policy do not transfer full copyright to publishers. Instead, they retain a non-exclusive right to authorize OA through PMC. Consequently, publishers do not acquire full copyright in articles arising from NIH-funded research. Instead, at most, they acquire all but the right the author has retained. Publisher arguments that start with references to “their copyrighted articles” are already too sloppy deal justly with the actual policy.

(6) Publishers are failing to distinguish between (a) acquiring fewer rights from authors than they would like and (b) diminishing protection for the rights they do acquire. If they have a grievance, it's the former, not the latter. However, if one of these is inconsistent with copyright law, it's the latter, not the former.

(7) The NIH “mandate” is a mandatory condition in a voluntary funding contract, exactly as legitimate and consistent with copyright law as the mandatory conditions in publishing contracts.

(8) The mission of the NIH is to advance the public interest in medical research and health care, not the private interest of the publishing industry. Because the Judiciary Committee can set policy, its mission is not as clear-cut. But insofar as the committee sets policy on copyright, shouldn't its mission be less about creating artificial scarcity to publicly-funded research than promoting the progress of science and useful arts?

What's next?

Congress adjourned on September 26—not counting the special session to deal with the financial crisis—and isn't scheduled to reconvene until January 3, 2009. But it may come back during the break to deal further with the financial crisis, appropriations, or other hot issues. The Conyers bill is not one of those hot issues, and Howard Berman told Library Journal Academic Newswire that the subcommittee wouldn't take any action on the bill until at least January. However, Conyers or another proponent might attach the language to a bill that is moving toward a vote, during or after the recess.

When Congress reconvenes in January, we'll have a new Congress and soon after a new President as well. Whatever happens in the White House, Congress will have larger Democratic majorities than it has now. That won't necessarily help OA. John Conyers is a Democrat and John Cornyn (for example) is a Republican.

It helps us that the bill hasn't been endorsed by Howard Berman (D-CA) or Howard Coble (R-NC), the Chair and Ranking Minority Member of the Subcommittee on Courts, the Internet, and Intellectual Property. Richard Jones reports that Coble said the bill was “not your typical copyright issue” and that he “needed more time to learn and think about” it. According to Greg Piper, Coble hasn't endorsed the bill “because he had heard that some countries imposed similar requirements without violating IP treaty obligations.”

Berman took a stronger stand, saying that the Conyers bill would “turn back the clock” by overturning the NIH policy.

It also helps us that the bill may reflect turf rivalries among House committees more than member opposition to the NIH policy. In introducing the bill, Conyers was more passionate in defending the “sacred jurisdiction” of his committee, and denouncing the Appropriations Committee for bypassing his committee last year, than in opposing the NIH policy.

On the turf question, by the way, it's relevant that William Patry, former Copyright Counsel to the House Judiciary Committee, believes it's “absurd” to think that the NIH policy raises copyright issues or that it had to be reviewed by the Judiciary Committee.

It helps that nobody seems to want ownership of the bill. Conyers also told the Library Journal Academic Newswire that the bill was coming out of Berman's office, while Berman said it was coming out of Conyers’ office. (LJAN 9/16/08.)

Nevertheless, Conyers is a long-time opponent of the NIH policy and his opposition should not be doubted.

It helps that 46 law professors said there was no copyright problem with the NIH policy, and 33 Nobel laureates defended the NIH policy and charged that “the current move by the publishers is wrong.” (Law profs 9/8/08 and Nobelists 9/9/08.)

It helps that the US Copyright Office does not endorse the bill. Ralph Oman is a former Register of Copyrights (from 1994), a supporter of the bill, and was a witness at the September 11 hearing. But that only made other witnesses wonder about the current Register of Copyrights, Mary Peters. According to Andrew Albanese in Library Journal, the Copyright Office confirmed that Peters was not asked to testify. Sources in the office told Albanese that “the Copyright Office is … not … persuaded by publishers’ arguments regarding the NIH public access policy, and sees the recently introduced bill as unnecessary.”

It actually helps that subcommittee members were so uninformed about academic publishing and the NIH policy. It discredits the publisher lobbyists and suggests that a little more information in the right places could turn the tide. For example, some members didn't know that scholarly journals didn't pay authors or peer reviewers. Others didn't know that the NIH paid out $100 million/year to support the costs of publishing journal, which is about 30 times more than it spends implementing its public access policy.

In the end, however, it doesn't matter whether the Conyers bill arises from turf rivalries among House committees or objections to the NIH policy, and it doesn't matter whether the objections to the NIH policy arise from copyright problems or an unabashed desire to protect a powerful private interest at the expense of the public interest. The bill could pass anyway. Nor do divisions within the committee matter if proponents of the bill can attach its language to another bill and bypass the committee. That's why we have to continue our efforts to defend the NIH policy and block any effort to overturn it, from any direction.

If you're an American citizen and you contacted your Congressional delegation before the recent recess, thank you. Be prepared to do so again in the new session.

It would help if sympathetic publishers would speak out even now, during the recess. The Association of American Publishers division of Professional/Scholarly Publishing (AAP/PSP), the DC Principles Coalition (DCPC), and the American Association of University Presses (AAUP) have all endorsed the Conyers bill. In their letters to Congress, each group pretended to speak for its members. Rockefeller University Press has already spoken out, disavowing the AAUP statement and noting that the association did not consult its members before endorsing the bill.

University faculty, librarians, and administrators should ask their university presses to take a similar step. Publishers willing to do so should issue a public statement and send copies to their Representative, Senators, and the leadership of the House and Senate Judiciary Committees.

In August 2007, I made this assessment:

There is a rising awareness of copyright issues in the general public, rising frustration with unbalanced copyright laws, and rising support for remedies by governments (legislation) and individuals (Creative Commons licenses and their equivalents). Copyright laws are still grotesquely unbalanced, and powerful corporations who benefit from the imbalance are fighting to ensure that [these laws] are not revised in the right direction any time soon. But in more and more countries, an aroused public is ready to fight to ensure that they are not revised in the wrong direction either, something we haven't seen in the entire history of copyright law. However, this only guarantees that the content industry will have a fight, not that users and consumers will win.

Already the opposition to the Conyers bill is proving this thesis, though the outcome is still up in the air. Even governments are acknowledging the change, which may be why those seeking the next ratchet-click of copyright maximalism have had to draft the Anti-Counterfeiting Trade Agreement in secret.

If the Conyers bill is defeated, or dies without a vote, we will not have mobilized for nothing. We will have saved a good policy from repeal. We will have made it harder for the publishing lobby to come back. It has now had its hearing, and won't be able to complain that it never got one. It got its shot at a copyright argument before the Judiciary Committee, and won't be able to complain that it never got one. It won't be out of options or give up, but it will be at least one step closer to adaptation.

Here are the basic links for the Conyers bill and the September 11 hearing on it:

Fair Copyright in Research Works Act (H.R. 6845) (Conyers bill)

(the final colon is part of the URL)

An image scan of the text of the bill

Open Congress page on the H.R. 6845

The hearing on the bill conducted by the House Judiciary Committee, Subcommittee on Courts, the Internet, and Intellectual Property, September 11, 2008

The video webcast of the hearing

Transcript of the hearing (from a private company, not OA)

House Judiciary Committee

Subcommittee on Courts the Internet and Intellectual Property


Also see the sequel to this article, “Re-introduction of the bill to kill the NIH policy,” SPARC Open Access Newsletter, March 2, 2009.

No comments here
Why not start the discussion?