Opening Access to Research
by Peter Suber
We all understand why free online music sharing is controversial. Musicians make a living by selling their work, and widespread unauthorized sharing could slash their revenue. File sharers respond with evidence that obscurity is more costly than piracy, for those below the rank of superstar, and that unauthorized online sharing can actually boost sales. But instead of entering that debate here, let’s just note its existence and take a detour around it.
Imagine a group of authors who do not make a living by selling their work, and who actually authorize free online sharing. They don’t take this unorthodox path because they are rich. They do it because their topics, genres, purposes, incentives, and institutional circumstances lead them to write for impact, not for money. Their careers and reputations have more to gain from the larger audience of interested readers than the smaller audience of paying customers.
You won’t find many novelists or journalists in this group. But you will find a growing number of researchers from every field of the sciences and humanities.
Researchers do not make a living by selling articles to peer-reviewed scholarly journals. In general, scholarly journals don’t buy articles from authors or pay royalties. Scholars may write books for money, but they write journal articles for impact. A growing number of them realize what this means in the digital age, and decide to authorize free online sharing of their peer-reviewed journal articles.
Most researchers make a living by working at universities, which pay salaries. The primary effect of those salaries is to enable faculty to put food on the table, of course, and work as professors rather than paint houses or deliver pizzas. But an important side-effect of those salaries is to insulate faculty from the literary market. Musicians could go hungry if they spent months recording an album of avant-garde music of interest to just a handful of friends around the world. But salaried faculty will not go hungry if they spend months writing an article on a cutting-edge topic of interest to just a handful of colleagues around the world. Faculty have no financial reasons to worry about picking narrow topics, defending unpopular ideas, or opposing conventional views. They are free to focus on what is likely to be true rather than what is likely to sell.
By design or accident, this system protects serious inquiry. If university salaries didn’t insulate faculty from the literary market, and if scholarly journals paid royalties on research articles, then faculty incomes would be tied to the popularity of their ideas. In that world, profit-seeking would distort truth-seeking, more than it does in this world.
When the first scholarly journals were launched in London and Paris 350 years ago, they quickly superseded letters and books as the premier method for distributing new research. Journals gave authors a wider audience than private letters, and a more rapid public time-stamp than books. The faster turnaround and authoritative time-stamp helped establish an author’s priority over others who might be working on the same questions. These advantages more than made up for the fact that journals didn’t buy articles or pay royalties.
The system worked well until the prices of print journals began to rise faster than inflation in the 1970’s. Prices have risen faster than inflation, and faster than library budgets, continuously for almost four decades. Even the wealthiest academic libraries are forced to make budget-driven cancellations every year. Harvard does so, for example, and recently told its faculty that journal prices are untenable and unsustainable. This is sometimes called a journal pricing crisis, but could just as well be called a research access crisis.
One more piece of context: In 2010, Elsevier, the largest publisher of scholarly journals, reported a profit margin of 36%. In the same year, ExxonMobil reported only 28%.
In February 2002 the Budapest Open Access Initiative gave a name — “open access” (OA) — to free online sharing of research. It described strategies for achieving OA, and called for it in every field and country. Here’s the BOAI definition:
By “open access” to [peer-reviewed research literature], we mean its free availability on the public internet, permitting any users to read, download, copy, distribute, print, search, or link to the full texts of these articles, crawl them for indexing, pass them as data to software, or use them for any other lawful purpose, without financial, legal, or technical barriers other than those inseparable from gaining access to the internet itself. The only constraint on reproduction and distribution, and the only role for copyright in this domain, should be to give authors control over the integrity of their work and the right to be properly acknowledged and cited.
In short, OA literature is digital, online, free of charge and free of most copyright and licensing restrictions. Ideally, the only restriction on use is an obligation to attribute the work to the author.
Why allow even this restriction? The purpose of OA is to remove barriers to the scholarly uses of scholarly literature, without harming scholars. There’s no legitimate scholarly need to suppress attribution to the texts we use. And in any case, suppressing attribution would hurt authors. If they don’t get paid for their articles, at least they should get credit. Their impact and careers depend on that.
We could turn the question around: Why remove any restrictions at all? The answer is to share knowledge and accelerate research. Barrier-free access helps readers find and retrieve the research they need, and helps authors reach readers who can apply, cite and build on their work. Knowledge has always been a “public good” in the theoretical sense that consumption doesn’t deplete it (it’s “nonrivalrous”) and consumption is available to all (it’s “nonexcludable”). OA makes knowledge a public good in practice.
OA is made possible by copyright-holder consent and the internet. We don’t expect copyright-holder consent from musicians, novelists, movie-makers, or other creators who hope to make a living by selling their work. (Despite this, we do get consent from some of them.) The real opportunity for OA is among scholars who write journal articles for impact, not for money.
There are many ways to deliver OA, but two vehicles dominate. OA can be delivered by peer-reviewed journals (in the jargon, “gold” OA) or by repositories (“green” OA).
OA journals can use any of a dozen different business models for paying their bills. (OA does not presuppose that publishing is costless.) OA publishers can be for-profit or non-profit. Some use traditional methods of peer review, in order to tweak just the access variable of scholarly journals. But others use new and innovative models of peer review taking advantage of the digital network for gathering and disseminating peer judgments. More than 8,000 peer-reviewed OA journals are now listed in the Directory of Open Access Journals.
OA repositories are simply online collections or databases of digital content. Some aim to disseminate the research output of a field (“disciplinary” or “subject” repositories), while others aim to capture the research output of a university, lab, or foundation (“institutional” repositories). With permission, they can host articles already peer-reviewed and published by journals. More than 60% of conventional, subscription journals give standing permission for authors to deposit their peer-reviewed manuscripts in OA repositories. OA repositories can also host content not published in journals, such as unrefereed preprints, conference presentations, theses and dissertations, datasets, audio, video, source code, and content digitized from print or microfilm. More than 2,100 OA repositories are now listed in the Directory of Open Access Repositories (OpenDOAR) and more than 2,900 in the Registry of Open Access Repositories (ROAR).
About 30% of peer-reviewed journals today are OA. That’s a large fraction compared to ten years ago, and the growth rate is accelerating. But it’s still a small fraction of the full range of journals. Hence, a policy requiring researchers to submit new work to OA journals would limit their freedom to submit work to the journals of their choice. That’s why there are no “gold” OA mandates anywhere today.
However, there are a rapidly growing number of “green” OA mandates, both at universities and funding agencies. Some universities request or encourage faculty to deposit their new research articles in the institutional repository. Others, like Southampton and Salford, require deposit in the repository. If the publisher does not give permission for immediate OA, then the deposit remains “dark” until an embargo period expires and the piece may be switched to OA. At other universities, like Harvard and MIT, faculty not only agree to deposit their new peer-reviewed manuscripts in the repository, but also grant the university non-exclusive rights to make the deposits OA. Because faculty grant these rights for future publications, their action takes priority over their future publishing contracts. Because faculty may opt out, case by case, they remain free to decide for or against OA for any given publication. Because the policy shifts the default to OA, it changes behavior on a large scale and the opt-out rate remains in the low single digits.
More than 150 green OA mandates at universities are now listed at the Registry of Open Access Repositories Mandatory Archiving Policies (ROARMAP).
If you’re familiar with faculty passion to preserve their autonomy and prerogatives, you may suspect that most university OA policies were adopted by administrators. But the reverse is true. Most were conceived, drafted, pushed, and adopted by faculty themselves. In fact, the Open Access Directory counts more than 35 cases of unanimous faculty votes for institutional OA policies, including the policies at Harvard and MIT.
A growing number of research funding agencies require green OA for the peer-reviewed articles arising from the research they fund. The Wellcome Trust has done this since 2005, and the Research Councils UK since 2006. The Wellcome model is also followed by the US National Institutes of Health (NIH), the largest funder of non-classified research in the world. On this model, when grantees publish articles based on funded research, they must retain the right to make the peer-reviewed manuscript OA, and they must deposit that manuscript in a designated OA repository. Publishers who believe the costs or risks exceed the benefits need not publish Wellcome-funded or NIH-funded authors, of course. I don’t have data for Wellcome-funded authors. But 100% of surveyed publishers accommodate the NIH policy. Not one refuses to publish NIH-funded authors on account of the agency’s OA policy.
More than 50 green OA mandates at funding agencies are now listed at ROARMAP.
On a single day last month three major announcements transformed OA policy in the UK. First, the Research Councils UK revised their OA policies. They now prefer gold to green OA, and are willing to pay publishers to provide it. (The funds will be awarded to universities in block grants, and universities will use their own criteria to decide which authors and journals are worth supporting and to what degree.) Second, David Willetts, UK Minister for Universities and Science, accepted the OA recommendations of the Working Group on Expanding Access to Published Research Findings (informally called the Finch Group after its convenor, Dame Janet Finch).
These recommendations also prefer gold OA to green. And finally, the Higher Education Funding Council for England announced plans to require OA for all research submitted to the next Research Excellence Framework in 2014.
In one way or another, the UK is about to require OA for all publicly-funded research. Denmark did so the month before. And the day after the UK announcements, the European Commission announced its own sweeping OA policy for EU-funded research and urged member states to adopt similar policies.
Most people who surf the web don’t realize that the internet was developed by researchers in order to share research, and that commerce was only allowed online years later. But the original research purpose has not been forgotten, and committed researchers have steadily moved the free online sharing of research from the periphery to the mainstream of scholarly communication.
Today most universities, libraries, and research funding agencies are providing OA, experimenting with it, or considering it. Its only opponents have been academic publishers. But today most academic publishers are also providing OA, experimenting with it, or considering it. Even the ones continuing to lobby against it are hedging their bets by taking steps toward adaptation. This is what deep institutional change looks like.
About the Author:
Peter Suber is the Director of the Harvard Open Access Project, Faculty Fellow at the Berkman Center for Internet & Society, Senior Researcher at the Scholarly Publishing and Academic Resources Coalition (SPARC), and Research Professor of Philosophy at Earlham College. His latest book is Open Access (MIT Press, 2012).