Beginning in the fall of 2010, Aaron Swartz repeatedly logged on to MIT’s campus network and used an automated script to download nearly 5 million articles from JSTOR, one of the largest digital archives of scholarly journals in the world. When they discovered Swartz’s actions, both JSTOR and MIT faced a choice—should they help prosecute Swartz as a nefarious hacker, or should they forgive him for skirting the law while pursuing his activism?
Last summer, after he agreed to return the articles he’d downloaded, JSTOR dropped its civil case against Swartz. MIT took a different course. According to Swartz’s friends and supporters, the university played a key role in federal prosecutors’ efforts to prosecute Swartz of a litany of computer fraud charges. On Jan. 11, facing the near certainty of jail time, Swartz committed suicide. Shortly afterward, MIT’s president released a heartfelt public statement. “It pains me to think that MIT played any role in a series of events that have ended in tragedy,” wrote L. Rafael Reif. He appointed Hal Abelson, a respected computer science professor, to lead an investigation into the university’s participation in the Swartz case.
Abelson has said that he will release his report in a few weeks’ time. I’m hoping that the investigation will go beyond merely excoriating the school for its shameful role in Swartz’s prosecution. If MIT truly wants to atone for joining the federal case against Swartz, it should do something much grander: It should pledge to spend its money, prestige, and moral authority to launch a multiuniversity campaign to free every scholarly article from behind pay-wall archives like JSTOR. In other words, MIT should pledge to finish the project Swartz started.
Making academic articles available to everyone is one of the most direct ways for MIT to fulfill its public-spirited mission to expand the world’s access to knowledge. Over the last decade, MIT, like other big-name schools, has put thousands of its lectures and course materials online, allowing anyone, anywhere to access it for free. It has even begun offering certificates to people who take its online courses. But liberating academic articles from pay walls would be a much greater, more lasting contribution. If every scholarly work were free and searchable, teachers, schoolchildren, university students, and brilliant autodidacts everywhere (people like Swartz himself) would be able to use the Internet as a true source of learning.
Freeing academic articles would also help universities reclaim their place in public debates. Much of the work produced in academia is never seen by anyone outside that cloistered world, because everyone who’s not affiliated with a university is cut off from access. The fact that most people don’t have access to legitimate research is probably not the only reason that so many of your fellow citizens hold dubious beliefs about climate change, evolution, vaccines, or other things that make you facepalm—but it doesn’t help. By killing pay walls, MIT could make the truth more relevant.
But if all that’s too dreamy for you, here’s a simpler reason for MIT to stop the pay-wall madness: It will save money. The world’s colleges now collectively spend at least $10 billion and probably more than $20 billion every year on subscriptions to academic journals and archives like JSTOR. Even worse, those costs are rising at an astronomical rate—by one calculation, the amount that a typical college library spends on annual journal subscriptions rose by more than 300 percent between 1986 and 2005, much faster than inflation, tuition, and most university budgets. (Note that this was during a period when many journals went electronic—a time when you’d expect their costs and, thus, their prices to go down, not up.) These prices keep rising because the market for journals is inelastic—since there’s no substitute for any specific journal, whatever price it charges, universities feel like they’ve got to keep paying. This is all explained very well in a paper called “The High Cost of Scholarly Journals (And What To Do About It),” which I’d recommend you read if it weren’t behind a pay wall.
The amount universities spend on journals is especially perverse when you consider that most of the research in those journals was produced by scholars affiliated with and supported by universities, government agencies, and philanthropic endowments, all of whom have an interest in spreading scholarship far and wide. When you stop to think about it, the whole process looks Rube Goldbergian: People who work for universities and are funded by the public are giving their work away to journals for free—and then the journals are charging universities to buy it back. They’re making enormous profits from the scheme, too. For instance Elsevier, one of the leading publishers of scientific journals, routinely reports profit margins of around 37 percent.
MIT could stop the whole business with a few bold steps. First, it should declare that, within three years’ time, its libraries will cease subscribing to all academic journals and archives that do not make their articles available online to everyone. Second, MIT should require all of its faculty, grad students, and other affiliated researchers to submit their work only to open-access journals. Third, MIT should instruct its deans and other officials to no longer look favorably upon the mere fact of publication in a “prestigious” journal when making hiring and tenure decisions. Instead, promotions should be based on the quality of a person’s work, wherever it’s been published. (This sounds obvious, but most people in academia will tell you that where you publish is just as important as what you publish.)
Finally and perhaps most importantly, MIT should encourage other universities to participate in this effort. Specifically, it should establish a fund that pays for the true costs of publishing academic journals. Call it the Aaron Swartz Memorial Open-Access Fund. Instead of paying exorbitant subscription fees to for-profit journals, universities would instead contribute to the fund. (The amount would be a function of a school’s size and research budget.) Journals would draw from the fund according to how often their work is accessed. It’s not unlike the compulsory license system that pays musicians when their work is covered or played on the radio, except instead of allowing for more poppy renditions of Elvis tunes, this fund would let anyone in the world access any academic article at any time.
I didn’t come up with this plan by myself. I’ve filched many of these ideas from open-access advocates, especially Michael Eisen, a biologist at UC Berkeley who’s been a leading crusader on the issue. Eisen lays the blame for the sorry state of scholarly publishing at the feet of university presidents, whom he calls “feckless” on the issue of open access. Over the past five years, many other participants in the academic publishing world have offered plans for reforming the industry. In 2008, the National Institutes of Health, the world’s largest funder of medical research, adopted a “public access policy” that requires all papers funded by the agency to be made freely available a year after publication. Faculties at major universities have also come out in favor of open access, with groups at MIT, Harvard, Stanford, and several UC schools adopting guidelines that encourage faculty to contribute their work to open journals. And yet most journals are still closed.
“The problem is that university administrators have not made the cognitive leap—they haven’t joined the issue, and they’re really to blame for the problem,” Eisen says. MIT’s faculty open-access policy is a case in point. Like many other schools’ guidelines, it includes an “opt-out,” meaning that if a faculty member wants to submit his work to a closed journal, he can get the policy waived. Many journals—including some of the most prestigious—require authors to opt out of their school’s open-access policies. And many faculty members do just that in order to get published. Since 2009, when MIT’s open-access guidelines went into effect, only a third of its faculty’s articles have been made available online. The rest—about 14,000 papers—are stuck behind a pay wall. Eisen argues that this will continue until university presidents take a strong stand in favor of open access, especially by adopting open-access policies without an opt-out.
Skeptics of open-access rules raise a couple of major objections to such policies. First, there is some cost to producing academic journals—and if they’re completely open, their revenues will plummet, and many will go belly up. Things would be especially dire for journals published by nonprofit organizations like academic societies. Every year, the American Historical Association spends about $500,000 to produce the American Historical Review, a much-cited history journal. It sets its subscription fees low enough to cover that cost. Since the 1990s, the association has experimented a few times with lifting its pay wall—and each time it saw a steep decline in subscriptions, Robert Townsend, the AHA’s deputy director, told me.
One of the ways that open-access journals in the sciences support themselves is by asking researchers to pay a fee when submitting their work. (For instance, the Public Library of Science, which publishes several open journals, asks for as much as $2,900 from authors; PLOS waives the fee for authors who can’t pay, however.) Townsend argues that such a plan wouldn’t work well in the humanities, where researchers labor under much tighter budgets than in the sciences. “A typical research grant from the [National Endowment for the Humanities] is about $500 or $1,000, enough to pay for one trip into the field and nothing more,” Townsend says.
Eisen concedes that asking researchers to pay for publication isn’t the best way to move toward open access. That’s just the only funding mechanism possible right now. And it shows why universities like MIT—which are the main source of revenue for journals—have to lead the fight. If schools developed a new, more efficient way to pay for journals—to pay for the cost of peer review and editing but not for marketing and profits—then universities will save money while ensuring access. It would be a win for everyone, except for-profit journals.
After Swartz’s death, Eisen published a blog post titled, “How Academia Betrayed and Continues To Betray Aaron Swartz.” In that essay, he argued that “the real way to honor Aaron Swartz is to combat this pervasive institutional fecklessness and do everything in our power to make sure no papers ever end up behind pay walls again.” I asked him if he’d heard from any university officials in response.
He hasn’t. “I wish I’d seen any evidence that this was a wake-up call for somebody to say, ‘How ridiculous was it that this stuff wasn’t freely available in the first place?’ But I haven’t,” Eisen says. “I’ve seen lots of anger from people who are already on this side, but I haven’t seen a single thing from universities recognizing that in some deep way they were responsible.”
It’s your move, MIT.