Rogue Downloader's Arrest Could Mark Crossroads for Open-Access Movement

Michael F. McElroy, Zuma Press, Newscom

Aaron Swartz is accused of downloading over four million articles from JSTOR.
July 31, 2011

This past April in Switzerland, Lawrence Lessig gave an impassioned lecture denouncing publishers' paywalls, which charge fees to read scholarly research, thus blocking most people from access.

It was a familiar theme for Mr. Lessig, a professor at Harvard Law School who is one of the world's most outspoken critics of intellectual-property laws. But in this speech he gave special attention to JSTOR, a not-for-profit journal archive. He cited a tweet from a scholar who called JSTOR "morally offensive" for charging $20 for a six-page 1932 article from the California Historical Society Quarterly.

The JSTOR archive is not usually cast as a leading villain by open-access advocates. But Mr. Lessig surely knew in April something that his Swiss audience did not: Aaron Swartz—a friend and former Harvard colleague of Mr. Lessig's—was under investigation for misappropriating more than 4.8 million scholarly papers and other files from JSTOR.

On July 19, exactly three months after Mr. Lessig's speech, federal prosecutors unsealed an indictment charging that Mr. Swartz had abused computer networks at the Massachusetts Institute of Technology and disrupted JSTOR's servers. If convicted on all counts, Mr. Swartz faces up to 35 years in prison.

The arrest instantly became a new focal point in the long-running debate about how to restructure scholarly publishing. But some people in Cambridge are also asking questions about whether Harvard, which employed Mr. Swartz, and MIT, where his alleged misdeeds took place, should have done more to prevent the episode from spiraling into a major federal prosecution.

Mr. Swartz's case has become a cause célèbre among hundreds of programmers, scholars, and activists who have worked with him during the last decade. (Mr. Swartz came to prominence in 2000, when, at the age of 13, he helped write an early version of RSS, a hugely popular Web-syndication format that feeds content into services like Google Reader.) These activists say the 35-year sentence looming over Mr. Swartz is a prime example of the irrationality and cruelty of the existing copyright regime. If Mr. Swartz abused his guest privileges on MIT's network, they say, the matter should have been settled as a civil dispute, not handed over to federal law enforcement.

"MIT has a duty to get down on its knees and beg that this prosecution be dropped," says Richard M. Stallman, a Boston-based programmer and prominent "free culture" advocate who attended graduate school at MIT in the 1970s.

But others say Mr. Swartz appears to have acted recklessly in this episode, putting his liberty and his promising career at risk while achieving nothing on behalf of open access.

"If the indictment is accurate, this behavior seems to have been illegal, immoral, and ineffective," says Stuart M. Shieber, the director of Harvard's Office of Scholarly Communication. Mr. Shieber believes the current structure of scholarly publishing is unsustainable and does not serve the public well, but he says he cannot imagine how Mr. Swartz's exploit could have helped the situation.

Caught in a Closet

In Cambridge last week, friends and acquaintances of Mr. Swartz said they were horrified by the thought that he might spend three decades in prison. But they said they had no clear idea what he might have been up to at MIT.

"I knew nothing about this until I read about the indictment," says John H. Summers, a historian and former Harvard lecturer who was recently named editor of The Baffler, an independent journal of cultural criticism. Mr. Swartz volunteered to redesign the magazine's Web site and has written an essay on "content farming" for the forthcoming issue. The two men had met several times over the last three months, but Mr. Summers said Mr. Swartz betrayed no sign that he was under the shadow of a federal grand jury. "He has a very cool temperament," Mr. Summers says.

But Mr. Swartz had known he was in trouble since January 6, when he was arrested after being spotted removing a laptop from a wiring closet in the basement of MIT's Building 16, where he had allegedly tapped directly into the core of MIT's network in order to avoid JSTOR's security measures.

That basement, which lies underneath the university's anthropology department, is a chilly, cement service corridor that most MIT students probably never see.

What exactly was Mr. Swartz doing here last fall? He and his lawyer did not reply to requests for interviews, and no one else has come forward with direct knowledge of his project. Likewise, officials at MIT declined multiple requests for comment about how Mr. Swartz's downloading was detected and how the university handled the affair. Most discussions of the case now center on two documents: the federal indictment and a police report that was published last week by Wired.

The purported facts in those documents have not yet been contested in court. But if they are broadly accurate, the story seems to have gone like this: Last fall, Mr. Swartz began an appointment as a research fellow at Harvard's Edmond J. Safra Center for Ethics, which Mr. Lessig directs. (The two first crossed paths in 2002, when, at the age of 15, Mr. Swartz was hired as a programmer for Creative Commons, a project Mr. Lessig started to simplify copyright.)

Mr. Swartz was brought to the Safra Center to study questions of open government, but he spent some of his free time on other matters. Beginning in late September, according to the indictment, Mr. Swartz used automated software to download huge quantities of papers from JSTOR, far exceeding the limits imposed by JSTOR's terms of service. Though he had access to JSTOR at his Harvard office, Mr. Swartz carried on this activity two miles away at MIT, where he had no affiliation. He logged onto MIT's guest network under the name "Gary Host," which collapses to "ghost" as a username.

Mr. Swartz's downloading was detected by JSTOR within 48 hours, and the service began to block an increasing range of IP addresses at MIT. The university then tried to obstruct Mr. Swartz's laptop specifically, by barring the Media Access Control address, or MAC address, that the network had assigned to his computer. But Mr. Swartz managed to evade all those measures. On October 9, when he was simultaneously using two computers to download JSTOR material, the "pace was so fast that it brought down some of JSTOR's computer servers," according to the indictment.

For several days thereafter, JSTOR cut off access to MIT's entire network. Heidi McGregor, a spokesperson for Ithaka, the nonprofit education-technology foundation that merged with JSTOR in 2009, says such events have been extremely rare in JSTOR's history. Citing the continuing criminal case, she declined to clarify exactly how long MIT was suspended or why JSTOR eventually was willing to restore the university's access. She also declined to explain whether the server failures on October 9 affected international access to JSTOR.

Even after the October disruption to the university, Mr. Swartz persisted. At some point in November, the indictment charges, he plugged his laptop directly into the basement wiring closet at MIT. He allegedly used external hard drives to store his JSTOR downloads, periodically visiting the wiring closet to pick up his harvest.

On January 4, MIT police officers were notified that a member of the university's technical-security staff had discovered a laptop in the wiring closet. That morning, a team including a Secret Service agent and police officers from Cambridge and Boston visited the site and installed a Webcam. The camera caught Mr. Swartz visiting the closet that afternoon, concealing his face with a bicycle helmet. Two days later, he was spotted again as he removed his laptop from the closet, and this time he was arrested as he rode his bicycle across campus. (On top of the federal charges that were unsealed on July 19, Mr. Swartz faces a state charge of felony breaking and entering. Not long after the January arrest, he left his fellowship at the Safra Center.)

According to Ms. McGregor, JSTOR believes it has recovered all of the millions of files that Mr. Swartz downloaded. The two parties signed a settlement in which Mr. Swartz assured them that he had not posted the files online or made copies for anyone else.

She adds that JSTOR never contacted any law-enforcement authorities about the matter. The decision to pursue criminal charges, she says, was not JSTOR's.

Motivations for 'Hacktivism'

Why did Mr. Swartz go to such lengths?

He has a longstanding interest in making large-scale data sets available for public analysis. He has pursued that goal through a Web site known as, and in a well-publicized 2008 project, he downloaded an enormous amount of data from the federal courts' PACER system. That episode earned him an FBI file, but he was not arrested.

What was his goal at MIT? Had he been planning to use the huge cache of JSTOR data to analyze corporations' financial support for scholarly research? That seems plausible, as he had helped a friend do a similar study of academic legal research in 2008. Had he planned to create a public database of information about the JSTOR papers, as he and his colleagues at the Open Library project have done with books? Or had he actually planned to feed the full text of the JSTOR library into file-sharing sites? A 2008 "Guerilla Open Access" manifesto written by Mr. Swartz proclaimed, "We need to take information, wherever it is stored, make our copies and share them with the world. ... We need to download scientific journals and upload them to file sharing networks."

Mr. Shieber, of Harvard, says the wholesale file-sharing scenario seems most plausible to him, assuming that the basic facts in the indictment are correct. If Mr. Swartz had wanted to look at corporate financial support, Mr. Shieber says, he probably would have examined a much smaller range of JSTOR data—say, economics papers published since 1975.

Mr. Summers, who has spoken extensively with Mr. Swartz since the arrest, says he believes his intention was to analyze the structure of academic research, not to commit wholesale copyright violations. In any case, Mr. Summers says, it is absurd for the U.S. attorney to treat Mr. Swartz's actions as analogous to stealing cars. "The idea that we should pay no attention to context and motive—that's just a dereliction of intellectual duty," Mr. Summers says. (The U.S. attorney's office declined to comment.)

At MIT, meanwhile, faculty members are debating whether the university was too quick to bring in federal authorities.

"What Aaron Swartz did was a clear violation of the rules and protocols of the library and the community," says Christopher Capozzola, an associate professor of history and acting associate dean of the school of humanities, arts, and social sciences. "But the penalties in this case, and the sources of those penalties, are really remarkable. These penalties really go against MIT's culture of breaking down barriers."

Mr. Stallman, the programmer and MIT alumnus, says he is mystified by the police report's suggestion that Secret Service agents were brought in just hours after the laptop was discovered. "At best—if they didn't know what the laptop was doing—it was an overreaction," he says. "Surely MIT people can examine a laptop without police help."

Over in Cambridge at Harvard, Mr. Lessig declined multiple requests for an interview.

Ms. McGregor, of JSTOR, said her colleagues were dismayed by his April lecture in Switzerland. Mr. Lessig did not give enough credit to JSTOR's efforts to provide low-cost access to universities in developing countries, she said. And the $20 price for the California Historical Society paper, she said, was not JSTOR's choice. The approximately 800 publishers that participate in the JSTOR system decide individually how much they will charge for access for people who are not logging in from institutions with JSTOR subscriptions.

One such person is Mr. Summers. He has taught at Harvard and at Boston College, but he presently has no academic affiliation with either institution. Two of his scholarly papers are available at JSTOR; if you're logging in from outside an affiliated university, it will cost you $38 to buy them.

"What Aaron's case begs us to remember is that universities are supposed to be public, not-for-profit institutions," Mr. Summers says. "They owe a standing moral debt to the public."