Over on TechDirt, they're asking this question: Why, two days before Aaron Swartz' arrest, did the Secret Service take over handling of the case? Their question comes from Marcy Wheeler's excellent reporting on new documents disclosed in his case:
According to a suppression motion in his case, however, two days before Aaron was arrested, the Secret Service took over the investigation.
On the morning of January 4, 2011, at approximately 8:00 am, MIT personnel located the netbook being used for the downloads and decided to leave it in place and institute a packet capture of the network traffic to and from the netbook.This was accomplished using the laptop of Dave Newman, MIT Senior Network Engineer, which was connected to the netbook and intercepted the communications coming to and from it.Later that day, beginning at 11:00 am, the Secret Service assumed control of the investigation. [my emphasis]
In fact, in one of the most recent developments in discovery in Aaron’s case, the government belatedly turned over an email showing Secret Service agent Michael Pickett offering to take possession of the hardware seized from Aaron “anytime after it has been processed for prints or whenever you [Assistant US Attorney Stephen Heymann] feel it is appropriate.” Another newly disclosed document shows the Pickett accompanied the local cops as they moved the hardware they had seized from Aaron around.
Her point is well-taken: According to the Secret Service website, they get involved in computer crime investigations in limited situations, and downloading academic papers from JSTOR fits none of the criteria.
But if you go and look at the testimony of the Secret Service before Congress, you might find the answer there. In April 2011, Special Agent Pablo Martinez testified before the Senate Judiciary Subcommittee on Crime and Terrorism. The purpose of his testimony was to inform the subcommittee about measures to coordinate and investigate computer crimes which could result in economic loss. Buried in that testimony, there is this:
The Secret Service developed a multifaceted approach to combating cyber crime by: expanding our Electronic Crimes Special Agent Program; expanding our network of Electronic Crimes Task Forces; creating a Cyber Intelligence Section; expanding our presence overseas; forming partnerships with academic institutions focusing on cybersecurity; and working with DHS to establish the National Computer Forensic Institute to train our state and local law enforcement partners in the area of cyber crime. These initiatives led to the opening of 957 criminal cases and the arrest of 1,217 suspects in fiscal year 2010 for cyber crime related violations with a fraud loss of $507.7 million. The arrest of these individuals prevented an additional loss estimated at $7 billion dollars and involved the examination of 867 terabytes of data, which is roughly the equivalent of 867,000 copies of the Encyclopedia Britannica. As a result of these efforts, the Secret Service is recognized worldwide for our investigative and innovative approaches to detecting, investigating and preventing cyber crimes.
Could one of those partners have been MIT? Did the Secret Service and FBI fear that their partnership work in the area of combating computer crime and online fraud would be revealed?
On the question of partnerships, let's have a look at Janet Napolitano's speech at MIT and what she had to say about the government relationship with the university:
MIT, of course, is the source of one of the great national-security collaborations, in the form of the Lincoln Laboratory, which currently works with DHS on several projects. I have seen a good example of an impressive DHS-MIT project, which is called the Imaging System for Immersive Surveillance – or "ISIS." The prototype camera, which is at Boston's Logan Airport, greatly expands our ability to spot suspicious objects and quickly analyze any breaches in security.
Marcy asks whether MIT asked the Secret Service to get involved*. From these two pieces of information, it's unclear whether or not they actually asked, or it was simply routine in cases where there was not just one (perceived) breach on the network, but a persistent effort to breach and download information. Accept for the moment that federal law enforcement agencies make wingnuts' paranoia look mild. In my limited experience, they're paranoid and make presumptions of major guilt over minor things, leaving the accused with the unpleasant task of proving their innocence.
Since MIT is a collaborator on national security via their Lincoln laboratory, I think it's entirely possible that at the point where the feds were brought into the case, they were fearful that JSTOR was simply a dry run for their national security scholarship; hence, the Secret Service involvement.
There are other explanations as well, but they all seem to lead back to some kind of belief that Aaron Swartz was doing more than simply downloading journal articles from JSTOR. In short, I believe they feared he was going to steal and publish "national security secrets." Take this little piece Special Agent Martinez' subcommittee testimony with regard to the Secret Service CERT liason program:
The primary goals of the program are: to broaden the Secret Service's knowledge of software engineering and networked systems security; to expand and strengthen partnerships and relationships with the technical and academic communities; to provide an opportunity to work closely with CERT-SEI and Carnegie Mellon University; and to present the results of this partnership at the quarterly meetings of our ECTFs.
In August 2004, the Secret Service partnered with CERT-SEI to publish the first ever "Insider Threat Study" examining the illicit cyber activity in the banking and finance sector. Due to the overwhelming response to this initial study, the Secret Service and CERT-SEI, in partnership with DHS S&T, are working to update the study. An updated study, expected to be released in late 2011, will analyze actual incidents of insider crimes from inception to prosecution. The research team will share its findings with federal, state, and local law enforcement, private industry, academia and other government agencies.
It's always the damned banks, isn't it? But the point here is that there's clearly academic-government research partnerships, forensic partnerships, and policy partnerships. They want to guard those, and with some good reason.
The problem I see is that Aaron Swartz didn't touch any data from those partnerships, yet from investigator to prosecutor, they treated him as though he were the modern-day Rosenberg spy team rolled up in one really, really smart young guy. Instead of reading what he wrote in the Guerrilla Open Access Manifesto, where he clearly said that he believed information should be free, and he intended to free it, they leapt to a conclusion and then reached for every fragment they could find to support it. As Marcy points out, the government invoked that manifesto for the very first time in their response to Swartz' Motion to Suppress:
In other words, precisely at the moment the government defended all the searches it did of Swartz, it (for the first time, I believe) introduced a new descriptor (in addition to the adjectives “writer, hacker, and activist”): Swartz wrote the Guerilla Open Access Manifesto.
The reference is particularly odd, being introduced (though not elaborated on) in this brief defending the investigative approach used by MIT and then the government. It effectively invokes First Amendment protected speech to justify investigative tactics.
Of course, in the end, no Monday-morning quarterbacking will restore Aaron Swartz. But if it leads us to a place where we might have a real conversation about prosecutorial and investigatory overreach, perhaps it will at least begin to facilitate a long-overdue conversation about why people mistrust government with reason, what safeguards government should implement to protect the rights of citizens, and why we must reject celebrating prosecutors who want to "make examples" of others.
Now that the government has dropped the charges against him posthumously, I am hoping his attorneys will release all of the pleadings and documents related to his case. It's time to have an honest conversation about this notion of "homeland security", who exactly is being protected (besides banks, of course), and how it can come to pass that the full force of the FBI and Secret Service came down on the head of a smart, passionate person who simply wanted to publish academic papers for everyone to access.**
*Update: As I was going through the new documents released in the case more carefully, I came across a report of new discovery released by Deputy US Attorney Heymann. It indicates that MIT police were involved in the investigation from day one and remained involved. That tracks with reports that MIT, unlike JSTOR, was not willing to drop the charges against Swartz and actually assisted in the investigation.
What is most heartbreaking, however, is that the US Attorney's office, even after they had all of the facts and knew there had been no breach of national security, would not agree to any sentence that did not include prison time. They had also been warned that Swartz was a suicide risk, and showed no heart at all.
Andy Good, Swartz’s initial lawyer, is alternately sad and furious.
“The thing that galls me is that I told Heymann the kid was a suicide risk,” Good told me. “His reaction was a standard reaction in that office, not unique to Steve. He said, ‘Fine, we’ll lock him up.’ I’m not saying they made Aaron kill himself. Aaron might have done this anyway. I’m saying they were aware of the risk, and they were heedless.”
Good says the hard-line attitude was not unique to his case, but the facts were. People who steal things usually benefit financially from it. Swartz downloaded the material because he believed that such information — in this case, much of it research underwritten by taxpayers — should be free to the public. He did it not to make a buck, but to make a point.
Swartz and his lawyers were not looking for a free pass. They had offered to accept a deferred prosecution or probation, so that if Swartz pulled a stunt like that again, he would end up in prison.
Marty Weinberg, who took the case over from Good, said he nearly negotiated a plea bargain in which Swartz would not serve any time. He said JSTOR signed off on it, but MIT would not.
“There were subsets of the MIT community who were profoundly in support of Aaron,” Weinberg said. That support did not override institutional interests.
There it is, in black and white. An "almost done deal", except it was MIT who walked away from it. They should be ashamed.
**Update 2: To clarify, downloading JSTOR academic papers and publishing them was an attempt on Swartz' part to provide public access to work the public has paid for via their tax dollars and support for academic endeavors. If one doesn't have access to JSTOR, it's costly to get it and access that information. To call this some kind of a "hack" is a complete misnomer. He downloaded publicly funded research with the goal of letting the public have access to it. That's all he did. No hacking and no security breaches. To say the government and MIT overreacted would be an understatement.