Column for Internet World
By Mike Godwin
March 1995
For David LaMacchia, Christmas came four days late. It was December 29th when federal Judge Richard Stearns decided that the MIT undergraduate, who'd been charged with conspiring to trade millions of dollars of illegal copies of software over the Internet, had in fact committed no crime at all.
It had been nearly a year since LaMacchia, the kind of student whose dorm room is so cluttered that he can barely make it from bed to computer terminal, had been notified by officials at MIT that he was under federal investigation. "I guess I got really scared when they told me I was responsible for 30 million dollars' worth of software violations," LaMacchia said early this year. (Later estimates put the figure at closer to $1 million.)
But the question of whether LaMacchia had in fact committed any crime at all turned out to be a difficult one for authorities to answer. Says LaMacchia's lawyer, Harvey Silverglate: "The government is attempting to assert control over this burgeoning thing called the Internet." As a result, he says, it "spasmodically overreacts in order to set an example, to deter behavior the government doesn't like."
LaMacchia's case had been touted as the largest single case of software piracy ever reported. According to the goverment, the student had set up a bulletin-board system on some MIT workstations that could be used to trade unauthorized copies of commercial software. The problem for the feds was that the copyright laws didn't criminalize LaMacchia's conduct. Section 506 of Title 17 (the Copyright Act) does offer a criminal provision, of course:
"Criminal offenses
"(a) Criminal infringement.--Any person who infringes a copyright willfully and for purposes of commercial advantage or private financial gain shall be punished as provided in section 2319 of title 18."
The language is straightforward enough, but it was difficult with the available facts for the government to show that David LaMacchia had ever received, as a result of his BBS operation at MIT, any "commercial advantage or private financial gain." So, as a fallback measure, they chose to indict him on a charge of conspiracy to commit wire-fraud.
I first learned about the case early in 1994, when Brian LaMacchia (David's brother, who's also an MIT student) called me at the Electronic Frontier Foundation, looking for a referral. (I sent him to Harvey Silverglate.) Although no charges had then been filed, it was clear to me after talking to Brian that the government would have some trouble making a case against his brother -- I had written about the difficulties the government has in computer-crime contexts when it tries to go beyond the Copyright Act to prosecute an alleged infringer with a more general criminal statute, such as the interstate-transportation-of-property statute or the wire-fraud statute. In a column in this magazine ("When Copying Isn't Theft: The Government Stumbles in a Hacker Case, " Nov./Dec. 1993), I had pointed out that the Supreme Court had held, in a 1985 case called Dowling v. United States, that copyright infringements must be prosecuted, if at all, under the criminal provisions of the Copyright Act. Specifically, the Court had held that Congress's well-deliberated penalty provisions under the copyright laws precluded any attempt by prosecutors to "fill the gaps" with more general-purpose criminal laws.
And yet it was precisely this sort of gap-filling that the Boston U.S. Attorney's office seemed to be attempting in the LaMacchia case. What on earth were they thinking?
What's worse, the difficulties they were going to have with the Dowling decision would be compounded by the fact that, even apart from Dowling, the government couldn't easily show that wire-fraud had taken place. Let's look at the elements of wire-fraud as discussed in one federal case:
"The elements of wire fraud are (1) formation of a 'scheme to defraud,' and (2) use of interstate wire communication to further that scheme. . . . Congress did not define 'scheme or artifice to defraud' when it first coined that phrase, nor has it since. . . . Instead that expression has taken on its present meaning from 111 years of case law [since the phrase was first used in the mail-fraud statute of 1872]....It's clear from the discussion of wire fraud in the Lemire case that a defendant, to be convicted of wire fraud, must be shown beyond a reasonable doubt to have made a misrepresentation, or to have failed to disclose material facts that he had a duty to disclose. Yet the LaMacchia indictment did not allege either a misrepresentation or a nondisclosure, and the facts as related in the indictment didn't even imply one. To criminal lawyers experienced at dealing with mail-fraud and wire-fraud cases, it didn't seem merely that the government had written its indictment badly (that's a problem the government can correct with a superseding indictment). Instead, it looked as if the government had no case at all -- no criminal-copyright case, and no scheme to defraud. In the absence of a fraudulent scheme, LaMacchia couldn't be conficted of conspiracy to carry out such a scheme."At the core of the judicially defined 'scheme to defraud' is the notion of a trust owed to another and a subsequent breach of that trust. But '[n]ot every breach of a fiduciary duty works a criminal fraud'
In their attempts to delineate which breaches of duty rise to the level of criminal fraud, courts have used various limiting doctrines. Some, including this court, have required that the fraud be 'active'--that the fiduciary utilize his trusted position to obtain a benefit for himself at the expense of the person whose trust he breaches. ... Other courts have required that the breach be accompanied by some material non-disclosure or misrepresentation to the party owed the duty. ... The crux of these requirements is that the wire fraud statute makes criminal only breaches of duty that are accompanied by a misrepresentation or nondisclosure that is intended or is contemplated to deprive the person to whom the duty is owed of some legally significant benefit." -- United States v. Lemire.720 F.2d 1327 (D.C. Cir. 1983), cert denied, 467 U.S. 1226 (1984)
Now, I'm not the greatest lawyer in the world, but it seemed clear to me after reviewing both Dowling and the federal fraud cases that the government shouldn't have brought this case at all -- that LaMacchia had committed no crime, regardless of the claim that his BBS had been used for largescale software copyright infringement. So, I then had tosk, how do we explain the government's decision to prosecute David LaMacchia?
After some reflection, I came up with five theories to the explain the government's actions. Feel free to pick any of them (or any selectoin of them) that you find most credible:
When I first started posting these theories on the WELL, and explaining how I thought that, under the law, LaMacchia had committed no crime, I was surprised at how much criticism I received from people who *couldn't accept* either a) that the government was mistaken to bring the case, or b) that I seemed to care so much about it. After all, they said, even if the laws don't quite reach LaMacchia's (alleged) conduct, shouldn't he be punished anyway? And surely this case raises no important issues for a civil libertarian!
The questions assumed that LaMacchia had deliberately set up the system to facilitate software piracy, but I knew that, in addition to claiming that the indictment was legally flawed, LaMacchia and his attorneys also steadfastly deny the government's factual allegations. David LaMacchia characterizes himself as a sysop who was being blamed for conduct of his users--to LaMacchia, his case illustrates a serious risk to system operators on the Net: to what extent will they be held criminally responsible for the acts of their users? Computer networks, both public and private, have become an important forum for public discourse and activity--make sysops responsible for their users' actions, and you may give them an incentive to quit operating forums altogether.
Similarly, there's a fine line between legitimate deterrence and a chilling effect on Constitutionally protected speech. A prosecution like the one in the LaMacchia case may deter some software pirates, but at what price?
Even some law-enforcement officials (notably Scott Charney of Department of Justice) have expressed concerned that well-meaning efforts to police cyberspace may have an undue chilling effect on First Amendment prerogatives -- will we feel free to speak freely if we think there's a cop constantly inspecting what we say for criminal content?
But for all that there's concern about Constitutional rights, the government also feels a certain pressure to put the lid on activities it sees as antisocial even when the laws don't directly address the conduct in question--hence the LaMacchia case. While some copyright holders were pleased to see the government step in and attempt to convict LaMacchia of copyright-related offenses, civil libertarians were disturbed at what they saw as a tactical use of the government's ability to prosecute. "This prosecution represents another attempt to mold the criminal law according to what the Justice Department wants rather than what Congress has intended," says Silverglate." In his decision to dismiss the LaMacchia case, Judge Stearns cited a Supreme Court opinion to similar effect: "[i]t is the legislature, not the Court which is to define a crime, and ordain its punishment." Stearns, who assumed at least for the purposes of his decision that LaMacchia had committed the actions alleged in the indictment, suggested that Congress, if it wishes, could amend the copyright laws to enable prosecutions of cases like LaMacchia's. But even though he disapproves of LaMacchia's alleged actions, he also noted that expansive interpretations of the criminal law "would serve to criminalize the conduct of not only persons like LaMacchia, but also the myriad of home computer users who succumb to the temptation to copy even a single software program for private use. It is not clear that making criminals of a large number of consumers of computer software is a result that even the software industry would consider desirable."
Stearns' opinion reminded me a bit of the the dialog between Thomas More and his son-in-law in the Robert Bolt play "A Man For All Seasons." The son-in-law says he'd willingly cut down every law in England to get at the Devil. More responds: And when the Devil turns on you, what protections would there be, every law in England being cut down?
A lot of the arguments I saw on the WELL reminded me of More's son-in-law -- some users hoped to find some way to stretch the law to make it fit LaMacchia (whom they assumed to be guilty of *something*). When I pointed out that LaMacchia was denying the factual allegations and had some pretty good legal arguments as well, I got a lot of grief -- "You're just focusing on technicalities!" "You're simply trying to pick apart this indictment on technical grounds!"
What some people call a concern with legal technicalities and legal pedantry is what civil libertarians call a concern with due process. And is it really a "technical grounds" argument to say that Congress didn't make the defendant's behavior a crime?
That line from Thomas More keeps echoing in my head -- if the government stretches the law and ignores precedent to reach every instance of conduct it disapproves of, what protections will be left for us?