William Stuntz’s Misguided Theory of Privacy and Transparency
posted by Daniel Solove
William Stuntz (law, Harvard) has long been advancing thoughtful provocative ideas about criminal procedure. I’ve always found Stuntz to be insightful even when I disagree (and I have disagreed with him a lot). Stuntz’s recent essay in The New Republic entitled Against Privacy and Transparency has me not just disagreeing, but doing so rather sharply.
Stuntz begins with an interesting historical generalization. He argues that privacy and transparency (open government) “seem like quintessentially liberal ideas,” although historically they had long been conservative ideas. Stuntz notes that the call for greater government transparency “flowed from pro-business conservatism” because it made it hard for an activist government to alter the status quo. He argues that privacy helped make it hard to regulate big business during the progressive movement in the early 20th century. Stuntz observes: “Privacy, once the right’s favorite right, became the left’s friend thanks to the civil rights movement. In a time when J. Edgar Hoover was spying on Martin Luther King Jr. and Southern sheriffs were enforcing America’s own version of apartheid, police snooping had a decidedly right-wing cast.” As for transprency, “Vietnam and Watergate made the left suspicious of government power generally and executive power in particular. When liberals looked for a way to make Richard Nixon’s imperial presidency a little less imperial, they stumbled on weaponry that Taft’s Republicans had used against Harry Truman: force the president to disclose as much as possible.”
The historical picture is far more complicated than the one Stuntz paints. Justice Louis Brandeis, one of the leading liberals in the early 20th century, was one of the main proponents of privacy and transparency, and he was strongly in favor of New Deal politics. Indeed, it was Brandeis who wrote the famous article, The Right to Privacy in the Harvard Law Review that gave birth to the privacy torts; it was Brandeis who penned the powerful dissent in Olmstead v. United States, 277 U.S. 438 (1928) where the Court held that the Fourth Amendment didn’t cover wiretapping; and it was Brandeis who wrote the famous line in favor of transparency, “Sunlight is said to be the best of disinfectants.” Stuntz is right when he acknowledges that privacy and transparency have strong roots in conservative thinking. But they also have strong roots in liberal thinking, and they are not concepts that have been passed like a baton from the conservatives to the liberals.
But this is not the part of Stuntz’s essay that makes my blood boil. It is his main thesis, where he argues:
Today, the danger that American democracy faces is not that rulers will know too much about those they rule, nor that too many decisions will be made without public scrutiny. Another danger looms larger: that effective, active government–government that innovates, that protects people who need protecting, that acts aggressively when action is needed–is dying. Privacy and transparency are the diseases. We need to find a vaccine, and soon.
Huh? The problem with our government stems from privacy and transparency? To justify this startling conclusion, Stuntz argues that:
[D]ifferent forms of evidence-gathering are substitutes for one another. Anything that raises the cost of one lowers the cost of all others. The harder it is to tap our phones, the more government officials will seek out alternative means of getting information: greater use of informants and spies, or perhaps more Jose Padilla-style military detentions with long-term interrogation about which no court ever hears, or possibly some CIA “black ops,” with suspected terrorists grabbed from their homes and handed over to the intelligence services of countries with fewer qualms about abusive questioning. In an age of terrorism, privacy rules are not simply unaffordable. They are perverse.
Stuntz’s logic seems to be that we should let the government invade our privacy to a significant degree, because if we don’t, the government will resort to even worse things. The argument that if you stop somebody from doing something bad, they’ll do something even worse can be used in almost any situation to defeat almost any law or regulation. Using this logic, one might argue that we should let thieves steal, because if we don’t, then they’ll resort to even worse crimes. The argument proves way too much, and as a result, winds up proving nothing in the end. Moreover, the kinds of information gathering techniques Stuntz lists as examples of “alternatives” rest on very uneasy legal and constitutional ground. Perhaps one of the reasons they have occurred is because of a lack of adequate transparency and a lack of sufficient checking of the Executive Branch. But Stuntz, however, sees transparency as part of the problem.
Stuntz has many more arguments which are worth responding to.
He argues that transparency makes it harder for government officials to do something, and doing something is better than doing nothing: “For most officials most of the time, the key choice is not between doing right and doing wrong, but between doing something and doing nothing. Doing nothing is usually easier–less likely to generate bad headlines or critical blog posts.”
Is this really true? Government officials often always try to do something — the problem is that the “something” they try to do isn’t the result of an informed and thoughtful policy analysis but often a cheap gimmicky solution that will grab headlines. Stuntz frequently complains (correctly, in my opinion) of the excessive increase in criminal laws, and in this essay, he writes that “American criminal codes have metastasized.” Why has this happened? It’s not the result of the government doing nothing. It is because of the strong sentiment that the government must do something, and it’s easy to enact more criminal laws. So the choice for officials isn’t between doing something or nothing — it’s between doing something symbolic versus doing something meaningful but more nuanced and complicated. When it comes to security, the symbolic measures often have high civil liberty costs with very little security payoff. Left unexplored are the many more meaningful alternatives where the benefits might outweigh the costs. Perhaps transparency and privacy are just what the doctor ordered — they make it harder for government officials to resort to quick easy measures and force them to think harder and consider measures that will have a security payoff that outweighs the costs. I have blogged a lot about these issues in connection with the NYC subway searches:
Stuntz goes on to argue that we shouldn’t restrict government information-gathering but should focus on restricting the use or disclosure of information after the government gathers it. Stuntz begins by observing:
Every year, tens of millions of Americans fill out their tax returns, giving the IRS a tremendous amount of information about their finances. The affront to privacy mostly goes unnoticed. Why? Because anonymity matters more than privacy.
I don’t think that the issue here is anonymity. It’s confidentiality. There are strict rules and strong expectations that tax information will be kept confidential. Also, filing tax information is mandatory. Even if a person cares deeply about her privacy, if she doesn’t turn over tax information to the IRS, she’ll wind up in jail. So I don’t know what Stuntz’s proposition about filing tax returns really proves other than the fact that people don’t want to go to prison.
Stuntz then states that two “key propositions follow” from this insight into our filing our taxes as required by law:
First, the more people whose lives the government invades, the better. When targets are few, anonymity disappears. If there were 100 tax forms filed instead of 100 million, the IRS might do more snooping than is healthy. The more phones are tapped, the less freedom is threatened.
So we should want the government to invade our privacy more in order to protect it? Stuntz’s argument may sound like a profound paradox, but it is really little more than doublespeak. This statement might have been true in the days long before computer technology and before Herman Hollerith invented the punch card machine. But in today’s day and age, we have sophisticated computer technology that can analyze zillions of bits of data in nanoseconds. As Bob O’Harrow notes in his book No Place to Hide (2005), the database company ChoicePoint has “more than 250 terabytes of data regarding the lives of about 220 million adults.” (p. 145). Modern data mining techniques make it much less likely that a piece of data will exist as a needle in a haystack. So I find it almost silly when Stuntz says that more wiretapping and more invasive information gathering equals less of a threat to freedom.
Stuntz goes on to argue:
Second, the initial invasion of privacy isn’t the problem; subsequent disclosure is. The true image of privacy intrusion is not some NSA bureaucrat listening in on phone calls, but rather Kenneth Starr’s leaky grand jury investigation, which splashed a young woman’s social life across America’s newspapers and TV screens. That is the nightmare worth protecting against. The best way to stop the nightmare from happening is to limit not what information officials can gather, but what they can do with the information they find.
Why isn’t the initial invasion of privacy the problem? Stuntz mentions the leaks from Kenneth Starr’s investigation as the big problem, but what about calling Monica’s mother to testify, which outraged many? What about the subpoenas to Kramerbooks for Monica’s book purchases? There were many dimensions of Starr’s investigation that struck people as problematic — they brought attention to the extremely powerful tools currently in a prosecutor’s arsenal. Moreover, there are many forms of problematic government information gathering. Would J. Edgar Hoover’s dossiers have posed no problem so long as he couldn’t leak them? Why not put a surveillance camera in all of our homes so long as the government is limited in its use and disclosure of that data? It is certainly true that the Fourth Amendment focuses heavily on government information gathering and little to none on how the government uses the information once it has been collected. But why not regulate both collection and use?
Stuntz argues that a better approach to transparency is to “require limited disclosure: say, to the key congressional committees and to any courts designed to supervise the relevant process.” This is what the law does with regard to foreign intelligence gathering — the Foreign Intelligence Surveillance Act (FISA) requires a court order from a secret court to engage in surveillance. The Bush Administration, however, has violated the FISA and has authorized warrantless wiretaps by the NSA. And it has done so through a lack of transparency. It is only now that the information about the NSA surveillance has been released that the public has the opportunity to analyze the merits of the Bush Administration’s actions. As for congressional oversight, recent events make me skeptical of Congress serving as a meaningful check on the Executive Branch. Perhaps when Congress is controlled by a different party than the Executive it can function as a better check, but one of the lessons we’ve been learning over the past decade or so is that Congress doesn’t care to do much checking when the White House is occupied by the same party.
Stuntz concludes his essay by observing: “We have too much privacy, and those who govern us have too little.” Unfortunately, Stuntz has it exactly backwards.