Last month, Viviane Reding, the European Commissioner for Justice, Fundamental Rights, and Citizenship, proposed a new framework for data protection in the EU. One of the most controversial elements of this proposal is the “right to be forgotten.” The basic idea is that individuals should have control over their data on the Internet, including the ability to correct inaccurate information and the ability to remove information that they no longer wish to have publicly available. While this sounds innocuous, the right to be forgotten entails as its corollary the duty to forget or to erase, which suggests censorship to advocates of free speech. Peter Fleischer, Google’s Global Privacy Counsel, has described the effect of the EU proposal as making him “the most powerful censor on the planet.”
The right to be forgotten creates a clear tension between two values that many of us concerned with civil liberties share: the protection of free speech and the protection of privacy rights on the Internet. On one level, this conflict points to a cultural and legal difference between the United States and the European Union in terms of the relative valuation of free speech protections and privacy protections. The challenge on this level is how to reconcile legitimate differences of valuation in a medium that easily spills over national borders. We have seen several instances in which Internet companies have been forced to accommodate widely varying degrees of free speech protection, and perhaps the right to be forgotten can be forced into this framework. The long-term effect may be to create stronger national boundaries on the Internet–a move away from the idealism of its early days, but perhaps something that we should have expected as the Internet matures into a space that is central to commercial, political, and social life.
The more interesting question, however, concerns the nature of forgetting as more information is put on digital media and made accessible from anywhere with an Internet connection. The origins of the right to be forgotten are in European laws protecting those convicted of crimes, including a German law from 1973 that limits the publication of the names of criminals who have completed their sentences. Such laws were intended to signal that serving a sentence erased a criminal’s debt to society, but have raised new problems with the availability of information on the Internet.
Fundamentally, the accessibility and permanence of personal information on the Internet creates an unprecedented shift in an individual’s control over self-presentation. Information deemed unflattering can stick around, as can more favorable information. The right to be forgotten would give the individual control over self-presentation by creating the power to order that personal information be erased. This goes beyond well beyond the passive act of “forgetting.” To the extent that the right to be forgotten encompasses the right to prevent others from speaking truthfully about oneself, it creates a clear conflict with basic free speech protections.
But we should not be so quick to dismiss privacy protections on free speech grounds. A more limited version of this proposed right may protect the ability to control information about oneself without infringing on the speech of others. Solutions may be more architectural than legal—for example, making unwanted information available, but less accessible, requiring some expenditure of effort. We ought to be able to build an information regime that protects the interests that we value without prematurely sacrificing one value for another out of expediency and failure of imagination.
As it currently stands, the conflict between the right to be forgotten and free speech rights may devolve into a conflict between two sets of actors. Those who demand that their data privacy be protected may skew toward those on the margins: those with criminal records who find that the ease of a Google search prevents them from starting over, or teenagers who find that the consequences of “youthful indiscretion” may attach for longer than they expected. Those in favor of free speech may skew toward the new media companies. Given these sets of actors, the outcome of any conflict between these competing rights seems foreordained. If we take seriously the challenge of the EU’s proposed right to be forgotten, we can make debates about speech and privacy trade-offs productive.
I don’t think the right to be forgotten has to extend as far as to cause the kind of conflict with free speech that you identify. A right to be forgotten could only include your right to information that you as an internet user originally created or made available (your Facebook profile, email contacts, photos, etc) rather than to information created by others based on publicly available information about a user. For example, Anthony Weiner should have the right to delete his Twitter account and have all of his previous tweets, photos, and anything else associated with the account permanently deleted. That right doesn’t extend to him being able to demand that any one who wrote about his ill-advised photo-tweet should have to delete references to it. In an era of click-through contracts, it will require government regulation to prevent users from signing over rights to their own information as part of the terms of use of social networks and other web services.
Such a right could be coupled with the ability of users to recover monetary damages for reputational harm for information released that was intended to be private or after they have indicated that the information should be “forgotten.” A recent Slate article argued for a mandatory online reputation insurance regime: http://www.slate.com/articles/technology/future_tense/2012/02/_right_to_be_forgotten_how_facebook_google_and_other_companies_can_protect_internet_user_privacy_.2.html While I wouldn’t go as far as ex ante insurance, a legal regime that recognize the right to be forgotten would be necessary for those harmed by the release of stored data to recover for that harm.
It’s true that the right to be forgotten doesn’t need to be so broad as to create conflicts with free speech. But the language of the proposal is ambiguous on this point. It defines the data to be covered by the right as “any information relating to a data subject” (Article 4(2)), but also includes a provision allowing data to be preserved “for exercising the right of freedom of expression” (Article 17(3)(a)), but only to the extent that the member states of the EU choose (Article 80).
Peter Fleischer, of Google, described three scenarios in which this right might be exercised. First, to information that the individual posts. Easy enough. Second, to information someone else re-posts. A little fuzzier, but it would seem to be covered under the right to be forgotten. Third, to truthful information expressed by a third party. The broad reading of “personal data” under Article 4 would seem to cover even this, but the exception in Article 17 would allow individual states to preserve such information for journalistic or artistic purposes. The language of the right to be forgotten presumes the right of erasure in even this third scenario, though it allows that member states can choose otherwise.
As for reputational harm, accidental releases of information, etc., I agree that some basic right to be forgotten could be an important foundation for creating a legal framework to govern interaction on the internet. But there are still a few kinks to work out — such as figuring out where in Fleischer’s framework we want to draw the line.
Apparently this topic is popping up in American courts regarding whether it is libel or defamation to publish information about an arrest or conviction record that has been legally erased. Discussion at the Volokh Conspiracy: http://www.volokh.com/2012/08/08/is-it-libel-to-say-someone-was-arrested-when-the-arrest-record-has-been-erased/
That’s pretty interesting. It makes me wonder what it means to expunge a record if your search engine of choice can produce the record without any heavy lifting on the part of the searcher. I’m uncomfortable with the Orwellian idea of requiring a publisher to remove references to an expunged record from existing articles, but if we want to allow for the possibility that arrest records can be “erased,” it seems to me that there ought to be some barrier between the casual internet searcher and the “erased” record.
I’m basing this on the feeling that if someone went to the library and looked up old papers, those references to an expunged arrest should still be there and should not have been cut out. So the record remains available to the determined searcher. But the ease with which we can find documents on the internet creates a problem when the purpose of a law is to remove documents from public perusal.
I think that striking the right balance between availability of information and privacy requires putting some requirement of minimal exertion of effort between the searcher and the hidden document to be sought. How to implement that, however, is another question.
I think it’s a different between a record being legally erased and actually erased. The expungement of a record is only relevant within a legal proceeding, not in public discourse more broadly. The person can deny the arrest or conviction if they are asked about it under oath; if they are convicted again, it can’t be used against them in sentencing; etc. There is no difference between being able to find a newspaper reference to an arrest or conviction at the library or on the internet. Prior to the internet, being able to go to the library and look it up might have been considered too easy, because the person didn’t have to go to the newspaper and access the actual archives. The distinctions need to be permanent (like the difference between news and police records) not technology dependent (like the difference between the library and the internet).
Here’s where I get to demonstrate my total ignorance of libel and defamation laws.
While the distinction you make is important, I’m concerned about how to implement protections for records that a) may have no legal effect because they have been “legally erased” but b) have spillover effects because they are easily accessible and purport to have some legal effect because they were written before they were legally erased. I assume that judges or other actors within the legal system understand that the expunged record would have no force, but what about other actors? It’s their eyes that I’m concerned about here.
I don’t mean to suggest that there is some radical difference between finding a record in the library versus finding it online, but rather that in combing through an archive the searcher almost necessarily acknowledges the historical/temporal position of the documents, while I strongly suspect that that awareness is less present in a casual google search. It’s a matter of how the way in which we interact, physically, with the archive in question then informs our understanding of the content. And to say that while there remains an important distinction between the moment of publication and the moment of access in the archive, the relationship between those two moments strikes me as problematic on the internet.
I’m unclear which actors you are concerned about. Employers? Journalists? Casual acquaintances?
I would think that anyone Googling a person and accessing an old newspaper article is going to pay attention to the date of that article. Regardless of whether the record has been legally expunged, the event still happened, and therefore the news report of it is still valid. If I get asked about it in a job interview and I say it never happened, I’m lying, even though I could say the same thing in court and not be perjuring myself.
I just think that to say that someone is libeling or defaming you because they make a true statement about your past, regardless of what legal records say, is a dangerous precedent to set. The NJ court quoted in the Volokh piece is dead on when they express their concern over creating an Orwellian scheme in which a change in records is the same thing as a change in historical fact.
I guess to address your question slightly differently than the way you are asking it: surely the purposes for which we expunge records includes not just the interest of erasing the legal effect created by the existence of a record, but also to allow the individual attached to that record to just move on (particularly in the case of minors, say). It is this latter concern that motivates the “Right to Be Forgotten.” And it gets creepy, and runs smack into basic free speech principles, when this turns into a system for the mandatory erasure of said records. So I agree totally that, as you say above, “to say that someone is libeling or defaming you because they make a true statement about your past, regardless of what legal records say, is a dangerous precedent to set.” So perhaps libel is not the right framework here.
But I still don’t think that ends the matter. I continue to think that there is space between the problematic solution you describe of deleting historical fact and the (also problematic, I think) position of encouraging total recall. The solution (or first step toward one) that I outlined above is an attempt to recreate a space for information to be recallable with sufficient concentration. I’m analogizing to memory because I don’t know how else to think about it. How to structure a legal framework that accounts for this is also something I don’t know. But if we accept that the records of a minor may be expunged, partly so that they just don’t follow him or her around unnecessarily, the problem of having access to factual statements relating to those records is not one of truth or falsity but one of degrees of presence. And tools such as libel law may be too tailored to the former issue to be of much help with the latter. (And maybe we also have disagreement about whether the latter is a problem at all.)
Anyway, some links:
http://www.bbc.com/future/story/20120315-mastering-the-art-of-forgetting
http://www.nytimes.com/2010/07/25/magazine/25privacy-t2.html?pagewanted=all
http://www.theatlantic.com/technology/archive/2012/05/forget-about-it-making-the-internet-more-like-our-brains/256832/
And a gratuitous Nietzsche quote courtesy of the first link: “Without forgetting it is quite impossible to live at all”