A Google employee helps direct people as they register for Google I/O 2014 in San Francisco on June 24, 2014. (Jeff Chiu/AP)
The European Court of Justice recently decided that Google had to implement a limited version of the so-called right to be forgotten, which allows individuals to request that Google searches not link to embarrassing and/or potentially misleading information. A recent blogpost by a BBC editor says that Google has informed the BBC that one of its stories, concerning Stan O’Neal, a former boss of Merrill Lynch, will no longer come up in “certain searches” on Google. This has led to uproar and indignation among technology commentators. Marc Andreesen, creator of Netscape and prominent Silicon Valley venture capitalist, has claimed that the E.U. is mandating the “memory hole” from George Orwell’s “1984.” CNBC has claimed that an “estimated 50,000
These stories are exaggerated. The CNBC claim seems to rest on a basic error. More than 50,000 people have requested that some details be expunged, but there are no good numbers on how many requests have been granted. Google has said that the process of assessing the requests has just begun. It’s likely that the numbers of successful requests will be a small fraction of the number of people asking.
As for the broader legal issues, it isn’t at all clear that E.U. law requires Google to censor this result. Indeed, there’s good reason to believe that Google isn’t required to censor these results. The E.U. official in charge of judicial affairs has explicitly said that the court’s ruling does not apply to journalism.
The ruling does not elevate the right to be forgotten to a “super right” trumping other fundamental rights, such as the freedom of expression or the freedom of the media.
The court ruling says specifically that the rights of the individual looking for privacy do not override the public interest in getting information in situations where:
for particular reasons, such as the role played by the data subject in public life, … the interference with his fundamental rights is justified by the preponderant interest of the general public in having, on account of inclusion in the list of results, access to the information in question.
Put more plainly, people don’t have the right under E.U. law to have their embarrassments forgotten where there’s a public interest in having the embarrassments known.
So why are so many stories like this floating around? One possible factor behind the alarmism is the principal-agent relationship between E.U. authorities and Google. Political scientists and economists talk about principal-agent problems in which one actor (a principal) tries to delegate some responsibility to a different actor. For example, shareholders delegate the running of firms to managers. Congress relies on specialized agencies to administer policies that it has set. The problem is that the principals and the agents may want different things. Perhaps the managers want fancy offices and higher salaries than are in the shareholders’ interests, or the agencies want to subvert Congress’s intentions to protect their client industries. Principals can create various means of control and monitoring to try to get their agents to do their bidding, but these are imperfect, and often expensive.
The European Court of Justice decision has created a principal-agent problem from hell. Google is acting as the agent. It now has to implement the Court’s understanding of European privacy law, and act as a kind of ersatz privacy regulator deciding which requests to accommodate, and which to deny. The problems with this are twofold. First, the court’s ruling is murky, providing little explicit guidance, and second, Google really, really hates its new duties.
Over time, the first problem will get easier, as specialized data officials provide rulings that clarify when the right to be forgotten applies. But Google has strong incentives to push to make sure that the right to be forgotten is interpreted as narrowly as possible, to minimize the burden that falls on it. Hence, Google has some plausible motivations for interpreting the new mandate in ways that are both extremely conservative (so that the ECJ can’t complain) and extremely controversial (to generate political pressure to dial back the right to be forgotten as far as possible).
This might — or might not — explain why the BBC got a notification from Google that it was excluding the web page from certain searches. For sure, the ECJ hasn’t asked Google to send out that kind of notification — that’s Google’s own decision. Furthermore, it’s likely that courts or privacy regulators would hold that Stan O’Neal’s right to privacy is outweighed by the public interest in knowing about his business record. He is just the kind of very prominent person who will likely find it hard to push back against a public interest argument. If, as later updates have suggested, it’s an individual commenter whose right is being upheld, it’s even trickier – it’s hard to see how someone’s blog comment could reasonably lead to the disappearance of an entire post (and if it were possible, it would obviously create some rather problematic incentives for people who wanted to bury controversial stories quickly).
Even so, Google may have incentives to accede to the request without complaint — and to publicize that it is so doing — because it knows that this is likely to send journalists into a frenzy. Even if the ECJ can press Google into service as an unpaid regulator, it can’t force Google to regulate in the exact ways that it would like Google to. And Google, like the Good Soldier Svejk in Jan Hasek’s novel, can perhaps interpret the court’s mandate in ways that formally stick to the rules, but in practice actually undermine it. There are, of course, other possible explanations for Google’s actions — it may be that there are excellent private reasons why Google is acceding to this request. But for sure, the controversy surrounding the request helps Google to push back (as it wants to push back) against strong interpretations of European privacy standards.