Books I have co-authored

“Delete” – a new thought-provoking book about the Net and Information Policy

What will happen to society and the individual if everything is remembered forever and nothing is forgotten? This thought-provoking question is addressed by Viktor Mayer-Schönberger (information policy professor, National University of Singapore) in the book “Delete: The Virtue of Forgetting in the Digital Age” (Princeton Univ. Press 2009). His pessimistic answer is that:

“Perfect, comprehensive digital memory denies human beings the ability to grow, to change, and to evolve over time. That is deeply worrying.”

His argument is similar to that of the privacy advocates’, though he approaches it from a new angle. In a world with perfect retrievable memories people can be haunted and destroyed by something from their past or from their private lives at any time. What is innocent information today can be incriminating evidence in an uncertain future with another government, a changed cultural value system, or if the person moves on to a new phase in his or her life.

In such a world people would self-censor, refrain from expressing their opinions, and be fearful about sharing their thoughts, words and photo albums with their friends.

Mayer-Schönberger illustrates this concept with a story about the student teacher Stacy Snyder. Snyder had passed all her exams and had earned all her credits but was refused her teaching certificate because she had put an innocent costume party photo of herself on her MySpace page with the caption “drunken pirate”. Snyder offered to remove the picture but to no avail.

The 25 year old student teacher who had her future ruined by a MySpace picture

The 25 year old student teacher who had her future ruined by a MySpace picture

An additional example is the Canadian psychotherapist Andrew Feldmar who was permanently barred from entering the US when a customs officer did a Google search and found an article he wrote in an academic journal ten years ago where he mentioned that he had used LSD around 1965. As Schönberger points out, what happened 40 years ago has nothing to do with Feldmar’s life or the person he is today.

There are numerous other examples of people who have had their lives shattered because something from their private sphere or their distant past resurfaced on the internet and was taken out of context.

Another of Mayer-Schönberger’s arguments is that a perfect memory is a curse, not a blessing. Rare medical cases where people are unable to forget (and therefore remember exactly what happened every day of their lives) show that these people become overwhelmed with trivial details and have difficulty making normal everyday decisions.

The book is a first attempt to address this issue. The author’s idea is to formulate a framework for how to shift the default mode in the information society back to slowly forgetting (the natural human state for millennia), instead of remembering. The methods he identifies could focus either on the power aspects or on bringing the time aspects back into the equation. He identifies six methods for managing the problem:

  • Digital Abstinence (stay away from the Net, or at least don’t put everything about yourself on Facebook)
  • Privacy Rights (these are the rights to request that information be removed from corporate and public databases, websites, etc.)
  • Privacy DRM (the idea of tagging pieces of digital information with limitations such as date for deletion, noindex instructions for search engines, restrictions for who can access this information, etc.)
  • Cognitive Adjustment (learn to devalue old information and understand that it is less relevant and credible)
  • Full Contextualization (keep all information, but by including the context of single pieces of information the risks of misinterpretation will be minimized)
  • Information Ecology (laws for record keeping and compulsory deletion of old database records)

When Mayer-Schönberger discusses the methods he points out that privacy rights is a useless tool since individuals almost never actively request that their information be removed or take their claims to court. Information Ecology rules are more efficient as they protect all information. However, the data retention laws that were initiated after 9/11 move in the opposite direction.

The author proposes that the default in the digital world should be changed to slowly forgetting instead of remembering. Perhaps one should receive a prompt for when one wants an item to expire every time a document is saved or something is put on the internet. This would serve as an important reminder about the temporal nature of information.


I must commend Mayer-Schönberger for his insightful and thought-provoking arguments. I support his program for mandated expiry dates of old database records held by large organizations and possibly prompting for expiry dates when you save a document, etc. However, I am not convinced that his program for institutionalized forgetting will solve the problem.

To some extent I think he is barking up the wrong tree. Instead of proposing an elaborate system for forgetting in the digital world, the problem would be significantly reduced in a more relaxed and trusting culture.

The first problem is that we live in a society driven by the logic of the tabloids, rife with hysterical moral panics. Background checks and diligence create a self-reinforcing process, where the mindset of the trash media feeds into the mainstream corporate environment. If you devote significant resources to uncovering incriminating information you will most likely find something minor. To justify the costs of the investigation it is much easier to disregard the context and act on the information rather than be lenient. This pro-panic bias feeds itself. If a major company or government agency rejects a qualified candidate for a senior role because of some 15 year old nude pictures it sets the tone and other organisations will follow their lead.

The second problem is that moral panics feed into the legislative process, which results in disproportionate laws. If lawmakers could deliver a professional, rational and proportional legal system many of the problems addressed by Mayer-Schönberger would disappear on their own. For example, the statue of limitation rules for minor crimes would make it moot if someone used drugs more than 10 years ago. Policies for the “morality” of teachers should be handled by professional managers, not by politicians and political appointees with a need to pump themselves up and show toughness against alleged “immorality”.

The third problem is that information is power. As long as incriminating material can be a bargaining chip that can give you a hold on your enemies there will be incentives to hoard information. Even if the full program for forgetting were implemented, bad faith information hoarders would ignore it. But this problem would dissolve by itself if society was more relaxed. The more bans and restrictions that are not respected, the more hypocrisy there is in a society, and the more room for blackmail.

My personal belief is that the tidal wave of digital information is stronger than any program for forgetting. The problems brought up by Mayer-Schönberger might very well lead to a repressive, paranoid, self-censoring society, where people are afraid of surveillance, backstabbing, and attacks in the same way as in the former Soviet Union. (I find this negative development less likely outside of the US and the UK. In France, for example, infidelity by leading politicians is a non-issue.)

But even in the US/UK, it is possible to envision a positive development, driven by the forces of demography as one generation is replaced by the next. Over time the Facebook generation will grow up and their new life on the Net will initially clash with the value system of the older generation. For example, the teenage practice of sending sexual text messages or pictures via the mobile phone (sexting) is widespread (2025% according to surveys). In the next 10-15 years there will be a number of scandals in which hard working, bright career lawyers are denied partnership because a raunchy nude picture that was taken when they were a teenager resurfaced during a background check. A number of aspiring public figures will have their lives destroyed when the tabloids realize how easy it is to uncover damning material by paying hard cash.

I believe that the public guardians of “morality” will eventually lose this battle in the same way as the prohibitionists lost their battle to keep alcohol illegal. In 2030 most people under 40 will have been to a costume party, played with a camera in the bedroom during their teen years, or have a friend who took pictures of them drinking alcohol in their dorm room even though they were under 21. At some point, the sympathies will shift from the zealots to the “culprits” as the new generation gets fed up with the older generation’s hypocritical moral panics. Recruiters will eventually realize that a ten year old Facebook picture in which the candidate is drunk and surrounded by friends does not disqualify him. Rather, it can be seen as an indication of social skills. In 2030 the newly appointed prosecutor will be forgiven for any ten year old “scandal” that has resurfaced about his dorm room shenanigans. If society is flooded with “scandalous” material about almost everyone, eventually no one will have the energy to care or be upset by it. Ergo, the material will be forgotten. (As a bonus we will have a more relaxed society with discredited tabloids and fewer moral panics.)


More information can be found in his book, in a video from his seminar at Harvard, from CBC and NPR radio interviews with transcripts. Blogs about Mayer-Schönberger’s ideas are Reputationdefender, Harvard Berkman CenterPrinceton University Press, Only Dead FishReuters Great Debate,  and the Daily Kos.

Comments are closed.