Delete: The Virtue of Forgetting in the Digital Age
By Viktor Mayer-Schonberger
Hardcover: 256 pages, $24.95
Princeton University Press: Princeton, NJ
October 2009
Since the beginning of time, for us humans, forgetting has been the norm and remembering the exception. Because of digital technology and global networks, however, this balance has shifted. Today, with the help of widespread technology, forgetting has become the exception, and remembering the default. How and why this happened, what the potential consequences are for us individually and for our society, and what--if anything--we can do about it, is the focus of the book.
If the measure of a successful book is how long you spend thinking about its premises and conclusions long after you've finished it, then Viktor Mayer-Scholberger's Delete a blockbuster. Even after I decided after much thought that I disagreed with his theories, they are theories that haunt, raising questions about personal identity, memory and ultimately our relationships with our own individual histories.
A couple strands of the book are straightforward and non-controversial. Early on, Mayer-Schonberger runs through a quick overview of the rise of curating human knowledge, from oral storytelling and memorization through clunky tablets and unwieldy papyrus to popular books. It's a good, quick sketch of the evolution of documentation, which led, the author argues convincingly, to a "shared memory" and a "community of belonging." So far, so good.
He then begins to make a strong case for enforcing informational privacy, that our digital trail should belong to us and that few of us realize how many of our details are available to corporations and government. For those of us who are aware of how just how long everything is stored, particularly those of us who use social media or comment and write on the web, the author has a specific concern: "... fostered by our knowledge of the vast digital memories that Google and others have about us, we may thus become overly cautious about what we say--in other words, the future has a chilling effect on what we do in the present."
It is along around here that skepticism began to seep in for me. Mentioning this in passing, note it, propose a couple of solutions and move on? Sure, why not. But if this were a novel, passages such as the following would be but a foreshadowing of greater preoccupation--I would argue nearly obsessed preoccupation-- to come: "... once we realize that information can reach anyone, we'll err on the side of caution and if in doubt censor ourselves rather than risk incalculable damage."
First off, in a world where the younger adult generations are nearly all putting the occasional foolish artifact online, surely the sting of misguided photos, raunchy quips and party videos will fade. Already we're seeing that effect in the public arena; witness the barely noted reaction to the Rep. Ike Skelton's admonition last week to telling Todd Akin to stick it up his ass. There was a small, bemused flurry that dissipated almost immediately. Why? Because we're inured by now. Overexposed. It's a yawn. After hearing Mike Duvall boasting of his sexual conquests in Sacramento, everything that comes afterwards is bound to be weak tea indeed. And that's the point: If we recognize that everyone has "gotcha" moments that were ... well, got, surely that should immunize most of us from any but the most egregious documented incidents.
Mayer-Schonberger disagrees. In fact, he spends quite a bit of time fretting over how our lives are overdocumented, and behaviorally constrained as a result. His hypothesis is that it currently is easier to "save" memories than to discard them, and early on he uses the example of digital photography to illustrate this:
Assuming it takes only three seconds for a person to look at an image and decide whether to preserve it or not, and that she values her own time at a current average wage, the "cost" of the time alone that it takes to decide exceeds the cost of storage (including having a second copy on a backup device). With such an abundance of cheap storage, it is simply no longer economical to even decide whether to remember or forget. Forgetting--the three seconds it takes to choose--has become too expensive for people to use.
And so, the author argues, the modern person stands upon a pile of digital material -- photos, text files, emails, IM's--that will doom her or him to being locked in to being the same static person, forever.
He takes information privacy considerably further than I'm willing to, and it's unclear, ultimately, if his drastic measures are meant to protect us from outsiders who would misuse our information ... or from our selves. There's something creepy about the lengths he thinks most of us should want to go to in order to shield our current and future selves from our past selves. Consider this proposal, in which everything we ever do on a computer -- every file, every search -- should come with us deciding when it should expire and disappear. And keep in mind this the same author who argued earlier that the cost-benefit analysis we make when downloading photos from a digital camera is too time expensive at three seconds per decision:
Users when saving a document they have created, would have to select an expiration date in addition to the document's name and location on their hard disk. Users wouldn't be able to save the file without specifying an expiration date, much like how they can't save a file without selecting a name for it. Based on these preferences, the users' computers would do the rest: managing expiration dates and clearing out expired files, perhaps once a day. Of course, users would have a little software utility to change the expiration dates in case they discover information has lost its value earlier than expected, or remains important and useful beyond its originally envisioned lifespan.
Or imagine this scenario, every time a photo is taken:
Suppose somebody takes a digital picture of you. If seen as an information transaction, the expiration date for such a picture ought to be set jointly between the picture taker and yourself. Rather than haggling over it in person, the negotiations over expiration dates could be done electronically. Each digital camera could have a built-in process to select expiration dates (perhaps through an easy preset). Before taking a picture, the camera sends out a "picture request." Imagine further that we carry with us small "permission devices" (the size of key rings) that when receiving such a "picture request" respond with the owner's preferred expiration date. The device could even have a little selector to quickly change preset expiration dates, perhaps from "zero" (if one does not want the picture added to digital memory at all) to "three years from now" (a medium term duration) to "100 years from now" (for memorable events). When the picture is taken, it's stored with the shortest of all expiration dates received (including that of the picture taker himself). The camera could even provide a visual cue as to whether somebody's device said "zero" or similar extreme compared to others in the picture. Such a solution could be easy enough to utilize for most users, and yet make it possible that pictures we do not want stored long-term would become forgotten over time.
This strikes me as a ludicrously paranoid and impractical proposal. There's so much absurdity wrapped up in this idea that it's hard to know where to begin. First off, what happens to crowd shots? Do people just disappear if their device doesn't make an agreement with the photographer? Does it cancel the whole picture? What happens when different people in a shot have different expiration dates? In a country with a yawning digital divide, it's hard to envision all of us carrying around these devices. And what happens if you lose it? Etc.
The strangeness of the photo solution bears an uncanny resemblance to beliefs that were around at the beginning of photography, that you could have your soul stolen by the picture-taker. This undercurrent of superstition runs throughout the book, with pages and pages spent on cookie "agreements" online. But the most disturbing aspect is twofold: reflected throughout is a basic distrust of humans to put their pasts in perspective if concrete evidence is around. Near the beginning of the book he uses the fictional example of a woman named Jane who is contacted by an old friend she hasn't seen in a while, tries to remember the cafe in which they last met, culls through emails to find out, and runs across an email disagreement they'd had that she'd forgotten. So much ill feeling is rediscovered that she considers canceling the reunion.
Forgive me, but Jane is silly. And if people really would cancel a proposed coffee date because they were re-reminded that eight years earlier they had an email tiff, they are not the sort of people to be taken seriously (and certainly not the sort of people the world should be designed around). Why not assume that Jane would be bemused at the forgotten falling out, and would put it in perspective: Wow, imagine getting so worked about a small thing! Shake your head and move on, Jane. That's what grown-ups do.
And that is my major objection to the premise at the foundation of the book: That we are fools, most of us, who need to be protected from our pasts. I heartily disagree. As the mother of two adult daughters who've kept daily journals since middle school, I can attest that (a) they do not consult the musings of their formal selves about current decisions; (b) when they take out their journals and read them, they have a good laugh at how passionately they felt about incidents that later became unimportant. In other words, instead of their past constraining their current behavior, reminders of past phases help them now to recognize that no matter how big a deal something is today, October 18, 2009, five years from now, it will most likely be a minor footnote in a larger life. This, I would argue, is actually a motive in favor of preserving digital memory. Otherwise, we can be inclined to argue that we have always been [fill in your own blank -- hard-working, sober, reliable, etc.]. Recognizing that we have grown and changed also allows us to free the people up around us from being perfect all the time. Especially, I would argue, the young people in our lives. Presented with evidence that we too once adored silly books, music and political beliefs fosters a tolerance for the paths of change that those around us will undertake too.
Just as we as a society become more tolerant of silly college Facebook party photos several years old, so too will we gain perspective about our own individual lives, recognizing growth and perspective from our wealth of documentation of our lives.
To be fair, Mayer-Schonberger briefly -- very briefly -- toys with this possibility:
It is a striking idea: permanent presence of all information would lose much of its sting if we could consciously disregard old facts and accept that humans are ever-changing beings, never locked into a particular state. Such cognitive adjustment would eliminate the potential danger of clouded decision-making or failing to act in time.
This would appear to be making the same point I argued above, but he undermines it by quickly following up with: "Unlike advocates of a rapid cognitive adjustment, I am skeptical such change is attainable within a couple of human generations." Really? Even when it's clear that culturally, for the most part, we've became used to a heck of a lot in a mere generation: gay couples, swearing in movies, women wearing pants instead of skirts. That's in addition to, as I've argued above, barely batting an eye when a politician curses on camera.
And my final objection to the author's reasoning can be summed up with this quote, about using expiration dates for personal files:
There may be some situations in which individuals are not best suited to choose expiration dates themselves--for example, when individuals are unaware of the potentially harmful consequences of setting too long or too short an expiration date. In such situations, lawmakers may set upper ceilings for expiration dates, or mandate fixed expiration date for certain types of information and/or certain kinds of transactions.
This distrust of the individual to serve as guardian of one's own remembered life--as well as one's public one--is where the author and I part ways permanently. Granted, I confess to being far less concerned about my own informational privacy than I am about societal cleansing of uncomfortable truths, and the author makes legitimate (although I'd still argue verging on the clinically paranoid) arguments for safekeeping keys to identity. But this is exactly why I kept mulling over this book long after finishing it, wanting to have a conversation with the author and a vigorous debate over his premises and conclusions--the thesis is just enough grounded in real concerns to have a ring of truth. This would be a great selection for a book group read, sure to get people chiming in from all sides of the informational divide.