The Internet Is Not Always ‘Forever’

By Charles C. W. Cooke

Friday, April 23, 2021

 

The promise of the Internet was freedom — from censorship, from gatekeepers, from monopolies. Without choke points, early architects promised, the truth would not merely out; it would deposit itself into a permanent archive, which, over time, would serve as a boon to historians and hoarders alike. Hence the pithy claim that “the Internet,” unlike real life, “is forever.”

 

In many important ways, this promise has come to pass. And yet, because a number of the Internet’s core functions have been so aggressively centralized, it has also proven itself to be a remarkably efficient machine for the rewriting of history. Whether this second tendency will turn out to have been a blip or an augury remains an open question.

 

Prior to the existence of the Internet, the second edition of any product inevitably came in a different box. The morning edition of a newspaper was physically discrete from the evening edition. A book with the typos corrected could be placed next to the error-ridden version on the same shelf. An edited or remastered movie would be released as its own DVD. This, though, is no longer true. Today, the second editions of digital products often completely replace their predecessors. If a downloaded video game is found to have issues after publication, the developers simply produce a patch, upload it, and have the end users overwrite their mistakes. The same goes for online news, which is often updated without correction, and for all manner of Web-based software. You may, if you are fortunate, be able to see what Twitter looked like in 2014. But you cannot use the Twitter of 2014. It is gone, forever.

 

Often, this practice is innocuous; nobody, after all, wants a broken copy of Doom. But what happens when public figures begin to realize that they can use the iterative nature of the Internet to rewrite their own histories — or, at least, to make it difficult to track the changes that they have made? This is not a hypothetical inquiry. At present, the New York Times is engaged in a dispute with its union over whether it would be acceptable for its Web team “to retroactively correct the bylines of transgender journalists who change their names.” The union wishes to alter the work certain writers did on the grounds that it was “wrong” all along and must be corrected after the fact to better comport with cosmic reality; the Times opposes this project.

 

I do not wish to get into the merits of the union’s argument here. Rather, I want to note that, until a few years ago, it could quite obviously not have made the demand it is making now, and to propose that the fact that such a demand is now possible has extraordinary implications for our understanding of history. Clearly, the Times cannot go back and collect up all the copies of its newspaper that contain articles by writers who have subsequently changed their names. On its website, though, it can in effect do exactly that. And, if it does, most readers will never have a clue that it has done so.

 

In other industries, the union’s request is already being granted as a matter of course. Until recently, the actor now known as “Elliot Page,” a man, was instead known as “Ellen Page,” a woman. Indeed, for years, Page’s public name was “Ellen,” a fact that was reflected on IMDB, on Netflix, on Wikipedia, and so on. Last year, though, this changed — and, with it, so did the history of cinema. Today, Apple Movies, Amazon Prime Video, Hulu, and others list the star of the movie Juno — which is about a girl who gets pregnant — as “Elliot Page,” even though this was not Page’s name when the movie was made and distributed.

 

Again, I am not especially interested in arguing over the merits of Page’s core claim; in the grand scheme of things, what our thespians wish to be known as is just not that important a question. Instead, I am wondering how this capability could be used for more sinister ends as it becomes more widespread. At some point, physical copies of products such as newspapers, albums, movies, and video games will die out in favor of wholly ephemeral versions that are stored on remotely run servers and licensed with a key. And, when they do, those who run the services that provide them are going to enjoy incredible power over history. Historians looking back at the life of Muhammad Ali will not find it difficult to pinpoint the moment that he ceased to be Cassius Clay. But if he’d been born in 2020, they very well might.

 

Pretty much everyone in the world has an incentive to change something about their history. Newspapers that make mistakes have an incentive to edit their pieces without acknowledgment; politicians who consider a given speech or appearance embarrassing have an incentive to erase it from the record; corporations that upset their customers have an incentive to make their mistakes disappear. Because people still make copies, because the Web thrives on hostile interaction, and because major events are still widely covered, it remains difficult for individual players to completely paint over the past. But it would be foolish to pretend that the digital age does not make such an endeavor easier than it has been at any other point in human history, and it would be foolish, too, to pretend that there are no examples of mass compliance with the revised narratives that result. The New York Times will eventually cave to its employees’ demand, and, when it does, everyone else in its orbit will, too. Perhaps, then, a new slogan will be in order: “All the News That’s Fit to Print — and Alter without Notice, If Necessary.”