Trust Me, I'm Lying: Confessions of a Media Manipulator

by Ryan Holiday

  These were the old rules:

  1. If the outlet is legitimate, the stories it breaks are.

  2. If the story is legitimate, the facts inside it are.

  3. It can be assumed that if the subject of the story is legitimate, then what people are saying about it probably is too.

  These rules allow one journalist to use the facts brought forth by another, hopefully with attribution. This assumption makes researching much easier for reporters, since they can build on the work of those who came before them, instead of starting from the beginning of a story. It’s a process known as the “delegation of trust.”1

  The web has its own innovation on the delegation of trust, known as “link economy.” Basically it refers to the exchange of traffic and information between blogs and websites. Say the Los Angeles Times reports that Brad Pitt and Angelina Jolie are splitting up. Perez Hilton would link to this report on his blog and add his own thoughts. Then other blogs would link to Perez’s account and maybe the original Times source as well. This is an outgrowth from the early days of blogging, when blogs lacked the resources to do much original reporting. They relied on other outlets to break stories, which they then linked to and provided commentary on. From this came what is called the link economy, one that encouraged sites to regularly and consistently link to each other. I send you a link now, you send me a link later—we trade off doing the job of reporting.

  The phrase “link economy” was popularized by Jeff Jarvis, who you met here earlier. His credentials as a blogger, journalism professor at the City University of New York’s Graduate School of Journalism, and author of books such as What Would Google Do? have made him incredibly influential. Unfortunately, he’s also an idiot, and the link economy he advocates is a breeding ground for manipulation.

  The link economy encourages blogs to point their readers to other bloggers who are saying crazy things, to borrow from each other without verification, and to take more or less completed stories from other sites, add a layer of commentary, and turn it into something they call their own. To borrow a term from computer science, the link economy is recursive—blogs pull from the blogs that came before them to create new content. Think of how a mash-up video relies on other clips to make something new, or how Twitter users retweet messages from other members and add to them.

  But as the trading up the chain scam makes it clear, the media is no longer governed by a set of universal editorial and ethics standards. Even within publications, the burden of proof for the print version of a newspaper might differ drastically from what reporters need to go live with a blog post. As media outlets grapple with tighter deadlines and smaller staffs, many of the old standards for verification, confirmation, and fact-checking are becoming impossible to maintain. Every blog has its own editorial policy, but few disclose it to readers. The material one site pulls from another can hardly be trusted when it’s just as likely to have been written with low standards as with high ones.

  The conditions on which the delegation of trust and the link economy need to operate properly no longer exist. But the habits remain and have been mixed into a potent combination. The result is often embarrassing and contagious misinformation.

  Like the time when Crain’s New York e-mailed me to ask if American Apparel would be closing any of its stores in Manhattan because of the financial crisis. No, I replied emphatically. No. So they found a real estate agent who didn’t work for American Apparel to say we might. Headline: “American Apparel likely to shed some NY stores” (even though my quote in the article said we wouldn’t). The Crain’s story was linked to and used as a source by Jezebel, and then by New York magazine’s The Cut blog, then by Racked NY. AOL’s Daily Finance blog turned it into a slideshow: “10 Leading Businesses Shuttering Stores Because of Downturn.” None of those sites needed to ask me any questions, since Crain’s had asked and answered for them—they could just link.* A week later, for unknown reasons, Crain’s republished the article under a new headline (“Unraveling American Apparel Could Put NYC Stores on the Block”), which, after showing up on Google Finance, started the same chain over again.2

  More than a year later every one of those stores is still open. The links still point to the same bad story.

  A few years back a young Irish student posted a fake quotation on the Wikipedia page of composer Maurice Jarre shortly after the man died. (The obituary-friendly quote said in part, “When I die there will be a final waltz playing in my head that only I can hear.”) At the time, I’m not sure the student understood the convergence of the link economy and the delegation of trust. That changed in an instant, when his fabricated quote began to appear in obituaries for the composer around the world.

  It’s difficult to pinpoint where it started, but at some point, a reporter or a blogger saw that quotation and used it in an article. Eventually the quote found its way to The Guardian, and from there it may as well have been real. The quote so perfectly expressed what writers wished to say about Jarre, and the fact that it was in The Guardian, a reputable and prominent newspaper, made it the source of many links. And so it went along the chain, its origins obscured, and the more times it was repeated, the more real it felt.

  This is where the link economy fails in practice. Wikipedia editors may have caught and quickly removed the student’s edit, but that didn’t automatically update the obituaries that had incorporated it. Wikipedia administrators are not able to edit stories on other people’s websites so the quote remained in The Guardian until they caught and corrected it too. The link economy is designed to confirm and support, not to question or correct. In fact, the stunt was only discovered after the student admitted what he’d done.

  “I am 100 percent convinced that if I hadn’t come forward, that quote would have gone down in history as something Maurice Jarre said, instead of something I made up,” he said. “It would have become another example where, once anything is printed enough times in the media without challenge, it becomes fact.”3

  The proponents of the link economy brush aside these examples. The posts can be updated, they say; that’s the beauty of the Internet. But as far as I know there is no technology that issues alerts to each trackback or every reader who has read a corrupted article, and there never will be. The evolution of a news story is a lot like biological evolution. It jumps around, cross contaminates, and occasionally develops at the same time in multiple places. It’s impossible to track or correct.

  Senator Eugene McCarthy once compared the journalists covering his 1968 presidential campaign to birds on a telephone wire. When one got up to fly to a different wire, they’d all follow. When another flew back, the rest would too. Today this metaphor needs an update. The birds still follow one another’s leads just as eagerly—but the wire need not always exist. They can be and often are perched on illusions, just as blogs were when they repeated Maurice Jarre’s manufactured remarks.

  THE LINK ILLUSION

  In the link economy, the blue stamp of an html link seems like it will support weight. (As had the links to The Guardian story containing the false quote.) If I write on my blog that “Thomas Jefferson, by his own remarks, admitted to committing acts considered felonious in the State of Virginia,” you’d want to see some evidence before you were convinced. Now imagine that I added a link to the words “acts considered felonious.” This link could go to anything—it could go to a dictionary definition of “felonious acts” or it could go to a pdf of the entire penal code for the state of Virginia. Either way, I have vaguely complied with the standards of the link economy. I have rested my authority on a source and linked to it, and now the burden is on the reader to disprove the validity of that link. Bloggers know this and abuse it.

  Blogs have long borrowed on the principle that links imply credibility. Even Google exploits this perception. The search engine, founded by Larry Page and Sergey Brin when they were Stanford students, copies a standard practice from academia in which the number of citations a scientific paper gets is an indic
ator of how influential or important it is. But academic papers are reviewed by peers and editorial boards—shaky citations are hard to get away with.

  Online links look like citations but rarely are. Through flimsy attribution blogs are able to assert wildly fantastic claims that will spread well and drive comments. Some might be afraid to make something up outright, so the justification of “I wasn’t the first person to say this” is very appealing. It’s a way of putting the burden all on the other guy, or on the reader.

  People consume content online by scanning and skimming. To use the bird metaphor again, they are what William Zinsser called “impatient bird[s], perched on the thin edge of distraction.” Only 44 percent of users on Google News click through to real the actual article. Meaning: Nobody clicks links, even interesting ones. Or if they do they’re not exactly rigorous in pouring over it to make sure it proves the point in the last article they read.

  If readers give sites just seconds for their headlines, how much effort will they expend weighing whether a blog meets the burden of proof? The number of posts we read conscientiously, like some amateur copy editor and fact-checker rolled into one, are far outpaced by the number of articles we just assume are reliable. And the material from one site quickly makes its way to others. Scandalous statements get traction wider and faster—and their dubious nature is more likely to be obscured by the link economy when it’s moving at viral speed. Who knows how many times you and I have passed over spurious assertions made to look legitimate through a bright little link?

  A BROKEN PHILOSOPHY

  May becomes is becomes has, I tell my clients. That is, on the first site the fact that someone “may” be doing something becomes the fact that they “are” doing something by the time it has made the rounds. The next time they mention your name, they look back and add the past tense to their last assertion, whether or not it actually happened. This is recursion at work, officially sanctioned and very possible under the rules of the link economy.

  Under these circumstances it is far too easy for mistakes to pile on top of mistakes or for real reporting to be built on lies and manipulations—for analysis to be built on a foundation of weak support. It becomes so easy, as one reporter has put it, for things to become an amalgam of an amalgam.

  The link economy encourages bloggers to repeat what “other people are saying” and link to it instead of doing their own reporting and standing behind it. This changes the news from what has happened into what someone said the news is. Needless to say, these are not close to the same things.

  One of my favorite books is Kathryn Schulz’s Being Wrong: Adventures in the Margin of Error. Though media mistakes are not the subject of the book, Schulz does do a good job of explaining why the media so regularly gets it wrong. Scientists, she says, replicate each other’s experiments in order to prove or disprove their findings. Conversely, journalists replicate one another’s conclusions and build on top of them—often when they are not correct.

  The news has always been riddled with errors, because it is self-referential instead of self-critical. Mistakes don’t occur as isolated incidents but ripple through the news, sometimes with painful consequences. Because blogs and the media have become so interdependent and linked, a lapse of judgment or poor analysis in one place affects many places.

  Science essentially pits the scientists against each other, each looking to disprove the work of others. This process strips out falsehoods, mistakes, and errors. Journalism has no such culture. Reporters look to one-up each other on the same subjects, often adding new scoops to existing stories. Meanwhile, people like Jeff Jarvis explicitly advise online newspapers and aspiring blogs not to waste their time trying “to replicate the work of other reporters.” In the age of the link, he says, “this is clearly inefficient and unnecessary.” Don’t waste “now-precious resources matching competitors stories” or checking and verifying them like a scientist would. Instead, pick up where they left off and see where the story takes you. Don’t be a perfectionist, he’s saying; join the link economy and delegate trust.

  When I hear people preach about interconnectedness and interdependence—like one reporter who suggested he and his colleagues begin using the tag NR (neutral retweet) to preface the retweets on Twitter that they were posting but not endorsing—I can’t help but think of the subprime mortgage crisis. I think about one bank that hands off subprime loans to another, which in turn packages them and hands them to another still. Why are you retweeting things you don’t believe in?! I think about the rating agencies whose job was to monitor the subprime transactions but were simply too busy, too overwhelmed, and too conflicted to bother doing it. I think of falling dominoes. I wonder why we would do that to ourselves again—multiplied many times over in digital.

  Of course replication is expensive. But it is a known cost, one that should be paid up front by the people who intend to profit from the news. It is a protection and a deterrent all at once. The unknown cost comes from failure—of banks or of trust or of sources—and it is borne by everyone, not just the businesses themselves.

  When Jarvis and others breathlessly advocate for new concepts they do not understand, it is both comical and dangerous. The web gurus try to tell us that this distributed, crowd-sourced version of fact-checking and research is more accurate, because it involves more people. But I side with Descartes and have more faith in a scientific approach, in which every man is responsible for his own work—in which everyone is questioning the work of everyone else, and this motivates them to be extra careful and honest.

  The old media system was a long way from perfect, but their costly business model, so derided by these web gurus, pushed for at least a semblance of scientific replication. It found independent confirmation wherever possible. It advocated editorial independence instead of risky interdependence. It is expensive, sure, and definitely unsexy, but it is a step above the pseudo-science of the link economy. It was certainly better than what we have online, where blogs do nothing but report what “[some other blog] is reporting …” where blogs pass along unverified information using the excuse, “but I linked to where I stole it from.”

  To simply know where something came from, or just the fact that it came from somewhere else, does not alleviate the problems of the delegation of trust. In fact, this is the insidious part of the link economy. It creates the appearance of a solution without solving anything. Some other blog talked to a source (don’t believe them; here’s the link) so now they don’t have to. That isn’t enough for me. We deserve better.

  I happened to get lucky that CNN decided not to run their poorly sourced story. I appealed to their reason and humanity and it worked. Nearly two years have passed since then. To this day I consider the incident a fluke, and I assume I will never be so lucky again. And neither will anyone else.

  * One blogger from AnnArbor.com did e-mail me. He asked, “Since AA was closing stores in NYC, would we be closing in Ann Arbor too?” No. No! It didn’t stop him, either.

  XVII

  EXTORTION VIA THE WEB

  FACING THE ONLINE SHAKEDOWN

  IN BYGONE DAYS A COMPANY MIGHT HIRE A PR MAN TO make sure people talked about their company. Today, even a company with little interest in self-promotion must hire one, simply to make sure people don’t say untrue things about their company. If it was once about spreading the word, now it’s as much about stopping the spread of inaccurate and damaging words.

  When the entire system is designed to quickly repeat and sensationalize whatever random information it can find, it makes sense that companies would need someone on call 24/7 to put out fires before they start. That person is often someone like me.

  One of my first big contracts was a ten-thousand-dollar gig to handle a group of trolls who had been vandalizing a company’s Wikipedia page and filling it with lies and rumors. These “facts” were then showing up in major newspapers and on blogs that were eager for any gossip they could find about the company. How do we just make it stop?, th
e company pleaded. We just want to be left alone.

  It’s the same predicament Google found itself in when Facebook hired a high-profile PR agency to execute an anonymous whisper campaign against them through manufactured warnings about privacy. Bloggers of all stripes had been pitched, with the idea of building enough buzz for the grand finale: editorials in the Washington Post, Politico, USA Today, and the Huffington Post. Like my client, Google was stunned senseless by the plot. Imagine a $200 billion company saying, Make it stop. We just want to be left alone. But they were effectively reduced to that. “We’re not going to comment further,” Google told reporters during the firestorm of controversy. “Our focus is delighting people with great products.”

  Sure, go ahead and focus on that Google, but it doesn’t matter. Once this arms race has begun, things can’t just go back to normal. It escalates: A company sees how easy it is to plant stories online and hires a firm to attack its competitor. Blindsided by the bad publicity, the rival hires a firm to protect itself—and then to strike back. Thus begins an endless loop of online manipulation that can cost hundreds of thousands of dollars. And that’s the easiest of the PR battles a company may have to face.

  Consider what happened to the French yogurt giant Danone, which was approached by Fernando Motolese, a video producer in Brazil, with two hypothetical videos.

  One, he said, was a fun spoof of their yogurt, which was designed to improve digestive health and, um, other bodily functions. The other, he said, was a disgusting version of the first video, with all the indelible scatological images implied by such a spoof. He might be more inclined to release the first version, he said, if Danone was willing to pay him a fee each time it was seen.