:format(webp)/cdn.vox-cdn.com/uploads/chorus_image/image/55575267/GettyImages_118313163.0.jpg)
"Epidemic!" "Hate-based criminality!" "Threat to democracy!"
This is how some of the most powerful politicians in three of the world's largest economic and military powerhouses have described what has been broadly defined as "fake news."
Hillary Clinton referred to the phenomenon of fake news as an "epidemic" about a month after the impeccably credentialed candidate lost a seemingly unlosable election to a scandal-plagued reality show host with the highest unfavorable rating of any presidential candidate in history. At a ceremony honoring the former Nevada Sen. Harry Reid on his retirement, Clinton said fake news "is a danger that must be addressed and addressed quickly."
She referred in particular to the incident at a popular Washington, DC, pizzeria called Comet Ping Pong — where a deranged and armed man duped by an outlandish conspiracy theory originating on Reddit decided to mount his own “investigation.” He ended up opening fire while looking for a child sex-trafficking ring that online fantasists asserted was operated out of the eatery by longtime Clinton confidante John Podesta.
At Recode's Code Conference in Ranchos Palos Verdes last month, Clinton went even further, citing unnamed studies that she claims determined "the vast majority of the news items posted [on Facebook] were fake” — a massive overstatement. Clinton also laid the blame of the proliferation of fake news on the "millions of bots" following Donald Trump's Twitter account. While it's true Trump's following does contain millions of bots, that's pretty much standard for any well-known celebrity, including Clinton.
Concerns about fake news — a term whose meaning has proved to be flexible, but which originally referred to false news claims presented in an official-looking way, often online, in a way intended to deceive readers — have crossed the pond as well. The UK’s House of Commons Culture, Media, and Sport Committee has launched an investigation into fake news that aims to pin down a definition of the term and determine whether and how social media companies ought to be compelled to prevent its proliferation. The committee's chair, Conservative MP Damian Collins, said that the “phenomenon of fake news is a threat to democracy and undermines confidence in the media in general."
Meanwhile in Germany, Justice Minister Heiko Maas recently unveiled a draft of a law called the Network Enforcement Act, or NetzDG, which specifically cites "the experience in the US election campaign" as one reason for a crackdown on "punishable false reports (‘fake news’).” One stick used to discourage such false reports would be fines on social media companies of up to 50 million euros (roughly $53 million). The act refers to “hate-based criminality” as the one of the driving forces behind fake news, and asserts that self-regulation on the part of companies “has proven to be insufficient.”
NetzDG, which was approved by the German cabinet but still needs to be ratified by parliament before September's elections if it is to become law, offers perhaps the clearest window yet into what it might take to police the internet in the way that people like Hillary Clinton would like — and the example is far from reassuring. The proposed law imposes such requirements as providing an "easily identifiable, directly accessible, and continuously available" method for users to report illegal content (say, a link button) which in Germany would encompass many forms of speech that would be protected by the First Amendment in the United States, including racism (which is defined broadly) and Holocaust denial.
As Emma Llanso of the Washington, DC-based Center of Democracy and Technology's Free Expression Project notes, NetzDG wouldn’t just cover social media companies — which would be bad enough — but any company that helps people “exchange or share any kind of content with other users or make such content accessible to other users.” That would include Gmail and web-hosting services. It would also require these to monitor not just “false reports” but a number of other kinds of speech banned under German law, including a whopping 24 provisions of German law that also prohibit things like "defamation of the President, the state, and its symbols," "defamation of religions," and "distribution of pornographic performances." There's a whole lot of subjectivity crammed into this draft law.
:format(webp):no_upscale()/cdn.vox-cdn.com/uploads/chorus_asset/file/8799703/GettyImages_632641364.jpg)
Complaints over such materials would have to be addressed "immediately" by the companies. If deemed "clearly illegal” by corporate staff, the offending content would have to be removed within 24 hours; more ambiguously illegal content would have to be removed within seven days. But there's more — a lot more.
The social media websites must maintain a copy of the illegal material, expunge all other copies from the site, and prevent them from being re-uploaded (a near impossibility, considering the speed and scope of the internet), and explain the reasons for any action, or inaction, to the complainant. Senior management must submit "monthly reviews" on their compliance with the law and provide mandatory quarterly reports of all complaints and subsequent actions taken in response. It's exhausting just reading all the requirements, but I encourage you read concise summary of the draft law here on the legal privacy blog of the law firm Alston and Bird.
Among Western democracies, Germany is particularly sensitive (and not without good reason) about allowing the proliferation of things like Holocaust denial and racist or bigoted expression. That is to say, Germany does not allow such speech, which can be punishable with both fines and prison time. In the United States, the First Amendment protects the right to almost all speech — with essentially only direct incitements to violence, credible threats, libel, and slander left undefended (and very high bars to prove the last two in court). One way to look at this is Germany is demanding that private companies merely enforce existing law. But another perspective holds that corporations should not be deputized as de facto censors in service of the government.
A law as sweeping as Germany's NetzDG would be untenable in the US, and is very likely so broad as to be unenforceable anywhere, including Germany. This is essentially what Facebook asserted in a statement released last month, in which the company said it "shares the federal [German] government’s concern regarding hate speech and false news online," but that NetzDG "is not the right way to achieve these political goals." The company's statement added, "The draft law provides an incentive to delete content that is not clearly illegal when social networks face such a disproportionate threat of fines.” Through fear of being out of compliance with the law, in short, Facebook might err on the side of censorship.
Google and Facebook ramp up private efforts to curb fake news
The social media behemoth knows its reputation is tied to the quality of content on the site, and it recently announced its plans to hire 3,000 new employees to join an already-existing team of over 4,500 fact-checkers and monitors to police violent and other over-the-top offensive content. That’s on top of a user-driven system that allows for objectionable content to be flagged and removed. (According to the Pew Research Center, half of all Americans rely on Facebook as a source of news.)
In December 2016, Facebook also began outsourcing judgment on "questionable" news items to fact-checking media outlets including Snopes, ABC News, PolitiFact, Factcheck.org, and the Associated Press, who have the ability to essentially bury "disputed" content down users' news feeds, with the intention of making "fake news" harder to be seen.
Internationally, in the run-up to the elections in France and the United Kingdom, Facebook also tried to mitigate its potential role as an enabler of deliberately misleading items — meant to sway public opinion in the run-up to the elections in France and the United Kingdom — by deleting tens of thousands of fake accounts, according to the BBC, as well as by running "Tips for spotting false news" in full-page advertisements in major newspapers in both countries. However, there are limits to how far Facebook’s shareholders are willing to go. By a vote of more than 6 billion to 50 million (each Class B share counts for 10 votes), they recently weighed in against creating a study on how fake news proliferates on Facebook.
Google, whose shareholders also rejected the idea of such a study, recently announced plans to rework its search engine to prevent it from directing people to bogus, defamatory claims — such as Barack Obama's attempted coup against President Trump, or President Trump's popular vote victory over Clinton. It will also ask 10,000 evaluators to draw on an updated quality control manual to ferret out fake news. These guardians of the search engine are supposed to adorn such content with Google's "Lowest" rating, making it among the last things to appear in a user search.
Here's where it gets tricky, both for Facebook and Google. There are some cases in which the falseness of a news claim will be inarguable, as with Comet pizzeria and the Democratic sex ring. But in other cases, one person's fake newsmaker is another person's bold truth-teller. Will the new and improved Google search system penalize climate change skeptics, for example? Will the in-house evaluators at these companies downgrade the many uncorrected op-eds blaming Trump for tacitly encouraging the wave of over 100 bomb threats at Jewish centers from January to March of this year (which turned out to be mostly the work of an Israeli-American teenager and a disgraced leftist journalist)?
Will supermarket tabloids with "Bat Boy" covers be exempt? What about religious-based material, which is often inflammatory and almost never grounded in reality? Was the "Pizzagate" shooter relying on fake news any more than the nutcase who in 1989 slugged the legendary astronaut and senator, John Glenn, in the jaw on the steps of the Smithsonian because he was "guided" by his own fevered interpretations of the Christian Bible?
A representative at Google declined to answer questions along those lines. Instead, she directed me to an April blog post touting the company's "latest quality improvements for Search." In that post, the company asserts "it's become very apparent that a small set of queries in our daily traffic (around 0.25 percent), have been returning offensive or clearly misleading content" — Holocaust denial, for example. Beyond labeling such pages as low-quality, it does not appear that Google is looking to get into the business of policing op-ed pieces.
But that still leaves the door open for Facebook — which has been accused, in the past, by former employees, of suppressing news from right-leaning sites. Conservative critics have also raised not unreasonable concerns that some of the more popular fact-checking outlets often bring their own political biases to the table. Even the act of choosing which stories to fact-check is frequently informed by the fact-checkers' politics.
A proposal to revive laws banning “criminal libel”
In a recent article titled "Free Speech and Fake News," the philosopher Peter Singer declared fake news to be "a threat to democratic institutions.” And he hyperbolically implied that admitted performance artist Alex Jones's repeating of the farcical "Pizzagate" story presented a "clear and present danger" to the republic and is thus unprotected by the First Amendment. (In a video that was viewed more than 400,000 times on YouTube before it was taken down, Jones referring to “all the children Hillary Clinton has personally murdered and chopped up and raped.”) Singer dismisses former Supreme Court Justice Louis Brandeis's proposed remedy fighting such fallacies ("more speech, not enforced silence") as "naïve," and proposes making libel — normally a local, civil court matter — into a federal criminal offense.
Singer notes that criminal libel was used in Britain "for centuries" before its abolition in 2010. What he fails to note is that when it was finally done away with, it was artists, writers, and left-leaning politicians who spearheaded the effort to repeal. As novelist Will Self put it at the time, "This has always been a shape-shifting law, capable of being employed as a cudgel against satirists, incendiarists, malcontents and revolutionaries alike."
:format(webp):no_upscale()/cdn.vox-cdn.com/uploads/chorus_asset/file/8799751/GettyImages_457061257.jpg)
Considering the fact that President Trump has repeatedly mused about "opening up libel laws" to go after his political enemies, namely, the New York Times, perhaps it is Singer who is being naïve. The distinguished philosopher is suffering from a failure of imagination if he fails to foresee how this awesome expansion of criminal liability for speech could be abused by bad actors in power — even though Trump's election appears to partly motivate his call to increase the power of government to crack down on speech.
It isn't only fake news that's driving calls for social media reform. Following the Facebook Live streaming of an elderly man's murder earlier this month, Jesse Jackson suggested a 30-day "moratorium" on Facebook's streaming app to serve as a "time out" so the company can find ways to keep its users from making it "a platform to release their anger, their fears, and their foolishness."
The live-streaming of a murder was horrifying, but short of getting rid of the app itself, it's not clear how Facebook could have stopped its brief, though wide, dissemination. (Once alerted to the video’s existence, the company promptly removed it from the site and vigorously monitored attempts to repost it.) And, as Alyssa Rosenberg astutely explained, in the Washington Post killers have long found a way to document their crimes and project to an audience — take the Virginia Tech mass murderer's videotaped rants (which were aired by NBC News) or the disgruntled ex-TV reporter who two years ago stalked two former colleagues until they were broadcasting live before gunning them down in cold blood.
No one wants their children or grandparents opening Facebook and reading — and believing — false news stories suggesting that presidential candidates have committed crimes, or are suffering from terminal illnesses, or have been endorsed by the pope. And a recent bombshell report in the Washington Post demonstrates that Vladimir Putin himself sees the value in using fake news as an effective weapon in the game of geopolitical sabotage.
But as I consider the potential unintended consequences of cracking down on fake news, I keep coming back to Facebook’s mishandling of the iconic photo of a naked 9-year-old Vietnamese girl running from the United States’ napalm bombing of her village during the Vietnam War. After a Norwegian author published the photo as part of a commentary on the evils of war, Facebook took the photo down, saying it violated standards regarding nudity on the site. Based on the decision of some unknown Facebook employee, the powerful photo was swept up and thrown down the objectionable content memory hole. (Facebook restored the image only after a protest that involved thousands of users posting the photo on their own pages.)
And as Brooke Borel points out at FiveThirtyEight, fake news has existed in the United States since the country’s beginning. You might include, in that category, partisan publications hurling invective at their opponents — painting them in the absolute worst light, sometimes stretching facts —and "penny presses" publishing "humbugs,” deliberately false stories meant to serve as entertainment, right alongside the real news. Borel also makes the smart point that feel-good fake stories are at least as likely to proliferate in mainstream culture as sweaty alt-right conspiracy theories, which further confuses the issue of policing "fake news."
This past March, California Democrats were scheduled to hold hearings on a proposed bill called the California Political Cyberfraud Abatement Act, which would have made it "unlawful for a person to knowingly and willingly make, publish or circulate on an Internet Web site…a false or deceptive statement designed to influence the vote." As Dave Maass of the Electronic Frontier Foundation put it in a blog post opposing the bill, such a law would not only be flagrantly unconstitutional, but would also provide the added benefit to our already overcriminalized society by making it "illegal to be wrong on the internet if it could impact an election.”
The bill was quietly tabled without a public hearing shortly after EFF's public pushback, but the fact that lawmakers let it get even that far is troubling.
To many, Trump's election may feel like such a cataclysmic event that the continued existence of the country requires reconsidering our robust protections for free speech — even if that means some nonsense, even malign nonsense, sees the light of day. That would be a mistake — every bit as much as it was a mistake for the wounded American populace to have willfully gone along with the many civil liberties violations carried out in the name of our security in the days, months, and years after 9/11.
Anthony L. Fisher is a journalist and filmmaker in New York whose work has also appeared in the Daily Beast, The Week, New York Daily News, and Reason. Fisher wrote and directed the feature film Sidewalk Traffic, available on major video-on-demand platforms.
The Big Idea is Vox’s home for smart discussion of the most important issues and ideas in politics, science, and culture — typically by outside contributors. If you have an idea for a piece, pitch us at thebigidea@vox.com.
