Hillary Clinton and friends spent $1.2 billion this election and all they got was a lousy Jill Stein recount. Political scientists may mark 2016 as the year everything changed. Maxims about conducting polls, raising money, and winning elections are in flux. New paradigms abound, foremost in the art, process, and control of political speech. A growing recognition also exists of the importance of the psychology of persuading voters.

More than ever, online political communication mattered. In the brave new world, a candidate’s tweet, a YouTube video from some obscure group, a cheeky Facebook post, or an underground trolling operation could move the needle. High-budget television and radio advertising campaigns became less effective. Big-spending campaigns and super PACs employing commission-earning consultants often languished or were irrelevant.

Traditional advertising can still help unknown candidates gain name identification or respond to a specific attack. And it can be effective when targeted toward specific demographics, such as Republican primary voters. But it gets brushed aside by organic grassroots enthusiasm generated online. There the crowd ignores image-driven advertising; memes and witticisms delivered in 140 characters rule. Ironically, the old-fashioned rally, perhaps the oldest American campaign tactic, is still an enthusiasm generator.

Cyberspace has also reduced the relevance of government censors who oversee old-style political communications. But it also raises questions about whether the new boss, Silicon Valley tech giants, will be worse than the old. As political speech evolves, new threats emerge as old ones cling to power. The fight for free political speech isn’t over, but it has changed, and we must remain vigilant lest our freedom slip.

About 87 percent of Americans use the Internet and 74 percent use social media, the Pew Research Center recently found. For campaigns, this means that the communications gatekeeping function is as obsolete for campaign consultants as it became for network-news executives last century.

The Internet is the most free-speech-friendly space in history. When the Supreme Court described the country’s political-speech ideal a half-century ago in New York Times Co. v. Sullivan, it wasn’t thinking of cyberspace, but it might as well have been. The Court declared “a profound national commitment to the principle that debate on public issues should be uninhibited, robust, and wide-open, and that it may well include vehement, caustic, and sometimes unpleasantly sharp attacks on government and public officials.”

The Internet embodies that commitment like nothing else. Would-be citizen journalists and pundits with nothing more than an Internet connection and a laptop place their wares in the same arena as do corporations, governments, traditional media companies, and advocacy groups with seven-figure budgets.

The result has been remarkable. People with otherwise ordinary lives and common jobs can exert enormous influence in the political marketplace. Online superstars gain hundreds of thousands of Twitter followers. Groups with shoestring budgets can get millions of views with clever YouTube videos such as the blockbuster “Crush on Obama,” featuring “Obama Girl” (2007). The marketplace chooses winners and losers without regard to advertising budgets or institutional power. For the first time in history, people can obtain information unfiltered by the lens of the state, the editorial judgments of news editors, or the biases of television producers.

The result isn’t perfect. Entrepreneurs searching for easy money peddle “fake news.” Scam PACs juice contributions through emotionally charged e-mails sent to millions with a click. Trolls threaten and harass anonymously.

But freedom isn’t supposed to be perfect. It’s just supposed to be free. More than any other medium, the Internet ensures our founding doctrines: that we remain the ultimate sovereign, the final arbiter of our culture and politics, the deciders of our national destiny.

To its credit, the Federal Election Commission has encouraged this. Ten years ago, the FEC established rules exempting online communications from its regulatory ambit, nurturing the online revolution. That decision to abstain has practical as well as doctrinal benefits. As FEC commissioner Lee Goodman explained:

I cannot imagine how the Federal Election Commission will begin to regulate hundreds of thousands of blogs, YouTube videos, chat rooms, emails and links, and all sorts of Internet-based political discussion because of how vast political discussion on the Internet currently is. . . . [The FEC is being asked] to establish an Internet review board where a room full of government bureaucrats sit on a daily basis and troll the Internet for political commentary — to identify online commentators who did not register or report their expenses in connection with their website, and to issue subpoenas seeking information about their expenditures. I know of no other way that the FEC could regulate the hundreds of thousands of posts on the Internet, absent such a review process.

This laudable view, however, is not universally shared. The FEC is a pinpoint in larger battles about government efforts to control information flows. But to would-be online political speakers, its power is paramount. Seldom does the commission summarily dismiss online-speech cases by applying its Internet exemption. It has done so only once in the past three years. In fact, some commissioners have declared war on it and now want the FEC to police the Internet. These commissioners hold forums, propose rulemakings, and scold the commission for its benighted (in their view) lack of enthusiasm for online regulation.

They use “transparency” as a cudgel to stifle the FEC’s own rules. But transparency has costs on the Internet just as elsewhere. No one proposed that Trump’s election-night protesters register with a bureau, and neither should the government force them to do so when they protest online. This disclosure battle forces speakers large and small, prominent and obscure, to negotiate the FEC’s bureaucratic maze on issues that should be simple.

A small Ohio group posted YouTube videos attacking President Obama and Senator Sherrod Brown. Another criticized presidential candidate Rand Paul for his position on the Iran deal. CREW (Citizens for Responsibility and Ethics in Washington), a once-respected watchdog, posted politically themed press releases on its website. The commission should have cited the Internet exemption and quickly discarded complaints about such activity. Instead, all those cases ended in 3–3 deadlocks, costing the groups time, effort, and legal fees.

FEC obduracy also stifles innovation and hampers democracy. Last year a company called Democracy Rules created an online platform that matched users in a virtual meeting place with philosophically aligned candidates and groups. For a small fee, the company amplified donations of like-minded people and increased their impact. These innovations remove middlemen PACs and vendors who charge hefty fees and legal expenses to troll small-dollar donors performing the same function. Democracy Rules empowers citizens and promotes First Amendment ideals. Yet instead of embracing the effort, the three Democrats on the FEC sought to bring the company under its thumb.

For the first time in history, people can obtain information unfiltered by the lens of the state, the editorial judgments of news editors, or the biases of television producers.

The FEC’s 3–3 split among commissioners limits the damage that any faction can do to speech and innovation. And theoretically, the government must remain viewpoint-neutral. But as political speech moves away from the ability of government to regulate it, new challenges emerge. Technology giants that provide online platforms can ban viewpoints or constrain speech for any other reason. These threats are a new frontier in the never-ending battle to keep speech free.

President Obama recently decried the “wild west” media atmosphere in which outlets are “pumping out all kinds of crazy, toxic stuff.” He called for a “curating function” whereby media would be filtered for “truthiness.” The First Amendment forbids government to so curate, so he was suggesting that Google, Facebook, Twitter, and other major aggregators filter information before letting it reach the masses’ less discerning ears.

The president’s paternalism and hyperbole about the impending threat to democracy is having an effect. The political-media establishment daily denounces subversive and apocryphal online stories. But any “curating” by Silicon Valley giants, given their sometimes-pronounced biases, raises questions about fairness, arbitrariness, secrecy, and censorship, not to mention the danger that such top-down control on the flow of information poses to their profitability or viability.

Any curating functions would inevitably encourage the biases of those performing it. That would quickly devolve into censorship of politically harmful speech not about what is “untrue.” The end goal for this speech policing isn’t censorship of “news” that the pope endorsed Trump but rather the reinstatement of a filter that the Internet has stripped away.

Facebook announced that it would police “truthiness” via Snopes, PolitiFact, Factcheck.org, and ABC News. Some of these organizations have well-known and exposed biases. Snopes is a left-wing operation. In one investigation, not one employee with a conservative background could be found. Its managing editor, Brooke Binkowski, tweeted that Brexit supporters were “pandering to racist mouth-breather ‘Britain First’ types.” Some Snopes employees migrated from the far-left site Raw Story. The Right also distrusts PolitiFact for its liberal spin and malleability of “facts” to fit preferred narratives.

Would-be academic curators fare no better. A Massachusetts professor absurdly designated RedState, The Blaze, and the Independent Journal Review as purveyors of “fake news.” She admitted a bias but argued that the sites “should be considered in conjunction with other news/info sources due to their tendency to rely on clickbait headlines or Facebook descriptions, etc.” In other words, these sites do what every other online source does — amp headlines for views. A CNN personality ludicrously described Breitbart and Fox News as providing “a different set of facts about the world.”

“Real” fake-news outlets, which fabricate stories for clicks and the attendant ad revenue, do exist. But they compose a tiny percentage of posted material. National Public Radio found one person making $10,000 to $30,000 each month generating clicks with salacious headlines. These operations exist to feed peoples’ preexisting biases; they actually persuade no one. The practice is unseemly, but the ultimate curator is the people.

Hillary Clinton conveniently blamed her defeat on “fake news,” among other culprits. But the Democratic nominee ran her own fake-news operation through acolyte David Brock. The mercurial operative paid online trolls to counter negative media stories or opinions about Clinton. He also oversaw a continual media distraction operation through Shareblue, a site that existed to shame journalists into providing Clinton only glowing media coverage. As the New York Times described it, “In the sprawling Clinton body politic, Shareblue is the finger that wags at the mainstream news media.” Stories about papal endorsements are no more dubious than Clinton’s troll operation.

The push to “curate” “fake news” along with other disfavored sources like “hate speech” has dire consequences for political communications. Before announcing its partnership with Snopes and PolitiFact, Facebook had already admitted tweaking its news feed to suppress conservative stories. And some employees wanted candidate Trump’s provocative speech about Muslims banned. Facebook has outsourced its censorship operation to contractors overseas who spend seconds determining a post’s value. Whatever tack Facebook decides on will have consequences for its 1.2 billion users.

Twitter had previously embraced its “wild west” political aggregator role, billing itself “the free speech wing of the free speech party.” But, as with Facebook, its biases have become policy, without clear rules. Twitter has arbitrarily purged or suspended right-wing provocateurs including Milo Yiannopoulos and James O’Keefe as well as the libertarian law professor and commentator Glenn Reynolds. All had to guess at terms-of-service violations and reinstatement policies. Twitter has discriminated against ideologies including Richard Spencer’s neo-Nazi-inspired movement. Suspending accounts for ideological reasons and without forewarning is a dangerous game; it may propel odious movements to greater popularity.

In Europe, social-media companies censor what the European Union considers “hate speech.” A similar agreement here would further erode waning trust in public institutions and present tech companies with intractable problems. It would harm our principle of free speech and imbue political discourse with the stifling conformity that prevails on many college campuses.

The Internet is an incredible human achievement. It has done more to ensure freedom and human rights than any government agency ever could. But such freedom is insecure when would-be censors exist either inside or outside government. We can guard against them by ensuring that government policy and technological processes preserve the right to speak, offend, ridicule, hate, or lie without sanction.

Congress should codify Internet freedom — as in many important respects it already did when it passed the Communications Decency Act in 1996 — by amending the FEC’s enabling statute. Further rulemaking on the scope of the Internet exemption granted by the FEC would reduce the power of recalcitrant agencies to ignore it.

Platforms should find technological fixes to thwart those who intimidate or harass. One online guru proposed 23 technological fixes that would increase users’ power to regulate their experience on Twitter. Facebook could ensure similar tools, although online threats decrease when anonymity does.

Innovation, shame, and market share are the tools to combat private censors.

Online providers should improve their terms-of-service policies on free speech or the lack thereof. They should be more transparent in how the policies are enforced. Most important, if they reduce a user’s speech, the companies should, when asked, tell the user how and why they took that action. Failure to do so might show bad faith, which could remove the law’s immunity for their decisions.

But we must also accept that the Internet reflects the society and people who use it. Banning odious ideologies or con men only makes them more attractive to the disaffected. Nothing will stop trolls, hucksters, and charlatans from existing online. But their presence should concern us less than the presence of “curators” seeking to silence them. Internet failings are human failings. They can’t be banned, but they can be exposed.

The media should attempt to regain their lost credibility. According to a Gallup poll from September, Americans’ trust in the media is at its nadir. Refocused efforts at unbiased reporting and ideological diversity would lessen the impact of “genuine” fake news. But when CNN never has a map that shows Donald Trump winning, when news outlets are continuously exposed for their bias, when reporters give advance copies of stories to their subjects, when failure is rewarded with promotions, when ideological conformity is enforced, people will continue to look for alternatives.

#related#Innovation, shame, and market share are the tools to combat private censors. A slew of high-profile defectors, coupled with the risk of being burdened with a reputation for censorship, will cause platforms to rethink meddling. Tumblr, known for its toxic brand of “Tumblr feminism,” is struggling. Its parent company, Yahoo, wrote down $230 million this year for the network of micro-blogging sites. Twitter has had its own profitability struggles.

Facebook is skating on thin ice. It should reflect that the nation is almost evenly divided politically. If a curating operation leans left or right, it risks losing a big chunk of users and therefore revenues.

Who we are as people is defined by whom we answer to. The ultimate responsibility for policing our speech should be ours alone, unencumbered by bureaus or, to the extent possible, technology executives. We must constantly remain vigilant. Whatever technological changes and advances come, we must remain the ultimate arbiter of our destiny.