Testimony before Subcommittee on Communications and Technology United States House of Representatives
December 1, 2021
Research Fellow, Technology Policy
The Heritage Foundation
My name is Kara Frederick. I am a Research Fellow for Technology Policy at The Heritage Foundation. The views I express in this testimony are my own and should not be construed as representing any official position of The Heritage Foundation.
Social media companies were ostensibly created to democratize access to information and give ordinary citizens a voice in the public square. Today, reality has proven much different.
In Facebook’s case, as recently as 2019, representatives of the company came out strong in their stated desire to “stand up for freedom of expression.” CEO Mark Zuckerberg’s refrain that “more speech” is the best antidote to “bad speech,” his repeated desire to not be an arbiter of truth, and additional support for free speech during a 2019 address at Georgetown University articulated a different vision for Facebook’s content-moderation practices than those implemented since.
Actions taken by Facebook and other Big Tech companies like Google and Twitter in the past two years demonstrate a marked departure from these values.
Special privileges from the government—in the form of immunity from civil liability under Section 230 of the 1996 Communications Decency Act—enable the betrayal of these values.
Big Tech companies have moved far afield from Congress’ original intent in crafting the legislation and are now acting as publishers of information—retaining the benefits of being a publisher without the legal responsibility. As Federal Communications Commission (FCC) commissioner Brendan Carr notes, Section 230 amounts to a “unique regulatory legal advantage for one set of political actors.” These actors, primarily Big Tech companies, are abusing this privilege through continued viewpoint censorship, opaque content-moderation decisions, inconsistent application of vague rules, and a lack of transparency and genuine recourse for their users.
The outlook is not sanguine. A lack of accountability and the “sweeping immunity” conferred on Big Tech by broad interpretations of Section 230 over the years has emboldened these companies to abuse their concentrations of power and increase their symbiotic relationships with the federal government, constrict the digital lives of those who express specific political views, and sharpen digital surveillance on ordinary Americans. Put simply, Big Tech companies are not afraid of the American people or meaningful checks on their abuse of power. And it shows.
To help rectify this, targeted reforms of Section 230 should be enacted—at bare minimum. The “twenty six words that created the internet” have outlived their original purpose. Section 230 is in need of a refresh.
Impact on the Body Politic: Information Access, Manipulation, and the Culture of Free Speech
The scale and reach of these digital platforms render them transformative—they control the flow of information in such an expansive way as to fundamentally shape the public square. This is embodied in Big Tech’s willingness to leverage their market dominance to shut off direct access to digital information, engage in a demonstrated pattern of information manipulation, and adversely impact America’s culture of free speech.
Information Access. Practically, these companies’ positions as global oligopolies translate into the ability to limit access to information. Not all censorship is created equal. Censorship conducted at the cloud hosting infrastructure or Internet service provider (ISP) level severely curtails direct access to digital viewpoints and actors who run afoul of these providers. By controlling these lower levels of the technical stack upon which many other levels (like digital platforms and applications) depend, a small group of unelected tech executives can pull the plug on entire companies—with very limited recourse.
The case of Twitter competitor Parler in January of 2021 illustrates how this can happen. Google and Apple removed Parler from their stores at the application layer of the technology stack after the January 6 Capitol riots. But it was not until Amazon Web Services declined to host the Parler platform at a lower level of the stack (cloud hosting infrastructure) that it suddenly ceased to exist on the Internet as originally conceived. Parler, then the most downloaded application in Apple’s app store and a Big Tech competitor popular with conservatives, has yet to reach even 4 percent of the users it gained immediately before its deplatforming.
Information Manipulation. Information manipulation by Big Tech companies is also rampant. The litany of suspensions of ordinary Americans by Big Tech platforms for expressing right-leaning political views grows by the day. The Media Research Center (MRC) in October 2021 determined that Twitter and Facebook censor Republican Members of Congress at a rate of 53-to-1 compared to Democrats in Congress. The same month, The Wall Street Journal published leaked documents that revealed Facebook created and used two internal tools that suppressed right-wing content, media traffic, and reach on its site in the aftermath of Donald Trump’s 2016 presidential victory. Facebook’s research concluded that if the tools were removed, “very conservative” traffic would increase by tens of percentages. The platform was still using one of these tools in October 2021. Further, in 2020, an independent media company assessed that Google likely suppressed conservative media outlets during the lead up to the 2020 presidential election. In one example, the company’s analysis of third-party data suggests that Breitbart’s Google search visibility shrank by 99 percent during the 2020 presidential election cycle, compared to the same period in 2016.
This occurs as the Taliban, Iran’s supreme leader, North Korean officials, and Chinese Communist Party spokesmen spew propaganda, antisemitism, and genocidal rhetoric on the American-owned platforms of their choosing.
Such manipulation has direct political implications. In November 2020, another MRC study found that one in six Biden voters claimed they would have modified their vote had they been aware of information actively suppressed by tech companies—information unfavorable to President Biden and supportive of President Trump. In October of this year, a poll by McLaughlin & Associates indicated that 52 percent of Americans believe social media companies’ suppression of the Hunter Biden laptop story constituted election interference.
Culture of Free Speech. At a more abstract level, despite arguments focused on Big Tech as private companies within their rights to ban users, America’s culture of free speech is at issue. Distinctly American norms and values flow from a set of rights guaranteed by the Constitution and embedded in our culture, which includes free speech. Even the American Civil Liberties Union (ACLU)—not the friends of free expression they once were—is worried about this infringement by tech companies. For instance, the ACLU’s Senior Legislative Counsel stated in January that
it should concern everyone when companies like Facebook and Twitter wield the unchecked power to remove people from platforms that have become indispensable for the speech of billions.
Russian dissident Alexei Navalny, German chancellor Angela Merkel, and Mexico’s president Andrés Manuel López Obrador—none of whom are particularly sympathetic to American conservatives—also spoke out about the threats this abuse of consolidated power poses to freedom of expression writ large.
Further, this culture of free speech is threatened by the increasing symbiosis between Big Tech and the government. By its own admission, the Biden Administration is pressuring private tech companies to take down content under a broad and politically biased definition of “misinformation.”The Administration is also vocal about its plans to push Big Tech companies to ban users across all private platforms if banned by one company.
Americans are thus right to point to an erosion of trust in these companies’ ability to be neutral gatekeepers of information. For example, a summer 2020 Pew poll found that roughly three-quarters of adult Americans believe “social media sites intentionally censor political viewpoints that they find objectionable.”
A healthy body politic depends on the genuine exchange of ideas. It is 2021 and the verdict is in: Big Tech companies’ willingness to suppress specific points of view is corrosive to a free, American society.
Material Effects on Children
In addition to political impacts, internal Facebook research revealed in 2021 that Big Tech’s practices result in measurable, deleterious effects on young citizens. As companies compete for younger and younger portions of the market, the “race to the bottom” is in full swing. Newer companies are vying for footholds as others hemorrhage users in this demographic. For instance, The Wall Street Journal reporting on Facebook’s internal research suggests that in the United States, the number of teens who use Facebook every day fell by 19 percent in the past two years and was projected to fall an additional 45 percent by 2023. In a similar period, Facebook set a multi-year goal to create products specifically for preteens, considering them a “valuable but untapped audience.”
Later, in March 2021, Facebook revealed it intended to create an Instagram for children under 13 years old. (It already has a 6- to 12-year-old–focused Messenger app.) YouTube Kids invoked children as young as three years old in its rollout in 2015. Up and comers like TikTok, owned by a Beijing-based parent company, are deliberately courting younger markets. According to Statista, 62 percent of TikTok users in the United States are between 10 to 29 years old in 2021 and skewing younger. Last year, a Pew Research Center survey found that among 9-to-11-year-olds, 30 percent of their parents claim their child uses TikTok, 22 percent of parents estimate their kids use Snapchat, 11 percent Instagram, and 6 percent assess their children are on Facebook.
This is significant due to the toxic effects of these platforms on American youth. According to Facebook’s own research from 2019 to 2020, 32 percent of teenage girls said that “when they felt about their bodies, Instagram made them feel worse,” 6 percent of teen Instagram users who reported suicidal thoughts traced their emergence directly to Instagram, and 1 in 3 teen girls said Instagram made their body image issues worse.
These platforms also portend broader social impacts. According to The Wall Street Journal, teenage girls in the United States, the United Kingdom, Canada, and Australia are likely developing verbal and physical tics by watching influencers on TikTok who exhibit the same habits. In one case, Texas Children’s Hospital reported incidents of approximately 60 teens with these behaviors, compared to one to two cases pre-pandemic. Every reported incident of an uptick involved the use of TikTok, according to the Journal’s assessment. Author Abigail Shrier also documents social media’s influence on social contagions de rigueur, stating these sites offer an “endless supply of mentors” to fan the flames of gender dissatisfaction in teen girls.
Companies like Facebook are aware of these impacts yet continue to double down and expand efforts targeted at children. Google-owned YouTube easily absorbed a $170 million fine by the Federal Trade Commission (FTC) and the state of New York in 2019 for collecting data on children younger than 13 without parental permission. In fact, after research on Instagram’s toxic effects on young girls was made public, the head of Instagram declared publicly that “building ‘Instagram Kids’ is the right thing to do.”
Americans should be aware of and push back against these efforts to tear at the moral fabric of our society at any cost, starting in kindergarten. As Representative Bill Johnson (R–OH) said at a congressional hearing in March 2021: “Big Tech is essentially giving our kids a lit cigarette and hoping they stay addicted for life.”
Underpinning all of this is the fact that these tech companies continue to benefit from special liability protections from the federal government.
In attempts to hold Big Tech accountable for this behavior, Congress should avoid one pitfall in particular: calls to further suppress content based on politically expedient and expansive definitions of “harm” and “dis/misinformation.” Conservatives must hold these companies to account for their censorship, not encourage more censorship at the hands of Big Tech and the government.
The entity that defines what content is “harmful” or “dis/misinformation” is critical. Right now, Big Tech’s current definitions are often mutable, overly broad, and used as a catchall for “views we don’t like.” For example, the theory that COVID-19 could have originated from a Wuhan Institute of Virology leak in China was suppressed by an array of digital platforms yet is now considered part of acceptable discourse. The New York Post’s Hunter Biden-Ukraine laptop story was labeled Russian misinformation and actively censored by Facebook and Twitter before Hunter Biden went public and failed to deny his ownership of the laptop.
Pushing companies to crack down on free expression and expanding “oversight” by the government to police speech will continue to restrict free inquiry and accelerate the digital stratification of America already underway.
Holding Big Tech accountable should result in less censorship, not more.
Further, this focus on “harm” comes at the expense of bigger problems. As Facebook’s vice president of state public policy noted this year, the platform “[allows] people to share information about how to enter a country illegally or request information about how to be smuggled.” Human traffickers, foreign Islamist terrorists and their cheerleaders, and drug cartels also proliferate on the platform, in violation of these companies’ own policies. Policing such content should be prioritized over targeting of nebulous conceptions of “misinformation.”
The Way Forward
Any proposals to hold Big Tech companies accountable should empower citizens to redress the imbalance between the companies and their users.
At a minimum, Congress should:
Use the First Amendment as the standard from which all Section 230 reforms flow.
- American lawmakers have a duty to protect and defend the inalienable rights given to us by God and enshrined in our Constitution by the Founders. Rights—to include free speech—that specific tech companies, in conjunction with the government, are actively and deliberately eroding. The argument that private companies do not bear free speech responsibilities ignores current overt collaboration and communication between government and Big Tech companies that work together to stifle free expression. While private companies may have the right to exclude content from their platforms, they should not be immune from civil liability for altering content posted by others. This includes adding truth banners or labeling, suppressing, or removing content that is entirely legal but which they consider “objectionable.”
Any reforms to Section 230 should proceed with this in mind.
Reform Section 230 in the following ways.
- Strip immunity if tech companies censor based on political views. When tech companies act as publishers and restrict content based on political opinion or association, these actions are inconsistent with the liability protections codified within Section 230 or Congress’ intent when it passed the legislation. Big Tech companies should not receive liability protection against lawsuits in such instances.
- Strike the phrase “otherwise objectionable” from Section 230(c)(2)(a).
- Build in a sunset clause to address appropriate Section 230 reform at least every 7 years.
Hold tech companies accountable for inconsistent application of policies via the judicial system.
- Examine discrepancies between advertised mission statements, terms of service, and policies and implementation as possible cases of fraud and/or “breach of contract.” If companies claim they are for free speech and are acting in a politically neutral manner, then they should prove it. They should be held accountable if they, in fact, are not.
- Companies should implement a user-friendly appeals process to provide prompt and meaningful recourse for users who are wrongfully targeted for their speech.
- Companies should report content-moderation methodology and decisions to the FTC, with a public availability component. For example, some tech companies post quarterly, public reports on the enforcement of their community standards. These should be more granular and cover content-moderation decisions, practices, and behaviors related to free expression (e.g., wrongful or erroneous take-downs, such as Amazon’s prohibition of ads for Heritage scholar Mike Gonzalez’s book BLM: The Making of a New Marxist Revolution). Other major tech companies should follow suit. As the Biden White House admits its collusion with Big Tech to remove content, this push is a good starting point to incentivize transparency.
- With regard to enforcement, the FTC and like government agencies exist, in part, to enforce the law. Policymakers should be able to use these organizations to address and deter political discrimination without expanding and weaponizing federal government power.
- Remove liability protection when companies censor based on political views. Allow American citizens genuine recourse in court to prove their case against Big Tech censorship.
- Incentivize algorithmic transparency. Too often, Big Tech companies hide behind their algorithms to justify censorship. Yet programmers and other personnel dictate the design and implementation of those algorithms. Users have a right to information on this main ingredient of the product they use. Transparency reports should include how these companies’ algorithms operate and affect users, including details on the impact of any ad-hoc changes within reporting periods.
- Account for the scale and reach of tech companies. Big Tech companies, aided by Section 230’s liability shield, flourished in the late 1990s and early 2000s. They were able to cement first-mover advantage in a technical sense—accruing data and refining algorithms, etc., over decades to improve the products Americans use today. Changes to Section 230 should not cement these technical advantages (which, in some instances, have compounded after their initial accrual with government help) and entrench a handful of companies within the market at the expense of competitors. Instead, reforms should be made with the understanding that changes to legislation will impact new entrants or smaller companies differently than Big Tech companies.
- Articulate the expectation of accountability, transparency, and fair treatment for all users. This can look like:
Institute additional mechanisms for user control.
- U.S. tech sector leaders must make firm commitments to implementing efforts like third-party algorithm use, and not just hint at them in front of Congress.
Look outside Washington, DC.
- While these adjustments are well-suited to address this moment through a conservative framework, focused Section 230 reform is not a silver bullet to righting Big Tech’s wrongs. A full spectrum of action is necessary. A simultaneous, multipronged approach to securing freedom of expression in the digital world includes promoting the principles of federalism through constitutional state legislative action, efforts by technologists to build platforms where freedom of expression is protected, as well as revivifying civil society efforts to promote transparency within these companies.
The government gave tech companies an inch with Section 230, and they took a mile. It is past time to rein in their ability to squelch valuable—indeed essential—public discourse by clarifying the law.
The Heritage Foundation is the most broadly supported think tank in the United States. During 2020, it had hundreds of thousands of individual, foundation, and corporate supporters representing every state in the U.S. Its 2020 operating income came from the following sources:
Program revenue and other income 14%
The top five corporate givers provided The Heritage Foundation with 1% of its 2020 income. The Heritage Foundation’s books are audited annually by the national accounting firm of RSM US, LLP.