At the end of May, President Trump
issued an Executive
Order demanding action against social media sites for
“censoring” conservatives. His Department of Justice made
a more
specific proposal in mid-June. Clearly coordinating
with the White House, Sen. Josh Hawley introduced
a bill that same morning, making clear that his “Limiting
Section 230 Immunity to Good Samaritans Act” is
essentially the administration’s bill — as called for in
the May Executive Order. The administration is expected to make its
next move next week: having NTIA (an executive agency controlled by
Trump loyalists and advised by a former law professor intent on
cracking down on tech companies) ask the FCC to make rules
reinterpreting Section 230 to do essentially the same thing as the
Hawley bill. These two approaches, both stemming from the Executive
Order, are unconstitutional for essentially the same reasons: they
would put a gun to the head of the largest social media websites,
forcing them to give up editorial control over their services if they
want to stay in business.
The First Amendment would not allow
Congress to directly require websites to be politically
“neutral” or “fair”: the Supreme Court has
recognized that the First Amendment protects the editorial discretion
of websites no less than newspapers. Both have the same right to
decide what content they want to carry; whether that content is
created by third parties is immaterial. Hawley’s bill attempts
to lawyer over the constitutional problem, using an intentionally
convoluted process to conceal the bill’s coercive nature and to
present himself as a champion of “free speech,” while
actually proposing to empower the government to censor online content
as never before.
Instead of directly meddling with
how websites moderate content, Hawley’s bill relies on two
legal sleights of hand. The first involves Section 230 of the
Communications Decency Act of 1996. That law made today’s
Internet possible — not only social media but all websites and
services that host user content — by protecting them from most
civil liability (and state criminal prosecution) for content created
by third parties. Given the scale of user-generated content —
with every comment, post, photo and video potentially resulting in a
lawsuit — websites simply could not function if Section 230 did
not immunize them not just from ultimate liability but from the
litigation grindstone itself. Hawley knows that all sites that host
user content depend on Section 230, so he’s carefully crafted a
bill that turns that dependence against them — to do something
the First Amendment clearly forbids: to force them to cede editorial
control over their services. (Here’s a redline showing how Hawley’s bill would amend Section 230.)
Second, Hawley claims that his bill
“protects consumers” by holding companies to their
promises. In reality, it defines “good faith” so broadly
that “edge providers” would face a constant threat of
being sued under consumer protection and contract laws for how they
exercise their editorial discretion over user content. Given the
fines involved ($5,000/user plus attorneys’ fees), a single
court decision could bankrupt even the largest tech company.
No one should have any illusion
about what Hawley’s bill really does: use state power to
advance a political agenda. The bill’s complicated structure
merely masks the elaborate ways it violates the First Amendment.
Conditioning 230 immunity on opening yourself up to legal liability
under consumer protection law is a Rube-Goldberg-esque legal
contraption intended to do what the First Amendment clearly forbids:
forcing websites to host user-generated content they find
objectionable.
How the Hawley Bill
Works
Section 230(c)(1) says: “No
provider or user of an interactive computer service shall be treated
as the publisher or speaker of any information provided by another
information content provider.” These have been called the The
Twenty-Six Words That Created the Internet. When
websites and services are sued for third party content they host,
Section 230 allows them to cheaply get lawsuits against them thrown
out with a motion to dismiss. Consequently, lawsuits are far rarer
than they would be in a world without 230. Section 230(c)(1) ensures
that those who create content are the ones to be sued. Courts resolve
nearly all 230 cases under this provision.
Republicans have insisted angrily
that all of Section 230 was intended to depend on a showing of
good faith, including political neutrality; however, the plain text
of the statute is clear. Only Subsection 230(c)(2)(A) requires such a
showing — and the statute’s operative language doesn’t
mention neutrality. As Justice Neil Gorsuch recently declared,
“When the express terms of a statute give us one answer and
extratextual considerations suggest another, it’s no contest.
Only the written word is the law, and all persons are entitled to its
benefit.” Bostock v. Clayton County, 590 U.S. ___
(2020). By proposing to amend Section 230(c)(1) to require both good
faith and neutrality, Trump’s DOJ and Hawley both
concede that the President’s Executive Order and other
Republican clamoring for immediate legal action are simply wrong
about the current state of the law.
The real aim of Hawley’s bill
is to force the largest social media services to change how they
treat content that serves the “MAGA” political agenda —
e.g., not labeling Trump’s tweets, allowing far-right
provocateurs to engage in bannable conduct, treating Diamond and Silk
or Gateway Pundit as the journalistic equivalents of The New York
Times. The bill is almost perfectly tailored to do just that
while avoiding damage to smaller, alternative social networks favored
by conservative activists for their “anything goes”
approach to content moderation.
Hawley’s bill applies only to
“edge providers”: websites or services with 30+ million
annual unique users, or more than 300 million unique global users, in
the past year, and more than $1.5 billion in global revenue. To
maintain 230(c)(1) protections, they would have to attest to “good
faith” — essentially, political neutrality — in
their content moderation practices. Thus, an edge provider has to
choose between two litigation risks: If it “voluntarily”
exposes itself to suit for the “fairness” of its content
moderation, it cedes editorial control to judges and regulators. If
it surrenders Section 230 protections, it risks being sued for
anything its users say — which may simply make it impossible
for them to operate.
Trump’s Executive Order asks
the Federal Communications Commission to collapse Section 230’s
three distinct immunities into a single immunity dependent on “good
faith” — and then define that term broadly to include
neutrality and potentially much more. The Hawley bill does roughly
the same thing by requiring large “edge providers” to
promise “good faith.” Both would change the dynamics of
litigation completely: A plaintiff with a facially plausible
complaint would (1) prevail on a motion to dismiss, (2) get
court-ordered discovery of internal documents and depositions of
employees to assess “good faith” (however that term is
expanded), and (3) force the company to litigate all the way through
a motion for summary judgment. Whether or not the plaintiff
ultimately wins, this pre-trial phase of litigation is where the
defendant will incur the vast majority of their legal costs —
and where plaintiffs force settlements. Multiply those costs of
litigation, and settlement, times the millions or billions of pieces
of content posted to social media sites every day and you get “death
by ten thousand duck-bites.” Fair v. Roommates, 521 F.3d
1157, 1174 (9th Cir. 2008). That’s why Judge Alex Kozinski (a
longtime conservative champion once short-listed for the Supreme
Court) declared: “section 230 must be interpreted to protect
websites not merely from ultimate liability, but from having to fight
costly and protracted legal battles.” Id.
Having to prove good faith to
resolve litigation would kill most social media websites, which exist
to host content by others. Ironically, it’s possible that the
best established social media sites with the biggest legal
departments might cope; they might even be grateful that Hawley’s
bill had made it impossible for new competitors to get off the
ground. At the same time, if (c)(1) is no longer an immunity from
suit but merely a defense raised only after great expense, websites
across the Internet would simply turn off their comments sections.
Today, Section 230 doesn’t
define “good faith.” Courts assessing eligibility for the
230(c)(2)(A) immunity have defined the term narrowly. See e.g.,
BFS Fin. v. My Triggers Co., No. 09CV-14836 (Franklin Cnty.
Ct. Com. Pl. Aug. 31, 2011) (allowing antitrust claims); Smith v.
Trusted Universal Standards in Elec. Transactions, 2011 WL
900096, at *25–26 (D.N.J. Mar. 15, 2011). Hawley’s bill
would add a five-factor definition of “good faith” in a
new Subsection 230(c)(3). These factors would give plaintiffs ample
room to declare that an edge provider had been politically biased
against them. Inevitably, courts would have to analyze the nature of
third-party content, comparing content that had been removed with
content that had not in order to judge overall patterns.
To maintain 230 protections, an edge
provider must also agree to pay up to $5,000 damages to users if it
is found to have breached its (compelled) promises of “neutrality.”
Three hundred million users times $5,000 is $1.5 trillion dollars,
exceeding the entire market cap of Google. The bill also adds
attorneys fees, threatening to create a cottage industry of
litigation against edge providers. The mere threat of such massive
fines will fundamentally change how websites operate —
precisely Hawley’s goal.
Perhaps most important is what the
bill doesn’t say: unlike Trump’s Order, Hawley’s
bill doesn’t directly call on the FTC or state AGs to sue
websites for bias. But make no mistake; his bill would weaponize
federal and state consumer protection laws to allow politicians to
coerce social media into favoring their side of the culture wars. The
FTC might hesitate to bring such suits, because of all the
constitutional problems discussed below, but multiple Republican
attorneys general have already made political hay out of
grandstanding
against “liberal San Francisco tech giants.”
They would surely use Hawley’s bill to harass edge providers,
raise money for their campaigns, and run for governor — or
Senate.
A New Fairness
Doctrine — with Even Greater First Amendment Problems
The Original Fairness Doctrine
required broadcasters (1) to “adequately cover issues of public
importance” and (2) to ensure that "the various positions
taken by responsible groups" were aired, thus mandating the
availability of airtime to those seeking to voice an alternative
opinion. President Reagan’s FCC abolished these requirements in
1987. When Reagan vetoed Democratic legislation to restore them, he
noted that “the FCC found that the doctrine in fact inhibits
broadcasters from presenting controversial issues of public
importance, and thus defeats its own purpose.”
The Republican Party has steadfastly
opposed the Fairness Doctrine for decades. The 2016 Republican
platform (re-adopted verbatim for 2020) states: “We likewise
call for an end to the so-called Fairness Doctrine, and support
free-market approaches to free speech unregulated by government.”
Yet now, Hawley and Trump propose a version of the Fairness Doctrine
for the Internet that would be more vague, intrusive, and arbitrary
than the original.
In Miami Herald Publishing Co. v.
Tornillo, 418 U.S. 241 (1974), the Supreme Court struck down a
1913 state law imposing a version of the Fairness Doctrine on
newspapers that required them to grant a “right of reply”
to candidates for public office criticized in their pages. The Court
acknowledged that there had been a technological “revolution”
since the enactment of the First Amendment. The arguments made then
about newspapers, as summarized by the Court, are essentially the
same arguments conservatives make about digital media:
The result of
these vast changes has been to place in a few hands the power to
inform the American people and shape public opinion…. The
abuses of bias and manipulative reportage are, likewise, said to be
the result of the vast accumulations of unreviewable power in the
modern media empires. The First Amendment interest of the public in
being informed is said to be in peril because the ‘marketplace
of ideas’ is today a monopoly controlled by the owners of the
market.
Id. at 250. And yet, the
court struck down the law as unconstitutional because:
a compulsion to
publish that which “‘reason’ tells them should not be
published" is unconstitutional. A responsible press is an
undoubtedly desirable goal, but press responsibility is not mandated
by the Constitution and like many other virtues it cannot be
legislated.
Id at 256.
“Government-enforced right of access inescapably ‘dampens
the vigor and limits the variety of public debate.’" Id.
at 257. Critically, the Court rejected the intrusion into the
editorial discretion “[e]ven if a newspaper would face no
additional costs to comply,” because:
A newspaper is
more than a passive receptacle or conduit for news, comment, and
advertising. The choice of material to go into a newspaper, and
the decisions made as to limitations on the size and content of the
paper, and treatment of public issues and public officials —
whether fair or unfair — constitute the exercise of editorial
control and judgment.
418 U.S. at 258. The Trump/Hawley
Fairness Doctrine would impose the very same intrusion upon editorial
judgments of edge providers. In addition, determining whether a
website has operated “fairly” would be “void for
vagueness since no editor could know exactly what words would call
the statute into operation.” Id. at 247.
The Supreme Court upheld the
Fairness Doctrine for broadcasters in Red Lion Broadcasting Co. v.
FCC, 395 U.S. 367 (1969), but only because the Court denied
broadcasters full First Amendment protection: “Although
broadcasting is clearly a medium affected by a First Amendment
interest, differences in the characteristics of new media justify
differences in the First Amendment standards.” The same
arguments have been made about the Internet, and the Supreme Court
explicitly rejected them.
When the Court struck down Congress’
first attempt to regulate the Internet, the Communications Decency
Act (everything except Section 230), it held: “our cases
provide no basis for qualifying the level of First Amendment scrutiny
that should be applied to this medium.” Reno v. American
Civil Liberties Union, 521 U.S. 844, 870 (1997). The Court has
since repeatedly reaffirmed this holding. While striking down a state
law restricting the purchase of violent video games, Justice Scalia
declared: "the basic principles of freedom of speech and the
press, like the First Amendment’s command, do not vary when a new and
different medium for communication appears.” Brown v.
Entertainment Merchants Assn., 564 U.S. 786, 790 (2011). In
short, Red Lion represented an exception, and even that
exception may not survive much longer.
Social Media Aren’t
Public Fora, So the First Amendment Protects Them
The President’s Executive
Order attempts to sidestep the Supreme Court’s consistent
protection of digital speech by claiming that social media are
effectively “public fora” and thus that the First
Amendment limits, rather than protects, their editorial discretion —
as if they were extensions of the government: “It is the policy
of the United States that large online platforms, such as Twitter and
Facebook, as the critical means of promoting the free flow of speech
and ideas today, should not restrict protected speech.” The
Order also cites the Supreme Court’s decision that shopping
malls were public fora under California’s constitution
in Pruneyard Shopping Center v. Robins, 447 U.S. 74, 85-89
(1980).
But Justice Kavanaugh, leading the
five conservatives, explicitly rejected such arguments last year:
“merely hosting speech by others is not a traditional,
exclusive public function and does not alone transform private
entities into state actors subject to First Amendment constraints.”
Manhattan Community Access Corp. v. Halleck, 139 S. Ct. 1921,
1930 (2019). Pruneyard simply doesn’t apply to social
media.
Trump’s Order cites the
Supreme Court’s recent decision in Packingham v. North
Carolina, 137 S. Ct. 1730, 1737 (2017) (social media “can
provide perhaps the most powerful mechanisms available to a private
citizen to make his or her voice heard”), but omits the
critical legal detail: it involved a state law restricting the
Internet use of convicted sex offenders. Thus Packingham
changed nothing: the First Amendment still fully protects, rather
than limits, the editorial discretion of website operators under
Miami Herald and Reno.
Hawley’s Bill
Imposes an Unconstitutional Condition
Hawley’s bill turns on one
underlying legal claim more than any other: that Section 230 is a
special privilege granted only to large websites, and withholding it
does not violate the First Amendment. The factual claim is false: the
law applies equally to all websites, protecting newspapers,
NationalReview.com, FoxNews.com and every local broadcaster from
liability for user comments posted on their website in exactly the
same way it protects social media websites for user content. The
legal claim is also wrong.
The Supreme Court has clearly barred
the government from forcing the surrender of First Amendment rights
in order to qualify for a benefit or legal status. In Agency for
Int’l Dev. v. All. for Open Soc’y Int’l, Inc., 570 U.S. 205
(2013), the Court said that the government couldn’t condition
the receipt of AIDS-related funding on the recipients’ adoption
of a policy opposing prostitution (a form of compelled speech). Much
earlier, in Speiser v. Randall, 357 U.S. 513, 518 (1958), the
Court made it clear that denying a tax exemption to claimants who
engage in certain forms of speech effectively penalizes them for that
speech — essentially fining them for exercising their First
Amendment rights.
Using Section 230 to coerce social
media companies into surrendering their First Amendment rights is no
different. Consider how clearly the same kind of coercion would
violate the First Amendment in other contexts. Pending legislation
would immunize businesses that re-open during the pandemic from
liability for those who might be infected by COVID-19 on their
premises. Suppose the bill included a provision requiring such
businesses to be politically neutral in any signage displayed on
their stores — such that, if a business put up, or allowed a
Black Lives Matter sign, they would have to allow a “right of
reply” in the form of a sign from “the other side.”
The constitutional problem would be obvious and in no way ameliorated
by the “voluntary” nature of the immunity program.
Social Media
Companies Can’t Be Forced to Risk Being Associated with Content
They Find Objectionable
The case against unconstitutional
conditions and public forum status is even clearer for websites than
it would be for retailers or shopping malls, for two reasons. First,
social media companies are in the speech business, unlike
businesses whose storefronts might incidentally post their own speech
or host the speech of others. Reno makes clear that websites
enjoy the same First Amendment right as newspapers, and “[t]he
choice of material to go into a newspaper, and the decisions made as
to limitations on the size and content of the paper, and treatment of
public issues and public officials — whether fair or unfair —
constitute the exercise of editorial control and judgment.”
Miami Herald, 418 U.S. at 258.
Second, Pruneyard emphasized
that shopping malls could “expressly disavow any connection
with the message by simply posting signs in the area where the
speakers or handbillers stand.” But users will naturally assume
speech carried by a social network reflects their decision to carry
it — just as Twitter and Facebook have been attacked for not
removing President Trump’s tweets or banning him from their
services.
Disclaimers may actually be less
effective online. Consider the three labels Twitter has applied to
President Trump’s tweets (the first two of which provoked the
issuance of his Executive Order).
The first
example not only fails to clearly “disavow any
connection with the message,” it is also ambiguous: it could be
interpreted to mean there really is some problem with mail-in
ballots.
Similarly, Twitter applied a “(!)
Manipulated Media” label to Trump’s tweet
of a video purporting to show CNN’s anti-Trump bias. Twitter’s
label is once again ambiguous: since Trump’s video claims that
CNN had manipulated the original footage, the “manipulated
media” claim could be interpreted to refer to either Trump’s
video or CNN’s. Although the label links to an “event”
page explaining the controversy, the warning only
works if users actually click through. It’s far from clear to
many users that the label is actually a link that will take them to a
page with more information.
Finally, when Trump tweeted,
in reference to Black Lives Matter protests, “when the looting
starts, the shooting starts,” Twitter did not merely add a
label below the tweet. Instead, it hid the tweet behind a disclaimer.
Clicking on “view” allows the user to view the original
tweet:
And yet Twitter has still been
lambasted for not taking the tweet down completely, a decision
interpreted by some as an acceptance of the validity of such an
extreme position.
Further, disclaimers risk creating
increased liability; indeed, they may trigger lawsuits from scorned
politicians. For example, labeling (and hiding) Trump’s tweets
provoked issuance of the Executive Order. In the end, the only truly
effective way for Twitter to disavow Trump’s comments would be
to ban him from their platform — precisely what the Hawley bill
aims to deter.
In this sense, the Trump/Hawley
version of the Fairness Doctrine is hugely more intrusive than
the right of reply in the original Fairness Doctrine; it puts edge
providers in the doubly unconstitutional position of (a) hosting
content they do not want to host and (b) being afraid even to label
it as content they find objectionable.
Why the Hawley Bill’s
Good Faith Requirement Violates the First Amendment
To maintain 230 immunity, edge
providers would be required to promise to moderate content in “good
faith” — which the Hawley bill defines very loosely as
“honest belief and purpose…fair dealing standards, and…[no]
fraudulent intent” — in other words, political neutrality
(and more). The bill adds this to Section 230’s list of
exceptions: “Nothing in this section shall be construed to
impair or limit any claim for breach of contract, promissory
estoppel, or breach of a duty of good faith.’’ Thus, an
edge provider’s compelled “promises” could be
enforced by the Federal Trade Commission, state AGs, or private
plaintiffs under various federal and state consumer protection laws
and common law contract theories. These enforcement mechanisms raise
slightly different legal issues, but they all violate the First
Amendment in essentially the same way: state action interfering with
edge providers’ exercise of editorial discretion.
Consumer Protection
Law Can’t Police “Fairness” Claims
Republicans used to oppose
weaponizing consumer protection laws against media companies. In
2004, MoveOn.org and Common Cause asked the FTC to proscribe Fox
News’ use of the slogan “Fair and Balanced” as a
deceptive trade practice. Republican Chairman Tim Muris responded
pithly: “I am not aware of any instance in which the [FTC] has
investigated the slogan of a news organization. There is no way to
evaluate this petition without evaluating the content of the news at
issue. That is a task the First Amendment leaves to the American
people, not a government agency.”
Similarly, the Hawley bill would
necessarily embroil the FTC, state AGs, and judges in “evaluating
the content … at issue.” Media companies aren’t
exempt from consumer protection or antitrust laws, but the First
Amendment makes suing them for how they exercise their editorial
discretion extremely difficult, if not impossible — which is
why the FTC has never attempted to police marketing claims about
editorial practices the way it polices marketing claims generally.
As Chairman Muris noted, general
statements about “fairness” or “neutrality”
simply are not verifiable. This is why the Ninth Circuit recently
dismissed Prager University’s deceptive marketing claims
against YouTube. Despite having over 2.52 million subscribers and
more than a billion views, this right-wing producer
of “5-minute videos on things ranging from history and
economics to science and happiness,” sued
YouTube for “unlawfully censoring its educational videos and
discriminating against its right to freedom of speech.”
Specifically, Dennis Prager alleged
that roughly a sixth of the site’s videos had been flagged for
YouTube’s Restricted
Mode, an opt-in feature that allows parents, schools
and libraries to restrict access to potentially sensitive (and is
turned on by fewer than 1.5% of YouTube users). The Ninth Circuit
ruled:
YouTube’s
braggadocio about its commitment to free speech constitutes opinions
that are not subject to the Lanham Act. Lofty but vague
statements like "everyone deserves to have a voice, and that the
world is a better place when we listen, share and build community
through our stories" or that YouTube believes that "people
should be able to speak freely, share opinions, foster open dialogue,
and that creative freedom leads to new voices, formats and
possibilities" are classic, non-actionable opinions or puffery.
See Newcal Indus., Inc. v. Ikon Office Sol., 513 F.3d 1038, 1053 (9th
Cir. 2008). Similarly, YouTube’s statements that the platform will
"help [one] grow," "discover what works best,"
and "giv[e] [one] tools, insights and best practices" for
using YouTube’s products are impervious to being
"quantifiable," and thus are non-actionable "puffery."
Id. The district court correctly dismissed the Lanham Act claim.
Prager Univ. v. Google LLC,
951 F.3d 991, 1000 (9th Cir. 2020). Websites can’t be
sued today for making statements that may sound like offering
neutrality — contrary to Republican claims that they should be,
and Trump’s call for such lawsuits in this Executive Order. The
Hawley bill implicitly concedes this point.
But simply forcing edge providers to
be more specific in their claims about neutrality will not
overcome the ultimate constitutional problem. Puffery includes
“claims [which] are either vague or highly subjective.”
Sterling Drug, Inc. v. FTC, 741 F.2d 1146, 1150 (9th Cir.
1984) (emphasis added). It would be difficult to imagine a more
subjective marketing claim than one about “good faith,”
“neutrality” or “fairness.” Ultimately, the
reason consumer protection law does not attempt to police marketing
claims about neutrality is not their lack of specificity but their
subjectivity.
In theory, the FTC might be able to
base a deception case on certain very clear, objective claims
about editorial practices; that category of deception, however, would
be narrow — the use of human moderators to evaluate particular
pieces of content or to decide which topics are “trending,”
or the application of community standards to elected officials, for
example. These deception cases would do little to address the
complaints of conservatives, and even such narrow complaints might be
unconstitutional.
Consumer Protection
Law Can’t Police Non-Commercial Speech
The FTC can police marketing claims
for being misleading to the extent they “propose a commercial
transaction.” Central Hudson Gas & Elec. Corp. v. Public
Service Comm’n of New York, 447 U.S. 557,561 (1980);
Virginia State Bd. of Pharmacy v. Virginia Citizens Consumer
Council, Inc., 425 U.S. 748, 762 (1976). Community standards
documents do much more than that: they are essentially statements of
values, comparable to Christian retailer Hobby Lobby’s
statement
that the company is committed to “[h]onoring the Lord in all we
do by operating the company in a manner consistent with Biblical
principles.”
Such statements are non-commercial
speech, which is fully protected by the First Amendment under strict
scrutiny even when it is misleading. United States v.
Alvarez, 567 U.S. 709 (2012). To overcome strict scrutiny, the
government must show that the bill is (1) necessary to address a
compelling government interest (2) to which the law is narrowly
tailored, and (3) that the government uses the least restrictive
means possible to address that interest. Reed v. Town of Gilbert,
576 U.S. 155, 163, 171 (2015). In Miami Herald, the court
noted that Florida’s interest in “ensuring free and fair
elections” was a “concededly important interest,”
but had to yield to the “unexceptionable, but nonetheless
timeless, sentiment that liberty of the press is in peril as soon as
the government tries to compel what is to go into a newspaper."
418 U.S. at 260. The bill also fails on the second two prongs of
strict scrutiny,
If the Hawley bill passes, the Trump
Administration will undoubtedly argue that edge providers’
community standards are ads for their services. But when speech has
commercial aspects that are “inextricably intertwined”
with other fully protected speech, that speech is generally fully
protected. Riley v. Nat’l Fed’n of the Blind of N.C.,
Inc., 487 U.S. 781, 783 (1988). For example, corporate statements
endorsing Black Lives Matter receive First Amendment protection even
when embedded in marketing claims.
Courts are generally reluctant to
label content as commercial speech because that denies the speech
full First Amendment protection. Although community standards and
terms of service may “refer[] to a specific product,”
they in no way resemble traditional advertising — two of the
factors courts assess in drawing the line between commercial and
noncommercial speech. Bolger v. Youngs Drug Prods. Corp., 463
U.S. 60, 66-67 (1983). The third factor, the profit motive —
which Hawley harps on in his public statements — is not
dispositive: “If a newspaper’s profit motive were
determinative, all aspects of its operations—from the
selection of news stories to the choice of editorial position—would be subject to regulation if it could
be established that they were conducted with a view toward increased
sales.” Pittsburgh Press Co. v. Pittsburgh Comm’n on Human
Relations, 413 U.S. 376, 385 (1973) (emphasis added).
Pittsburgh Press makes clear
that statements about the way publishers exercise their editorial
discretion are fundamentally different from statements about the
health benefits of drug products, for example.
Even if a court decided to treat
community standards as commercial speech, the government would still
face an uphill battle. “The party seeking to uphold a
restriction on commercial speech carries the burden of justifying
it,” Bolger, 463 U.S. at 71, n. 20, and “must
demonstrate that the harms it recites are real, and that its
restriction will in fact alleviate them to a material degree.”
Edenfield v. Fane, 507 U.S. 763, 771 (1993). Because the
government’s interest in regulating commercial speech lies in
its misleading or false nature, it would have to show that statements
about a website’s editorial practices are misleading. General
claims about “fairness,” however, are simply not
verifiable.
Why the Government
Can’t Compel Disclosures about Editorial Policies
Compelling edge providers to change
what they say about their community standards violates the First
Amendment even apart from enforcement of such claims. As a condition
for maintaining 230 protection, the Hawley bill requires edge
providers to (1) “describe any policies … relating to
restricting access to or availability of [user-generated] material”
and (2) “promise that the edge provider shall … design
and operate the provided service in good faith.” The first
requirement seems hands-off: it does not directly dictate what an
edge provider’s terms of service must say. But this is simply a
trick of clever drafting: this requirement does not need to be
specific, because the second requirement (“good faith”)
will, in practice, govern both. The two inquiries will collapse into
one, allowing complaints about both the fairness of content
moderation practices as compared to community standards, and the
adequacy of those standards.
As a result, companies would (1)
make their community standards as opaque or unspecific as possible
and (2) minimize transparency about content moderation generally
(e.g., avoiding public statements or reporting on content
removals). But relying on “good faith” does not solve the
compelled speech First Amendment problem.
Suppose that, instead of suing to
enforce Fox News’ “Fair and Balanced” slogan in
2004, Congressional Democrats had proposed a bill like Hawley’s:
just replace “community standards” with “editorial
standards” and apply the bill to cable programming networks
over a certain size. It would be obvious that the government cannot
compel traditional media companies to “describe any policies …
relating to [selection] of [programming] material.”
By contrast, the government may
(and does) compel food manufacturers to disclose ingredient lists and
nutritional information. The First Amendment permits such mandates
because they apply to statements of objective fact, not the
disclosure of opinions. This is why the seemingly simple age-based
ratings systems for video games and movies have evolved as purely
private undertakings. Behind each label is an editorial judgment, an
opinion, about how to apply rating criteria. The government
can compel neither the rating system overall, nor specific
disclosures about the contents of specific films, nor disclosure of
the rating methodology. By the same token, it cannot compel websites
to disclose their editorial methodologies, whether implemented by
humans or algorithms. Brown, 131 S. Ct. at 2740.
The Hawley Bill Is
Designed to Chill the Exercise of Editorial Discretion
The Hawley bill proposes four
criteria for assessing a website’s “good faith.”
The first two concern “selective enforcement,” whether by
humans or algorithms. But what purports to be a regulation only of
marketing claims would actually, inevitably embroil regulators and/or
judges in evaluating the editorial discretion of edge providers —
conduct that would clearly qualify for the full protection of the
First Amendment as non-commercial speech under Miami Herald.
Twitter’s alleged political bias in applying its community
standards is no more actionable under consumer protection law than
would be Fox News’ political bias in its editorial policies.
The third criterion — “the
intentional failure to honor a public or private promise made by, or
on behalf of, the provider” — appears to preserve
consumer protection claims, but its aim is significantly broader. In
Barnes v. Yahoo!, Inc., 565 F.3d 560 (9th Cir. 2009), the
court allowed the plaintiff’s suit against Yahoo! to proceed.
Barnes sued the company for failing to stop her ex-boyfriend from
posting revenge porn. The court ruled that the company had
essentially waived its Section 230 immunity when its Director of
Communications promised the plaintiff she would “personally
walk the statements over to the division responsible for stopping
unauthorized profiles and they would take care of it.”
This promissory estoppel theory was
limited to the particular facts of that case: a clear promise made
directly to a specific user. The Hawley bill’s “public
or private promise” language could be read to allow plaintiffs
to set aside Section 230 immunity and sue edge providers for far more
general statements about content moderation practices that would
never qualify for promissory estoppel. By holding companies to every
past statement, the Hawley bill aims to stop companies from changing
their content moderation policies over time as new challenges emerge
— a critical dimension of any company’s editorial
discretion.
The fourth criterion — “any
other intentional action taken by the provider without an honest
belief and purpose, without observing fair dealing standards, or with
fraudulent intent” — seems tailor-made for a law school
exam on the “void for vagueness” standard. In particular,
it is considerably more expansive than the narrow standard the
Supreme Court set forth in Central Hudson Gas Elec. v. Public
Serv. Comm’n, 447 U.S. 557 (1980), for regulating commercial
speech: “there can be no constitutional objection to the
suppression of commercial messages that do not accurately inform the
public about lawful activity.” In other words, the Court allows
the regulation of commercial speech only because of its effects,
not its intent. Applying a subjective, rather than an objective
standard, would make litigation significantly easier. Thus, this
criterion would not be constitutional even if it were applied solely
to commercial speech. But as we have already seen with the Fox News
example, there would be no way to apply this standard “without
evaluating the content … at issue,” as FTC Chairman
Muris put it.
The Bill
Unconstitutionally Targets Specific Websites
The bill applies to “edge
providers,” defined as providers of a website, mobile
application or web application with more than $1.5 billion in global
revenue and more than 30 million U.S. users or more than 300 million
global users, that have accessed the site by any means in the past
year. This tailors the bill to apply to just a handful of services:
Google (Alphabet), Apple, Facebook (including Instagram and Whatsapp)
and Amazon (the so-called “GAFA”) as well as Twitter,
eBay, Microsoft, Apple, and TikTok (because the revenue threshold is
global). Reddit, Flickr, and Etsy would meet the user thresholds but
not the revenue thresholds. Wikipedia wouldn’t be covered
because it’s a non-profit.
What may at first seem like a
sensible way to focus the effect of the bill actually creates a host
of problems. First, it’s possible that, despite posing an
existential threat to “Big Tech” companies, Hawley’s
bill could actually protect them from competition. By penalizing
smaller market entrants for getting too big, Hawley’s bill
creates an incentive for small players to get bought-out by their
“big tech” counterparts before crossing Hawley’s
size threshold — big companies better equipped to handle the
legal risks Hawley’s bill would create.
The bill’s scope raises three
distinct constitutional problems. First, singling out a small group
of websites provides further reason for applying stricter scrutiny.
“Minnesota’s ink and paper tax violates the First Amendment not
only because it singles out the press, but also because it targets a
small group of newspapers…. And when the exemption selects
such a narrowly defined group to bear the full burden of the tax, the
tax begins to resemble more a penalty for a few of the largest
newspapers than an attempt to favor struggling smaller enterprises.”
Minneapolis Star, 460 U.S. at 591-92. Applying taxes only to
large newspapers “poses a particular danger of abuse by the
State.” Arkansas Writers’ Project, Inc. v. Ragland, 481
U.S. 221 (1987).
Hawley’s bill poses a “danger
of abuse” by focusing on only the largest social networks —
all of the ones conservatives complain about being biased
against them — while excluding sites with a laissez-faire
approach to content moderation, where extremist right-wing content
has been allowed to flourish, such as Reddit.
The relatively high revenue threshold excludes Reddit as well as
other popular social media sites like Yelp (business reviews), IMDB
(movie reviews), Fandom (a hosting platform), and Pinterest. The user
threshold also excludes smaller social networks that have become
gathering places for the Alt Right, like Gab (1.8
million monthly users users) and Minds (1.25
million users total).
The bill might apply to
websites for traditional media, but even this is difficult to
predict. Websites the largest newspapers and cable channels all meet
the monthly user threshold, but won’t qualify for the revenue
threshold if separate corporate digital divisions are treated as the
“edge providers” covered by the bill. In theory, it might
be possible to “pierce the corporate veil” to argue that
the parent companies’ revenue should be counted, but this is
not what the bill says — which further suggests the bill
is tailored to social media sites. In any event, including some large
traditional media websites in its scope wouldn’t come anywhere
near making the bill broad enough to avoid the concerns of
Minneapolis Star or Arkansas Writers’ Project.
Second, the bill applies only to a
particular subset of Internet media — websites, apps and
services that host user content, not services like Netflix or non
Internet media. On its own, this all but ensures that the bill would
be subject to strict scrutiny — which it would surely fail. See
Turner Broadcasting System, Inc. v. FCC, 512 U.S. 622 (1994)
(“Regulations that discriminate among media … often present
serious First Amendment concerns.”); Minneapolis Star
Tribune, Co. v. Minnesota Commr of Revenue, 460 U.S. 575, 583
(1983) (a tax applied only to newspapers).
Arguably, a bill that applied
equally to all “interactive computer service providers”
would be less problematic because it would not single out a
“small group” of sites for what amounts to punishment.
Abandoning user count or revenue thresholds would avoid the problem
of retaliatory targeting, but additional First Amendment problems
would remain.
Hawley’s Bill
Would Backfire Against Conservatives
It’s impossible to anticipate,
ex ante, the net effect of the law upon the decision-making of
each social media service — i.e., whether they will do
more or less moderation, and whether conservatives would actually
benefit overall. The chief purpose of Section 230 was to avoid the
“Moderator’s Dilemma,” created by Stratton
Oakmont, Inc. v. Prodigy Services Co., 1995 WL 323710 (N.Y. Sup.
Ct. 1995). The court held Prodigy more liable because it
actively engaged in content moderation to create a “family-friendly”
service. If edge providers fear that removing certain content may
increase their legal risks, they will moderate less. On the other
hand, they may calculate that more moderation will allow them
to claim a more consistent approach.
That the same law could produce
diametrically opposite results is not at all unusual in First
Amendment jurisprudence. This is precisely the constitutional problem
with vague laws: they are both unpredictable and highly subject to
manipulation by those charged with enforcement.
Empowering the government to
determine political neutrality cuts both ways. Discouraging edge
providers from moderating incendiary or abusive speech from the right
will have the same kinds of effects on the left. Democrats will just
as easily claim “bias” when speech they like is removed.
Consequently, social media sites will hesitate to take down content
from Antifa or radical anti-police activists
for fear that a Democratic FTCor state attorney general will sue
them.
More generally, if Republicans start
suing edge providers for failing to deliver on the claim of
neutrality required by the new Hawley bill, you could count on
Democrats — when they have the chance — to start suing
social media operators for not living up to other provisions in their
community standards. Consider Twitter’s Community Standards:
Twitter has made an editorial
decision not to remove tweets posted by President Trump that seem to
violate all of these prongs (minus the one about child sexual
exploitation). The First Amendment clearly protects their right to
make that decision, but if the government could hold a company to
such statements about its editorial practices, as Hawley claims,
without violating the First Amendment, why couldn’t a
Democratic FTC make the same argument about Twitter not living up to
its promise to enforce its community standards? Indeed, Facebook has
been heavily
criticized by groups on the left for failing to do
more to take down racist content that may even incite users to
violence.
For better or worse, the First
Amendment prevents the government from forcing Facebook, Twitter or
any other social media sites to change how they favor, disfavor, or
remove user content. But if Hawley’s bill were somehow to pass
now, it could just as easily be used by a Biden administration to
pressure social media sites to take down right-leaning content in the
years it would take for the complex legal questions outlined here to
work their way through the courts.
The “Problem”
for Republicans Isn’t 230, but the First Amendment
In the end, Republicans’
complaints aren’t really about Section 230, but about the First
Amendment. Yes, Section 230 protects websites from liability for user
content — “death by ten thousand duck-bites.”
Roommates, 521 F.3d at 1174. While the Hawley bill and Trump’s
Executive Order both make edge providers liable for what users say,
this is only a means to an end; their real focus is not on the
decision made by edge providers to host potentially unlawful content,
but on their decision not to host content they deem
objectionable. That decision is one the First Amendment protects as
fully for websites as it does for newspapers or Fox News.
Trump, Hawley and other Republicans
would do well to remember what President Reagan said when he vetoed
legislation to restore the Fairness Doctrine back in 1987:
We must not
ignore the obvious intent of the First Amendment, which is to promote
vigorous public debate and a diversity of viewpoints in the public
forum as a whole, not in any particular medium, let alone in any
particular journalistic outlet. History has shown that the dangers of
an overly timid or biased press cannot be averted through
bureaucratic regulation, but only through the freedom and competition
that the First Amendment sought to guarantee.
Republicans should ask themselves:
“WWRD—What Would Reagan Do?” The answer should, by
now, be clear: “Congress shall make no law…”
