The two imperatives of the First Amendment

The Constitution embraces the Lockean principles of free speech and private property as "different aspects of an indivisible concept of liberty." Intel v. Hamidi, 30 Cal. 4th 1342 (2003) (Brown, J, concurring). But concerns about viewpoint-based censorship on social media expose a tension between the two. Do free speech principles prohibit political bias on media like YouTube and Twitter, or do these sites enjoy a property right to exclude whatever material they wish?

Two speech models

The First Amendment encompasses both a civic and autonomy imperative. Speech is the "essence of self-government," which requires an "uninhibited, robust" exchange of ideas. Garrison v. Louisiana, 379 U.S. 75 (1964). Property rights therefore cannot justify censorship absolutely; police could not constitutionally arrest a women for distributing religious writings on the street of a company-owned town. Marsh v. Alabama, 326 U.S. 501 (1946). The California Supreme Court, observing that shopping malls were replacing the business district, extended this principle to hold the state constitution protected students seeking petition signatures at malls, whose prominence as public meeting places rendered them "an essential and invaluable forum" for exchanging ideas. Robins v. Pruneyard Shopping Center, 23 Cal. 3d 899 (1979). The U.S. Supreme Court found this construction of state law permissible (though not mandatory), as the mall owner's property interest did not permit him to exclude speakers. PruneYard Shopping Center v. Robins, 447 U.S. 74 (1980). Cyberspace is the successor to streets and malls. Packingham v. North Carolina, 137 S. Ct. 1730 (2017).

On the other hand, the First Amendment protects speaker autonomy, and protects the right to not speak or support speech. Championed by Thomas Jefferson ("To compel a man to furnish contributions of money for the propagation of opinions which he disbelieves and abhors is sinful and tyrannical"), it underlies establishment clause restrictions on compelled funding of religious institutions, and protects individuals from expressing or supporting speech against their will. Janus v. AFSCME, 138 S. Ct. 2448 (2018); NIFLA v. Beccera, 138 S.Ct. 2361 (2018).

The Supreme Court thus struck down a law forcing newspapers to provide a "right to reply" to candidates they had criticized. Miami Herald Publishing Co. v. Tornillo, 418 U.S. 241 (1974). Despite the statute's purpose of ensuring "an electorate informed about the issues," the Supreme Court protected the autonomy of a "privately owned newspaper to advance its own political, social, and economic views." PruneYard distinguished Miami Herald, as the mall had no comparable interest in expressing a particular viewpoint.

These civic and autonomy imperatives collided when college students objected to being forced to contribute to a speakers' fund, due to disagreement with some of the speech presented. Board of Regents v. Southworth, 529 U.S. 217 (2000). As the fund promoted a "free and open exchange of ideas," PruneYard, rather than Miami Herald, governed the decision. So long as the program was viewpoint-neutral, the Supreme Court would exempt it from the general rule forbidding compelled support for speech. The students were not sponsoring any particular speaker but the forum as a whole.

The Communications Decency Act

Congress relied on the Marsh/PruneYard model in creating the Communications Decency Act. 47 U.S.C. Section 230. Citing the internet's role as "a forum for a true diversity of political discourse," Congress rejected the Miami Herald model, and categorized "interactive computer services" as not the publishers or speakers of content provided by other parties. Congress thus distinguished such services from newspapers and radio/television broadcasters, which may be held "liable for publishing or distributing obscene or defamatory material written or prepared by others." Blumenthal v. Drudge, 992 F.Supp. 44 (D.D.C. 1996). Sites were more like the U.S. Postal Service or telephone providers, common carriers which do not monitor communications, and are not liable when customers use them to arrange drug deals or plan a murder. To preserve the internet's "vibrant and competitive free market," Congress declined to impose affirmative burdens on internet services, so they would not be "liable for their failure to edit, withhold or restrict access to offensive material disseminated through their medium." "Publisher" status is a two-sided coin. The Miami Herald concurrence observed "freedom of the press is not a freedom from responsibility for its exercise," so the paper's choosing what to publish and what to suppress entailed responsibility if it published defamatory or other unprotected content. See also Cal Const., art. I, sec. 2 (emphasis added): "Every person may freely speak, write and publish his or her sentiments on all subjects, being responsible for the abuse of this right." Publishers bear responsibility for the ultimate product, even when it comes from another source.

Common carriers, by contrast, bear no such liability because they do not "edit, withhold or restrict access." Southworth suggests this viewpoint-neutrality is a condition for fora's exemption from otherwise applicable rules.

Distortions and double standards

Two cases pending before the 9th U.S. Circuit Court of Appeals are testing -- and distorting -- these principles. Prager University v. Google, LLC, 18-15712; Gonzalez v. Google, LLC, 18-16700. PragerU sued Google/YouTube for restricting its conservative videos. Google/YouTube's briefing justifies this suppression by asserting Miami Herald protects "editorial judgments ... of how ... or even whether to present, particular content." The converse appears in Gonzalez, where relatives of terror victims sued YouTube for posting ISIS videos facilitating an attack. After championing its "editorial judgment" against PragerU, the same defendant argued six months later, "Section 230 forbids ... treat[ing] Google as the 'publisher or speaker' of content posted by others." ISIS does not violate its standards, but Dennis Prager does.

In sum, when it comes to censoring speech, Google contends it may transmit or reject content as it pleases, like a newspaper. When it comes to liability, Google claims it is not a speaker, just a passive conduit for others' voices.

But perhaps even content from other sources becomes the speaker's own through editing choices. Southworth overrode the general rule against forced subsidies by considering the forum holistically, and not by reference to its parts. Similarly, presenting some speech and suppressing other speech (or even rearranging the order of the feed) creates a distinctive, new message from that expressed by the original creators.

For example, Middle East journalists have acknowledged they fear losing access to sources within Gaza for their reporting, or worse, suffering violent reprisals, so they self-censor. If a broadcaster/publisher presents an Israeli police response but omits the preceding violence, it conveys a different message than if it showed both.

Not just speakers but audiences lose when selective information distorts the debate. For example, Twitter suspended the account of Mary Ann Mendoza, whose son was killed by a man in the U.S. illegally (and who had failed to appear for his sentencing hearing on a prior conviction) unless she deleted, "@Kamala Harris[,] What law can I break and have you defend me so staunchly? Provide me sanctuary from our laws?" Looser immigration and bail policies present both benefits and costs, but if the public hears only the former, it distorts the process of self-government, and the resulting laws we enact.

Twitter claimed Mendoza had violated its "hate speech" standards, which some consider necessary to protect vulnerable populations. But the tweet actually criticized a United States senator, and the First Amendment's most basic function is enabling scrutiny of governmental policies. Continental Europe, by contrast, developed its speech restrictions in order to shield the government and powerful institutions from challenge. Keiter, "Balancing a 'Right to Be Forgotten" with a "Right to Remember," 13 Cal. Legal Hist. 421 (2018); see also Reed v. Town of Gilbert, 135 S. Ct. 2218 (2015) (Alito, J. concurring): "Limiting speech ... favors those who do not want to disturb the status quo. Such regulations may interfere with democratic self-government and the search for truth."

Speech is simultaneously receiving more protection than ever from the Supreme Court, and less from internet gatekeepers. One year after the Supreme Court held the government could not deny a trademark to the band The Slants for being offensive to Asian-Americans (Matal v. Tam, 137 S. Ct. 1744 (2017)), Twitter banned using transgender persons' "biological" rather than "social" pronouns, or birth name (e.g. referencing "Bruce Jenner" as "him"). This conflict presents competing interests, and both deserve protection.

We cannot be forced to express an idea we reject as a condition for driving, even on our own property. Wooley v. Maynard, 430 U.S. 705 (1977) (state could not condition driving on displaying "Live Free or Die" license plate). Packingham suggests that "surfing the web" is just as important as driving to full social participation. Though Section 230(a)(4) announced minimum regulation worked "to the benefit of all Americans," silencing "wrong" pronoun users contravenes that principle. Nevertheless, forcing Twitter to express what it considers the "wrong" pronoun likewise infringes its rights -- if it is expressing its own ideas as a speaker. But a common carrier like Verizon asserts no comparable infringement when its network communicates controversial ideas.

The solution may lie in the compromise established in education. Public schools may not force students to pray or salute the flag, but receive full funding from the government, whereas religious schools have more discretion in regulating student speech, but do without most of the funding otherwise available. A social media site should likewise be categorized as public or private, a utility or publisher. The former would be a true platform for all ideas without censorship, and immune from challenge for not editing or withholding material. The latter would be free to maintain its values in communicating the content expressed on it, but responsible for its choices. By fostering an "uninhibited" exchange of ideas while preventing their involuntary expression, this would fulfill both First Amendment imperatives. 

Mitch served in the Chambers of California Supreme Court Justice Janice Rogers Brown and filed a successful amicus brief in NIFLA v. Beccera.