Social Media Common Carrier

Social Media Platforms as Common Carriers?

I finally have a presentable draft of this article, forthcoming in the Journal of Free Speech Law; I'll be posting excerpts over the next couple of weeks.


You can also read the article in PDF, or read all the posts that have been posted so far on this thread. I still have time to make edits, and I'd love to hear what corrections or suggestions or counterarguments people have. Here, to start with, is the Introduction:

[* * *]

Say that the U.S. Postal Service refused to allow the mailing of KKK, Antifa, or anti-vax publications.[1] That would be unconstitutional,[2] however much we might appreciate the desire of USPS managers to refuse to participate in spreading evil and dangerous ideas. And though UPS and FedEx aren't bound by the First Amendment, they too are common carriers[3] and thus can't refuse to ship books sent by "extremist" publishers.[4]

Likewise for phone companies, whether land-line monopolies or competitive cell phone providers.[5] Verizon can't cancel the Klan's recruiting phone number, even if that number is publicly advertised so that Verizon can know how it's being used without relying on any private information.[6] To be precise, the companies need not be common carriers as to all aspects of their operation: They can, for instance, express their views to their customers in mailings accompanying their bills, without having to convey others' views.[7] But they are common carriers as to their function of providing customers with telephone communications services.

And this seems to me to be a valuable feature of our regulatory system, not just an odd side effect of common carrier law. Certain kinds of important infrastructure, under these rules, are available equally to all speakers, regardless of the speakers' ideologies. Government enterprises (such as the post office) shouldn't decide which organizations or ideas should be handicapped in public debates. And neither should large private businesses, such as phone companies or package delivery services.

That is important even as to groups and viewpoints that are seen as extreme. But it is especially important as to viable political candidates, ideas, or media outlets that are serious competitors in democratic life. When elections are closely divided, even small interference with various groups' ability to affect public opinion can make a big difference in outcomes.[8] FedEx and Verizon shouldn't have the power to thus affect elections by refusing to carry certain views.

On the other hand, say the Los Angeles Times refuses to run an ad promoting the KKK, or promoting Antifa, or opposing vaccination. There is good reason to support the Times' right to do this. People read the Times in part precisely because of its editorial judgment, its ability to winnow the good and sensible views out of the vast mass of nonsense and folly; treating the Times as a common carrier would make it useless. And indeed the Times would have a First Amendment right to refuse to publish whatever material it chooses.[9]

The same would likely happen if a bookstore refused to distribute books like that. Perhaps both the newspaper and the bookstore might be condemned as unduly narrow-minded, if they go too far in excluding such material, at least unless they promote themselves as being ideologically focused. But for material that is seen as sufficiently extreme, newspapers' and bookstores' rejecting such material is quite normal.

The question, of course, is where we might fit the various functions of social media platforms.[10] This Article will offer some (often tentative) thoughts on this questions. I'll begin by asking in Part I whether it's wise to ban viewpoint discrimination by certain kinds of social media platforms, at least as to what I call their "hosting function"—the distribution of an author's posts to users who affirmatively seek out those posts by visiting a page or subscribing to a feed.

I'll turn in Part II to whether such common-carrier-like laws would be consistent with the platforms' own First Amendment rights, discussing the leading Supreme Court compelled speech and expressive association precedents, including PruneYard Shopping Center v. Roberts; Turner Broadcasting System v. FCC; Rums­feld v. FAIR; Miami Herald Co. v. Tornillo; Wooley v. Maynard; Pacific Gas & Electric Co. v. Public Utilities Commission; Riley v. National Federation of the Blind; Hurley v. Irish-American Gay, Lesbian & Bisexual Group; NIFLA v. Becerra; Boy Scouts of America v. Dale; and Janus v. AFSCME. (I discuss elsewhere whether such laws, if enacted on the state level, would be barred by 47 U.S.C. § 230(c)(2)(A) and the Dormant Commerce Clause.[11]) And then I'll turn in Part III to discussing what Congress may do by offering 47 U.S.C. § 230(c)(1) immunity only for platform functions for which the platform accepts common carrier status, rather than offering it (as is done now) to all platform functions.

On balance, I'll argue, the common-carrier model might well be constitutional, at least as to the hosting function. But I want to be careful not to oversell common-carrier treatment: As to some of the platform features that are most valuable to content creators—such as platforms' recommending certain posts to users who aren't already subscribed to their authors' feeds—platforms retain the First Amendment right to choose what to include in those recommendations and what to exclude from them.

And I also don't want to oversell the label "common carrier." I think the analogy to certain familiar common carriers, such as phone companies and package delivery services, is helpful; but it's only an analogy. Even if it proves to be a helpful analogy, there's little reason to think that all the details of common carrier law ought to be fully adopted for social media platforms, or that the threshold for regulation should be defined by traditional common carrier rules.[12]

Other analogies can also be helpful: As Part II.A will argue, the clearest First Amendment analogs would be cable must-carry rules and rights of access to the real estate of shopping malls and universities.[13] Justice Thomas has recently suggested that public accommodation laws might be useful analogies as well;[14] indeed, some courts have recently treated media web sites as places of public accommodations for purposes of disability law,[15] and laws in some jurisdictions already ban discrimination based on political affiliation or ideology.[16] The point is simply that the insights behind how certain communication and distribution services—and certain forms of property more generally—may and may not be regulated could also be helpful for thinking about various functions of social media platforms.

[1] Assume that it wouldn't need to open sealed envelopes, because the nature of the material is clear from the identity of the mailer or from the cover of an unwrapped magazine.

[2] See Lamont v. Postmaster General, 381 U.S. 301, 307 (1965).

[3] See, e.g., FedEx Corp. v. United States, 121 F. App'x 125, 126 (6th Cir. 2005).

[4] See 49 U.S.C. § 13101(a)(1)(D) (setting forth general policy against "unreasonable discrimination"), § 14101(a) (requiring common carrier to provide "transportation or service on reasonable request"); Mitchell v. United States, 313 U.S. 80, 94–95 (1941) (interpreting predecessor to this statute as banning race discrimination by common carriers, because such discrimination "would be an invasion of a fundamental individual right" if done by the government). They may sometimes be expected to monitor shipments for illegal content, see, e.g., U.S. Dep't of Justice, UPS Agrees To Forfeit $40 Million In Payments From Illicit Online Pharmacies For Shipping Services, Mar. 29, 2013,, but they can't block materials simply because they don't like the ideas expressed within them.

[5] 47 U.S.C. § 202(a); Genevieve Lakier, The Non-First Amendment Law of Freedom of Speech, 134 Harv. L. Rev. 2299, 2317 (2021); Michael Kent Curtis & Eugene D. Mazo, Campaign Finance and the Ecology of Democratic Speech, 103 Ky. L.J. 529, 557 (2015).

[6] Christopher Yoo, The First Amendment Rights of Common Carriers: Net Neutrality, Privacy, and Beyond (forthcoming 2021), notes cases holding that allowed phone companies some power to block certain uses of their services, chiefly for dial-a-porn. But those all involved statutes that the courts read as specially limiting the common carrier obligation. Carlin Commc'ns, Inc. v. Mountain States Tel. & Tel. Co., 827 F.2d 1291, 1293–95 (9th Cir. 1987) (state law "prohibiting the distribution of sexually explicit material to minors," which was seen as "embod[ying]" a "public policy" of "protecting minors from 'adult entertainment'"); Carlin Comnc'n, Inc. v. S. Bell Tel. & Tel. Co., 802 F.2d 1352, 1361 n.5 (11th Cir. 1986) (47 U.S.C. § 201 exclusion of the service from common carrier obligations); Network Commc'ns v. Mich. Bell Tel. Co., 703 F. Supp. 1267, 1275 (E.D. Mich. 1989) (same); Info. Providers' Coal. for Def. of First Amend. v. FCC, 928 F.2d 866, 877 (9th Cir. 1991) (47 U.S.C. §§ 233(b), (c)(1) exclusion of "indecent" communications).

[7] Pacific Gas & Elec. Co. v. Pub. Util. Comm'n, 475 U.S. 1 (1986), so held as to public utilities generally, and there is no reason why this analysis would be different for common carriers.

[8] Cf. Jonathan Zittrain, Engineering an Election, 127 Harv. L. Rev. F. 335 (2014).

[9] Miami Herald Pub. Co. v. Tornillo, 418 U.S. 241, 258 (1974).

[10] For earlier examples of calls to treat social media platforms as common carriers, see, e.g., K. Sabeel Rahman, Regulating Informational Infrastructure: Internet Platforms as the New Public Utilities, 2 Geo. L. Tech. Rev. 234 (2018); Adam CandeubBargaining for Free Speech: Common Carriage, Network Neutrality, and Section 230, 22 Yale J. L. & Tech. 391, 433 (2020); Tunku Varadarajan (interviewing Richard Epstein), The 'Common Carrier' Solution to Social-Media Censorship, Wall St. J., Jan. 15, 2021; Richard Epstein, Should Platforms Be Treated as Common Carriers? It Depends–Perhaps (working paper). For a very early suggestion along those lines, though not aimed at modern social media platforms, see David J. Goldstone, A Funny Thing Happened on the Way to the Cyber Forum: Public vs. Private in Cyberspace Speech, 69 U. Colo. L. Rev. 1, 40–47 (1998).

[11] See Adam Candeub & Eugene Volokh, Interpreting 47 U.S.C. § 230(c)(2), 1 J. Free Speech L. __ (2021); Eugene Volokh, Does 47 U.S.C. § 230(c)(2) Violate the First Amendment? (in draft); Eugene Volokh, State Social Media Mandates and the Dormant Commerce Clause (in draft).

[12] Social media platforms today aren't common carriers under some traditional definitions of the term, because they don't hold themselves out as "neutral conduits of information." Matthew Feeney, Are Social Media Companies Common Carriers?, Cato Inst. (May 24, 2021, 3:39 pm),; Berin Szóka & Corbin Barthold, Justice Thomas's Misguided Concurrence on Platform Regulation, Lawfare (Apr. 14, 2021, 10:30 am), But cases such as Rumsfeld v. FAIR and Turner Broadcasting v. FCC show that access mandates may be imposed even on institutions—such as universities and cable operators—that are far from neutral conduits in many of their operations (e.g., defining their curriculum, hiring faculty, organizing conferences, or selecting what channels to include) and that may seek to be nonneutral in further ways (say, in selecting who may recruit on campus).

And even an entity that publicly announces that it wants to exclude some people may still be required to include them, whether this is because it's treated as a common carrier, a public utility, or as subject to a sui generis public access regime—after all, one facet of common carrier status is that the common carrier is barred from discriminating. See, e.g., 47 U.S.C. § 202(a). Common carrier status, for instance, barred railroads from discriminating among passengers based on race, even before such discrimination was expressly forbidden by public accommodations statutes. See, e.g., Mitchell v. United States, 313 U.S. 80, 97 (1941). Likewise, PruneYard was required to allow leafleters even though it had an express "policy not to permit any visitor or tenant to engage in any publicly expressive activity, including the circulation of petitions, that is not directly related to its commercial purposes." PruneYard, 447 U.S. at 77. This was likewise true for the universities in Rumsfeld, and the cable systems in Turner.

Likewise, telephone companies were barred from excluding people based on mere suspicion that they were using the service for illegal purposes, such as gambling. Andrews v. Chesapeake & Potomac Tel. Co., 83 F. Supp. 966, 968–69 (D.D.C. 1949); Nadel v. N.Y. Tel. Co., 170 N.Y.S.2d 95, 98 (N.Y. Sup. Ct. 1957). One rationale for this bar was that, "Public utilities and common carriers are not the censors of public or private morals, nor are they authorized or required to investigate or regulate the public or private conduct of those who seek service at their hands." Pa. Publications v. Pa. Pub. Util. Comm'n, 36 A.2d 777, 781 (Pa. 1944) (cleaned up); People v. Brophy, 120 P.2d 946, 956 (Cal. App. 1942); Commonwealth v. Western Union Tel. Co., 67 S.W. 59, 60 (Ky. 1901).

[13] See infra Part II.A.1.

[14] Biden v. Knight First Am. Inst. at Columbia Univ., 141 S. Ct. 1220, 1224, 1227 (2021) (Thomas, J., concurring).

[15] See Winegard v. Crain Commc'ns, Inc., No. 20-CV-01509 (AJN), 2021 WL 1198960 (S.D.N.Y. Mar. 30, 2021); Sullivan v. BDG Media, Inc., 71 Misc. 3d 863 (N.Y. Sup. Ct. 2021); Nat'l Fed'n of the Blind v. Scribd Inc., 97 F. Supp. 3d 565 (D. Vt. 2015); Nat'l Ass'n of the Deaf v. Netflix, Inc., 869 F. Supp. 2d 196 (D. Mass. 2012); see also Sullivan v. LLC, No. 18-CV-1939 (JPO), 2019 WL 1299966 (S.D.N.Y. Mar. 21, 2019) (treating such a media site as a place of public accommodation, though in a case where defendant did not contest that proposition). For more on whether web sites that are ancillary to the sale of other goods and services are places of public accommodations for purposes of federal disability discrimination law, see Gil v. Winn-Dixie Stores, Inc., 993 F.3d 1266, 1277 (11th Cir. 2021) (saying no); Robles v. Domino's Pizza, LLC, 913 F.3d 898, 905–06 (9th Cir. 2019) (saying yes); Blake E. Reid, Internet Architecture and Disability, 95 Ind. L.J. 591, 597–99 (2020) (summarizing the dispute).

[16] These jurisdictions include D.C., Seattle, Madison, Ann Arbor, Champaign-Urbana, the counties containing Detroit and Ft. Lauderdale, and several other cities, counties, and territories, and possibly also California. See Eugene Volokh, Can Places of Public Accommodation Exclude People Based on Their Politics?, Volokh Conspiracy, Apr. 8, 2021, at 5:46 pm, Some of the laws ban only discrimination based on party affiliation, but others ban discrimination based on broader political beliefs as well. The main federal public accommodations law, Title II of the Civil Rights Act of 1964, doesn't currently treat social media platforms as places of public accommodation, Lewis v. Google LLC, No. 20-16073, 2021 WL 1423118, (9th Cir. Apr. 15, 2021), and in any event doesn't ban discrimination based on ideological belief; but the question would be whether it, and similar laws, should be extended.

NEXT: Kelo, Originalism, and Public Use

Editor's Note: We invite comments and request that they be civil and on-topic. We do not moderate or assume any responsibility for comments, which are owned by the readers who post them. Comments do not represent the views of or Reason Foundation. We reserve the right to delete any comment for any reason at any time. Report abuses.

  1. Say that the U.S. Postal Service refused to allow the mailing of KKK

    I believe southern postmasters did refuse to distribute abolitionist mailings, and memory sez there was some kerfuffle in Congress to forbid the USPS in general from accepting them for mailing even in northern states; but the details escape me.

    1. The point of the argicle is self evident. Morecommunication happens on these platformsthan on others. They have 90% of the platform market. They are quasi-governmental organizations, with procedures, with accusations, procedures, punishments, appeals from those procedures. They are immunized. They are bigger and richer than governments. They are promoting the interests of the Chinese Commie Party to get access to its markets. They are all knowing, even with phone turned off, in a 100% surveillance state. They reported your body noisesand movementsto world headquarters without permission. Their Terms of Use are unconsxionabe in procedure and in substance. They are contracts of adhesion like taxes.

      They allowed billions of federal crimes on their platforms. They themselves have committed millions of crimes inflating viewerships to advertisers. Half their viewers are not human.

      Instead of regs and laws, they should be seized in civil forfeiture. They should be auctioned off like the Ferrari of a drug dealer. Repeat unless they straighten out.

      Passing laws is too risky, since they can just buy out entire legislatures with jobs to relatives of lawmakers. What Eugene is implying is a total waste of time.

      They are our enemy.

      1. They do act as sovereign nations which can ignore the Bill of Rights…and try and influence elections. If they are essentially Bolshevik fifth columnists which one can make an argument they are, then as enemies of liberty their leadership should be deported (Eastern Europe/Old Commie Land would be appropriate), assets confiscated and new management which allows all opinions should be appointed…

        1. Titus: That reminds me. Greatest fraud heist in human history. By their ownership of the media and of the Democrat Party, the Dem Gov lockdowns, took down our economy to defeat Donald Trump. The hysteria, and over reaction netted them an election, and $1.7 trillion in added wealth. The billionaires of China scored $2 trillion.

          They dropped the world GDP by $4 trillion, and killed 100 million people by starvation, the biggest, fastest mass murder in human history for their own enrichment.

          Eugene should stop kidding around, doing that dumb, lawyer, weasel, incremental thing.

          1. Eugene does not really understand even recent history. Immunity grows an enterprise. Liability shrinks it. These are unauthorized industrial policy by the lawyer profession. Congress wanted to grow the internet at the time it enacted Section 230 in 1996. Now it is big, and does not need protection nor growth.

            The same was done with trains, with charities, marriages (long ago), tribunals, reporters, states. All are gigantic now, and can all become liable. Liability is the royal road to shrinking an enterprise. The lawyer did a great job of destroying shipbuilding, the American family, manufacturing.

            1. No matter what those unconscionable Terms of Use say, the companies have converted the personal information of billions of users without compensation. I estimate the value of that information to be half of revenues. An good, non-frivolous, aggregate claim should retrieve that value for the users. This is a crime of theft. The government should no longer suborn theft.

              1. I get a picture of my mail ahead of delivery every day. It is a small step to convert the information in the sender and recipient areas of the mail into advertising information, to enhance Post Office profits.

                1. Ohio AG lawsuit against Google for bias in search results to directing viewers to Google products.


        2. They do act as sovereign nations which can ignore the Bill of Rights

          No; they act as private citizens to whom the Bill of Rights doesn’t apply.

    2. Yes, John Calhoun was famous for initiating this ban. At the time, it was simply obvious to any progressive person that these religious fundamentalists who were agitating against queer folk and their peculiar institutions were motivated solely by animosity, pure hatred, and not by any sort of rational or legitimate political motive. The good queer folk in the South had a right to feel safe from such evil people’s hate.

      Very progressive stuff, that John Calhoun. His admirers in the left are imitating him to this day.

      1. Well, that’s a stretch.

        1. Horseshoe theory seems closer to fact every day, regardless of your opinion.

  2. Telephone companies are I think required to discontinue service to people who use their phone to harass people.

    1. I’m pretty sure that’s not right. Even if you are an evil harasser, you still have a need and right to call 911 in case of emergency, contact your representative or senator by phone, etc. Your specific harassing behavior can be enjoined but forbidding you from using the phone even for legitimate purposes would seem to be classic prior restraint.

  3. A few comments.

    1). Great article so far.
    2) Just an FYI, when posting the article to the blog, leave out the footnotes. They read well in the .pdf, but in the blogspace the formatting is confusing.
    3) Page 17 ” Nor have they, to my knowledge, blocked major candidates’ speech during an actual campaign”

    -On October 15th, 2020, Twitter briefly suspended the “TeamTrump” campaign account and several related accounts, when they posted about Hunter Biden’s laptop.

    1. 2) Just an FYI, when posting the article to the blog, leave out the footnotes. They read well in the .pdf, but in the blogspace the formatting is confusing.

      I disagree. What browser are you using? I am using Edge and the formatting of the above footnotes is not at all confusing.

      1. I was confused.

        What’s a footnote? Are there headnotes and armnotes too?

      2. Re Footnotes: Chrome

        Re: Formatting footnotes.

        Again, this is preference. But the Professor has some rather long footnotes here. In the .pdf, it’s clear what they are, through the smaller text and placement on the same page. They can be referenced on the same page easily, in the context for which they are meant.

        In the blog, if you’re reading quickly, they almost read like part of the article. I can have the entire screen be one footnote. There’s no font size differentiation. And flipping from the end of the article to the beginning is more time consuming and makes it easier to lose track of the context.

        1. And flipping from the end of the article to the beginning is more time consuming and makes it easier to lose track of the context.

          Remember that you can click on the number in the body to be taken to the footnote and on the number at the footnote to be taken back to the body.

  4. Sped through the rest of the article.

    1) In regards to the rest, also reference IIlhan Omar’s GOP opponent Danielle Stellabeing banned prior to the 2020 election.

    2) A second, unconsidered argument is the fiscal and political pressure other governments can exert on the major social media organizations in order to present their viewpoint, while suppressing dissident viewpoints. Recently, YouTube has taken down the Xinjiang videos which demonstrate the genocide occurring in China.

    The Chinese government has a long history of using its power to exert control over the media. Don’t refer to Taiwan as independent. Don’t refer to the Chinese government in a negative viewpoint. The following is a 94 page report on how China exerts it influence on Hollywood to censor undesired viewpoints.

    Now, if China was to (and likely currently does) exert that same influence on the social media organizations, to effectively eliminate dissemination of information, it would be corrosive to society. In this manner, the common carrier laws would be a benefit to the major social media corporations. The companies could point to these laws and say “We have to keep up the posts…they’re the law”. Absent that however, the companies will take down or remove the posts which offend China, so as to keep an advantageous relationship with the Chinese government.

    1. Now, if China was to (and likely currently does) exert that same influence on the social media organizations, to effectively eliminate dissemination of information, it would be corrosive to society.

      Since China bans the major social media organizations (Facebook and Twitter, at least), it would seem unlikely that it could exert a whole lot of influence on them.

      1. One would think. Yet they all kowtow to China. Zuck was over there a few years ago begging them to let him in.

        Twitter is filled with CCP propaganda and agents. Which propaganda then gets parroted by US media members like Chris Hayes.

        YouTube censored, removed, blocked, banned video evidence and testimonials regarding the concentration camps and torture of Uyghurs!

      2. It’s not just social media companies. Did everyone see all the Charlie Munger praise for communists over the weekend? Our “elites” are buckwild out of control.

      3. “Since China bans the major social media organizations (Facebook and Twitter, at least), it would seem unlikely that it could exert a whole lot of influence on them.”

        Well, it’s not as clear cut as you would think. For example, Hong Kong (owned by China) is currently a major point of contention between the social media companies (which are currently legal there…for now) and the Chinese Government (Or Hong Kong government, depending), in regards to new laws around social media. The social media companies REALLY don’t want to pull out, but they also don’t want to kowtow to the new laws. Other compromises might be made. Then there’s the billions the Chinese government spends in advertising on Facebook…

        So, there’s give and take there. The Chinese have a lot of carrots to offer to get their way. And some sticks as well.

  5. Mail delivery is not the same. Only the recipient can read the material delivered by mail. Forgive me, but I fail to see how a privately-held platform should be forced to allow disinformation or other noxious speech it opposes. The remedy for those who want to use Twitter or whatever platform to distribute their message is simple: Start your own.

    1. Disinformation is simply information I don’t agree with. Using that excuse just shows that you suffer from invincible ignorance. Because it shuts down debate.

      And “start your own” is pure unadulterated BS. Look at the recent history with Parler.

      1. That’s a nonsense definition, as if disinformation doesn’t exist.

      2. You must be referring to the issue where Parler had a legal contract with AWS and refused to honor their commitment and were resultingly kicked off the AWS service until they decided to comply with the legal contract they initially signed?

        1. Amazon’s position was roughly “the contract said they had to sing and dance on demand, and we didn’t like their dance”.

          1. And I am sure Amazon’s quick leap into action at the behest of politicians who have already unrolled their Dexter carving kits had absolutely nothing to do with it.

            1. There is certainly that aspect as well.

          2. That’s a bald-faced lie. Does your reputation not mean anything to you?


            Parler didn’t even bother to dispute the alleged facts regarding content which was promoting violence on their platform, in violation of the AUP with AWS. However here you are, pretending that those facts are somehow disputable, and proving to anyone who reads this that you’re completely full of shit and not to be trusted.

          3. Assuming for the sake of argument that you had accurately characterized Amazon’s position: so what? If you sign a contract saying you’ll do something, and then you don’t do it, then there will be consequences for your breach.

            Parler could’ve tried to prove that they didn’t actually breach the contract, but that was a complete loser of an argument, because they did.

    2. And yet you allow the government to confiscate private property (rental housing) for over a year because it’s “necessary?” And you require bakers to bake cakes, photographers to snap pictures, etc. Pls spare me the “private property” crap when it only goes one wAy. And as far as build your own, why does that not apply to across the board to all institutions. Let taxpayers fund the institutions ( schools, banks, services, businesses) they believe in?

      1. Because banks and housing are not the same as a website platform that an elementary school kid can create at recess these days.

      2. By “you” here I assume you mean Donald Trump? You know, the person who put the eviction moratorium in place.

      1. Because cars and website posts are the same things.

        1. Cars and website posts are not.

          Cars and monstrous web sites and multi-billion dollar cloud/server website hosts are more similar.

    3. re: “Only the recipient can read the material delivered by mail.” – post cards.

      re: “disinformation” – The whole point of freedom of speech is that we don’t trust people like you (or me) to decide what is “disinformation”. We trust readers to make that determination.

      re: “start your own” – Start your own postal service. Or water utility. Or even your own social media company. We’ll wait. Give us some personal data on how easy that is.

      We recognize common carriers because natural monopolies exist. The same network effects arguably make some social media companies into natural monopolies, too. The fact that a monopoly is natural means that we can’t simply break it up – but there are still dangers to it. Common carrier obligations are one way to balance those dangers.

  6. I’m not comfortable at all with using common carrier law to force social media companies to publish opinions they don’t agree with. I cannot understand what is conservative or libertarian about that solution. I’d much rather see conservatives build social networks and let the market decide which ones are more tolerant of opinion diversity.

    1. I tend to agree. I think antitrust law is the better focus for dealing with social media platforms. That, in my view, makes more sense to ensure competing voices are heard. If we want a marketplace of ideas, then just make sure it’s a free market. Or I suppose free-ish, since I’m not that naive.

    2. Really if you are going to stick the nose of regulation under the tent in a far less intrusive way it would be better to require the major website hosts to be content neutral as to who they offer hosting services to rather than trying to dictate what each individual platform must publish.

      Basically if you’re going to host a Communist site you gotta carry the Nazi’s if they want to buy space for a suite too.

      If you host Facebook then you cannot deplatform a conservative equivalent.

    3. I am skeptical of the common carrier theory too.

      Much better would be to have a social media tax, to tax profits at 75% for social media companies, defined as presenting content and advertising on users pages.

      And a an offsetting tax credit for presenting users content without censoring for political viewpoint, or “fact checking” or warning about “misinformation”.

      1. ” tax profits at 75% for social media companies, defined as presenting content and advertising on users pages.”

        That won’t work. Somehow these companies never have a profit to tax…

        1. “What good is electricity in the home?”

          “Senator, in 20 years, you’ll be taxing it.”

          And taking credit for it.

          And saying without infrastructure it wouldn’t exist, so government created it and can have detailed control over it, you should be grateful you wretch we don’t alter the deal further.

          1. Jump to the end: “How grateful? (Waggles fingers)”

          2. We’re really concerned about the issues now, not 20 years from now, when the next administration changes the law again.

    4. This comment is completely on target. I would think at least a few libertarian/conservative types would be wise enough to know that if they build a system to regulate this, it is going to be used against them in the end.

      It would be far better to fight to address why it is difficult to start competing businesses rather than regulate.

      1. “It would be far better to fight to address why it is difficult to start competing businesses rather than regulate.”

        This is a common argument, but it’s misleading. Let’s address it in the context of current common carrier laws.

        Current common carrier laws require Fed Ex or UPS, for example, to take your package and deliver it, regardless of its political contents, so long as you pay for it (There are certain practical limits).

        But, why do they exist? Why can’t Fed Ex just reject your package full of Antifa T-shirts? I mean, can’t anybody just start their own transportation business? All you need is a cheap car and some gas, right?

        1. “Current common carrier laws require Fed Ex or UPS, for example, to take your package and deliver it, regardless of its political contents, so long as you pay for it (There are certain practical limits).”

          This is not clearly the case. You should read the Professor’s support for the proposition, which is a vague statute and then a case interpreting its predecessor statute as prohibiting *discrimination on the basis of race*, not discrimination based on speech. It’s not at all obvious that UPS or Joe’s Interstate Trucking Co or whatever would be required to carry a package with a box that has “UPS is garbage” or “to hell with America” printed all over the outside.

          1. UPS or Fed Ex can have consistent packaging standards…which they do. The CONTENTS of the package, they are required to carry, regardless of political message. So, if the package contained a bunch of T-shirts that said “UPS is Garbage,” yes, they would be required to take the package.

            1. So Joe’s Interstate Trucking, do they need to have an established packaging standard that excludes external printing of “Joe’s Interstate Trucking is garbage” in order to actually refuse? In your imagined view of the law obviously, because as I pointed out, that is not established law.

        2. This is a common argument, but it’s misleading. …

          Yes, large markets controlled by a few players are going to distort transactions and market entry is not frictionless. We are not going to disagree here and this is fairly accepted economics. I do think that suggested remedies are shortsighted and betray a real lack of understanding of the underlying problem.

          But, why do they exist?

          They exist because the problems stated before exist and have historically been solved by knee jerk politics. It’s very much like someone solving a leak in a water tank by hammering on pieces of wood and when the leak resumes at the edge of the board, you hammer on another piece of wood. When someone tells you what you are doing is stupid, you simply retort “… but I have to stop the leaks” and before you know it you end up with a horrible Frankenstein’s monster of a regulatory system that is still leaking and simply results in a ton of wood on the side of the tank.

          Yes, the leaks exist. They need to be addressed, but they need to be patched cleanly at the source and that source is market entry. Rather than write another ton of regulations that don’t solve the underlying problem and simply allow the bigger corporations to bury the small ones in red tape, work on giving the small fish enough shallow water that they can grow in. All additional regulation is going to do is prevent entry of anyone new and provide a source of grift for the usual suspects.

    5. ” to force social media companies to publish opinions they don’t agree with. ”

      The problem is, that’s not really what’s going on.

      What’s going on is people are self-publishing on the platform (Twitter, YouTube, Facebook, etc), and the platforms are then finding opinions they disagree with, and taking them down. It’s a key, but important difference.

      1. Yeah, for the most part, it’s a lot like email. Regular folks are privately communicating among themselves and their closest 200 friends, and Facebook comes in and stomps on whatever it doesn’t like.

    6. Yes. Though with Parler it seems that those with unpopular views wanting a social network must also develop their own infrastructure from the ground up. Their own web hosting capacity, cell phone operating systems (so they control the stores), CDN, anti-DDOS, etc. I never actually looked at Parler (I don’t do twitter either), but wasn’t happy with what happened to them anyway.

    7. I’m not comfortable at all with using common carrier law to force social media companies to publish opinions they don’t agree with. I cannot understand what is conservative or libertarian about that solution.

      Oddly, (some of) the same conservatives who vehemently objected to net neutrality now want to force social media companies to host user content they disapprove of.

  7. I agree completely that platform censorship is becoming a serious problem. But I am not quite [yet?] convinced that labeling some elements of Facebook or Twitter as “common carriage” is going to improve anything.

    The core business model of both Facebook and Twitter is to make its users feel engaged, so they continue using the platform and thus continue to be targets for advertising.

    The proposed “hosting function” model is essentially the creator-subscriber model: one person creates a post, and others “subscribe” to it. Both Twitter and Facebook support that model directly. Requiring common-carrier treatment for this model is at least plausible. If I don’t like someone, I can unsubscribe.

    Twitter in fact is built almost entirely around the creator/subscriber model. BUT Twitter is also under considerable pressure to encourage overall “appropriate” behavior on their platform, and this gives them a powerful *business* incentive to engage in content moderation. Twitter also has serious business concerns about stamping their imprimatur on “false information”, which may cause other users to leave. This can be compared to the LA Times subscriber model: the paper knows that publishing some content will cause some readers to cancel their subscriptions. Twitter’s primary business concern is maximizing their user’s continued interest in the platform.

    For Facebook, this applies even more strongly: Facebook needs to be seen as a place that its users want to return to (even if it makes them mad, I suppose). Facebook has a business interest in not inciting users to cancel their accounts.

    Unlike Twitter, Facebook also has a huge volume of non-subscriber business, where views are at the whim of the Facebook Content Sharing Algorithm. If *I* post, my Friends will see that post **only if the Content Sharing Algorithm determines that their their seeing my content will further increase their Facebook engagement.**

    As with Twitter, Facebook doesn’t want to give their “ordinary” users reason to cancel their accounts. And on Facebook, most of their traffic lives outside the creator/subscriber model. In more practical terms, FB doesn’t want any content that causes its users to lose engagement. And, as with Twitter, they have a business interest in the moderation of *all* their content, so users don’t start to dread what they might see there. (And even if I do not subscribe, for example, to Mussolini’s feed, his presence their would increase the probability that I will run into other posts by his supporters.)

    These are not marginal issues. Facebook Inc has expressed concern that large classes of users are migrating away from, hence their investment in Instagram.

    I think the best Facebook analog of USPS/UPS package delivery [p 3] is Facebook Messenger: delivery of a specific item to a specific recipient. Package delivery is close to the Twitter model, but isn’t a perfect fit.

    Mostly, I find the LA Times analogy apt: the newspaper has both a business reason and a First Amendment right not to run content they disagree with, and not to accept that content as advertising either.

    Perhaps the best arguments for common carriage can be made for email (a lot like FB Messenger) and for web hosting. But, as you write [p 9], email is *not* considered a common carrier. If it is not, **I have trouble seeing why Facebook or Twitter subscriptions should be**.

    (The telephone is intrusive in a way that asynchronous communications are not. So it is a poor analogy.)

    As for web hosting, there are business reasons there too why platforms might not want to be common carriers: flooding attacks. A flooding attack on one hosted website typically interferes with at least some other sites hosted by the same provider. And regardless of that attack risk, if AWS is not a common carrier, I have trouble seeing why Facebook and Twitter should be. Nobody thinks received *Amazon’s* imprimatur.

    The bottom line is that I would rather see us start handing out common-carrier status to email and web hosting (with anti-spam provisions for email). Or at least to include these in the overall argument.

    Finally, relating to Section III, I believe that the entire point of Section 230(c)(1) is to protect sites that *do* engage in content moderation or editing. Prodigy did this, Compuserve did not. Compuserve wasn’t liable because of that. So if a site is a common carrier, I don’t see why they need Section 230 at all. (Which, to be fair, sounds like what you say as well.)

    Disclaimer: I have only skimmed the article at this point. Especially section II.

    Peter Dordal, Loyola Univ Chicago CS Dept

    1. My problem with your description of Twitter ahd FaceBook is that you describe how they were, and not how they are now. Both have clamped down hard on what they call “misinformstion”, but is better described in most cases as moderate to conservative speech. Instead of providing evidence and arguing ligically, the left today yells “you lie” and “misinformation”, and these platforms comply, silencing that moderate to conservative speech.

      I have moved away from these platforms, as have very many to the right of center. It seems counterproductive for them. They are actively chasing off users. I pretty much use FB only to track my kid, and see the pictures of their life that they post. So, I haven’t cut the cord yet…

      Where’s the profit there? Maybe the companies believe that the loss from moderate to conservative users is worth the cost, when the venefit is the imposition of a progressive reality on the western world. Or maybe SJWs have “captured” these companies, they have becone such echo chambers as a result, and they are so orofitable, that they dobt realky notice the revenue that they are losibg to their censureship.

      Or, possibly we look at this nationally, while these companies look ihternatiinally, and they see that they can make far more money catering to countries, like China, that benefit greatly when moderate to conservative speech in this country is suppressed. There are almost 4x as many Chibese as there are Americans, and out of those, maybe half agree, at least politicalky, with their heavy handed censorship of their polit8cal opponents abd enemies

    2. “Facebook also has a huge volume of non-subscriber business, where views are at the whim of the Facebook Content Sharing Algorithm…… most of their traffic lives outside the creator/subscriber model. ”

      Not really. You pick your friends. You pick your nose. But you can’t pick your friend’s nose. Point is, you only see on Facebook those people that you choose to friend or follow (and who you continually choose to remain friends or followers of).

  8. Look all these social media companies are just selling a digital bulletin board..yes the poster isn’t paying directly, the advertiser is but they are selling you a service (as they sell your contact info so that can be viewed as “payment” by the customer). So can Twitter or FB then discriminate against you? I know the CRA of 64 does allow political discrimination/viewpoint discrimination perhaps that is where the debate is needed. Today the Govt allows social media companies to discriminate (as the sellers)..maybe we should revisit the CRA sections 2/8 and decide if govt has any role in deciding what sellers and buyers agree to…should sellers like buyers be able to discriminate for whatever reasons they like? If not and sellers have to sell..then FB/Twitter and the rest can’t censor at all…

  9. Never thought I’d see the day when Eugene Volokh, long known as a Free Speech hawk, would come out in favor of the regulation of speech, in order to serve a conservative misinformation network.

    Hate speech? Nope. Sexually-harassing speech? Nah. Bullying? What even is that? But de-platforming got his guns firing.


    1. And yet, as you well know, he’ll allow unadulterated swill like yours to remain on this site, regardless of how many wide-eyed innocents who aren’t actually familiar with you might happen by and think you’ve said something cogent and thoughtful. And that’s why you spent the time typing it.

      1. The Volokh Conspiracy regularly and happily publishes vile racial slurs, and evocations of violence toward liberals, yet censors comments poking a bit of fun at conservatives.

        And this is the best conservative legal academic can offer.

        This explains the course of the American culture war.

        Carry on, clingers.

    2. I guess you feel that your ox is being gored.

      How about discussing the law as EV presents it rather than your druthers.

      1. To be precise, I think such common carrier regulations may be constitutional in some situations precisely because I don’t view them as regulation of speech, and certainly not as content- or viewpoint-based regulation of speech. (I discuss that in detail in Part II.)

        1. Yes – Part II takes for granted that we ought to treat social media platforms as “common carriers,” at least with respect to their “hosting function,” and then pencils out a way that one might constitutionally do so.

          Searching for an argument that social media companies are actually “common carriers” with respect to their “hosting function,” one does not find one in Part I. A number of vague policy arguments are made – with a parade of horribles drawn from the right-wing blogosphere. You wave at a host of Supreme Court precedents that cut against your policy arguments with a series of just-so distinctions.

          Eugene, at the core of your whole argument is an equivocation: “compelled hosting.” You use this term to draw a connection between forcing a social media platform to “host” speech with which it disagrees and cases where the use of someone else’s private property to engage with speech activities can be compelled. And you use it to distinguish social media platform “hosting” from compelled publishing and broadcasting of speech with which a newspaper or broadcaster to disagree. But nowhere do I see this connection/distinction based on the essence of what social media platforms do, which by any objective standard seems to be publishing. I submit content to Facebook, which Facebook then makes available to other users pursuant to the terms of our user agreement. That’s it.

          Instead, your rhetorical strategy is to avoid this difficult question by pumping intuitions with convenient analogies and emotional appeals. It is clear from your writing on this that you are struggling with this question, analytically – the amount of heat and just-so distinctions reminds me of a term paper in law school I had to bring to a close when I found the caselaw didn’t fully support the grand theoretical argument I thought I had come upon.

          It is awful, and incoherent. That’s not to say that it won’t manage to carry the day, somehow, in this bizarre world we live in, where Free Speech scholars write in Free Speech journals about how it is now “Free Speech” to compel ISPs to carry state propaganda. But it is incredibly sad.

      2. I would love to have nothing but the time and resources to spend my day reading Eugene’s lengthy article, refer to the underlying cases, check some of his cited secondary sources, etc., and engage in a more involved, earnest discussion of the merits.

        Unfortunately, I work for a living, and can only read the arguments as Eugene presents them here, maybe skimming the papers he posts, and respond accordingly.

        As for my “ox” – yes, I am alarmed that Republican legislators and governors in this country are trying to regulate Facebook and Twitter so as to coerce them into carrying their own lies and propaganda. It’s exactly like the kind of heavy-handed response to social media we’re seeing throughout the autocratic world. The president of Nigeria last month shut down and banned Twitter when they had the temerity to delete a tweet by him threatening to engage in an all-out civil war with Nigerian insurgents in the country’s south.

        In the section of his article where Eugene runs through a parade of horribles where Facebook has blocked “conservative” speech, he cites: an unfounded conspiracy theory floated during the campaign; another conspiracy theory about the origins of the coronavirus in China; an apparent attempt to direct harassment at leaders of the BLM; some racist dogwhistling from a disgraced former lawmaker; and a video containing COVID misinformation. (I note that Eugene chooses to describe all of these in highly misleading fashion.)

        So, yes, I am alarmed at the idea that conservative politicians may use the law and the Constitution, with the support of conservative “free speech” scholars like Eugene, to force social media platforms to carry their conspiratorial attacks on their political opponents or to propagate scientific misinformation about global pandemics in order to serve narrow political ends. Facebook and Twitter are already designed in ways that inflame fears and vindicate false beliefs; laws forcing them to do so do not bode well for the future of ordered democracy.

        1. “as to coerce them into carrying their own lies and propaganda”
          But they should carry your sides lies and propaganda. Your criticism is easily as emotion laden as you claim EV’s is. An example, “another conspiracy theory about the origins of the coronavirus in China” when in fact that theory is at least plausible (though I do not accept it as there is no evidence that SARS-CoV-2 is human engineered).

          Okay, I see how that works. My issue is not so much with partisan speech but with censoring opinion that is critical of “accepted authority” even when that speaker is presenting solid scientific evidence.

          Of course we should thank you for wanting to protect the stupid people who cannot tell the difference.

          1. Your comment is an example of precisely the kind of mentality that conservative politicians want to create and mold through social media.

            I at no point defended Facebook’s attempts to control or shape public debate by cracking down on right-wing conspiracy theories and misinformation. As it happens, I am also annoyed by the way that Facebook decides for itself what we should and shouldn’t be able to see. The mainstream media has been doing this, as well, in some misguided, self-righteous attempt to “protect us from ourselves,” as you might put it.

            What alarms me is requiring Facebook to carry that harmful, intentionally misleading material. Facebook should be free to choose what material it carries, while I – annoyed as I am by their editorial decisions – am fully entitled to stop using its services. Which I largely do not do.

            But you do not see that nuance. You just assume I must be “for” what you are “against,” and so you line up like a dutiful soldier, making all kinds of assumptions about what I must believe and support. That kind of polarization is helpful to your conservative overlords, who appreciate your uncritical support as they pursue their own narrow interests, and they seek to force Facebook, Twitter, et al., to assist them in that larger project.

            As for the “lab leak” theory – some important attention needs to be paid to what we know about it, our sources, why messages were being circulated when they were. Trump sought to promote the lab leak theory when it was generally tied to conspiracy theories about China being engaged in “gain in function” research and served as a useful way to deflect attention from Trump’s several missteps in handling the pandemic. The media responded accordingly – as though this were just another Trump lie. Now, we have some limited, unverified intelligence that a few lab workers had COVID symptoms earlier than had previously been known. Nothing about “gain in function,” nothing conclusively tying the infections to in-lab exposures, nothing like that – just some circumstantial evidence to support a plausible but still unverified claim. So nothing about what we now know shows that earlier media skepticism was off the mark, or even that over-zealous.

            But – again – the kind of glossing over the details is the same thing that Eugene does, in his own piece.

            1. “their editorial decisions”

              Their editorial decisions.

              Their free speech rights.

              Their speech.

              See what this is about?

              As I understand it, nobody is discussing requiring Facebook, et al to carry speech that they disagree with. Not as a bare directive. Instead, the discussion is whether rules would be applied to that effect in connection with special protections and immunities absolving them of responsibility for that content.

              It seems to me that you and Facebook, et al, want to have it both ways. They would like to say that they are just a tool, an “interactive computer service,” just a transmission line that people use to communicate their messages. It’s not their speech so they’re not responsible for it.

              On the other hand, they want to say that the content on their platforms, the sum of their many “editorial decisions” and careful content-based and politically motivated curation of what people see . . . is an expression of their views! An exercise of their deeply cherished free speech rights.

              1. As I understand it, nobody is discussing requiring Facebook, et al to carry speech that they disagree with. Not as a bare directive. Instead, the discussion is whether rules would be applied to that effect in connection with special protections and immunities absolving them of responsibility for that content.

                You don’t understand it. People are indeed discussing requiring Facebook, et al. to carry speech that they disagree with. Indeed, the state of Florida recently passed a law to do just that (though a federal judge immediately enjoined it).

                And in fact this very post of EV’s is about classifying these websites as common carriers, which indeed is about forcing them to carry speech that they disagree with.

                It seems to me that you and Facebook, et al, want to have it both ways. They would like to say that they are just a tool, an “interactive computer service,” just a transmission line that people use to communicate their messages. It’s not their speech so they’re not responsible for it.

                Nope. You don’t understand this either. This isn’t “both ways.” They are not saying that they are just a tool, just a transmission line. § 230 is not about forcing them to choose between blindly transmitting other people’s speech or having liability; it’s about them not having to make that choice.

                1. You don’t understand it

                  Funny, from my perspective (and many others it seems), it appears that you are the one who doesn’t understand it.

    3. Never thought

      That much is obvious.

    4. ” Never thought I’d see the day when Eugene Volokh, long known as a Free Speech hawk, would come out in favor of the regulation of speech, in order to serve a conservative misinformation network. ”

      Welcome to the club, pal. Artie Ray Lee Wayne Jim-Bob Kirkland may have been a hayseed, but he had this guy pegged long ago.

      The Volokh Conspiracy Board Of Censors must find this article hilarious.

    5. “would come out in favor of the regulation of speech, in order to serve a conservative misinformation network.”

      Ha, thanks comrade

  10. A few things you may want to consider:

    The “common carrier” analogy seems inapt for one reason you don’t really address; unlike every other common carrier, social media platforms as such do not actually “carry” anything. They do not “carry” goods, nor people, nor, as it were, information. They are not a telephone company, they are an answering service. They are not UPS or Fedex or USPS, they are a mail order company. All that they do happens on their property, on their servers. The messages arrive to them by common carrier, and leave by common carrier, but the social media site itself is not a common carrier.

    A similar objection can be lodged regarding the treatment of them as a utility (as many common carriers are). Unlike every other utility I can think of, they do not directly rely on the government’s allocation of a scarce resource for its existence. Whether it be allocation of limited space in public easements across private property for the laying of cables or pipes, or allocation of limited spectrum for radio communication, utilities rely on the exercise of government power to acquire a limited resource necessary to their proper functioning. A social media site does not. The government doesn’t need to allocate *them* computer equipment, or bandwidth, or electricity, they just need the same access to them that everyone else has, and nobody’s demanding equal access to posting information on consumer laptops.

    My final point for now is this: platforms on the internet are absolutely everywhere, there are virtually no material barriers to entry to starting one, certainly none that are government imposed. This right here, the comments section of Reason, is a kind of social media platform. What people want when it comes to Facebook and Twitter is *the kind of community* they’ve fostered. Their tools and software are part of it, yes, but so too are their content and moderation policies. A certain kind of speech environment is what they’re selling. If enough people wanted an unregulated model, it would exist. It doesn’t (or at least isn’t popular) not because the government wont allocate it spectrum, or because Facebook or Twitter or Reddit or YouTube or TikTok or Snapchat or Linkedin or any of the others use some imagined monopoly power to crush that alternative, it’s because people aren’t attracted to it. You say social media isn’t consumed as a coherent product, and that it’s not sold for its viewpoint, but that’s simply false. The outer bounds of their product and their viewpoint on what their product stands for may be far wider than a newspaper, but it is not infinitely so. Hence, the moderation.

    1. Indeed. The closest thing I can think of to an unmoderated platform is something like 4chan or 8chan. I don’t know how popular those truly are, though I imagine a lot of extremists like it there. I can’t imagine compelling Twitter or Facebook to host stuff like that just because it’s speech.

      1. The funny thing is, even 4chan moderated for content, and not just illegal content (though I don’t know if 8chan or its successor, 8kun, do).

    2. At some point, you might give up and fight fire with fire.

      These are the people who wanted common carrier status for ISPs to forbid them from advantages to their own streaming services. So “the other guys” have no problem with it.

      The rest are simple games of which chunks of the internet should be subsumed. Pound the principles. Pound the law. Pound the table. Get your way. Force your will on society. Principles don’t matter.

    3. Telephone companies do not “carry” anything, either, unless you consider calls and other forms of messages to be carried. Long distance telephone companies do not require government allocations of space or equipment. And so on and so forth. Your distinctions prove too much (namely, that long-eatablished common carriers shouldn’t be considered common carriers).

      1. AT&T absolutely carries information over their telephone wires, laid down on public easements obtained by governments, in order to get the messages from point A to point B. It’s what they do.

        1. Then Facebook and YouTube carry information as well. Smaller distances perhaps….

          1. What public rights of way do they carry it over?

            1. The same public rights-of-way used by broadcasters

              1. I was unaware that Twitter was allocated broadcast spectrum. What frequencies do they use?

                1. No they don’t but you clearly know nothing about the way traffic moves over the internet via many rights of way owned my many entities including the government.

                  1. Moreover not all of those links are hard wired (though most are), and what you call the “broadcast spectrum” is just a portion of the electromagnetic spectrum that the government laid claim to control.

                  2. What makes you claim I know nothing about how the traffic moves over the internet? I’m extremely familiar with it. You’re the one under the weird impression that Twitter, in its role a social media company hosting content, is involved in that actual transportation, rather than just being a user of information transportation provided by others (as we all are). The access ISPs are like common carriers, social media sites are nothing like them. It’s like claiming you’re a common carrier because you use the mail.

                    1. No, it’s not like claiming someone is a common carrier because they use the mail.

                      It’s like claiming a company is a common carrier because people hire them to deliver mail, even if the company subcontracts part of that transit to other organizations (as the postal service sometimes does).

                    2. “It’s like claiming a company is a common carrier because people hire them to deliver mail, even if the company subcontracts part of that transit to other organizations (as the postal service sometimes does).”

                      Who hires Twitter to deliver messages? Does Twitter give you your internet connection? Are they your access ISP? How is Twitter delivering anything anymore than a user is?

    4. “My final point for now is this: platforms on the internet are absolutely everywhere, there are virtually no material barriers to entry to starting one,”

      This is somewhat of a misleading argument.

      Let’s take Fed Ex or USPS as common carriers. The counter argument there is “Anyone can take a package from one place to another, there’s no real barrier, so there shouldn’t be common carrier laws!”. Or take a Train company. “Anyone can walk on their own two feet from one place to another, there shouldn’t need to be common carrier laws!”.

      What makes Twitter, Facebook, and YouTube so effective are the network effects. The ability to have and reach a large number of people simultaenously, through the network, that offers greater efficiencies and speed of communication.

      Likewise, Fed Ex and USPS are so useful, because of their massive network effects. The have a large number of people, vehicles, and so on which allow for reaching a large number of people simultaneously. If you want to mail out a million campaign flyers to a million people, you use USPS. A “one-man shop” would take forever. Likewise, if you want to reach a large audience on the internet, you use YouTube or Twitter. A new “one man shop” simply doesn’t have the network effects in place.

      Those network effects that YouTube and Twitter have, likewise allow them to skew the information pipeline, as they wish. In addition, they allow it a certain influence to crush new competitors.

      1. My argument for not treating them as a common carrier is a separate argument from my final point. Of note, your asserted reason for why we treat common carriers as common carriers is utter nonsense. Just within the narrow class of common carriers that UPS and FedEx (USPS is quite different, what with its statutory monopoly on non-urgent letters and bulk mail and quasi-governmental status) are, aka motor carriers, there are currently 1,813,445 federally registered entities, many of whom do have nothing more than a single vehicle. And that’s just at the federal level. Of those, over half offer their services “for hire”. Over a third of those maintain operating authority to transport property for the general public. That’s over 300,000. Just at the federal level. It’s hardly about anything akin to “network effects”. It’s about transporting (in this case, interstate) on the public rights of way.

        1. So, in the analogy, I was deliberately using NATIONAL (or international) level common carriers as a comparison to NATIONAL (or international) level social media companies. As opposed to LOCAL level common carriers. There’s a reason for this.

          There is a real local level physical system. And it has effects. For example, there may only be one or two tow truck companies (which are common carriers) in a local area. Because of this, they exist as a monopoly or duopoly in the given area, and a separate “system” exists there, which allows for abuses. If the only tow truck company in town won’t tow your car because they don’t like you, you have a problem. Multiply that by a million small towns, and you see the reason.

          There is effectively real no “local” internet. There is no real independent internet for Jonesburg, AL or Anytown, IN. Facebook owns most of the local “neighborhood” groups. If there was a local internet, there may be a lot more interesting case law. But since it only exists on the national level, it should be the national level comparisons that are made.

          1. I do not see how a local tow truck would generally be considered a common carrier under the federal scheme I’m discussing.

    5. My final point for now is this: platforms on the internet are absolutely everywhere, there are virtually no material barriers to entry to starting one, certainly none that are government imposed.

      Not true. Examine what happened to Parlor with a critical eye as to exactly what happened. Multiple elements of their ecosystem turned on them. The most critical element, a way to get paid, did so as well. The reason that the critical elements of infrastructure are hard to replace is the government makes entry into those markets difficult allowing a cartel to freeze out outsiders.

      1. Parler failed for a number of reasons, most of which were related to it being a half-assed attempt, which is why, despite their promise of information privacy, scraping the locations of the posts of all their users was a trivial matter for security researchers. Amazon kicked them off for not following its policies, but there is no reason that Parler couldn’t have chosen to simply buy their own servers. I run my own public facing servers, including a service that allows third-party content, it cost me no more than the domain name registration fee, my consumer-grade internet service, and whatever my two old laptops are worth. Everyone with a connection to the internet can visit them if they wish.

        But from the sounds of it, your real problem is with discrimination by payment processors, which absolutely do have government-imposed barriers to entry. Maybe you should focus on them.

        1. … Amazon kicked them off for not following its policies, …

          Which as far as I can tell are so nebulous that anybody can be booted out of a contract at any time by Amazon. Probably not something the government should be trying to solve, but a demonstration that you are an utter fool if you are running a business on Amazon AWS because you believe it is “reliable”.

          Maybe you should focus on them.

          Indeed. That was the entire point of the post. The tweeks to common carrier law are not going to solve the problem. You need to take dead aim at Operation Chokepoint and the whole mess of government malfeasance that even makes this possible.

          1. While I agree with you that people ought to have access to payment processing (particularly given the substantial involvement of the government in the core payment processing systems of the ACH Network and SWIFT), we’re going to run smack into that great destroyer of civil liberties known as the war on drugs.

            1. We don’t have a free market, so the entire paradigm based on that is outmoded. The Federal Reserve buys corporate bonds and equities now. Payment processors routinely snuff out anyone whose politics they don’t like. We have an oligarchy of crony capitalists.

    6. But they do carry information, and they carry it on pieces of silicon they own just as the telephone companies carry it on copper wires they own.

      1. They do not carry information any more than you do. They process it and they store it. They, like you, hire someone else to transport their responses to the messages you send to them. A telephone or cable company, acting as an access ISP, does the transporting, and does it over wires laid on easements they received from the government, or radio spectrum they’re allocated by the government. You may not understand how social media or the internet works, but your actual expectations of what Twitter will and wont do are evident, you don’t expect them to actually transport the information from you and then back to you, which is why you hire an access ISP to do that.

    7. You may be confused about how social media works. Nobody is ever required to see anything they don’t want to see (except ads). You determine who your friends are, who you follow, and what content you choose to see.

  11. While you conclude that social-media providers censorship users or original posts is not protected by the First Amendment, but censorship of comments might be protected by the First Amendment. How about retweets and links to original posts?

    I ask because what strikes me as being different about social media is the very loud bullhorn it provides to a speaker and I am not sure if or how First Amendment law treats that aspect of speech. Perhaps First Amendment law should take that factor into account and protect to some degree the provider’s right to limit the size of the bullhorn by (for example) allowing them to censor links and retweets.

    1. “Hi. I’m from the government. We are going to wreck a hundred billion dollars’ worth, or more, of your trillion dollar companies, by trashing section 230, or perhaps breakup, or both, unless you censor harrassment.


      “Hey! Did you note our political opponents are posting harrassing tweets? Wink wink!”

      Except there was no wink wink.

      facebook briefly refused to do this, because the people need to see what their politicians are saying, or democracy is a joke. He caught holy hell for it.

      Please people stop saying these companies are exercising free speech options of their own free will.

      The Republicans tried all this but in a vain attempt to stop it, push it in the opposite direction. Fight fire with fire.

      I despair because there should be no fire in the first place.

  12. Have not read the article so just a few preliminary comments from the post.

    1. Perhaps the term “social media platform” by itself is not the right term for this purpose because there are aome social media platforms that have (in my view) obvious common carrier characteristics and others that simply don’t. This blog, for example, is clearly a social media platform but doesn’t have the attributes of a common carrier. One thing needed is size; tiny platforms don’t really matter. A second thing needed is use for conveying ordinary messages of the sort that might once have been conveyed by mail, telegraph, or telephone. Facebook clearly does this. But not all aspects of Facebook do. A possible third thing is generality of function. This blog is highly specialized. Facebook, Twitter, etc. are not. I see borderline cases. Reddit has size and generality without private messaging.

    2. One has to take care in the definitions. I suspect a tech giant like Facebook could divide itself into ten thousand or a hundred thousand nominally separate but seamlessly integrated companies, each below a size threshold, and each specializing in some different subject, without the slightest change in the way anything appears to the user. That is, one has to think about the ways a tech giant could reorganize itself to evade a common carrier law.

  13. Here is a solution: have all social media ban for a period of 90 days all left wing speech the right disagrees with and see how they feel about it at the end of the trial period. I think we might see some bipartisan support develop….

    1. Whining, ankle-nipping, grievance-consumed, stale-thinking clingers are among my favorite culture war casualties.

      And the core of the Volokh Conspiracy.

  14. I was quite surprised by [1] and [2]. I had assumed the Post Office was formed (and put in our Constitution!) for the same reasons as Britain’s Royal Mail, whose founding charter is quite explicit about using its monopoly power to block and/or spy on those who would use the mail to exchange obscene or seditious writings.

    In fact, were this not true, then I cannot see how US courts allow the Postal Service to have and enforce the Private Express Laws that put Spooner’s company out of business.

    1. jdgalt1: In the 1800s and early 1900s, the Post Office was indeed seen as permitted to refuse to carry speech that it chose not to carry (even speech that was otherwise constitutionally protected). But starting with the 1940s and then especially clearly in Lamont v. Postmaster General (1965), the Court rejected such censorship power on the post office’s part.

  15. A problem with beating the legal bushes for analogies is that you end up arguing by analogy—an always-suspect method, except in courtrooms, where it is prized. Why do that in this instance?

    Social media “platforms,” are—and have always been—examples among a long-recognized class of businesses: publishers. Whatever analogy anyone can offer, it will prove more distant from the mark than the obvious alternative of accepting that social media “platforms” are publishers, and ought to be governed by legal standards customarily applied to that class of activity.

    It is foolish to suppose some policy model other than publishing is a choice freely available, to be made at pleasure. To make legal policy based on such a presumption, in the hope that doing so will alleviate this or that objection to “platform” publishing practices, will just bring more trouble.

    You might get away with that as a matter of philosophical rumination. As a matter of law, the analogies will kick in with a vengeance.

    And those analogies will not be the ones loosely applied across disparate classes of businesses—like the ones we see in online ruminations. They will instead be courtroom analogies, cited to prove the essential identity of institutions in one business class—the publishing class—to which media companies of all kinds in fact belong, by virtue of their shared methods of doing business.

    To make laws for Facebook—to obligate it as a common carrier, for instance—would fairly quickly, by process of closely-reasoned legal analogy, impose similar obligations on book publishers, newspapers, cable news companies, and a host of other publishing types. Lawyers will insist that all those can be shown to practice business alike, to use the same methods, and thus must be treated alike before the law. Press freedom would thus be heavily burdened.

    No one doubts that policy problems, political problems, social disruptions, and other baneful outcomes have been delivered by social media business practices and editorial choices. Those were not caused because social media publishers were inherently different than other publishers. They were caused because Congress mistakenly elected to award to online publishers, and not to others, a special privilege to avoid civil damages.

    That difference created such a potent business advantage that the online class swiftly became nationally dominant, to a degree never before seen in the publishing industry. All the troubles and tangles folks try now to fix flow directly from that one unfortunate congressional choice—to treat some members of an essentially identical business class unequally with the others. It was a choice which unexpectedly empowered online publishers to write their own rules, and to seize disproportionate influence.

    Putting all the members of the publishing class of businesses back under the same umbrella, by applying legal standards alike to all of them, is the wisest route out of the present contentious media crisis. Governed alike they will compete alike, and keep each other in check.

    Adding further arbitrary legal differentiations among closely similar businesses—businesses which continuously prove their shared identity by competition in the same publishing marketplace, and by non-competition in, for instance, the common carrier marketplace—is the opposite of wisdom. Doing that will lead to more trouble, not less.

    1. If they are publishers, government just not unmoor that role from legal responsibility and liability. It did that through Section 230. I suppose we could repeal Section 230, but treating social media oligopolists as common carriers seems less likely to have pernicious effects.

      1. I doubt it. Common Carrier is a terrible regulation in just about every instance. Amending Section 230 to force Facebook et al to have accountability if they want its safe harbor protections is a much easier fix. For example, require certain user-friendly standards be implemented in user agreements such as notice, appeal, right to court review, and a requirement that standards be clear, and set out at the outset of a user’s recruitment. A site shouldn’t be able to get large by catering to, say, hardcore gamers only to, when they have profited from that relationship, discard them in favor of a demo they now find better via mass bans.

        1. Amending Section 230 to force Facebook et al to have accountability if they want its safe harbor protections

          This is gibberish. This is not “amending” § 230; this is repealing it. The entire point of § 230 is to eliminate this condition.

          A site shouldn’t be able to get large by catering to, say, hardcore gamers only to, when they have profited from that relationship, discard them in favor of a demo they now find better via mass bans.

          Why the hell not?

          1. Whoops, sorry about screwing up the formatting.

          2. “This is gibberish. This is not “amending” § 230; this is repealing it. The entire point of § 230 is to eliminate this condition. ”

            No, the point of 230 is to let them make honest mistakes via understaffing, or simply just missing something as oversight. When a user posts something defamatory. The point is not to allow random moderation and unclear and unfair EULAs.

            “Why the hell not?”
            Because of network effects being almost 100% of the value of any social network, original users, particularly power users are as important to the success of the company as the CEO, angel investors, etc. Without them they are nothing. They should frankly get stock options.

            1. No, the point of 230 is to let them make honest mistakes via understaffing, or simply just missing something as oversight. When a user posts something defamatory.

              There is no textual or historical evidence for this claim.

              The point is not to allow random moderation and unclear and unfair EULAs.

              I don’t know what an “unfair” EULA is, but the whole point of 230 was that they would not have to justify their moderation decisions.

              Because of network effects being almost 100% of the value of any social network, original users, particularly power users are as important to the success of the company as the CEO, angel investors, etc. Without them they are nothing. They should frankly get stock options.

              Oh, so this was a troll comment. Nevermind.

    2. SL starts with the same false premise as always and slippery slopes it into trouble for Free Press.

      Sorry SL, if people still valued your opinion, you wouldn’t have to be here arguing this same tired and forced argument every day.

  16. …the essential identity of institutions in one business class—the publishing class—to which media companies of all kinds in fact belong, by virtue of their shared methods of doing business.

    Online platforms, whether Facebook, Twitter, Reddit, Snapchat, or message boards like the comments section of articles here, are not doing business the same way as newspapers, radio and TV stations, or book publishers. The act of publishing inherently involves editorial choices being made prior to the dissemination of the published material. Moderation of user-generated content is not the same as such full editorial control. (Even live broadcasts give the producers some control over what is seen and heard. They have choices over who is interviewed, how to respond to what an interviewee says, what to point their cameras at, etc.)

    In my understanding, Section 230 was legislated because online platforms like message boards (the somewhat popular predecessors to current social media that existed at the time) are in an in-between space that didn’t fit any legal standards available. An online space where users can post messages visible to all other users is not publishing those messages in the same sense that print media is published. Nor are these messages being broadcast over airwaves within a limited spectrum (as whey_standard points out above). The number of users is also often vastly larger than any ability of the platform owner to moderate the content in real time. The company would generally have to rely on other users to flag comments that they think should be subject to moderation, and material posted by users that the company would not want on their platform could persist for a considerable amount of time before being taken down.

    What would it mean to treat online platforms the same as all other media companies, in terms of liability? Would Twitter potentially be liable for defamation if it didn’t remove a tweet from a user fast enough when the target of an allegedly defamatory statement wants to them as well as the person that tweeted it? Would Twitter have to avoid almost all moderation in order to avoid such liability? How would that apply to smaller platforms? Or to outlets that are definitely publishers, like Reason and the Volokh Conspiracy themselves, but that also host the comments of readers?

    You have some good points, Stephen, and gave us some important things to consider. But it seems like you are arguing to do away with Section 230 entirely. And you seem to base this on an assertion that social media platforms are doing business in the same manner as publishers, which I do not believe is accurate.

    1. SL is jealous that he is no longer one of the “top men” responsible for deciding what is ok to publish.

      1. Nonsense, Vinni. If I owned a publication, I would insist that I be the one to decide what to publish in my publication. I don’t at all want to decide what goes in your publication. That’s up to you. But if you want government to intervene, and force me to publish what you want me to publish, then you are an enemy of press freedom. It’s that simple.

        I do suspect that many of the hostile comments I get on this subject come from people who are enemies of press freedom. You have been pretty hostile. Is that you?

        1. Well, since this isn’t about the rights of publishers, you’re just plain wrong, again. As usual.

          Facebook is not “The Press”. Repeat this, until it sinks in.

    2. Stephen is partisan, Because big tech is censoring conservatives he’s searching for a reason why it’s Ok. If the reverse were true he’d change sides.

      Obviously the limited liability was meant too shield companies form lawsuits and allow free speech. But the obvious is not anymore

      1. Big tech is not censoring conservatives. It does occasionally decline to be associated with blatant bigots, repulsive liars, delusional kooks, and advocates of violence.

        Clingers hardest hit. But not all conservatives are lying, insurrectionist, delusional, violent bigots. Not nearly.

      2. Wreckingball, you, along with plenty of others, seem to misunderstand the publishing principles involved. We hear plenty of complaints about, “censorship.” That is not how folks customarily think about editing by a private publisher. Problem is, because the business models used by social media companies mostly do without editing, it looks more sinister when something they initially allowed published gets taken down. But they remain private publishers, and the constitution gives them a right to not publish anything they want to not-publish. When they take something down after publishing it, they really are only exercising their constitutional right to edit. It’s a terrible look, and a great argument for getting rid of Section 230, which enabled the whole publishing-without-editing business model, but it isn’t, “censorship.”

        1. That is not how folks customarily think about editing by a private publisher

          And, since we aren’t dealing with private publishers, your broken record screeds are entirely irrelevant.

    3. JasonT20, thanks for a thoughtful comment. On this subject, this blog (and the nation) could use many more of those than it gets.

      Your arguments are not without basis, but your points about inherent characteristics of publishing put the focus in the wrong place. Important as it is to society, the editorial function has never been the defining attribute of publishing. The history of publishing is full of unedited materials, and unedited contributions to opinion—some as exalted as Tom Paine’s Common Sense, others as utilitarian and forgettable as a local all-advertising freebie in a resort area.

      A list of characteristics which does define a considerable part of the publishing class of business activity includes these 3 points:

      1. A publisher assembles useful or expressive content to attract an audience from among the public at large.

      2. The publisher curates the audience by adjustments to its content, with an eye to making the audience marketable to particular would-be advertisers who want to reach with their own messages the kinds of people the curation attracts.

      3. The publisher monetizes that audience by selling access to it to the interested advertisers.

      That set of characteristics by no means exhausts the entire publishing class. But it is a widely-practiced set of characteristics which largely excludes non-publishing types of businesses. Common carriers, for instance, do not do those things. And almost any business which does do those things is clearly a publisher. So that list of characteristics, where it applies, usefully functions as a practical guide to sort out publishers from the others.

      Note that when you check their business models against the list, social media companies typically check every box. Try to think of any other class of businesses which does that.

      A final proof that social media companies are publishers becomes evident when you look at their market effects. It is almost exclusively the class of other publishers which have suffered the effects of social media companies’ competitive success. A massive transfer of advertising revenue from newspapers and other classes of publisher to social media companies is a defining characteristic of the current crisis. Side effects have included destruction (for want of supporting revenue) of a considerable fraction of the nation’s news-gathering capacity. The nation’s public life has suffered accordingly.

      If competition from social media companies is the principal cause of so much trouble among other publishers, it can hardly be the case that social media companies are not publishers themselves. The market they affect defines the class of business activity they practice.

      You are correct that I advocate complete repeal of Section 230. And I do not doubt that supporters of Section 230 are probably correct when they predict that doing that would notably shrink the scope, influence, and economic impacts of the giant social media companies.

      That does not bother me at all. Most of the well-justified complaints about social media company dominance would be ameliorated if that happened.

      Where I differ with you is that I think your assumption is mistaken that such changes in the social media presence online would equate to an important loss either to publishing, or to the ability of most would-be commenters to get published online. The key fact to keep in mind is that online publishing is vastly more economically efficient than ink-on-paper publishing, and somewhat more efficient than other kinds, including broadcasting.

      Back in the day, the task of informing Los Angeles of Sunday’s news fell largely—extremely largely—to the Los Angeles Times. To do it, the publisher collated, manufactured, and distributed weekly a paper artifact with the approximate aggregate weight of a navy cruiser.

      Think what it means economically to avoid the expense of that gigantic effort. The savings are sufficient to support a many-fold increase in editorial effort. Thus, it is not true, or even reasonable to suppose, that a prior editing requirement would wipe out all forums with online comments. For example, the New York Times already reviews editorially and publishes daily comments numbering in the thousands—and doubtless does so at the cost of only a negligible fraction of its budget. Get rid of Section 230, and its gigantic ad revenue siphon, and there would once again be economic space in the publishing marketplace for competition among smaller publishers who could afford to edit.

      Editing to avoid civil liability is by no means as complicated as many Section 230 supporters insist it is. Lawyers are almost never needed. The key is that most comments are readily recognized as liability-free opinion, which requires no more effort to approve than to read. Questionable but potentially interesting assertions of fact can always be set aside for further investigation. A single editor can typically approve more than a thousand comments a day. Even small publications could afford to hire editors to review as many comments as their readership is ever likely to generate.

      Other advantages of a publishing market free from social media giantism would include ease of entry for would-be competitors. In today’s comment wars, you often see folks dismissed who complain about social media censorship. They get told to start their own platforms if they don’t like the choices already on offer. They quite rightly treat that as a taunt, rather than as realistic advice. But with a renewed diverse and profuse publishing ecosystem—with more money available to smaller publishers from advertisers—starting new publishing enterprises would once again become a realistic possibility.

      As a matter of history, the only safe harbor ever found for press freedom has been creation of a publishing marketplace made up of thousands of diverse private publishers, representing every conceivable thread of opinion, and all struggling against the others to distinguish themselves by the unique characteristics of their offerings, and the quality of their content. For generations, this nation enjoyed just such a publishing marketplace, and the result was regarded world-wide as an ornament of civilization.

      Now, Section 230 has put that press freedom seriously at risk. When a free speech fundamentalist as zealous as Eugene Volokh calls for government intervention in publishing, it is a signal that the time has arrived for a root-and-branch re-evaluation. Repeal Section 230 and start over.

      1. Now, Section 230 has put that press freedom seriously at risk.

        Section 230 does not affect the press. Never has. Never will. Because, get this, Section 230 is not about private publishers, or the press. It’s about you, me, EV, sharing our opinions and thoughts in a, purportedly, public space (a tangled web of publicly and privately owned), and letting the owners of that space not be liable for your, my, or EV’s, speech, because they aren’t the publishers (repeat this until you get it), until they decide to act like publishers, at which time, they should no longer maintain special protections from liability for affirmatively deciding what speech they will allow to be published.

        Notice how, at no time, does anything like NYT or CNN come into play here (their comment sections, nominally, but that’s entirely separate from their primary business of being “the press”).

  17. A waffley solution looking for a non-existent problem. . . .

    1. This is a prominent Federalist Society project. They are putting several of their top men on it.

  18. “People read the Times in part precisely because of its editorial judgment, its ability to winnow the good and sensible views out of the vast mass of nonsense and folly…” Professor Volokh, that assumes facts, not in evidence.

  19. They are common carriers. They have limited liability in exchange for not censoring. But that last part seems to have been lost.

    1. They transport nothing, not even information, and the most famous liability shield they have (47 USC 230(c)) *is explicitly for “censoring”*.

      1. Uh nope-er-roo buddy. They don’t have to worry about censoring because they are shielded. Thanks for playing

        1. I’m not sure what you think you’re responding to, I didn’t claim they weren’t shielded from liability for what others post, I asserted that they are explicitly shielded for “censoring” activity, despite your claim that they somehow agree not to censor. 47 USC 230(c)(2) quite explicitly says “No provider or user of an interactive computer service shall be held liable on account of any action voluntarily taken in good faith to restrict access to or availability of material that the provider or user considers to be obscene, lewd, lascivious, filthy, excessively violent, harassing, or otherwise objectionable, whether or not such material is constitutionally protected”.

          If they agreed not to censor, that provision wouldn’t make sense.

      2. “They transport nothing, not even information,”
        Nonsense. But then you do not know the technical meaning of information. Wie schade.

        1. They don’t. I’m familiar with what information is, you appear to not be familiar with what “transport” means.

    2. “They have limited liability in exchange for not censoring.”

      That is literally the opposite of what Section 230 does. It explicitly gives them limited liability WHEN THEY CENSOR.

  20. But I’m expecting some of this to go in reverse. The Post Office may start censoring what it delivers and who it delivers too because of racism and info that criticizes the regime. You know we wouldn’t want “misinformation” going around willy nilly.

    Utilities may censor customers. Remember when dear leader Newsom was going to (maybe he did) cut off utilities to those who disobeyed his unconstitutional Covid orders.

    And the courts like trained seals will clap and say of course this is allowed because we’re in an “emergency” (fill-in the blank, Covid, racism, climate change)

  21. A slightly but by no means entirely off topic comment.

    There has always been an elite contrilling who gets access to the public soapbox. But up until perhaps a generation ago, that elite has to some extent had the values and culture of an aristocracy. Access decisions were largely in the hands of a group of professionals – academics, journalists, publishing houses, and others – which generally didn’t, couldn’t, and didn’t expect to make really large sums of money, and were motivated by an actual interest in the content rather than by technology, money, or power for its own sake. Hollywood, where you really could make large sums, was something of an exception, but it too tended to cultivate the aura of an aristocracy, and mostly at least played at keeping to its values.

    So one thing entities like Facebook represents is a control of political discourse by an elite whose fundamental interests have nothing to do with academic, journalistic, or artistic interests in their subjects, have no concern for ideals like justice, truth, or beauty, and come from a background which more or less values power for its own sake. Technology, after, is considered wiz-bang and valued not for its artistry but because of its capacity for power.

    America hasn’t really faced being run by an elite like that. Past elites had also been power-hungry, but at least pretended the aristocratic ideals of a society where it was once considered bad form, selfishness, to be seen apearing to seek an office like the presidency oneself. One got humbly drafted by ones colleagues.

    In this respect, there’s a remarkable similarity between the culture of politicians like Trump and that of the leaders of the tech giants, despite their many glaring differences. Both from a culture whose currency deals much more directly and overtly in power for its own sake than in the past. Both see the values of the old artistocrat culture, including the concept of having responsibilities for others, as being more or less for little people, and not for people like themselves who seek to be truly great. Truly great people are unencumbered by having to consider others’ needs. And they are unencumbered by seeing what they ostensibly do as having any value in itself, or as serving any other purpose than helping them get what they want.

    Even a world where there is no art, only propanda, reflects a shadow of the elite culture. Here there is neither art nor propaganda, but only marketing, content existing only as a marketing opportunity.

    We’ve never lived in a world like that before.

    In much the way a politician would once have been ashamed to admit he was seeking office solely for personal gain, a generation or two ago a communications company would be ashamed to say that its sole purpose was to curate an audience for advertisers. It would at least pretend that it had some sort of public purpose – facilitate communications, teach, entertain, do art. It would hesitate to judge content solely by its potential for product placement.

    No more.

  22. To further elaborate on the limits of the common carrier model:

    A large number of the controversies around the social media platforms aren’t about whether or not the platforms will host content or not, but what conditions they place on the hosting.

    A common complaint that conservatives have is with “fact checks” that the platforms do alongside certain content. Even if the platforms were common carriers, it seems like it would be a clear violation of their free speech rights to say that they’re not even allowed to add their own commentary alongside speech that they’re carrying.

    Another complaint that we see a lot is that the platforms will refuse to run certain advertisements (Marsha Blackburn was making a big fuss about this). But outside of some government-run advertising platforms (e.g., advertising hosted by transit agencies), it’s commonplace for broadcasters and others who host advertising to have all sorts of restrictions on the type of content they’re willing to allow in advertisements.

    Or look at the PragerU lawsuit against YouTube. Their complaint was basically twofold: 1) that YouTube age-restricted some of their content, and 2) YouTube demonetized some content. In this case, PragerU was arguing that not only must YouTube carry their content, but they shouldn’t be allowed to decide that some content was inappropriate for children or that YouTube should be forced to put advertisements next to their content and share the profits with PragerU. Both of these positions seem very far away from the sort of thing that the common carrier framework is going to encompass.

    1. Just for a minor clarification, it wasn’t even that YouTube actually age restricted anything. It was that YouTube flagged some of PragerU’s content such that individual users who restricted their own viewing wouldn’t see it. So PragerU was essentially saying that it was unlawful to rate a movie PG-13 because some people would choose not to go to PG-13 movies.

      1. That’s incorrect. YouTube makes the age restricted videos completely unviewable by default. The only way you can view the video is to create an account with YouTube, confirm that you are of age, sign in with your credentials, and view the video as an identified account holder.

        Presumably this makes it easier to track those who are consuming wrongthink.

        1. That’s not really right. Restricted Mode is off by default. You are conflating Restricted Mode — which is what PragerU sued about — with age restrictions.

          “Restricted Mode is an optional setting that has been available since 2010. It’s used by a small subset of users, such as libraries, schools, and public institutions, who choose to have a more limited viewing experience on YouTube. Restricted Mode is turned off for viewers by default. Learn how to turn Restricted Mode on or off.”

          (You can test this yourself by opening YouTube in an incognito window. You’ll see that RM is off, and that you can watch videos that have been flagged as restricted.)

          “Is Restricted Mode the same as age-gating or age-restricting videos?
          No, a video that is not available in Restricted Mode is not necessarily age-restricted.
          Age restricted content may not be appropriate for all audiences. Content will not be made visible to users who are:
          Signed out
          Are under 18 years of age
          Have Restricted Mode turned on
          Learn more about age-restricted content.”

          So, yes, actually-age-restricted videos are not viewable unless one logs in. Age restrictions are pretty much for exactly the purpose that moral scold conservatives would want them to be for. Not for political views, but profanity, sexually suggestive stuff, drug use, physically dangerous things that kids Shouldn’t Try At Home.

          1. “Signed out
            Are under 18 years of age
            Have Restricted Mode turned on”

            Odd way to refute ML’s comment that “The only way you can view the video is to create an account with YouTube, confirm that you are of age, sign in with your credentials”

            1. I’m sorry if nuance is tough for you.

          2. Point taken. So YouTube is restricting PragerU videos from being seen by those in libraries, schools, public institutions, homes, families and others who use restricted mode to limit viewing to age appropriate content.

            Of course, there is nothing age inappropriate about those videos. That’s just a pretext for engaging in politically motivated censorship (censorship that they have a right to engage in, I assume).

            Some might say that PragerU’s heavily neocon viewpoints are objectionable and not to be trusted, but that’s not age-related.

            Regardless, there is a lot of reporting that YouTube initially very early on applied the “actual age restricted” restriction to some PragerU videos, and that they have done likewise for other political content. Not sure how much there is to that or how much difference it really makes.

Please to post comments