Listen & Download
The recent concurrence by Justice Thomas in Biden v. Knight First Amendment Institute has raised new questions about how we might think about restrictions on speech and debate on social media. Where private, concentrated control over online content and platforms exists, can a solution be found in doctrines that limit the right of a private company to exclude?
While there is historical precedent for regulating communications networks in a similar manner as traditional common carriers, are social media platforms best understood as communications networks that “carry” information from one user to another? Or have they created a business model built more on “curated” speech that to some degree reflects their own expressive interest in acceptable debate and discussion? And how should we think about possible state regulatory efforts to regulate private companies in this way?
- Adam Candeub, Professor of Law, Michigan State University
- Geoffrey A. Manne, President and Founder, International Center for Law and Economics
- Olivier Sylvain, Professor of Law, Fordham University
- Charles M. Miller, Deputy Chief Counsel, Ohio Attorney General’s Office
- Moderator: Hon. Gregory G. Katsas, Judge, United States Court of Appeals, D.C. Circuit
* * * * *
As always, the Federalist Society takes no position on particular legal or public policy issues; all expressions of opinion are those of the speaker.
Dean Reuter: Welcome to Teleforum, a podcast of The Federalist Society’s Practice Groups. I’m Dean Reuter, Vice President, General Counsel, and Director of Practice Groups at The Federalist Society. For exclusive access to live recordings of practice group teleforum calls, become a Federalist Society member today at fedsoc.org.
Alida Cass: Welcome to The Federalist Society’s showcase discussion series on free speech and social media. This afternoon, June 25, we will discuss the question of common carrier regulation of digital platforms. I’m Alida Cass, Vice President for Strategic Initiatives at The Federalist Society and Director of the Freedom of Thought project. This project is a new initiative of The Federalist Society to consider emerging challenges to freedoms of thought, speech, and conscience. As always, please note that all expressions of opinion are those of the experts on today’s event.
We are fortunate to have Judge Katsas with us to moderate this discussion series. Judge Katsas was appointed to the D.C. Circuit in December 2017. After graduating from Harvard Law School, he served as a clerk to Judge Edward Becker on the Third Circuit and to Justice Clarence Thomas on the Supreme Court. For 16 years, he practiced at Jones Day where he specialized in appellate and complex civil litigation. He also served as Assistant Attorney General and as Deputy Counsel to the President. Before joining the bench, Judge Katsas argued more than 75 appeals, including 3 cases at the Supreme Court.
I’m going to turn this over to Judge Katsas to introduce our panelists. After our speakers give their opening remarks, we will turn to audience questions, time permitting. If you have a question, please enter it in the Q&A feature at the bottom of your screen.
With that, thank you for being with us today. Judge Katsas, the floor is yours.
Hon. Gregory Katsas: Welcome, everyone, to this third of our six part series on free speech and social media. These panels are a rollout of The Federalist Society’s Freedom of Thought project. I’ve mentioned it a little before on the prior panels. If you have any interest in learning more, please check it out on the website.
Our last panel focused on Section 230 of the Communications Decency Act. Just to recap a little bit, that statue provides two distinct protections for internet platforms. The first, set forth in 230(c)(1), tracks a rule that was historically associated with common carriers like telephones and telegraphs, which is that the platforms can’t be treated as publishers and can’t be held liable for third party speech that they do carry on their platforms.
The second rule in Section 230(c)(2) provides that platforms can’t be held liable for blocking third-party speech that they find objectionable. And that rule tracks First Amendment protections that are normally associated with speakers and editors, folks like newspapers and parade organizers. And it extends even if the government itself couldn’t prohibit the speech at issue.
So 230 provides an immunity, but that begs the question of immunity from what? Our next panel will explore antitrust law as a source of regulation or liability.
Today, we’re going to explore another proposal which was recently made by Justice Thomas in his concurrence in the cert denial in the Knight First Amendment case, and that is to treat different kinds of interactive computer services, whether they’re social media platforms or search engines or whatever, as common carriers under state law. And that invokes the historic treatment of entities that have been conduits for speech like telephones and telegraphs or places that are public accommodations like inns and taverns. The Thomas concurrence relied pretty heavily on a law review article by Professor Adam Candeub, and it provoked a lawsuit by the State of Ohio to have Google Search declared a common carrier under Ohio law.
We are very pleased to have with us today both Professor Candeub and Chip Miller, who is Ohio’s lead counsel in that lawsuit. And together, they are obviously on the cutting edge of both the academic and the real world push to apply this common carrier model to modern internet platforms. We also have the opposing viewpoint very ably represented on this panel. We have Geoff Manne and Olivier Sylvain, who are deeply skeptical of this push and whose views are informed by their own very deep experience in academia, and as consultants, and in private practice. So we’re hoping for a very lively debate informed by a robust and healthy diversity of viewpoints.
On to traditional introductions. Adam Candeub is a professor at the Michigan State University College of Law and a senior fellow at the Center for Renewing America. He began his career as a law clerk to Chief Judge J. Clifford Wallace on the 9th Circuit and as an associate at Jones Day. In government, he served as a lawyer in the Federal Communications Commission and as a Deputy Assistant Secretary and Acting Assistant Secretary at the National Telecommunications and Information Administration in the Department of Commerce. He’s widely published in the areas of communications law and antitrust.
Geoff Manne is the president and founder of the International Center for Law and Economics, a research center based in Portland, Oregon. He’s a distinguished fellow at the Northwestern Law School’s Searle Center on Law, Regulation & Economic Growth. In 2017, he was appointed to the FCC’s Broadband Deployment Advisory Committee, and he recently served on the FCC’s Consumer Advisory Committee. He’s an expert in law and economics, particularly in areas of antitrust, telecommunications, consumer protection, intellectual property, and technology. He served as a law clerk for Morris Arnold on the Eight Circuit, and he practiced at Latham & Watkins.
Charles Miller is counsel to the Ohio Attorney General, and so he represents various Ohio officers and agencies in litigation. His cases have touched on major matters involving state election laws, tax subsidies, and utility ratemaking. And as I mentioned, he now serves as lead counsel in Ohio v. Google LLC, which seeks to declare that Google Search is a common carrier under Ohio law. He previously has served as a judge on the Ohio First District Court of Appeals, as a litigation partner with Keating, Meuthing & Klekamp—I hope I got that right—and as a law clerk to Justice Maureen O’Connor at the Ohio Supreme Court.
Olivier Sylvain is a Professor of Law at Fordham University. He teaches administrative law, legislation and regulation, and information law, including data protection. His scholarship is focused on communications law, including artificial intelligence and the Communications Decency Act. At Fordham, he’s the director of the McGannon Center for Communications Research and the academic director of the Center for Law and Information Policy. He is the president of the New York affiliate of the ACLU, and he sits on the advisory committee for the Cyber Civil Rights Initiative. Before entering academia, he practiced law at Jenner & Block.
We’ve scheduled these panels for an hour, but with one extra panelist on this one, we’re hoping to run a little bit over that. Without further ado, let me turn the floor over to Professor Candeub.
Prof. Adam Candeub: Well, thank you so much, Judge Katsas. And I’m looking forward to an interesting discussion. For most of our history, government, whether at the federal or state level, has regulated dominant utilities as well as communications and transportation systems to ensure they do not discriminate in providing service to their users. Varying significantly over time, over different types of industries, and in different jurisdictions, this wide body of law is generally known as common carriage. As Judge Katsas pointed out, many now advocate, perhaps most notably Justice Thomas, that imposing common carriage regulation on the dominant internet platforms may be a good idea.
Just briefly going over some sort of generic objections to this proposal and responses I think might be a good way to be a springboard for further discussion. So from a legal perspective, perhaps the most important, the strongest objection to imposing common carriage type regulation on the platforms would view the regulation as violating the First Amendment because it would be a government imposition upon the network’s editorial control and judgement.
This view was expressed quite powerfully in a dissent in United States Telecommunications Association v. FCC, the dissent by then Judge Kavanaugh when he was on the D.C. Circuit. And this case involved the FCC’s network neutrality rules, which are essentially a nondiscriminatory mandate not on social media platforms but on broadband internet service providers. And Justice, then Judge Kavanaugh said that in dissent that this requirement that broadband service providers treat all providers of content equally without discrimination violated the First Amendment because it would impose upon the ability of broadband service providers to decide what goes on their servers and decide what sort of content they want to transmit.
But I think it’s fair to say that, at least right now, Justice Kavanaugh’s view is in the minority and, in fact, has to overcome really centuries of precedent, not just old precedent from the 17th and 18th century, but really more recent statements from the Supreme Court. As mentioned, well, as implicated earlier, for centuries, we’ve had common carriage for telegraphs, for telephones, for satellites, for virtually every type of communications technology. And we’ve never had First Amendment objections, or successful ones, to them, so this will be a new departure.
But besides ancient history, the Supreme Court in cases like Denver Area Educational Telecommunications Consortium has always viewed that the rights of speakers, or First Amendment rights of speakers tend to dominate as compared to those of platforms and transmitters of speech, and the rights of businesses and other private entities to exclude individuals for expressive reasons is limited. Cases like Pruneyard Shopping Center, which some wise court savants may roll their eyes at, stated that unequivocally, but also more recent cases which I think are much stronger, cases like Turner Broadcasting and Rumsfeld v. FAIR clearly show that businesses and private entities don’t have some sort of basic integral right to exclude other people for expressive purposes.
Moving away from the law, more from practical policy issues, many argue that social media is not a simple provision like electricity or phone service but rather is by its nature, by its essence, a curated experience that requires platforms to discriminate for and against certain types of content, and that’s what people, in fact, want when they go on to Facebook.
I guess my response would be common carriage is a very flexible set of rules. It’s not one thing. It’s not ratemaking. If you go back to the 15th century, it really has a basis in feudal law and obligations of services available to everybody. In the early 19th century and later 19th century, it takes on a more requirement of fair dealing and openness to all.
And it really wasn’t until the 20th century that it sort of merged with public utility regulation and had this very sort of dirigiste, very rigorous, very government involved approach so that, therefore, you could declare under common carriage regulations much lighter types of regulations, such as imposing strict disclosure of regimes and requirements on the platforms so that they have to tell us how they’re promoting or hiding content.
Or one solution, which I think would be the happiest of all arrangements, is simply to take the blocking and screening and content moderation function away from the centralized platform and give it more to the users. That seems to be the best of both worlds.
I’m running out of time, so I just want to say in response to economic arguments, people are concerned that this sort of regulation is inherently inefficient, that it will impose costs upon these wonderful, innovative aspects of our technology, of our economy. But I would argue that democracies require social institutions like robust public discourse in order to survive. There’s a very strong empirical literature showing that free speech is important in lowering levels of government corruption and increasing levels of government trust.
And let’s face reality. Recent events like Twitter’s treatment of the New York Post’s stories on Hunter Biden, to right now, YouTube’s ban on discussing COVID treatments like ivermectin, decisions which really could have cost hundreds of thousands of lives after all the dust has cleared, really show that the platforms are not committed to a robust public discourse.
So I would argue that whatever horrific inefficiencies common carriage type regulations might impose upon the economy, we have to balance them against the institutions that we seek to preserve. And I hope I’m under time.
Hon. Gregory Katsas: You are under. That’s great.
Prof. Adam Candeub: Good.
Hon. Gregory Katsas: Geoff?
Geoffrey Manne: Well, I’ll use Adam’s extra time, then.
Prof. Adam Candeub: A libertarian approach, right?
Geoffrey Manne: Yeah, exactly, until someone turns me off.
Hon. Gregory Katsas: You can owe him for the extra time. Go ahead.
Geoffrey Manne: I wanted to just thank The Federalist Society and my fellow panelists for having me. And I also wanted to apologize for the shell motif and fan behind me. I’m not normally in such an environment, but I’m visiting my family in Florida today, so when in Florida.
Let me start where Adam left off, I think, and let’s start by talking about this idea of lacking robust public discourse. I think it’s almost — I think it’s virtually impossible to suggest that there is not robust public discourse in the online and offline worlds today. I think, if anything, the online platforms that would be the subject of common carriage or other kinds of regulation are enormously responsible for that robust public discourse.
You can always find a couple of isolated examples here and there, but they go virtually no way toward undermining the notion that there’s robust public discourse. The amount of discourse today, the ease with which people can access it, and the extent of it, and the massive competition that is out there today, the number of platforms online, the number of newspapers and broadcast outlets, it would take a month to just name them all. It’s kind of absurd to me to think that there’s any problem with robust public discourse today.
So it leads me to ask the question what exactly it is anyone wants to accomplish by imposing a common carriage standard on online platforms in particular. And it concerns me when I think I have an answer to that. I’ll certainly ask Adam to elaborate on this later. I didn’t really hear a specific outcome that he wanted other than something vague like a more robust public discourse and presumably the avoidance of the couple of examples that he mentioned.
What concerns me is that it’s hard to differentiate the push for a common carriage type regime here from what seems to me to be an objective of using the government to try to force politically preferenced speech in contexts in which the First Amendment very clearly prohibits it. That’s not to say that common carriage is inconsistent with the First Amendment. There are, of course, plenty of contexts in which it’s allowed. But when it comes to communications, it is very firmly circumscribed and in ways that I don’t think are even close to being applicable here. Again, I won’t belabor it. We can talk more about it later. But looking back at Pruneyard and Turner and FAIR, I just don’t think you have the conditions that are similar to the ones here.
And most importantly is the fact that communication is extraordinarily diverse on online platforms, and that’s by design. These platforms that are being complained about — let’s just pick Twitter as a general example, but it’s important to note that they’re all widely differentiated. The content that appears on Twitter would be utterly useless if it weren’t for moderation and the differentiation among the content that appears. To have a discourse, you have to have differing points of view, and you have these communications, these debates that happen on Twitter all the time. But obviously, Twitter could become a cesspool of all kinds of horrible content were it not for content moderation by Twitter and, obviously, the other platforms like it.
I don’t think anyone wants — I hope no one wants the platforms to be unable to engage in content moderation, which is why, again, I fear that the common carriage type proposals are an effort not to undermine the notion of content moderation but simply to impose a kind of preferred content moderation around the margins that the proponents care about. What bothers me most about this is that there are problems on these platforms. There are types of content that appear that are indeed unlawful, harmful, and potentially problematic. But they don’t strike me as the kinds of things that people who are promoting a common carriage kind of regime actually care about.
Adam’s examples went to the New York Post thing and misinformation around COVID. Those are not examples of libelous communication or tortious communication. Those are examples of kinds of communication that some people would like to see more of, but, very notably, other people would like to see less of, which gets me to the next point about common carriage which is common carriage is inherently a government-enforced kind of regime.
Adam mentioned a couple of ways that he would like to see it implemented, and I want to come back to those, but — well, actually, I’ll address them right now. Those aren’t common carriage. The kinds of things that Adam mentioned, the kinds of ceding more control to users, for example, you don’t need a common carriage approach to impose those kinds of obligations on platforms. So again, I worry that — and it may be useful to do so. I don’t think I would support the exact restraints that Adam mentioned, but I do support quite a few potential constraints that go beyond what Section 230 — that would impose some liability beyond Section 230’s immunity.
But there’s a massive space between common carriage, mandating carriage of all like situated comers similarly in a nondiscriminatory fashion, and imposing some constraints or some regulations on the carriers of that content. I don’t see why a common carriage approach is in any way required to impose, for example, the kinds of obligations that are well established in the common law that have to do with a duty of care to prevent harm by the least cost avoider in certain contexts. We have myriad examples of that, none of them occurring in what would be typically considered common carriage type situations. And yet, I think they are far more applicable and relevant to the social media and online platform context.
Again, just to underscore the point, my fear is the preference for a common carriage approach is because those proposing it want to impose obligations on these platforms to carry speech in a way that the First Amendment simply doesn’t permit. They don’t like what they see, for some reason, and they’d like to see more or less of the kind of content that is currently available on these platforms. Leaving aside that all of this content is generally available in lots of other contexts, I think that is not an appropriate objective for the government to pursue, nor do I think it’s an advisable one for any of us to pursue.
I haven’t been following the time, but I assume I have a couple more minutes, right?
Hon. Gregory Katsas: About three more minutes.
Geoffrey Manne: Yeah, perfect. I wanted to talk a little bit more about what I think is the biggest problem with the application of mandatory nondiscrimination, mandatory carry kind of regime on these platforms. And that is the extent of differentiation and the extent of moderation that’s required on these platforms.
Common carriage is imposed in circumstances where you tend to see commodity-like products being transported in, if you can apply the term, commodity-like ways, a railroad transporting lumber or any other goods. I know there are some examples where it’s slightly more complicated, but I would certainly like to note that when you get to those, like cable, for example, when you get to the communications context and you start to bump up against speech issues, the kinds of constraints that are allowable under common carriage are far, far more limited.
But most importantly is this, that the kind of result you get from imposing common carriage is uniformity. And the very sine qua non of these platforms is the lack of uniformity. There is such a mismatch between the underlying logic of common carriage, and the kinds of outcomes that common carriage contemplates, and what these platforms do that I just don’t really understand how anyone thinks that it’s appropriate.
I do understand the notion that they think there’s a public interest here, and I believe that that’s correct. There are lots of circumstances in the world today, in the U.S. today, in which we have industries and products and activities that are absolutely imbued with a public concern. The food supply, education, healthcare, these are not — there is no common carriage requirement imposed on any of these, and yet, we do regulate them in ways that we think is important because we do think there are important state interests involved.
It seems to me like it’s perfectly plausible to do such a thing in this context if there is indeed a basis for doing so, but that the mismatch with a common carriage based approach to that is so stark that I find it hard to believe that we’re even talking about it.
I do look forward to hearing Chip talk about his complaint against Google. I think it’s notable that it’s not a social media platform context in which they’re bringing their suit, and I have lots of thoughts about it, and I’m sure we’ll get to it in the Q&A, but I’m very curious to hear why anyone thinks that’s a common carriage issue.
So I will stop there, having used a least a little bit of Adam’s time, and thanks for the dispensation.
Hon. Gregory Katsas: All right. That’s a very natural point to segue over to Chip.
Charles Miller: All right, thank you. And I was really hoping that Adam would yield his time to me since we’re on the same side of this, but it didn’t happen.
Geoffrey Manne: He didn’t yield. I just took it.
Charles Miller: I know. All right, so as stated —
Prof. Adam Candeub: — I wanted to be nondiscriminatory.
Charles Miller: There we go. Geoff, what Ohio is doing here is different a bit than what’s been discussed so far in this series, which is we’re not focused on social media. We’re focused on Google Search and what Google Search does in its role on the internet, and how we think that that looks and operates and acts like a common carrier.
So I’m going to see what I can do about sharing my screen here and seeing if this will pop up as I thought that it would. Is it sharing now? Okay, well, I had switched over to this other — I switched over to my laptop, and I thought that I’d be able to do it better from this than my work computer, but since Apple is so good about security settings, it’s not allowing me to turn that on.
What I was going to show you was a nice little graphic that talks about the share of internet search that Google has. And historically, going back for the last 10 years, Google has had a minimum of 80 percent of all search that’s occurred in this country and globally, and it’s increased over time. Right now in the United States, Google Search has the market dominance of about 88 percent of search that occurs in the United States. Globally, it’s 92 percent, and on mobile platforms, it’s 94, 95, maybe 96 percent with the most recent statistics that you’re looking at.
Everyone says, “Well, hey, there are competitors out there. You can use Bing.” Bing has about 2 percent, 2 to 3 percent of the market of search. And so it’s simply not been proven to be an effective competitor.
And there’s talk in this realm about network effects, but one of the effects that’s out there for Google is the way that algorithms work, they become more refined and better the more that they’re used. And Google has such a lead on its competitors out there about conducting search that I think there’s agreement even by Microsoft that Google’s algorithms produce better search results.
In addition to dominating search, Google dominates the internet. Currently, right now, on a basis of rankings of use of websites in general across the world, the most visited website in the world is Google.com. For a given month, they’ve got 92 billion visits. Number two is YouTube with 34 billion visits. That’s also a Google property. And if you take those two Google properties and combine them, they receive more site visits than the next 48 of the largest 50 websites’ visits combined. And Bing receives about the same number of site visits as Google.com.br for Brazil at about 1 billion a month. And so the numbers simply are not close. And so we have this kind of monopolization that’s essentially going on by Google where it controls much of the web traffic.
And we’ve heard discussions about the public utility and common carrier—I guess I should focus on common carrier—a common carrier approach out there and what it does and what it means. And essentially, Google Search operates like common carriers. If you go back to the example of ferries over a river, what does Google do on the internet? It ferries to you the information and the products that you’re looking for across the sea of data that’s out there on the internet.
And there’s some discussion out there about, well, Google is a service provider. It’s not the infrastructure. But if you look at the infrastructure that’s out there, most people have multiple avenues available of accessing the internet. You can use the telephone company. You can use fiber. You can use cable. And with that, you can also simply use your cell phone and link the cellular networks that are out there. So there are all these different networks that are available for you to achieve the internet, but all of those lines of communications, all of those rail lines, if you will, all go to a central hub. And the central hub is Google because people go to Google to determine where to go next. And how Google directs everyone and where it directs them is of critical public importance.
I can go on with other analogies to other existing common carriers that are out there, but that establishes the framework that we’re looking at this and why Google is really the center of the internet and how people experience it. And so with the Google Search being out there and being central to what everyone does, our position is that that centrality to the internet should come with heightened duties. And as we know, within common carrier status comes this duty to not unreasonably discriminate. It’s not to discriminate at all; it’s not to discriminate unreasonably. And that unreasonable aspect of the test is really what’s critical to us.
Google, as a common carrier, has the ability to rank pages. It has the ability to use its algorithms to determine what is of most interest and most relevant to someone who’s searching. And it does that quite well. But what Google has done over time is it’s changed the way that it displays these results. And instead of prominently displaying the relevant results, what it does now is it interlaces those results with what Google wants you to see, and what Google wants you to do, and highlights Google products.
So if you search for a flight from where I am in Cincinnati out to where I want to go in Montana for vacation this summer, when it comes up on the search result page, you’re going to see Google flights near the top all the time, and not just as a natural search and not just as an ad, but actually integrated into the search result pages. So it’s this vertical preferencing that Google is doing that common carriage law would not permit. If Google is going to have integration in its web results, great, it can do it, but it has to offer that integration to Travelocity or Orbitz or others so that it’s offering those opportunities equally to competitors of its vertical products. That’s the concept that we’re looking at here.
We’re doing this as a case where we’re seeking first to establish that Google is a common carrier under Ohio law. And there are recent cases out there in Ohio of common law designations of public utilities and the common carriers that are outside of the public utility regulatory ground. So we intend to utilize that to get that designation, and then once we go beyond that, we’ve got to look at this duty on the commercial side essentially to avoid self-preferencing.
There are some other issues out there that folks have raised and has been discussed about Section 230 and how that can apply, and the Dormant Commerce Clause, and specifically with respect to that, Google can display its results in a way that simply it makes these changes in Ohio. It localizes its results to everyone all the time anyway, so it can choose to do this in a way that doesn’t have external effects outside of Ohio.
I’ve got, obviously, a lot more to say about our case, but I know that I’ve hit my time limit here, so I’ll turn it back to Judge Katsas.
Hon. Gregory Katsas: Olivier, you’re up.
Prof. Olivier Sylvain: Thank you, Judge Katsas. And I also want to thank The Federalist Society for inviting me, an unlikely invitee since I’m not generally one to subscribe to Federalist Society positions. But as I’ve said to the panelists, our discussions about big tech these days have found odd bedfellows among a variety of people. And in that vein, I’m enthusiastic about engaging in the conversation here. I can’t help also observe the oddity of being on an all-male panel. It’s been a long time.
Let me say a couple of things with regards to what I think Chip did allude to, but Chip stuck to the common carriage line. But with regards to what progressives have been talking about, they’ve been talking about public utility regulation. And some of the things that Adam mentioned, since he and I are both communications kind of folks, we think about transparency in the same way as we might think about disclosure of tariffs and nondiscrimination in the context of public utility regulation. But with regards to common carrier, it’s definitely — these are in the mix as well.
But I’d like to observe that there is not a lot of clarity when we talk about public utility versus common carrier regulation in the ether, generally. I will observe that the Communications Act—and Adam can correct me if I’m wrong—has a definition of common carrier, but it’s kind of circular, and the only thing it does is exclude broadcasting. So what we have instead is the common law, which is what Adam has written about and on which Justice Thomas relies in his opinion in the Knight case. So I just want to set that out also.
Reddit does is one way. Parler does it one way. Twitter does it one way. YouTube does is one way. I think Geoffrey and I are pretty aligned on this, and I don’t want to duplicate this, but I do want to observe that to the extent these companies are not putting themselves out as serving the public indiscriminately, it’s hard for me to square the common carrier analysis in this setting. And in any case, we talk about this as a market for moderation. This is the sort of thing that I would presume longstanding conservatives would be interested in, different opportunities to hear different points of views, again, echoing what Geoffrey has said.
But we can dig down even more, and I’d like to talk about the First Amendment in this setting. And the real question is whether there’s editorial discretion that resembles the sorts of things that are protected in the First Amendment. And I want to admit that I’m not sure on this question, but I know what the cases say. I don’t know as a normative matter what happened, but I know what the cases say. And it’s going to be really tough to overcome a line of cases that, with regard to search in particular, that underscore that this is what the courts have found to be a space where there is editorial discretion.
The kind of ranking that Chip has described was by the Western District of Oklahoma in the Search King v. Google case. For that court, it was editorial decision making, and so the First Amendment could be an affirmative defense against a claim of tortious interference because Google can make decisions about how to rank, even if it’s ostensibly anticompetitive.
Now, that’s another interesting problem, and I think Chip’s points on this are interesting. I’d like to return to them later. But that’s a case in which — a foundational case in some regards, even though it’s a district court opinion that’s not from California on this, pretty foundational in setting out why the First Amendment is protective of ranking decisions. They’re a recommendation from Google that would be different from, say, a recommendation from Bing, again, cabining out the competition antitrust problem that Chip identified.
Another important which speaks to something I think this audience is especially interested in is also editorial discretion that is biased against certain political points of view. And in the Zhang v. Baidu case out of the Southern District of New York, the court says even a Chinese language search engine that is forced to do certain kinds of moderation decisions by virtue of being Chinese may suppress and censor political dissidents in the results of their search, ironically, under the First Amendment. There’s a kind of discretion that they can do that, and that’s ostensibly, again, going to Geoffrey’s point, there’s another opportunity — you can find these dissident sites or these dissident sources through Google or some other avenue.
Now, that’s — for me, the First Amendment resolves it as to search. I think social media is a little bit more complicated, but I’m still — the cases suggest otherwise with regards to whether this is allowed. And the one that stands out is the Prager University case, Prager University v. Google out of the Ninth Circuit where the court said that social media — when Prager University brought a case saying that YouTube Google was censoring against conservative views in violation of the First Amendment, the court said, “No, you don’t have a First Amendment right here.”
First of all, this is a private actor, citing the Halleck case from the Supreme Court — and we can talk about the Halleck case if we have time, but they’re also not operating a traditional government function. They’re not running elections, and they don’t run a company town. And those are the areas in which we might impose some — entitle individuals to a free speech right, but no so here. And that’s, again, the same reason that Geoffrey mentioned and that I want to echo here, that there is a market for it, even if there is a clear competition problem, again, that I’ve been cabining.
I do want to observe that the Pruneyard case is really interesting. Adam mentioned it. I know he’s probably written about it. Justice Thomas mentions is in his opinion, and my eyes, as with many others, opened wide when he cited to the Pruneyard opinion. But that is a case decided under California constitutional law. That is not a case that jibes with U.S. constitutional law with regards to whether people have a right to speak on private property. So I assume this is a generative and fruitful conversation we can have as a panel.
I want to return to this idea that common carriage can be a lot of different things, which is what Adam says. And it’s actually a very, I think, a fruitful way to think about government regulation. It doesn’t have to be carriage as such.
Now, the term is common carriage, so we might presume that that is part of the formula, but let’s cabin that out and think about transparency and disclosure requirements, which are public utility regulations. I’m actually open and interested, for what it’s worth, in those kinds of interventions. They do raise questions about compulsion of speech, which is as violative of First Amendment rights as censorship. And so that’s an interesting problem and question, but compulsion disclosure in these settings might be appropriate, for what it’s worth. I want to convey that that’s something that is interesting to me.
But we still have to get over the question if we are going to be treated this way is whether they are truly indiscriminately putting themselves out there, that they are really takers for all comers, or is it — I don’t know how to phrase that. That they are allowed to take everyone, that they’re taking everyone that wants to come.
And again, to get back to one of the points that Geoffrey made, is this really what advocates of reform want, that you would include pornographers, other kinds of speakers — forget pornographers; people who are engaged in ostensibly lawful speech but clearly dangerous speech. And that gets to the point about moderation. A lot of these companies have made decisions about the kind of content they allow presumably because in the market for moderation, there is a consumer interest in making sure that whatever site they visit is censored or is moderated. There’s actually a consumer demand for it. And I actually think that’s where that’s coming from. But again, it’s a market for moderation where you can find what you want.
This reminds me, actually, I forgot to mention one of the interventions that Adam mentions that would be fruitful is taking blocking and censoring out of the hands of the social media companies and into the hands of users. This is — if you read the cases as I read them, that would be censorship. The government imposed that obligation. That is the sort of thing that would indeed be intruding on the intermediary’s right to speak, citing back to the Tornillo case where the right of reply statute is struck down in Florida. And I’m not just saying that because Geoffrey’s there.
I think that the thing that Justice Thomas’s opinion does that opens conversation up is enumerate the circumstances and conditions for common carrier regulation, the sort of thing that Adam has written about. And so one of them is market power, and that is an interesting thing. But I’m not sure it’s sufficient because in the cases, the Turner cases and even in the other cases, what is also important is market gatekeeper, that these are entities that are actually gatekeeper to the distribution of content like a cable company. And that’s why that is treated under intermediate scrutiny because of its incidental effects on speech.
But the real problem that Congress is trying to address, and that the FTC has implemented, the right regulation is cable company’s role as a gatekeeper, not just because it is a market actor. And I think that helps in some ways think about the Ohio case a little bit. But because Chip focused so much on market share and size, this makes me think antitrust is really not gatekeeping as such, but antitrust really is the intervention. And I know Judge Katsas says that’s a discussion for next time, but I do want to flag that.
My last point is taking up something that Chip said. He said that we should hold these companies accountable, at least impose some heightened sense of duty. My position, and this will return to the panel you just did, is that they should actually be subject to some legal duty. Under Section 230, they are not. They are immune for so many of the decisions they make, even if they are distributing information and content that is unlawful. And that’s not a 230 discussion, but I don’t even think we should be thinking about hiding duties just yet. I think we need to get over the question of whether they are following any legal duties that are ostensibly the sort of thing that all other entities in the market must follow.
I think I’ve probably gone long enough. I’ll stop here.
Hon. Gregory Katsas: Olivier, a couple minutes over. Adam, a couple minutes short. So let’s do three minute follow-ups. And Adam, I’ll give you five if you want it.
Prof. Adam Candeub: Thank you, Judge Katsas. I appreciate that. So just quickly, I think I’m somewhat surprised by Geoffrey’s remark that we have this wonderful diversity of public media. I think, as I mentioned, when President Trump was kicked off, the former President of the United States, when we can’t talk about serious scientific issues in an honest way, that’s not a public forum that I’m particularly interested in.
The question about cesspools and we want to market in moderation, as Olivier pointed out, first of all, one, I’m not sure that’s true in the sense that we don’t — these platforms exhibit tremendous network effects. Second, it’s not clear there’s not collusion going on. I think what happened with Amazon Web Services and Gab with the breaking of their contract suggests that these platforms will not allow a platform moderation — a market for moderation to develop, and so that may not be the answer. If we accept these competitive predicates, then I think regulation is perhaps the best approach.
The idea — again, this is Olivier’s point that search is itself a First Amendment protected activity. In studying those two district court cases, I’m not sure that’s the case. If you look at Google with it’s massive, as Charles pointed out, it’s massive and really impossible to reproduce search histories, you can really conceive it not as an expressive activity but as like the ultimate index of information in our society.
It’s sort of like the Yellow Pages for everything that’s out there, and it’s the best Yellow Pages because Google has clarified it and perfected it through all the AI and huge amounts of data. So in that respect, it’s not their expressive activity. It’s a unique property, much like the bridge over the St. Louis River was in antitrust law. So I’m not sure I agree with that.
Again, the fear of what — I think there is actually a tremendous area of agreement here with Geoff and with Olivier about other things beyond common carriage. As I said, if you go through the centuries of common carriage laws, it meant many things in different times. I think whether that’s disclosure, whether that’s some type of transparency, I think that would be fine.
And last point, and I’ll end quickly, is on discrimination. Telephone companies, airlines, they all have nondiscrimination requirements. They can’t deny you a telephone because you’re a Republican or you’re a Democrat or you’re a Satanist. They work fine. I just don’t see the end of the world if we have some sort of nondiscrimination requirement of the sort that virtually every institution deals with.
Am I under again, Judge Katsas?
Hon. Gregory Katsas: We’re under. You have two more minutes if you want them.
Prof. Adam Candeub: No, I will give them to Charles this time.
Hon. Gregory Katsas: All right. Geoff?
Geoffrey Manne: I can’t help but laugh at the remark that Adam made about these platforms not allowing for a market for moderation to emerge. As he was saying it, I was repeating in my head these 350 platforms will not allow a market for moderation to emerge. There is a massive market for moderation here.
There are an enormous number of platforms, and it’s not, of course just platforms. There’s traditional media. There are — here’s a good example. We’re going to — you’ve discussed Trump on this video. Google is not going to prohibit this video from being shown. Even if Google did, there would be a number of other avenues of reaching it. FedSoc hosts it on its own site. It doesn’t just host it on YouTube. And you don’t have to go through Google Search to find it. And even if you did, even if all of that were true, there are about a billion discussions of President Trump and of the COVID treatment issue available out there.
You can point to a couple of examples where certain platforms have been imposing their own community standards that might have precluded that conversation, but you simply cannot demonstrate that the conversation isn’t happening. It’s happening all over the place, just maybe not in the particular place you want it to happen.
But the First Amendment has never protected your preference for speaking in a certain way in a certain place. There are ways in which it does do that, of course, but in this respect, the idea that you should have the right to speak on Twitter when there are hundreds if not thousands of other outlets available to speak on is not something that I think the First Amendment would in any way countenance the government mandating.
I want to go to the point about discrimination really quickly and just say I think one of the really important things that is missed here is the notion that — so as Adam said, and I don’t mean to suggest that Adam was being absolutely literal with this. It’s a convenient way of talking, but it’s indicative. So he talked about you’re a Republican or a Democrat and about platforms or traditional common carriers, a telephone not being able to discriminate on that basis.
The issue here is not the speaker. It’s the content. These platforms are all about providing moderated platforms that present an agglomeration of content that fits with their preferences for how that content be provided. As far as I know, there is no effort to squelch speech, even on these platforms, even though they could, on the basis of who’s speaking. It’s generally the content of the speech itself.
And the thing is that that’s what makes these platforms so robust. That’s what makes common carriage so inapplicable to them. They are not transmitting unified commodity-like pieces of product like you get in traditional common carrier settings. They are providing a platform for massively diverse and always moderated discussions to take place.
And it’s not, again, as far as I know, about squelching the political preferences of the speakers. It is in some cases about cutting off some of the speech that happens to be correlated with the political preferences of some of the speakers, but it is in every case excluded or constrained because of perfectly transparent community standards they impose. Never is it because they are Republican. It’s generally because they’re lying or they’re suborning violence.
Now, you can disagree about whether the content is actually doing that or not, and I understand the kinds of concerns that a lot of people have. But I just don’t think it’s fair to suggest that the platforms, even if they could in any reasonable way, are imposing sort of their own political preferences. I think that that’s rather a function of the correlation that’s being observed rather than an intentional effort on their parts.
Hon. Gregory Katsas: Chip?
Charles Miller: All right. Thank you. It becomes a bit frustrating in these debates when we have the folks that are kind of opposing the common carrier view of this, I think, being kind of less disciplined in saying, ah, those things don’t apply for these reasons and trying to kind of, I feel, hold us to a very, very tight standard of what does a common carrier look like in the law now, and so it must be that.
For example, we’ve got this concept of monopoly. Well, there’s discussion about well, you don’t have to use Google. There are other things out there. You can use Bing. Sure, it doesn’t work as well, but you can go there. You can go find WebCrawler if you really go and look for it. It still exists somewhere. But when you get to common carriage, as opposed to public utility, the monopoly isn’t important.
Think of the number of trucking companies that you pass any time you’re on the highway. There are all kinds of trucks that are going by, and by all kinds of different companies, and some are individuals, some are conglomerates, but they’re all common carriers. And with respect to the content, we say that, hey, well, with Google, what it’s producing is it looks different for whatever search you do, and it might look different depending upon who you are doing the search because you’re in a different location, or you’ve exhibited in the past that you prefer local coffee shops to Starbucks, so they’re going to show you the local stuff.
Well, in trucking, when you carry eggs, you transport those a lot differently than when you transport, say, hazardous materials, and there are different regulations that go with that. And there are refrigeration regulation that go with that some things. And when you’re an air carrier, the regulations for doing scheduled service of passengers is different than that of doing charter service for passengers is different than that for carrying cargo. And so in all of these realms, there are different things to transport, but that doesn’t mean that they’re not common carriers. They are common carriers.
And with respect to the size of Google being so predominant means that maybe antitrust is the way to go, we all know that the antitrust law out there is so focused upon consumer harm that that’s a different type of scenario than what we’re talking about here. There is consumer harm that’s at play, but how it’s demonstrated is a bit different.
But common carriage does apply. With respect to common carriage, as was stated, the question is do you accept all comers? Google does. There’s no question it accepts all comers. On the user side, you go to that website, you type in what you want, it’ll produce a result. You pay for that? You kind of do. You don’t exactly pay for it with currency, but you pay for it with data. And on the other side, to get listed, Google goes and crawls the whole web. It looks everywhere, and you get listed in those white pages, and then they rank it.
And again, what we’re looking for, the ask that we make, isn’t to alter that algorithm and to say that there’s anything wrong with the algorithms. Google’s algorithms are super refined. They work just fine. But what Google does is Google doesn’t trust the algorithms for itself, and it makes sure that it puts its reviews at the top because it wants to keep that information. It tries to capture the clicks, and it does a very, very good job.
Over 50 percent of searches, and some say up to two-thirds of searches that happen on Google are captured clicks. They never leave Google properties. They’re either resolved on the Google side itself, they’re sent to YouTube instead of other video providers, or they’re sent to Google Travel or other verticals that they have. And it’s Googles abuse of its close to monopolization of internet search that it’s using to grow its other businesses and give itself an unfair competitive advantage in these other lines that we’re concerned about and that we think common carriage is perfectly reasonable.
Hon. Gregory Katsas: Olivier?
Prof. Olivier Sylvain: I went a couple minutes over, so I’m going to try to stay short, okay? How’s that, Judge Katsas? I’m not known for that, so I’m going to try hard.
First, on the definition of common carrier, this is not a definition that I made up. This is a definition from the Communications Act. And with regards to what the sorts of things could count as being regulated, I wasn’t necessarily advancing the position I have. I was talking about what the courts have said.
As for the antitrust question, I do think there is an interesting, worthwhile antitrust conversation that is not the common carrier conversation. And the European commission has done precisely what I believe Chip is describing with regards to comparison shopping at Google, fining Google for privileging its own comparison shopping site under competition law, not as common carriage regulation.
Finally, I think the nondiscrimination problem, to just amplify the point, is if you impose a nondiscrimination principle, it isn’t just that you’re taking all comers, including the stuff most of us don’t want to see. But because of the ways in which these automated decision making systems operate in consideration of the things that they accelerate and distribute, we’re going to get the most explosive, the most harmful and incendiary content getting distributed. That’s how I experience the nondiscrimination principle working.
Hon. Gregory Katsas: Let’s talk a little bit about the connections, if any, between the antitrust concepts and the common carrier concepts because I know, Chip, you said we don’t necessarily need a finding of market power to regulate as common carriers. But a lot of concern in this area does seem driven by the sense, right or wrong, that the Googles and the Twitters and the Facebooks of the world have enormous control over speech. Justice Thomas said that, among others.
So how should we think about the relevant market here and alternatives for speech? Is it an antitrust concept, and if so, what is the relevant market? And Geoff, you said there are lots of opportunities for speech. That’s clearly true, broadly defined, but if you’re, let’s say, a public figure who wants to communicate in real time with millions of your followers and Twitter won’t carry you, Facebook won’t carry you, and yeah, maybe you can book yourself on Fox News or publish in National Review, but is that a reasonable alternative in antitrust concepts or forum concepts or whatever standard should govern in this area?
Geoffrey Manne: Can I address that? I’d love to weigh in a little bit on the antitrust issues or the antitrust relevant issues. I think a big thing that I just have to say before I address your question, Judge Katsas, is the notion that Chip raised of this being not — of the problem of antitrust being that it adopts a consumer-centric consumer welfare approach, and that their effort is to pursue something different.
And my question is why? What is the harm that you think you’re protecting against? The consumer welfare standard at least adopts a standard of harm that looks at whether there is injury to consumers. The European commission’s Google Shopping case is a great example of precisely doing not that, of doing the opposite of that. There is no consideration of consumer harm in the Google Shopping case. There is just a sort of imposition of a kind of preference for nondiscrimination without any means to justify its basis.
Charles Miller: There is consumer harm in these contexts.
Geoffrey Manne: Which is what?
Charles Miller: Well, the consumer harm occurs a couple different ways. One, because Google can do this slight of link, if you will, it can send customers to sources that are not the best for them. For example, when it prefers Google reviews over, say, TripAdvisor or Yelp or something like that, it could be sending people to less quality content. So they’re sending them to things that provide problems for them.
Geoffrey Manne: It’s, of course, not sending them anywhere. What’s interesting about that is it’s a preferencing —
Charles Miller: — Well, wait, wait, wait, wait, wait. So that’s part of the problem. There’s another aspect —
Hon. Gregory Katsas: — I’m going to intervene here. One at a time. Chip, you want to finish the thought?
Charles Miller: Yes. And so the other aspect of consumer harm that exists is the lack of innovation. If Google is able to suck up everything and steer everything to Google, then you’re going to have less competitors and less innovation, and so that’s eventually going to harm customers. My point is it’s difficult to establish all of that in the antitrust realm where we can deal with that in common carriage.
Hon. Gregory Katsas: Okay. Geoff?
Geoffrey Manne: It is difficult to establish long-term, speculative, innovation-related harms. And that’s by design in the antitrust context. And the places where we’ve imposed common carriage restraints are, I think, cases where — actually, I don’t want to speak to those because I think a lot of them are faulty as well.
But the point really is to say just because you can identify a theoretical possibility of harm doesn’t mean that you’ve justified imposition of this constraint, which, by the way, acts precisely in the nature of a constraint on innovation. The thing that you’re complaining about in particular is an innovation by Google to start offering direct answers to queries at the top of its search result pages rather than the ten blue links that take you out elsewhere. That innovation arguably wouldn’t or couldn’t occur under the scheme that you’re suggesting. It’s hard to know in any meaningful way whether that loss of innovation would be greater or less than the theoretical prospective loss of innovation that you’re talking about.
I think on the market definition point you asked about, Judge Katsas, just really quickly, I just wanted to say I think it’s an interesting question, and I suppose one that doesn’t apply in the common carrier context in the sense that if the relevant conditions are met, if you’re holding yourself out as a provider to all comers, which is debatable, and if the government decides and is able to impose the constraint, I don’t think there’s any reason that it has to justify it by reference to the kind of relevant markets you would have to in antitrust.
But that’s precisely why I think it’s a problem. I think that in the antitrust context, this should be relevant. I’m not clear that it would be in any particular case, but it very clearly should be relevant when you’re talking about these kind of speech opportunities. Google Search might be something different. But in the common carrier setting, I can’t imagine that there’s any sort of — I know there’s no obligation to take account of that, and that’s one of the things that strikes me as problematic about trying to apply it in this context.
Hon. Gregory Katsas: Olivier or Adam? Olivier, can you hear us?
Prof. Olivier Sylvain: Yeah.
Prof. Adam Candeub: Olivier, would you like to say something?
Prof. Olivier Sylvain: Can you hear me?
Hon. Gregory Katsas: We hear you.
Prof. Olivier Sylvain: Can you hear me? If not, Adam, please go ahead. My connection is unstable. I apologize. I don’t have much to add anyway, so, Adam, go ahead. If my connection clears up, I’ll jump in.
Prof. Adam Candeub: Okay. Thank you, Olivier. I’d just say this debate is really quite old. It is a rehash of the debate that Harvard Professor Bruce Wyman had with Cornell Professor Charles Burdick in 1910. What is the basis of common carriage? And Wyman said, “No, it has to do with industries that are open to — that offer themselves at all, and it’s about making sure that everybody’s able to participate fully and completely in society.” Burdick said, “Oh, no, no. there’s this antitrust market power component.”
And of course, the B+ answer, which I think is correct, is they’re both right. There’s an element of both in this rather amorphous regulation. I’m eager from comments from the audience as well, but one is common carriage allows for regulation without the rather rigorous and perhaps even too rigorous demonstrations that antitrust requires, but also allows for this notion that we as members of society have to participate fully in certain types of activities like communicating to each other.
And I think it also recognizes that network effects are quite powerful. There’s a reason why we don’t really see a tremendous diversity of firms with equivalent market power. We see a power law in which pretty much most areas of activity are dominated by a handful. And that’s because we all like to talk to everybody, and powerful networks are better. The more powerful, the more go on them, so there’s a push toward that.
But going back to Judge Katsas’ point is what is the market here? And again, I think we have more work to do. I’m not quite sure what the market is. That’s the thing about Google. It sort of gets its tentacles into everything and is the market for nothing, and yet, everything, which is not a good answer. But I think this is something that we have to come up with better answers for because I’m not quite sure I have one.
Hon. Gregory Katsas: Let me try to squeeze in one more question quickly, which is it sounds very appealing to say in the abstract that let’s just impose a light touch “no unreasonable discrimination” rule, and maybe we can particularize that to “don’t deplatform speakers based on disagreement with their political viewpoint.” That all sounds great, but when you try to operationalize this, how intrusive are we going to be? And I think I read — I think it was your article, Adam. I think someone on the regulatory side of this said something like we’ll probably need a federal agency to operationalize these judgements.
Prof. Adam Candeub: Not me. I hope not.
Hon. Gregory Katsas: How worried should we be with the prospect of the FTC or an Ohio commission pulling the curtain out to examine how Google does its algorithms or whatever?
Prof. Adam Candeub: I’ll be very brief. My preference is actually because I think judges have such good insights and we can rely upon their judgement and that common carriage law was imposed by judges for most of the 19th century. It really was until the 1920s. And I think if we kept toward judicial remedies, I think by nature the rules would be simple and easier to apply because you just couldn’t get more complicated.
Hon. Gregory Katsas: But is it harder because we are regulating speech?
Geoffrey Manne: Exactly, yes. That’s not to say that the courts can’t address that, but I do think it is —
Hon. Gregory Katsas: — Geoff, when regulating airlines, it’s intrusive, but so what? When we’re regulating Google or the cable company or whoever, they have their speech interests too. And do we care more about that, or do we care about the interests of —
Charles Miller: — That’s why we’re talking about this as common carriage today. If we were talking about it as regulation, we wouldn’t need to talk about what operative legal theory we’re regulating under. We’re just talking about what power we’re regulating under. And so a common carriage approach is designed to do that case by case analysis via common law and the courts and use the jurisprudence to get to the right result and carefully weigh all of those issues.
Prof. Olivier Sylvain: Judge Katsas, can I weigh in with the 30 seconds we have left?
Hon. Gregory Katsas: Sure.
Prof. Olivier Sylvain: I’m sorry about my messed up connection there. So even if just because you label something a common carrier doesn’t mean that you can proceed, you still have to go over the legal analysis and question. And I do think it matters that it’s the way Judge Katsas has set out this proposal as to be provocative, it’d be addressed to the heartline of what’s protected under the First Amendment. It’s not like lumber.
But I will say, to offer a fig leaf, there are things that happen on these intermediary sites that are not political speech but that are commercial speech. And that is the sort of thing we might want to look into, particularly if the content is driven by the sorts of political economy that Adam and Chip are interested in.
Hon. Gregory Katsas: I’ve been told that we have a hard stop of 2:20, and it is now 2:20, so I’m going to turn the floor back to Alida.
Alida Cass: Thank you very much, Judge Katsas, and to all our panelists. On behalf of The Federalist Society, I want to thank our experts for the benefit of their time and expertise today. And I want to thank our audience for calling in and participating. We welcome listener feedback by email at [email protected].
I hope you will all join us next week on Thursday, July 1st at 1:00 p.m. for the next discussion in this series when we will consider the antitrust questions relating to speech and digital platforms. As always, keep an eye on our website and your emails for announcements about upcoming teleforum calls and virtual events. Thank you all for joining us today. We are adjourned.
Dean Reuter: Thank you for listening to this episode of Teleforum, a podcast of The Federalist Society’s Practice Groups. For more information about The Federalist Society, the practice groups, and to become a Federalist Society member, please visit our website at fedsoc.org.