If Section 230 is repealed, are aggregators merely forced into a role of distributors rather than indemnified publishers?

Question

According to the article by the Electronic Frontier Foundation (EFF) "Section 230 is Good, Actually"

Section 230 says that any site that hosts the content of other “speakers”—from writing, to videos, to pictures, to code that others write or upload—is not liable for that content, except for some important exceptions for violations of federal criminal law and intellectual property claims.

They go on to say,

Basically, Section 230 means that if you break the law online, you should be the only one held responsible, not the website, app, or forum where you said the unlawful thing.

Just to be clear, that isn't strictly correct, right? A quick wikipedia search shows that in Cubby, Inc. v. CompuServe Inc.

As a distributor, CompuServe could only be held liable for defamation if it knew, or had reason to know, of the defamatory nature of the content. As CompuServe had made no effort to review the large volume of content on its forums, it could not be held liable for the defamatory content.

In one source online, I even saw Section 230 described as

Congress enacted § 230 to remove the disincentives to self-regulation

So which is it? Does the removal Section 230 mean a site will be held liable for the content created by others? Or, does it mean that a site will only be held liable if they engage in optional moderation or acts of "publishing" (such as filtering and removal of offensive material). It seems like it's still the creator that is liable. And you can still have a free and open moderation-free internet in a post-Sec 230 world?

Handling spam

Update: On spam, the same case above Cubby, Inc. v. CompuServe Inc., 776 F. Supp. 135 (S.D.N.Y. 1991) states that,

CompuServe's ultimate right under the contract to remove text from its system for noncompliance with its standards merely constitutes control over the result of CCI's independent work. This level of control over the Journalism Forum is insufficient to rise to the level of an agency relationship.

With this text, I don't think it's sufficient to say that in a post-Section 230 world removal of content would be defamation per se, it seems it has satisfy some standard.

Answer 1

you can still have a free and open moderation-free internet in a post-Sec 230 world

Sure, but remember what moderation-free means: no moderation whatsoever. That means no removal of offensive content like trolling, profanity-laden or racist rants, or even outright spam.

Stack Exchange, for example, gets thousands of attempted spam posts a day, despite the fact that very few of them actually get through, and the ones that do are usually quickly removed. Section 230 protects Stack Exchange's ability to do this without incurring liability for what users post. Imagine a Stack Exchange in which spamming was allowed.

So I would say that the EFF's statement is substantially accurate, in that the ability of sites to perform such moderation is fairly essential to their ability to function as communities. The fact that they could avoid liability by not moderating is not relevant if it would make the site unable to function properly.

Answer 2

The quote from the EFF:

Section 230 says that any site that hosts the content of other “speakers”—from writing, to videos, to pictures, to code that others write or upload—is not liable for that content, except for some important exceptions for violations of federal criminal law and intellectual property claims.

Is a little over-broad.

In traditional US defamation law there are specific liabilities attached to being the "publisher" of something. In general, a publisher of defamatory content was a joint tort-feasor with the author, they were jointly and severally liable for anything defamatory in the work. This rule was inherited from English common law. It was based on the assumption that a publisher would be fully aware (or should be) of the content of all publications that the publisher published, and that the publisher was often a much better target of litigation than an author.

In the late 20th-century this rule was changed for digital publishers by court action. In Cubby, Inc. v. CompuServe Inc. and other similar cases a publisher was not held liable for defamation unless the publisher knew or have good reason to know of the defamatory content.

In addition to defamation, there are other ways for publishers to be liable for published content. Torts such as invasion of privacy and copyright infringement, and crimes such as publication of classified information and publication of obscene matter could result in publisher liability. None of these were affected by cases such as Cubby, although the logic of Cubby might be extended to some or all of them.

Section 230 says that an online service provider or user shall not be treated as the publisher, or as the speaker, of anyone else's content. Thus none of the liabilities traditionally imposed on a publisher (or a speaker) apply, even if the host carefully moderates user content. That is significantly broader than the Cubby rule. Repeal of this section might return the law to the Cubby rule under which a service provider escaped liability for defamation provided that it had no good reason to know of the content, but might not escape copyright liability, or various other forms of liability, and might be liable if the provider did routinely review content. (Other laws such as the DMCA would still give some protection to hosts, I would suppose.)

The suggestion in the question that repeal of section 230 would have little legal effect is not thus correct.

However, section 230 does not mean that there is no theory of liability that could apply to a host or service provider. And it would not prevent a specific statute imposing liability on a host as such (not as a publisher) in limited circumstances. That is where I think the EFF is over-simplifying. But their statement is probably closer to the facts of the matter than many other current statements are.

Answer 3

I worked with forums that hosted User Generated Content (UGC) ... both before DMCA landed and also shortly after, before it was certain how courts would interpret DMCA.

Yes, lack of a well-tested Section 230 had a chilling effect on site monitoring and oversight.

Naturally, we "ate our own dogfood" and read and participated in the UGC. So we saw lots of stuff that we'd rather not be on the forum. And it would've been easy to hop over into the admin tools and action it.

But it was put to us plain, by counsel, that we could do that a little... but we couldn't be seen doing it enough that someone might argue that our involvement constituted moderation. In short, we were told to save administrative intervention for only the most extreme cases, or those reported/flagged by non-staff users. One of our own staffers couldn't sit at home and report TOS violations, even on their own time, or this would be seen as the left and right hands of the same entity (us).

So yes, we couldn't administer our own forums, at least not by modern standards.

Answer 4

It seems everyone believes that liability for speech is broken down into two parties,

• The poster,
• The distributors/aggregators/publishers (which we will distinguish later),

Further, broadly, that

• Content distributors/aggregators/publishers should not be liable,
• Except in some conditions (as defined by behaviors),
• Section 230 eliminates some of those conditions (there are substantial exceptions),

Note, that Cubby, Inc. v. CompuServe Inc. was further refined in in Stratton Oakmont, Inc. v. Prodigy Services Co. which enumerates behaviors that can be considered above,

• Does the service "[hold] itself out as an online service that exercised editorial control"
• Does the service have "automatic software screening programs". Note, in this case, the phrase was in reference the removal of content based on a list of obscene terms.
• Does the service have an agency relationship with its editors, which "Guidelines which Board Leaders are required to enforce" would establish.

As case-law establishes these terms, when a "distributor" engages in these behaviors they become a "publisher".

So in a post-Section 230 world software screening content would not be permitted. Also, community moderators, which can still exist mustn't have an agency relationship with the distributors/aggregators/publishers, and must maintain their own independence. And the distributors/aggregators/publishers must not say it is providing value over the contributions of its users.

In the case of StackExchange, it seems some obvious violations would be policies not decided by the community itself (like who can run for elections) and the use of a network-ban, and further any actions by paid "Community Managers" that impact the content or community.

Response to comments

David Siegel writes this,

I would disagree with several aspects of the above answer. First, it is not that liability is divided into two parts, it is that author and publisher were jointly liable and either could be sued. Third, sec 230 deals with criminal as well as civil liability at least in some cases. Fourth, it is not at all clear that aggregators or distributors who were NOT also publishers ever had liability, and if they did section 230 does not help them. (A bookstore is a distributor but not a publisher or author. An ISP might be a distributor.) ... Fifthly the case-law rules you cite were still very much under development when sec 230 was passed, and were not yet settled law. Had it not been pass there would probably have been further developments. If 230 were repealed, there might well be further developments. Indeed something not unlike 230 might emerge from caselaw. Or it might not. Sixth it is not at all correct to say "screening content would not be permitted." Merely that screening would carry increased legal risks. Newspapers have bourn such risks for hundreds of years, as have other publishers. ... Seventh, not everyone agrees tht publishers should be exempt[t from liability. at least some of those pushing repeal of 230 want publishers to be potentially liable, so that they will police/restrict posters. –

1. I never said the division was mutually exclusive, I only said it exists.
2. (there is no number 2 [sic?])
3. I have corrected this though I don't think that's relevant to the argument.
4. I didn't distinguish in my answer between "distributors/aggregators/publishers", however I acknowledged these different terms jointly. I did this, stylistically, because I wanted to focus on the differences which I identified as "conditions" in the very next section. Contrast this to your own answer, where you misidentify CompuServ as a "digital publisher". However, I've tried to more cleanly draw this distinction.
5. I agree there is little case law pre-Section 230. That's not a reason to ignore the case law that exists.
6. In the context of law, not permitted and "incur legal risks" are the same. I physically can shout "fire" in a movie theater. There may also be repercussions and legal remedies available to those who acted on it.
7. Sure, but if you're arguing for a change in liability the question becomes to whom that change applies: this is the first answer to incorporate that distinction (whether you want to use the terms established in case law of "distributor" and "publisher", or merely distributors that do not engage in behaviors, and those that do). That said, the loudest people pushing this (Trump, et al) are lashing out against the policing that exists.