Judge Goes Rogue and Rejects Snap’s Section 230 Defense for [Reasons]–Neville v. Snap

The plaintiffs are parents of Snapchat users who purchased fentanyl from other Snapchat users and suffered overdoses. They sued Snapchat for a wide ranges of tort claims. To get around the clear Section 230 barrier to those claims, the plaintiffs attempted the now-standard Lemmon v. Snap workaround of claiming to sue over Snap’s first-party product design choices, not the drug dealers’ content:

Plaintiffs disavow any claim based upon Snap’s activities as a publisher of the third-party (drug seller) content. That is, plaintiffs do not contend that Snap is liable for failing to eliminate or otherwise moderate some or all of the third-party drug sellers’ content. Instead, plaintiffs’ sole focus, they say, is on (1) Snap’s alleged independent tortious conduct and (2) Snapchat (a “feature-packed social media app.”) as a defective product.

The court rejects most of Snapchat’s motion to dismiss.

Section 230

The court’s Section 230 analysis runs for 13 pages (pages 9-22), but it has an inappropriately laconic denouement.

Hacking the Precedent

The key to the judge’s unexpected ruling is how the court manipulates the precedent it purports to consider.

The court acknowledges it must follow Section 230 rulings from the US Supreme Court (a null set following the Supreme Court’s dodge in Gonzalez v. Google) and California Supreme Court. With respect to the latter, the most recent case, Hassell v. Bird, gets a single non-substantive cite. The venerable Barrett v. Rosenthal opinion only gets a couple of citations, including an acknowledgement that the court rejected Section 230’s publisher/distributor distinction (more on that in a moment). The court made a big show of recounting the applicable precedent, but then only barely engaged with the binding precedent.

The court also acknowledges that it’s bound by California state appellate opinions but says it’s free to choose among conflicting decisions. In particular, the court acknowledges the Doe II v. MySpace ruling, which applied Section 230 to allegations that the service didn’t do enough to keep users safe. The Doe II opinion addressed many of the issues before this court, including whether the claims were based on third-party content or first-party design. Yet, the court feels free to disregard this extremely relevant and analogizable decision because allegedly Liapes v. Facebook reached a contrary outcome, even though Liapes dealt with a different set of facts and issues. [Note: Liapes v. Facebook is on appeal to the California Supreme Court, and I filed a letter in support of granting review in the case.]

So this is how the court “hacks” the precedent to reach its desired result. It marginalizes the California Supreme Court opinions, cherrypicks less relevant California state appellate decisions, and pulls in non-binding federal caselaw only when it suits its narrative. These moves enable the court to treat this case as jurisprudentially novel when it really is not–and, in turn, reach an outcome that obviously conflicts with the precedent.

Statutory Analysis

The court says the “Plaintiffs’ theories here do not purport to hold Snap liable for failing to remove some or all of the drug sellers’ third-party content from Snapchat.” The plaintiffs are claiming this, but is it correct? As indicated by Doe II v. MySpace, the Fifth Circuit wranglings in Doe v. Snap, and many other cases, it’s impossible to disaggregate the plaintiffs’ legal theories from Snapchat’s publication of the drug dealers’ content.

To buttress its assessment that the plaintiffs are suing over first-party design and not third-party content, the court says:

a “but for”/”based on”/”flows from” test is not consistent with a plain meaning analysis of the words Congress chose to employ. If Congress had intended to immunize all interactive computer services from liabilities “based on” third-party content, there are straightforward elocutions to express that intention. But that is neither what Congress did nor what Congress could have done consistent with the policy statements in subdivision (b) of section 230. Instead, Congress chose to invoke words of art drawn from common law defamation-liability distinctions between “publishers” and “speakers,” on the one hand, and, apparently, “distributors” on the other.

Wait, what? The statute expressly says that the courts can’t treat Snapchat as the publisher of third-party content. That’s why so many courts, such as Doe II v. MySpace, have cut through pretextual arguments about first-party design when the substance is to impose liability for third-party content. Also…why is the court talking about the publisher/distributor distinction when it acknowledged that Barrett v. Rosenthal (remember, the binding California Supreme Court decision) rejected that distinction?

[This de novo statutory interpretation, decontextualized from the precedent, reminded me of the rogue “plain language” appellate court decision in Daniel v. Armslist, which the Wisconsin Supreme Court reversed.]

Review of the Precedent

Moving on from its de novo statutory review, the court cites the Stratton Oakmont, Zeran, and Dyroff cases. The Dyroff facts are VERY close to this case: the Dyroff plaintiff sued a service because a user connected with a drug dealer on the service, bought fentanyl-laced heroin, and overdosed. The Ninth Circuit said that Section 230 applies to that case, so it seems like the precedent helps Snapchat. Yet, after describing the Dyroff case, the court moves on without applying it at all.

The court advances the uncontroversial statement that “Some federal appellate and trial courts have disagreed with Zeran and its progeny’s broad immunity construction of section 230.” The court says those courts have interpreted the word “publisher” to be consistent with its narrower meaning from defamation law.

Hold up. The Barrett v. Rosenthal California Supreme Court opinion–another reminder that it’s binding precedent on this court–spent a lot of time discussing the definition of “publisher,” and none of it supports the interpretation this judge wants to make. The court doesn’t discuss Barrett here at all. Instead, the court valorizes:

(1) Judge Katzmann’s DISSENT in the Second Circuit’s Force v. Facebook opinion (the court praises that dissent as “influential”–though surely not more influential in California state courts than Barrett v. Rosenthal…?)

(2) the Ninth Circuit’s mess of a decision in Gonzalez v. Google–which remains in limbo after the Supreme Court’s remand of that case, and

(3) Justice Thomas’ DISSENT to cert denial in the Malwarebytes v. Enigma case–recall that’s the screed he wrote “without the benefit of briefing.”

This is what it looks like when a judge goes rogue. The judge bypasses on-point precedent to embrace sketchy secondary sources.

The court then makes another uncontroversial statement: “Limitations on the breadth of section 230 immunity are found in reported decisions deriving from the (1) distinctions between a defendant’s independent conduct as opposed to only posting third-party content and/or (2) when the interactive computer service involvement with the third-party content rises to the level of “development” or even partial creation of that content.” The court recounts, without analysis, the precedents of Lemmon v. Snap, Lee v. Amazon, Doe v. Internet Brands, AM v. Omegle (recall, a C.D. Cal. judge just expressly disagreed with that ruling), and Liapes v. Facebook. Why did the judge recite these cases–several of which aren’t binding on the court–and not the hundreds of conflicting opinions, including the opinions that have expressly disagreed or limited these? ¯\_(ツ)_/¯

The Court’s “Analysis”

After navigating this selective tour through Section 230’s text and precedent, the court’s entire application of the law to the facts in this case:

The court agrees with Snap that it, as an “interactive computer service,” cannot be liable as a publisher of the third-party drug sellers’ content. But the allegations of the SAC do not purport to impose liability upon Snap for publishing or failing to moderate the drug sellers’ content (i.e., to “treat Snap as a publisher”) but instead on account of its alleged independent tortious conduct – independent, that is, of the drug sellers’ posted content.

The allegations assert conduct beyond “incidental editorial functions” for which a publisher may still enjoy section 230 immunity. Additionally, the court finds that the alleged attributes and features of Snapchat cross the line into “content” – as the Liapes and Lee courts found, too. The court rejects, as did the Ninth Circuit in Barnes, Snap’s assertion of “general immunity” under its “based on”/”flows from”/”but for” reading of the scope of section 230.

Recall, Snap demurs to “[a]ll causes of action alleged in the SAC – which are based on content created and exchanged by third parties – [as] barred as a matter of law under Section 230 of the Communications Decency Act.” The demurrer is overruled on that ground as to counts 1, 2, 3, 4, 5, 6, 7, 8, 9, 11, 12, 13, 14, 12 15, 16.

The demurrer is sustained without leave to amend on this ground as to count 10 for “aiding and abetting.” Paragraph 931 of the SAC makes clear that plaintiffs seek to impose liability upon Snap because it “encouraged and assisted each of the above referenced Snapchat Drug Dealers in their use of Snap’s unique product features to sell deadly counterfeit pills …. ” Unlike their other counts, which focus on Snap’s alleged independent tortious acts, count 10 is too intimately tied to Snap’s publication of the drug sellers’ third-party content to operate outside the zone of section 230 immunity.

So, where’s the analysis? This is a typical error by 1L students. The court just states its conclusions without connecting the law to the facts. The court conclusorily says that the allegations relate to “independent tortious conduct,” conduct beyond ‘incidental editorial functions,’ and attributes/features that cross the line into “content.” But how exactly do they do that? Where is that line? Do all of Snapchat’s features and attributes cross the line? Could Snapchat have designed its service in a way that would have defeated these arguments? Instead of explanations, the court simply points to the plaintiffs’ allegations and says “good enough.”

Even more baffling, the court says the aiding-and-abetting claim is covered by 230 because it’s “too intimately tied to Snap’s publication of the drug sellers’ third-party content.” How and why is aiding-and-abetting too tied to content publication, but other claims (such as negligence) aren’t? (Note: the Taamneh decision suggests that the aiding-and-abetting claim would fail without Section 230, but that’s not how the court resolves it).

Products Liability

Having rejected Section 230 for [reasons], the court turns to claims’ merits. I won’t go through all of them because the judge’s resolution was similarly conclusory for most of them.

With respect to the products liability claims, Snapchat pointed out that it’s a service, not a tangible product, so “products” liability doesn’t apply to it. The court sidesteps the extensive precedent on this topic because:

The tangible product versus (intangible) service test is a false dichotomy as applied to Snapchat, at least as Snapchat is described in the SAC. As noted, even the parties struggle to find language with which to categorize Snapchat [Eric’s note: I discuss this more below], but neither “product” nor “service” are up to the job….Snap’s arguments rely heavily on tangibility – a concept the court finds unhelpful to the analysis…

What? The tangible/intangible distinction is foundational to Internet Law and has been the subject of decades of academic discussion and court opinions. [UPDATE: See, e.g., this 50-state survey on the topic of “whether intangible items – chiefly computer software, website algorithms, and other electronic information – is treated as a “product” for purposes of imposing strict liability on their creators.”] Yet, somehow the court thinks it’s a “false dichotomy” and “unhelpful.” OK, if the judge is going to disregard the extensive precedent on this topic, the judge surely has a clever new solution, right? (emphasis added)

The court sees the issue as follows: is Snapchat, with its features, functionalities, and its place in our society and economy in 2023, something (pick your descriptor: product, service, interactive platform, app, website, software, tool) [Eric’s preferred descriptor: PUBLISHER] for which strict products liability does or should apply? Or, using the 25-year-old Third Restatement’s formulation, is the context of Snapchat’s distribution and use sufficiently analogous to the distribution and use of tangible personal property to warrant its being treated under the law of strict products liability?

The court’s answer is: not enough information yet to tell, and the question cannot be resolved on demurrer

It’s pretty shocking for a judge to blow past decades of law on the topic only to conclude “I don’t know, it’s really hard.” Disregarding the precedent is what makes it so hard…

In a footnote, the judge adds:

The unhelpfulness of the tangibility test is further revealed by considering whether Snapchat, to the extent it exists as a collection of electrons flitting among silicon atoms, has mass and thus could potentially be “tangible” by a profoundly sensitive measuring device. The answer, absurdist and theoretical, is “yes.” An electron weighs 9.11 x 10e-28 grams

I’m sure the judge was stone-cold sober when he wrote this, but it’s the kind of statement I might expect if someone wrote a judicial opinion while tripping on acid. Perhaps the judge should have edited out its “absurdist” answer.

The Nomenclature Issue

The court observes that the parties use a range of descriptors to characterize Snapchat, including “service,” “app,” “product”, “tool,” “interactive course of conduct,” “platform,” “website,” and “software.” Nomenclature battles are common in Internet Law, because they directly affect the applicable analogies and that can have significant substantive effects. To me, the best way to characterize Snapchat isn’t even mentioned: “publisher.” Snapchat’s only contribution to the victims’ harms came from Snapchat’s publication of third-party content. That’s why so many similar cases have resolved on Section 230.


This case shows how Lemmon v. Snap is wreaking havoc on Section 230 jurisprudence. The plaintiffs said they weren’t suing on third-party content even though their allegations clearly targeted how Snapchat published third-party content. The plaintiffs simply repackaged Section 230-preempted arguments using different words. The Lemmon v. Snap decision puts a lot of burden on judges to recognize and reject such doctrinal equivalencies. This judge either failed that analytical test or engaged in judicial activism based on his normative beliefs. Either way, if repackaging like this can work, plaintiffs can always bypass Section 230, and its value as an immunity diminishes greatly.

This case will surely be appealed to the California Appeals Court. This opinion has many sharp edges that warrant the appellate court’s attention, starting with its non-engagement with the binding California Supreme Court decisions in Barrett and Hassell. However, in the past couple of years, some California appellate courts have swung hard against Section 230, especially in the Amazon cases. That reduces my hope that the appellate court will save Section 230.

Case citation: Neville v. Snap, Inc., Case No. 22STCV33500 (Cal. Superior Ct. Jan. 2, 2024)