…but I do not think it means what you think it does. This morning, the Supreme Court said as much regarding "contributory infringement" in Cox Comm. Inc. v. Sony Music Entertainment, Inc., No. [20]24–171 (25 Mar 2026). The hard part of understanding this opinion — which is largely consistent with techbros' preferences — is determining which word is drawing Inigo's ire.
Justice Sotomayór's concurring-in-the-judgement opinion has the better of both the doctrinal analysis and focus: It's not the meaning of "contributory infringement" that is at issue, but of "intent." Both Justice Thomas's seven-justice controlling opinion and Justice Sotomayór's (for herself and Justice Jackson) focus on failure to prove intent as sufficient basis to, in this instance, absolve Cox Communications. Where the opinions differ is on a question not actually necessary to the result: Whether the misnamed "secondary liability"1 is limited to common-law concepts of "contributory" and "vicarious" liability established under the 1909 Act (and, less clearly, its predecessors) and imported into interpretation of the 1976 Act.2 Justice Thomas would limit all secondary liability to acts/omissions fitting comfortably within contributory and vicarious infringement as presently defined; Justice Sotomayór would be open to other varieties of secondary liability that do not, although such other varieties are (in her opinion) disclaimed as unnecessary to decide this case.
The main difficulty here is that, in the abstract, Cox Communications deserves to lose — but not on a traditional meaning-of-copyright-infringement-at-common-law (or under the 1909 Act) analysis. That Justice Sotomayór's opinion has the better of the argument is illustrated by a hypothetical welcome package for new subscribers that includes a link to the Tor browser and a tutorial on BitTorrent, and asking whether such a welcome package would constitute sufficient proof of intent to proceed on a contributory-infringement theory. More to the point, though, is whether assistance by technical support personnel in response to specific customer inquiries ("My connection isn't reaching The Pirate Bay, what am I doing wrong?") would also implicate corporate intent. Instead, though, failure to actually implement a statutory requirement intended to both protect copyright holder interests and provide a pre-liability-consideration safe harbor for service providers — the DMCA, and in particular § 512(i) — is treated as utterly irrelevant to the liability question, instead of intimately intertwined with it.
But everyone wanted easy answers. So that's what they got, even though they didn't actually answer the question lurking underneath. Everyone wanted a bright-line rule in a field of constantly evolving standards (the very definition of "the Progress of Science and useful Arts"). Cox gets this wrong by answering the wrong question. I can't offer "the right" answer, but I can say that this Court was too solicitous of non-copyright-related uses of internet connections in evaluating the meaning of "intent" as to copyright-related uses of internet connections.3 Indeed, the Betamax, Tasini, and Grokster opinions necessarily rest on the premise that considerations of what infringes copyright, and what regulates too much, must limit themselves to copyright contexts and defenses/privileges within copyright contexts.
- The corrolary/opposite of "secondary" is or should be "primary," if one is being linguistically consistent. That, however, is not the term of art; instead, we speak of "direct" as the opposite of "secondary," when "secondary" itself would be better called "indirect." That this runs right back into Inigo's objection to "inconceivable" is both nerdy wordplay and a conceptual objection to the way the misnaming influences the way one thinks about the underlying concepts.
- The opinions are silent on whether this limitation is consistent with US obligations under the Berne Convention, to which the US acceded in 1988. On the one hand, this is entirely understandable because "consideration of treaty obligations" was not in the record. On the other hand, Justice Thomas's opinion is overreaching by declaring a limitation on alternative theories not actually argued. Under ordinary circumstances, the best way to deal with this is via amending the statute — probably § 501 but possibly by tweaking 106 (separately or together). My trust in Congress — and, in particular, Congress's ability and willingness to even listen to actual creators as distinct from transferee copyright holders — is less than 4'33".
Interestingly and reflexively with the theme of this blawg piece, almost all common analyses fail to engage with the title itself:
4'33" = (4x60)+33" = 273"
that is absolute zero, –273[.15]C, which at the time of Cage's composition (1952) was just creeping into awareness outside of chemical physics. This leads to the deeper question of whether the piece is about "absolute silence at absolute zero" or "ambient/background noise at absolute zero". That, however, is a "two cultures" argument for another time, however much it also exposes the unconsidered-case problems with the Cox rationale itself.
- Here, I don't think the Court needed to issue an "advisory opinion" — a doctrine that I think has long outlived its usefulness, because we're not operating in a vacuum any longer. It would have been sufficient to explicitly call for intentional consideration by Congress, instead of relying upon Congressional silence plus being "loath to expand such liability beyond those precedents" (slip op. at 7, emphasis added), which is rather self-contradictory and neglects that there were both a predecessor to this Act in the same space and externally-imposed treaty obligations. It most emphatically is the place of the Court to tell Congress to do the hard parts of the homework; an "advisory opinion" would be doing the homework for Congress. This is especially so when "loath to expand" is against not specific categorization in the statute, but complete Congressional silence as to what "infringement" means (see § 106): Congress didn't do its homework because it didn't recognize that this question was in it, or was delegating that question to the teacher. Which turns on what "silence" means…