Within an hour or so of the Supreme Court handing down its landmark ruling in Spokeo v. Robins, 136 S. Ct. 1540 (2016), I had declared the case to be the end of TCPA class litigation.

In a nutshell: a violation of the TCPA does not necessarily cause harm. Indeed a violation of the statute cannot cause harm unless a recipient does not want or welcome the content of the call. And not all uninvited calls will be unwanted. Thus there is a clear gap between the protections of the statute and the harm it was designed to prevent. Spokeo tells us, therefore, that not all violations of the TCPA will give rise to Article III standing and the criteria for determining whether or not a call is actionable will rest within the mind of a called party. As such TCPA class actions must wilt and die beneath the foul-smelling haze of individualized standing inquiries turning on the subjective mindset of class members. In short, Spokeo is to TCPA class litigation as RAID is to cockroaches.

But sometimes cockroaches aren’t so bad. Even I must recognize that sometimes calls are so annoying and improper that forcing someone to sit through them necessarily causes harm. Fake political survey calls that are actually sales pitches for Caribbean Cruise Line, Inc. likely fit the bill. (Alas, another client I’ll never do work for.) And a TCPA class action suing to prevent millions of such calls might be a cockroach big enough and worthwhile enough to keep around for a while. And that is exactly the case Judge Kennelly faced in Aranda v. Caribbean Cruise Line, Inc., 2016 WL 4439935 (N.D. Ill. Aug. 23, 2016).

And so the Court has a problem. On the one hand I have no doubt that the Court easily recognized how the application of Spokeo to a TCPA class action ought to work. Indeed the analytic framework adopted in Aranda is nearly flawless in that it mirrors my own. (Please keep the snickering to a minimum.) But on the other hand the Court was facing what may be the biggest baddest cockroach ever certified. A case involving millions of folks that (allegedly) received really inappropriate phone calls that were designed to deceive them in addition to annoy them. And all for the purpose of selling cruises to a disinterested populace—which was literally what the TCPA was designed to prevent in the first place.

On top of that, as Judge Kennelly tellingly chronicles in the Aranda decision itself, the defendants (allegedly) responsible for these destructive calls have been fighting the Court tooth and nail for years. The class was originally certified two years ago. And—after an unsuccessful trip to the Seventh Circuit and losing dual summary judgment motions—the Caribbean Cruise-aligned Defendants still asked the Court to reconsider and de-certify the class on the basis of Spokeo. Talk about dead on arrival. By this point in the litigation the Court had no doubt reviewed thousands of pages of evidence and briefing and spilled gallons of cyberink writing lengthy and well-considered opinions painting Defendants into a corner. Indeed, the Seventh Circuit specifically adopted Judge Kennelly’s reasoning in his earlier certification decision—Birchmeier v. Caribbean Cruise Line, Inc., 302 F.R.D. 240, 250 (N.D.Ill.2014)—as part of its own landmark ascertainability decision in Mullins v. Direct Digital, LLC, 795 F.3d 654, 657 (7th Cir. 2015). No way a little Supreme Court decision was going to undo all that hard work. That ship had sailed.

So if this is a classic case of bad facts making a good judge make bad law, why did he have to make such bad law? Intellectual honesty. The Court’s reasoning is right all the way up until it is—in my humble opinion—wrong. But even when it was wrong it was still wrong in the right way. Here’s what I mean.

The court correctly addresses Spokeo all the way through. For instance, it correctly identified that if a violation of the TCPA “just as easily” may not cause harm then “it is not sufficient simply to allege that the statute… was violated.” Aranda at * 5. The decision also correctly analyzes whether the procedural protections of the statute are “attenuated” from the harms Congress sought to prevent so that a gap exists between the procedure and the substance. Id. That is, the Court correctly tees up the metaphysical question—is there a gap between the protections of the statute and the harm it was designed to prevent? This is the same rigid black-and-white analytic framework I proposed. There are only two options—either there is a gap between the statute’s protections and the harm it was designed to prevent or there is not. It’s a binary problem. Only a “1” or a “0” permitted. A yes or no. And the results flow starkly from that determination. If there is a gap—no more TCPA class actions, ever. They’re gone. If there is not a gap—Spokeo is essentially valueless in TCPA cases. There is no middle ground.

Unfortunately the Court’s answer to the big question is the opposite of what one would expect: “There is no gap—there are not some kinds of violations of Section 227 that do not result in the harm Congress intended to curb…” Aranda at * 5.

But that doesn’t seem quite right. If your mom calls you without your consent from her smartphone she just violated the TCPA. And of course you were not harmed by that call. So of course there’s a gap. The TCPA prevents uninvited calls but lots of times we may welcome a call from some attractive stranger, or a charming creditor that has our best interests at heart, even if we didn’t expressly request the call in so many words.

But of course the Court could not acknowledge that gap in Aranda. To do so would be to admit that the entire case needed to be de-certified. And then the bad guys would win (allegedly). After all, with respect to the at issue in Aranda there probably isn’t a gap. All of those calls likely melted someone’s brain a little and brain melting is a form of concrete harm for Article III purposes.

But that’s where the intellectual honesty part comes in. The Court couldn’t simply say these calls cause harm but other calls might not. That’s inconsistent with the proper (and very rigid) analytic framework set forth above. Again, either there is a gap or there isn’t. And there couldn’t be a gap in Aranda. So once the Court made up its mind that “no gap” had to be the answer in that particular case, the remainder of the decision wrote itself. The Court also and necessarily had to find the protections of the TCPA are “substantive, not procedural” and that “unsolicited telephonic contact [necessarily] constitutes an intangible, concrete harm.” Aranda, at *5-6. And so we have a very well-considered decision reaching a very unlikely result—that calling someone without their permission necessarily causes them harm.

Again common experience suggests otherwise. Calling a person walking down the street is no different than saying “hi” to that person in the moment. And yes, it might momentarily distract him or her from catching Pokemon, but is that really a judiciable harm? And maybe all uninvited telemarketing calls received by a person in the quiet environment of one’s home necessarily cause harm—maybe—but surely not all uninvited calls for any purpose made under all circumstances do. And so we’re back to that pesky gap. And that means TCPA class actions are dead.

So what is to be made of Aranda for those of us still fighting the good fight? The heart of this decision remains the Court’s finding that “the receipt of unsolicited telemarketing calls…by their very nature invade the privacy and disturb the solitude of their resipience.” Aranda at *5. For those of us that primarily defend informational callers, therefore, we have a pretty stout basis to distinguish Aranda.”That was a telemarketing case” we can say to the Court with an air of nonchalance “Romero provides the proper analysis here.” And most courts would probably agree.

But the real truth is that the Aranda court was facing a RAID-resistant supercockroach and that—in all likelihood—is why the decision came out the way it did. And of course there’s a gap.