HTTP 200 OK
Allow: GET, PUT, PATCH, DELETE, HEAD, OPTIONS
Content-Type: application/json
Vary: Accept
{
"resource_uri": "https://www.courtlistener.com/api/rest/v4/audio/106506/?format=api",
"id": 106506,
"absolute_url": "/audio/106506/dennis-donnelly-v-des-moines-register-and-tribune-co/",
"panel": [],
"docket": "https://www.courtlistener.com/api/rest/v4/dockets/72034895/?format=api",
"date_created": "2026-09-22T14:39:24.791533-07:00",
"date_modified": "2026-09-22T14:57:13.088285-07:00",
"source": "C",
"case_name_short": "",
"case_name": "Dennis Donnelly v. Des Moines Register and Tribune Co.",
"case_name_full": "",
"judges": "",
"sha1": "40cb41a3b4eb4104823f6d2b41308af538b4cc8c",
"download_url": "https://media-oa.ca8.uscourts.gov/OAaudio/2026/9/253451.MP3",
"local_path_mp3": "mp3/2026/09/22/dennis_donnelly_v._des_moines_register_and_tribune_co._cl.mp3",
"local_path_original_file": "mp3/2026/09/22/dennis_donnelly_v._des_moines_register_and_tribune_co..mp3",
"filepath_ia": "",
"ia_upload_failure_count": null,
"duration": 1941,
"processing_complete": true,
"date_blocked": null,
"blocked": false,
"stt_status": 1,
"stt_source": 1,
"stt_transcript": "25-3451 for the Southern District of Iowa, Dennis Donnelly v. Des Moines Register and Tribune Company et al. All right, Mr. Sir, we'll hear from you first. Good morning, Your Honors. May it please the Court? I think I've already reserved three minutes for rebuttal, if that's okay. You may, but you'll need to stop on your own because the clock will continue to run if you continue to speak. And the clock has started. There we go. Great. Good morning. So I know we're all excited to talk about the First Amendment today, but I think we should probably start with jurisdiction because it is, after all, jurisdictional. So the Class Action Fairness Act was enacted by Congress with a particular purpose. It was to prevent a litigation gamesmanship tactic that would take class action consumer protection cases against nationwide products and try to get around federal courts by suing with a class of people in a particular state under that state's law. And so we passed the Class Action Fairness Act. But Congress recognized at that time that there are class actions that belong in state court because they deal primarily with issues and victims in that state. And so there are a trio of exceptions, two of which are mandatory, that I believe are relevant to removal in this case. I represent a putative class. We have exclusively state law claims. And those claims deal with something that happened in Iowa to Iowans under Iowa law about an event in Iowa. So it seems to me it belongs in Iowa state court. The district court disagreed. And I think the two primary questions the court has to answer today are new to this court. There's no Eighth Circuit precedent interpreting either the primary defendant portion of CAFA's first removal provision or the principal injuries prong of the second provision at issue. So turning to the primary defendant prong, I actually think this is perhaps the easier of the two questions to resolve. And the best way to think about it, I think, is that a primary defendant has to be something more than a significant defendant. And that's because the two terms are used near one another in the same statutory section. And so courts point especially to the Eastern District of Louisiana's opinion in Caruso, which I think goes through this in pretty good detail. The Fifth Circuit's decision in Watson, I think, is good on this, although not quite as detailed. But the basic point is it seems correct that a significant defendant is less than a primary defendant. So the question here is, is Gannett, as the corporate owner of the Des Moines Register, a primary defendant? And my answer is no. It is the corporate parent. It is appropriate to be here because it sets standards for its employees at the company. But it is not a primary defendant in that it did not have a major role in the events that are at issue. Well, didn't the complaint allege that they were responsible for coming up with the poll and set standards for the use of polls? No to the first and yes to the second. So Gannett, as the corporate parent, did not sponsor the poll. The Des Moines Register Company, which is a separately incorporated Iowa company that's owned by Gannett, was responsible for both conducting the poll through its contracted pollster and then publishing the poll. What Gannett did, you're correct, Your Honor, is set standards for its newsrooms. And one portion of those standards is that it charges its employees to be careful with, among other things, polling data. We obviously believe they weren't careful here. But I would submit that setting standards for a nationwide corporation is not being a primary defendant, right? It's being a corporate parent. And there's a string cite in our brief of probably 6, 8, 10 cases. Do you hold them accountable for the damages you seek, the relief you seek? Yes. Iowa is a comparative negligence state. So eventually I think that would be a question for the jury. But yes, they are here because we hope to get money out of them, if that's what you're asking. So I think the primary defendant prong is best seen in light of the significant defendant prong nearby. The second sort of question that the case presents in this regard is the principal injuries prong. And that is, I think, probably a little bit more of a messy state of the law nationwide. There's kind of a Third Circuit case. There's obviously a Fifth Circuit case, Cheapside Minerals, which is what the district court relied on most extensively. And I'm always hesitant as a lawyer to come in and say, y'all should create a circuit split. I realize that is asking a lot. But respectfully, I think the Fifth Circuit got it wrong in this case. And if it's any consolation, I think the Fifth Circuit kind of created a circuit split with the Third Circuit anyway. The Third Circuit, I think, takes a more appropriate approach to defining principal injuries, which is to say that the injuries need to be concentrated in the forum state, but they do not need to be exclusive to the forum state. The Fifth Circuit's interpretation in Cheapside Minerals is that every single member of the putative class needs to be in the forum state in order to qualify for the exception. That, I think, is contrary to the legislative history, but I realize that's not the best guide in this instance. It's just the guide that the Third Circuit relied on in, I'm going to say it, Voting Jar. I think that's right. I think the better way to think about it as a matter of text is that it's the principal injury to the class, which is to say that the class as a whole needs to be injured in the forum state, but it's not that every single member of the class needs to be in the forum state. I think if you look at some of the examples you'll see in district court cases interpreting the provision, you know, they're thinking about something like an explosion at a factory, right, where a whole bunch of workers are injured. And so the event takes place in the forum state, but those workers, you know, might be in multiple neighboring states and commute in, something like that. And so the injury to the class takes place in the forum state, but not What's the injury alleged here? So the injury alleged here is that the Des Moines Register committed consumer fraud and that the injury to consumers is their subscription price that they paid. One of the elements of Iowa consumer fraud is that you have to have a specific dollar amount associated with the injury. And here that dollar amount, that injury, is the subscription price. And there are people across the country that paid the subscription price? Gannett has submitted an affidavit saying that, yes, Your Honor. I think those are my big thoughts on jurisdiction. I'm happy to answer any other questions on it. Otherwise, I know we're all really here for the fun part, which is not Iowa consumer protection law. It's the First Amendment. So in that case, I'll transition us to the First Amendment and talk about, I think, the two cases that matter most in determining this case. The first is the Animal Legal Defense Fund case. I think this Court's opinion in Animal Legal Defense Fund sets very clearly that actual malice is the correct standard for pleading a claim based on false speech in a private damages context. The other side, I think, is out to make this sort of categorical immunity for newspapers and pollsters and others to say that there are very limited exceptions to the First Amendment, such as defamation. And if you don't fit in one of these extremely narrow buckets, then it's all fair game. And I just don't think that that position can be reconciled with Animal Legal Defense Fund. I think that opinion is very clear that historic torts, including the torts that we've pled here, are not preempted by the First Amendment. It simply sets a standard that you have to plead actual malice. And that, I think, the district court agreed on, that actual malice is the standard. She simply, I think, erred in finding that we hadn't pled enough to show actual malice. So that's the second important case in determining the question we've presented, and that's Nelson Auto Center. All right, what Nelson Auto Center says is that in order to plead actual malice, you need to introduce enough in your pleadings to raise a reasonable inference. It's not to say you're going to win. It's just enough to say you've raised a reasonable inference of malice. Federal Rule of Civil Procedure 9 says malice can be pled generally. Our burden is eventually to show that the Des Moines Register and the other defendants either knew subjectively and intentionally that the poll was wrong, or at least had subjective substantial doubts about what they were doing. That's our burden at trial or at summary judgment. In order to get there, though, we don't have a window into the defendant's mind at this stage. There's no smoking gun e-mail that's already public. Those sorts of things come out during discovery. Or ultimately, we have depositions, and we, through the course of depositions, are able to ask, did you have subjective doubts about whether or not this information was accurate, and did you decide to run with it anyway? So the question for this Court right now on a motion to dismiss is, have we pled enough to raise a reasonable inference that they acted either knowingly or recklessly subjectively? Well, how do you do that? Well, what we have done is try to stitch together as many different elements or indicia of fraud, elements or indicia of subjective doubt as possible. And I think there are two in particular that I would highlight for this Court. One is just math. There's a one in 3.5 million chance that the poll was an honest mistake. That is to say, polling is social science, right? That on the day the poll was taken, the voters of Iowa believed this about the election within 3.4 percent, 95 out of 100 times, right? There's a margin of error at a confidence level. 3.4 percent, 95 out of 100 times. And so if you look at being off by 16 points, there's a political scientist who's done this, who's a former member of the Iowa State faculty, the chances of it being an honest mistake, that they just called the wrong people who were misrepresentative of the people of Iowa is one in 3.5 million. It seems like running with something that has that sort of likelihood of being erroneous would raise a significant reasonable inference of subjective doubt or knowledge. The second piece of information is that at the time, both Dr. Selzer and the Des Moines Register described the poll as stunning, shocking, surprising. Well, often when we use those words in our regular everyday lives, when something is shocking to us, sometimes our response is to say, gee, I wonder if it's true, right? Like that is a very normal human response when you get information that you don't expect, information that is contrary to everything else you know about a topic. When you describe it as shocking, you wonder is it the information is accurate or is it that everything else I know about the world is inaccurate, right? That's a very human response. And so when the Des Moines Register itself and Dr. Selzer herself have described the poll in this way, it seems like very good evidence for us at a pleading stage to say it was a reasonable inference that they had subjective doubt about it and they decided to run it anyway. Obviously, there are a number of other facts that we have in our complaint, which we lay out in our brief, but I would point to those two in particular as sort of the best to say that there's a reasonable inference under Nelson that we have raised this. And I see I'm three minutes on the nose, so I am happy to take any questions. Otherwise, I'll save my time for rebuttal. Very well. Thank you, sir. Thank you for your argument. Mr. Yoshimura, we'll hear from you. Thank you, Your Honor. May it please the Court, I'll be discussing the jurisdictional issues and time permitting some of the Iowa claim issues, and I'll reserve the rest of the time for Mr. Korn Revere to speak with you about the First Amendment. When it comes to CAFA, respectfully, I think Mr. Soar's summary of Congress's intent was incomplete. If you look at Section 2 of the legislation that contains the finding and purpose of CAFA, in which Congress stated that its purpose was to, quote, restore the intent of the framers of the United States Constitution by providing for federal court consideration of interstate cases of national importance under diversity jurisdiction, and specifically to prevent plaintiff's attorneys from keeping cases of national importance out of federal court. This case absolutely belongs in federal court. It's a class action on behalf of a putative class spanning all 50 states, concerning the nationwide publication of a poll concerning the U.S. presidential election in 2024. That certainly fits the bill. And, in fact, the subject matter jurisdiction of the federal courts is not contested. The structure of CAFA provides for elements to confer jurisdiction. Those have been conceded. What we are talking about today are these so-called exceptions, which, to be clear, do not divest the federal courts of jurisdiction. Rather, they're a mandatory abstention doctrine that's built into the statute. And the reason that that distinction is important is because it informs the framework in which we consider those exceptions. In the Westerfeld case from this court back in 2010, that framework is set out very clearly. First, we assume that the statute contemplates broad federal jurisdiction and that the exceptions are narrow, which is why there's a burden-shifting mechanism, and it's Mr. Sirr's obligation to prove the elements of the exceptions. But, importantly, and as confirmed by Judge Smith, your opinion in the Hargett case in 2017, we resolve any doubt as to the applicability of those exceptions against remand. And as the Westerfeld case said, that's because we're avoiding using the exceptions as jurisdictional loopholes to evade CAFA. So, as Mr. Sirr said, the two issues here are whether Gannett is one of the primary defendants in this case and whether the principal injuries were incurred in Iowa or in additional places besides Iowa. Does the record indicate what percentage of non-Iowa plaintiffs are a part of this class or would be a part of it? It does not yet. And the district court's ruling was clear that that's because she doesn't need to know that in order to reach conclusion on those two elements. Neither of them require a quantitative determination of how many members of the class are imputed in this petition. That's because when, I guess, we can speak specifically to the principal injuries prong because that's where the quantitative and qualitative dichotomy comes into play. The Cheapside case that Mr. Sirr mentioned, we would say, is the model for this court to follow. In that case, appended by Judge Judith Jones out of the Fifth Circuit, she did a very robust textual analysis of the statute to determine exactly what principal injuries means in context. And what she noted is that the quantitative part of the exception is a different element, and that's the two-thirds of the members being in the state requirement. And so Judge Eppinger in this case below said, I actually don't need to know the answer to that because under the Fifth Circuit case, the question about principal injuries has to do with the qualitative nature of the injuries here. And we've just heard from Mr. Sirr. I think it's pretty clear the only injury at issue are the subscription payments that subscribers from around the country and all 50 states paid to the Des Moines Register. Because of that, there's really only one type of injury if we're doing the qualitative analysis. There aren't multiple injuries that we need to weigh. Because there's just the one, the question is, where was that injury felt? And I think Mr. Sirr's reply brief notes that we're talking about the injury to the class as a whole. That injury occurred everywhere in the country. It doesn't matter if it was concentrated in one state or another because the principal injuries prong is not the quantitative part of the exception. It's the qualitative part of the exception. So as Judge Jones found, based on that textual analysis, when some plaintiffs sustain their primary injuries in the form state but others do not, this element, the principal injuries element, is not satisfied. Counsel, are there any types of injuries other than those relating to real property that would be able to satisfy this exception the way they're interpreting it? It's an interesting question. I think that most of the cases are about property. And, in fact, the example that Judge Jones wrote about in the Fifth Circuit case had to do with that exact issue. I think in this particular case, we don't need to cross that bridge because there's only one type of injury and only the one injury. But why would principal mean what you're arguing in light of the two-thirds prong? Why would Congress write it that way? Because quantitatively, they want a certain threshold of the members to be within the state. And then qualitatively, they want to make sure that the injury actually occurred in that state. But aren't you interpreting principal in a quantitative way by looking at a number of how many people are in the state? No. So what the Fifth Circuit is suggesting, and we argue it's the best way to do it, is to look at the number of injuries and then rank them amongst each other, which ones are the principal for the case. So the example that Judge Jones wrote about in the Fifth Circuit case is, you can, for example, imagine a case where bodily harm occurs in Texas and then there are related medical bills to be paid in Florida. There you've got two different types of injury, and so you have to weigh them against each other. What you're not doing is checking how many class members are in these places and then saying how many dollars are associated with them. That quantitative piece is left to the first prong. If I could switch gears real quick to the Homestead exception. So you've talked a lot about the Fifth Circuit under the other exception, but on the Homestead exception, you haven't talked a lot about the Watson case. Doesn't the Watson case say that your primary defendant can't meet everybody, and that's essentially what you're arguing in this case, that all defendants are primary defendants? I don't think that the Watson case stands for that necessarily. And, in fact, the Vodenichar case, the Hunter case, a number of other cases that we've cited state the opposite, that depending on the number of defendants and what they're alleged to have done, all of them may be primary defendants or only some of them. Certainly, it can't just be a singular defendant. There's not one primary defendant because the statutory text says that there can be multiple ones. We think that under any definition, whether it is some or all, Gannett has to be a primary defendant in this case. Judge Smith, as you were asking about the allegations in the petition, I'd just like to point you to a few of them. In paragraph 86, plaintiff has alleged that Gannett exercises editorial control over its affiliates to an extent unseen in other chain newspapers, and that the tie between Gannett and the register is specific to the poll in paragraph 72. And the causes of action, they specifically allege that Gannett is liable based on its editorial control of the register and that it seeks to hold Gannett liable for its influence over its local paper's newsroom decisions. And, in fact, the petition then cites a case called Sandman that confirms that that's the intent here. So Gannett is not merely on the hook as a parent. It's being alleged to have caused the publication of the poll. It is a primary defendant, and the exception does not apply in this case. For that and the reasons stated in our papers, we would ask you to affirm the district court's order in full. Thank you. Very well. Thank you for your argument. Mr. Korn Revere, we'll hear from you. Thank you, Your Honor. Your Honor, may it please the Court. The district court was correct when it described plaintiff's theory of liability as absurd and dismissed all claims with prejudice. Now, the plaintiff's argument essentially is that any subscriber can sue a publication if they become frustrated in believing that their news product isn't accurate. Now, just imagine how broad a theory this is. It means that subscribers to Fox News can sue if they believe that the product isn't fair and balanced, or a New York Times subscriber could sue the newspaper if he becomes convinced that the Times doesn't, in fact, print all the news that's fit to print. But the law doesn't permit such claims. That's why in Brandt v. The Weather Channel, the court dismissed a claim for a bad weather forecast that resulted in injury. The court described the claim as a novel and unprecedented expansion of the scope of tort law with no limiting principle, and the same is true here. That's why the plaintiff can't name a single case supporting the theory of liability in this case, and every court that has dealt with the claim that news should be treated as fraud has dismissed the claim. That includes Hollander v. CBS News, where the court dismissed wire fraud claims based on allegedly false and misleading statements about then-candidate Donald Trump. It also includes Washlight v. Fox News, which dismissed claims under the Washington Consumer Protection Act based on claims that Fox News had falsely reported about COVID-19. And it really doesn't matter which of the theories of liability that Mr. Donnelly tries to pursue. None of them provide a basis for liability, and all fail at the threshold. There simply isn't a false statement here based on an inaccurate poll. Now, Mr. Donnelly claims that the poll results were way off, and using various adjectives to describe how far off, and that it violates professional standards even to publish an outlier result. In fact, if you look at pages 37 to 38 of his opening brief, he says that a shocking result alone should be enough to show recklessness on a motion to dismiss. This is really nonsense, and it really just is an indication that Mr. Donnelly does not know how polling works. As a matter of fact, if you look at the complaint, paragraphs 101 to 104, it tries to make the claim that publishing an outlier violates the American Association for Public Opinion Research Code of Ethics, citing only snippets of that. Well, as it turns out, the association filed an amicus brief in this case, and they disagree with that conclusion, stating things like outliers are statistically inevitable, that stunning outliers may actually accurately reflect voter trends that other polls have missed, and that there is no way to know at the outset of a poll whether or not a given outlier is wrong or whether it captures something real that other polls have missed. And simply citing to the margin of error published with the poll shows that a poll is a deviation from its stated margin of sampling error. It can't unilaterally be characterized as a falsifiable statement of fact. There simply is no false statement here. And it's why the district court correctly concluded that the results of an opinion poll are not an actionable false representation, merely because the anticipated results differ from what actually occurred. So are you suggesting, then, that the court not need consider the First Amendment because you say the complaint doesn't plead the elements of the torts? Is that the point you're making? I think it would be a good thing for the court to issue a ruling on the First Amendment claims, as the district court did. But it doesn't need to because all of these claims fail for failing to satisfy the elements of the claim, beginning with whether or not there's a false statement of fact. But all of those cases, like the Brandt case, also deal with the First Amendment issues after dealing with the failure of a state law claim because they talk about what would happen if state law permitted such a claim. And if it did, it would bump up directly against the First Amendment. Now, Mr. Suram tries to rely on cases saying that they can, if it is a cognizable cause of action, then falsity can lead to liability. And he goes straight to the actual malice test. But even on the allegations that he has made, as the Gannett briefs point out, this fails under reckless disregard, that the facts that he has alleged don't add up to that. And essentially what he's saying is that if it fails to meet professional standards, then it can be evidence of actual malice. Why would actual malice be the correct standard here? Well, we've said that it's not. Actual malice is a strategic protection that applies to areas of unprotected speech. And so we have, from the beginning, resisted the idea that actual malice shall be applied. You don't get actual malice if you don't state a claim. But if you do, as Mr. Sur has asked, then it fails even under that standard. But our argument has been that this is not, okay, first it's not a false statement, but even if it were, that it wouldn't be a cognizable claim under the First Amendment. And that's because the case is based on a singular fallacy. And that is they claim that all frauds, if all frauds involve false speech, that any false speech can then therefore be a fraud. But the Supreme Court has disagreed with that proposition. In Madigan v. Telemarketing Associates, for example, it says just labeling an action as one for fraud doesn't make it so. And that a claim should be quickly dismissed if it doesn't plead the elements of fraud. And that's been our argument here all along, that there is no general First Amendment exception for false speech in general unless it falls into one of the previously recognized categories. And this one does not. Counsel, why not? Because isn't this different than the way you're trying to characterize it? The newspaper wasn't just reporting on news at a press conference or something. They were creating the news through what plaintiffs have said is a product. So why is that not different than the cases that you're relying on? They were trying to measure voter sentiment, as newspapers and pollsters have done for decades. And they made their best estimate. But having that forecast miss the mark and be inaccurate isn't a false statement and isn't the basis for liability. I know that my time is up. Unless you have any other questions, I will ask the Court to affirm the decision below. Very well. Thank you for your argument. We'll hear Roberta. Good morning again. I'll try to zip through just maybe four or five quick points. I'll start, Judge Smith, with the question you asked earlier. Is the exception limited to just property? And the two examples I would point to from our briefing, there's one case in California and there's another case in Montana. The California case dealt with donors to a nonprofit. The Montana case dealt with retirees. And even though most of the donors lived in California and even though most of the retirees lived in Montana, the courts nevertheless found that the principal injury was in those forum states. So I recognize it's district courts and not the Fifth Circuit, but I think those are useful examples that it's not just limited to real property. Perhaps moving on to Mr. Kornerbeer's arguments. First, he says there's no limiting principle. Actually, there are two limiting principles. The first is the actual malice standard. But the second is that a statement has to be falsifiable. To say all the news that's fit to print is not a falsifiable statement. To say that we have acted in accord with our ethical principles, one of which is that we are careful with polling results, well, that's a question you can take to a jury. To say that the information we have presented is accurate, well, that's a falsifiable statement that's different than kind of mere corporate puffery. So is actual malice the correct standard for these torts as opposed to a defamation case? So I think this Court's, I hope this is the right answer, so I think this Court's holding in Animal Legal Defense Fund is that when there is a cognizable legal harm, regardless of what tort label we apply to it, the correct standard for false speech is actual malice. In other words, we don't have particular torts like defamation and slander, and then anything else is fair game, right? So in Animal Legal Defense Fund, it's trespass. In the second Animal Legal Defense Fund case, it's lying on an employment application. Neither of those have been recognized by the United States Supreme Court in any precedent as exceptions to the First Amendment or cases where actual malice applies. Yet this Court says, I think correctly, that defamation and slander, the sorts of cases the Supreme Court has recognized, are just examples of the underlying principle, and the underlying principle is that harm to private interests can be pursued through private cause of action, and the way to do that is that you have to, the way that we protect the First Amendment interest while also allowing somebody to pursue an injury to their private rights is we apply actual malice. Do you think the Animal Legal Defense Fund case talks about actual malice? No, actually it doesn't. I think the Animal Legal Defense Fund case rejects Mr. Korn Revere's assertion that there are particular narrow categories like defamation and that any other private tort action is preempted by the First Amendment. I recognize you wrote it, so you're the authority on it. Well, it says what it says. I'm not sure it says what you say it says, but we'll have to look at that. That's fair enough. Thank you very much for your argument. Thank you, sir. Thank you to all counsel. The case is submitted and the Court will file a decision in due course."
}