Oral argument — United States v. Jose Belmonte Cardozo

Our second case is 25-4239 United States v. Cardozo. Mr. Richman. Good morning, your honors. May it please the court, my name is Todd Richman. I represent Jose Belmonte Cardozo. As I see it, there are essentially three issues, interrelated issues in this case. First, whether a manual search of an electronic device at the border requires at least articulable individualized suspicion. Second, whether at least that level of suspicion existed here and whether it related to an ongoing transnational offense. And third, depending on the answer to those questions, whether the search here was conducted, excuse me, was authorized by binding precedent. All of that, of course, assumes that this even is a border search. If the court were to find that it is, was a search done to further the government's generalized interest in law enforcement, then the standard Fourth Amendment law applies and I'd submit that the Agba Kayin would control what would have been required. So why would you say it's not a border search? Well, if the court were to follow the analysis that the Agba Kayin court followed, that this was essentially a search directed at domestic crime. It occurred at the border. But, you know, if you read the first paragraph of the government's brief, it makes very clear what they think this case is about. None of it is, you know, the facts that they were investigating or wanted to investigate Mr. Belmonte Cardozo. Well, if the search occurred at the border, that would seem to me simplistically that that's a border search. Now, it might be a routine search or a non-routine search, but I don't see how you get away from the fact this is the first entry into the country. I understood, Your Honor, and I didn't intend to argue that at any length. Agba Kayin was at the border like this and the court found that the mere fact that it was at the border did not make it a border search and held that standard Fourth Amendment law applied. So there is the fact that it's at the border is not in itself determinative. But I was intending to move past that. I was not. It's not my principal argument. I agree it turns principally, or at least our argument has turned principally on, is it a routine or non-routine search at the border? And that, of course, relates to the level of privacy intrusion. And our argument is that... Am I correct that all of the circuit courts of appeal that have examined this have determined similar searches to be routine? They have. A few of those are based on essentially pre-Riley law. They're all post-Riley decisions. A few of them were bound by pre-Riley law, though, I think. And one, of course, is a circuit. I'm sorry to cut you off. Well, I don't know. I mean, they may have had some pre-Riley law, but Riley's the Supreme Court. And if that changes things, you have to follow the Supreme Court. And the decisions that Judge Agee mentioned, I think, all took place post-Riley. I think there are, yes, several post-Riley. So I agree with that. I would submit that those cases did not give proper weight to the level of privacy intrusion that cell phones, modern cell phones... So in basic terms, you would say all those other cases are wrong? I think this court should independently examine it. I do think they're wrong. Every modern device, whether you search it manually or forensically, every modern smartphone, in a brief search, you're immediately going to see all these categories of information. Browsing history, location history, text messages, contacts, photos, videos, and more, of course. The list could go on and on. And every one of those, even in a brief search, implicates and can tell about a person's most personal concerns. Political beliefs, religious beliefs, medical information, financial information, sexual preferences, and on and on. All of that is the same in a manual or forensic search. Well, the difference, though, is that, and maybe this isn't such a difference, as significant a difference as it may have been some years ago, but there is a difference between a manual search, which just takes a little bit more time, and a forensic search, which can, you know, is limited only by the power of the computing technology that's being used. So, as a practical matter, I think the government argues that, yeah, it's possible, I suppose, to intrude as deeply as you suggest into a phone at the border in terms of a manual search, but the practical reality is that a law enforcement officer simply is not going to have the time to do that. So, Your Honor, I take issue with that on a couple of levels. I would submit that largely what we're talking about in a difference between a manual search and a forensic search, it's a different methodology. It's not particularly different in terms of the level of privacy intrusion. But there is some technological differences, aren't there? I mean, doesn't a forensic allow to recover, you know, deleted information? So, they're not one-to-one. Some do, although the one in Coleshoose did not. The one in Coleshoose was literally limited to the same information a user could see. But what I would point the court to first is, right, Well, the defendant in Cole's search never challenged the initial search. Correct. Just the forensic search. Which we held was non-routine. Right. Well, when you look at the Supreme Court's cases that, you know, limited, but from what we have, where they found a non-routine search, it's involved not property, but the person. And so, if you have a body cavity search or detention for medical purposes to get somebody to expel drugs, they found that to be non-routine. But I'm not aware of the court ever finding a property-related search to be non-routine. I think that's an accurate observation. I don't think it should control. If you look at Riley, which is obviously a different warrant exception, but Riley was about manual searches. The Riley search occurred in 2009, 17-year-old technology versus today. They made a categorical rule that even then, based on a 2009 smartphone, that even a brief manual search of it, even limited to the information related to the arrest, was too much, not 30 seconds. But the court there, and I know you can see this, wasn't weighing the counter-interest in the protection of the border, which the court has said that interest is at its zenith in terms of the sovereign being able to protect and prevent and deter contraband and other things from coming across. So, you've got to balance those two interests, right? And Riley doesn't do that because it's not a border search case. That is absolutely true, and it's a different exception. I would submit, as a number of courts have said, that digital contraband is categorically different, not that there's not an interest in intercepting it, but if you intercept all the drugs coming into the country, you're going to largely stop the drug problem. If you intercept all the stuff that's coming in on phones, you're almost going to have no impact on the circulation of child pornography. One other separate issue, back to the privacy interest. I appreciate your Riley argument. Riley certainly, you know, is, as you note, is different in terms of the exception and not at the border. But even on the privacy interest, I'm trying to think about it a little bit. I think it helps use, and certainly people have an interest in all the information on their phones. I get that. But in the border search Supreme Court case, I think the Supreme Court said that people generally have less privacy interest when they're crossing the border than they may be when they're just engaged in normal activity. I mean, wouldn't that aspect of the privacy interest still be in play? I think that's certainly part of the weighing, absolutely. But cell phones, I think as both Carpenter and Riley make clear, they're a category of their own because of just the absolute depth and breadth of information that one can get, even just scanning, you know, holding the phone in your hand and scanning through it if you have the passcode. And they have on-board capabilities that are essentially forensic. You can look through them in just a couple of minutes and get to all the most private information you might be looking for. So I take Your Honor's point, but the privacy interests are just extraordinary. They're singular. I think there's almost nothing in modern society that is greater. But if we were to focus on the facts of this case, this was a traditional manual search, wasn't it? It was. A very limited one because it didn't take very long for the officer to find what she was looking for. So I would disagree that it's particularly limited because it was relatively quick. It was limited in time. It went right to the place a person would be most likely to store their most intimate information. And, you know, I think that the correct rule is a categorical rule like the court adopted for search-as-an-instance arrest because even a quick search can immediately go to the most intimate information a person has. It's hard to draw a line at one minute, two minutes, four minutes when it's too intrusive if the court starts down that road. I would submit that the technology itself is just – is essentially the technology itself as recognized in COLSUs required individualized suspicion. And my point really is that this search, in terms of the level of privacy intrusion, is equal to the search in COLSUs. Can I turn your attention to the good faith exception? We've had this discussion here which suggests that this is a very challenging issue. We had COLSUs which left the question open in this circuit. We've got the – before us today, we've got the competing interests of the sovereign at the border versus the very legitimate privacy concerns enunciated by Raleigh. And then we've got the significant number of circuit cases, outside circuit cases, that would seem to suggest that this was okay. So given all of that, why isn't – why doesn't good faith apply in this context? It seems like there was a vacuum here. The officer did what he or she thought was reasonable given that legal vacuum. So why should we punish the law enforcement agency at this stage given the vacuum that existed at the time on this very complex issue? I understood. I agree there's a bit of a vacuum. The test is whether, and this is from Davis Supreme Court, the test is whether the contract is authorized by binding precedent. So the out-of-circuit cases, I would say that . . . What about COLSUs saying the uniform body of law when you don't have binding precedent? I think COLSUs was, as I understood the good faith ruling there, was that in this circuit we had never before required reasonable suspicion of ongoing transnational offense. And so these searches had been permitted before that without any particular issue. Can I push back on that restriction that you just defined as to whether a reasonable officer would have known that this search was legal given or in light of binding precedent? So we have the Stevens case that says that the – it seems like the relevant question is whether a reasonably well-trained officer would have known that the search was illegal in light of all the circumstances. And the circumstances being here, the lack of any definitive ruling by our court, COLSUs leaving the question open, the competing interests, the significant number of out-of-circuit cases saying that this is okay. So what about Stevens? Is that just wrong? I will acknowledge I'm not particularly familiar with Stevens, but, I mean, I have always understood the cases that we cited all say it's a question of whether they're acting in the realm of what's authorized. Maybe it's two sides of the same coin. It is. And I think that the good faith should be limited to acting within the realm of what's authorized because otherwise there's just sort of a general permission to do whatever as long as this court hasn't explicitly ruled against it. But if you say that you're acting within the realm of what's authorized, then there's no reason to apply the good faith exception. I mean, if the officer did what was allowed by law or went beyond the law, there's no good faith, no reason to apply good faith. Well, I think if this court, you know, holds, for example, as it did in COLSUs required the ongoing transnational offense requirement, that hadn't existed before, it would exist going forward, but good faith would save the search otherwise. And similar in, you know, Mr. Riley, for example, I don't think got to take advantage of the ruling in his favor in his case because it was only announced, you know, years after his arrest. So I would submit that good faith should be limited to what has been approved and not simply picking around in gray areas and trying to do whatever we can do in a gray area and hoping that the court will approve it. All right. You've got some time left for rebuttal, Mr. Richmond. Thank you. Thank you. Mr. Sawyers. Good morning, your honors, and may it please the court. Reed Sawyers for the United States. I do want to briefly start with my friend's newfound argument that this isn't a border search at all, which for good reason he did not raise below or in his briefs. And that is so because as Chief Judge Diaz recognized in the Congo, child sexual abuse material is prototypical digital contraband, which is core to the government's authority under the border search doctrine. Recognizing that this is a border search, the government believes that ICCE is binding precedent on this score, which authorized manual border searches of electronic devices without reasonable suspicion. The defendant in that case did raise both a First and Fourth Amendment argument. You can see that most clearly in his reply brief where he says, for example, although a routine border search may be reasonable under the Fourth Amendment, searching computer files, diskettes, and data, even at the border, is not routine and not permissible under the First Amendment. I'm sorry, which case are you quoting from now? This is ICCE's, the 2005 case written by Judge Wilkinson, in which he said that the manual border search of a computer did not require reasonable suspicion. Well, I thought, didn't the appellant in that case just raise a First Amendment claim? So as I was explaining, if you read his reply brief, for instance, as well as his opening brief, he talks about both the First and Fourth Amendments. And if you look at ICCE's itself, at 503, it describes him as raising a First and Fourth Amendment argument. At 505 Note 1, it expressly addresses the Fourth Amendment and finds the search there to be reasonable for purposes of the Fourth Amendment. Then at 505 to 506, it has an extended discussion of border search doctrine. And then after ICCE's, in cases like Kolsue's, this court, district courts in this circuit, like Sabanci, other courts of appeals, interpreted ICCE's to be addressing border searches, to be addressing the Fourth Amendment question. And then my friend below likewise interpreted ICCE's that way. At 34, he describes it as finding a manual search of a computer is a routine search. He interpreted it the same way in his opening brief. Technically, that's a waiver. I agree it's a bit weird to say you waived whether a prior case is finding precedent. But at the least, the fact that he only discovered his newfound interpretation of ICCE's in his reply brief shows that that is not the fairest interpretation of ICCE's. So in your view, the ICCE's decision puts our circuit in the same camp with all the others? Yes. And of course, even if this court thought it was an open question, you'd have ICCE's as, I guess, strong dicta, Kolsue's as strong dicta, saying the distinction between forensic and manual searches is totally manageable. And then by my count, I believe, seven or eight, depending on whether you think the Eighth Circuit Zhang decision was an actual holding, or just saying it likely would adopt this position. But virtually all the court of appeals to address it, and virtually all the court of appeals have said that manual searches do not require reasonable suspicion. So counsel, what if you had a manual search? They seized the phone, and from a time period, took a month looking through it like in the Kolsue's case. But did it all manually, you know, took a month, wrote down, had a big old long report of everything they found. I guess what I'm trying to figure out is whether, you know, is there any point that something that is done manually has enough characteristics of a forensic search that they're functionally the same? Or if we're using our hand versus a machine, is that determined? So three answers to that, Your Honor. And the first is I do just want to make clear, there's no evidence that has ever happened. There's no reason to think it would ever happen for all the practical reasons the government described in this brief. Likewise, CBP policy, and I should say, we didn't file a 28-J on this, but CBP has come out with a new policy, effective January 1st, 2026, which as far as I can tell is materially indistinguishable, just adds a little more detail, which is why we didn't think it was worth a 28-J, but just wanted to bring that to the Court's attention. It's available on Google. You can just search for it online. But the policy likewise makes clear that searches have to be carried out expeditiously and in a reasonable manner. But to address the thrust of Your Honor's question, this court in Colsus said that a search aside, a seizure has to be reasonable, and a search reasonable at its initiation has to remain reasonable in scope and duration. Likewise, Chief Judge Diaz in Nkongo held that apart from a forensic search, a non-routine seizure to conduct a forensic search independently requires reasonable suspicion. And this court has general Fourth Amendment cases like Pratt saying that delay in conducting a search or seizure violates the Fourth Amendment. So I think this court would almost certainly say that a 30-day manual search transgresses the Fourth Amendment because it is not reasonable in scope and duration. Of course, this case does not require the court to delineate the bounds of that exception because the CSAM was discovered within two minutes by Officer Oliphant. But I don't think that as a matter of practicality, as a matter of policy, and as a matter of Fourth Amendment law, it's true that you can engage in these infinite searches with an infinite number of CBP officers, which I think is one of the major concerns motivating both my friend and the outlier Second Circuit district court decisions which are currently on appeal. I think the other major way in which my friend and those decisions go wrong is this idea that the government does not have a significant interest in the interdiction of digital contraband because it invariably exists on the cloud. And I think that's not true as the Zhang case, which involved the attempted exportation of trade secrets from Monsanto, demonstrates. If you were trying to smuggle classified material to Russia, if you were trying to smuggle trade secrets to China, there are obvious reasons why you would have that only on your phone. Likewise here, Belmonte Cardoso, you can see this at JA237, had the CSAM, the record suggests, only on his phone, he would capture it from Snapchat, where it's automatically deleted, on one phone, capture it onto a second phone, and then he attempted to smuggle that into the United States, and there's no reason to think it existed anywhere else but his phone. So this idea that there's no use in the interdiction of digital contraband is just factually incorrect. Can I ask you about the... Let's assume for a moment, I know you disagree with this, that there is some modicum of reasonable suspicion that is required at the border. I know that's not the government's argument, and you have a response to why the reasonable suspicion would suffice here anyway, but I was troubled by the sort of paucity of the evidence in this case that led to the search. You've got, apparently, for lack of a better term, catch a sexual abuser month. They were engaged in that kind of activity, the law enforcement officers were. You've got a generalized report with respect to Bolivia of a heightened activity with respect to CSAM evidence, and then you've got these five cash app transactions, none of which identify, link Mr. Cordoza specifically to any one particular transaction, and a honing in on the SNAP transaction would suggest that that might have been shorthand for an exchange of CSAM activity for money, but again, none of it linked directly to him, so a one-in-five chance that it was him. How does that amount to reasonable suspicion? So a couple points on that, Your Honor. First, the overall standard we're applying is not stringent. As this Court made clear in cases like McCoy... Doesn't it have to be particularized, though? And how is a one-in-five chance that he was the one involved with the SNAP transaction particularized to him? So I will address that. I just do want to say all that needs to be done with the totality of the information available to the officers is eliminate a substantial fraction of innocent travelers, not all innocent travelers, and the way we think about the report is that if you look at JA-155, 156, as well as JA-149, Officer Oliphant testified this report was not done just based on the fact that these five individuals were engaged in transactions with minors, but also the fact that those transactions were deemed suspicious based on the memo lines. And Officer Oliphant confirmed that suspicion generally because she saw at least one of the memo lines was related to SNAP, and she knew from her training and experience that SNAP was an app where CSAM is heavily elicited. But the way we think about that is she was generally confirming the validity of this report, which is not just they reported people engaged in transactions with minors, but they reported people engaged in suspicious transactions with minors based on the memo lines. And she wasn't able to check every single memo line, but she did see one memo line that was very suspicious to her, and that confirmed the entirety of the report, which is both that these people are engaged in transactions with... Does that really work when one of the memo lines... I mean, it seems... And I get the point about SNAP, but as Chief Judge Eah says, that's just one of them. And some of the other memo lines are food and stuff. Is it the government's position that that's suspicious or that's indicative of this type of behavior? So the report itself is not in the record, so I'm a bit... I don't want to, you know, talk outside the record, but there's a reason we have to view the facts in the light most favorable to the government here. And Officer Oliphant testified that the individuals were found suspicious based on the memo lines, and then she also provided one example of a suspicious memo line. So I think that is sufficient on the low bar for reasonable suspicion, which, again, just requires you eliminating a substantial portion of innocent travelers, which, frankly, just the reports of the five individuals alone would likely be sufficient to meet that standard. I, of course, agree that it would be better, there would be more suspicion, if, you know, each of those individuals had been linked to... The report had linked each of them to a bunch of specific, suspicious memo lines. But I think there's clear testimony in the record that they were reported both based on having transactions with minors and those memo lines being suspicious. So I think viewed in the light most favorable to the government... Don't you have to... I mean, it may be enough. I mean, the report indicates that each of the five, including the defendant here, was engaged in what was a suspicious transaction. So that much of it seems particularized or individualized. The question is... He was coming to a country with heightened concerns about this type of activity. Probably not, maybe not enough on its own, but those two things may be enough. I'm just trying to figure out how much you can really make of a... Even construing the record in the light most favorable to the government, we got a one in five chance that the snap applies, it seems like. And you also have these things that don't relate, it doesn't seem, from the memo lines at all to this type of activity. I mean, do you think you need the snap memo line reference? I don't necessarily think we need the snap memo line reference, but I do think that even without tying it specifically to Belmonte Cardozo, it does validate in the mind of a trained officer like Oliphant, who is exclusively focused on the interdiction of CSAM. It does set off those alarm bells in her mind, confirming that these are indeed suspicious transactions suggestive of the purchase of CSAM, not just a list of everybody who engaged in transactions with minors. But of course, our primary contention is that the government's primordial authority is at its zenith at the border, and that reasonable suspicion is not required here. I'm happy to answer any additional questions the court has. If not, we'll rest on our briefs and ask the court to refer. Can I ask you about good faith? Because we had a back and forth with your colleague as to what the standard should be. Is it that an officer needs to work within the bounds of binding precedent as it exists, or is it that there was no authority that suggested that the conduct was illegal under the circumstances, or is there a difference between those two? I think there probably is a slight difference, and I think your reading of Stevens is correct. It's clear that officers have to know that it's illegal. I think my friend is sort of trying to collapse good faith to only where it's 100% clear it's authorized by circuit precedent. But if that were the case, good faith would almost never apply. The only areas where good faith would apply is when circuit precedent is overruled on bonk or by the Supreme Court, and I don't think good faith is that narrow. Stevens, for instance, I believe, held that the attachment of a GPS device to someone's car was authorized by a more general precedent from the Supreme Court, NOTS, which was about tracking using a beeper. Likewise, in Kolsuse, this court talked about the general legal landscape. And again, we would urge this court to address the merits and to hold that reasonable suspicion is not required for manual border search, as we think the court has already done in Ickes, and all the other courts of appeals to address this question have done. But at an absolute minimum, it is still the case that no court of appeals has ever held that reasonable suspicion is required for a manual search. And if this court could read Ickes as binding precedent, in Kolsuse, if district courts could read it that way, in cases like Sabanchi, I think at the very least, a reasonable officer could read it that way. So whether you're relying on Ickes, whether you're relying on the general legal landscape, whether, as in Stevens, you're relying on more general pronouncements from the Supreme Court, like Ramsey's, and all the other Supreme Court border search cases, I think there's absolutely good faith here. But again, we would urge you to address the issue on the merits rather than just holding that there's good faith. But only address it if we agree with you, right? Like Chattery, we want to win, and you got me there. Yes, we'd rather you do good faith if you were planning to rule against us. I will not lie about that. Thank you very much. Thank you, Your Honor. Well, I agree with the government in suggesting that the court should reach the merits of the standard for manual searches at the border. Whether we agree with you or not. Obviously, I'd like the court to agree with me. I would like to go, because I didn't discuss the reasonable suspicion facts at all earlier. The government, I think, has sort of added some things into the record in saying that she reviewed this and determined it was suspicious, that Cash App had determined it was suspicious and that she corroborated that by looking at a little bit of the report. Every time she used the word suspicious in referring to this report from Cash App, she said, potentially suspicious evidence of possible CSAM. She never said that I knew that they had done any analysis to know what this really was. She just knew that these names had been floated in front of her. With respect to Mr. Belmonte, all she knew was he'd engaged in at least one transaction between 10 and 15 months earlier with an account that was owned by a miner, but you may or may not have known that for an unknown reason in an unknown amount with an unknown memo line, where some of the memo lines were things like food and stuff. I would submit that's just not particularized in any way and cannot establish individualized suspicion by any standard. Same with the evidence about Bolivia, which is just sort of generalized. Well, they have a significant problem. Well, so does the U.S., so how does that, without any quantification, that doesn't exclude a single member of the traveling public. Was the U.S., if the report was the one that was in place at the time, which was the 23 report, not the 24 report, is the U.S. in the Tier 2 watch? No, but... So, it's not the same at the time as the U.S.? No, it's not, but the reasoning, if you do go into the depths of the report, the reasoning that Bolivia was in the watch list that year was because it simply hadn't submitted its report yet, I guess, by the time of the publication for that year. It was a lack of data. And, of course, the main point of the report is a report on countries' efforts to address this problem, not a report on the quantification of the problem. And, of course, it's generalized as to human trafficking, not child sex trafficking. And there's no indication there was ever anything on the merits of its efforts or on the merits of the extent of its problem that it was ever anything other than Tier 2, which is the vast majority of the world. So, it just doesn't exclude any of the traveling public. And the officer testified to that. She testified that she has no idea if the rates are any higher in Bolivia than Virginia, where Dulles Airport is. So, I would submit that that evidence does not possibly reach the standard required by this court in Feliciana and McCoy about excluding a substantial portion of the public. And going to where the government started, the government raised ickies and claiming that there's a Fourth Amendment discussion there. The enthroning comment that the government has put so much weight on is a passing comment in a section that's clearly about the First Amendment. The court made perfectly clear in the paragraphs immediately before and after that this was ruling on a request for a First Amendment exception to the normal standard for border searches, which it just assumed was... So, it kind of looked at ickies as if it had answered that question? I don't think it did because Colsus, I think, in two places, suggested that manual searches was an open question in this circuit. So, it couldn't have thought of Colsus that way. And I would also say that Agba Kayin, good faith ruling, didn't look at ickies that way because Agba Kayin was a case where they held a warrant should have been sought. But it was upheld on good faith because the search itself was pre-Colsus and the reasoning a warrant should have been sought was based on Colsus. They didn't say, and we've approved these searches under ickies. So, again, it was just because Colsus had intervened and so they applied good faith there. They never relied on ickies as the basis for saying that this court was in breach of conduct. So, I would ask the court to reverse. I'd submit there's simply not reasonable suspicion here and the court should require reasonable suspicion. Thank you, Mr. Richmond. Thank you. I want to thank both lawyers for their fine arguments in this interesting case. We'll come down and greet you and adjourn the court sine die. This honorable court stands adjourned. Sine die. God save the United States and this honorable court.