# Oral argument — Smith v. Navient Solutions > Briefs, arguments, decisions, and more. URL: https://www.frixlaw.com/law-library/documents/argument%3A70215 ## Record - **Collection:** Oral argument - **Document type:** Oral argument - **Published:** May 21, 2020 ## Text Okay, Mr. Chamey, want to go ahead? Absolutely. Um, good morning, Your Honors. My name is David Chamey and I represent the appellant, Greg Smith. And as I told Greg when we started, I'd like to reserve three minutes for rebuttal. Okay. Thank you. Um, Your Honors, I'm going to start with, of course, at any point in time, if you have questions, let me know, but I want to start with what appears to be, um, a split in district courts and circuit courts that is essentially found based on an interpretation of a very specific portion of the statute that defines what an autodialer is. The specific portion of the statute that says an autodialer has the capacity to store or produce telephone numbers to be called using a random or sequential number generator and then to dial those numbers. Now, the district court in this case, like other courts have done, has decided that the only systems that qualify as autodialers are the systems that make up phone numbers, that generate random numbers and then dial them. And some courts have specifically used and explained this logic, even though by doing so you have created suffusage in the statute, the word store loses meaning, but they've explained it by the definition of the last antecedent rule. They've relied on the last antecedent rule and the comma to control how judges and courts now are interpreting a statute. Mr. Schimler, if I can interrupt, let me tell you that that rule has come up many, many years ago. It came up in an en banc argument. I find that one of the least appealing arguments I've ever heard, and I know that you're not making the argument, you're challenging it, but one of the least appealing arguments I've ever heard. Lawyers and judges and congressional people are more likely to suggest these things, do not write these things with Stumpkin, I guess it is, I'm sorry, I'm sorry, behind me, Stumpkin-Wagner in one hand was quoted in one place in the Chicago Manual of Style, which is not how writing is done, and we invent this fiction to gloss over the fact that it was just sloppy drafting and an ambiguity cut to the statute. That has nothing to do with the intent of placing the damn comma where it is, it was just something that happened in drafting the statute. Someone who drafted this thing, probably from the staff of Trichardt Law School, stuck in a comma there without thinking about, well, they'll put this in here, then that will be interpreted using the last antecedent rule, and therefore the meaning is clear. That's pure fiction, pure fiction. And I agree wholeheartedly, Your Honor, but for the, what I'd like to do is just say that not only is it fiction, the Supreme Court and Justice Scalia and Barnhart v. Thomas discussed the last antecedent rule, and I believe that they are somehow bound by this. That would be in contravention to what our Supreme Court says in a unanimous decision, and I'll give you the site, it's 124 S. Court 376 Act 380. Justice Scalia says this rule is not an absolute and can assuredly be overcome by other indicia of meaning. All right, Mr. Chairman, it's Judge Bevis. So you're right, your battle here is between the canon of the last antecedent and surplusage, and Judge McKee is right, we're not just looking at the abstract, we've got to look at them in context. So Judge Sutton's perspective is when we say a court can reverse or affirm using the precedence at hand, as a matter of common sense, they're not just saying the affirmance is with the precedence at hand. Ordinarily, we'd understand that the reversal or affirmance is with the precedence at hand. Or if we said, you know, an actor in acting or dancing for the pleasure of the crowd, the acting just naturally goes with the dancing. That's the common sense insight underlying the canon of the last antecedent. Now, at first I was kind of persuaded by that, by the, you know, your argument. Your argument is fundamentally that it makes no sense to put store in this statute, right? All the work is being done by produced. I want to hear what you have to say in response to this, because both of these canons are defeasible if that's just the way an English reader would read it. What made a difference to me yesterday was when I came across some comments that Noble submitted to the FCC, where they said, back in the 90s, there were two different ways that auto-dialers worked. Some of them just generated numbers and instantly called them. That would just be producing. But others of them, they generated a number, they called the first number. Then they generated another number in the same phone exchange, stored it, and they checked whether the second number was the same as the first by storing it and using a flag. And if it wasn't the same as the first, they called it. And if not, they knocked it out. So some auto-dialers used storing and producing. Others of them just used producing. When I learned about that technology, it didn't seem so ridiculous to mention storing and producing if there were two different ways these auto-dialers worked at the time. I don't know if you understand this or if you've looked at that technology or that discussion, but it came up in Judge Barrett's opinion for the Seventh Circuit. What's your response to that, that there were two kinds of auto-dialers and some of them used storing and some of them didn't? It's still surplusage, but it's not ridiculous surplusage to cover the waterfront to mention both things that were involved in some of the auto-dialers. Well, I think, Your Honor, the word or causes a problem for that interpretation because the statute says store or produce, suggesting two different functions. Well, yes, but it is also true that we have precedent that says, you know, sometimes when you say or, you mean and or, and sometimes you mean it exclusively. So some of them are storing and producing. Some of them are just or producing. So I agree, you didn't have to put store in there, but it's not nonsensical to put store in there. Well, that would require us to ignore other portions of the statute, Your Honor. And here's what I mean by that. Congress requires express consent to the person you're calling. Why would you, you're not going to have express consent to call numbers you're making up. Well, wait a second. Judge Bevis, again, in B1A, you know, the express consent applies if you're using an auto-dialer or an artificial or pre-recorded voice. So a bunch of these typos about the government debt or the consent, et cetera, make perfect sense if they're applying to the artificial or pre-recorded voice that's part of that same B1A. So I don't think it's nonsense. And let me turn this around. What does seem to make some sense to me is why the ban on calling 911 numbers or, you know, if it's about, you know, purely pre-programmed stored numbers, the 911 and the hospital numbers wouldn't be in there. It's only if the auto-dialer is randomly or sequentially calling, you know, hospital numbers. It's not doing it intentionally. It's doing it accidentally. So I don't think B1A favors you. And if anything, that cuts the other way. Because the hospital exemption they were dealing with in the 90s would only have been for these sequential or random number generators, not intentionally. Five minutes. Your Honor, the point here is that in the 90s, there were systems out there that were making up numbers. There were also systems that were calling stored lists. So in the 90s, Congress would have had more of a concern for these random sequential number generators that did exist. They no longer exist. That did exist and were calling a bunch of numbers made up all out of whole cloth, thin air, and would reach hospitals or 911. That issue, though, is no longer an issue. And even in the 90s, you still had to deal with the other sort of dialing systems that were calling from stored lists. So and the prerecorded voice calls, you can't say that, for example, storing or producing or the consent, which is a completely different section, only applies to a portion of B1A. It applies to the whole. And so, you know, because that would require you to literally, I guess I'd say, ignore the last antecedent rule in a way. No, it applies to the whole. It's not needed as to the automated telephone dialing system, but it is needed as to the other port. But the point is that it's not ridiculous to say consent is required there because consent still does some work in that parenthetical. Well, while I agree it does some work, I think you have to ask, does it make sense that – you'd have to say that consent only applies to a portion of the next section because you wouldn't need it. You wouldn't need it if it only applied to the portion that talked about making – dialing numbers that were totally made up. So – It's in that parenthetical. It's a separate parenthetical. The separate parenthetical is emergency calls and unconsented to calls. They stuck that in to limit the rest of the phrase and they didn't – they wouldn't have needed to stick it in if it were just an automated telephone system, but it is needed as to the artificial pre-recorded voice. And so, you know, again, when they stick something in in a parenthetical, they didn't need to break it out to be, you know, super exacting into a different subsection. They just – there's a ban on auto dialers of pre-recorded voices and then there's a carve out for emergency numbers or consent. And the carve out wouldn't have been needed if it were just a ban on auto dialers, but it's not just a ban on auto dialers. It covers more than that. And so consent is doing work and the ban on emergency calls is doing some work. But – and I don't disagree with that, Your Honor. I actually think that that's accurate, but I don't think that actually cuts against the opponent's argument here. Two minutes. As to whether or not consent is logically needed to call. Obviously, because it's my position that you're going to be – if people are calling from stored lists, then you're going to only want to be able – you're only going to be able to do that to people who've given you their consent. And while – Mr. Kamey, is there any canon that helps you beyond supportage? Is your argument really resting on the idea that this word store is not needed in the work, the consent is not doing any work, et cetera? Is that the core of your argument? Well, I have another argument, but the core of the argument here is that the authority – the Supreme Court authority, when you're reading statutes, is you have to give – you have to look to see is there evidence to give – to let you understand what congressional intent was. And so not only are we talking about what Congress had put in the statute, but we have to look at what the FCC has done since 2003 on the issues of predictive dialers. I want you to address that, and also address ACA, the D.C. Circuit's opinion, and our opinion in Dominguez 2, and to what extent did the FCC guidelines before 2015 survive as we – our opinion in Dominguez 2 stated? What happened to Dominguez 2 after ACA? Well, so as far – the way – I read ACA, and everyone who says that ACA International, by striking the expanded definition of capacity, has somehow, by implication – By the Supreme Court. Yeah, but it's by implication, striking that potential capacity definition has somehow struck the prior orders. I ask them, respectfully, how can the 2015 order survive, at least portions of it survive, while by implication the entirety of the 03 and 08 orders are gone, when what the ACA court did, the D.C. Circuit did in that case, is just to say, look, you've gone too far by saying any potential future capacity will create an autodialer. If it could be – and of course the cell phone AT – cell phone concerns of – because basically an ATDS is really just a computer system with software. And so it's not just – It's kind of a cell phone, isn't it? It's a computer system with software. That's right. And so that's why it's important to look at, well, did ACA actually, did the D.C. Circuit actually do anything to say anything affirmatively saying, look, we're striking this concept of predictive dialers being autodialers? And they didn't. In fact, what the D.C. Circuit said related to what appeared to be inconsistencies in the FCC's prior orders is that, look, you just got to make it clear. When they said you can't insist on the words random or sequential number generator and simultaneously say that calling from stored lists, predictive dialers, which aren't random or sequential number generators, are also autodialers. So you've got to come up with a clearer definition of what an autodialer is and how you use the word capacity. But it is logical. I'm sorry. No, that argument has some force. I'll bring it to that. I'm concerned about something else, though, that has gotten some play in the FCC and in these cases, which is the circuits that have gone with the grammar-centric approach in favor of Navient are worried that every, you know, if all the statute requires is storing and dialing numbers, then every iPhone, every phone with the capacity to redial or call forward, every iPhone that sends an auto response, sorry, I'm driving, call you back, could be swept in. And that's a good reason not to sweep all of those things in and to require more than just storing numbers, but to require numbers that have some kind of robo aspect to them, like they've been generated using a random or sequential number generator. So what do you say about that? We should be troubled if this is going to sweep in all the smartphones. I would say to that point specifically, that's not what's happening here. And if a smartphone, if somebody decided to download an app onto their smartphone and blast out thousands and thousands of text messages, then they would be using an autodialer. Okay, fair enough. So how about these common functions of auto responses on iPhones? How is that not covered by storing iPhone store numbers and the dial numbers? Yep. And so that's very similar to the Dominguez case as far as what was happening in that case. Here in Dominguez, people were uploading their phone numbers to this machine so they could get notifications, email notifications. And the system in that case, which wasn't storing numbers provided to it by Yahoo to call, it was only responding to affirmative. It was reacting to a request by what it believed to be the end user who had previously owned that number. That's the same as when somebody reaches out to you, they've consented to a call back, right? So an auto response while I'm driving would not be me blasting out an automated message. It would be me responding to a phone call. Yeah, that simply would not be the same. And here, what Navient was doing, and they've acknowledged in their own facts, they were calling from a stored list, Mr. Smith, and they were calling him without even agents on the call. These are agentless calls. All right, fair enough. But I still want to hear a response to Judge McKee's concern, which is what is the status of Dominguez? Dominguez postdates all of these FCC rules. So first of all, does Dominguez bind us in saying the FCC rules are off the table? And second, Dominguez took as a premise that the focus here was on using a random or stored number generator. So what do we do with Dominguez? Well, first of all, the only thing Dominguez said regarding the auto dialer issue explicitly, and I've read it, is it said that an ATDS must have the present capacity to function as an ATDS, as an auto dialer. I've gone through it multiple times looking for language that says anything about what happens if the calls are being made from a stored list automatically. There's no discussion of that. The system at issue in Dominguez is completely different than the system at issue here. The process, what was happening, the function of that machine was completely different. The reason Dominguez was looking at whether that system could randomly or sequentially generate numbers was because everyone admitted it wasn't calling anybody or texting anybody off of a stored list automatically. So the court in Dominguez didn't need to consider whether what was happening in this case, whether or not that would be an auto dialer. And it did not do so. In fact, I think the Ninth Circuit got it wrong when it implied what Dominguez stood for. I don't know what he's been doing. I don't know what he's been reciting to what we said in Dominguez. I think it started with Judge Sutton, and what they're saying he said, I don't think we ever said. And I agree. I didn't see it in the case. I've looked through it. I've looked through the order. I've looked through the opinion. And it appears that courts have taken the language from Dominguez, you know, conversation and discussion, and sort of created their own, well, here's what we think they meant, and have started to say, well, this is what Dominguez stood for. But it's not, from reading the opinion, that's just not there. But I think you reserve some time for the panel. Yep, I'm going to stop. According to my random sequential dialer, you have 18 minutes and 51 seconds now. We'll hear from you again. Thank you, Your Honor. Mr. Ehrlich? Thank you so much. I'd like to start, I guess, by diving into Dominguez, and then I'd like to come around to the statutory interpretation points, if I could. And I would say that when you understand the background of how Dominguez came to be, I believe the case is directly on point. I think it does, it is precedent that this court should follow. Yahoo! was making the argument that the system could not generate random or sequential numbers. Now, it is true that the system is different than the system here. But as it relates to the language of the statute that we're talking about, the only thing that's relevant to the argument is whether, at least as similarities between the cases, is whether the system could generate random or sequential numbers and then dial them. That is the defense that Yahoo! made. It obviously went up and back to this court a couple times, given the issues of ACA. And ultimately, when it came back to this court the second time, the issue was, in light of ACA, does the system have the actual current capacity to do a function? And that function was material to the court's decision. Because in order to decide the latent versus current capacity issue, the question is capacity to do what? And that is a question that this court answered. The court said, absent from the experts that the plaintiffs had was any discussion of how the system actually could or did generate random telephone numbers to dial. And the court's quote was, because it does not shed a light on the key factual question actually at issue in this case, whether the system functioned as an autodialer by randomly or sequentially generating telephone numbers and dialing them, the expert report lacks fit or relevance and was therefore properly excluded. And the conclusion was that Dominguez couldn't find any evidence to create a disputed fact that whether the system had the present capacity to function as an autodialer by generating random or sequential numbers and dialing those numbers. But we did confirm that the 2015 SEC guidance was gone. And my concern is, absent the 2015 guidance one, what happened to the guidance before 2015? And if that guidance is still in place, what does that do to Navient's case here? Assume that it is still in place. Well, I mean, obviously there's a disagreement. I don't believe it is in place. And so even if it were, I think at the end of the day, I think it's Justice Scalia had a quote along these lines. A regulator can be the sorcerer's apprentice, but never the sorcerer. So at the end of the day, the statute... Well, at the court's deadline, I would agree with that. After some time, I'm not so sure. But go ahead. So my point would be that fundamentally the statute is the controlling place to look, which I think is what this court did in Dominguez when it effectively said, OK, let's just take a look at the statute. I do agree with... I would point out that both Mark's, the Seventh Circuit case, and I think it's Dan Halick, and then the Glasser case in the Eleventh Circuit, all agreed that the ACA decision invalidated not just the 2015 guidance. Mr. Hurwitz, let me ask you about that, because I've read through this a couple of times. It's not free from doubt, but the core of what Shreema Batson said the ACA was, as Mr. Chairman said, the SEC is talking out of two sides of its mouth. And on the one hand, it says we're reaffirming these earlier orders. And on the other hand, it makes a reference to stored list numbers. Now, it's true in this discussion it also cites the 2003 order. But the specific infirmity that troubles it is the ambiguity that they're taking both positions at once, and that's only in the 2015 order. So the logic of ACA really seems to be that the infirmity is in the 2015 order, not the earlier orders. I'd like to know if you have a basis for disagreeing with that, and then you can go on. You're right that we still have to do our work at Chevron step one here, but I'm not sure that's the best way to read ACA. Thank you. Yeah, the court actually did address the idea of whether it could revisit and look at prior guidance, because one of the things that the FCC was arguing is that the D.C. Circuit lacked the power to review and evaluate what the FCC had ruled in 2003 and 2008 because there was never a timely appeal from those orders. And the quote was, the agency reasons that the issue was resolved in prior agency orders, specifically declaratory rulings in 2003 and 2008, concluding that the statutory definition of an ATDS includes predictive dialers. According to the commission, because there was no timely appeal from those orders, it's too late to raise a challenge by seeking review of a more recent declaratory ruling. We disagree. The reason it disagreed is because the commission's 2015 attempted to incorporate in everything it had said before. You're right. You're right. The earlier section of the opinion says this is all on the table, but when they do the work of striking something down, they strike down for inconsistency. And so I think it's only that part of the 2015 ruling that's gone. But I see your side. It's not – it's a colorable, respectable argument. Maybe we can move on to Dominguez. The thing that bothered me about Dominguez is that the – this was never briefed or argued in Dominguez. Dominguez is all about the sufficiency of the evidence, these experts on the table. It was about current capacity versus latent capacity. And that's why Dominguez is only focusing on the 2015 rule, because only the 2015 rule gets to whether it has to have the latent capacity or the current capacity. And they're not really focusing on current capacity to do what. You're right. The current capacity was presumed to be the current capacity to, you know, dial random or sequential numbers. But it wasn't really contested in Dominguez. So what do we do if they're – they assumed this, but they really assumed it for purposes of argument. They didn't – it wasn't contested there. They didn't squarely hold it. It was contested below, of course. And that was the thing that led to all this litigation. And by the time it got up to the court, you know, I can appreciate the fact that the folks were concerned about this current versus potential capacity. But the court can't resolve that issue, in my view, until it understands what the capacity is, the capacity to do what. So I don't think there's a way to separate the finding that it's the capacity to generate random or sequential numbers from the court's resolution of whether it's potential or current. Because you can't answer the question until you've already accepted as a baseline what it is that that capacity means, the capacity to do what. And I think that's why both Marx, as well as the 7th and the 11th Circuits, have read Dominguez, as have district courts within the 3rd Circuit, to say that you can't escape the possibility of the random or sequential number generation part of the decision. All right. Could we move on to the statute itself? I mean, this has been helpful. But I do want to know. So I think the problem with your adversary's position is this weirdness about storing. It has to take a direct object that has to jump over and take the numbers to be dialed without taking the phrase that modifies it. The problem for you is that the word store could have been omitted from the statute, and it would mean exactly the same thing you're arguing now, right? That surplusage should give us pause. Why did they put store in there if the produce does all the work? I think it's because the word store actually does work under the way it was understood at the time the statute was written. And there's a good discussion of this, again, both in the 7th Circuit and the 11th Circuit cases, where at the time that this was created, there was equipment that could generate random or sequential numbers and then hold them for periods of time and then be able to cross-check those to other numbers, for example, and not even dial immediately. And then there's other equipment that generates the number and instantaneously produces that number into the dialer and then dials it. So I don't agree that it's true that the word does no work. I think under the common understanding of that equipment at the time it was created, it actually was a real thing. Now, I know that the appellant has argued that, well, that equipment doesn't exist anymore. I guess my report to that is, if random or sequential number dialers, generators don't exist anymore, the statute did what it was intended to do, which is effectively get rid of them. And I think the legislative history of the statute does explain why, for example, there's a 9-11 exception and a hospital room exception, because there were genuine concerns that random or sequential number generation from telemarketers was tying up hospital lines, emergency lines. If whole neighborhoods were getting hit, then it wasn't possible to make outgoing calls. It wasn't possible to call the police. So I think that's all consistent with the understanding of the work that the random or sequential number generation language is doing. And then there's other principles of statutory interpretation that come in. We've talked about the last antecedent rule. The point's well taken that it's not inviolable, but it is something, it is a factor, and it has been recognized by the Third Circuit and the Supreme Court as a fact. Five minutes. The other piece of work, the other piece of work, yeah, go ahead. What do we do with that? As I see it, this is a battle between, on the one hand, surplusage. Store is pointless. That's their best argument, if not their only. And your best, if not only, argument is, look, this phrase at the end, it's got to modify numbers to be dialed, and it's bizarre for store to take that as an object without taking the phrase that modifies it. So there's a battle between these two canons. Number one, which wins in the abstract? And number two, is the battle between the two so clear that we resolved as a chamber on step one, or could we get to step two if our understanding is that these earlier SEC number ruins are still alive and breathing, as we see it? Well, there are surplusage arguments that both sides can make, right? The appellant argues that the word store is surplusage. I've spoken to that, right? I don't think it was under the understanding of the statute as it was created. But their reasoning, I would argue, eliminates significantly more of the statute, because the word produce does no work, neither does the phrase using a random or sequential number generator. All of the statutes that you can say- You can't store a number using a random or sequential number generator. What you're storing has to be generated. So produce is still working. You can't store something, obviously, until it exists. And that is the act of production, how production using a random or sequential number generator. So it's still there. Well, I guess it depends what you mean by the word produce, which I actually looked up for purposes of this argument. And it's one of these insidious words in the English language that has numerous definitions. I mean, if I make a cake in my kitchen, I have produced the cake, but I've also produced it to you when I show it to you. And so it depends how you want to understand what that word means. I think one way to understand the distinction is to say that in this context, produce means to immediately take and present, to take and present something. And it has to mean something different than generate, because otherwise those words would mean the same thing. So I think the meaning of that word is slightly different. And if all that is required, for example, is to store numbers in a list and then have them sent to a dialer, one could argue there is really nothing being produced. The list was generated. The numbers were generated. And then they're going to be— Where do they come from? Where do the numbers come from under your reading of it? They're generated by a random or sequential number generator. But you're saying that's not production? It has to mean something different than production. So, for example, in my hypothetical, if I bake a cake, I've generated a cake. If I give it to you, I've produced it to you, in the same sense of producing documents in a litigation. I haven't created those documents. They existed. I've just produced them to you. So I see that as slightly different than generate. But I would say, at just a more basic level, is if you had to compare whose interpretation of the statute does more violence to it, I think it's the appellant's. Because it would mean that the phrase using a random or sequential number generator is meaningless. And all that would be required is that a list could be created and dialed. And it would give no content to, effectively, half of that statutory provision. And I think that if you're weighing surplusage on the one hand versus striking out an entire part of a statute, that you would have to err on the side of accepting some surplusage at the end of avoiding eliminating half of the statute. And Sutton made this point. There's going to be surplusage here no matter how much you might assume. It's just a horribly drafted statute. Yes, Your Honor. I would agree that if one were starting from scratch, this isn't the way you would want to write the statute. And obviously, the courts all across the country have struggled with it. But I think that the way that the district court resolved it is consistent with an understanding of the statute at the time it was created. It gives effect to most of the statute. And it gives effect to the key provision that we're talking about here. I think it is consistent with this court's conclusion in Dominguez. Just like in Dominguez, we had a situation here where the plaintiff's expert couldn't give an opinion as to whether the equipment could generate random or sequential numbers. That is the exact way that the case came before this court in Dominguez. And so for all those reasons, I do believe that the district court reached the correct result. Anything else? No, I have no questions. Okay. Your staff is too long in this moment. I'm not suggesting that that means you necessarily have to take them. But if there's anything else you want to say, Mr. Horowitz, go ahead. No, I think I wanted to just address briefly this idea of the consent exception and just make a couple comments about that. It isn't true that the consent provision has no meaning if the statute requires random or sequential number generations for a few reasons. First of all, assume you're getting services, A, from a provider, let's say your cell phone company. And they have your number and you've consented to have them contact you in conjunction with service A, and then they want to sell you service B. But they also want to sell service B to all your neighbors. And so they start sending out messages using a random or sequential number generator to both you and your neighbors. Well, you don't have a claim because you've consented, but your neighbor does. And so I don't believe that the consent defense is really that the appellant raises is all that applicable for that reason. The other point is the one from Glasser, which is obviously it still does work because of the artificial or prerecorded voice limitation. And then the third point would be, and again, this was raised in Glasser, the statute applies to equipment that has the capacity to generate random or sequential numbers. Not specifically whether it was used for that particular call. So if I have a, let's call it a predictive dialer that can generate random or sequential numbers, I just didn't use that functionality, but the equipment absolutely has that capacity. Even our district court below recognized that that would fit within the statute. And the court was very careful to rule that predictive dialer, he's not saying predictive dialers can never be an ATDS. If the question is, can they randomly or sequentially dial? Mr. Hurwitz, Judge Bevis, one follow-up question to the Chevron discussion from earlier. So you've made a surplusage argument that the surplusage might be worse on the other side, might not. Your adversary would disagree. But we've got these dueling canons. We've got dueling applications of canons. How do we decide whether there's enough at Chevron step one, using the traditional tools of federal construction, that it doesn't even go to the agency, when we've got the canons on either side? Is the fact that we've got dueling positions here, I mean, of course, the sides always disagree. That's enough to create ambiguity. But how do we tell when the battle of canons is close enough to kick this to Chevron and potentially to the agency if the SEC orders are still on the table? Right. So on step one, it's my position that the surplusage argument that they're relying on is really not so grave that it does any great violence to the statute, because the statute does say random or sequential number generation. And those words must be given effect. So I guess my initial reaction is I'm not moved by the surplusage argument. It's all that grave, even at step one. If you get beyond that, of course, I have concerns that there is no applicable guidance right now from the FCC as a result of ACA. And I realize that the other side has a different interpretation. And I realize that courts have also wrestled with that concept. But if you get to that point, I don't know that there's any valid agency guidance to get us through at that point, which would kind of just take you back to looking at the statute. Okay. Thank you. Thank you. Mr. Shim, you have some time. Thank you, Your Honor. Just I'll keep it short. I think it's incorrect to state that Congress intended to do away with autodialers when it drafted the statute. I mean, that's clearly not what Congress intended to do. Congress intended to allow autodialers to exist and wanted autodialers to exist. They just wanted people to use them respectfully, respectfully, with consent. So I don't think the fact that, as Mr. Hurwitz states, that surpluses now exist because the word produce means nothing. No, produce meant something. Systems did, in fact, work that way. There's no disputing that. Everybody agreed. And the fact that those random sequential number generators are no longer in use, that I think is a good thing. But the effect of the phone calls, the robocalls, has not changed. So if Congress intended to get rid of robocalls, then the statute did not do its job because, as we all know, you can watch John Oliver, you can look at the lawsuit filings on the issues, or you can look at the number of complaints to the FCC about robocalls. It's a big one. You can go home for a few hours and wait for the phone to ring. That's right. You can just wait for your phone to ring, probably ringing as we speak. So I take exception with that. And the only other thing is to say, as Judge Bevis' question regarding the Chevron analysis, I obviously believe that there is clearly ambiguity in the statute. That has been borne out simply. Even in the courts that have agreed with our position on what an auto dialer is, have acknowledged the ambiguity. So I'm not going to argue there's no ambiguity. Of course, there is poor crafting of the statute. However, there is no ambiguity with respect to what Congress intended and what Congress has said and what the agencies have said over 15 years. And I understand Judge Bevis' concern about the government exemption. But getting Congress to act to amend the statute is a big deal. It's hard to do, especially now. But in 2016, they actually passed an amendment to this statute. This is during all of the litigation. The FCC order was still good. And Congress didn't go in and say, hey, the FCC's got this wrong. The ACA appeal was up. They didn't say, oh, the FCC's been getting this wrong all along. In fact, they doubled down. They said, no, that is what we meant. In fact, we're going to give the government an exemption. So not that the government exemption matters. In fact, two circuit courts have said it's unconstitutional. I happen to agree with that. But it's irrelevant. What it stands for today is what did Congress think an auto dialer was when they drafted it? And what do they think it is today? And their exemption for the government-backed debt collection shows that they intended other debt collections, who are obviously calling from a stored list, to be covered by the TCPA and what an auto dialer is. I think the ambiguity is resolved not only by agency action, but also by clearly what Judge Scalia said is other indicia of meaning. And we have an avalanche of that here. Thank you very much, Mr. Samuel. Taking that around the advisement, very interesting case. Thank both of you for your arguments. Thank you, Your Honor. Thank you. Could we get a transcript? Yes, could we get a transcript, gentlemen? And speak to Mr. King and he'll instruct you on how to go about doing that. Very good. Thank you so much. Thank you. --- Source: Frix Law Library, https://www.frixlaw.com/law-library/documents/argument%3A70215. Public record. Not legal advice.