UNITED STATES OF AMERICA v. IGNACIO SANCHEZ MENDOZA III (2026)

Ignacio Sanchez Mendoza, III, pled guilty to receipt of child pornography in violation of federal law. Due to a prior state conviction for possession of child pornography, Mendoza faced an enhanced statutory imprisonment range for his federal offense. The relevant sentencing enhancement, 18 U.S.C. § 2252A, applies when the previous conviction is one “relating to” possession of child pornography under federal law.

Mendoza argued that the enhancement was inapplicable because the elements of his prior state conviction were broader than those of federal possession of child pornography. The district court disagreed. The court then sentenced Mendoza to the heightened statutory minimum and ordered him to pay restitution to fourteen victims. We affirm as to the application of the sentencing enhancement, and we affirm the restitution order as to eleven of the fourteen victims. But we vacate the restitution order as to three of the victims and remand for further proceedings.

I.

The Department of Homeland Security Investigations discovered that a particular IP address had accessed a site on the dark web that was known to offer subscribers access to child sexual abuse and exploitation material, with a specific focus on “hurtcore” and other pain-and death-related contents. A search of the subscriber's IP address revealed his name and location: Ignacio Mendoza, in San Antonio, Texas. Law enforcement officers executed a federal search warrant at Mendoza's residence and seized several of his electronic devices. A forensic review of those devices revealed Mendoza's digital stash, which contained thousands of “image files and video files containing penetration, bondage, bestiality, infants and toddlers.”

Mendoza pled guilty to receipt of child pornography in violation of 18 U.S.C. § 2252A(a)(2). That offense carries a statutory imprisonment range of five to twenty years. See 18 U.S.C. § 2252A(b)(1). But if the offender “has a prior conviction under ․ the laws of any State relating to ․ possession ․ of child pornography,” then the imprisonment range becomes fifteen to forty years. Id. (emphasis added). As Mendoza conceded at sentencing, he has a prior state conviction from 2004 for possession of child pornography in violation of Texas Penal Code § 43.26.

However, Mendoza challenged the applicability of § 2252A(b)(1)'s sentencing enhancement. Before his sentencing hearing, he filed a written motion for a ruling on the enhancement, which the district judge denied. At sentencing, Mendoza renewed his objection to the enhancement and argued that his prior state conviction was not one “relating” to his “new federal conviction.” The district court overruled Mendoza's objection, finding that “the enhancement [did] apply because Mr. Mendoza [has] done this before.”

The district court then addressed restitution. The Presentence Report (PSR) catalogued restitution requests from thirteen victims, and by the time that the sentencing hearing began, there was a request from a fourteenth. The Government argued that Mendoza should be ordered to pay restitution to all fourteen victims, for a total of $94,000. As the Government explained, the Government had “submit[ted] all the restitution packets to defense counsel,” and “most of those had victim impact statements.” The court noted that it believed it had “seen all that,” and while admonishing Mendoza, the court had read excerpts from several of the victim-impact statements. Mendoza's counsel confirmed that the statements were available to Mendoza but added that Mendoza had not read them because his counsel had determined that “it'd be traumatic and painful for [Mendoza]” to do so. As for the restitution amounts, Mendoza's counsel merely replied that “the record speaks for itself.”

At the end of the sentencing hearing, the district court sentenced Mendoza to the enhanced statutory minimum of fifteen years in prison and ordered that he pay the full $94,000 in restitution. Mendoza's counsel reiterated the objection to the sentencing enhancement, and the district court confirmed that Mendoza had the right to appeal the court's application of the enhancement.

Mendoza now appeals, on two grounds. First, he challenges the application of § 2252A(b)(1)'s sentencing enhancement. Mendoza argues that his Texas conviction is not one “relating to” federal possession of child pornography because the Texas statute's definition of child pornography is broader than the federal definition. Second, Mendoza challenges the district court's restitution order. He argues that court erred by ordering restitution under a statute that did not authorize it and by failing to conduct the proximate-cause analysis required by the correct statute as well as Supreme Court precedent.

II.

We review de novo the statutory-interpretation question of whether a prior conviction is one “relating to” the possession of child pornography for purposes of 18 U.S.C. § 2252A(b)(1)'s sentencing enhancement. See United States v. Northington, 77 F.4th 331, 334 (5th Cir. 2023).

Mendoza did not object to the district court's restitution orders, so we review the restitution issues for plain error. United States v. Rosbottom, 763 F.3d 408, 419 (5th Cir. 2014). To prevail under that standard, Mendoza must show “(1) an error, (2) that is clear or obvious, and (3) that affected his substantial rights.” United States v. Curry, 125 F.4th 733, 736 (5th Cir. 2025) (citations omitted). “Even if he makes such a showing, this court has discretion to correct the error only if it seriously affects the fairness, integrity, or public reputation of judicial proceedings.” Id. (cleaned up).

III.

We first address (A) whether Mendoza's prior Texas conviction for possession of child pornography is one “relating to” the “possession ․ of child pornography” for purposes of 18 U.S.C. § 2252A(b)(1)'s sentencing enhancement. Then, we examine (B) whether the district court plainly erred in how it ordered restitution.

A.

The relevant sentencing enhancement increases an offender's statutory imprisonment range if he “has a prior conviction under ․ the laws of any State relating to ․ possession ․ of child pornography.” 18 U.S.C. § 2252A(b)(1) (emphasis added).¹ To determine whether a prior conviction qualifies for the sentencing enhancement, our court applies the “categorical approach.” United States v. Wikkerink, 841 F.3d 327, 331 (5th Cir. 2016) (citing United States v. Taylor, 495 U.S. 575, 602 (1990)). That requires us to compare the statutory elements of the previous conviction with those of the generic federal offense. Id. at 331–32. The prior conviction applies to enhance the current offense's sentence only if the prior conviction's statutory elements “are the same as, or narrower than, those of the generic federal offense.” Id. at 332 (quoting Descamps v. United States, 570 U.S. 254, 257 (2013)).

Mendoza argues that Texas's definition of child pornography is broader than the federal one, and therefore § 2252A(b)(1)'s enhancement is inapplicable under the categorical approach. The statute at issue criminalizes the possession of any “visual depiction of a child engaging in sexual conduct.” TEX. PENAL CODE § 43.26. The penal code defines “sexual conduct” as “actual or simulated ․ lewd exhibition of the genitals, the anus, or any portion of the female breast below the top of the areola.” TEX. PENAL CODE § 43.25(a)(2). The relevant federal statute defines “child pornography” as “any visual depiction ․ of sexually explicit conduct.” 18 U.S.C. § 2256(8). It defines “sexually explicit conduct” as “actual or simulated ․ lascivious exhibition of the anus, genitals, or pubic area of any person.” Id. § 2256(2)(A).

Mendoza observes that the Texas definition goes beyond the federal definition by including “any portion of the female breast below the top of the areola.” He reasons that Texas's definition is thus categorically broader than the federal one, meaning that his prior Texas conviction is not one “relating to” the generic federal offense of possession of child pornography for purposes of § 2252A(b)(1)'s sentencing enhancement.

We disagree. Mendoza's argument rests on an insupportably narrow reading of the phrase “relating to.” Supreme Court precedent instructs that the ordinary meaning of “relating to” is usually a broad one. Morales v. Trans World Airlines, Inc., 504 U.S. 374, 383 (1992). As the Court has explained, that phrase means “to stand in some relation; to have bearing or concern; to pertain; refer; to bring into association with or connection with[.]” Id. (quoting BLACK'S LAW DICTIONARY 1158 (5th ed. 1979)); see also Coventry Health Care of Mo., Inc. v. Nevils, 581 U.S. 87, 96 (2017) (“Congress characteristically employs the phrase to reach any subject that has ‘a connection with, or reference to,’ the topics the statute enumerates.”) (citing Morales, 504 U.S. at 384). The Court has also applied that broad understanding of “relating to” in the context of the categorical approach. See Pugin v. Garland, 599 U.S. 600, 607 (2023).

Section 2252A does not define the phrase “relating to.” So for Mendoza's argument, the question therefore becomes whether § 2252A's use of “relating to” bears its ordinary meaning or takes on a different one specific to this statute. See BP Am. Prod. Co. v. Burton, 549 U.S. 84, 91 (2006) (“Unless otherwise defined, statutory terms are generally interpreted in accordance with their ordinary meaning.”); ANTONIN SCALIA & BRYAN A. GARNER, READING THE LAW: THE INTERPRETATION OF LEGAL TEXTS 69 (2012) (“Words are to be understood in their ordinary, everyday meanings—unless the context indicates that they bear a technical sense.”).

While we have yet to address this question in a published opinion, this court rejected Mendoza's exact argument in a recent unpublished opinion, which we find persuasive if not precedential.² See United States v. Taylor, No. 24-20303, 2025 WL 2206975 (5th Cir. Aug. 4, 2025). In Taylor, the court determined that while the Texas definition's inclusion of “any portion of the female breast below the top of the areola” makes it broader than the federal definition, § 2252A's use of “relating to” requires “only a connection to the specified conduct, not an element-by-element match.” Id. at *3 (citing United States v. Hubbard, 480 F.3d 341, 349 (5th Cir. 2007)). This court went on to hold that the defendant's conviction under Texas Penal Code § 43.26—the same statute under which Mendoza was previously convicted—clearly satisfied § 2252A's “relating to” requirement. Id. Taylor's reasoning applies with equal force in this case.

Most other circuits that have engaged with this issue are of the same accord.³ As those courts have explained, other statutory clues confirm that § 2252A employs the ordinary meaning of “relating to.” See, e.g., United States v. Liestman, 97 F.4th 1054, 1060 (7th Cir. 2024) (en banc).⁴ Different sentencing enhancements use narrower language than § 2252A. For example, 18 U.S.C. § 924(e)(2)(A)(ii) uses the connective term “involves,” which the Supreme Court has interpreted narrowly as meaning “necessarily entails.” Id. (citing Shular v. United States, 589 U.S. 154, 158 (2020)). Additionally, Congress amended nearby provision § 2241(c) expressly to require the narrower connection now pressed by Mendoza. Id. at 1061. This indicates that Congress recognizes how to require a strict match between a prior conviction and federal law for purposes of a sentencing enhancement. United States v. Portanova, 961 F.3d 252, 257 (3d Cir. 2020). Yet in § 2252A, Congress chose to use “the broadest of connecting language.” Liestman, 97 F.4th at 1061.

In short, nothing in § 2252A's text or context suggests that Congress “intended to reference only state child pornography offenses that match the federal child pornography offense.” United States v. Bennett, 823 F.3d 1316, 1323 (10th Cir. 2016). Rather, the phrase “relating to” as used in § 2252A “takes on its usual broad meaning,” such that “a state definition need not be a perfect match with the federal definition of child pornography.” United States v. Trahan, 111 F.4th 185, 192 (1st Cir. 2024). The prior conviction must only “stand in some relation to” or have a “connection with” the federal offense. United States v. Mayokok, 854 F.3d 987, 993 (8th Cir. 2017) (quotation omitted).

Resisting the straightforward interpretation espoused by a majority of circuits, Mendoza argues that we should instead construe § 2252A's use of “relating to” more narrowly. He chiefly relies on Mellouli v. Lynch, 575 U.S. 798 (2015), for support. In Mellouli, a statute subjected an alien to deportation if the alien was “convicted of a violation of a state law relating to a controlled substance (as defined in [21 U.S.C.] § 802).” Id. at 802 (quoting 8 U.S.C. § 1227(a)(2)(B)(i)) (original brackets omitted). The Supreme Court construed “relating to” narrowly there, despite acknowledging that the term is usually read broadly. Id. at 811–12. According to the Court, this was because the text clearly pegged the meaning of “controlled substance” to those encompassed by § 802 and because the historical background of the statute demonstrated that Congress had long required a direct link between an alien's crime and a particular controlled substance. Id. at 812–13.

Mellouli does not meaningfully support Mendoza's position. As the Seventh Circuit explained, Mellouli turned on “statutory and historical features” of the statute at issue, and those features have “limited relevance” to the proper interpretation of § 2252A. Liestman, 97 F.4th at 1065. Instead, § 2252A must be interpreted “in the light of its own statutory context,” and as explained above, that context “supports a broad understanding of ‘relating to.’ ” Id.; see also Trahan, 111 F.4th at 194–97 (distinguishing Mellouli).

At bottom, Mendoza contends that his Texas conviction for “possession” of “child pornography” is not one categorically “relating to ․ possession ․ of child pornography” under 18 U.S.C. § 2252A(b)(1). But his argument rests on an unduly narrow interpretation that would effectively read “relating to” out of the sentencing enhancement. “We must assume that Congress chose the words ‘relating to’ for a purpose.” Hubbard, 480 F.3d at 348. So we join the majority of circuits that have considered the question and conclude that “relating to” as used in § 2252A(b)(1) retains its ordinary, broad meaning. Applying that reading here, the district court properly applied § 2252A(b)(1)'s sentencing enhancement to Mendoza's federal offense.

B.

Mendoza contends that the district court plainly erred in two ways when ordering restitution: (1) by ordering restitution under an inapplicable statute; and (2) in failing to conduct a proximate-cause analysis as required by the correct statute and Supreme Court precedent. We discuss each issue in turn.

1.

The PSR, adopted by the district court, recommended restitution pursuant to 18 U.S.C. § 3663A.⁵ But that statute does not authorize restitution in Mendoza's case because restitution for child-pornography offenses is governed by 18 U.S.C. § 2259, not § 3663A. United States v. West, 137 F.4th 395, 402 (5th Cir. 2025). As Mendoza argues, the district court thus ordered restitution under an inapplicable statute, thereby committing an error, one that was plain or obvious. Id.

But that alone is insufficient to merit reversal: Mendoza offers no argument as to how this error either affected his substantial rights, or, if left uncorrected, would “seriously affect the fairness, integrity, or public reputation of judicial proceedings.” Puckett v. United States, 556 U.S. 129, 135 (2009). Put differently, Mendoza provides no reason to surmise that the district court would have deviated from ordering the same restitution amounts had the court applied § 2259 instead. Mendoza's failure to address the third or fourth prongs of plain error review dooms the first of his two challenges to the restitution orders. See United States v. Sanches, 86 F.4th 680, 687 n.3 (5th Cir. 2023); United States v. Capistrano, 74 F.4th 756, 772 (5th Cir. 2023).

2.

Nonetheless, we must evaluate Mendoza's restitution order under the proper statutory authority to assess Mendoza's second challenge. See West, 137 F.4th at 402. Section 2259 commands that the district court “shall order restitution” in “the full amount of the victim's losses.” 18 U.S.C. §§ 2259(a); 2259(b)(1). The Supreme Court has observed that this “mandatory restitution scheme” is consistent with the statute's “penological purposes.” Paroline v. United States, 572 U.S. 434, 457 (2014). The statute aims to ensure that offenders are not left with the “mistaken impression that child pornography possession ․ is a victimless crime.” Id. at 458. Indeed, “every viewing of child pornography is a repetition of the victim's abuse” because the “unlawful conduct of everyone who reproduces, distributes, or possesses the images of the victim's abuse” contributes to the problem: Demand creates supply. Id. at 457. Section 2259's language thus communicates “Congress' clear intent that victims of child pornography be compensated by the perpetrators who contributed to their anguish.” Id.

Accordingly, the Supreme Court has determined that courts should order restitution that “comports with the defendant's relative role in the causal process that underlies the victim's general losses.” Id. at 458. Restitution is appropriate under § 2259 “only to the extent the defendant's offense proximately caused a victim's loss.” Id. at 448. If a court orders a defendant to pay restitution without determining the degree to which his conduct proximately caused the victim's loss, then the restitution “necessarily exceeds the statutory maximum.” United States v. Winchel, 896 F.3d 387, 389 (5th Cir. 2018). And a court's failure to undertake this analysis meets the requirements for plain error. See West, 137 F.4th at 403.

Paroline emphasized that the proximate-cause inquiry, while essential, is not an analytical straitjacket. Rather, “a court must assess as best it can from the available evidence the significance of the individual defendant's conduct in light of the broader causal process that caused the victim's losses.” Paroline, 572 U.S. at 459. The Paroline Court listed several factors as “rough guideposts” for the proximate-cause inquiry:

the number of past criminal defendants found to have contributed to the victim's general losses;

reasonable predictions of the number of future offenders likely to be caught and convicted for crimes contributing to the victim's general losses;

any available and reasonably reliable estimate of the broader number of offenders involved (most of whom will, of course, never be caught or convicted);

whether the defendant reproduced or distributed images of the victim;

whether the defendant had any connection to the initial production of the images;

how many images of the victim the defendant possessed;

and other facts relevant to the defendant's relative causal role.

Id. at 460. But those factors “need not be converted into a rigid formula.” Id. Important here, Paroline also noted that the proximate-cause analysis is particularly difficult when the defendant is “one of thousands who have possessed and will in the future possess the victim's images but who has no other connection to the victim.” Id. at 449.

In Mendoza's case, the district court adopted the PSR, which listed thirteen of the fourteen victims and included their respective restitution requests. The Government adduced the fourteenth victim and her restitution request at Mendoza's sentencing hearing. The Government then submitted thorough victim-impact statements describing the pain and trauma experienced by each of the fourteen children. The district court ordered the requested restitution amounts for all fourteen victims.

Mendoza argues that the district court did not conduct a Paroline proximate-cause analysis whatsoever. That is largely incorrect. True enough, the district court nowhere mentioned Paroline—but our precedents impose no magic words requirement for restitution calculations. See United States v. Teijeiro, 79 F.4th 387, 394–95 (5th Cir. 2023) (upholding restitution orders even when “the district court did not address Paroline”). The question is whether, as to each of the fourteen identified victims, the district court did enough to satisfy Paroline's proximate-cause inquiry.

For eleven of the victims, the Government submitted not only extensive victim-impact statements but also detailed letters written by the victims' attorneys that included breakdowns of each victim's estimated total financial loss, the harm suffered by the child, and the long-term psychological impacts of that harm on that child.⁶ All eleven letters specifically referenced Paroline, and most of them analyzed the Paroline factors in depth. While the district court did not explicitly state that it relied on the victim-impact statements and attorneys' letters, the court stated that it had “seen all that” when the Government noted that it had submitted the restitution packets and accompanying requests to both the court and Mendoza's counsel. And the district court read directly from several of these statements at the sentencing hearing while admonishing Mendoza. Moreover, as Mendoza's counsel acknowledged, the statements were available to Mendoza; counsel simply stated that “it'd be traumatic and painful” for Mendoza to read them.⁷

While the record could surely have been clearer, this is sufficient to support the restitution orders for these eleven victims on plain error review. The victim-impact statements thoroughly documented the abuse inflicted upon each of the children. And as was the case in Teijeiro, the Government submitted “detailed information from professionals regarding the victims' economic losses and future treatment expenses.” 79 F.4th at 391. The restitution amounts requested, and ordered, were generally a fraction of the victims' estimated losses.⁸ Cf. id. at 395. And “there is no indication of duplicative recovery.” Id. at 396. Considering the victim-impact statements, the attorneys' letters, and the victims' restitution amounts compared to their estimated losses—all of which were before the court and provided to Mendoza's counsel before sentencing—the restitution ordered by the district court for these eleven victims satisfied Paroline.

Mendoza's attempted rebuttals are unconvincing. He primarily relies on our decision in West to argue that the district court failed to undertake the required proximate-cause analysis. In West, the district court ordered a defendant to pay $6,000 in restitution after he pled guilty to child pornography offenses. 137 F.4th at 398. Although the victim's mother submitted a victim-impact statement, the mother did not request restitution, and the PSR reflected that omission. Id. Yet the PSR recommended restitution anyway, apparently because the statute mandated it, and the district court adopted the PSR. Id. Our court vacated the restitution order because the “record in th[e] case” was “devoid of any proximate-cause analysis performed by the district court as required by Paroline.” Id. at 400. There was only “a free-floating restitution order of $6,000, untethered to any determination of loss suffered by the victim in this case.” Id.

If anything, West sharply contrasts with Mendoza's case. First, the victims in this case did request restitution, as evidenced by the PSR and the written submissions provided to the district court. More importantly, the requested restitution amounts were not “free-floating” dollar figures “untethered to any determination of loss.” Id. To the contrary, the victim-impact statements and attorneys' letters exhaustively catalogued the victims' losses—physical, psychological, and economic—and tied the amounts requested to those losses. Unlike in West, there is no difficulty in determining how the district court arrived at the precise restitution amounts that it ordered for these eleven victims: It clearly relied on—and in part recited—the information in the victim-impact statements and attorneys' letters.

Mendoza next invites us to compare the record in his case to that in West, observing that unlike in West, his PSR did not include any victim-impact statements. He insists that there was no proximate-cause analysis “in the record, certainly not an explicit one,” and he asserts that the term “proximate cause” appears “nowhere in the pleadings, the PSR, or the sentencing transcript.” Mendoza's contention seems to be that no sufficient Paroline analysis could have taken place because the district court's record did not include the documentation in support of the restitution orders.

We are unpersuaded by this fixation on the district court's docket. It is true that the victim-impact statements and attorneys' letters were not addended to Mendoza's PSR or otherwise filed of record in the district court.⁹ But we fail to see why those documents' absence from court's docket changes the outcome in Mendoza's case, especially on plain error review. Mendoza never disputes that he had access to the documents; nor does he question their veracity or whether the district court indeed relied on them. And he does not argue the proposition—much less support it—that an otherwise adequate Paroline analysis is rendered infirm because materials on which the court relied in ordering restitution, while authentic and available to the defendant, were not filed of record at the time.

For good reason: That proposition appears nowhere in our precedent. To the contrary, the thrust of Paroline is that “a court applying § 2259 should order restitution in an amount that comports with the defendant's relative role in the causal process that underlies the victim's general losses.” 572 U.S. at 458. For these eleven victims, that standard has been met, if untidily, at least for the purposes of plain error review.

In his reply brief, Mendoza adds one last argument: None of the victim-impact statements or attorneys' letters “say anything at all about Mendoza's conduct or his relative contribution to the losses the victims have suffered.” In his view, the “restitution packets provided little to no basis” for the district court to assess Mendoza's role in the victims' harm. He therefore concludes that the documents are “[not] enough to support the individual restitution awards in this case.”

This is a new argument. Before filing his reply brief, Mendoza's position had consistently been that the district court did not perform Paroline's proximate-cause analysis at all, not that the district court performed that analysis incorrectly. There is a difference between these two claims. See United States v. Alfred, 60 F.4th 979, 982 (5th Cir. 2023). And arguments first raised in a reply brief are not properly before us. See McCaig v. Wells Fargo Bank (Texas), N.A., 788 F.3d 463, 479 n.7 (5th Cir. 2015) (“Arguments first raised in a reply brief are waived.”).

The upshot: For eleven of Mendoza's victims—“Pia,” “Dipper,” “Taylor,” “Maria,” “Lily,” “Sarah,” “Jordan,” “April,” “John Doe 3,” “John Doe 4,” and “Sloane”—the district court's restitution orders satisfied Paroline. On plain error review, we affirm as to those eleven awards.

That leaves three remaining victims: “Donatello,” “Anna,” and “Ali.” We conclude that as to these three, there is insufficient indication that the court conducted a Paroline analysis.

The PSR indicates that “Donatello” requested $5,000 in restitution out of his estimated total loss of $20,000. The PSR also shows that “Anna” requested $3,000 in restitution, but it did not list her estimated total loss. And “Ali,” the fourteenth victim, is not listed in the PSR. The Government disclosed her at sentencing and stated that she requested $5,000 in restitution, but there was no reference to her estimated total loss. The Government submitted victim-impact statements for all three, but not letters from their attorneys detailing these victims' estimated financial losses or harm suffered by each child.

Without more, and unlike the other eleven victims, there is no indication that the district court synthesized the available information regarding these victims into the proximate-cause analysis required by Paroline in ordering restitution. To be sure, their victim-impact statements are no less tragic and troubling than those of the other victims. But those statements alone do not constitute a proximate-cause analysis. Two of the victims did not include their estimated losses, and even for the one who did, the only documentation of his loss is a lone number in the PSR with no further support. We are left with little indication as to why the district court ordered the restitution amounts it did, other than that those amounts are what the victims requested.

We therefore vacate the restitution order as to “Donatello,” “Anna,” and “Ali” and remand for further proceedings. On remand, the district court may, in its discretion, allow the Government to “present new evidence justifying the order[s] of restitution.” West, 137 F.4th at 403. And we note that “this may well lead to a restitution order requiring [Mendoza] to pay far more than [what] he was initially ordered to pay.” Id. at 403 n.3.

IV.

We conclude that the use of the phrase “relating to” in 18 U.S.C. § 2252A(b)(1) retains its ordinary, broad meaning. In doing so, we align with the majority of circuits to have considered the question. Under that reading, the district court properly applied the statute's sentencing enhancement to Mendoza's federal offense, based on his prior state conviction “relating to ․ possession ․ of child pornography.” 18 U.S.C. § 2252A(b)(1). We therefore AFFIRM the district court on that issue.

The district court obviously erred in ordering restitution under an inapplicable statute. But that error did not affect Mendoza's substantial rights because there is no indication the district court's award would be any different had it applied the correct statute.

On substance, for eleven of the fourteen victims, listed supra, the district court's restitution order satisfied Paroline's required proximate-cause analysis on plain error review. The victim-impact statements and attorneys' letters relied upon by the district court, and provided to Mendoza and his counsel, provide sufficient support for the restitution amounts awarded, irrespective of whether those documents were filed of record in the district court. We therefore AFFIRM the district court's restitution order as to those eleven victims.

However, for the remaining three victims, also listed above, the district court reversibly erred by failing to conduct Paroline's required proximate-cause analysis. We therefore VACATE the restitution ordered for those three victims and REMAND for further proceedings consistent with this opinion.

“I think I've seen all that—or somewhere.” The majority assures us that this remark adequately substitutes for Paroline's rigorous restitution analysis. See Paroline v. United States, 572 U.S. 434, 459–62 (2014). Yet where “[t]he record ․ is devoid of any proximate-cause analysis,” the district court has “no basis upon which” to order restitution under 18 U.S.C. § 2259. United States v. West, 137 F.4th 395, 400, 403 (5th Cir. 2025) (emphasis added). Such an error necessarily “affects a defendant's substantial rights,” and we must vacate the award on plain-error review. Id. (citations omitted). Surely then, we must vacate the restitution award here. The majority says no.

I disagree.

No doubt, our court in Tejeiro affirmed a restitution award on plain error review, even though “the district court did not address Paroline.” United States v. Teijeiro, 79 F.4th 387, 394–96 (5th Cir. 2023) (citation omitted). But there, “the district court adopted the PSR, which included the estimated total loss of each victim and an addendum with victim impact statements.” Id. Since the documents incorporated into the PSR reflected a Paroline analysis that could sustain the restitution awarded, the panel affirmed. Id.

The majority attempts a Tejeiro-esque reconstruction of the district court's Paroline analysis—but on a materially different record. Here, the restitution documents were not attached to the PSR, nor were they filed in the district court. The only reason we have them before us is that the government supplemented them on appeal. This is hardly equivalent to restitution documents filed with, and then incorporated into, the PSR, which the district court explicitly adopted.

The closest the district court came to a causation analysis was a remark that it had “seen all that—or somewhere,” and a few references to the victim impact statements when admonishing the defendant. This satisfies the majority that the documents the government furnishes on appeal are, in fact, the documents before the district court. In other words, the majority finds that the documents are authentic. But “as an appellate court, we cannot take evidence or hear matters initially. We are dependent entirely on the record made in a trial court.” In re Montes, 677 F.3d 415, 416 (5th Cir. 1982) (per curiam). This is why “[f]or decades this court has ․ decline[d] to consider the merits of issues based on new evidence furnished for the first time on appeal.” United States v. Rudolph, 103 F.4th 356, 362 (5th Cir. 2024). Because the majority resolves the Paroline documents' authenticity, it exceeds our role as a reviewing court.

The fact remains that there is no evidence in the record that the district court ever analyzed proximate cause. And, although the government could have filed the Paroline documents under seal in the district court, it never did. See West, 137 F.4th at 402–03 (“The burden of demonstrating the amount of the loss sustained by a victim as a result of the offense shall be on ․ the [g]overnment.” (citation modified) (quoting 18 U.S.C. § 3664(e))). The majority obligingly fixes these errors on appeal. In this, I cannot join them. I would vacate the restitution award entirely.¹

FOOTNOTES

1.   The full text of the statute reads:Whoever violates, or attempts or conspires to violate, [§ 2252A(a)(1), (2), (3), (4), or (6)] shall be fined under this title and imprisoned not less than 5 years and not more than 20 years, but, if such person has a prior conviction under this chapter, section 1591, chapter 71, chapter 109A, or chapter 117, or under the Uniform Code of Military Justice or the laws of any State relating to aggravated sexual abuse, sexual abuse, or abusive sexual conduct involving a minor or ward, or the production, possession, receipt, mailing, sale, distribution, shipment, or transportation of child pornography, or sex trafficking of children, such person shall be fined under this title and imprisoned for not less than 15 years nor more than 40 years.18 U.S.C. § 2252A(b)(1) (emphasis added).

2.   “An unpublished opinion issued after January 1, 1996 is not controlling precedent, but may be persuasive authority.” Ballard v. Burton, 444 F.3d 391, 401 n.7 (5th Cir. 2006) (citing 5TH CIR. R. 47.5.4).

3.   See United States v. Trahan, 111 F.4th 185, 192 (1st Cir. 2024); United States v. Portanova, 961 F.3d 252, 258 (3d Cir. 2020); United States v. Liestman, 97 F.4th 1054, 1062 (7th Cir. 2024) (en banc); United States v. Mayokok, 854 F.3d 987, 993 (8th Cir. 2017); United States v. Bennett, 823 F.3d 1316, 1322 (10th Cir. 2016); but see United States v. McGrattan, 504 F.3d 608, 615 (6th Cir. 2007); United States v. Reinhart, 893 F.3d 606, 614 (9th Cir. 2018).

4.   Some circuits addressed 18 U.S.C. § 2252(b)(1)'s identically worded enhancement rather than the text of § 2252A(b)(1), at issue here. But neither party has asserted a difference between the two. “Given the identical operative language, we assume that those circuits would apply the same analysis to § 2252A's enhancement.” Trahan, 111 F.4th at 192 n.5.

5.   Section 3663A(a)(1) requires, in relevant part, that “when sentencing a defendant convicted of an offense described in subsection (c), the court shall order, in addition to ․ any other penalty authorized by law, that the defendant make restitution to the victim of the offense[.]”

6.   The pseudonyms of these eleven victims are “Pia,” “Dipper,” “Taylor,” “Maria,” “Lily,” “Sarah,” “Jordan,” “April,” “John Doe 3,” “John Doe 4,” and “Sloane.”

7.   As further discussed above the line, and contra the dissent, post, at 21, Mendoza has never contested the authenticity of the restitution materials submitted by the Government to both the court and Mendoza, or asserted that the record on appeal somehow departs from what was provided prior to sentencing. Nor does he challenge that the district court indeed relied on the documents in formulating its restitution order.

8.   For example, “Pia” requested $7,000 in restitution despite having an estimated loss of nearly $2 million; “Lily” requested $10,000 in restitution despite having a loss of over $6 million; and “April” requested $10,000 in restitution despite having a loss of about $2.7 million.

9.   The documents are available to us because we granted the Government's motion to supplement the record on appeal.

1.   I concur with the majority's opinion to the extent it rejects Mendoza's challenge to his sentencing enhancement under 18 U.S.C. § 2252A.

CORY T. WILSON, Circuit Judge: