Opinion
No. 22-10509
08-28-2023
UNITED STATES of America, Plaintiff-Appellee, v. Lawrence F. CURTIN, Defendant-Appellant.
Jorge Roberto Delgado, Brittany Bull Panuccio, DOJ-USAO, Southern District of Florida, Daniel Matzkin, Lisa Tobin Rubio, U.S. Attorney Service - Southern District of Florida, Jason Wu, U.S. Attorney Service - SFL, Miami, FL, for Plaintiff-Appellee. Srilekha Jayanthi, Michael Caruso, Federal Public Defender's Office, Southern District of Florida, Miami, FL, Bernardo Lopez, Federal Public Defender's Office, Sowmya Bharathi, Akerman, LLP, Fort Lauderdale, FL, Lawrence F. Curtin, BUTNER, NC, for Defendant-Appellant.
Appeal from the United States District Court for the Southern District of Florida, D.C. Docket No. 1:21-cr-20102-wfjg-1 Jorge Roberto Delgado, Brittany Bull Panuccio, DOJ-USAO, Southern District of Florida, Daniel Matzkin, Lisa Tobin Rubio, U.S. Attorney Service - Southern District of Florida, Jason Wu, U.S. Attorney Service - SFL, Miami, FL, for Plaintiff-Appellee. Srilekha Jayanthi, Michael Caruso, Federal Public Defender's Office, Southern District of Florida, Miami, FL, Bernardo Lopez, Federal Public Defender's Office, Sowmya Bharathi, Akerman, LLP, Fort Lauderdale, FL, Lawrence F. Curtin, BUTNER, NC, for Defendant-Appellant. Before Wilson, Newsom, and Luck, Circuit Judges. Newsom, Circuit Judge:
Lawrence Curtin—who has long struggled with serious mental-health issues—has a pattern of threatening judges. This case arises out of a threat that he recently made against a federal magistrate judge in his hometown of Fort Pierce, Florida. Curtin was convicted in federal court of (1) mailing a threatening communication, in violation of 18 U.S.C. § 876(c), and (2) threatening a federal official, in violation of 18 U.S.C. § 115(a)(1)(B). For his crimes, he was sentenced to 60 months in prison. Curtin now challenges his convictions and sentence on five grounds. After careful consideration, we affirm.
I
The story underlying this appeal begins in 2012, when Curtin was injured in a car wreck. The accident eventually spawned four lawsuits and, more troublingly, two letters threatening judges—including the one underlying the convictions at issue here. Here are the details: Curtin initially filed but lost a personal-injury action in Florida state court. He followed up with back-to-back civil suits in federal court. Both cases were initially assigned to Magistrate Judge Shaniek Maynard, who recommended that they be dismissed. Curtin separately complained to the Florida Judicial Qualification Commission about the handling of his original case by state-court Judge Janet Croom. The commission referred Curtin for prosecution on the ground that his complaint contained a threat—it invoked the "Biblical law which states an 'eye for an eye' " and expressed Curtin's view that he had an "obligation . . . to stop Croom." The charges were ultimately dismissed, however, when a Florida court found Curtin incompetent to stand trial.
Continuing his litigation flurry, Curtin filed yet another federal suit challenging the state-court decisions—this time adding the "Florida State Court System" as a defendant, alleging that it was (or its members were) part of an organized-crime conspiracy. That case, too, was assigned to Judge Maynard, who again recommended dismissal. The objection that Curtin lodged in response to Judge Maynard's report and recommendation forms the basis of this case. Judge Maynard interpreted Curtin's objection as containing a threat to her and her family. Here's the key passage:
WHERE IN MY JUNE 23, 2018 LETTER DO I THREATEN DEATH OR BODILY HARM TO [JUDGE] CROOM? NO WHERE! My June 23, 2018 letter as you will note is addressed to the judicial qualification commission (JQC). YOU DO NOT ADDRESS A LETTER TO THE JQC TO THREATEN A JUDGES [SIC] PERSON. YOU ADDRESS IT TO THE JQC TO THREATEN A JUDGES [SIC] POSITION. Maynard is unable to understand this. I also named Maynard as an addier [sic] and abetter. Maynard knew about the defendants refusing my heart medication in an effort to kill me yet SHE COVERED IT UP.
https://www.youtube.com/watch?v=a2vUNuX5Hgl
It is obvious from the totality of words in the song including its title that I am
threatening Maynard with death and bodily harm. Also by holding onto the hand of the preacher of color that I am threatening Maynard who is a woman of color with death.
The clip is accessible here. See Video, Doc. 198 (https://www.ca11.uscourts.gov/media sources).
The YouTube clip featured a video of Curtin listening to the gospel hymn "Road to Glory" in a church, approaching the pulpit, and taking a black preacher's hand. Judge Maynard is black, and her father was a pastor who had been working in the Fort Pierce community—where both Curtin and Judge Maynard lived—for about 20 years.
A jury convicted Curtin of mailing a threatening communication, in violation of 18 U.S.C. § 876(c), and of threatening a federal official, in violation of 18 U.S.C. § 115(a)(1)(B). He now appeals, raising several challenges to his convictions and sentences. We will take up Curtin's contentions in turn, and we will provide additional factual and procedural detail as necessary.
II
Curtin first contends that there was insufficient evidence to convict him. We review sufficiency-of-the-evidence challenges de novo, United States v. Kelly, 888 F.2d 732, 739-40 (11th Cir. 1989), making "[a]ll factual and credibility inferences" in the government's favor, United States v. Cooper, 203 F.3d 1279, 1285 (11th Cir. 2000). "In order to find the evidence sufficient, we need not exclude every reasonable hypothesis of innocence or find the evidence wholly inconsistent with every conclusion except that of guilt, provided that a reasonable factfinder could find that the evidence establishes guilt beyond a reasonable doubt." Kelly, 888 F.2d at 740. We consider Curtin's challenges to each of his convictions separately.
A
To obtain a conviction under 18 U.S.C. § 876(c), the government must prove beyond a reasonable doubt that the defendant (1) knowingly sent a message through the mail, (2) knew that the mailing contained a "true threat," and (3) intended (or at least knew) that the statement would be viewed as a threat. 18 U.S.C. § 876(c); cf. United States v. Oliver, 19 F.4th 512, 517 (1st Cir. 2021) (reciting § 876(c)'s elements). Curtin concedes that he knowingly mailed the objection to Judge Maynard's report and recommendation. The questions, therefore, are whether he knew that the mailing contained a true threat and intended that Judge Maynard would view it as such.
Pointing to Elonis v. United States, 575 U.S. 723, 135 S.Ct. 2001, 192 L.Ed.2d 1 (2015), the government suggests that § 876(c)'s third element requires proof of purpose or intent. See Br. of Appellee at 10; cf. also United States v. Mabie, 862 F.3d 624, 632 (7th Cir. 2017) (accepting the government's "conce[ssion] that § 876(c) is a specific-intent crime, requiring proof that [the defendant] sent his letter 'for the purpose of issuing a threat, or with knowledge that the communication [would] be viewed as a threat' "). That may be right. We note, though, that Elonis dealt with a statute, 18 U.S.C. § 875(c), that contained no explicit mens rea element. See 18 U.S.C. § 875(c) ("Whoever transmits in interstate or foreign commerce any communication containing any threat . . . ."). Section 876(c), by contrast, begins with the phrase, "Whoever knowingly . . . ." As the Supreme Court has observed, "courts ordinarily read a phrase in a criminal statute that introduces the elements of a crime with the word 'knowingly' as applying that word to each element." Flores-Figueroa v. United States, 556 U.S. 646, 652, 129 S.Ct. 1886, 173 L.Ed.2d 853 (2009); see also United States v. Bachmeier, 8 F.4th 1059, 1063-64 (9th Cir. 2021) ("Other provisions in § 876 criminalize actions 'with intent to extort,' but subsection (c) contains no such language."); United States v. Chapman, 866 F.3d 129, 134 (3d Cir. 2017) (adopting a knowledge standard for § 876(c) after Elonis, but without citing it). Ultimately, we needn't decide the mens rea issue. Whether § 876(c) requires purpose or intent, or something less, like knowledge, the evidence here—which we'll discuss in detail—is sufficient.
" 'True threats' encompass those statements where the speaker means to communicate a serious expression of an intent to commit an act of unlawful violence to a particular individual or group of individuals." Virginia v. Black, 538 U.S. 343, 359, 123 S.Ct. 1536, 155 L.Ed.2d 535 (2003). "The speaker need not actually intend to carry out the threat"; rather, "a prohibition on true threats 'protect[s] individuals from the fear of violence' and 'from the disruption that fear engenders,' in addition to protecting people 'from the possibility that the threatened violence will occur.' " Id. at 359-60, 123 S.Ct. 1536 (alteration in original) (quoting R.A.V. v. City of St. Paul, 505 U.S. 377, 388, 112 S.Ct. 2538, 120 L.Ed.2d 305 (1992)).
The Supreme Court's recent decision in Counterman v. Colorado, 600 U.S. 66, 143 S. Ct. 2106, 216 L.Ed.2d 775 (2023), which held that a state need only prove that a defendant acted with a mens rea of at least recklessness in order to satisfy First Amendment concerns, is irrelevant here. As explained in text, the record evidence sufficiently demonstrates that Curtin acted with a mens rea of at least knowledge, which surpasses recklessness.
Curtin's mental state, of course, is provable through circumstantial evidence. See Liparota v. United States, 471 U.S. 419, 434, 105 S.Ct. 2084, 85 L.Ed.2d 434 (1985); United States v. Hurley, 755 F.2d 788, 790 (11th Cir. 1985). With respect to § 876(c)'s second and third elements, the strongest items of evidence illustrating Curtin's state of mind are his own words. In plain terms, Curtin stated in his objection that he thought "[i]t [was] obvious from the totality of words in the song including its title that [he was] threatening Maynard with death and bodily harm" and, further, that he was "threatening Maynard who is a woman of color with death." And as if to underscore his seriousness, Curtin buttressed his words with all caps accusations (e.g., "SHE COVERED IT UP"), underlined emphasis ("Maynard is unable to understand this"), and an explicit reference to Judge Maynard's race.
Curtin now implies that he was being sarcastic and insists that, in any event, he didn't intend the statement as a threat. Perhaps. All that matters for present purposes, though, is that it wouldn't be unreasonable for a factfinder to conclude, as the jury here clearly did, that Curtin meant what he said and that he meant to threaten Judge Maynard. And indeed, we have already rejected a sufficiency-of-the-evidence challenge in similar circumstances—there, where a defendant called his threat to assassinate the President "political hyperbole." United States v. Callahan, 702 F.2d 964, 965-66 (11th Cir. 1983) (distinguishing Watts v. United States, 394 U.S. 705, 707-08, 89 S.Ct. 1399, 22 L.Ed.2d 664 (1969)).
For similar reasons, we needn't get bogged down in the parties' competing interpretations of the song, "Road to Glory"—whether it was, as Curtin insists, an innocuous attachment or instead, as the government says, a broadening of the threat to include Judge Maynard's family. Given the applicable standard of review, and construing the facts in the government's favor, we have no trouble concluding that a reasonable jury could have found that the video corroborated rather than mitigated Curtin's plain-language threats.
In sum, there was ample evidence to support the jury's determination that Curtin knowingly sent a true threat and intended that it would be viewed as such—and thus to convict him under § 876(c).
B
For the same reasons, there was sufficient evidence to convict Curtin of threatening a federal judge in violation of 18 U.S.C. § 115(a)(1)(B). That statute requires the government to prove beyond a reasonable doubt that the defendant (1) "threaten[ed] to assault, kidnap, or murder" a federal judge (2) with "intent to retaliate" against her "on account of the performance of official duties." Id. As we have explained, a jury could reasonably have concluded from the text and context of Curtin's objection that he meant to threaten Judge Maynard with "death" and "bodily injury"—i.e., "murder" and "assault" within the meaning of the statute. And because he lodged his objection in response to Judge Maynard's report and recommendation, the jury could also have reasonably concluded that he threatened her in an effort to "retaliate" against her "on account of [her] performance of official duties."
III
Curtin next argues that the district court erred when it denied his motion to dismiss the indictment on the ground that the government violated 18 U.S.C. § 4241(d). In relevant part, that provision states that a district court may commit a defendant to the "custody of the Attorney General" to be hospitalized for "treatment in a suitable facility" if the "court finds by a preponderance of the evidence that the defendant is presently suffering from a mental disease or defect rendering him mentally incompetent." Id. Importantly here, the statute goes on to say that the defendant's hospitalization is authorized only for "a reasonable period of time, not to exceed four months, as is necessary to determine whether there is a substantial probability that in the foreseeable future he will attain the capacity to permit the proceedings to go forward." Id. § 4241(d)(1).
Here are the facts relevant to Curtin's § 4241(d) challenge: Curtin was originally arrested and detained on August 24, 2020, and a week later he was denied release under the Bail Reform Act on the ground that he was a "danger to the community." See 18 U.S.C. § 3142. Several months later, on November 24, the district court ordered Curtin hospitalized pursuant to § 4241(d), as all agreed that Curtin was not at that time competent to proceed. Unfortunately, because of what the government has described as "logistical challenges and a backlog at [its] psychiatric facilities," Curtin didn't arrive at the hospital until March 22, 2021. At the end of the statutory four-month period on July 22, Curtin requested transfer back to the detention facility in Miami, and the next day he moved to dismiss his indictment on the ground that he had been hospitalized too long. A few weeks later, with Curtin still in the hospital, the district court held that the government had indeed violated § 4241(d) by keeping him beyond the four-month mark. Even so, the court rejected Curtin's contention that dismissal was the proper remedy. Instead, it ordered Curtin discharged from the hospital and returned to detention. In mid-September, the court and the parties received a psychiatric report from the government hospital explaining that Curtin's competency had been restored. The judge accordingly found Curtin competent to proceed.
We review the denial of a motion to dismiss an indictment for abuse of discretion. See United States v. Castaneda, 997 F.3d 1318, 1325 (11th Cir. 2021) (quoting United States v. McPhee, 336 F.3d 1269, 1271 (11th Cir. 2003)). Of course, "[a] district court by definition abuses its discretion when it makes an error of law." Koon v. United States, 518 U.S. 81, 100, 116 S.Ct. 2035, 135 L.Ed.2d 392 (1996).
As best we can tell, Curtin makes three § 4241(d)-related arguments. First, and most fundamentally, he contends (as he did below) that he was hospitalized beyond the statute's four-month deadline and that the only proper remedy for that over-detention is dismissal of the indictment. Second, he contends that the district court miscalculated the length of his hospitalization, which he says should be deemed to have begun with the commitment order on November 24, 2020, rather than when he physically arrived at the facility on March 22, 2021. Finally, he asserts that the government doctors' report detailing their competency findings should have been submitted within the four-month period.
We can make quick work of Curtin's second and third arguments. As for the former, we conclude that Curtin invited—and thereby waived any claim with respect to—the error that he now presses. In the district court, Curtin filed a reply expressly "agree[ing]" that "the four-month period authorized under § 4241(d)(1) began on March 22, 2021 and ended on July 22, 2021." Doc. 78 at 2. And with respect to the latter, there is simply no firm footing in § 4241(d)'s text for a requirement that psychiatric findings be released or received within the four-month period. That provision prescribes a "reasonable period of time, not to exceed four months," in which the government may "hospitalize the defendant." 18 U.S.C. § 4241(d) (emphasis added). To be sure, it goes on to say that the hospitalization's purpose is to "determine" whether the defendant's competency can be restored, but it would stretch the text too far to hold that it imposes a hard deadline for the delivery of the report memorializing that determination. Cf. United States v. Magassouba, 544 F.3d 387, 407-08 (2d Cir. 2008) ("No one's interests—not the parties', not the court's, and not the public's—are well served by encouraging undue haste in § 4241(d)(1) evaluations.").
That leaves us with Curtin's principal objection—that he was actually, physically hospitalized for longer than the four months that § 4241(d) permits. The government concedes that it violated § 4241(d). See Br. of Appellee at 22 ("Again, the parties agree that [the government hospital] held Curtin for more than four months."). The dispute concerns the appropriate remedy. For his part, Curtin insists that, having concluded that he was hospitalized in violation of the statute, the district court should have dismissed the indictment against him.
We disagree. As an initial matter, there's no particular reason to think—at least none has been explained to us—that dismissal would be the (or even an) appropriate response to a § 4241(d) violation, there being no clear logical relationship between the wrong (over-detention) and the remedy (expungement of the entire case). Nor has Curtin pointed us to any precedent that supports his dismissal request. United States v. Donofrio, 896 F.2d 1301 (11th Cir. 1990), on which he principally relies, is off-point. To be sure, we held there that § 4241(d)'s four-month deadline is "mandatory," but we said nothing about the appropriate remedy, let alone that a violation would warrant dismissal of the indictment. See id. at 1302.
Finally, nothing in Dolan v. United States, 560 U.S. 605, 130 S.Ct. 2533, 177 L.Ed.2d 108 (2010), can be read to suggest that dismissal would be the proper remedy for a § 4241(d) violation. The Supreme Court there sorted statutory deadlines into three categories—jurisdictional limitations, claims-processing rules, and time-related directives—and then set out to determine the effect of each with respect to the particular "action to which the statute attached the deadline." Id. at 609-10, 130 S.Ct. 2533. We needn't decide here the precise nature of § 4241(d)'s four-month deadline, because even if it were jurisdictional, it would at most have deprived the district court of the authority to perform "the action to which the statute attache[s] the deadline"—namely, an inmate's continued hospitalization. Section 4241(d), that is, authorizes and limits hospitalizations; it does not authorize, or purport to limit, prosecutions. The most, therefore, that Curtin could get out of Dolan would be an order requiring his release—which is exactly what he got.
In sum, Curtin has offered no persuasive justification for dismissing his indictment as a means of remedying the admitted violation of § 4241. The district court did not abuse its discretion in denying his motion to dismiss.
To the extent that Curtin raises a due process challenge to his over-hospitalization, we reject it for two reasons. First, as explained in text, Curtin was originally detained under the Bail Reform Act, 18 U.S.C. § 3142, and that detention would presumably have continued even if he had never been hospitalized pursuant to § 4241(d). Second, Curtin wasn't detained long enough to violate the Due Process Clause under Jackson v. Indiana, which held only that an incompetent defendant can't constitutionally be confined beyond "the reasonable period of time necessary to determine whether there is a substantial probability that he will attain that capacity in the foreseeable future." 406 U.S. 715, 738-39, 92 S.Ct. 1845, 32 L.Ed.2d 435 (1972) (finding a due process violation when an incompetent defendant had been detained more than three years). Compare, e.g., Magassouba, 544 F.3d at 419 (holding that an inmate's 19-month detention under § 4241(d) did not violate the Due Process Clause).
IV
Curtin next challenges the failure of the entire bench of the Southern District of Florida to recuse sua sponte from his case. Different judges from that district presided over Curtin's pretrial proceedings for more than a year, including those resulting in his § 4241(d) hospitalization, before, in September 2021, his lawyer requested that the court recuse itself and transfer proceedings to the Middle District of Florida. The motion was unopposed, and trial occurred in the Middle District. Curtin now contends that all Southern District judges should have sua sponte recused from his case earlier.
We review recusal decisions for abuse of discretion. United States v. Berger, 375 F.3d 1223, 1227 (11th Cir. 2004). But when a party fails first to seek recusal in the district court—assuming of course that, as here, he challenges a district court's failure—we review only for plain error. See Hamm v. Members of Bd. of Regents, 708 F.2d 647, 651 (11th Cir. 1983). For Curtin to surmount the high plain-error bar, he must show not only that an error occurred, but also that it was plain, that it affected his substantial rights, and that it "seriously affects the fairness, integrity, or public reputation of judicial proceedings." United States v. Olano, 507 U.S. 725, 732-36, 113 S.Ct. 1770, 123 L.Ed.2d 508 (1993) (alteration accepted) (quotation omitted).
We needn't decide whether any of the Southern District's judges erred by failing to recuse because we conclude that any error, if one occurred, wasn't plain. "It is the law of this circuit that, at least where the explicit language of a statute or rule does not specifically resolve an issue, there can be no plain error where there is no precedent from the Supreme Court or this Court directly resolving it." United States v. Lejarde-Rada, 319 F.3d 1288, 1291 (11th Cir. 2003).
Neither the recusal statute nor our general pronouncements about it provide a hard-and-fast rule of the sort that might satisfy the plain-error standard. The statute provides that "[a]ny justice, judge, or magistrate judge of the United States shall disqualify himself in any proceeding in which his impartiality might reasonably be questioned." 28 U.S.C. § 455(a). We have described § 455(a)'s standard, in general terms, as asking whether an "objective, disinterested, lay observer fully informed of the facts underlying the grounds on which recusal was sought would entertain a significant doubt about the judge's impartiality." United States v. Scrushy, 721 F.3d 1288, 1303 (11th Cir. 2013). At least on its own, that fact- and context-specific standard doesn't "specifically resolve" the recusal issue in Curtin's favor.
Nor does the most analogous case, In re Moody, 755 F.3d 891 (11th Cir. 2014), "directly resolve" matters. For starters, the Moody Court itself emphasized—as just explained—that "[r]ecusal decisions under '§ 455(a) are extremely fact driven and must be judged on their unique facts and circumstances more than by comparison to situations considered in prior jurisprudence.' " Id. at 895 (quoting Nichols v. Alley, 71 F.3d 347, 351 (10th Cir. 1995)). The Court then proceeded to consider four factors in concluding that a panel of Eleventh Circuit judges didn't have to recuse from proceedings involving a defendant convicted of a former Eleventh Circuit judge's murder: (1) None of the three judges whose recusal the defendant sought served on the circuit at the time of the murder; (2) none was clerking on the court at the time; (3) no judge "enjoyed a close personal or professional relationship with [the victim judge] or with any member of his immediate family"; and (4) the two circuit judges who had taken part in the State of Alabama's prosecution had already recused. Id. at 895-96. Nothing in Moody's fact-intensive analysis "directly resolve[s]" the question whether all judges in a federal judicial district must recuse in a case involving a threat to another judge in the district.
Curtin contends that Moody "rejected a more stringent test from other circuits requiring recusal only when the presiding judge was also a potential victim." Br. of Appellant at 43-44. We disagree. To the contrary, the Moody panel at least implicitly rejected the broad rule that Curtin proposes—namely, that all judges in a district must recuse after one is threatened. See 755 F.3d at 896. Instead, the panel adopted a narrow interpretation of a relevant Seventh Circuit decision: "As we read it, Nettles stands for the proposition that . . . judges [who] were potential victims of the alleged attack" "must recuse." Id. (citing In re Nettles, 394 F.3d 1001 (7th Cir. 2005)); see also id. (citing Clemens v. U.S. Dist. Ct. for Cent. Dist. of Ca., 428 F.3d 1175, 1179-80 (9th Cir. 2005)).
The district court did not plainly err by failing to sua sponte recuse from Curtin's case.
V
Finally, Curtin contests both the procedural and substantive reasonableness of his sentence. We consider each challenge in turn.
A
Curtin first alleges that the district court committed procedural error by miscalculating his Guidelines range. The Supreme Court has explained that "significant procedural error[s]" include "failing to calculate (or improperly calculating) the Guidelines range, treating the Guidelines as mandatory, failing to consider the § 3553(a) factors, selecting a sentence based on clearly erroneous facts, or failing to adequately explain the chosen sentence—including an explanation for any deviation from the Guidelines range." Gall v. United States, 552 U.S. 38, 51, 128 S.Ct. 586, 169 L.Ed.2d 445 (2007). When we review for procedural error, we consider a "district court's interpretation of a sentencing guideline provision or term de novo." United States v. Dougherty, 754 F.3d 1353, 1358 (11th Cir. 2014).
Curtin contends, in particular, that the district court miscalculated his Guidelines range when it declined to apply a four-point reduction under U.S.S.G. § 2A6.1(b)(6), which decreases the base offense level if "the offense involved a single instance evidencing little or no deliberation." We disagree; we think it clear that Curtin's threat involved ample "deliberation." Curtin (1) composed his objection; (2) tailored it to his victim, in particular with an emphasis on her race; (3) took the time and energy to embed a video in it; (4) tried to deliver it in person, to no avail; and then (5) mailed it. That is more than enough to defeat § 2A6.1(b)(6)'s application. The district court did not err in concluding that Curtin failed to qualify for the four-point reduction.
B
Curtin separately argues that his 60-month sentence is substantively unreasonable—principally, he says, because the judge considered an improper factor during the hearing.
We review a criminal sentence's substantive reasonableness under an abuse-of-discretion standard, even when, as here, it is above the Guidelines range. See Gall, 552 U.S. at 51, 128 S.Ct. 586. A district court commits substantive error, and abuses its discretion, "when it (1) fails to afford consideration to relevant factors that were due significant weight, (2) gives significant weight to an improper or irrelevant factor, or (3) commits a clear error of judgment in considering the proper factors." United States v. Irey, 612 F.3d 1160, 1189 (11th Cir. 2010) (quoting United States v. Campa, 459 F.3d 1121, 1174 (11th Cir. 2006) (en banc) (Birch, J., dissenting)). "The party challenging the sentence bears the burden of establishing that the sentence is unreasonable . . . ." United States v. Early, 686 F.3d 1219, 1221 (11th Cir. 2012). "[A] major variance does require a more significant justification than a minor one." Irey, 612 F.3d at 1196.
As evidence that the judge considered an improper factor, Curtin points to the following statement, which the judge made toward the beginning of the sentencing hearing:
And I do want to say for the record there's some dispute in the record as to the phrase "road to glory." I will say I don't put any stead in it and don't base any sentence on it, but I'm very, very familiar with what I would call Evangelical-type Protestant churches. I grew up in one. And until I could get out of it, we went to church three times a week, Sunday morning, Sunday night, and Wednesday night. And "glory," the reference to "glory," in that milieu means heaven. It doesn't mean glory like we might think George Washington got glory at the end of the Revolutionary War. Protestants in that type of church when they
say "glory," that means heaven. Sometimes they say, my mother's passed; she's in glory. So again, I don't put any stead on it. I don't hinge any sentence on it, but to the extent that's a dispute in the record, I have to tell you what my experience is, and I'm quite certain about that.
The judge said more than that, though. Toward the end of the hearing, the judge emphasized two other reasons as justifying the sentence—namely, Curtin's pattern of threatening behavior and his focus on Judge Maynard's race:
And pursuant to Title 18 United States Code 3551 and 3553, it's the judgment of the Court that Mr. Curtin is committed to the Bureau of Prisons for an upward variance of 60 months. The reason why is this is a pattern. When I get letters that we're going to send a microwave machine or ask permission to blow the brains out of Chief Judge King and that we want sweat, perspiration to poor [sic] from Judge Croom's armpit, I don't know why it is that we keep sending letters to judges. We have an ongoing, disturbing conversation in writing with Judge Bert Jordan—Adalberto Jordan. I'm so bad at that. This is a pattern, and the pattern escalated and it escalated, and I wasn't really impressed with this tincture, if that's the word, this piece of reference to race in this threat.
You know, this is an older white gentleman threatening a young African-American female. And I guess there was some reason to comment on race, but from the victim's point of view, there's a little bit of history in Florida, and that little—it wasn't, you know, like terrible, but that little reference was gratuitous and added to what was a clear threat. So the upper variance is because of the pattern, the need to respect the law, which is grossly disrespected, and the need for public safety.
And he says, well, he hasn't hurt anybody. Well, yes, these threats are hurtful. Someone gets a letter where someone is suggesting we are going to blow their brains out or I'm going to threaten Magistrate Judge Smith in Orlando with bodily injury and death, they're very hurtful. Thank God it hasn't escalated further. So it's public safety and the repeated pattern that I see over a course of many years, almost to the point where marshals beat the path out there to do a threat assessment every time one of these letters comes.
The judge ultimately imposed a 60-month sentence, nearly doubling the government's requested 33 months, which was already "at the top of the guidelines" range. Curtin's attorney generally objected to the "reasonableness of this sentence under 18 U.S.C. §§ 3551 and 3553" and, more specifically, requested that Curtin be designated to a facility that could care for his serious physical-health needs and objected to the "upward variance given . . . the lack of consideration of his mental health issues." The judge responded that he would recommend the suggested facility and explained his decision: "I'm very aware of his health. And frankly, if it weren't for that he would have gotten a higher sentence." Curtin's lawyer once more objected to a "lack of consideration of his mental health issues." Again, the judge responded: "Oh, I have taken that into consideration. I have taken everything. And it is my conclusion based upon the statutory factors that this sentence is sufficient but not greater than necessary to respond to those statutory factors in 18 U.S.C. [§] 3553 . . . . [T]his is a long, ongoing pattern . . . ."
Curtin now argues that his sentence is substantively unreasonable on the grounds that the district court (1) considered an improper factor when he commented on the religious terminology in the YouTube video, (2) failed to properly weigh Curtin's physical- and mental-health issues, and (3) over-emphasized Curtin's history of sending threatening letters to judges.
We can quickly dispense with Curtin's second and third substantive-reasonableness challenges. As to the second, the district judge properly considered Curtin's physical- and mental-health issues and history. The judge specifically stated that he had accounted for them and, indeed, said that he would have imposed a higher sentence if it weren't for them. As to the third, we find that the judge appropriately considered and weighed Curtin's repeated threats to judges. As we will explain, that pattern of threats, when combined with the evidence of racial animus, amply supports the judge's above-Guidelines sentence.
Before we get there, though, we must address Curtin's first challenge—his contention that the judge considered an improper factor when he commented on the religious language and imagery in the "Road to Glory" song. This, we think, presents a closer question. On the one hand, the § 3553(a) factors clearly don't include reliance on one's own personal religious experience. See 18 U.S.C. § 3553(a). In fact, the Sentencing Guidelines make clear that consideration of religion generally (without specifying the defendant's or the judge's) is an improper factor. See U.S.S.G. § 5H1.10. And the judge did go on at some length about his personal experience with evangelical lingo. On the other hand, the judge expressly disclaimed any reliance on that experience, twice stating that he didn't "put any stead" in it and, separately, that he didn't "hinge any sentence on it." Cf. United States v. Schwarzbaum, 24 F.4th 1355, 1364 (11th Cir. 2022) (stating that we would "take the district court at its word" when it described its sentence).
We needn't decide here whether the judge considered an impermissible factor because any error was harmless. See United States v. Williams, 456 F.3d 1353, 1362 (11th Cir. 2006) (stating that we assess harmlessness after the party challenging the sentence carries his "initial burden" of establishing that the court considered an "impermissible factor"), abrogated on other grounds, Kimbrough v. United States, 552 U.S. 85, 128 S.Ct. 558, 169 L.Ed.2d 481 (2007). "[A] district court's consideration of an impermissible factor at sentencing is harmless if the record as a whole shows the error did not substantially affect the district court's selection of the sentence imposed." Williams, 456 F.3d at 1362.
Here, as already explained, the judge expressly considered two other factors when making his sentencing decision, both of which he cited as bases for an upward departure: (1) Curtin's history of making threats; and (2) the evidence of racial animus. See, e.g., Doc. 212-16 at 1 (showing Curtin writing to Judge Sterling Johnson, a district court judge in New York, for permission to "blow [U.S. District Court Judge] James Lawrence King's brains out of his head" because "King had a secret meeting where he incited others to do this to me and it is clearly a precept of Emmanuel Kant's 'Let the Punishment fit the crime.' "). We think that these two factors alone pass muster as "significant justification[s]" for the upward variance, Irey, 612 F.3d at 1196, especially given the deference we give district courts in sentencing matters. So even if the district judge's reference to his own religious experience was improper, it didn't "substantially affect [his] selection of [Curtin's] sentence"—and, therefore, was harmless. Williams, 456 F.3d at 1362.
A sentencing judge may impose an upward variance based on uncharged conduct—here, the pattern of threats—when it relates to the defendant's history or characteristics. See United States v. Overstreet, 713 F.3d 627, 637-38 (11th Cir. 2013).
VI
For the foregoing reasons, we AFFIRM Curtin's convictions and sentence.
Newsom, Circuit Judge, concurring:
Among the many issues that this case presents, one recurs with some frequency, and our treatment of it has always struck me as a little odd. Our precedent has (albeit haphazardly) categorized a criminal defendant's argument that the district court considered an impermissible factor in imposing a sentence as a challenge to the sentence's "substantive" reasonableness, rather than an allegation of "procedural" error. See Maj. Op. at 1311. That didn't—and for reasons I'll explain, still doesn't—make much sense to me. So I decided to look into it.
The deeper I dug, though, the more problems I uncovered. As it turns out, our sentencing precedent is pretty hopelessly conflicted—not only with respect to the categorization of particular sentencing-related challenges as "substantive" or "procedural," but also with respect to the rules that govern the preservation of those challenges for appeal and, as a result, the standards by which we review alleged sentencing errors. In the pages that follow, I hope to (1) unmask the contradictions in our precedent and (2) briefly propose a better way of classifying and adjudicating sentencing-related challenges.
I
As presented to us, the sentencing issues in this case appeared to be pretty straightforward. Curtin raised two familiar sentencing challenges: In fixing his 60-month prison sentence, Curtin complained, the district judge had both miscalculated his Guidelines range and considered an impermissible factor.
What started as benign curiosity about the genesis and rationality of our treatment of a sentencing court's consideration of an impermissible factor as a substantive error turned into a pretty wild ride. Along the way, I discovered that our precedent is confused—and frankly, just sloppy—in at least two respects: (a) how we've gone about categorizing sentencing-related challenges as substantive or procedural and (b) what we've said a defendant needs to do to preserve those challenges. I'll try to explain those bodies of law (such as they are) in turn. Then, having done that, I'll offer in the next Part a few ideas for setting things straight.
A
I'll start with our classification of sentencing challenges as substantive or procedural—and, in particular, with the challenges that Curtin has presented here. First, Curtin contends that the district court miscalculated his Guidelines range by refusing to apply a four-point reduction under U.S.S.G. § 2A6.1(b)(6). See Maj. Op. at 1311. That, the Supreme Court has told us, is a quintessential "procedural" error. Most notably, in Gall v. United States—which for all intents and purposes minted the "substantive" and "procedural" labels—the Court offered an illustrative (if partial) list of "procedural error[s]," as follows:
[T]he appellate court . . . must first ensure that the district court committed no significant procedural error, such as [1] failing to calculate (or improperly calculating) the Guidelines range, [2] treating the Guidelines as mandatory, [3] failing to consider the § 3553(a) factors, [4] selecting a sentence based on clearly erroneous facts, or [5] failing to adequately explain the chosen sentence—including an explanation for any deviation from the Guidelines range.552 U.S. 38, 51, 128 S.Ct. 586, 169 L.Ed.2d 445 (2007) (emphasis and enumeration added). So, Supreme Court precedent is clear: An argument that the sentencing court "improperly calculate[ed] the Guidelines range" is—objectively, verifiably—an allegation of "procedural error." Id. Happily, our precedent has consistently tracked the Supreme Court's own. In United States v. Pugh, we accurately recited the Supreme Court's catalogue of "procedural error[s]"as including the "fail[ure] to calculate (or improperly calculating) the Guidelines range." 515 F.3d 1179, 1190 (11th Cir. 2008). So far as I can tell, we've since remained faithful to that clear—and correct—holding. See, e.g., United States v. Shaw, 560 F.3d 1230, 1237-38 (11th Cir. 2009); United States v. Dougherty, 754 F.3d 1353, 1358-59 (11th Cir. 2014).
The story with respect to the second error that Curtin alleges—that, in imposing his sentence, the district court considered an impermissible factor, see Maj. Op. at 1311-13—is more complicated. With respect to the classification of that issue, our precedent is both embarrassingly inconsistent and, to the extent that it can be deemed to provide an answer, wrong. So far as I can tell, Pugh had the first word about how to characterize an impermissible-factor challenge like Curtin's. There, quoting a pre-Gall decision, we said that "[a] sentence may be substantively unreasonable when the district court . . . bases the sentence on impermissible factors." 515 F.3d at 1191-92 (alteration in original) (emphasis added) (quoting United States v. Ward, 506 F.3d 468, 478 (6th Cir. 2007)). Just months later, though, we said in United States v. Livesay that the consideration of an impermissible factor constitutes "procedural" error. 525 F.3d 1081, 1092 (11th Cir. 2008). But any rule that might have emanated from Livesay appears not to have taken hold because sitting en banc two years later we reiterated the view that a sentencing court's consideration of an impermissible factor implicates substantive (rather than procedural) reasonableness. See United States v. Irey, 612 F.3d 1160, 1189 (11th Cir. 2010) (en banc). And to the extent that Pugh, Livesay, and Irey left any doubt, United States v. Lebowitz, 676 F.3d 1000 (11th Cir. 2012), seemed to resolve it. There, we held—again, and quoting Irey—that a sentence may be substantively unreasonable when the district court "gives significant weight to an improper or irrelevant factor." Id. at 1016 (quoting Irey, 612 F.3d at 1189).
The settlement, though, was short-lived. Here's a rough chronological post-Lebowitz summary of our published, precedential caselaw regarding impermissible-factor challenges:
| 2014 | United States v. Vandergrift, 754 F.3d 1303, 1308 (11th Cir. 2014) (procedural) |
|---|---|
| 2015 | United States v. Rosales-Bruno, 789 F.3d 1249, 1256 (11th Cir. 2015) (substantive) |
| 2015 | United States v. Cavallo, 790 F.3d 1202, 1236-38 (11th Cir. 2015) (substantive) |
| 2017 | United States v. Alberts, 859 F.3d 979, 985 (11th Cir. 2017) (procedural) |
| 2022 | United States v. Coglianese, 34 F.4th 1002, 1007 (11th Cir. 2022) (procedural) |
| 2022 | United States v. Butler, 39 F.4th 1349, 1355 (11th Cir. 2022) (substantive) |
| 2023 | United States v. King, 57 F.4th 1334, 1340 (11th Cir. 2023) (procedural) |
| 2023 | United States v. Oudomsine, 57 F.4th 1262, 1266 (11th Cir. 2023) (substantive) |
Totally unsurprisingly, our unpublished caselaw is just as dissonant, if not more so. Compare, e.g., United States v. Bryant, 809 F. App'x 609, 617 (11th Cir. 2020) (categorizing the consideration of an impermissible factor during sentencing as a procedural error); United States v. Cooper, 779 F. App'x 588, 593 (11th Cir. 2019) (same); United States v. Fox, 650 F. App'x 734, 738-39 (11th Cir. 2016) (same); United States v. Salas-Argueta, 249 F. App'x 770, 772 (11th Cir. 2007) (same), with, e.g., United States v. Scott, 496 F. App'x 992, 995 (11th Cir. 2012) (categorizing the consideration of an impermissible factor during sentencing as a substantive error); United States v. Perez, 396 F. App'x 590, 592 (11th Cir. 2010) (same); United States v. Vidal, 275 F. App'x 873, 877 (11th Cir. 2008) (same). Ordinarily, that wouldn't bother me so much—unpublished decisions aren't precedential, so their mistakes don't infect our jurisprudence more generally. For good or ill, though, the fact is that we decide the vast majority of sentencing appeals in unpublished opinions. Accordingly, there's a serious risk that unless we straighten out our published precedent, the inconsistencies will continue to persist—underground, as it were, and by dint of raw inertia.
When faced with the sort of intra-circuit split that our improper-factor opinions reveal, our "precedent about precedent" privileges the first-decided case. United States v. Madden, 733 F.3d 1314, 1319 (11th Cir. 2013). Here, that means Pugh, as supplemented by our ensuing en banc decision in Irey—both of which clearly held that a court's consideration of an impermissible factor renders a defendant's sentence substantively unreasonable. So despite our flip-flopping, the law of this circuit is clear.
As I see it, though, it's also clearly wrong. Here's why: Pugh, again, is the fountainhead case. Quoting and citing a handful of pre-Gall decisions from other circuits, the Pugh panel noted in dicta that " '[a] sentence may be substantively unreasonable when,' " among other reasons, the district court "bases the sentence on impermissible factors." 515 F.3d at 1191-92 (quoting Ward, 506 F.3d at 478, and citing United States v. Ausburn, 502 F.3d 313, 328 (3d Cir. 2007), United States v. Willingham, 497 F.3d 541, 543-44 (5th Cir. 2007), and United States v. Boleware, 498 F.3d 859, 861 (8th Cir. 2007)). The panel never offered any of its own reasons why consideration of an impermissible factor should be viewed as bearing on a sentence's substantive (as opposed to procedural) reasonableness. Rather, it just repeated what others had said in the days leading up to Gall.
Pugh (and its progeny) erred in categorizing a district court's consideration of an impermissible factor as a substantive error. I say so for several reasons. For starters, the Supreme Court in Gall introduced its list of illustrative "procedural error[s]" with the phrase "such as," thereby indicating that it wasn't meant to be exhaustive. 552 U.S. at 51, 128 S.Ct. 586. Relatedly, the consideration of an impermissible factor at sentencing is just the mirror image of one of the "procedural error[s]" that the Gall Court specifically identified: "failing to consider the [proper] factors" as set out in 18 U.S.C. § 3553(a). Id. I don't see any meaningful difference—in terms of distinguishing substantive from procedural error—between failing to consider the proper factors and considering an improper factor. Accordingly, it seems to me that we had it exactly right when we said—only to sub silentio reverse ourselves later—that if, by dint of Gall, a sentencing court's failure to consider the proper factors constitutes a procedural error, "it is only logical" that a court's consideration of an improper factor likewise constitutes procedural, rather than substantive, error. Vandergrift, 754 F.3d at 1308.
Finally, it just makes sense that a district court's consideration of an improper factor would be a procedural, rather than substantive, error. At the risk of oversimplifying, in legal lingo—as in language more generally—the term "procedure" refers to "[a] specific method or course of action," Procedure, Black's Law Dictionary (10th ed. 2014), and a "procedural" legal rule is distinguished from a "substantive" rule this way: "The rules that prescribe the steps for having a right or duty judicially enforced, as opposed to the law that defines the specific rights or duties themselves," Procedural Law, id.; see also e.g., Substantive Law, id. ("So far as the administration of justice is concerned with the application of remedies to violated rights, we may say that the substantive law defines the remedy and the right, while the law of procedure defines the modes and conditions of the application of the one to the other." (quoting John Salmond, Jurisprudence 476 (10th ed. 1947, Glanville L. Williams ed.))). In short, the procedural aspects of a sentencing determination refer to the various decisional inputs—the "methods," "steps," etc.—and the substantive aspect of the determination reflects the output—i.e., the product of those inputs. Maybe I'm too simple, but it just seems pretty straightforward to me.
Cf. Erie R. Co. v. Tompkins, 304 U.S. 64, 58 S.Ct. 817, 82 L.Ed. 1188 (1938), and its progeny.
And to be clear, our miscategorization of impermissible-factor challenges is indicative of a larger problem: We have likewise erroneously classified as substantive other errors that, in fact, are procedural in nature. For instance, we have held that a district court's "fail[ure] to afford consideration to relevant factors that were due significant weight" can render a sentence substantively unreasonable. E.g., Irey, 612 F.3d at 1189. So too, we have suggested, at least, that a court's placement of "undue weight" on a single § 3553(a) factor constitutes a substantive, rather than procedural, error. See, e.g., Oudomsine, 57 F.4th at 1267. Both of those, though—like the miscalculation of the Guidelines range and the consideration of an improper factor—implicate the propriety of the "methods" and "steps" by which the court determines a defendant's sentence, not the reasonableness of the sentence itself. They are input errors, not output errors. Thus, they are—to my way of thinking, at least—procedural, not substantive.
* * *
At this point, one might ask, "So what?" Fair question, but as it turns out, the categorization of an error as substantive or procedural actually matters. It may well affect the rules governing the preservation of appellate challenges—although, for reasons I'll explain in the next section, our preservation precedent is so jacked up that it's hard to tell. But in addition—as a "for instance," but one that seems to bear directly on this case—the classification decision affects how we review for harmlessness. Although we nominally review both substantive and procedural sentencing errors for harmlessness, see United States v. Williams, 456 F.3d 1353, 1362 (11th Cir. 2006), abrogated on other grounds, Kimbrough v. United States, 552 U.S. 85, 128 S.Ct. 558, 169 L.Ed.2d 481 (2007) (substantive); United States v. Rodriguez, 398 F.3d 1291, 1300 (11th Cir. 2005) (procedural), the rules and standards that we apply in doing so are materially different.
An alleged substantive error—like (under our precedent) a sentencing court's consideration of an impermissible factor—is deemed to be harmless whenever "the record as a whole shows the error did not substantially affect the district court's selection of the sentence imposed." Williams, 456 F.3d at 1362 (11th Cir. 2006). And importantly, the defendant bears the burden of proving that the error was not harmless—i.e., that it did, in fact, "substantially affect" his sentence. So if the defendant can't affirmatively show that the error didn't come out in the wash of the "totality of the circumstances" test that governs substantive-reasonableness challenges, he won't be entitled to a remand. In short, he'll lose.
For an alleged procedural error, by contrast, the roles, burdens, and standards are reversed. The government bears the burden of proving harmlessness, and the "standard is difficult to satisfy." United States v. Mejia-Giovani, 416 F.3d 1323, 1327 (11th Cir. 2005). The government can't prevail by pointing to uncertainty or ambiguity about the error's effect on the defendant's sentence; rather, it must affirmatively show "that the error did not affect the sentence, or had but very slight effect." United States v. Campa, 529 F.3d 980, 1013 (11th Cir. 2008) (remanding for resentencing because we were uncertain whether the sentence would have been the same absent the error) (internal quotations omitted). Indeed, the government's burden to prove a procedural error's harmlessness is so high that we've said it's "as difficult for the government to meet . . . as it is for a defendant to meet the third-prong prejudice standard for plain error review." Id. (internal quotations omitted).
Put simply: With respect to substantive errors, the tie goes to the government; for procedural errors, it goes to the defendant. So in a potentially large number of mushy-middle cases, where we're just unsure about an error's precise effect, our categorization of it as substantive or procedural could be outcome-determinative. And we need look no further than this case for a demonstrative: Pursuant to our impermissible-factor precedent, we categorized Curtin's challenge to the district judge's invocation of his own religious experience as an allegation of substantive error. See Maj. Op. at 1312-13. Then, having done so, and pursuant to our applicable harmless-error precedent, we concluded that, in the totality of the circumstances, the judge's statement didn't matter; it was outweighed by his reliance on valid considerations. See id. at 1313-14. Had Curtin's impermissible-factor challenge been classified instead as an allegation of procedural error—as I think it should have been, frankly—the burden would have shifted, and it's at least possible that the result would have flipped.
To sum up, then: We've been inconsistent in how we've categorized impermissible-factor (and other similar) challenges, and, to the extent we seem to have settled on a characterization of such challenges as substantive rather than procedural, I think we've gotten it wrong.
B
Unfortunately, there's more confusion where that came from—in particular, in the rules that we've employed to determine whether a defendant has properly preserved his sentencing challenges.
Recall that Curtin's attorney initially lodged only a general, non-specific objection to his sentence: "[W]e object to the reasonableness of this sentence under 18 U.S.C. 3551 and 3553." As the majority opinion notes, he later amended his objection to challenge more specifically the judge's upward variance on the ground that it reflected a failure to account for Curtin's mental-health issues. See Maj. Op. at 1319. Conspicuously, though, he made no specific mention of the particular issues that he now raises on appeal: the miscalculation of Curtin's Guidelines range and the judge's consideration of an improper sentencing factor. Although the parties haven't focused on the preservation issue before us, I'm not at all sure that either of Curtin's challenges was properly preserved for appellate review by way of his lawyer's blanket "reasonableness" objection.
Turns out our caselaw is just as unsure. Worse, our caselaw is a grab-bag—there's a little something in it for everyone. Based on existing precedent, a party can argue—and a panel might well conclude—pretty much whatever it wants concerning whether a sentencing-related challenge was properly preserved for appellate review. Here's a sampling.
With respect to errors that we have characterized (rightly or wrongly) as bearing on a sentence's substantive reasonableness, we have recognized that a general, blanket objection suffices to preserve the error for appeal. See, e.g., United States v. Carpenter, 803 F.3d 1224, 1232 (11th Cir. 2015) (indicating that a defendant's boilerplate objection to the "substantive and procedural reasonableness of [his] sentence" adequately preserved a substantive-reasonableness challenge and accordingly reviewing under the ordinary abuse-of-discretion standard). And yet we have also held, to the contrary, that a specific objection is required to preserve a substantive-reasonableness challenge. See, e.g., United States v. Cavallo, 790 F.3d 1202, 1237 (11th Cir. 2015) (reviewing a substantive-reasonableness challenge for plain error after concluding that the defendant's general objection failed to preserve the error).
And sentencing-related errors that we've deemed procedural? More of the same—which is to say more inconsistency. Unsurprisingly to me, we have plenty of decisions holding that a boilerplate objection isn't good enough. See, e.g., United States v. Parks, 823 F.3d 990, 996 (11th Cir. 2016) (applying a plain-error standard to a blanket objection to procedural unreasonableness); United States v. Johnson, 694 F.3d 1192, 1195 (11th Cir. 2012) (suggesting that plain-error review applies when a defendant makes only a blanket objection that his sentence is "procedurally unreasonable"). Somewhat more surprisingly, we also have cases holding, to the contrary, that a blanket objection will preserve a procedural-error challenge. See, e.g., Carpenter, 803 F.3d at 1232, 1238 (observing that the defendant "preserved before the district court" a procedural-error argument simply by objecting "to the substantive and procedural reasonableness of [his] sentence"); see also, e.g., Coglianese, 34 F.4th at 1007 (suggesting the same).
And if our published caselaw is bad, then our unpublished caselaw—where, again, most sentencing appeals are handled—is even worse. Compare, e.g., United States v. Rivera, 812 F. App'x 962, 963 n.1 (11th Cir. 2020) (holding that a blanket objection to a sentence's substantive unreasonableness does not suffice to preserve the argument for appeal), and United States v. Jones, 752 F. App'x 858, 859 n.1 (11th Cir. 2018) (holding that a "general objection to the reasonableness of [the defendant's] sentence" does not preserve a procedural-error argument for appeal and reviewing for plain error), with, e.g., United States v. Caulton, No. 21-11035, 2021 WL 4787151, at *2 (11th Cir. Oct. 14, 2021) (holding that a blanket objection to a sentence's substantive unreasonableness suffices to preserve the challenge for appeal), and United States v. Beasley, 562 F. App'x 745, 753 (11th Cir. 2014) (suggesting that a blanket objection to an alleged procedural error suffices to preserve the challenge for appeal).
All of this inconsistency matters—not only to our law, but to real people. Divergent preservation rules beget divergent standards of review, which in turn may well beget divergent results in similar cases. Defendants whose sentencing-related challenges are deemed to have been properly preserved get ordinary abuse-of-discretion review. That's a deferential standard, to be sure, but it's not toothless; it is well established, for instance, that "[a] district court by definition abuses its discretion when it makes an error of law." Koon v. United States, 518 U.S. 81, 100, 116 S.Ct. 2035, 135 L.Ed.2d 392 (1996). But those whose challenges are deemed not to have been preserved confront the plain-error standard, which requires proof not only that "an error occurred" but also that it "was plain," that it "affected [the defendant's] substantial rights," and that "it seriously affected the fairness of the judicial proceedings." United States v. Ramirez-Flores, 743 F.3d 816, 822 (11th Cir. 2014). By any measure, that's a higher bar.
* * *
Along at least two vectors, then—how we've classified sentencing-related challenges as substantive or procedural and what we've required to preserve such challenges for appeal—our precedent is confused and internally inconsistent. So much so, I fear, that any panel could theoretically categorize any error however it chooses, analyze preservation in whatever way it thinks makes sense, and (thereby) apply any of multiple standards of review—all with the full backing of what purports to be binding circuit precedent. But if "precedent" has an essence, surely it's that "[l]ike cases should be decided alike." Brian Garner et al., The Law of Judicial Precedent 21 (2016). Given the sorry state of our sentencing caselaw, I think there's a substantial risk that like cases are not being decided alike. To the contrary, and speaking only for (and of) myself, I fear that familiarity—so many sentencing appeals, and so many of them decided in unpublished decisions—has bred inattention, and perhaps even apathy.
1 think it's time to convene the full Court to set our law straight. In the meantime, I have a few tentative thoughts about how we might improve our approach to these important issues. I turn to those now.
II
If we want to begin to rationalize our sentencing precedent, we need to return to the Supreme Court's post-Booker sentencing decisions—especially Gall and its progeny. They tell us a lot about both (1) the process for categorizing and evaluating sentencing-related challenges and, by extension, (2) the rules that should govern the preservation of those challenges. Ultimately, I think the Supreme Court's decisions demonstrate, first, that policing the line between procedural and substantive errors serves important purposes, and second, that the traditional preservation rules apply irrespective of an error's categorization.
A
First, what does Supreme Court precedent say about how appellate courts should classify and examine sentencing-related challenges? As I've already noted, Gall effectively invented the modern taxonomy of "substantive" and "procedural" sentencing issues. Here's the key passage, part of which we've already reviewed:
Regardless of whether the sentence imposed is inside or outside the Guidelines range, the appellate court must review the sentence under an abuse-of-discretion standard. It must first ensure that the district court committed no significant procedural error, such as failing to calculate (or improperly calculating) the Guidelines range, treating the Guidelines as mandatory, failing to consider the § 3553(a) factors, selecting a sentence based on clearly erroneous facts, or failing to adequately explain the chosen sentence—including an explanation for any deviation from the Guidelines range. Assuming that the district court's sentencing decision is procedurally sound, the appellate court should then consider the substantive reasonableness of the sentence imposed under an abuse-of-discretion standard. When conducting this review, the court will, of course,
take into account the totality of the circumstances, including the extent of any variance from the Guidelines range. If the sentence is within the Guidelines range, the appellate court may, but is not required to, apply a presumption of reasonableness. But if the sentence is outside the Guidelines range, the court may not apply a presumption of unreasonableness. It may consider the extent of the deviation, but must give due deference to the district court's decision that the § 3553(a) factors, on a whole, justify the extent of the variance. The fact that the appellate court might reasonably have concluded that a different sentence was appropriate is insufficient to justify reversal of the district court.552 U.S. at 51, 128 S.Ct. 586.
That paragraph tells us three very important things. First, despite what a long line of our own cases would suggest, "procedural reasonableness" isn't the first step in the Gall analysis—and, for that matter, isn't a thing at all. See, e.g., United States v. Sanchez, 586 F.3d 918, 935 (11th Cir. 2009) (internal quotations omitted) (quoting United States v. Hunt, 459 F.3d 1180, 1182 n.3 (11th Cir. 2006) ("[A] sentence may be reviewed for procedural or substantive unreasonableness.")); Dougherty, 754 F.3d at 1358 ("procedural reasonableness"); United States v. Waters, 937 F.3d 1344, 1358 (11th Cir. 2019) ("procedural reasonableness"); Oudomsine, 57 F.4th at 1264 ("procedural reasonableness"). To the contrary, a careful reading of Gall makes clear that the Supreme Court there identified two types of challenges: (1) those alleging a "procedural error"; and (2) those challenging the sentence's "substantive reasonableness." 552 U.S. at 51, 128 S.Ct. 586 (emphasis added). Our "procedural reasonableness" cases blur that line and, in so doing, erroneously imply that procedure-related sentencing inquiries are more amorphous than they are. Under Gall, procedural error is a binary: it either occurred or it didn't.
Second, both in Gall and since, the Supreme Court has prescribed a distinct order of battle: An appellate court should "first" assure itself that the district court committed no procedural error, and "then" review the sentence itself for substantive reasonableness. Gall, 552 U.S. at 51, 128 S.Ct. 586. As the Court recently reiterated, "[i]f the trial court follows proper procedures and gives adequate consideration to these and the other listed factors, then the question for an appellate court is simply, as here, whether the trial court's chosen sentence was 'reasonable' or whether the judge instead 'abused his discretion in determining that the § 3553(a) factors supported' the sentence imposed." Holguin-Hernandez v. United States, — U.S. —, 140 S. Ct. 762, 766, 206 L.Ed.2d 95 (2020) (quoting Gall, 552 U.S. at 56, 128 S.Ct. 586); accord, e.g., Rosales-Mireles v. United States, — U.S. —, 138 S. Ct. 1897, 1904, 201 L.Ed.2d 376 (2018); Molina-Martinez v. United States, 578 U.S. 189, 136 S.Ct. 1338, 1345, 194 L.Ed.2d 444 (2016).
Third, the Supreme Court's decisions make clear that a reviewing court can apply a "presumption of reasonableness" only at Step 2, when reviewing the sentence for substantive reasonableness—not at Step 1, when reviewing for procedural error. In Gall's predecessor, Rita v. United States, the Supreme Court had held that appellate courts may apply a presumption of reasonableness to sentences that fall within the Guidelines. 551 U.S. 338, 347, 127 S.Ct. 2456, 168 L.Ed.2d 203 (2007). The Guidelines, the Court said, "seek to embody the § 3553(a) considerations," so it's fair to presume that a within-Guidelines sentence reflects a reasonable application of § 3553(a) in a particular case. Id. at 350, 127 S.Ct. 2456. Building on Rita, Gall then prescribed its two-step decisional calculus and reiterated that the presumption of reasonableness applies only at the second step, after the reviewing court has satisfied itself "that the district court's sentencing decision is procedurally sound." 552 U.S. at 51, 128 S.Ct. 586. All of this, as the Supreme Court explained in Rosales-Mireles, "makes eminent sense." 138 S. Ct. at 1910. The district court, it said there, is ultimately charged "with determining whether, taking all sentencing factors into consideration, including the correct Guidelines range, a sentence is sufficient, but not greater than necessary." Id. (internal quotations omitted). But if "the district court is unable properly to undertake that inquiry because of " a procedural error—including, in that case, "an error in the Guidelines range"—then the resulting sentence "no longer bears the reliability that would support a presumption of reasonableness on review." Id. (internal quotations omitted).
The Supreme Court's foundational sentencing precedent also helps to explain the divergent harmlessness standards that apply to substantive and procedural errors. When a defendant alleges a procedural error at Step 1, so to speak, and an appellate court can't tell whether it affected the district court's decision, the sentence isn't "reliab[le]," and remand is required. Rosales-Mireles, 138 S. Ct. at 1910. Things are different at Step 2. So long as the district court committed no procedural error and the totality of the circumstances reveals that an alleged substantive error didn't "substantially affect" the resulting sentence, we're safe to affirm. See Williams, 456 F.3d at 1362.
Once one appreciates that reviews for procedural error and substantive reasonableness are "entirely separate inquir[ies]," id., and, importantly, that the former must precede the latter, it's easy to see how our miscategorization of procedural errors as substantive—most notably, but not exclusively, a district court's consideration of an improper sentencing factor—undermines Gall's ordering principle and skews the decisional calculus. Instead of applying reasonableness review only after we've ensured that the district court's decisionmaking process was procedurally sound by reference to all procedural prerequisites, we have demoted and devalued some of those prerequisites by merging them into (and really submerging them in) the secondary, substantive prong of the analysis.
B
Next: What do the Supreme Court's sentencing decisions tell us about error preservation? Two things, I think. First, and perhaps most fundamentally, an objection on "reasonableness" grounds does nothing—because it's a non sequitur. "Reasonableness," the Court recently reiterated, is neither a trial-court objection nor "the substantive standard that trial courts apply under § 3553(a)." Holguin-Hernandez, 140 S. Ct. at 763. Rather, it's an appellate standard of review: It is simply "the label we have given to the familiar abuse-of-discretion standard that applies to appellate review of the trial court's sentencing decision." Id. at 766 (internal quotations and emphasis omitted); see also Pugh, 515 F.3d at 1188 ("[T]he Court 'expressly equated' reasonableness review 'with the old abuse-of-discretion standard used to review sentencing departures.' "). So, to object, as Curtin's trial counsel did here, to the "reasonableness" of the district court's sentencing decision is an empty gesture—it's the equivalent of saying to the district judge, "I object because you abused your discretion." Neither puts the district court on notice that it has committed any particular error—just that it was wrong in some vague, unspecified way.
Which leads to the second preservation-related lesson that I think we can draw from the Supreme Court's sentencing decisions—namely, that at least as matters currently stand, they don't do anything to alter the usual preservation rules. The Court was recently "ask[ed] to decide" both "what is sufficient to preserve a claim that a trial court used improper procedures in arriving at its chosen sentence" and "when a party has properly preserved the right to make particular arguments supporting its claim that a sentence is unreasonably long." Holguin-Hernandez, 140 S. Ct. at 767 (emphasis omitted). The most it was willing to venture was that the defendant there "properly preserved the claim that his 12-month sentence was unreasonably long by advocating for a shorter sentence and thereby arguing, in effect, that this shorter sentence would have proved 'sufficient,' while a sentence of 12 months or longer would be 'greater than necessary' to 'comply with' the statutory purposes of punishment" within the meaning of § 3553(a). Id.
I think the only fair reading is that the Supreme Court has, at least to this point, left the ordinary preservation rules intact: A criminal defendant who wishes to preserve a claim of error for appellate review must object to the district court's ruling and state "the grounds for that objection." Fed. R. Crim. P. 51(b). A challenge is properly preserved only if it "articulate[s] the specific nature of [the defendant's] objection . . . so that the district court may reasonably have an opportunity to consider it." United States v. Zinn, 321 F.3d 1084, 1090 n.7 (11th Cir. 2003). "A sweeping, general objection is therefore insufficient." Coglianese, 34 F.4th at 1009-10. Nothing in the Supreme Court's sentencing decisions—Holguin-Hernandez or otherwise—suggests that those rules are inapplicable to sentencing challenges.
Accordingly, to preserve for appeal the argument that the district court committed a quintessential "procedural error"—say, a miscalculation of the Guidelines range—a defendant must lodge an objection, in reasonably clear terms, either in a pre-hearing sentencing memorandum or contemporaneously at the hearing itself. See, e.g., Irey, 612 F.3d at 1223 n.44 (holding that arguments made in sentencing memoranda or raised during the sentencing hearings are preserved). The same rule should apply, I contend, to an argument that the sentencing judge considered an improper factor—anything less than a specific objection would fail to alert the district court to the alleged error. So too for all other input-based, procedural errors: Specific error, specific objection.
And what if the defendant's challenge is what I'd call a "true" output-related substantive-reasonableness challenge—for instance, that the district court's chosen sentence is just (my paraphrase) "too long"? There, I think—and as Holguin-Hernandez seems to confirm—all the defendant needs to say is, in effect, "My sentence is too long." See 140 S. Ct. at 766. To be clear, though, the tolerance for greater generality doesn't stem from the fact that the defendant's challenge there is properly, formally categorized as substantive rather than procedural. Instead, it results from a straightforward application of the usual preservation rules: Given the nature of that particular defendant's allegation, that's all he needs to say in order to state "the grounds for [his] objection," Fed. R. Crim. P. 51(b), and thereby give the district court a fair "opportunity to consider it," Zinn, 321 F.3d at 1090 n.7. A different stripe of substantive-reasonableness challenge would likely require a different objection. A defendant taking aim, for instance, at a district court's imposition of a particular condition of supervised release would presumably need to train his fire specifically on that condition. See, e.g., id. at 1087, 1089-91 (holding that "if a defendant fails to clearly articulate a specific objection [to the special conditions of supervised release] during sentencing, the objection is waived on appeal and we confine our review to plain error").
Applying the usual preservation rules to all sentencing challenges—of whatever stripe—dovetails with the practical considerations that the Supreme Court has emphasized in its sentencing decisions. For one, it has stressed that district courts "have an institutional advantage over appellate courts in making [sentencing] determinations." Gall, 552 U.S. at 52, 128 S.Ct. 586. "The sentencing judge is in a superior position to find facts and judge their import under § 3553(a) in the individual case. The judge sees and hears the evidence, makes credibility determinations, has full knowledge of the facts and gains insights not conveyed by the record." Id. at 51, 128 S.Ct. 586 (internal quotations omitted). Given those indisputable realities, it makes perfect sense that we would require a defendant to make his case clearly in the district court so that the judge has the first crack at correcting his own mistakes before an appellate court steps in.
For another, application of the usual rules furthers the goal of ensuring consistency and uniformity between and among individual defendants where it matters most: criminal adjudication and punishment. See, e.g., United States v. Booker, 543 U.S. 220, 264, 125 S.Ct. 738, 160 L.Ed.2d 621 (2005) ("Congress sought to 'provide certainty and fairness in meeting the purposes of sentencing, [while] avoiding unwarranted sentencing disparities . . . [and] maintaining sufficient flexibility to permit individualized sentences when warranted.' " (alterations in original) (quoting 28 U.S.C. § 991(b)(1)(B))). Consistency in sentencing is a worthy aim—one that, it seems to me, is achieved not only through the consistent application of the Sentencing Guidelines, see id. at 253-54, 125 S.Ct. 738, but also through the uniform application of the rules that govern appellate review.
* * *
Clearly, I got more than I bargained for in this case—and, by extension, so did you. What I found, though—and what I hope I've demonstrated—is that our sentencing precedent is a crazy quilt. First, we've been freakishly inconsistent in our characterization of sentencing challenges as "substantive," "procedural," or (tellingly) both. And in important respects, even where we have settled into something that might be called a pattern, we've chosen poorly. Nowhere is that more true, in my view, than in our classification as substantive of what is to me the self-evidently procedural challenge to a district court's consideration of an improper sentencing factor. Second, we've been just as erratic in our pronouncements about what suffices to preserve sentencing-related challenges: One day, a boilerplate objection will suffice, the next day it won't.
Enough is enough. We should convene the full Court to restore some order. And when we do, we should take our cue from the Supreme Court's own sentencing decisions. To start, we should hold that all (as I've called them) "input"-related challenges are allegations of "procedural error" and should be assessed at the outset, before turning to evaluate, as a matter of "substantive reasonableness," the district court's "output"—i.e., the sentence itself. And when determining whether a defendant has properly preserved his sentencing-related challenge—whether substantive or procedural—we should apply the usual rules and ask whether he specifically stated the grounds of his particular objection, in a manner that clearly put the district court on notice of its alleged error.