PUBLISHED

UNITED STATES COURT OF APPEALS FOR THE FOURTH CIRCUIT

No. 18-4609

UNITED STATES OF AMERICA,

Plaintiff - Appellee,

v.

KIRK RUSSELL MARSH,

Defendant - Appellant.

Appeal from the United States District Court for the Eastern District of Virginia, at Alexandria. Anthony John Trenga, District Judge. (1:17-cr-00122-AJT-1)

Argued: September 20, 2019 Decided: December 9, 2019

Before GREGORY, Chief Judge, and THACKER and HARRIS, Circuit Judges.

Motion to dismiss appeal granted by published opinion. Judge Harris wrote the opinion, in which Judge Thacker joined. Chief Judge Gregory wrote an opinion concurring in part, dissenting in part, and dissenting in the judgment.

ARGUED: James R. Theuer, JAMES R. THEUER, PLLC, Norfolk, Virginia, for Appellant. Aidan Taft Grano, OFFICE OF THE UNITED STATES ATTORNEY, Alexandria, Virginia, for Appellee. ON BRIEF: G. Zachary Terwilliger, United States Attorney, Katherine L. Wong, Assistant United States Attorney, OFFICE OF THE UNITED STATES ATTORNEY, Alexandria, Virginia, for Appellee. PAMELA HARRIS, Circuit Judge:

Kirk Russell Marsh pleaded guilty to identity theft and fraud. At sentencing, the

district court failed to advise Marsh of his right to appeal, in what Marsh identifies as a

violation of Rule 32(j) of the Federal Rules of Criminal Procedure. Marsh ultimately did

file an appeal, but not until well after the time limit for that filing had expired under Rule

4(b) of the Federal Rules of Appellate Procedure. The government promptly moved to

dismiss Marsh’s appeal as untimely.

Marsh contends that his late filing should be excused by the district court’s violation

of Rule 32(j). We disagree. Because Rule 4(b)’s deadline is a mandatory claim-processing

rule that must be strictly applied, the district court’s error cannot excuse Marsh’s untimely

filing of his notice of appeal, and equitable doctrines are unavailable to extend the deadline.

Our conclusion, however, does not render Rule 32(j) a nullity, as Marsh suggests; instead,

Marsh may seek to remedy the district court’s error in collateral proceedings. Accordingly,

we grant the government’s motion to dismiss Marsh’s appeal.

I.

Marsh was indicted on three charges: bank fraud, in violation of

;

aggravated identity theft, in violation of 18 U.S.C. § 1028A; and wire fraud, in violation

of

. The government alleged that Marsh used his position as a loan officer

at two federally insured banks to approve fraudulent loans to companies in which he had a

personal stake, creating false documents and bank records in the process.

2 Marsh and the government entered into a written plea agreement, under which

Marsh pleaded guilty to all three counts against him. The agreement also included an

appellate waiver, in which Marsh waived his “right to appeal the conviction and any

sentence within the statutory maximum . . . on any ground whatsoever other than an

ineffective assistance of counsel claim that is cognizable on direct appeal.” J.A. 13–14.

At his plea colloquy, Marsh acknowledged that he understood the appellate waiver in his

plea agreement and that he had spoken with his lawyer about whether he should waive his

right to appeal.

Marsh was sentenced on November 17, 2017. Following argument and rulings on

Marsh’s objections, the district court adopted the Probation Office’s Sentencing Guidelines

calculation, which led to a range of 78 to 97 months on the two fraud charges followed by

a 24-month consecutive sentence for identity theft. The district court then imposed a

below-Guidelines sentence of 54 months for the two fraud charges and a consecutive 24-

month sentence on the identity theft charge, for a total of 78 months imprisonment, to be

followed by a three-year term of supervised release. Critically, at no point during

sentencing did the district court advise Marsh of his remaining rights to appeal, as required

by Rule 32(j). See Fed. R. Crim. P. 32(j)(1)(B) (“After sentencing – regardless of the

defendant’s plea – the court must advise the defendant of any right to appeal the

sentence.”).

The court entered judgment the same day it sentenced Marsh, on November 17,

2017. Under Rule 4(b) of the Federal Rules of Appellate Procedure, a defendant’s notice

of appeal in a criminal case generally must be filed within 14 days of the entry of judgment.

3 See Fed. R. App. P. 4(b)(1)(A)(i). It was not until August 15, 2018, however – 283 days

after the entry of judgment – that Marsh noted his pro se appeal by submitting it through

the mail system of the Bureau of Prisons. The government promptly filed a motion to

dismiss Marsh’s appeal as untimely, which Marsh opposed. This court then issued an order

for supplemental briefing on the issue of timeliness, directing the parties to address whether

the district court’s non-compliance with Rule 32(j) or equitable tolling excused Marsh’s

late filing of his notice of appeal.

II.

A.

At the outset, the government contends that the premise of this case is mistaken:

The district court’s failure to advise Marsh of his right to appeal, according to the

government, was not a Rule 32(j) error at all. Under Rule 32(j)(1)(B), when a defendant

pleads guilty, the district court, after imposing sentence, “must advise the defendant of any

right to appeal the sentence.” Fed. R. Crim. P. 32(j)(1)(B). But here, the government

argues, no discussion of Marsh’s appellate rights was required, because Marsh knowingly

and voluntarily waived the right to appeal in his plea agreement. We disagree.

The first problem for the government’s argument is that Marsh’s plea agreement

expressly preserves his right to appeal his sentence based on “an ineffective assistance of

counsel claim.” J.A. 13–14. Because the text of Rule 32 requires the district court to

“advise the defendant of any right to appeal the sentence,” Fed. R. Crim. P. 32(j)(1)(B)

(emphasis added), the district court was required to advise Marsh of his right, under the

4 plain terms of his plea agreement, to an appeal for ineffective assistance. As the Supreme

Court has cautioned, district courts “must be meticulous and precise in following each of

the requirements of Rule 32 in every case,” see Peguero v. United States,

(1999) (addressing failure to advise of appeal rights), and that includes advising a

defendant of even a narrow right to appeal.

Nor was the scope of the district court’s error limited to the appeal rights expressly

reserved in the appeal waiver. As we have held, a defendant who waives the right to appeal

nevertheless “retains the right to obtain appellate review of his sentence on certain limited

grounds,” even if those grounds are not specified in the plea agreement. United States v.

Attar,

(4th Cir. 1994). No appeal waiver, for instance, can bar a

defendant’s right to challenge his sentence as outside a statutorily prescribed maximum,

“or based on a constitutionally impermissible factor such as race.” United States v. Marin,

(4th Cir. 1992); see also Attar, 38 F.3d at 732–33 (plea waiver does not

bar appellate review of post-plea violation of right to counsel). And, of course, Marsh

retained the right to argue on appeal that he did not enter into his appeal waiver knowingly

and voluntarily, rendering that waiver invalid. See United States v. Johnson,

(4th Cir. 2005). In short, “[a]n appeal waiver does not always preclude an appeal,”

and the district court was required by Rule 32(j) to inform Marsh of his remaining,

though strictly limited, right to appeal.

We think the government’s contrary argument is inconsistent not only with Rule

32(j)’s text, but also with the importance the Supreme Court has attached to “[t]he

requirement that the district court inform a defendant of his right to appeal.” Peguero, 526

5 U.S. at 26. It is not enough, the Court has explained, that the defendant is somehow made

aware of his right to appeal, though that of course is critical. What is equally vital under

Rule 32(j) is that advice about appeal rights “comes from the court itself.” Id. at 27

(emphasis added). Relying on defense counsel is not a substitute, in part because there

may be practical obstacles to communication once the defendant is taken into custody or if

the post-sentencing “relationship between the defendant and the attorney [is] strained.” Id.

at 26. But also, as the Court emphasized in Peguero, requiring that the district court itself

advise a defendant of “any right to appeal” disabuses the defendant of any concern he

otherwise might have that the judge – “who may later rule upon a motion to modify or

reduce the sentence” – will see the filing of an appeal as an “affront.” Id. at 26–27.

Rule 32(j), in other words, serves “important functions” that include but go beyond

ensuring a defendant’s knowledge of “any right to appeal.” Id. at 26. Because the district

court did not advise Marsh that he retained the right to appeal on certain limited grounds,

it failed to comply with Rule 32(j).

B.

That brings us to the question on which we ordered briefing: Whether a district

court’s error in failing to inform a defendant at sentencing of a right to appeal can excuse

a defendant’s late filing of a notice of appeal under Rule 4(b). We conclude that it cannot,

and therefore we must dismiss Marsh’s appeal as untimely.

1.

We begin with the nature of the time limit prescribed by Federal Rule of Appellate

Procedure 4(b). Rule 4(b) requires a criminal defendant in Marsh’s position to file a notice

6 of appeal “in the district court within 14 days after . . . the entry of either the judgment or

the order being appealed.” Fed. R. App. P. 4(b)(1)(A). “Upon a finding of excusable

neglect or good cause, the district court may” extend that deadline, but by no more than 30

days. Fed. R. App. 4(b)(4). There was no extension in this case, and in any event, there is

no question that Marsh’s notice of appeal, filed 283 days after the date of judgment in his

case, was well outside any possible Rule 4(b) deadline.

As the parties agree, Rule 4(b)’s deadline is not jurisdictional. Instead, Rule 4(b) is

a nonjurisdictional claim-processing rule. See United States v. Hyman,

(4th Cir. 2018) (“[T]he late filing of a notice of appeal does not deprive the Court of subject

matter jurisdiction[.]”). That means that Rule 4(b)’s time limit, unlike a jurisdictional rule,

may be waived or forfeited if it is not timely raised. See id.; see also Hamer v.

Neighborhood Hous. Servs. of Chicago,

, 17–18 (2017) (describing distinction

between jurisdictional rules and nonjurisdictional claim-processing rules). But here, of

course, there has been no waiver; the government promptly moved to dismiss Marsh’s

appeal on timeliness grounds, and Marsh does not contend otherwise.

The fact that Rule 4(b)’s time limit is not jurisdictional, however, does not by itself

mean that a late filing can be excused, even for good cause – like a district court’s Rule

32(j) error. That is because some claim-processing rules are “mandatory,” meaning that

they are “unalterable if properly raised by an opposing party.” Nutraceutical Corp. v.

Lambert,

(2019) (internal quotation marks and citations omitted). And,

critically, it is clear, that Rule 4(b) falls into this “mandatory camp,” id.; we held as much

in Hyman, see 844 F.3d at 498, and the Supreme Court later agreed, see Manrique v. United

7 States,

(2017) (declining to decide whether Rule 4(b) is jurisdictional

but holding that the filing of “a timely notice of appeal . . . is at least a mandatory claim-

processing rule”). It follows that once properly raised, we must “strictly apply” Rule 4(b)’s

deadline, Hyman,

: As the Supreme Court has instructed, claim-processing

rules in the “mandatory camp” are “not susceptible [to] equitable approach[es],” like

equitable tolling, that otherwise might forgive a late filing. Lambert,

.

2.

Marsh does not dispute that Rule 4(b) is a mandatory claim-processing rule, with all

that entails. Nevertheless, he argues, the time limit in Rule 4(b) may be excused in his case

because of the district court’s Rule 32(j) error, or equitably tolled because he was not

advised by the district court of his right to appeal. We cannot agree.

At the outset, we note that the Supreme Court very recently (and only after briefs

were submitted in this case) addressed a very similar question. In Nutraceutical Corp. v.

Lambert, the Court held that a different nonjurisdictional 14-day appeal deadline – this one

for seeking an interlocutory appeal from an order respecting class certification under Rule

23(f) of the Federal Rules of Civil Procedure – is “mandatory” and thus not subject to an

“equitable approach” in the form of equitable tolling. Lambert, 139 S. Ct. at 714–15.

As the Court explained, the key to determining whether a nonjurisdictional claim-

processing rule like Rule 23(f) is “mandatory” so as to “preclude[] equitable tolling” is the

text, and, specifically, whether “the text of the rule leaves room for such flexibility.”

. That a rule is “phrased in an unqualified manner,” the Court clarified, is not enough

to show that it is mandatory.

. But what did “speak directly to the issue of Rule

8 23(f)’s flexibility,” in the Court’s view, was that a different rule, Rule 26(b) of the Federal

Rules of Appellate Procedure, expressly excepted the deadline for “the precise type of

filing” found in Rule 23(f) from a general authorization of extensions for good cause.

Because Lambert’s reasoning on this point is so critical here, we explain it in some

detail. Federal Rule of Appellate Procedure 26 governs deadlines for filings in the courts

of appeals. See Fed. R. App. P. 26 (“Computing and Extending Time”). Under Rule 26(b),

courts are given broad discretion to “extend the time prescribed by these rules” or by a

court order “[f]or good cause.” Id. at 26(b). That authorization, however, does not extend

to a few specified deadlines: “But the court may not extend the time to file: (1) a notice of

appeal (except as authorized in Rule 4) or a petition for permission to appeal; or (2) [a

notice of appeal or challenge to certain agency actions].” Id. (emphasis added). Because

Rule 23(f)’s deadline for petitions seeking permission to appeal class certification orders

falls within this “express carveout,” the Court concluded, the Rules “express a clear intent

to compel rigorous enforcement of [that deadline], even where good cause for equitable

tolling might otherwise exist.” Lambert,

.

The Court’s reasoning in Lambert controls this case. As noted above, this court

already has concluded that Rule 4(b)’s deadline for filing a criminal appeal is a mandatory

version of a claim-processing rule, Hyman,

, and the defendant does not

argue otherwise. And, crucially, the very same subsection of Rule 26(b)(1) that “single[s]

out” Rule 23(f)’s deadline “for inflexible treatment,” Lambert,

, also

singles out one other deadline: Rule 4’s deadline for the filing of an appeal. See Fed. R.

App. P. 26(b) (“But the court may not extend the time to file [] a notice of appeal (except

9 as authorized in Rule 4) or a petition for permission to appeal . . . .”) (emphasis added).

And though Rule 4(b) itself allows for one potential extension, under no circumstances

may that exception “exceed 30 days.” Fed. R. App. P. 4(b)(4). The upshot, then, is a

maximum period of 44 days after judgment in which to file a criminal appeal under Rule

4(b). And for precisely the same reasons given by the Court in Lambert, Rule 26(b)(1)’s

“express carveout” of the Rule 4 time limit reflects an intent to make that deadline

mandatory, and to preclude equitable approaches that otherwise might excuse a late filing.

Lambert,

.

As in Lambert, that conclusion is reinforced by the other governing appellate rules.

Rule 3 states that appeals “may be taken only by filing a notice of appeal . . . within the

time allowed by Rule 4.” Fed. R. App. P. 3(a)(1) (emphasis added). And as the Court

indicated in Lambert, although Rule 2 generally permits a court of appeals to “‘suspend

any provision of these rules in a particular case’” for good cause, it does so with a

“conspicuous caveat,” specifying that this authorization is available “‘except as otherwise

provided in Rule 26(b).’” Lambert,

(quoting Fed. R. App. P. 2). And

that brings us, like the Lambert Court, back to the plain text of Rule 26(b) – which, as

discussed above, expressly prohibits extensions of the notice of appeal deadline where no

source of authority for the extension appears in the text of Rule 4 itself, just as it prohibits

extensions of the Rule 23(f) deadline for petitions seeking permission to appeal. See

Lambert,

. Because the courts “may not disregard” a clear intent to make

a claim-processing rule mandatory, we must conclude that Marsh’s late filing under Rule

10 4(b) cannot be excused nor his deadline extended through “equitable approach[es],” like

the doctrine of equitable tolling, under Lambert.

.

We recognize, of course, the gravity of the stakes when what is at issue is a criminal

appeal. But nothing in the text of Rule 26(b) or the Court’s reasoning in Lambert allows

us to distinguish for these purposes between a civil appeal deadline under Rule 23(f) and a

criminal appeal deadline under Rule 4(b). On the contrary: Rule 26(b) groups the two

together in a single subsection, see Fed. R. App. P. Rule 26(b)(1), and Lambert relies in

part on United States v. Robinson,

(1960), see Lambert,

, a

criminal case in which the Court held, on the basis of an extension-of-time provision

similar to Rule 26(b)’s, that an untimely notice of appeal from a criminal judgment requires

dismissal “regardless of [the] excuse,” Robinson, 361 U.S. at 226–27. When the “rules

invoked” – whether criminal or civil – “show a clear intent to preclude tolling, courts are

without authority to make exceptions” – even in cases where a litigant may have been

“reasonably mistaken, or otherwise deserving,” Lambert,

.

3.

Perhaps in response to the Supreme Court’s recent decision in Lambert, Marsh now

focuses not on equitable tolling, but on a different equitable doctrine. As the Court

acknowledged in Lambert, it has left open the question of whether even a mandatory claim-

processing rule, not subject to equitable tolling, nevertheless might give way in certain

unique circumstances, see Lambert,

, where “the cause of the failure to

meet [a] Rule’s deadline [is] an erroneous ruling or assurance by the District Court itself,”

11 Carlisle v. United States,

(1996). 1 According to Marsh, this “unique

circumstances” exception applies in his case, because of the district court’s failure to advise

him of his right to appeal. We cannot agree.

Marsh’s specific objection, as described above, is to the district court’s

noncompliance with Rule 32(j)(1)(B) at sentencing. And as we have explained, Marsh is

correct that the district court’s omission of information about his remaining appellate rights

constituted a violation of Rule 32(j). But the unique circumstances doctrine is a narrow

and “sharply honed” exception. See Carlisle,

(Ginsburg, J., concurring).

And at a minimum, it requires some affirmative statement by the district court – later

proved incorrect – that a filing will be timely. See

(unique circumstances doctrine

inapplicable in absence of “an erroneous ruling or assurance by the District Court itself”);

at 435–36 (Ginsburg, J., concurring) (describing unique circumstances exception as

“based on a theory similar to estoppel” that applies when a litigant is “misled by [a] trial

court statement or action”) (citation omitted). Indeed, we have rejected application of the

unique circumstances exception on precisely that ground, refusing to excuse an untimely

filing where the litigant could not show “that he relied on an affirmative assurance by the

district court that his [motion] was sufficiently timely[.]” Alston v. MCI Commc’ns Corp.,

(4th Cir. 1996) (emphasis added); see also Panhorst v. United States, 241

1 Lambert also noted a possible exception, even to a mandatory claim-processing rule, where “an insurmountable impediment” makes a timely filing impossible. 139 S. Ct. at 717 n.7 (citing Fed. R. App. P. 26(a)(3), addressing computation of time when “the clerk’s office is inaccessible”). Marsh has not relied on any such exception here.

(4th Cir. 2001) (rejecting unique circumstances exception in part because

“the district court . . . did not provide specific assurance” that filing was timely) (internal

quotation marks and citation omitted).

This “limited” understanding of the unique circumstances exception, see Panhorst,

241 F.3d at 370, is consistent with the cases that gave rise to the doctrine, decided by the

Supreme Court in the early 1960s. First, in Harris Truck Lines, Inc. v. Cherry Meat

Packers, Inc.,

(1962) (per curiam), the Supreme Court excused an otherwise

untimely appeal where a litigant had received from the district court, and then relied on, a

14-day extension of the time to appeal that subsequently was deemed invalid by the court

of appeals. In light of the “unique circumstances” presented – the “obvious great hardship”

to a party who relies to his detriment on an incorrect ruling by a district court – the Supreme

Court ruled that the litigant was entitled to file within the district court’s erroneous

extension of the time limit for appeal.

. Two years later, in Thompson v. INS,

(1964) (per curiam), the Court again applied the unique circumstances

exception, where a litigant relied on a trial judge’s “explicit statement” that his motion for

a new trial was filed “in ample time,” so that it would postpone the deadline for filing an

appeal,

. The district court proved to be incorrect, but by then, the appellant

already had filed an appeal “within the assumedly new deadline” but outside the actual

deadline, and for that reason, the Supreme Court concluded, the late filing must be excused.

.

13 The Supreme Court has not actually applied the unique circumstances exception

since the time of Harris and Thompson. 2 What it has done in the intervening years is

overrule Harris and Thompson, “to the extent they purport to authorize an exception to a

jurisdictional rule.” Bowles v. Russell,

(2007). And while it has left

open the question of whether some form of the unique circumstances exception might

continue to apply to nonjurisdictional but mandatory claim-processing rules like Rule 4(b),

see Carlisle,

; Lambert, 139 S. Ct. at 717 n.7 (Court had “no occasion to

address” whether motion might be deemed timely had district court “misled [party] about

the appeal filing deadline”), it also has described that doctrine in exceptionally narrow

terms, tailored to the precise factual circumstances presented by Harris and Thompson.

The unique circumstances doctrine, the Court has explained, can apply “only where a party

has performed an act which, if properly done, would postpone the deadline for filing his

appeal and has received specific assurance by a judicial officer that this act has been

properly done.” Osterneck v. Ernst & Whinney,

(1989); see also

Panhorst, 241 F.3d at 372–73 (applying Osterneck formulation and rejecting application

of unique circumstances exception).

Whatever the precise status and scope of the unique circumstances exception, we

do not think it can be stretched to accommodate this case. The district court’s Rule 32(j)

2 In 1964, the Court applied the exception for a third and final time, in a one-sentence per curiam opinion, where a district court erroneously had issued an extension of the ten- day filing deadline for a Rule 59(b) motion. See Wolfsohn v. Hankin,

(1964) (per curiam).

14 error meant that Marsh was not informed by the court of his remaining appellate rights.

But notwithstanding this serious omission, the district court did not convey incorrect

information, through a ruling or other assurance at the sentencing hearing to which Marsh

objects, that caused Marsh to believe that his appeal was timely or that the deadline for

appeal had been extended. Cf. Carlisle,

(Ginsburg, J., concurring)

(describing doctrine as applying where district court “convey[s] . . . incorrect”

information). 3 And it seems clear – and Marsh does not argue otherwise – that the still-

narrower formulation of Osterneck, requiring that a party actually take some step that could

permit the district court to extend an appeal deadline and that the district court endorse that

step incorrectly, see

, could not apply here. Accordingly, Marsh’s untimely

appeal cannot be excused under the unique circumstances doctrine.

3 At oral argument, Marsh suggested that he was affirmatively misled, for purposes of the unique circumstances exception, by the district court’s statement at his plea colloquy that he had waived his right to appeal both his convictions and his sentence. But that statement conveyed no information about the timeliness of any potential appeal, the matter at the heart of the unique circumstances doctrine. More fundamentally, the district court’s warning at the plea colloquy, designed to ensure that Marsh fully understood the gravity of his appeal waiver, see Fed. R. Crim. P. 11(b)(1)(N), was not in error. At sentencing, as we have held, the district court was required to advise Marsh of his remaining and limited appeal rights. But at the plea colloquy stage, its job was to impress upon Marsh the magnitude of the much larger category of rights he was waiving. Qualifications and caveats about expressly preserved and unwaivable grounds for appeal at the plea acceptance stage risk confusion for the defendant, and a district court does not err by saving those advisements for sentencing.

15 4.

Finally, Marsh argues that unless we can find a way to excuse the late filing of his

appeal, we effectively nullify the requirements of Rule 32(j). As Marsh sees it, this is a

perfect Catch-22: If a district court fails to advise a defendant of his appeal rights and the

defendant nevertheless files a timely appeal, then the Rule 32(j) error is harmless; if as a

result of the district court’s error the defendant does not file a timely appeal, then we must

dismiss the appeal instead of reviewing the error. Absent direct review, in other words,

there can be no meaningful remedy for a Rule 32(j) violation.

We disagree. The Supreme Court has made clear that there is a remedy for a Rule

32(j) violation, and that it comes in the form of vacatur of a sentence on collateral review.

The Court considered precisely this question in Peguero v. United States, after a district

court failed to advise the petitioner of his right to appeal his sentence under the then-

applicable version of Rule 32(j). 526 U.S. at 24–25. The petitioner, who never filed a

direct appeal, later moved under

to set aside his sentence, which would

allow for an appeal after resentencing. See

at 24–25, 29. Because the record firmly

established that the petitioner had “full knowledge of his right to appeal” notwithstanding

the district court’s error, the Court concluded, he could not show prejudice and thus was

not entitled to collateral relief.

. But the Court also made clear that “a court’s

failure to give a defendant advice required by the Federal Rules is a sufficient basis for

collateral relief” when the petitioner is prejudiced by that error because he does not have

independent knowledge of his right to appeal.

at 27–29 (emphasis added).

16 Nothing about Peguero’s prejudice analysis renders the promise of collateral relief

illusory. As the Second Circuit explained in applying Peguero, when a district court fails

to advise a defendant of his right to appeal in violation of Rule 32(j) and the defendant

seeks § 2255 relief, it is the government that “bears the burden of establishing harmlessness

by showing through clear and convincing evidence that the defendant either actually

exercised [the right to appeal], waived this right, or had independent knowledge of this

right.” Soto v. United States,

(2d Cir. 1999). And, importantly, it is

the loss of the right to appeal itself that establishes the prejudice necessary to merit

collateral relief. See

. It is not necessary, in other words, that the petitioner also

“demonstrate that he had meritorious grounds for an appeal” in order to prevail under

Peguero.

at 54–55 (quoting Peguero,

(O’Connor, J., concurring)).

Instead, a district court’s “failure to inform a defendant of his right to appeal, when

combined with the defendant’s lack of independent knowledge of and his actual failure to

exercise this right . . . leads to the loss of an important – albeit non-constitutional – federal

right which, by itself, is sufficient to merit collateral relief.” Soto,

. That

approach, we note, is consistent with the one we have taken to § 2255 ineffective assistance

claims based on counsel’s failure to file an appeal at a defendant’s request; in those

circumstances, we have held, the petitioner establishes prejudice through the loss of the

17 right to appeal itself, and need not also show meritorious grounds for an appeal. See United

States v. Peak,

, 41 (4th Cir. 1993). 4

Remedying Rule 32(j) errors through collateral relief is not a new idea in this circuit.

For years, we have vacated judgments of conviction in collateral proceedings because the

court failed to advise the defendant of the right to appeal as required by the Federal Rules

of Criminal Procedure. See, e.g., Paige v. United States,

, 781–83 (4th Cir.

1971). This case, of course, involves the right to appeal a sentence, not a conviction, but

the legal doctrines are the same. We cannot excuse Marsh’s untimely notice of direct

appeal. But if Marsh seeks habeas relief based on the district court’s Rule 32(j) error, and

the government cannot show that Marsh had independent knowledge of his limited right to

appeal his sentence, then Marsh would be entitled to relief in the form of a vacatur of his

sentence and entry of a new judgment “from which an appeal can be taken.” Peak, 992

F.2d at 42.

4 The government suggested at argument that a petitioner might be procedurally foreclosed from collateral relief by a failure to object to a Rule 32(j) error in an (attempted) direct appeal. That suggestion is hard to square with Peguero, in which the petitioner never filed a direct appeal at all, raising the error in question only on collateral review.

. And this is not surprising; as in the Strickland context, district court adjudication of a § 2255 petition usually will be “the forum best suited to developing the facts necessary to determining” whether a petitioner had independent knowledge of his right to appeal under the Peguero prejudice standard. See Massaro v. United States,

(2003).

18 III.

For the foregoing reasons, we dismiss Marsh’s appeal as untimely.

DISMISSED

19 GREGORY, Chief Judge, concurring in part, dissenting in part, and dissenting in the judgment:

This case is about two rules being violated. The first rule directs that “the court

must advise the defendant of any right to appeal.” Fed. R. Crim. P. 32(j)(1)(B) (emphasis

added). The second rule directs that “a defendant’s notice of appeal must be filed in the

district court within 14 days after . . . the entry of either the judgment or the order being

appealed,” or within 44 days if there is “excusable neglect or good cause.” Fed. R. App.

P. 4(b) (emphasis added). Neither party disputes that these rules were not followed.

Rather, the petitioner here claims that the district court’s failure to follow the first rule is

the reason the petitioner failed to follow the second. That is, the defendant failed to timely

file a notice of appeal because the district court did not inform him of his right to appeal.

The majority today decides, in effect, that appellate judges are required to enforce a

nonjurisdictional deadline even when a trial court caused a defendant to miss the deadline.

More confusingly, though, the majority’s holding does this while correctly noting the

important functions of Criminal Rule 32(j) and our history of correcting Criminal Rule

32(j) errors. Because I doubt that the drafters of the relevant procedural rules at play in

this case intended them to be read in a way that is more forgiving to judges than criminal

defendants—the people least able to protect themselves in our justice system—I must

respectfully dissent.

I.

Before explaining where we part ways, I note that the majority and I share some

common ground. We agree that the Government is mistaken in claiming that there was no

20 Criminal Rule 32(j) error here. See Maj. Op. Section IIA. The majority provides a great

description of the importance of Criminal Rule 32(j) and an explanation of why it was

violated in this case. We also agree—as do both parties in this case—that Appellate Rule

4(b) is not jurisdictional: it is a claim-processing rule. Our precedent has made this much

clear. See e.g., United States v. Urutyan,

(4th Cir. 2009) (“Appellate

Rule 4(b) is a court-prescribed, procedural rule.”). Finally, the majority and I agree that

both Appellate Rule 4(b) and Criminal Rule 32(j) come with some exceptions. For

example, unlike jurisdictional rules, Appellate Rule 4(b) can be waived or forfeited if not

raised in a timely fashion. Hamer v. Neighborhood Hous. Servs. of Chicago,

, 17 (2017). Likewise, Criminal Rule 32(j) is not a per se rule—if the government can

show that a defendant has independent knowledge of the right to appeal, then appellate

courts need not vacate a sentence because a judge failed to advise the defendant of this

right. Peguero v. United States,

(1999).

What separates the majority and I, then, is how we interpret the interplay between

these two important rules. Despite noting that Appellate Rule 4(b) is a claim-processing

rule, the majority holds that a petitioner cannot seek a remedy for a Criminal Rule 32(j)

violation on direct appeal. “Because Rule 4(b)’s deadline is a mandatory claim-processing

rule that must be strictly applied,” the majority writes, “the district court’s error cannot

excuse Marsh’s untimely filing of his notice of appeal.” Maj. Op. at 2. I cannot agree.

The way I see it, “[t]he procedural rules adopted by the Court for the orderly transaction of

its business are not jurisdictional and can be relaxed by the Court in the exercise of its

discretion when the ends of justice so require.” Schacht v. United States,

, 64

21 (1970). For this reason, we are not obliged to enforce Appellate Rule 4(b) when the

defendant was not informed about his right to appeal. Instead, we invoke Appellate Rule

4(b) only after the defendant was made aware of his right to appeal. As this case reaches

us, there is an insufficient record to validate Marsh’s claim that the trial court’s error was

the reason he failed to file a timely notice of appeal. Thus, I cannot subscribe to the

majority’s sweeping refusal to apply equitable doctrines when the result could be—

contrary to the ends of justice—that the defendant would bear the burden of a judge’s

mistake. Rather than dismissing Marsh’s appeal outright, I would remand this case to the

district court for an evidentiary hearing to determine whether the facts of this case call for

equitable doctrines to be applied. I consequently dissent.

II.

Following the majority, let’s begin with “the nature” of the time limit prescribed by

Appellate Rule 4(b). The majority starts by noting that Appellate Rule 4(b) is a

nonjurisdictional claim-processing rule. This typically means that it “does not set forth ‘an

inflexible rule requiring dismissal whenever’ its ‘clock has run.’” Holland v. Florida,

(2010) (citing Day v. McDonough,

(2006)). Unlike

jurisdictional rules, courts may apply equitable exceptions to soften time limitations.

Bowles v. Russell,

(2007). As the Supreme Court has put it, “[t]ime

requirements . . . are customarily subject to equitable tolling.” Irwin v. Dep’t of Veterans

Affairs,

(1990). Indeed, the Court has gone so far as declaring that

nonjurisdictional rules are “normally subject to a ‘rebuttable presumption’ in favor ‘of

22 equitable tolling.’” Holland, 560 U.S. at 645–46 (quoting Irwin,

).

Accordingly, the nature of a claim-processing rule does not suggest that it is “utterly

exceptionless.” Carlisle v. United States,

(1996) (Ginsburg, J.,

concurring) (“But like limitation periods generally, the 29(c)/45(b) constraint is not utterly

exceptionless.”). As court-promulgated, nonjurisdictional rules are subject to equitable

limitations.

Of course, this distinction between jurisdictional and nonjurisdictional rules is not

merely terminological. “Jurisdictional treatment of statutory time limits makes good

sense.” Bowles,

. This is because “[w]ithin constitutional bounds, Congress

decides what cases the federal courts have jurisdiction to consider.”

And since

“Congress decides whether the federal courts can hear cases at all, it can also determine

when, and under what conditions, federal courts can hear them.”

. This is

why jurisdictional time limits cannot be altered by courts. As the Supreme Court put it, if

jurisdictional rules are “too strict and technical, and likely to produce inconvenience or

injustice, it is for Congress to provide a remedy by altering the existing laws; not for the

court.” United States v. Curry,

(1848). The function of an appellate court

is not to correct what Congress has set as a jurisdictional limit of courts.

However, the Court has “treated the rule-based time limit for criminal cases

differently.” Bowles,

; see also Urutyan,

(comparing the

nonjurisdictional Appellate Rule 4(b) with Appellate Rule 4(a), which is backstopped by a

federal statutory deadline). This is because these rules, including Appellate Rule 4(b), were

“not enacted by Congress but [were] promulgated by [the] Court under the authority of

23 Congress to prescribe rules concerning the time limitations for taking appeals and applying

for certiorari in criminal cases.” Schacht,

. The nature of the time limit

prescribed by Appellate Rule 4(b), then, is like other claim-processing rules—it is

essentially designed to assist the court in its orderly transition of its business. See United

States v. Kwai Fun Wong,

, (2015) (“Time and again, we have

described filing deadlines as quintessential claim-processing rules, which seek to promote

the orderly progress of litigation, but do not deprive a court of authority to hear a case.”)

(internal quotation marks and citations omitted).

The majority acknowledges that Appellate Rule 4(b) is nonjurisdictional, but it

concludes that something about its nature precludes tolling. Why? Presumably, because

Appellate Rule 4(b) is a “mandatory” claim-processing rule. Since Appellate Rule 4(b) is

mandatory, the majority reasons, it is unalterable if “properly raised” by the Government.

By signifying a “claim processing rule” as mandatory, the majority suggests, a court

forecloses equitable approaches that may forgive late filings. Maj. Op. at 7-8.

The majority’s reasoning here is too quick. For starters, the majority does not tell

us why a rule is “properly raised” against a defendant who was previously unaware that he

had a right to appeal a judge’s sentence because the same judge failed to inform him of this

right. Moreover, the “mandatory” qualifier given to claim-processing rules is not meant to

categorically bar equitable approaches. For as long as the Supreme Court has referred to

claim-processing rules as mandatory, they have avoided claiming that mandatory claim-

processing rules foreclose equitable exceptions. Compare Manrique v. United States,

(declaring that “a timely notice of appeal . . . is at least a mandatory

24 claim-processing rule”), with Kontrick v. Ryan,

(2004) (“Whether the

Rules, despite their strict limitations, could be softened on equitable grounds is therefore a

question we do not reach.”). Indeed, just two terms ago the Court emphasized this point in

the same opinion. Compare Hamer, 138 S. Ct. at 17 (2017) (“Mandatory claim-processing

rules are less stern. If properly invoked, mandatory claim-processing rules must be

enforced, but they may be waived or forfeited.”), with id. at 18 n.3 (“We have reserved

whether mandatory claim-processing rules may be subject to equitable exceptions.”).

Importantly, a nonjurisdictional deadline does not preclude a court from hearing a case

“even when the time limit is important (most are) and even when it is framed in mandatory

terms (again, most are); indeed, that is so however emphatic[ally] expressed those terms

may be.” Kwai Fun Wong,

(internal quotation marks and citations

omitted).

Does it follow that nonjurisdictional deadlines are malleable in every respect? No.

“[E]quitable tolling is a discretionary doctrine that turns on the facts and circumstances of

a particular case and, therefore, does not lend itself to bright-line rules.” Rouse v. Lee,

(4th Cir. 2003) (en banc) (internal quotation marks and citations omitted);

see also Holmberg v. Armbrecht,

(1946) (“Equity eschews mechanical

rules; it depends on flexibility.”). This is why “circuit precedent requires a court to

consider the facts and circumstances of a particular case” when determining whether to

exercise the power to toll deadlines. Rouse,

. The upshot of all of this is

that, contrary to the majority’s portrayal, there is nothing about “the nature” of a mandatory

claim-processing rule that definitively precludes equitable tolling. Far from it. “Equitable

25 tolling is appropriate in those rare instances where—due to circumstances external to the

party’s own conduct—it would be unconscionable to enforce the limitation period against

the party and gross injustice would result.” Edmonson v. Eagle Nat’l Bank,

(4th Cir. 2019) (internal citations and quotations omitted). Therefore, I would remand

this case to determine whether the facts here present one of those rare instances where

injustice would result.

III.

The majority’s reliance on Nutraceutical Corp. v. Lambert,

(2019),

is also misplaced. In Lambert, the Court held that a deadline to file an interlocutory appeal

could not be tolled even when the party seeking the interlocutory appeal orally informed

the district court of his intention to seek reconsideration within the timeframe set out by

Civil Rule 23(f).

. In doing so, the Court explained that Civil Rule 23(f) was a

mandatory claim-processing rule that was not amenable to “the equitable approach that the

Court of Appeals applied.”

. But nowhere did the Court suggest that mandatory

claim-processing rules trump all other requirements invoked by the Federal Rules. To the

contrary, the Court went out of its way to make clear that the inquiry to determine whether

tolling is available must be guided by examining the “pertinent rule or rules invoked.”

. Taken collectively, the Court concluded that the relevant rules “express[ed] a clear

intent to compel rigorous enforcement of Rule 23(f)’s deadline, even where good cause for

equitable tolling might otherwise exist.”

.

26 When examining the relevant rules in this case, the majority’s analysis leaves out

the most important rule: Criminal Rule 32(j). So let’s include it now. A plain reading of

the Rules suggest that Criminal Rule 32(j) is a competing mandatory rule: “After

sentencing—regardless of the defendant’s plea—the court must advise the defendant of

any right to appeal the sentence.” Fed. R. Crim. P. 32(j)(1)(B) (emphasis added). And I

see nothing in the text of the Rules that express a clear intent to preclude tolling when

Criminal Rule 32(j) is not followed. Indeed, at times it appears to suggest the opposite.

Federal Rule of Criminal Procedure 2, for example, states that “[t]hese rules are to be

interpreted to provide for the just determination of every criminal proceeding, to secure

simplicity in procedure and fairness in administration, and to eliminate unjustifiable

expense and delay.” Fed. R. Crim. P. 2. The talk of “just,” “fairness,” and “unjustifiable”

speaks in an equitable tone. Perhaps this is why, as stated above, it has long been the

Supreme Court’s position that “[t]he procedural rules adopted by the Court for the orderly

transaction of its business are not jurisdictional and can be relaxed by the Court in the

exercise of its discretion when the ends of justice so require.” Schacht,

.

The conclusion my dear colleagues reach suggests that the Rules “express a clear intent”

to compel rigorous enforcement of Appellate Rule 4(b) when Criminal Rule 32(j) is

violated. But the Rules contain no language that calls for so harsh an interpretation.

Putting the Criminal Rule 32(j) violation to the side for the moment, there are other

pertinent differences between Appellate Rule 4(b) and Civil Rule 23(f). Contra Maj. Op.

at 11 (“But nothing in the text of Rule 26(b) or the Court’s reasoning in Lambert allows us

to distinguish for these purposes between a civil appeal deadline under Rule 23(f) and a

27 criminal appeal deadline under Rule 4(b).”). First, there are differences between the

emphatic character in which the Federal Rules speak about notices of appeal (which is the

subject here) and petitions for permission to appeal (which was the subject in Lambert). It

is true that Appellate Rule 26(b) mentions both. 1 Yet it gives a different text, and it takes

a different tone, when discussing a petition to appeal than it does when discussing a notice

of appeal. “Appellate Rule 26(b), which generally authorizes extensions of time, in turn

includes this express carveout: A court of appeals ‘may not extend the time to file . . . a

petition for permission to appeal.’” Lambert, 139 S. Ct. at 715. This is without exception.

By contrast, when the Rule discusses the timeframe to submit a notice of appeal it speaks

less authoritatively: “the court may not extend the time to file . . . a notice of appeal (except

as authorized in Rule 4).” Fed. R. App. P. 26(b)(1) (emphasis added). Having

exceptions—and thus speaking less emphatically—is contrary to the inflexible treatment

that Civil Rule 23(f) is singled out for. See Lambert, 139 S. Ct. at 715. Indeed, because

1 It is also true, as the majority points out, Maj. Op. at 10, that Appellate Rule 2 generally permits a court to suspend any provision of the rules for “good cause,” except as provided in Appellate Rule 26(b). And while there does not seem to be a clear test for good cause, I submit that a judge’s violation of a compulsory rule that causes a defendant to lose a statutory right is a more compelling reason to suspend a rule than the more lenient good cause standard. See Thompson v. E.I. DuPont de Nemours & Co.,

(4th Cir. 1996) (noting that the good cause standard is more lenient than excusable neglect); see also Britt v. Whitmire,

(5th Cir. 1992) (noting “the more lenient ‘good cause’ standard does not apply at all” when addressing an untimely notice of appeal). Thus, it would be consistent to compel rigorous enforcement of mandatory claim- processing rules when mere good cause exists (as in Lambert), but nonetheless suspend them “in those rare instances where—due to circumstances external to the party’s own conduct—it would be unconscionable to enforce the limitation period against the party and gross injustice would result.” Edmonson,

(internal quotation marks and citations omitted).

28 Appellate Rule 26(b) defers to Appellate Rule 4 to provide the boundaries for notices of

appeal, it follows that the governing rule is Appellate Rule 4 and not (the more inflexible)

Appellate Rule 26.

This point is further reinforced when we compare the advisory committee’s notes

for the two rules. See Mississippi Pub. Corp. v. Murphree,

(1946)

(noting that “in ascertaining” the meaning of the Federal Rules, the notes by the advisory

committee is “of some weight”). The advisory notes to Appellate Rule 4 include some

version of the noun “toll” no fewer than twelve times. Fed. R. App. P. 4 advisory

committee’s note. Whereas no form of the noun is mentioned in the advisory notes to Civil

Rule 23. Fed. R. Civ. P. 23 advisory committee’s note. This suggests that the drafters of

the Rules intended to leave more flexibility when enforcing Appellate Rule 4(b) than when

enforcing Civil Rule 23(f).

And for good reason. As Lambert itself reminds us, “interlocutory appeal is an

exception to the general rule that appellate review must await final judgment—which is

fully consistent with a conclusion that Rule 23(f)’s time limit is purposely unforgiving.”

Lambert, 139 S. Ct. at 716. Indeed, “[t]he purpose of Rule 23(f), in part, is to ensure that

interlocutory appeals of class certification decisions are heard and decided in a timely

manner, so as not to disrupt the proceedings at the district court level.” Gutierrez v.

Johnson & Johnson,

(3d Cir. 2008) (citing Fed. R. Civ. P. 23(f) advisory

committee’s note). Appeals to final judgments, however, serve a different purpose. And,

unlike an interlocutory appeal, “[p]ersons convicted in federal district courts have a direct

right to appeal.” United States v. Peak,

(4th Cir. 1993) (citing Coppedge

29 v. United States,

(1962)) (emphasis added). Thus, as one of our sister circuits

recently observed when interpreting Lambert, treating interlocutory appeals different from

appeals as of right “makes sense: a litigant who loses the opportunity to appeal a final

judgment forever loses the ability to appeal, but a litigant who loses the opportunity to file

an interlocutory appeal has another chance later.” Groves v. United States,

(7th Cir. 2019). Since interlocutory appeals are exceptional, there is no surprise that

“their limitations are ‘purposefully unforgiving.’”

(citing Lambert, 139 S. Ct. at 716).

In short, Lambert does not foreclose equitable tolling in this case. That case was

about “the rigidity of filing deadlines for interlocutory appeals even when they appear in

claim-processing rules.” Id. at 322 (citing Lambert, 139 S. Ct. at 714-15). The reasoning

there does not extend to criminal appeals as of right, where “the gravity of the stakes” are

much higher. And even if it does, there is nothing in Lambert that suggests that tolling is

categorically precluded when a judge is the reason the defendant missed the deadline. The

majority’s conclusion is inconsistent with the text of the relevant rules involved in this case

and “with the importance the Supreme Court has attached to ‘[t]he requirement that the

district court inform a defendant of his right to appeal.’” Maj. Op. at 5 (citing Peguero,

).

IV.

The majority also considers, and dismisses, Marsh’s attempt to rescue his appeal

under the “unique circumstances” exception. As the majority acknowledges, the Supreme

Court has left open the possibility of mandatory claim-processing rules being overridden

30 in certain circumstances. In Lambert, the Court did not address this issue because “nothing

the District Court did misled [Petitioner] about the appeal filing deadline.” Lambert, 139

S. Ct. at 717 n.7; see also Carlisle,

(Ginsburg, J., concurring) (noting that

there is a “sharply honed” exception that “covers cases in which the trial judge has misled

a party who could have—and probably would have—taken timely action had the trial judge

conveyed correct, rather than incorrect, information”). Since the petitioner’s delay was not

attributable to a mistake by the district court, there was no reason for the Court to consider

the unique circumstances exception.

To be sure, the Supreme Court has not precisely outlined the contours of the unique

circumstance exception. But appellate courts have interpreted the exception widely. See,

e.g., Hollins v. Dep’t of Corr.,

, 1326–28 (11th Cir. 1999) (applying the

unique circumstances doctrine when counsel relied on PACER system’s version of a

docket sheet, which failed to show the entry of final judgment); Sidebottom v. Delo,

(8th Cir. 1995) (applying the unique circumstance doctrine when petitioner

“was ‘lulled into inactivity’ by the district court’s representation,” in which “[t]he clear

inference” was that a notice of appeal could be filed later); Virgin Islands v. Gereau,

(3d Cir. 1979) (applying unique circumstances doctrine “where a defendant or

his counsel [was] affirmatively misl[e]d by some government authority as to the date by

which a Rule 35 motion must be filed”). But see Merrell-Nat’l Labs., Inc. v. Zenith Labs.,

Inc.,

, 790 (3d Cir. 1978) (recognizing the unique circumstance exception but

refusing to apply it primarily because the case involved “only the lesser hardship of losing

the right to appeal an interlocutory order”). The wisdom behind the doctrine is easy to see:

31 “[t]he unique circumstances exception evolved out of concern with the fairness of a

dismissal when the district court contributed to the party’s failure to take the steps

necessary to perfect the appeal.” Kraus v. Consol. Rail Corp.,

, 1365 (3d

Cir. 1990). In this sense, as the majority notes, the doctrine is “based on a theory similar

to estoppel.” Maj. Op. at 12 (citing Carlisle,

(Ginsburg, J., concurring)).

And viewing the doctrine in this manner makes even more sense with respect to claim-

processing rules—a court should be estopped from faulting a defendant for disrupting rules

designed for the orderly transition of a federal court’s business when a federal judge is the

reason for the disruption.

The record here presents a compelling case to apply the exception if, as Marsh

suggests, the district court was the reason Marsh failed to file a timely notice of appeal.

Marsh’s plea agreement waived any right to appeal “the conviction and any sentence within

the statutory maximum . . . on any ground whatsoever other than an ineffective assistance

of counsel claim that is cognizable on direct appeal.” J.A. 13-14 (emphasis added). During

the Rule 11 plea colloquy, the district court judge asked Marsh if he understood that he had

“waived [his] right of appeal under his plea agreement” and whether he had “spoken with

[his] lawyer about whether or not [he] should waive [his] right of appeal.” J.A. 52. Marsh

answered “Yes” to both questions. J.A. 52. Subsequently, the court again asked Marsh if

he understood that, by pleading guilty, he gave up his “[right] of appeal with respect to

both the convictions that would result from [his] guilty pleas, as well as any sentence that

is imposed based on those convictions.” J.A. 52 (emphasis added). Marsh again answered

“Yes.” Finally, the district court, after describing further rights Marsh would be giving up,

32 asked Marsh if he understood that he was waiving “all of these constitutional rights . . .

[a]nd all that would be left for the Court to do would be to decide what sentence to impose.”

J.A. 57 (emphasis added). Marsh yet again answered in the affirmative. J.A. 57. The

district court went on to explain to Marsh that the court was not bound by the Sentencing

Guidelines, the recommendation from the Government, or other conditions. J.A. 60-62.

Taken together, these statements would lead a defendant to believe he is being sentenced

entirely at the whim of the district court judge and must accept a penalty even if he believes

it is unlawful. But this is why we have Criminal Rule 32(j): to let the defendant know that

he could appeal a sentence he believes was unlawful. The sentencing judge here failed to

relay that message. And Marsh now claims that the failure to relay that message is the

reason he did not file a timely notice of appeal—that is, Marsh was misled by the

sentencing court’s actions.

The majority disagrees. On its view, the doctrine only applies when a judge makes

affirmative statements “about the timeliness of any potential appeal.” Maj. Op. at 15 n.3.

Since the district court “did not convey incorrect information, through a ruling or other

assurance at the sentencing hearing,” 2 the majority reasons, the unique circumstance

2 The majority does not appear to find anything misleading in the district court’s statements made during the plea colloquy stage. This is because, the majority writes, “[q]ualifications and caveats about expressly preserved and unwaivable grounds for appeal at the plea acceptance stage risk confusion for the defendant, and a district court does not err by saving those advisements for sentencing.” Maj. Op. at 15 n.3. One could quibble over whether a court should sacrifice accuracy in order to not confuse a defendant. But even if a court does not err by saving its advisements for sentencing, a court may err when it tells a defendant all the rights that are given up without later informing the defendant of the single right he retains. I see no substantive difference in the district court’s misleading

33 doctrine is inapplicable. Maj. Op. at 15. According to the majority’s view, the unique

circumstances may apply if a judge—who is not required to give a party any assurance—

tells a party that the appeal deadline is in 11 days when the deadline is actually in 10 days.

But it cannot apply if a judge fails to inform a defendant about his appellate rights

altogether—which the judge is required to do under Criminal Rule 32(j). See Maj. Op.

Section IIA. In other words, if a court tells you the wrong deadline to file an appeal, shame

on us. But if a court—neglecting its duty—fails to tell you about your right to appeal

whatsoever, shame on you. I don’t get it. 3

This distinction is illogical and frustrates the wisdom behind the doctrine. If an

appellate court is estopped from sending a defendant away when a court gives him the

wrong deadline to file an appeal, then an appellate court should be similarly estopped when

a court’s failure to follow a compulsory rule is the cause of the defendant’s failure to

comply with a deadline. I would therefore apply the unique circumstance doctrine in

situations where a court “misled a party who could have—and probably would have—

taken timely action had the trial judge conveyed correct” information about the party’s

right to appeal. See Carlisle,

(Ginsburg, J., concurring). Since Marsh

statements being made at the plea colloquy hearing than the sentencing hearing—both have the capacity to mislead a defendant. 3 It is not entirely clear how the majority would handle a third scenario: where a sentencing judge—acting in error—affirmatively tells a defendant during a sentencing hearing that the defendant has no right to appeal the sentence. Since such a statement would not be about the “timeliness of any potential appeal,” which the majority takes to be “at the heart of the unique circumstances doctrine,” the majority would presumably also consider this scenario to be outside of the doctrine’s reach. 34 alleges that this is the case here, I would remand this case to determine whether the facts

call for the unique circumstance doctrine to be applied.

V.

The majority ends its opinion by rightly noting that this Court has remedied

Criminal Rule 32(j) errors for years. Indeed, for nearly five decades many appellate courts

have emphasized the importance of a trial judge informing a defendant of his right to

appeal. 4 See, e.g., Peguero,

(“Trial judges must be meticulous and precise

in following each of the requirements of Rule 32 in every case.”); Paige v. United States,

(4th Cir. 1971) (noting a Criminal Rule 32 violation “requires a remand

for resentencing and reinstatement of the right to appeal”); United States v. Deans,

(3d Cir. 1971) (“[T]he mandatory time limit for perfecting an appeal does

not begin to run until the defendant is actually notified of his rights.”); United States v.

Benthien,

(1st Cir. 1970) (declaring that reinstating an appeal

whenever a trial court has not complied with Criminal Rule 32(a)(2) “insures that all

defendants will receive the protection the rule was intended to provide” and “serve[s] to

warn district judges of the necessity of strict compliance”). Perhaps in recognition of this

tradition, the majority’s analysis raises a more fundamental question: is it only appropriate

4 Criminal Rule 32 was amended in 2002. The requirement that the court advise the defendant of his right to appeal is currently found in Criminal Rule 32(j)(1)(B). Prior to the changes, however, the requirement could be found in Criminal Rule 32(a)(2). In addition, the earlier version of the Rule did not require the court to inform defendants who pled guilty of their right to appeal—it only required that judges inform defendants who pled not guilty of their right to appeal.

35 for petitioner to address Criminal Rule 32(j) errors through collateral relief —and not direct

appeal? That issue does not seem to appear in the pre-Kontrick cases, where appellate

courts treated notice of appeal deadlines as jurisdictional. Compare Kontrick,

(noting that courts “have been less than meticulous” in describing time prescriptions

as “jurisdictional”), with United States v. Robinson,

(1960) (describing

the prescribed time for taking an appeal as “mandatory and jurisdictional”). This is

understandable. If it is well known that a court lacks jurisdiction to hear an appeal, then it

should come as no surprise that a skilled litigant will forego the process of filing a direct

appeal and instead seek habeas relief—a court cannot provide a remedy when it is not in

its power to do so.

But, again, that is not the case here. There is no doubt that Appellate Rule 4(b) is a

court-prescribed deadline, Urutyan,

(“Appellate Rule 4(b) is a court-

prescribed, procedural rule.”), and this Court possesses the power to grant Marsh the relief

he seeks. See Holland,

(noting that nonjurisdictional deadlines do not “set

forth an inflexible rule requiring dismissal whenever its clock has run”) (internal quotation

marks and citations omitted). Thus, there is no reason he should only be able to seek relief

through a habeas petition if, as Marsh claims, a judge failing to comply with a compulsory

rule is the reason Marsh missed the deadline. It is well known that petitioners seeking

collateral relief have to clear a higher hurdle than on direct appeal. See, e.g., United States

v. Frady,

(1982) (“We reaffirm the well-settled principle that to obtain

collateral relief a prisoner must clear a significantly higher hurdle than would exist on

direct appeal.”). And the Supreme Court has long recognized that “[t]hose whose right to

36 appeal has been frustrated should be treated exactly like any other appellants; they should

not be given an additional hurdle to clear just because their rights were violated at some

earlier stage in the proceedings.” Rodriquez v. United States,

(1969);

see also Frady,

(“a collateral challenge may not do service for an appeal”).

Limiting relief of a Criminal Rule 32(j) error to a habeas petition would place an undue

burden on defendants who are only in that position because of a trial judge’s mistake. And

what makes matters worse is that petitioners may have to climb these hurdles without the

assistance of counsel. See Pennsylvania v. Finley,

, 555–56 (1987) (“Our

cases establish that the right to appointed counsel extends to the first appeal of right, and

no further.”); Hunt v. Nuth,

(4th Cir. 1995) (“[T]his Court has held that

the Constitution does not require counsel for defendants who attack their judgments under

.”). Given the perilous path Marsh—and others—would have to tread

because of a judge’s mishap, I cannot endorse the majority’s position that a petitioner can

only seek redress through a habeas petition. We can—and should—address the violation

here.

I respectfully dissent.

37