Williams v. Martinez

United States Court of Appeals
FOR THE DISTRICT OF COLUMBIA CIRCUIT

Argued September 14, 2009 Decided November 13, 2009

No. 08-5221

CRAIG ALLAN WILLIAMS,
APPELLANT

v.

R. MARTINEZ, WARDEN AND KENNETH L. WAINSTEIN, UNITED
STATES ATTORNEY FOR THE DISTRICT OF COLUMBIA,
APPELLEES

Appeal from the United States District Court
for the District of Columbia
(No. 1:08-cv-00971-ESH)

William Hoffman, appointed by the court, argued the cause
as amicus curiae in support of appellant. With him on the briefs
was David L. Cousineau.

Craig A. Williams, pro se, filed briefs for appellant.

Michael T. Ambrosino, Assistant U.S. Attorney, argued the
cause for appellees. With him on the brief was Roy W. McLeese
III, Assistant U.S. Attorney. R. Craig Lawrence, Assistant U.S.
Attorney, entered an appearance.
 2
Before: ROGERS, TATEL and BROWN, Circuit Judges.

Opinion for the Court filed by Circuit Judge TATEL.

Opinion concurring in the judgment filed by Circuit Judge
BROWN.

TATEL, Circuit Judge: Section 23-110 of the D.C. Code
establishes a procedure for collateral review of convictions in
the D.C. Superior Court and creates exclusive jurisdiction in that
court “unless” the remedy provided by that section is
“inadequate or ineffective.” In this case, we must decide
whether section 23-110 bars a prisoner convicted in Superior
Court from bringing a federal habeas corpus petition alleging
ineffective assistance of appellate counsel. Because the D.C.
Court of Appeals has held that challenges to the effectiveness of
appellate counsel may not be brought pursuant to section 23-
110, but must instead be raised through a motion to recall the
mandate in that court, we hold that section 23-110 does not
deprive federal courts of jurisdiction over habeas petitions
alleging ineffective assistance of appellate counsel.

I.
Although the background of this case is complicated,
involving as it does several proceedings spanning more than
fifteen years, see Williams v. United States, 878 A.2d 477 (D.C.
2005) (en banc); Williams v. United States, 783 A.2d 598 (D.C.
2001), understanding the issue before us requires knowing only
the following.

In 1990, a D.C. Superior Court jury convicted appellant
Craig Allan Williams of first-degree murder. Represented by
new counsel, Williams then appealed. During the pendency of
that appeal, Williams filed a motion for post-conviction relief
 3
pursuant to D.C. Code § 23-110, which provides that a “prisoner
in custody under sentence of the Superior Court claiming the
right to be released upon the ground that [] the sentence was
imposed in violation of the Constitution . . . may move the court
to vacate, set aside, or correct the sentence.” D.C. Code § 23-
110(a).

Consistent with its usual practice, the D.C. Court of
Appeals stayed Williams’s direct appeal pending the Superior
Court’s resolution of his section 23-110 motion. See Shepard v.
United States, 533 A.2d 1278, 1280 (D.C. 1987). Thereafter,
the Superior Court denied Williams’s section 23-110 motion,
and the D.C. Court of Appeals consolidated his appeal of that
denial with his direct appeal. In 1995, the D.C. Court of
Appeals affirmed Williams’s conviction.

Williams then filed a motion in the D.C. Court of Appeals
to recall the mandate affirming his conviction—the procedure
required in the District of Columbia to litigate the issue of
ineffective assistance of appellate counsel. See Watson v.
United States, 536 A.2d 1056, 1060 (D.C. 1987) (en banc). In
that motion, Williams complained that counsel on both his direct
appeal and his section 23-110 motion had rendered ineffective
assistance. The D.C. Court of Appeals summarily denied the
motion to recall the mandate.

Making the same ineffective assistance of appellate counsel
claim, Williams then sought habeas relief in federal court. The
district court dismissed Williams’s habeas petition for lack of
jurisdiction on the ground that section 23-110 provides the
exclusive remedy for collateral challenges to sentences imposed
by the Superior Court. Williams v. Martinez, 559 F. Supp. 2d
56, 57–58 (D.D.C. 2008).
 4
Williams appealed, and we referred his case to the district
court to determine in the first instance whether to issue a
certificate of appealability (COA). See 28 U.S.C. § 2253(c)(1)
(requiring a COA to appeal a final order in a habeas
proceeding); United States v. Mitchell, 216 F.3d 1126, 1130
(D.C. Cir. 2000) (holding that petitioners should seek a COA
from the district court before requesting one from the appeals
court). The district court declined to issue a COA, explaining
that for the reasons given in its opinion dismissing Williams’s
claim for lack of jurisdiction, Williams had failed to make “a
substantial showing of the denial of a constitutional right,” as
required for a COA. 28 U.S.C. § 2253(c)(2). Williams then
filed a request for a COA in this court, and we appointed amicus
curiae to present arguments on his behalf.

Because the district court denied Williams’s petition
without reaching the merits of his constitutional claim, we
review his request for a COA in two steps. We ask first whether
Williams has shown that “jurists of reason would find it
debatable whether the district court was correct” in dismissing
his petition for lack of jurisdiction, and second whether “jurists
of reason would find it debatable whether [his] petition states a
valid claim of the denial of a constitutional right.” Slack v.
McDaniel, 529 U.S. 473, 484 (2000). We take each step in turn.

II.
The answer to the first question—whether the district court
correctly dismissed Williams’s claim for lack of jurisdiction—
turns on the reach of section 23-110. Section 23-110(a)
authorizes a “prisoner in custody under sentence of the Superior
Court” to “move the court to vacate, set aside, or correct the
sentence.” D.C. Code § 23-110(a). Section 23-110(g) provides:
 5
An application for a writ of habeas corpus in behalf of
a prisoner who is authorized to apply for relief by
motion pursuant to this section shall not be entertained
by the Superior Court or by any Federal or State court
if it appears that the applicant has failed to make a
motion for relief under this section or that the Superior
Court has denied him relief, unless it also appears that
the remedy by motion is inadequate or ineffective to
test the legality of his detention.

D.C. Code § 23-110(g).

Williams contends that section 23-110(g) presents no bar to
his ineffective assistance of appellate counsel claim.
Specifically, he argues that because the D.C. Court of Appeals
prohibits prisoners from bringing challenges to the effectiveness
of appellate counsel under section 23-110—they may be raised
only through a motion to recall the mandate—his remedy under
section 23-110 is “inadequate or ineffective.” According to the
government, Williams, by focusing solely on the adequacy of
his remedies under section 23-110, “addresses the wrong
question.” Appellees’ Br. 25. As the government sees it, the
proper inquiry is not whether section 23-110 provides an
adequate remedy to test the legality of Williams’s detention, but
rather whether the “local remedy” taken as a whole does. Id.
Therefore, the government argues, because the D.C. Court of
Appeals provides an adequate local remedy to challenge the
effectiveness of appellate counsel, namely the opportunity to file
a motion asking the court to recall its mandate, section 23-110
bars Williams’s habeas petition.

We agree with Williams. Section 23-110(g)’s plain
language makes clear that it only divests federal courts of
jurisdiction to hear habeas petitions by prisoners who could
have raised viable claims pursuant to section 23-110(a). Recall
 6
that section 23-110(g) provides that a prisoner authorized to
apply for relief under section 23-110(a) may not bring a habeas
petition in federal court “unless it also appears that the remedy
by motion is inadequate or ineffective to test the legality of his
detention.” D.C. Code § 23-110(g). “[R]emedy by motion”
plainly refers to motions filed pursuant to section 23-110(a).
Although the D.C. Court of Appeals allows prisoners to
challenge the effectiveness of appellate counsel through a
motion to recall the mandate, such a motion—filed directly in
the D.C. Court of Appeals—is obviously not a “remedy by
[section 23-110] motion,” which is filed in the D.C. Superior
Court. D.C. Code § 23-110(g). Indeed, the D.C. Court of
Appeals itself has emphasized that a motion to recall the
mandate is an “independent” action separate and apart from a
section 23-110 motion. Wu v. United States, 798 A.2d 1083,
1091 (D.C. 2002). Thus, because the Superior Court lacks
authority to entertain a section 23-110 motion challenging the
effectiveness of appellate counsel, that section is, by definition,
inadequate to test the legality of Williams’s detention.
Accordingly, section 23-110 does not bar Williams’s habeas
petition.

Our decision in Streater v. Jackson, 691 F.2d 1026 (D.C.
Cir. 1982), supports this view. There, the D.C. Court of
Appeals dismissed Streater’s section 23-110 motion alleging
ineffective assistance of appellate counsel. Streater then
petitioned for habeas relief in the district court, which dismissed
his petition for failing to exhaust his local remedies. On appeal,
we recognized that Streater found himself in a bind: on the one
hand, section 23-110 was unavailable to him because the
Superior Court lacks authority to review appellate proceedings;
on the other hand, the D.C. Court of Appeals had at that time
failed to clarify what avenue of local relief remained open to
him. Accordingly, we instructed the district court to hold
Streater’s habeas petition in abeyance pending his application to
 7
the D.C. Court of Appeals to recall its mandate—the very
procedure subsequently adopted by the D.C. Court of Appeals
as the appropriate vehicle for mounting a challenge to the
effectiveness of appellate counsel. Id. at 1028; see Watson, 536
A.2d at 1060–61 (requiring that ineffective assistance of
appellate counsel claims be litigated through a motion to recall
the mandate). Of significance to the issue before us, we
clarified that after “a cogent ruling from the D.C. Court of
Appeals concerning local relief, if any, for Streater, the District
Court will be in a position to rule intelligently on his federal
petition for habeas corpus.” Streater, 691 F.2d at 1028. In
other words, Streater anticipated precisely the situation we
confront here—a federal habeas petition asserting ineffective
assistance of appellate counsel after the prisoner moved to recall
the mandate in the D.C. Court of Appeals—and seemed to have
assumed that the district court would have jurisdiction to
entertain that petition.

Blair-Bey v. Quick, 151 F.3d 1036 (D.C. Cir. 1998), further
reinforces this conclusion. In that case, a prisoner convicted of
violating the District of Columbia Code filed a federal habeas
petition challenging the procedures under which he was denied
parole by the D.C. Parole Board. We concluded that section 23-
110 presented no bar to the habeas proceeding, explaining that
the claim could not have been brought under section 23-110
because it did not “challenge [Blair-Bey’s] conviction or
sentence” as provided in section 23-110, and “section 23-110(g)
only bars us from hearing those claims that could have been
raised through section 23-110.” Id. at 1043; see also Neal v.
Director, 684 F.2d 17, 19 (D.C. Cir. 1982) (finding that section
23-110(g) does not preclude federal court jurisdiction over a
prisoner’s habeas challenge to his transfer between
penitentiaries because challenges to prison transfer procedures
fall outside the scope of section 23-110). Blair-Bey therefore
confirms that section 23-110(g) divests federal courts of
 8
jurisdiction only over habeas petitions by prisoners who, unlike
Williams, have an effective section 23-110 remedy available to
them.

Blair-Bey also speaks to the question, arguably left open in
Streater, whether the availability of an adequate local remedy
outside section 23-110 is sufficient to bar prisoners sentenced in
the District of Columbia from seeking federal habeas relief. In
Blair-Bey, as in this case, the prisoner had another means to
seek his release: section 16-1901 of the D.C. Code, which
provides a general habeas corpus remedy for prisoners confined
in the District. D.C. Code § 16-1901. Despite the availability of
that alternative procedure, however, we allowed Blair-Bey’s
federal habeas petition to go forward. Blair-Bey, 151 F.3d at
1047.

The section 16-1901 procedure at issue in Blair-Bey is
analogous to the mandate-recall procedure at issue here in that
both provide prisoners with a means to secure their release, and
both provide relief comparable to that otherwise available for
claims that fall within section 23-110’s scope. See Norris v.
United States, 927 A.2d 1034, 1038 (D.C. 2007) (explaining that
section 16-1901 and section 23-110 are both “designed to permit
challenges to unlawful custody”). Indeed, section 16-1901 is
perhaps more akin to section 23-110 because, like a section 23-
110 motion, a habeas petition under section 16-1901 is filed
directly with the Superior Court. See D.C. Code § 16-1901(c).
But just as the availability of section 16-1901 did not bar Blair-
Bey’s federal habeas petition, the availability of the mandate-
recall procedure does not divest the federal district court of
jurisdiction over Williams’s petition.

Nothing in Garris v. Lindsay, 794 F.2d 722 (D.C. Cir.
1986), on which the government relies, requires a different
result. In that case, Garris, a D.C. prisoner, argued in his direct
 9
appeal that he had been denied his Sixth Amendment right to
represent himself in his Superior Court trial. The D.C. Court of
Appeals rejected that claim, and D.C. law barred him from
relitigating the issue collaterally under section 23-110. Id. at
727. Garris then filed a federal habeas petition in which he
argued that because he was unable to take advantage of section
23-110, the district court had authority to entertain his petition.
Noting that “[i]t is the inefficacy of the remedy, not a personal
inability to utilize it, that is determinative,” we concluded that
the district court lacked habeas jurisdiction. Id. at 727. Here, it
is indeed the “inefficacy of the remedy” that presents the
problem. Section 23-110 was unavailable to Williams because
his constitutional claim—unlike Garris’s—falls outside that
statute’s scope.

In concluding that Williams may proceed with his habeas
petition, we are mindful that when Congress enacted section 23-
110 as part of the District of Columbia Court Reform and
Criminal Procedure Act of 1970, Pub. L. No. 91-358, § 210, 84
Stat. 608, it sought to vest the Superior Court with exclusive
jurisdiction over most collateral challenges by prisoners
sentenced in that court. See Swain v. Pressley, 430 U.S. 372,
378 (1977); Blair-Bey, 151 F.3d at 1045–46; see also Swain,
430 U.S. at 381–82 (explaining that “[s]ince the scope of the
remedy provided by § 23-110 is the same as that provided by §
2255, it is also commensurate with habeas corpus in all respects
save” administration by Article III judges). That said, Congress
also included the “inadequate or ineffective” exception,
indicating that it contemplated circumstances under which
prisoners sentenced in Superior Court could petition for habeas
relief in federal court. As noted above, supra at 7–8, we have
already recognized some exceptions under section 23-110(g),
and today we recognize another. The government warns that
allowing habeas petitions like Williams’s to proceed will “open
the floodgates to frivolous federal habeas claims.” Appellees’
 10
Br. 35. Although the government gives us no basis for thinking
that will happen, Congress can always close the “floodgates” if
the feared deluge comes to pass.

The concurring opinion correctly notes that D.C. prisoners
who challenge the effectiveness of appellate counsel through a
motion to recall the mandate in the D.C. Court of Appeals will
get a second bite at the apple in federal court. But that is a
consequence of section 23-110(g), which bars federal habeas
claims only when the section 23-110 motion is adequate or
effective. As explained above, a motion to recall the mandate
does not qualify as a motion pursuant to section 23-110.
Moreover, we allowed Blair-Bey to file a federal habeas petition
challenging his parole proceedings even though the D.C. Court
of Appeals had affirmed the Superior Court’s denial of his D.C.
habeas petition making precisely the same claim. Blair-Bey,
151 F.3d at 1038, 1047. We also recognize that if the D.C.
Court of Appeals wishes to avoid federal habeas review of
ineffective assistance of appellate counsel claims, it would have
to abandon its mandate-recall procedure and permit the Superior
Court to entertain such challenges under section 23-110—just as
we require such claims to be raised directly in the district court
pursuant to 28 U.S.C. § 2255. See United States v. Henry, 472
F.3d 910, 915 (D.C. Cir. 2007) (requiring federal prisoners to
raise ineffective assistance of appellant counsel claims pursuant
to section 2255); concurring op. at 5–6. But again, that is a
consequence of how Congress wrote section 23-110 and how
the District of Columbia’s highest court has interpreted that
provision of the D.C. Code. Although Congress could amend
the statute, we cannot, and on questions of District of Columbia
law this court defers to the D.C. Court of Appeals. See Blair-
Bey, 151 F.3d at 1050.

Given the foregoing, Williams has more than satisfied the
first step of his COA burden. He has shown not just that the
 11
district court’s habeas jurisdiction is debatable, see Slack, 529
U.S. at 484, but that under section 23-110 the district court in
fact has jurisdiction to entertain his habeas petition. We must
therefore decide whether Williams has presented a reasonably
debatable claim of the denial of a constitutional right, an issue to
which we now turn.

III.
Although criminal defendants enjoy a due process right to
the effective assistance of counsel during their first appeal as of
right, Evitts v. Lucey, 469 U.S. 387, 396 (1985), the Supreme
Court has made clear that defendants lack a constitutional
entitlement to effective assistance of counsel in state collateral
proceedings, Coleman v. Thompson, 501 U.S. 722, 752 (1991).
In this case, the government contends that Williams’s
ineffective assistance of counsel claims relate solely to his
section 23-110 motion, a collateral procedure to which no right
of counsel attaches. Accordingly, the government argues,
Williams has failed to allege the denial of a constitutional right
and so has no right to a COA. We disagree.

The government is certainly correct that Williams’s habeas
petition challenges the effectiveness of counsel in the section
23-110 proceedings. But the petition does not stop there. It
goes on to challenge the effectiveness of counsel in the direct
appeal as well. “Ground two” of the petition expressly alleges
the “[d]enial of due process and effective assistance of counsel
on first appeal of criminal conviction.” Pet. for Writ of Habeas
Corpus at 19.5, Williams v. Martinez, 559 F. Supp. 2d 56
(D.D.C. 2008) (No. 08-0971). Contrary to the government’s
claim, then, Williams has in fact asserted the denial of a
constitutional right—effective assistance of counsel in his direct
appeal.
 12
Having established that Williams has asserted a
constitutionally cognizable right in his habeas petition, we must
determine whether he has shown a reasonably debatable
infringement of that right. See Slack, 529 U.S. at 484. “The
COA determination under § 2253(c) requires an overview of the
claims in the habeas petition and a general assessment of their
merits.” Miller-El v. Cockrell, 537 U.S. 322, 336 (2003).
Because the district court found it unnecessary to reach the
merits of Williams’s habeas petition, however, we lack the
benefit of that court’s analysis. Moreover, the parties’ briefs
give little attention to the merits of Williams’s claim, and to the
extent that they do, they disagree on the precise nature of that
claim. Given all this, we are unprepared to decide whether
Williams has met his burden of stating “a valid claim of the
denial of a constitutional right.” Slack, 529 U.S. at 484; see id.
at 485, 489 (declining to address the merits of the COA analysis
where the parties had not briefed the issue). Accordingly, we
remand the case to the district court to consider the merits
component of the COA question, an evaluation that the court
should undertake in light of the standard set forth in 28 U.S.C. §
2254. See Miller-El, 537 U.S. at 349–50 (Scalia, J., concurring)
(explaining that a judge should deny a COA if all reasonable
jurists would conclude that the habeas statute bars relief); cf.
Madley v. U.S. Parole Comm’n, 278 F.3d 1306, 1308–09 (D.C.
Cir. 2002) (holding that District of Columbia courts are deemed
to be state courts for purposes of 28 U.S.C. § 2253); Coady v.
Vaughn, 251 F.3d 480, 484–85 (3d Cir. 2001) (requiring a state
prisoner to challenge his custody under 28 U.S.C. § 2254 rather
than § 2241).

For the foregoing reasons, we reverse the dismissal of
Williams’s habeas petition and remand to the district court for
further proceedings consistent with this opinion.

So ordered.
 BROWN, Circuit Judge, concurring in the judgment: I
agree it was error for the district court to dismiss Williams’s
habeas petition for lack of jurisdiction pursuant to D.C. Code
§ 23-110(g). However, the court interprets section 23-110 so
literally it confers routine jurisdiction in federal court for all
claims of ineffective assistance of appellate counsel by
prisoners under D.C. sentence. Since I believe this result
departs from congressional intent, I would read the statutory
scheme broadly to maintain federal jurisdiction solely as a
safety valve. Therefore, I concur only in the judgment.

I.

History matters here. Our current dilemma arises out of a
succession of procedural anomalies that can only be described
as “A Series of Unfortunate Events.” First, when section 23-
110 was enacted in 1970, the constitutional claim of appellate
ineffective assistance of counsel (IAC) did not exist. The
Supreme Court did not recognize a Sixth Amendment right to
effective assistance of counsel on direct appeal of a criminal
conviction until Evitts v. Lucey, 469 U.S. 387, 395 (1985).
Several years before Lucey, however, the District of Columbia
Court of Appeals (DCCA) considered an appellate IAC claim
in Streater v. United States, 429 A.2d 173 (D.C. 1980)
(Streater I). In Streater I, the DCCA held Streater, a D.C.
prisoner, could not raise the claim in a section 23-110 motion.
The DCCA determined the claim was “not within the purview
of § 23-110” because, among other reasons, Streater “has not
argued that his sentence was imposed in violation of the
Constitution of the United States.” Id. at 174. At the time,
prior to Lucey, this was a correct statement of the scope of
section 23-110. Relying on two federal circuit decisions
interpreting the federal habeas statute, 28 U.S.C. § 2255, the
DCCA also noted section 23-110 “provides no basis upon
which the trial court may review appellate proceedings.” Id.
 2

In Streater v. Jackson, 691 F.2d 1026 (D.C. Cir. 1982)
(Streater II), we reviewed the district court’s dismissal of
Streater’s habeas petition. Streater argued he was given the
“run around” by the D.C. courts. Id. at 1028. We observed
the DCCA had not “enlightened Streater as to the remedy, if
any, still open to him in the local courts.” Id. Reluctant to
meddle with the DCCA’s jurisdiction, we ordered the district
court to hold Streater’s petition in abeyance and invited the
DCCA to consider his motion to recall the mandate. Id. It
did. See Streater v. United States, 478 A.2d 1055 (D.C.
1984).

Fast forward to Watson v. United States, 536 A.2d 1056
(D.C. 1987) (en banc), in which the DCCA again faced an
appellate IAC claim, but this time post-Lucey. The DCCA
revisited the procedural question of “how one may challenge
previous counsel’s effectiveness on appeal.” Id. at 1059.
Turning to Streater I, the DCCA again rejected section 23-
110 as a procedure for raising this claim because “the
Superior Court should not have authority to rule on the
constitutionality of an appellate proceeding.” Id. at 1060
(citing Streater I, 429 A.2d at 174). The court also refused to
allow the claim to be brought under the District’s general
habeas statute, D.C. Code § 16-1901, for the same reason. Id.
(citing Streater II, 691 F.2d at 1028). Having dismissed these
statutory habeas remedies, the court identified an appropriate
procedural vehicle for the claim: “A motion in this court to
recall the mandate is the appropriate avenue to take in
presenting a Lucey challenge.” Id. (citing United States v.
Winterhalder, 724 F.2d 109, 111 (10th Cir. 1983)). In
reaching this conclusion, the DCCA observed that in Streater
II we had approved its decision to remove the claims from
section 23-110 in Streater I. Id. at 1060–61.
 3

II.

With this history in mind, I turn to the statutory provision
at issue, section 23-110(g), which provides:

An application for a writ of habeas corpus in
behalf of a prisoner who is authorized to apply for
relief by motion pursuant to this section shall not
be entertained by the Superior Court or by any
Federal or State court if it appears that the
applicant has failed to make a motion for relief
under this section or that the Superior Court has
denied him relief, unless it also appears that the
remedy by motion is inadequate or ineffective to
test the legality of his detention.

D.C. Code § 23-110(g). The court interprets this provision to
mean the federal courts have jurisdiction to entertain a D.C.
prisoner’s habeas petition whenever the prisoner is not
“authorized to apply for relief by motion pursuant to this
section.” See Op. at 5–6. Because in Streater I the DCCA
held appellate IAC claims cannot be presented in a section 23-
110 motion, the court correctly observes a prisoner with such
a claim is not “authorized” to file a section 23-110 motion;
thus, the court reasons, section 23-110 is “by definition,
inadequate” to address those claims. Op. at 6. While it is true
the statute’s plain words may be read to justify the court’s
holding, the result is clearly at odds with the statute’s purpose.

Enacted by Congress in 1970, section 23-110 was only
one provision of a sweeping legislative reform designed to
remove “local litigation” from the federal courts to the
District of Columbia’s judicial system. Swain v. Pressley,
430 U.S. 372, 375 (1977). As the Supreme Court noted in
 4

Swain, section 23-110 created for prisoners under D.C.
sentence a collateral procedure in the Superior Court
“comparable” to the habeas statute for federal prisoners,
section 2255. Id. Section 23-110, like section 2255, was
“intended to substitute a different forum [the Superior Court]
and a different procedure [section 23-110] for collateral
review” of D.C. prisoners’ sentences. Id. at 378. The
statute’s clear purpose was to shift initial habeas jurisdiction
for D.C. prisoners from the federal courts to the District of
Columbia courts. See Byrd v. Henderson, 119 F.3d 34, 36–37
(D.C. Cir. 1997).

This court now says the DCCA’s interpretation of section
23-110 requires federal courts to assume jurisdiction over
appellate IAC claims. The court in effect says, “The DCCA
made us do it.” Congress enacted section 23-110 with the
broad purpose of accomplishing the opposite result, and the
DCCA, with our encouragement, exercised its best judgment
in fashioning a procedural remedy for a heretofore
unrecognized constitutional claim. The DCCA therefore
could respond, “The D.C. Circuit made us do it.”

III.

The answer lies in section 23-110(g)’s “inadequate and
ineffective” clause—what we have called the “safety valve.”
Blair-Bey v. Quick, 151 F.3d 1036, 1042 (D.C. Cir. 1998). In
Swain, the Supreme Court upheld section 23-110 against a
Suspension Clause challenge, relying heavily on the last
clause of section 23-110(g): “That clause allows the District
Court to entertain a habeas corpus application if it ‘appears
that the remedy by motion is inadequate or ineffective to test
the legality of [the applicant’s] detention.’” 430 U.S. at 381.
Congress decided to replicate a “state” judicial system in the
 5

District of Columbia, with no interference from the federal
courts unless the state habeas remedy is deficient. Logically,
then, section 23-110(g) should be read purposively to require
federal courts to determine if the remedy, including a
substitute procedural mechanism like the motion to recall the
mandate, is “adequate and effective to test the legality” of the
prisoner’s detention.

The DCCA’s motion to recall the mandate procedure is
cumbersome, requiring the movant to meet a high initial
burden. See Watson, 536 A.2d at 1060. It is thus unclear
whether the recall-mandate procedure is an entirely adequate
and effective remedy. But we do not have to answer that
question in Williams’s case because the DCCA summarily
denied his motion to recall the mandate, his motion for an
explanation, and his petition for rehearing. The DCCA’s
failure to explain why it denied Williams’s motion leaves us
with no basis to determine whether it actually considered his
claim on the merits or rejected the claim solely because
Williams had failed to satisfy the initial burden for such
motions. We therefore cannot find the remedy afforded
Williams is adequate and effective, and section 23-110(g)’s
safety valve operates to allow the federal district court
jurisdiction to entertain his habeas petition.

IV.

With all this in mind, I see no reason for the court to
revisit the Streater question and reinterpret section 23-110(g).
The court today does a bait-and-switch: it rejects the mandate-
recall procedure and informs the DCCA the federal courts
have jurisdiction to hear Williams’s claim and others like it
unless the DCCA overrules its precedent and allows those
 6

claims to be heard by the Superior Court under section 23-
110.

It is true the procedural diktat the court imposes on the
DCCA today is identical to the procedure we have adopted for
ourselves. We “ordinarily” require federal prisoners to raise
appellate IAC claims collaterally pursuant to section 2255.
United States v. Henry, 472 F.3d 910, 914 (D.C. Cir. 2007).
But to give credit where credit is due, at the time the DCCA
adopted the motion to recall the mandate procedure—only a
year after Lucey—its decision was eminently reasonable.
Once the Supreme Court had authorized a new constitutional
claim to challenge events happening after the trial court
proceedings, it was logical to require the claim to be initiated
in the court before which the alleged defect occurred.

Moreover, when facing this same question, other federal
and state appellate courts have made the exact same choice.
See State v. Knight, 484 N.W.2d 540, 543 n.5 (Wis. 1992)
(noting Third, Ninth, and Tenth Circuits, Missouri, and the
District of Columbia employed motion to recall the mandate
procedure); see also United States v. West, 240 F.3d 456, 460
n.3 (5th Cir. 2001) (surveying approaches of federal and state
courts). Still other courts have answered the question
differently than either the DCCA or our circuit. For instance,
the Wisconsin Supreme Court, confronting a statutory
provision nearly identical to section 23-110, determined the
statutory provision was “‘inadequate or ineffective’” to
address appellate IAC claims. Knight, 484 N.W.2d at 544.
After considering the approaches taken by other appellate
courts, the Wisconsin Supreme Court held the claim should be
presented directly to the state appellate court in an original
writ of habeas corpus. Id. at 544–45; see also West, 240 F.3d
at 460 n.3 (observing Second Circuit recalls its own mandate
 7

from dismissal of direct appeal after district court has denied
section 2255 relief). We cannot fault the DCCA for doing
exactly what other courts have done when facing this thorny
procedural question.

Because the court has determined, as a matter of law, that
section 23-110 is inadequate and ineffective to raise appellate
IAC claims, a D.C. prisoner may now file a habeas petition
asserting this claim in the federal district court, and the court
will be obligated to review the claim on the merits. In light of
the exhaustion requirement of 28 U.S.C. § 2254(c), the
prisoner will first have to file a motion to recall the mandate
with the DCCA. But even if the DCCA recalls the mandate,
remands the record to the Superior Court for a factual hearing,
and then denies the prisoner’s claim in a decision on the
merits—clearly an adequate and effective remedy—the
prisoner still gets a second (or more accurately third) bite at
the apple in federal court. Some of these claims will be
summarily resolved pursuant to AEDPA’s deferential
standards, and it is unclear what the added burden on our
courts will be. For instance, Williams’s appellate IAC claim
appears to be little more than a string of tenuous arguments
nested like Russian dolls. A better use of our judicial
resources would be to assert jurisdiction only where the safety
valve is truly implicated.

At this late hour, rather than leaving the DCCA with a
Hobson’s choice, I would allow it to address the matter in the
first instance, as we did in Streater II. There we noted, “it is
apparent that the D.C. Court of Appeals is the tribunal best
situated to address Streater’s claim that he was denied
effective assistance of counsel in that forum.” 691 F.2d at
1028. Principles of federalism and comity also gave us pause
back then: “Mindful that relations between the District of
 8

Columbia and federal systems should not be ‘disturbed by
unnecessary conflict between courts equally bound to guard
and protect rights secured by the Constitution,’ we believe the
D.C. Court of Appeals should be invited to consider and rule
on the merits of Streater’s claim for post-conviction relief.”
Id. (quoting Ex parte Royall, 117 U.S. 241, 251 (1886)).
These concerns are heightened when, as is the case here, we
interpret a provision of the D.C. Code that is “an Act of
Congress applicable exclusively to the District of Columbia”
because “[w]e do not treat such local statutes as if they were
part of the United States Code,” and “[o]ur policy has been to
defer to the District of Columbia Court of Appeals on
questions of statutory interpretation.” United States v.
Edmond, 924 F.2d 261, 264 (D.C. Cir. 1991). If, as in
Streater II, we remanded Williams’s habeas petition to the
federal district court and ordered it held in abeyance, we
would allow the DCCA the opportunity either to recall its
mandate and address Williams’s claim on the merits or to
revisit the underlying question that has caused this court and
others so many procedural headaches.