by Kent Scheidegger*
Reading the headlines about Supreme Court decisions in April, one
would think the murderers were victorious across the board and that
the recent reforms in habeas corpus had been defeated. In
truth, the 1996 reform is being upheld largely as intended, and
the Supreme Court decisions have moved the law in the prosecution's
favor, compared to the center of gravity of the court of appeals'
positions prior to the present term.
Terry Williams v. Taylor
On April 18, the Supreme Court decided two habeas
cases, both named Williams v. Taylor. The Terry
Williams case involved the interpretation of the centerpiece
of the habeas reform portion of the 1996 Antiterrorism and Effective
Death Penalty Act: 28 U.S.C.
§ 2254(d). The critical battleground involved rulings
of state courts on the so-called "mixed questions of law and fact,"
i.e., the application of a general legal standard to the specific
facts of the case. Should a federal district or circuit court
nullify a state court decision on such an issue merely because it
disagrees with it, as a "higher" court would on a normal appeal, or
should the drastic remedy of collateral attack on a final judgment
be reserved for clearly wrong decisions, letting the original decision
stand if it is within the bounds of reasonable disagreement?
As this bill was debated in Congress, everyone involved understood
that the Hatch-Specter language ultimately adopted represented the
latter choice, and the Biden Amendment, specifically stated by its
author to be for the purpose of retaining de novo review of
such questions, was defeated. Yet, after the legislative smoke
cleared, various academics proposed a tortured construction of the
language to preserve the status quo ante and defeat the main
purpose of the reform. Not a single Court of Appeals bought
this argument. Though their formulations varied somewhat, every
case recognized that Congress had indeed changed the rule on mixed
Justice O'Connor's opinion is the opinion of
the Court on the main point, except that Justice Scalia characteristically
declines to join an incidental footnote on legislative history.
The opinion confirms that the basic analytical structure is the
one established in cases from the Fourth, Fifth, Seventh, and Eleventh
Circuits. That is, the phrase "contrary to ... clearly established
federal law" refers, for the most part, to the state court's selection
of the correct legal standard from existing Supreme Court precedent.
In practice, this will not be much different from the pre-AEDPA
case law requirement of Teague v. Lane, 489 U.S. 288
(1989), that new rules not dictated by existing precedent cannot
be created or applied retroactively on habeas corpus. The
application of existing standards to specific facts is governed
by the "unreasonable application" clause, which means what it says:
if the state court's application of the standard is reasonable,
the federal court respects it, even if the federal court would have
reached a different result in a federal prosecution before it on
The strange thing about this case is that four
Justices actually voted for the de facto repeal of the statute
in the guise of "interpretation," an interpretation unanimously
rejected by the Courts of Appeals. Even stranger, and more
frightening, it appears that up to the last minute Justice Stevens'
bizarre opinion had the vote of five Justices and was slated to
be the opinion of the Court. One would normally expect Justice
O'Connor's opinion to be the lead opinion of the case, as it is
the opinion of the Court on the main point and reaches the result
favored by the majority. Yet Justice Stevens' opinion occupies
that position. Justice O'Connor's opinion uses first-person
singular pronouns, which are typically used in concurrences and
dissents, not the opinion of the Court. Justice Stevens' opinion
says at one point, "Our disagreement with Justice O'Connor...."
That is majority language; a dissent would typically say, "My disagreement
with the Court...." The fifth vote must have switched very
near the end.
As close as we came to disaster, though, the
precedent is now established. The reform means what its supporters
intended it to mean. If our experience with Teague
v. Lane is any guide, the law will be respected and enforced
in some courts and shamelessly evaded in others, but on the whole
will provide a corrective force to the overreaching of federal judges.
On the specific case, the Court ruled that the
state courts had neither correctly nor reasonably applied the established
precedent on ineffective assistance of counsel, Strickland
v. Washington, 466 U.S. 668 (1984), and so the case goes
back to Virginia for a new sentencing hearing. The spin doctors
crowed that the door to the federal courthouse had not been completely
barred. Sure. Nobody said it had. Few ever wanted
Michael Williams v. Taylor
In the Michael
Williams case, the Court declined unanimously to overreach
in the other direction. Building on the precedent established
v. Tamayo-Reyes, 504 U.S. 1 (1992), Congress enacted
in 28 U.S.C.
§ 2254(e) that a habeas petitioner who has failed to develop
the factual basis of his claim in state court proceedings cannot do
so in federal court, except in cases meeting two conditions, essentially
both good cause and actual innocence. The question was whether
"failed" means simply did not, or whether it implies some degree of
fault on the petitioner's part. The Court's choice of the latter
interpretation was a win for the petitioner in this case, but it did
not change the law appreciably from the court of appeals decisions.
If the petitioner had no realistic chance to develop the facts in
state court, this subsection is inapplicable. The case goes
back to the district court for an evidentiary hearing.
Slack v. McDaniel
The third win for an individual murderer came April
26 in Slack
v. McDaniel. This case had many issues and
was argued twice in the term. The issue originally argued was
whether, when the first federal petition had been dismissed for failure
to exhaust state remedies, a second petition was subject to dismissal
under the "successive petition" rule. The high court said no,
overruling a dubious Ninth Circuit precedent on the point. The
more difficult question, considered on reargument, was whether and
how to apply the AEDPA's new limits on appeals, 28
U.S.C. § 2253(c), to this case.
Slack's federal petition had been filed before
AEDPA, but the appeal came after. Taking a chunk out of its
disastrous 1997 ruling in Lindh
v. Murphy, 521 U.S. 320, the Supreme Court said
the new law applies, contrary to the holdings of all but one of
the circuits. This is important in capital cases, because
a great many of them were languishing in district courts in April
1996, and all of these will be subject to the new limits.
The statute requires a certificate of appealability
to appeal from denial of habeas relief. Unlike the old certificate
of probable cause, the new certificate is issue-specific.
Again, this is particularly important in capital cases, as it narrows
the appeal from the blizzard of frivolous contentions typical in
such cases down to the very few that might actually have some merit.
A further interpretational question involved what the petition needs
to show to get the certificate when the petition was denied on procedural
grounds, without a ruling on the merits of the underlying claim.
The petitioner said he need only show an arguable claim of error
on the procedural point, a position adopted by several circuits.
The state boldly claimed that such issues are not appealable at
all. The Court held that he must show arguably correct positions
on both the procedural point and the merits. (On this point,
I found myself in the unusual position of having argued, as amicus,
a position more favorable to the petitioner than the one adopted
by the Court. I said a decent claim on the merits was enough.)
The Supreme Court expressed no opinion on whether
Slack had any case on the merits, so he goes back to the Ninth Circuit
to decide that. The press painted this case as a great victory
for the defense bar. That brings to mind Pyrrhus's comment,
"Another such victory ... and we are undone."
Edwards v. Carpenter
Not all the murderers prevailed in the Supreme
Court. In Edwards
v. Carpenter, decided April 25, the state won a
reversal of a Sixth Circuit ruling on procedural default.
Because the strict doctrine of res judicata
does not apply to habeas corpus, it is common for prisoners to make
attack after attack on the same judgment, each time thinking up
new claims and new arguments. To avoid endless litigation
of every case, most states have "procedural default" rules to the
effect that a claim must be raised at the first opportunity:
contemporaneous objection at trial, direct appeal for claims on
the trial record, or the first state habeas petition for claims
requiring facts outside the record. Claims made later must
typically be accompanied by a showing of good cause for not making
the claim earlier. In federal court, a claim defaulted in
state court cannot be considered unless the petitioner shows either
cause and prejudice or actual innocence. The procedural default
rule operates in tandem with the exhaustion rule, requiring defendants
to fairly and timely present their claims to the state courts first,
before turning to the federal courts. While this rule is often
criticized for its complexity, it comes closer to a fair balance
between the need to correct real injustices and the need for some
kind of finality than any alternative proposed.
The question presented in this case was the
role of ineffective assistance of appellate counsel as "cause" for
not raising the claim on appeal. (The claim was insufficiency
of the evidence, an unusual claim for a plea-bargain case.)
However, if Carpenter was going to attack his appellate counsel
as inadequate, the time and place to do that was his first state
habeas petition, and he did not include any such claim in that petition.
If claims omitted from the appeal could be revived without limit
by attacking the performance of appellate counsel, the procedural
default and exhaustion rules would cease to perform their function.
In a 7-2 opinion by Justice Scalia, the Court ruled that the usual
standard applies to the defaulted ineffective assistance claim,
whether it is stated as a free-standing claim or as cause for the
prior default. Carpenter might be able to meet that standard.
The fact he was unrepresented in his first state habeas is surely
relevant. But he has to show a good reason for an exception
to the usual rule of finality.
In federal courts around the country, the prosecution
is in a stronger position in May than it was in March. Don't
believe everything you read in the papers.
* Vice-Chairman, e-communications,
Federalist Society Criminal Law and Procedure Practice Group;
Legal Director, Criminal
Justice Legal Foundation
Criminal Justice Legal Foundation Briefs in Terry
Williams and Slack
Solicitor General's Brief in Slack
Abstracts of dueling law review articles in Columbia
Law Review on § 2254(d), by James
28 U.S.C. § 2254(d):
An application for a writ of habeas corpus on
behalf of a person in custody pursuant to the judgment of a State
court shall not be granted with respect to any claim that was adjudicated
on the merits in State court proceedings unless the adjudication
of the claim --
(1) resulted in
a decision that was contrary to, or involved an unreasonable application
of, clearly established Federal law, as determined by the Supreme
Court of the United States; or
(2) resulted in
a decision that was based on an unreasonable determination of the
facts in light of the evidence presented in the State court proceeding.;