A. Under Wainwright v. Sykes 433 U.S. 72 (1977), a prisoner bringing
a collateral attack on his conviction may be excused from the
consequences of failure to raise the claim at trial or on direct appeal
only by showing "cause" for that procedural default and "actual
prejudice" resulting from it. The court of appeals' holding in this
case that "cause" may be established by showing that the failure to
raise the issue was due to ignorance or inadvertence by counsel, rather
than a conscious and deliberate tactical choice, is flatly inconsistent
with Engle v. Isaac, 456 U.S. 107 (1982). There, the Court held that
the fact that counsel might have "overlooked" a claim or been "unaware"
of the basis for it does not constitute "cause" for the procedural
default, so long as counsel rendered constitutionally adequate
assistance.
B. The Court's rejection in Engle of the rationale adopted by the
court of appeals in this case was firmly rooted in the origins, nature,
and purposes of the "cause and prejudice" standard.
1. In Fay v. Noia, 372 U.S. 391 (1963), the Court held that
collateral attack would be barred by virtue of a procedural default at
trial only if the defendant had "deliberately bypassed" the opportunity
to raise his claim in the state proceedings. In Wainwright v. Sykes,
the Court abandoned that test and substituted in its place the narrower
"cause and prejudice" test. The court of appeals' holding in this case,
under which "cause" would be established unless counsel made a
deliberate tactical choice not to raise the claim, would reintroduce the
deliberate bypass standard that was explicitly rejected in Sykes.
Sykes rests on the principle that a constitutional claim may be
forfeited in a criminal case by the failure to make timely assertion of
that right at trial or on direct appeal and that such a forfeiture is
presumptively enforceable in a collateral attack on the conviction in
federal court. The existence of a presumptively enforceable procedural
default thus turns not on a subjective intent to waive the
constitutional claim, as under Fay v. Noia, but on the occurrence of an
objectively verifiable event: whether the defendant complied with a
contemporaneous objection requirement or a similar precondition to
preserving a claim. Similarly, under Sykes the existence of "cause" to
excuse an otherwise enforceable procedural default also turns not on the
subjective intent of the defendant or his attorney, but on the existence
of an objectively verifiable and external impediment to the defense's
developing or presenting the claim -- such as the refusal by the trial
or appellate court to entertain the claim or the impracticability of
raising it, interference by officials with the assertion of the right,
or the unavailability at the time of trial or appeal of the factual or
legal basis for the claim.
The deliberate bypass rule of Fay v. Noia was drawn from the
standards for determining whether there has been a valid waiver of a
constitutional right by the person whose right is at stake. Because the
constitutional rights on which counsel might base an objection at trial
or on appeal are not personal to the attorney, the court of appeals'
requirement that counsel have made a deliberate choice not to raise the
claim has no logical nexus to the standard for waiver of constitutional
rights on which Fay v. Noia rested. Moreover, because a lawyer acts on
behalf of the accused in a criminal prosecution and the law therefore
does not distinguish between the defendant and counsel insofar as
compliance with procedural requirements is concerned, there is no basis
under Sykes for distinguishing between the defendant and his attorney
for purposes of determining whether actual knowledge and a deliberate
decision are required in order to excuse a failure to comply with such
rules.
2. The court of appeals' holding in this case also finds no support
in the competing policy considerations that underlie the "cause and
prejudice" standard. From the state's perspective, the interest in
insisting upon a contemporaneous objection in order to avoid error and
assure finality, and to ensure that any needed retrial is promptly held,
is fully applicable irrespective of whether the procedural default at
trial or on appeal was due to a deliberate decision by counsel or to his
ignorance or inadvertence. There also would be substantial costs
associated with determining whether counsel actually had made a
deliberate tactical choice, because the habeas court would be required
to conduct the sort of potentially chilling post-trial inquiry into
counsel's tactics and motivations about which the Court was concerned in
Struckland v. Washington, No. 82-1554 (May 14, 1984), slip op. 19-20,
27.
Nor does the Court's decision in Reed v. Ross, No. 83-218 (June 27,
1984), support the court of appeals' rule. The court there held that
"cause" is established if the claim involved was novel at the time of
trial. However, whether a claim was not "reasonably available" and
therefore was novel under Reed v. Ross turns on an objective factor
external to the defense -- the state of the law at the time of trial or
appeal -- not on the subjective state of mind of the attorney at that
time. Moreover, where a claim is not novel, there is a substantial risk
that the failure to raise it was tactical. In this case, for example,
the fact that counsel initially raised the discovery issue in his
statement of issues on appeal suggests that he considered pursuing the
point but deliberately deleted it in order to concentrate on more
promising issues.
From the standpoint of the defendant, the fundamental fairness of his
trial would seem to be little affected by whether a particular claim was
lost through counsel's failure to realize its availability and potential
merit or through counsel's deliberate (but perhaps unwise) decision not
to press the claim. In either event the defendant stands equally
convicted, and any possibility that the outcome would have been
different had his lawyer acted differently is entirely unaffected by his
lawyer's motive (or lack of motive) for failing to assert the claim, as
long as the lawyer satisfied the constitutional standards for effective
assistance of counsel. As this Court made clear in Strickland v.
Washington, where counsel's performance satisfied these standards, the
trial must be regarded as having been fundamentally fair insofar as the
role of counsel is concerned.
ARGUMENT
INADVERTENCE OR IGNORANCE SHORT OF INEFFECTIVE ASSISTANCE OF COUNSEL
DOES NOT CONSTITUTE "CAUSE" FOR EXCUSING A PROCEDURAL DEFAULT
Under Wainwright v. Sykes, 433 U.S. 72 (1977), and its progeny, a
procedural default that bars litigation of a constitutional claim in
state court also forecloses federal habeas corpus relief absent a
showing of "cause" for that default and "actual prejudice" resulting
from it. The court of appeals held in this case that a prisoner can
satisfy the "cause" requirement by showing that the failure to raise the
claim at trial or on appeal was the result of inadvertence or ignorance
by his attorney, rather than being a conscious and deliberate tactical
decision. This is so, the court held, even though counsel's "momentary
lapse" did not cause his performance to fall outside the range of
ordinary attorney competence and therefore would not support an
independent claim of ineffective assistance of counsel under the Sixth
Amendment at trial /5/ or under the Due Process Clause on appeal. /6/
As a practical matter, the test announced by the court of appeals
amounts to a virtual elimination of the "cause" requirement and a return
to the "deliberate bypass" standard adopted in Fay v. Noia, 372 U.S. 391
(1963). It excuses nondeliberate defaults, not only in the case of
those critical decisions that must be made by the client personally, but
in connection with any question of trial or appellate strategy or
procedure that may result in a default.
Apart from the devastating effect adoption of this standard would
have on the effectiveness of procedural default rules in serving their
intended purpose, it would also pose serious problems of administration
for the habeas court. To begin with, in order to resolve the "cause"
issue under the Fourth Circuit's approach, the court on collateral
attack would be forced in most if not all cases to conduct a hearing and
make difficult determinations regarding counsel's trial and appellate
strategy and motives in an attempt to ascertain whether the default that
occurred actually was the result of ignorance or inadvertence rather
than deliberate choice. Beyond that, however, the "ignorance or
inadvertence" standard would not be easy to apply, since there are
numerous intermediate positions between complete unawareness of a claim
and a deliberate decision not to assert it despite an accurate
assessment of its merit. For example, if a lwayer considered making a
claim and decided not to do so, based upon an erroneous judgment that
the claim lacked legal merit, /7/ does the error amount to "ignorance"
excusing the default? If so, there would be few cases indeed in which
"cause" could not be shown.
A. This Court's Decision In Engle v. Isaac Establishes That
Inadvertence Or Ignorance On The Part Of Counsel Does Not Satisfy The
"Cause" Requirement
The court of appeals' holding in this case, which largely vitiates
the "cause" requirement, is directly inconsistent with this Court's
decision in Engle v. Isaac, 456 U.S. 107 (1982). In Engle, the Court
considered whether "cause" had been established for the habeas
petitioners' procedural default in failing to object to a jury
instruction that required them to bear the burden of proof on the issue
of self-defense. See 456 U.S. at 110-115. The Court stated that in
such a case "the defendant's counsel, for whatever, reasons, has
detracted from the trial's significance by neglecting to raise a claim
in that forum," and the Court acknowledged that "counsel's default may
stem from simple ignorance or the pressures of trial," as well as a
"deliberate" choice by counsel. Id. at 128-129 & n.34. These passages
make clear that the Court was addressing defaults that stemmed from
ignorance or inadvertence as well as those due to deliberate tactical
decisions.
The Court then reviewed the state of the law at the time the habeas
petitioners in Engle were tried (456 U.S. at 130-133), finding that they
did not then "lack the tools to construct their constitutional claim" on
the burden of proof issue (id. at 133). Against this background, the
Court concluded (id. at 133-134 (emphasis added)):
We do not suggest that every astute counsel would have relied
upon (In re Winship, 397 U.S. 358 (1970)) to assert the
unconstitutionality of a rule saddling criminal defendants with
the burden of proving an affirmative defense. Every trial
presents a myriad of possible claims Counsel might have overlooked
or chosen to omit respondents' due process agrument while pursuing
other avenues of defense. We have long recognized, however, that
the Constitution guarantees criminal defendants only a fair trial
and a competent attorney. It does not insure that defense counsel
will recognize and raise every conceivable Constitutional claim.
Where the basis of a constitutional claim is available, and other
defense counsel have perceived and litigated that claim, the
demands of comity and finality counsel against labeling alleged
unawareness of the objection as cause for a procedural default.
In this passage, the Court clearly rejected the contention that
counsel's inadvertence (as "overlooked" claim) or his ignorance
("unawareness") of the legal basis for the claim would constitute
"cause" excusing the procedural default, at least where the defendant
was provided the "competent attorney" that the "Constitution guarantees"
-- i.e., at least where counsel's omission did not amount to ineffective
assistance of counsel under the Sixth Amendment. /8/
The court of appeals' holding in this case simply cannot be squared
with the explicit holding of Engle. And, indeed, the court of appeals
made no effort to do so; it did not even cite or otherwise acknowledge
these dispositive portions of the Court's opinion. /9/ By contrast, the
other courts of appeals that have addressed the question since Engle
uniformly have held that inadvertence, ignorance or like error by
counsel short of constitutionally ineffective assistance does not
constitute the "cause" necessary to excuse a procedural default. /10/
In reaching its decision here the Fourth Circuit, with virtually no
independent analysis of the "cause" requirement, relied exclusively on
decisions of other courts of appeals that were rendered prior to Engle
and that no longer are good law even within their own circuits, as shown
by subsequent decisions of those courts rendered after Engle. /11/ The
court of appeals' decision here was thus an anachronism when it was
rendered and adopted by the en banc court.
B. The Rule Announced By The Court Of Appeals Is Inconsistent With
The Origins, Nature, And Purposes Of The "Cause And Prejudice" Standard
The court of appeals' holding that a procedural default may be
validly enforced only when the failure to raise or preserve a claim is
the product of the lawyer's deliberate tactical choice -- and
correspondingly that a habeas petitioner can establish "cause" for a
procedural default by showing that is was attributable to ignorance or
inadvertence by counsel-- also is inconsistent with the origins, nature,
and purposes of the "cause and prejudice" standard.
1. a. In Wainwright v. Sykes, 433 U.S. 72 (1977), the Court held that
the "cause and prejudice" standard, which previously had been applied in
Davis v. United States, 411 U.S. 233 (1973), and Francis v. Henderson,
425 U.S. 536 (1976), in the context of the failure to object to the
composition of the grand jury prior to trial, should also be applied to
other forms of defaults. The Sykes Court first described the standard
that had been announced in Fay v. Noia, 372 U.S. 391 (1963), under which
a federal district court was to deny relief only if the applicant had
"'deliberately by-passed the orderly procedure of the state courts'"
(433 U.S. at 83 (quoting 372 U.S. at 438)). This standard in Fay for
determining when a defendant was to be bound to a procedural default was
drawn from the standard the Court had applied in other contexts for
determining whether there had been a binding waiver of a constitutional
right such as the right to counsel or to a jury trial -- i.e. whether
there had been "'an intentional relinquishment or abandonment of a known
right or privilege'" (372 U.S. at 439, quoting Johnson v. Zerbst, 304
U.S. 458, 464 (1938)). In the Court's words in Sykes, a waiver under
Fay had to be "knowing and actual" (433 U.S. at 83).
In adopting the "cause and prejudice" standard in Sykes, the Court
left for resolution in future decisions the precise definition of that
standard, except to note that "it is narrower than the standard set
forth in dicta in Fay v. Noia, 372 U.S. 391 (1963), which would have
federal habeas review generally available to state convicts absent a
knowing and deliberate waiver of the federal constitutional contention."
433 U.S. at 87. It therefore is clear that the Court intended in Sykes
that the defendant could be held to his procedural default in the
original proceedings, and thereby precluded from raising his
constitutional claim on collateral attack, even if he was unaware of the
claim at the time the forfeiture occurred.
It would be wholly inconsistent with Sykes to hold that the habeas
petitioner nevertheless may establish "cause" by showing that his
default was not due to a knowing and deliberate tactical decision but
instead was due to inadvertence or ignorance; to do so would
reintroduce the deliberate bypass rule of Fay v. Noia. Yet that is what
the court of appeals has done in this case. The only difference is that
the court of appeals' decision focuses on whether the lawyer, rather
than the defendant personally, made a deliberate tactical choice. That
is not what this Court had in mind in Skyes. See Pet. App. 20 (Hall, J.
dissenting); Palmes v. Wainwright, 725 F.2d 1511, 1525 (11th Cir.
1984).
The background and circumstances of Sykes certainly do not suggest
that the Court intended to preserve the deliberate bypass rule in this
manner. See 433 U.S. at 94 (Burger, C.J., concurring). Justice
Brennan's dissent in Sykes pointed out that there was no basis for
inferring that either Sykes or his lawyer was even aware of the
existence of the Fifth Amendment claim (433 U.S. at 104); by repeatedly
focusing on the question (id. at 100, 101 n.3, 104, 105, 107, 110, 111,
112, 113, 114 & n.13, 115, 117), the dissent recognized that the logic
of the majority's holding would result in the preclusion of federal
habeas relief where the procedural default resulted from counsel's
inadvertence or negligence. A further confirmation of the intent of
Sykes in this regard derives from the Court's tracing of the origins of
its rule to Davis v. United States (see 433 U.S. at 84-85), because the
Court's opinion in Davis indicates (411 U.S. at 243-244) that neither
the defendant nor his attorney was actually aware of all the relevant
facts bearing on a possible objection to the composition of the grand
jury, even though these facts could have been ascertained at the time
that Fed. R. Crim. P. 12 required an objection to be lodged. See also
Shotwell Mfg. Co. v. United States, 371 U.S. 341, 361-364 (1963). /12/
b. In any event, the Fourth Circuit's holding in this case is
inconsistent with the premises and reasoning of Sykes. That decision
rests on the principle "that a constitutional right may be forfeited in
criminal as well as civil cases by the failure to make timely assertion
of the right before a tribunal having jurisdiction to determine it."
Yakus v. United States, 321 U.S. 414, 444 (1944). A forfeiture
therefore need not amount to a "knowing and actual" waiver of the
opportunity to raise the constitutional claim in state court in order to
be enforced by a federal court on collateral attack, as was required
under Fay v. Noia. See 433 U.S. at 83. Under Sykes, the procedural
default is presumptively enforceable on collateral attack even if it was
only a constructive "waiver" that resulted by operation of law from the
failure to comply with a procedural provision requiring a claim to be
presented at a particular time. The question of an enforceable default
thus turns not on subjective intent, but on the occurrence of an
objectively verifiable factor: whether the defendant complied with a
contemporaneous objection requirement or a similar procedural
precondition to preservation of a claim. Both logically and of
practical necessity, this rule of forfeiture applies whether or not
either the defendant or his lawyer actually recognized and considered
raising the claim.
Under Sykes, relief from an otherwise enforceable forfeiture may be
granted in a subsequent collateral attack proceeding only upon a showing
of some impediment to complying with the procedural requirement at trial
or on direct appeal that constitutes a legitimate and sufficient "cause"
for the default. But the absence of actual knowledge and a deliberate
decision by either the defendant or his attorney cannot be such "cause";
if it were, the distinction between loss of a claim through a default
and the relinquishment or abandonment of a right by a conscious and
affirmative waiver would be obliterated. Instead, like the existence of
the default itself, the existence of "cause" to excuse the default
ordinarily should be determined by reference to objectively verifiable
factors that are external to the defense.
Thus, for example, "cause" for a procedural default would be
established by showing that the court refused to entertain an objection
or an offer of proof, or that it was not practicable for the defendant
to make one (compare Reece v. Georgia, 350 U.S. 85 (1955), with Michel
v. Louisiana, 350 U.S. 91 (1955); cf. Rose v. Mitchell, 443 U.S. 545,
561, 563 (1979); Stone v. Powell, 428 U.S. 465, 489, 494 (1976)), that
the default resulted from "incapacity, or some interference by
officials" (Brown v. Allen, 344 U.S. 443, 486 (1953)), that the legal or
factual basis for the claim was not "reasonably available" with the
exercise of due diligence at the time it was required to have been
presented (Reed v. Ross, No. 83-218 (June 27, 1984), slip op. 15;
compare Davis v. United States, 411 U.S. 233, 243-244 (1973), and
Shotwell Mfg. Co. v. United States, 371 U.S. 341, 363 (1963)), or
perhaps that there was a unique feature of the procedural system or the
posture of the case that would have made the cost of avoiding the
procedural default unacceptably high under the circumstances (Clay v.
Director, Juvenile Division, Department of Correction, 749 F.2d 427,
434-435 (7th Cir. 1984)(Posner, J. concurring); cf. Sykes, 433 U.S. at
83 (quoting Fay v. Noia, 372 U.S. at 440)(noting the defendant's "grisly
choice" between accepting a life sentence and pursuing an appeal that
might have resulted in a death sentence); see also Riner v. Owens, 764
F.2d 1253, 1257 (7th Cir. 1985).
There perhaps are some other, exceptional circumstances in which
"cause" would be found for a procedural default, consistent with the
Court's assurance that the "cause and prejudice" standard will be
applied in such a way as to prevent a "miscarriage of justice" (Sykes,
433 U.S. at 91). But in providing this assurance the Court surely did
not mean to embrace a rule such as adopted by the Fourth Circuit here,
which would almost wholly vitiate the enforceability of procedural
defaults. Rather, the focus of the inquiry under the principles
announced in Sykes should remain whether there was a defect in the trial
or appellate process that brought about the default or some other
external impediment to developing or presenting the claim.
c. It would, moreover, be particularly inappropriate to reintroduce
the "knowing and deliberate bypass" standard in the context of counsel's
actions. As explained above, the deliberate bypass rule of Fay v. Noia
was drawn directly from the standard for finding a valid waiver of those
constitutional rights that may be lost only by an affirmative waiver.
Such rights are personal to the accused and therefore cannot generally
be waived without the personal participation of the accused. See 372
U.S. at 439. However, the constitutional rights on which counsel might
base an objection at trial or on appeal are not personal to the
attorney. A rule under which "cause" would be found for a procedural
default whenever counsel did not make a deliberate choice to forgo
proper assertion of the claim therefore has no logical nexus to the
standards for waiver of constitutional rights on which even the
deliberate bypass rule of Fay v. Noia was based. See Sykes, 433 U.S. at
93-94 (Burger, C.J., concurring). Accordingly if, as the Court held in
Sykes, a knowing and deliberate decision by the accused no longer is
required in order for him to be bound by a procedural default (in the
absence of some other sufficient "cause" for that default), it would be
perverse to require that the default nevertheless must have been the
product of a knowing and deliberate decision by counsel, as the court of
appeals held.
The Fourth Circuit's rule also distorts the relationship between the
accused and his attorney. The Sixth Amendment "speaks of the
'assistance' of counsel, and an assistant, however expert, is still an
assistant." Faretta v. California, 422 U.S. 806, 820 (1975). To be
sure, when the defendant chooses to have his lawyer present his case,
law and tradition allocate to the lawyer the power to make binding
decisions on all but such fundamental matters as whether to plead
quilty, waive a jury, testify, or take an appeal. See Jones v. Barnes,
463 U.S. 745, 751 (1983). The lawyer makes those decisions, however,
not in his own right, but as the representative of the accused. The
defense the lawyer presents therefore is, in contemplation of law, that
of the accused personally (Faretta, 422 U.S. at 820-821), who is bound
by counsel's conduct at trial or on appeal "except where 'the
circumstances are exceptional'" (Sykes, 433 U.S. at 91 n.14 (quoting
Henry v. Mississippi, 379 U.S. 443, 451 (1965)); see also 433 U.S. at
93-94 (Burger, C.J., concurring); id. at 94-95 (Stevens,J., concurring;
Estelle v. Williams, 425 U.S. 501, 515 n.4(1976) (Powell, J.,
concurring)). Because defense counsel acts on behalf of his client, and
because our criminal justice system does not distinguish between the
defendant and his attorney insofar as compliance with procedural rules
is concerned, there is no legal or logical basis under Sykes for a rule
that distinguishes between the defendant and his attorney with respect
to whether actual knowledge and deliberate decision is required in order
to enforce the consequences of a failure to comply with such rules.
2. For the reasons just discussed, the Court's holding in Engle that
counsel's unawareness of a legal claim that was available at the time of
trial or direct appeal does not constitute "cause" for a procedural
default was firmly rooted in the reasoning and purposes of Sykes itself.
As we now demonstrate, the soundness of that conclusion is reinforced
by reference to the policy considerations that animated the Court's
substitution of the "cause and prejudice" standard for the "deliberate
bypass" test of Fay v. Noia.
a. Sykes identified four substantial reasons for according
contemporaneous objection rules at trial a "greater respect" (433 U.S.
at 88) than they received under the deliberate bypass standard. First,
timely objection may either avert error or at least permit prompt
corrective measures and the prevention of further error, "thereby making
a major contribution to finality in in criminal litigation" (ibid.).
Second, when an individual is brought to court to stand trial,
"society's resources have been concentrated at that time and place in
order to decide, within the limits of human fallibility, the question of
quilt or innocence of one of its citizens"(id, at 90); "any procedural
rule which encourages the result that those proceedings be as free of
error as possible is thoroughly desirable, and the
contemporaneous-objection rule surely falls within this classification"
(ibid.). Third, "the rule of Fay v. Noia, broadly stated, may encourage
'sandbagging' on the part of defense lawyers, who may take their chances
on a verdict of not guilty * * * with the intent to raise their
constitutional claims in a federal habeas court if their initial gamble
does not pay off" (id, at 89); that collateral attack might occur at a
time when successful reprosecution would be more difficult or even
impossible. Fourth, if factual findings are necessary, "(a)
contemporaneous objection enables the record to be made with tespect to
the constitutional claim when the recollections of witnesses are
freshest, not years later in a federal habeas proceeding" (id, at 88).
See also Engle, 456 U.S. at 126-131.
In Reed v. Ross, the Court recognized that similar considerations of
finality, efficiency, and the facilitation of prompt retrials apply with
respect to procedural rules, such as the Virginia rule in this case,
that provide for the forfeiture of claims not raised on direct appeal.
The Court accordingly held that the "cause and prejudice" standard
applies to appellate defaults as well as trial defaults. Slip op. 9-10.
See also Leroy v. Marshall, 757 F.2d 94, 95 n.1 (6th Cir. 1985)
(collecting cases).
With the exception of the "sandbagging" factor, the concerns
identified in Sykes are fully applicable irrespective of whether the
procedural default was due to a deliberate decision by counsel or to his
ignorance or inadvertence. In either event, "(t)he defendant's counsel,
for whatever reasons, has detracted from the trial's (or appeal's)
significance by neglecting to raise a claim in that forum." Engle, 456
U.S. at 128-129. Moreover, although sandbagging in itself constitutes a
particular affront to orderly procedures because of its deliberate
nature, the concrete consequences of sandbagging (piecemeal review and
the prospect of a retrial at a much later date) can result just as
readily where the procedural default was due to counsel's unawareness.
Compare United States v. Agurs, 427 U.S. 97, 110 (1976).
In addition, there would be substantial costs associated with
attempting to identify those circumstances in which there actually had
been sandbagging or some other deliberate decision by counsel not to
raise the claim at the time required. The court of appeals contemplated
(Pet. App. 14-15) that resolution of that issue would require an inquiry
into counsel's "motivation." This in turn presumably would require an
evidentiary hearing, often long after trial, including testimony by
counsel (if he is still available and has a recollection of the
proceedings) regarding his state of mind and tactical judgments.
This Court has recognized the potential adverse consequences for the
independence of counsel and the functioning of the criminal justice
system that may result from such intrusive post-trial inquiry into
counsel's performance. See Strickland v. Washington, slip op. 19-20,
27. No doubt such an inquiry is sometimes inevitable under Strickland
when a defendant directly challenges his attorney's actions by bringing
an independent, constitutionally based claim of ineffective assistance
of counsel. But the Court should hesitate before it adopts a
formulation of the "cause" standard that would require an inquiry into
counsel's motivation where the attorney's performance is not the
ultimate focus of the challenge. These considerations weigh heavily
against any rule that makes the existence of "cause" depend on the
subjective knowledge and intent of counsel.
The Court did observe in Reed v. Ross that "the cause requirement may
be satisfied under certain circumstances when a procedural failure is
not attributable to an intentional decision by counsel made in pursuit
of his client's interests" (slip op. 12-13). The "certain
circumstances" with which Reed v. Ross itself was concerned involved a
constitutional claim that was novel at the time of trial but established
as meritorious by the time of the collateral attack proceeding, and the
Court held that "cause" exists in such a case if counsel had no
"reasonable basis" upon which to formulate the constitutional claim and
it therefore was not "reasonably available" to him. Id. at 13-15.
Resolution of that question turns not on the state of mind of the
attorney, but on the state of the law at the time of trial, which is an
objective external factor. See id. at 15-18. Compare Harlow v.
Fitzgerald, 457 U.S. 800, 815-819 (1982). The Court in Reed v. Ross did
not question the holding in Engle that counsel's actual unawareness of a
legal claim that was reasonably available at the time of trial does not
constitute "cause."
Nor does the Court's reasoning in Reed v. Ross support the court of
appeals' holding in this case. First, the Court observed that in the
case of a novel claim "it is safe to assume that (counsel) is
sufficiently unaware of the question's latent existence that we cannot
attribute to him strategic motives of any kind." Slip op. 13. A court
obviously cannot have the same confidence where a habeas petitioner
asserts that his attorney simply overlooked or was unaware of the legal
basis for a claim that was reasonably available at the time of trial and
had been identified and raised by other lawyers. See Engle, 456 U.S. at
133-134. Indeed, as Judge Hall observed in his dissent (Pet. App.
16-18), there is every reason to believe that counsel's actions in this
case were deliberate. The fact that counsel requested disclosure of the
statements at trial and initially identified the discovery issue in the
notice of appeal strongly suggests that there subsequently was a
conscious choice to abandon that issue in order to concentrate on what
might have seemed to be more promising claims. See Jones v. Barnes, 463
U.S. at 751-754. It is hard to fault this judgment. Although
respondent and the court of appeals now characterize the issue as a
Brady claim, counsel's request at trial appears to have been a
nonconstitutionally based discovery request relying on the rationale of
Jencks (see Pet. App. 17 (Hall, J., dissenting)), and his statement of
the issue in the notice of appeal was not framed in constitutional terms
(id. at 5). Counsel might have concluded that the prospects of success
on the discovery issue were not promising under state law. This is
doubly so viewing the matter in Brady terms; counsel could well have
concluded that the witness statements were unlikely to include any
evidence that was material and favorable to respondent, especially since
the trial court had reviewed the statements in camera prior to trial and
concluded that they contained nothing exculpatory.
Second, the Court in Reed v. Ross explained that a rule requiring
counsel to raise a novel question would serve little functional purpose,
because the novelty of the claim would probably lead the court to reject
it. Slip op. 13-14. By contrast, where a claim is not novel, there is
every reason to believe that the court may be receptive if the claim is
in fact meritorious. A rule that does not excuse a procedural default
based on mere inadvertence or ignorance reinforces counsel's duty and
incentives to be vigilant at the trial itself and on direct appeal in
bringing any such issues to the court's attention. /13/
b. The countervailing considerations that also "have come to be
embodied in the 'cause and prejudice' requirement" (Reed v. Ross, slip
op. 9) likewise do not support the court of appeals' holding. Those
considerations are (1) the prisoner's interest in having a federal forum
for the presentation of his claim (id. at 8) and (2) the role of the
writ of habeas corpus as "a bulwark against convictions that violate
'fundamental fairness'" (Engle, 456 U.S. at 126, quoting Sykes, 433 U.S.
at 97 (Stevens, J., concurring)).
Acknowledgment of a state defendant's abstract interest in presenting
his claim to a federal forum if it fails in state court offers little
guidance in giving content to the "cause" standard. That interest is
certainly exceedingly limited, if not nonexistent, when the defendant
was not sufficiently interested in his federal constitutional claim even
to raise or preserve it during the state proceedings. In any event, it
seems most unlikely that the prisoner's interest in presenting a
defaulted claim to a federal forum is much affected by whether his
lawyer's failure to raise and preserve the claim in state court was due
to a deliberate tactical decision or to ignorance or inadvertence.
That distinction also has no bearing on the fundamental fairness of
the conviction -- at least where, as the court of appeals necessarily
assumed in this case (Pet. App. 9), the attorney's performance satisfied
constitutional standards of effective assistance of counsel. This
conclusion is demonstrated by positing a situation involving two
defendants convicted at a joint trial in which an error occurred that
equally affected both. Suppose the attorney for one defendant
recognized the error and considered raising an objection but declined to
do so for tactical reasons, while the attorney for the other either
failed to notice the error or was unaware of the legal basis for an
objection. If actual prejudice could be shown to have resulted from the
error, then under the Fourth Circuit's holding, the second defendant
would be entitled to federal habeas relief but the first would not. Yet
it cannot seriously be contended that the fundamental fairness of the
two convictions is significantly different -- or that one defendant, but
not the other" is suffering from a "fundamentally unjust incarceration"
(Engle, 456 U.S. at 135) entitling him to access to a federal form --
simply because their attorneys had different subjective states of mind
when the error occurred at trial. This Court has stressed that the
definition of the terms "cause" and "prejudice" must be informed by the
role of the writ of habeas corpus in preventing miscarriages of justice.
Engle, 456 U.S. at 135; Sykes, 433 U.S. at 91. Because counsel's
state of mind has no direct relation to whether a miscarriage of justice
has occurred, the existence of "cause" should not turn on that factor.
This does not mean, of course, that the accused in a criminal
prosecution has no protection against errors by his attorney. In
Strickland v. Washington, this Court for the first time announced
standards to govern the determination whether counsel has rendered the
effective assistance required by the Sixth Amendment. Compare Sykes,
433 U.S. at 117-118 (Brennan, J., dissenting). If the defendant can
show that counsel's performance, including his failure to raise or
preserve an objection, was not within the range of reasonable competence
demanded of attorneys in criminal cases (Strickland, slip op. 17-21) and
that there is a reasonable probability that any errors by counsel
affected the outcome of the trial (id. at 24), he is entitled to habeas
relief on an independent claim of ineffective assistance of counsel.
But if counsel's performance was constitutionally effective under these
standards, the trial must be regarded as having been fundamentally fair
insofar as the role of counsel is concerned. Id. at 14-17, 19, 23-24,
26, 30; United States v. Cronic, No. 82-660 (May 14, 1984), slip op.
4-9. In that event, the habeas court should look elsewhere than to
counsel's performance in ascertaining whether the conviction was
fundamentally fair. Moreover, any attempt to identify the types of
errors by counsel that do not rise to the level of constitutionally
ineffective assistance but nevertheless do amount to "cause" for a
procedural default would lead to the very confusion concerning the
appropriate standards for evaluating attorney performance that this
Court's decision in Strickland had served to eliminate.
For the foregoing reasons, the Court should adhere to its view in
Engle that counsel's overlooking of a claim or his unawareness of the
legal basis for the claim does not constitute "cause" for a procedural
default, at least if the defendant received the effective assistance of
counsel that the Constitution guarantees. /14/
CONCLUSION
The judgment of the court of appeals should be reversed.
Respectfully submitted.
CHARLES FRIED
Acting Solicitor General
STEPHEN S. TROTT
Assistant Attorney General
ANDREW L. FREY
Deputy Solicitor General
EDWIN S. KNEEDLER
Assistant to the Solicitor General
KATHLEEN A. FELTON
Attorney
SEPTEMBER 1985
/1/ Under Virginia law, errors that could have been but were not
raised on appeal generally may not be raised for the first time on
collateral attack, unless there is a showing of ineffective assistance
of counsel in failing to raise the issue on appeal. See Slayton v.
Parrigan, 215 Va. 27, 29, 205 S.E.2d 680, 682 (1974), cert. denied, 419
U.S. 1108 (1975).
/2/ Of course, the Court made clear in Brady that the evidence must
be not only "material," but also "favorable to (the) accused" (373 U.S.
at 87). See also United States v. Bagley, No. 84-48 (July 2, 1985),
slip op. 6. There is no duty to disclose inculpatory evidence.
Weatherford v. Bursey, 429 U.S. 545, 559-560 (1977).
/3/ As this Court made clear in Evitts v. Lucey, No. 83-1378 (Jan.
21, 1985), slip op. 6, 11, 12-18, a claim of ineffective assistance of
appellate counsel actually arises under the Due Process Clause, not the
Sixth Amendment.
/4/ In remanding, the court rejected the contention that respondent
first should be required to exhaust his state remedies with respect to
his claim of attorney error on direct appeal. The court explained that
respondent relied on that error only to establish "cause" under
Wainwright v. Sykes for his failure to preserve the Brady issue on
appeal, not to establish an independent claim of ineffective assistance
of counsel, and that exhaustion of state remedies is not required on the
question of "cause" for a procedural default (Pet. App. 14).
/5/ Stirickland v. Washington, No. 82-1554 (May 14, 1984), slip op.
17-21.
/6/ Evitts v. Lucey, No. 83-1378 (Jan. 21, 1985), slip op. 6, 11,
12-18.
/7/ Since a non-meritorious claim would not survive the "prejudice"
prong of the "cause and prejudice" test, the "cause" prong is critical
to enforcing a default only in the case of claims that would have been
upheld had they been properly preserved.
/8/ In Engle, respondent Isaac argued (Br. 3-4, 8-10, 12) that trial
counsel had never considered or perceived the claim; respondent Hughes
argued (Br. 14) that there had been no "deliberate" or "strategic"
decision by counsel to build reversible error into the case; and
respondent Bell argued (Br. 49-52) that inadvertence, ignorance or
neglect by counsel, as distinguished from a tactical or strategic
decision, should constitute cause.
/9/ The court of appeals' only citation to Engle was in connection
with its discussion of the distinct question of exhaustion of state
remedies. See Pet. App. 14 n.8.
/10/ See Cantone v. Superintendent, New York Correctional Facility,
759 F.2d 207, 218 (2d Cir. 1985); Tsirizotakis v. LeFevre, 736 F.2d 57,
62 (2d Cir. 1984), cert. denied, No. 84-5305 (Oct. 1, 1984); United
States ex rel. Caruso v. Zelinsky, 689 F.2d 435, 443-444 (3d Cir. 1982);
Wiggins v. Procunier, 753 F.2d 1318, 1321-1322 (5th Cir. 1985); Weaver
v. McKaskle, 733 F.2d 1103, 1106 (5th Cir. 1984); Long v. McKeen, 722
F.2d 286, 288-289 (6th Cir. 1983), cert. denied, 465 U.S. 1106 (1984);
Webster v. Engle, 721 F.2d 566, 568 (6th Cir. 1983), Hockenbury v.
Sowders, 718 F.2d 155, 160 (6th Cir. 1983), cert. denied, No. 83-6218
(May 14, 1984); Williams v. Nix, 751 F.2d 956, 960 (8th Cir. 1985);
Thomas v. Auger, 738 F.2d 936, 939 (8th Cir. 1984); Dietz v. Solem, 677
F.2d 672, 673, 675 (8th Cir. 1982); Matias v. Oshiro, 683 F.2d 318, 321
n.3 (9th Cir. 1982); Baumann v. United States, 692 F.2d 565, 572 (9th
Cir. 1982); Palmes v. Wainwright, 725 F.2d 1511, 1525 (11th Cir. 1984);
Spencer v. Zant, 715 F.2d 1562, 1575 (11th Cir. 1983); Birt v.
Montgomery, 709 F.2d 690, 699 (1983), on rehearing, 725 F.2d 587, 597
(11th Cir. 1984) (en banc), cert. denied, No. 83-6830 (Oct. 1, 1984).
Similarly, In Runnels v. Hess, 713 F.2d 596 (1983), the Tenth Circuit
superseded its pre-Engle decision in the same case (653 F.2d 1359
(1981)) upon which the Fourth Circuit had relied in the instant case
(Pet. 11, 15). See also Clay v. Director, Juvenile Division, Department
of Corrections, 749 F.2d 427, 430-431, 433 n.4 (7th Cir. 1984)
(reserving the question whether attorney error short of ineffective
assistance can constitute "cause," and finding "cause" because counsel
rendered ineffective assistance under Sixth Amendment standards).
/11/ The court of appeals in this case relied upon pre-Engle
decisions of the Fifth, Sixth, Eighth, Ninth and Tenth Circuits. See
Pet. App. 10-12, 15. However, as indicated by the citations in note 10,
supra, each of these courts subsequently discarded its pre-Engle view;
the Fourth Circuit failed to take note of those developments in either
its panel or en banc opinion. Indeed, even the Fourth Circuit
previously had appeared to recognize that Engle rejected the very
rationale adopted in this case. See Honeycutt v. Mahoney, 698 F.2d 213
(1983); Wilson v. Procunier, 747 F.2d 251, 252 (1984), cert. denied,
No. 84-5852 (Feb. 19, 1985).
/12/ Similarly, in Tollett v. Henderson, 411 U.S. 258 (1973), decided
the same day as Davis, the Court held that the habeas petitioner had
forfeited his right to object to infirmities in the selection of the
grand jury even though the relevant facts were found "to have been
unknown to both respondent and his attorney" at the time of trial (411
U.S. at 266). See also id. at 273-276 (Marshall, J., dissenting);
Estelle v. Williams, 425 U.S. 501, 526 (1976) (Brennan, J., dissenting);
Francis v. Henderson, 425 U.S. at 549 n.2, 550 (Brennan, J.,
dissenting).
/13/ The Court also was concerned in Reed v. Ross that a holding that
required counsel to raise even claims having a novel legal basis would
lead to too many objections by counsel at trial or to the cluttering of
appellate briefs with far-fetched claims, in order to avoid procedural
defaults. Slip op. 14. This concern likewise has no application where
there is a reasonable basis for the argument in existing law.
/14/ Because the court of appeals in this case assumed that
respondent's attorney did not render ineffective assistance of counsel,
there is no occasion for the Court to decide in this case whether
counsel's failure to raise or preserve an objection constitutes "cause"
where his omission was sufficiently egregious to amount to ineffective
assistance of counsel under Sixth Amendment standards. Some courts of
appeals have held or suggested that "cause" would be established in such
circumstances. See, e.g., Clay v. Director, Juvenile Division,
Department of Corrections, 749 F.2d at 430-431; Tsirizotakis v.
LeFevre, 736 F.2d at 62; Birt v. Montgomery, 709 F.2d at 699; Baumann
v. United States, 692 F.2d at 572; United States ex rel. Caruso v.
Zelinsky, 689 F.2d at 443-444. The Fifth Circuit has held to the
contrary. Wiggins v. Procunier, 753 F.2d at 1321-1322; Weaver v.
McKaskle, 733 F.2d at 1106. The outcome should be the same under either
approach, because the standard for evaluating the attorney's performance
would be the same for purposes of showing "cause" as it would be for
purposes of establishing a Sixth Amendment claim of ineffective
assistance of counsel, and because the habeas petitioner must establish
actual prejudice whether he is attempting to overcome a procedural
default caused by counsel or seeking relief on an independent claim of
ineffective assistance of counsel.
WISCONSIN DEPARTMENT OF INDUSTRY, LABOR AND HUMAN RELATIONS, ET AL.,
APPELLANTS V. GOULD INC.
No. 84-1484
In the Supreme Court of the United States
October Term, 1985
On Appeal From The United States Court Of Appeals For The Seventh
Circuit
Brief For The National Labor Relations Board As Amicus Curiae
Supporting Appellee
TABLE OF CONTENTS
Question Presented
Interest of the National Labor Relations Board
Statement
Introduction and summary of argument
Argument:
The National Labor Relations Act preempts Wisconsin's
statutory debarment from state contracts of employers that
have violated the Act
A. States may not provide their own sanctions for
conduct that violates the Act
B. Wisconsin's statutory boycott of federal labor law
violators was intended to be and operates as a penalty for
unfair labor practices
C. The penalty imposed by Wisconsin is preempted by the
National Labor Relations Act.
QUESTION PRESENTED
Whether the National Labor Relations Act, 29 U.S.C. 151 et seq.,
preempts a state statute debarring employers from doing business with
the State if they have been found by courts of appeals to have violated
the Act three times within five years.
INTEREST OF THE NATIONAL LABOR RELATIONS BOARD
The question presented in this case is whether the National Labor
Relations Act preempts a state law debarring an employer who has
violated the Act from doing business with the State. Relying on San
Diego Building Trades Council v. Garmon, 359 U.S. 236 (1959), and its
progeny, the court of appeals held that the state statute is preempted
by the Act and therefore invalid under the Supremacy Clause because it
impermissibly supplements the remedial scheme established by Congress in
the Act. Formulating and implementing appropriate remedies for
violations of the Act are central to the mission of the National Labor
Relations Board. The Board accordingly has a significant interest in
the proper application to this case of the preemption doctrine, which is
designed to preclude interference with the uniform administration of
federal law by conflicting state rules and rememdies.
STATEMENT
1. The State of Wisconsin provides by statute that persons or firms
that have been found by orders of the National Labor Relations Board
enforced by courts of appeals to have violated the National Labor
Relations Act (NLRA), 29 U.S.C. 151 et seq., in three separate cases
within a five-year period shall be placed on a list of labor law
violators for three years. Wis. Stat. Ann. Section 101.245 (West Supp.
1984-1985). Persons or firms found in contempt of court for "failure to
correct" three separate violations of the Act within five years must
also be placed on the list. Ibid. The State "shall not purchase any
product known to be manufactured or sold by any person or firm included
on the list of labor law violators." Id. Section 16.75(8).
In 1982, the State placed appellee Gould Inc. on the list of
violators and announced that two contracts between Gould and the State
would be cancelled as soon as practicable and that the firm would not be
awarded any new state contracts until 1985. The State also notified all
of its current and potential vendors that they could not sell to the
State any product that contained components produced by Gould. Gould, a
Delaware corporation with its prinicpal place of business in Illinois,
was placed on the list because of the enforcement of four Board orders
against various divisions of the company, none of which were located in
Wisconsin. Gould no longer owned any of these divisions at the time
that it was placed on the list. J.S. App. 6-7, 50-53.
2. Gould filed this action in the United States District Court for
the Western District of Wisconsin, seeking injunctive and declaratory
relief with respect to the State's enforcement of its debarment statute
on the ground that the state law was preempted by the NLRA and therefore
invalid under the Supremacy Clause. /1/ On cross-motions for summary
judgment, the district court ruled that the Wisconsin statute is
preempted and permanently enjoined the State (through its officials)
from enforcing it against Gould (J.S. App. 45-83). /2/
The district court reasoned (J.S. App. 56-66) that whether the
Wisconsin debarment statute is preempted depends on a balancing of
relevant state and federal interests. The court rejected (id. at 66-74
& n.7) the State's assertions that its proprietary purchasing function
was furthered by the statute and that it was acting purely as a market
participant (id. at 72-73 (emphasis deleted)):
(The statute) is directed specifically towards effecting (sic) the
labor relations of those employers who are adjudicated federal
labor law violators. Defendants have not suggested that Sections
101.245 and 16.75(8) have any legitimate nonlabor purpose * * *.
It may be that the exercise of (the State's) proprietary
function is deeply rooted in tradition. However, in this
instance, the exercise of that function has the wholly
illegitimate purpose of discouraging federal labor law violations
by employers.
Indeed, the district court noted (id. at 73 (emphasis and citation
omitted; brackets in original)) that the State had conceded that it was
"'merely seeking to influence private conduct already prescribed (sic)
by federal labor policy.'" The "minimal" (id. at 74 (emphasis omitted))
legitimate state concerns were, in the district court's view, outweighed
by significant federal interests, such as the policy of uniform federal
remedies that were impinged by the law (id. at 74-82), and therefore the
debarment statute was preempted by the NLRA.
3. The court of appeals affirmed the district court's invalidation of
the Wisconsin statute (J.S. App. 1-44). /3/ Relying on San Diego
Building Trades Council v. Garmon, 359 U.S. 236 (1959), and its progeny
the court of appeals concluded (J.S. App. 8-21) that the statute was
preempted because it impermissibly supplemented the remedial scheme
established by Congress and the Board under the NLRA. Although the
debarment statute reflected the "laudable purpose" (id. at 14) of
promoting compliance with the NLRA, the State's decision to "bas(e) a
boycott on federal labor law violations, regardless of where the
violations occurred and whether they have been rectified, clearly
interfere(d)," in the court of appeals' view (id. at 15), with the
federal regulatory scheme. Wisconsin's "impos(ition) (of) a much
harsher sanction than Congress intended" (id. at 26-27) for federal
labor law violations, along with the possibility of other states
adopting debarment statutes of varying scope and operation (id. at 21),
demonstrated that "(a)llowing the states indiscriminate punishment of
recidivism impermissibly intrudes on the federal scheme of regulation"
(id. at 20 n.7).
The court of appeals rejected the State's argument that it was
legitimately furthering local policy as a market participant, not
regulating conduct within the purview of the NLRA. Because "the policy
the (state) law promotes is not efficient use of state funds but the
intent to effect compliance with the NLRA" (J.S. App. 31), the court of
appeals concluded that Wisconsin "simply is not functioning as a private
purchaser of services" (id. at 30), but rather is "actually regulat(ing)
employers' conduct" (id. at 27 (footnote omitted)). Even if Wisconsin
were acting as a market participant, the court of appeals distinguished
cases holding that such conduct does not run afoul of the "dormant"
Commerce Clause, because here "Congress has legislated decisively in the
area" (id. at 22).
INTRODUCTION AND SUMMARY OF ARGUMENT
It has long been settled that states may not provide their own
remedies for conduct that violates the National Labor Relations Act.
See, e.g., San Diego Building Trades Council v. Garmon, 359 U.S. 236
(1959); Garner v. Teamsters Local Union No. 776, 346 U.S. 485 (1953).
As the Court explained in Amalgamated Ass'n of St., Elec. Ry. & Motor
Coach Employees v. Lockridge, 403 U.S. 274, 287 (1971), such state
remedies must be preempted because the "nature of those remedies that
are and are not available is a fundamental part and parcel of the
operative legal system established by the National Labor Relations Act."
State sanctions for unfair labor practices encroach on the Board's
responsibility to develop and apply a unified remedial scheme that
effectuates the policies of the Act.
Wisconsin's debarment law was intended as a penality against
violators of the federal Act in order to deter future unfair labor
practices. Moreover, the state law clearly operates as such a penalty
by preventing employers from competing for the State's business. This
conclusion is reinforced by the substantial purchasing power of the
State: it blinks reality to suggest that the objects of the State's
boycott are not economically harmed because they may take their business
elsewhere.
The conduct at which the Wisconsin statute is directed -- actual
unfair labor practices -- is plainly within the core of Congress's
concerns in the NLRA. State penalties for such conduct are improper not
only because they are uninvited additions to what Congress envisioned as
a scheme developed and implemented by a single expert agency but also
because they are substantially at odds with the remedial scheme devised
by Congress. Under that scheme, the Board cannot impose flat penalties,
but rather is empowered to remedy unfair labor practices by restoring
the status quo ante and prohibiting the commission of similar practices
in the future. The Board's remedies are designed to effectuate
Congress's purpose to restore (insofar as possible) the flow of commerce
once a violation has been remedied; Wisconsin's law stands as an
obstacle to achieving that statutory purpose. /4/ Significantly,
Congress has refused to enact a federal contract debarment scheme, even
one carefully tailored, as Wisconsin's is not, to effectuate the
purposes of the NLRA.
The Wisconsin statute is not directed solely at matters of deep local
concern: to the contrary, it applies to unfair labor practices anywhere
in the country. None of Gould's violations, for example, took place in
Wisconsin. There is therefore no warrant for excepting the debarment
statute from the usual operation of the preemption doctrine. In any
event, this exception from the normal presumption of preemption was
crafted by the Court for a limited number of traditional local concerns
such as public safety and common law rights. It was never intended to
be available for invocation in the wholly conclusory fashion proposed by
the State in this case. Indeed, whenever Wisconsin has undertaken to
explain what its "deep local concern" might be, it has inevitably spoken
of a concern to affect labor relations, and that of course it may not
seek to do. That is precisely the subject of concern as to which the
federal statutory scheme is self-contained and exclusive.
The State seeks to avoid preemption by urging that it is acting
merely as a "market participant" as that term is used in "dormant"
Commerce Clause cases where Congress has not legislated. The very
triggering mechanism of the state law, however, shows that Wisconsin is
not acting as a market participant seeking to advance local economic
concerns, but rather as a regulator punishing conduct proscribed by
federal law. Moreover, even if the State's statute could be regarded
solely as market participation, the cases on which the State relies are
inapposite, because here Congress has legislated with preemptive effect.
The market participation doctrine is applicable to dormant Commerce
Clause cases only; it has never been invoked in the myriad situations
where state laws may conflict with some actual exercise of congressional
power -- such as designating an article of commerce as contraband which
neither the states nor other market participants may lawfully sell or
purchase. It is only in the dormant Commerce Clause cases that Congress
has not used its constitutional powers to define any national policy and
so a state, like any other "market participant," may participate on
whatever terms it chooses.
Here, however, Congress has expressed a comprehensive and detailed
policy on the subject matter of labor relations and it is the elementary
meaning of the Supremacy Clause that all levels of government are bound
by that policy. Of course private parties need not share that policy;
they may have quite different labor policies, whatever it may mean for a
private party to have a "policy" at all. But then private parties,
unlike the states, are not bound by the Supremacy Clause, but only by
the actual provisions of the law itself.
ARGUMENT
THE NATIONAL LABOR RELATIONS ACT PREEMPTS WISCONSIN'S STATUTORY
DEBARMENT FROM STATE CONTRACTS OF EMPLOYERS THAT HAVE VIOLATED THE ACT
The State of Wisconsin has acted to penalize employers that have
violated the federal labor law, in the hope of deterring future unfair
labor practices. The goal is commendable. Yet Congress intended that
remedies for violations of the NLRA be developed and administered by a
centralized expert agency, the National Labor Relations Board. Only in
this way can its purpose of promoting the flow of commerce and thus
benefiting employers and employees alike be achieved. For this reason,
the Court has long recognized that state attempts to supplement the
remedies for unfair labor practices applied by the Board are preempted
by the Act. While the state-imposed penalty here may be a novel one,
its purpose and effect are clear, as are the principles that apply.
Wisconsin's debarment statute is preempted by the NLRA and is therefore
invalid under the Supremacy Clause.
A. States May Not Provide Their Own Sanctions For Conduct That Violates
The Act
A repeated theme of the Court's labor law preemption decisions has
been that a state may not provide its own remedies for conduct that
actually or arguably constitutes an unfair labor practice under the
NLRA. /5/ More than 30 years ago, the Court, in holding a state remedy
preempted, observed that "when two separate remedies are brought to bear
on the same activity, a conflict is imminent." Garner v. Teamsters Local
Union No. 776, 346 U.S. 485, 498, 499 (1953). In San Diego Building
Trades Council v. Garmon, 359 U.S. 236 (1959), the seminal labor
preemption decision, the Court was "concerned with conflict in its
broadest sense; conflict with a complex and interrelated federal scheme
of law, remedy, and administration" (id at 243 (emphasis added)). Where
an employer's conduct actually violates the Act, "then the matter is at
an end, and the States are ousted of all jurisdiction" (id. at 245).
State remedies that supplement the NLRB's are preempted just as surely
as those that could be provided by the Board (id. at 247):
(S)ince remedies form an ingredient of any integrated scheme of
regulation, to allow the State to grant a remedy here which has
been withheld from the National Labor Relations Board only
accentuates the danger of conflict.
The Court amplified on this reasoning in reaffirming Garmon in
Amalgamated Ass'n of St., Elec. Ry. & Motor Coach Employees v.
Lockridge, 403 U.S. 274 (1971). The Court stated first that preemption
is not designed "merely to avoid disparity in the content of
proscriptive behavioral rules" (id at 287). Because "simple congruity
of legal rules does not in this area, prove the absence of untenable
conflict" (id. 290), the "potential for debilitating conflict" (id. at
289) is present even when a state applies the same rules as the NLRB
does. Remedies provided by states for conduct that violates the NLRA
conflict with Congress's purpose of "committ(ing) enforcement of the
(Act) * * * to a centralized (federal) agency" (id. at 288-289 n.5).
Preemption of state remedies is founded on the fact that the "nature of
those remedies that are and are not available is a fundamental part and
parcel of the operative legal system established by the National Labor
Relations Act" (id. at 287). See also, e.g., Local 926, Int'l Union of
Operating Eng'rs v. Jones, 460 U.S. 669, 684 (1983) (claim that
additional state remedies are not preempted "was squarely rejected" in
Garmon); Farmer v. United Bhd. of Carpenters, Local 25, 430 U.S. 290,
305-307 (1977) (damages cannot be awarded for aspect of tortious conduct
that would consititute an unfair labor practice).
The reason for concern with state remedies for conduct that violates
the NLRA is plain: Congress has committed to the Board "the primary
responsibility and broad discretion to devise remedies that effectuate
the policies of the Act." Sure-Tan, Inc. v. NLRB, No. 82-945 (June 25,
1984), slip op. 14; see, e.g., Phelps Dodge Corp. v. NLRB, 313 U.S.
177, 194 (1941) (Congress has "entrust(ed) to an expert agency the
maintenance and promotion of industrial peace"). Developing and
applying remedies to eliminate the "industrial strife" that has caused
"obstructions to the free flow of commerce" (Section 1, 29 U.S.C. 151)
is "peculiarly a matter for administrative competence" (Phelps Dodge
Corp., 313 U.S. at 194). Accordingly, the Board's remedies are "subject
to (only) limited judicial review" (Fibreboard Paper Prods. Corp. v.
NLRB, 379 U.S. 203, 216 (1964)). Only a unified system of remedies
applied by the Board can "give coordinated effect to the policies of the
Act." NLRB v. Seven-Up Bottling Co., 344 U.S. 344, 348 (1953).
The Board is, as the Court stated in NLRB v. Nash-Finch Co., 404 U.S.
138, 145 (1971) (footnote omitted; emphasis added), "the sole protector
of the 'national interest' defined with particularity in the Act." Laws
such as Wisconsin's are at odds with the congressional judgment that the
Board alone has the expertise to determine what measures to take in
response to unfair labor practices and that the policies of the Act are
best effectuated by avoiding a diversity of local procedures for dealing
with such conduct. Thus, even were it not the case that, as we show
below (pages 16-18). Wisconsin's law is inconsistent with the
essentially remedial nature of the scheme that the Board administers,
the State's effort to implement its own local "anti-recidivist" policy
frustrates the intent of Congress. The problem is, of course,
compounded by the prospect of other states enacting their own variations
on Wisconsin's law (see note 8, infra), thereby creating a pastiche of
state penalties imposed on top of the Board's remedial scheme. See
Lockridge, 403 U.S. at 286 (footnote omitted) (a primary factor in "the
enactment of a comprehensive national labor law, entrusted for its
administration and development to a centralized, expert agency * * * was
the perceived incapacity of common-law courts and state legislatures,
acting alone, to provide an informed and coherent basis for stabilizing
labor relations conflict and for equitably and delicately structuring
the balance of power among competing forces so as to further the common
good").
B. Wisconsin's Statutory Boycott Of Federal Labor Law Violators Was
Intended To Be And Operates As A Penalty For Unfair Labor Practices
1. The Wisconsin debarment statute penalizes a class of persons
defined solely in terms of their commission of unfair labor practices
proscribed by the NLRA. The State candidly admitted in the district
court that its law is intended to deter violations of the federal
statute (J.S. App. 73 (emphasis and brackets in original), quoting
Defendants' Br. 12, 18):
"The state has a deeply rooted policy of discouraging labor law
violations, and the State, as a market participator in the
purchase of goods and products, can so restrict its purchasers so
as not to promote labor law violators as an exception to NLRA
preemption . . . Wisconsin is not engaging in direct state
regulation of private labor conduct, but is merely seeking seeking
to influence private conduct already prescribed (sic) by federal
labor policy."
Similarly, the State argued in the court of appeals that "Wisconsin
has a deeply rooted interest in compliance with the labor laws" (J.S.
App. 14; see id. at 31). Thus, the policy behind the debarment statute
could not be plainer: the State is attempting to influence the conduct
of employers to prevent future violations of the NLRA by penalizing
those who have violated the Act in the past. /6/
The State is, as it admits (J.S. 19), acting out of "a policy
disgust" at dealing with unfair labor recidivists. By contrast, it does
not argue that its debarment statute is intended to further some
legitimate economic interest of the State. For example, the statute has
not been justified by a concern for product quality or uninterrupted
supply. Indeed, Wisconsin admits (Br. 30) that debarment is costly to
the State and results in "economic inefficiency." It is, moreover,
self-evident that the debarment policy can make a difference only when
the State's normal procurement criteria would otherwise cause it to deal
with the debarred company. Thus, the policy behind the statute is
related solely to the State's interest in punishing federal labor law
violators and not to any economic self-interest that would benefit an
ordinary participant in the market.
The practical effect of the statute also is plain: it operates to
penalize employers and their employees by barring them from competing
for the State's business. Gould faces the loss of $250,000 worth of
business with Wisconsin (J.A. 17, 20). Moreover, Wisconsin's debarment
sanction extends beyond the immediate employer that has violated the
NLRA. Pursuant to the statute, the State has refused to purchase from
any supplier products that contain components produced by employers on
the list of violators (J.S. App. 52-53), thus spreading the disruptive
effect of the penalty throughout the economy and increasing the
likelihood that the debarment statute will in fact influence the conduct
of employers, as it was intended to do.
The State of Wisconsin spent almost $7 billion in 1981 alone. /7/
While the economic effect on any particular firm would vary according to
such factors as its size and the percentage of its business attributable
to the State, a boycott by a consumer of this magnitude cannot help but
be significant in a large number of cases. The State certainly must
have though this to be true; otherwise, the debarment statute could not
have its intended effect of deterring violations of the NLRA. The
suggestion of amici National Governors' Association et al. (NGA) that a
boycott by Wisconsin "imposes no penalty" because labor law violators
are free merely "to take their business elsewhere" (Br. 12; see Br. 21)
is thus essentially fatuous. /8/
2. The State argues (Br. 31-33) that because its debarment statute
does not prescribed rules of conduct for private persons, it is not the
sort of state regulation addressed by the Court's labor preemption
cases. This argument is unavailing. In the first place, the cases make
it clear that states may not establish their own remedies for
violations, regardless of whether they have established their own
standards of conduct or merely incorporated those set forth in the NLRA
as interpreted by the Board (see pages 9-12, supra). /9/
Moreover, although the law at issue may be novel, its practical
impact on employers and employees is sufficient to bring it within the
principles already developed. The Court has cautioned against
"elevat(ing) form over substance" in labor preemption cases.
Allis-Chalmers Corp v. Leueck, No. 83-1748 (Apr. 16, 1985), slip op. 9.
/10/ Here, although the debarment statute is not in form identical to
the more typical state penalty, such as damages or a fine, in substance
it has the same effect: it imposes monetary costs on an employer that
has violated the NLRA. Surely Congress did not intend to permit
Wisconsin to impose a $250,000 fine on an employer that has been found
to have committed three unfair labor practices. Nor should it be
understood as having intended to allow the State instead to withdraw
$250,000 worth of business on the same factual predicate. /11/
This conclusion is reinforced by the "essentially remedial" nature of
the NLRA. Republic Steel Corp. v. NLRB, 311 U.S. 7, 10 (1940); accord,
Local 60, United Bhd. of Carpenters v. NLRB, 365 U.S. 651, 657-658
(1961). The Board for example, may not fine employers as a means of
deterring violations of the Act (Republic Steel Corp., 311 U.S. at 12):
(I)t is not enough to justify the Board's (remedy) to say that
(it) would have the effect of deterring persons from violating the
Act. That argument proves too much, for if such a deterrent
effect is sufficient to sustain an order of the Board, it would be
free to set up any system of penalties which it would deem
adequate to that end.
Thus, while the Board is free in any case to take whatever measures
are required to restore circumstances to what they would have been had
the unfair labor practices not been committed (Phelps Dodge Corp., 313
U.S. at 194) and to prohibit the commission of the same or similar
unfair labor practices in the future of (NLRB v. Raytheon Co., 398 U.S.
25, 27 (1970); NLRB v. Mexia Textile Mills, Inc., 339 U.S. 563, 567-568
(1950)), the Board is not empowered by the scheme devised by Congress to
issue an order that, like the Wisconsin debarment statute, operates as a
flat penalty for committing a given number of unfair labor practices and
that bears no necessary relation to the nature and consequences of the
unfair labor practices concerned.
Congress sought in the Act to "minimiz(e) disruption of interstate
commerce" by furthering "responsible and stable labor relations." Retail
Clerks Int'l Ass'n, Local Unions Nos. 128 & 633 v. Lion Dry Goods, Inc.,
369 U.S. 17, 27 (1962). It would be odd indeed to suppose that Congress
intended that the States could set up their own deterrent system of
financial penalties for violations of the NLRA when the Act itself "does
not prescribe penalties or fines" but rather seeks only "the protection
and compensation of employees" (Republic Steel Corp., 311 U.S. at 10).
Wisconsin's imposition of a penalty that Congress denied the Board would
quite plainly stand as an obstacle to the achievement of the purposes of
the NLRA.
C. The Penalty Imposed By Wisconsin Is Preempted By The National Labor
Relations Act
1. As an additional sanction for violations of the NLRA, Wisconsin's
debarment statute is preempted by the federal Act under the principles
enunciated by this Court time and again (see pages 9-12, supra). The
Court in Garmon declared (359 U.S. at 243-244) that a state law would
not be preempted if it addressed conduct that was "a merely peripheral
concern of the (NLRA)" or that "touched interests so deeply rooted in
local feeling and responsibility" that "compelling congressional
direction" would be required for preemption. See also, e.g., Local 926,
Int'l Union of Operating Eng'rs, 460 U.S. at 676. Neither exception is
available to the State here. /12/
a. Far from applying to "merely peripheral" conduct, the Wisconsin
law strikes at the heart of the NLRA: actual unfair labor practices as
determined by Board orders enforced by courts of appeals. As we have
explained (pages 9-12, supra), the coordinated remedial scheme developed
and applied by an expert federal agency is central to the achievement of
Congress's purposes in enacting the NLRA. A welter of state laws, such
as already is beginning to exist (see NGA Am. Br. n.1), each withdrawing
business for varying lengths of time from employers that have committed
varying numbers of unfair labor practices anywhere in the country, would
obviously disrupt the uniform administration of the Act that Congress
sought to acheive, to the detriment both of employers and employees.
/13/
Indeed, in considering and rejecting a proposed debarment scheme,
Congress recognized that debarment of labor law violators from
government contracts must be treated as an integral part of the federal
regulatory scheme. A federal contract debarment program for labor law
recidivists was proposed as part of the Labor Law Reform Act of 1978.
S. 2467, 95th Cong., 2d Sess. The bill's supporters made it clear that
such a measure must be confided initially to the discretion of the
Board's General Counsel acting under limits carefully defined by
Congress, if it is to effectuate, rather than thwart, federal labor
policy:
Of course, a debarment program cannot and should not be
utilized to achieve labor relations ends contrary to those
established by Congress. This bill prevents such practices by
placing initial control of the remedy in the Labor Board, which
has the responsibility of enforcing the unfair labor practice
standards already declared in Section 8 of the NLRA.
Under the bill, the General Counsel of the Board may issue a
complaint including the recidivism allegation which can lead to
debarment. The Committee's intent is that he not include this
allegation with respect to minor violations or those where
willfulness cannot be proven, or in other circumstances in which
debarment would be inappropriate.
S. Rep. 95-628, 95th Cong., 2d Sess. 30 (1978). The report goes on
to define "willfulness" and to make clear that only conduct that
violates an outstanding Board order is covered by the "debarment
sanction" (id. at 31). /14/ The period of debarment would be adjustable
to "effectuate the purposes of the NLRA" (ibid.); debarment similarly
could be rescinded to protect employees or to further the policies of
the Act (id. at 32).
The congressional consideration of this bill demonstrates the
functional connection between debarment and the operation of the federal
labor law scheme. The finely tuned approach suggested by the Committee
in order to avoid conflict with the NLRA contrasts sharply with the
mechanical Wisconsin law, which takes no account of the willfulness of
violations or other circumstances that may render debarment
inappropriate and indeed counterproductive to the goals of the Act.
That Congress has thus far not enacted a debarment sanction underscores
the need for preemption: "to allow the State to grant a remedy * * *
which has been withheld from the National Labor Relations Board only
accentuates the danger of conflict" (Garmon, 359 U.S. at 247). /15/
Indeed, this conflict is highlighted by the fact that the federal bill
was rejected at least in part because the debarment sanction was
considered incompatible with the NLRA. As Senators Hatch and Hayakawa
stated (S. Rep. 95-628, supra, at 68):
The (debarment) remedy is quite clearly punitive in nature. It
constitutes punishment against employers without providing any
compensation whatsoever for employees. Ironically, in its seal to
punish employers, the remedy will often result in a significant
loss of jobs -- occasioned by the loss of government contracts --
for the very employees it was designed to protect.
Moreover, the Board already has the power to tailor its remedies to
take into account the impact of recidivism on the administration of the
NLRA. See, e.g., J.P. Stevens & Co. v. NLRB, 417 F.2d 533, 538-541 (5th
Cir. 1969) (union afforded access to company bulletin boards and to a
list of names and addresses of employees; special requirements imposed
respecting distribution and reading of the Board's notice). The Board
may, for example, issue a "broad order" enjoining unlawful conduct
directed at parties other than those involved in the particular
proceeding where there is a history of repeat violations. See, e.g.,
NLRB v. Local Union, No. 3, Int'l Bhd of Elec. Workers, 477 F.2d 260,
268 (2d Cir.), cert. denied, 414 U.S. 1065 (1973). And, as the court of
appeals noted (J.S. App. 20 n.7), the Board may seek court enforcement
of its order even where compliance has taken place, if, in its judgment,
the remedial purposes of the Act so require. See 29 C.F.R. 101.13(b);
NLRB v. Mexia Textile Mills, Inc., 339 U.S. at 567-568; see also
Wallace Corp, v. NLRB, 323 U.S. 248, 254-255 (1944) (Board can go behind
settlement agreement to remedy subsequent violations of the Act). Like
the bill considered by Congress, the Board's carefully constructed
remedies for recidivism demonstrate the delicate balancing necessary to
tailor such a remedy to the goals of the Act, balancing that would be
wholly frustrated by the Wisconsin law.
b. Nor is the State acting solely with respect to matters of deep
local concern. The "local concerns" that may permit a state law to
stand in the face of potential conflict with the NLRA are for the most
part limited to a circumscribed set of traditional, core interests
behind state public welfare and safety statutes and common law. See,
e.g., Belknap, Inc. v. Hale, 463 U.S. 491 (1983); Linn v. United Plant
Guard Workers, Local 114, 383 U.S. 53 (1966); United Construction
Workers v. Laburnum Construction Corp., 347 U.S. 656 (1954). While the
State undoubtedly has an interest in how it spends its money, the
debarment statute does not further any state economic interest. To the
contrary, the statute concededly costs the State money by preventing it
from doing business with contractors whose bids provide the most
attractive combination of price, quality, and other features. /16/
By contrast, the extent of any legitimate concern that Wisconsin may
have for violations of the federal labor Act -- a motivating force
behind the debarment statute (see pages 12-13, supra) -- is
significantly weaker than its sphere of action in protecting its fisc
and citizens. The state law, moreover, is not limited to NLRA
violations within Wisconsin. To the contrary, it applies to unfair
labor practices occurring anywhere in the United States. Indeed, none
of the violations of the Act that formed the predicate for Gould's
debarment occurred in Wisconsin. /17/ This case is thus a far cry from
those involving conduct in violation of state law that took place within
the state's own borders. See, e.g., Belknap, Inc. v. Hale, supra
(misrepesentation and breach of contract); Linn v. United Plant
Workers, Local 114, supra (defamation). There is no need to require
"compelling congressional direction" (Garmon, 359 U.S. at 244) before
concluding that Congress did indeed intend to preempt such far-reaching
state sanctions.
2. To escape the obvious conclusion that state penalties for unfair
labor practices are preempted by the NLRA, the State, supported by amici
(NGA Br. 6-12; AFL-CIO Br. 19-22), argues (Br. 19-38) that it is acting
as a "market participant" rather than in its regulatory capacity and
that as such its actions may not be preempted. In making this argument,
the State and amici rely on a line of cases decided under the so-called
"dormant" Commerce Clause, addressing the limitations on state activity
burdening interstate commerce in areas where Congress has not itself
legislated. The argument fails for two reasons. First, Wisconsin is
not acting as a market participant as that term has been understood by
the Court. Second, even if it were, the dormant Commerce Clause cases
are inapplicable here, where Congress has exercised its commerce power
in the NLRA.
a. The few cases in which this Court has applied the market
participant doctrine, the State's activities were all designed to
"limit() benefits generated by a state program to those who fund the
state treasury and whom the State was created to serve." Reeves, Inc. v.
Stake, 447 U.S. 429 (1980). Thus, in Reeves, a state sold cement that
it had produced at a plant funded with state taxpayers' dollars only to
state residents in times of shortage. In Hughes v. Alexandria Scrap
Corp., 426 U.S. 794 (1976), a state's bounty program for abandoned
vehicles favored in-state processors. As the Court explained in Reeves
(447 U.S. at 442-443 n.16), the state in Alexandria Scrap was motivated
by the "desire to channel state benefits to the residents of the State
supplying them." Similarly, in White v. Massachusetts Council of Constr.
Employers, Inc., 460 U.S. 204 (1983), a municipal ordinance favored
municipal residents in construction projects financed in whole or in
part by city funds. Finally, in South-Central Timber Development, Inc.
v. Wunnicke, No. 82-1608 (May 22, 1984), the plurality noted (slip op.
15) that the state had acted "with the permissible state purpose of
fostering local industry." /18/
Here, Wisconsin has not acted with the purpose or effect of granting
its own residents preferential treatment in the distribution of State
funds. Rather, it is using its purchasing power for the purpose and
effect of punishing federal labor law violators. To accomplish this
result, the State refuses to purchase goods from vendors such as Gould
regardless of whether they are in-state or out-of-state and even when
they have made the lowest bid or are otherwise the most attractive.
This conduct bears little resemblance to that of a private actor in the
marketplace, for it furthers no economic interest of the State or its
residents. /19/ Once the State goes beyond conduct that would be normal
or economically rational for a private party in the market to achieve
other goals, it is no longer acting as a market participant. /20/ By
whatever method it has chosen, it is regulating.
b. Even if Wisconsin were acting as a market participant as that term
is used in the dormant Commerce Clause cases, that consideration is not
germane to whether its actions are preempted under the NLRA. The
Commerce Clause is a "self-executing limitation on the power of the
States to enact laws imposing substantial burdens on (interstate)
commerce." South-Central Timber, slip op. 5. At issue in every one of
the Court's market participant cases was this self-executing, "dormant"
aspect of the Commerce Clause. As the Court stated in Alexander Scrap
(426 U.S. at 810 (footnotes omitted; emphasis added)), "(n)othing in
the purposes animating the Commerce Clause prohibits a State, in the
abscence of congressional action, from participating in the market and
exercising the right to favor its own citizens over others." The Court
did not decide (id. at 810 n.19) "whether Congress could prohibit the
type of selective participation in the market" present in that case.
Here, Congress has acted pursuant to its commerce power in enacting
the NLRA. Its intent, as the Court has repeatedly recognized (see pages
9-12, supra), was to preempt the field of remedies for unfair labor
practices, and thereby to prohibit the states from acting to impose
penalties for such conduct regardless of the form of those sanctions.
There is nothing to suggest that Congress intended states to be able to
penalize labor law violators through the use of the public purse, when
the effect is to undermine the remedial scheme administered by the NLRB
just as if they had acted by more traditional routes. Indeed,
Congress's own rejection of a debarment sanction (see pages 19-21,
supra) suggests precisely the opposite.
In United Building & Constr. Trades Council v. Mayor of Camden, No.
81-2110 (Feb. 21, 1984), slip op. 10, the Court "decline(d) to transfer
mechanically into (the Privileges and Immunities Clause) context (the
market participant analysis fashioned to fit the Commerce Clause." Just
as the Privileges and Immunities Clause "imposes a direct restraint on
state action" (id. at 11) regardless of the form of that action so does
the NLRA, through the Supremacy Clause. That the dormant Commerce
Clause does not do so is simply irrelevant.
Amici NGA et al. make two arguments (Br. 6-12) to counter this
reasoning. First, they attempt to distinguish between Congress's power
to regulate the states and its intent to preempt state action. This
distinction lacks substance. If Congress has the power expressly to
prevent the states from enacting statutes such as Wisconsin's, it
certainly may exercise that power through its implicit intent to preempt
those statutes. The argument therefore collapses into NGA's second
contention, which is that Congress lacks the authority under the
Commerce Clause to set limits on a state's conduct as a market
participant. Amici assert that the dormant Commerce Clause cases stand
for the proposition that Congress has no power under the Commerce Clause
to legislate with respect to a state's market participation except
insofar as it has legislated in a like manner with respect to private
persons in the market. The State appears to make a similar argument
(Br. 29-30).
This argument fails for several reasons. First, the Court has
expressly stated that it was not addressing the applicability of the
market participant doctrine to circumstances (like the present) where
Congress has exercised its legislative power (see page 26, supra).
Second, the "Commerce Clause is a grant of authority to Congress, not a
restriction on the authority of that body." Massachusetts Council of
Constr. Employers, Inc., 460 U.S. at 213. It therefore does not follow
from the limitations on the reach of the dormant Commerce Clause that
Congress lacks the power to regulate states' market conduct. At this
late date, it is quite far-fetched to suggest that the NLRA is
unconstitutional insofar as it reaches the states qua market
participants. See generally NLRB v. Jones & Laughlin Steel Corp., 301
U.S. 1 (1937) (upholding constitutionality of the NLRA under the
Commerce Clause).
Indeed, the State and amici appear to admit as much; they argue
primarily that Congress cannot regulate state market participation to
any greater degree than it regulates private market conduct. But
certainly nothing in the Commerce Clause itself requires such identical
treatment of public and private entities. The Supremacy Clause makes
the policy of Congress the policy of the states as well. Accordingly,
as amicus AFL-CIO properly concedes (Br. 12-13), this Court has
recognized that governments and private actors are not similarly
situated -- that some conduct, for example, may be protected under the
NLRA "'against state, but not employer interference.'" Lodge 76, Int'l
Ass'n of Machinists v. Wisconsin Employment Relations Comm'n, 427 U.S.
132, 145 n.6 (1976) (quoting Cox, Labor Law Preemption Revisited, 85
Harv. L. Rev. 1337, 1346 (1972) (footnote omitted)). This is especially
true here, where the State is not acting as any rational private actor
could or would (see pages 22-26 & note 20, supra). Whatever the force
in other circumstances of amici's argument that Congress cannot regulate
state market participation to any greater degree than it addresses the
participation of private persons, it is largely inapposite here, where
the state's rule serves only a regulatory rather than a business purpose
and similar private conduct accordingly would be wholly aberrational.
/21/ Certainly Congress need not have addressed the idiosyncratic
private actor that boycotts labor law recidivists in order to be
understood to have intended to preempt state boycotts which so plainly
hobble Congress's purpose in the NLRA -- to promote commerce under a
remedial scheme developed and applied solely by the Board.
CONCLUSION
The judgment of the Court of appeals should be affirmed.
Respectfully submitted,
ROSEMARY M. COLLYER
General Counsel
JOHN E. HIGGINS, JR.
Deputy General Counsel
ROBERT E. ALLEN
Associate General Counsel
NORTON J. COME
Deputy Associate General Counsel
LINDA SHER
Assistant General Counsel
ELINOR HADLEY STILLMAN
Attorney National Labor Relations Board
I authorize the
filing of this brief.
CHARLES FRIED
Acting Solicitor General
SEPTEMBER 1985
/1/ Gould also argued that the statute violated its rights to due
process and equal protection under the Fourteenth Admendent by burdening
its right to seek judicial review of Board orders. The district court
did not reach this claim (J.S. App. 82 n.11), but it did conclude that
the argument was sufficiently substantial to justify an award of
attorney's fees under 42 U.S.C. 1988 (see J.S. App.8).
/2/ The district court dismissed three state agency defendants under
the Eleventh Amendment; the action proceeded against four state
officials under Ex parte Young, 209 U.S. 123 (1908) (J.S. App. 53-55).
Gould apparently abandoned any claim for damages (id. at 55 n.3).
/3/ The court of appeals reversed the district court's award of
attorney's fees to Gould under 42 U.S.C. 1988 (J.S. App. 38-44). Gould
did not cross-appeal from this portion of the judgment, and it is
therefore not before the Court. See, e.g., Bolger v. Youngs Drug
Products Corp., 463 U.S. 60, 64 n.5 (1983).
/4/ Indeed, an unfair labor practice adjudication is not necessarily
an indication of bad faith on the part of an employer; it can often
result from a good-faith effort to test the scope of rights where this
is unclear under the Act. Moreover, under the structure of the Act, it
is the only means whereby employers can challenge Board certification
decisions, such as bargaining unit determinations.
/5/ The Court's holdings have, of course, been grounded on Congress's
intent as expressed in the NLRA: "as in any preemption analysis, (t)he
purpose of Congress is the ultimate touchstone." Metropolitan Life Ins.
Co. v. Massachusetts, No. 84-325 (June 3, 1985), slip op. 22 (citation
and quotation marks omitted; brackets in original).
/6/ Before this Court, the State ignores its previous representations
and suggests (Br. 30-31, 45) that its debarment statute was intended
merely to further the State's interest in not giving its custom to
employers that have committed unfair labor practices. Even crediting
this change of heart, the structure and practical effect of the statute
ultimately must be controlling. See, e.g., Dean Milk Co. v. City of
Madison, 340 U.S. 349, 354 (1951).
/7/ Bureau of the Census, U.S. Dep't of Commerce, Statistical
Abstract of the United States 289 (104th ed. 1984). This figure
includes salaries of state employees and transfer payments, but it
suffices to demonstrate the scale of the State's purchasing power.
/8/ NGA notes (Am. Br. 2 n.1) that at least four other states have
passed legislation similar to Wisconsin's. These five states combined
spent more than $40 billion in 1981. Bureau of the Census, supra, at
289. The total amount spent by state and local governments in 1984 for
the purchase of goods and services (not including employee compensation)
was more than $190 billion. 65 Bureau of Economic Analysis, U.S. Dep't
of Commerce, No. 4, Survey of Current Business 13 (1985). It is
apparent that statutes such as Wisconsin's especially if they are
adopted by more and more states, can be expected to have quite a
significant economic impact on employers and employees throughout the
nation.
/9/ Surely Wisconsin could not prohibit recidivist labor law
violators from doing any business within the State. Such a law would be
preempted as to business both with private parties and with the State
itself. That Wisconsin's penalty has not swept so far, encompassing
only the latter prohibition, cannot reasonably be thought enough to make
it consistent with federal labor policy.
/10/ Plainly, the mere form of state action cannot exempt it from the
preemption doctrine. For example, a state statute granting a preference
in state contract bidding to labor law recidivists would obviously be
preempted, as it would tend to encourage labor law violations. Because
Wisconsin's law debarring recidivists conflicts (albeit in a different
way) with the achievement of federal labor policy, it too is preempted.
/11/ Amicus AFL-CIO relies on the plurality opinion in New York Tel.
Co. v. New York State Dep't of Labor, 440 U.S. 519 (1979), in arguing
(Br. 18-19) that a state's expenditure of public funds (there,
unemployment benefits) does not come within the traditional labor
preemption doctrine. Aside from the substantially different factual
situation -- here, the State's decision is predicated on unfair labor
practice findings and is intended to penalize violations of the NLRA --
the views of the plurality of three Justices in that case failed to gain
the acceptance of a mjaority of the Court. The case was decided on the
narrow ground of Congress's intent with respect to unemployment
compensation as expressed in the legislative histories of the NLRA and
the Social Security Act, which were passed within a short time of one
another. See id. at 540-546 (plurality opinion); id. at 546-547
(Brennan, J., concurring in the result); id. at 547-551 (Blackmun, J.,
concurring in the judgment).
/12/ Because Congress has actually addressed the employer's conduct
in the NLRA, these exceptions are not in fact directly applicable here.
See Brown v. Hotel & Restaurant Employees Int'l Union Local 54, No.
83-498 (July 2, 1984), slip op. 10-11; interests is, however, relevant
to whether Congress did intend to preempt a particular type of state
regulation. Metropolitan Life Ins. Co., slip op. 24 n.27. We will use
the "peripheral concern" and "local "interest" rubrics here only because
they seem a useful way of rebutting the State's contentions.
/13/ The State argues (Br. 40-44) that the Garmon preemption doctrine
applies only to prevent conflicting state and federal rules of conduct
and to protect the Board's primary jurisdiction to determine the
legality of conduct under the NLRA in the first instance. These are of
course grounds for preemption under Garmon, but not the only ones. It
is equally clear that preemption is necessary to prevent states from
adding to the federal remedial scheme, regardless of the consistency of
their subsantive rules with those of the Act. See pages 9-12, supra.
None of the cases cited by the State in any sense overules the
application of Garmon and Lockridge to state-supplied remedies for
conduct that violates the Act.
/14/ Accordingly, the enforcement of Board orders -- the basic
predicate for debarment under Wisconsin's law -- would not have been
sufficient for debarment under the federal bill. Only if the employer
were in contempt of an enforced Board order would the federal sanction
have applied.
/15/ Amici NGA would discount the relevance of Congress's refusal to
pass a federal debarment sanction on the ground that it does not help to
demonstrate Congress's intent in passing the NLRA (Br. 22 n.14).
Congress's actions subsequent to enactment of the NLRA, however, can
prove a helpful guide to the difficult determination of the preemptive
scope of the Act. See Brown v. Hotel & Restaurant Employees Int'l Union
Local 54, slip op. 15-16; NLRB v. Gullett Gin Co., 340 U.S. 361,
365-366 (1951). While the intent of a subsequent Congress may of course
differ from that of an earlier one, Congress's recent rejection of a
federal debarment sanction is quite germane to the practical question of
whether a state debarment penalty would interfere with the achievement
of the goals of the NLRA.
/16/ The motivation behind Wisconsin's statute suggests another way
in which it is incompatible with federal labor policy. The State has
acted to further what it perceives to be the interest of employees'
organizational rights by boycotting goods produced entirely or in part
by labor law violators. Congress, however, has expressly forbidden
unions from requiring employers to boycott goods of neutral employers.
See Section 8(e), 29 U.S.C. 158(e). Surely Congress did not intend to
allow the states (presumably at the behest of union advocates) to engage
in conduct so similar to the statutorily disfavored secondary boycott.
/17/ Moreover, Gould's divisions were responsible for their own labor
relations policies (State Br. 13, citing R. 2), and Gould no longer
owned the divisions that had committed the unfair labor practices at the
time that it was debarred (J.S. App. 7).
/18/ The plurality in South-Central Timber concluded (slip op.
14-17), however, that the state did not act as a market participant
because it imposed "downstream" restrictions on timber processing by its
customers. Similarly, here Wisconsin has imposed an "upstream"
restriction that none of its direct suppliers may themselves have
obtained supplies (as to the goods sold to the State) from Gould (see
pages 2, 14, supra). This factor alone is sufficient to take
Wisconsin's conduct outside the market participant doctrine.
/19/ The State does not contend, for example, that it is more
expensive or disruptive to deal with labor law violators. Cf.
Amalgamated Transit Union, Division 819 v. Byrne, 568 F.2d 1025, 1030
(3d Cir. 1977) (en banc) (State's refusal to subsidize transportation
firms that agreed to an uncapped cost-of-living clause in a collective
bargaining agreement furthered the State's interest in conserving scarce
financial resources.
/20/ As the State appears to recognize (Br. 30), a corporation's
management may not further its own social policies at the economic
expense of its shareholders. See, e.g., CFTC v. Weintraub, No. 84-261
(Apr. 29, 1985), slip op.5. A state, by contrast, may use its coercive
power to tax its citizens in order to finance economically inefficient
policies. Thus, contrary to the State's argument (Br. 29-30), Wisconsin
is not merely acting as General Motors would, and it is, for that reason
among others, not entitled to be excepted from the preemptive effect of
the NLRA. Moreover, even if Wisconsin could in this manner permissibly
further its local interest in compliance with the labor laws within its
borders, the law here sweeps more broadly, reaching violations occurring
anywhere in the United States.
/21/ The regulatory nature of the State's rule also wholly
distinguishes the question here from the Tenth Amendment concerns that
divided the Court in Garcia v. San Antonio Metro. Transit Authority, No.
82-1913 (Feb. 19, 1985), and its predecessor cases. In contrast to the
congressionally prescribed rules for state employment at issue in those
cases, the present case raises only a traditional preemption issue --
whether state-imposed sanctions regulating private conduct are
compatible with a federal scheme regulating the same private conduct.
Our position here, therefore, is entirely consistent with the views
expressed by all Members of the Court in Garcia.
STATE OF DELAWARE, PETITIONER V. ROBERT E. VAN ARSDALL
No. 84-1279
In The Supreme Court of the United States
October Term, 1985
On Writ Of Certiorari To The Supreme Court Of Delaware
Brief For The United States As Amicus Curiae Supporting Petitioner
TABLE OF CONTENTS
Interest of the United States
Statement
Summary of argument
Argument:
The erroneous restriction of a defendant's
opportunity to demonstrate bias on the part of a
prosecution witness does not require automatic
reversal without regard to prejudice
Conclusion
QUESTION PRESENTED
Whether an erroneous restriction upon a defendant's opportunity to
impeach an adverse witness by showing that he may be biased in favor of
the prosecution requires that the defendant's conviction be set aside
without any regard to whether the defendant was prejudiced by the error.
INTEREST OF THE UNITED STATES
The question presented by this case is whether an erroneous
restriction upon a defendant's opportunity to impeach a witness by
showing that he is biased in favor of the prosecution requires that the
defendant's conviction be set aside without any consideration of whether
the error was prejudicial. The Court's resolution of this issue will
apply equally to federal and state prosecutions.
STATEMENT
Following a jury trial in Kent County, Delaware, Superior Court,
respondent was convicted of first degree murder, in violation of Del.
Code Ann. tit. 11, Section 636(a)(1) (1979), and possession of a deadly
weapon during the commission of a murder, in violation of Del. Code Ann.
tit. 11, Section 1447 (1979 & Supp. 1984). Respondent was sentenced to
life imprisonment without the possibility of probation or parole on the
murder conviction and to a consecutive 30-year term of imprisonment on
the weapons charge (J.A. 2). On respondent's appeal, the Delaware
Supreme Court reversed his convictions on the ground that the trial
judge had improperly restricted cross-examination of prosecution witness
Robert Fleetwood in contravention of the Confrontation Clause of the
Sixth Amendment (Pet. App. A1-A23).
1. Shortly after midnight on January 1, 1982, Doris Epps was stabbed
to death in an apartment in Smyrna, Delaware, following a New Year's Eve
party. Respondent and Daniel Pregent, the only persons in the apartment
with Epps at the time that she was murdered, were arrested at the scene
of the crime and were charged with Epps' murder. At separate trials,
respondent was convicted and Regent was acquitted.
a. On December 31, 1981, Pregent and Robert Fleetwood held a joint
New Year's Eve party in their adjacent apartments (Pet.App.A2). The
party lasted from late morning until shortly before midnight, and more
than a dozen guests attended the party on and off during the course of
the day (ibid; II Tr. 91-94; J.A. 73-80, 121-123). The victim of the
murder, Doris Epps, joined the party at roughly 4:00 p.m. and stayed for
the remainder of the day (II Tr. 93, 118-119; III Tr. 45, 56; J.A.
78-79, 88). /1/ Respondent, an acquaintance of Fleetwood and Pregent (X
Tr. 28), stopped briefly at the party during that day (Pet. App. A2; II
Tr. 94; J.A. 79-80, 90).
At roughly 11:25 p.m., respondent returned to Pregent's apartment
(III Tr. 29-33; X Tr. 18-41). By that time, the party was over.
Pregent had quarreled with a female quest and had kicked a hole in a
hallway wall (III Tr. 48). Epps had become intoxicated and passed out
at about 10:30 p.m., and she was placed on a sofa bed in Pregent's
living room (II Tr. 111-113). Afterwards, Pregent got into the bed with
Epps, and everyone else left his apartment (II Tr. 107-108). Only
Fleetwood, Alice Meinier, and Mark Mood remained in Fleetwood's
apartment (III Tr. 49; IV Tr. 8-13; J.A. 95-96).
At approximately 11:30 p.m., a few minutes after respondent had
returned to Pregent's apartment, Fleetwood walked across the hall,
looked into Pregent's living room from the doorway, and saw respondent
sitting on the end of a sofa bed next to Pregent's feet (Pet.App.A3;
III Tr. 50). Fleetwood, who did not have a complete view of the bed,
did not see Epps or anyone else in the apartment; he returned to his
own apartment without speaking to respondent or to Pregent (III Tr.
50-52, 66-67; J.A. 82-85). Shortly before midnight, Meinier walked
across the hall from Fleetwood's apartment to check the clock in
Pregent's kitchen. /2/ The kitchen light was on and the clock indicated
it was 11:53 p.m. (IV Tr. 14, 51). Meinier immediately returned to
Fleetwood's apartment without looking into Pregent's living room, which
was dark (IV Tr. 14). Fleetwood fell asleep on his couch a few minutes
after nearby church bells had chimed in the New Year, but Meinier and
Mood remained awake talking in Fleetwood's apartment (III Tr. 53; J.A.
125-128).
Roughly one hour later, respondent knocked at Fleetwood's door, and
Meinier admitted him (IV Tr. 17). Respondent's shirt and hands were
splattered with blood, and he was holding a long, blood-covered knife
(id. at 17-20). Respondent stated that "he had gotten in a fight" but
that he "got them back" (id. at 19, 61-62). After Mood took the knife
from respondent's hand, Meinier suggested that respondent wash his hands
in the kitchen sink (id. at 21-22). Respondent placed his wrist watch,
which was blood soaked and had a piece of human tissue clinging to it,
on the counter next to the sink and washed his hands. Respondent then
said, "I think there's something wrong across the hall." Meinier went to
Pregent's apartment and discovered Epps' body lying in a pool of blood
on the kitchen floor. Mood then summoned the police (IV Tr. 65; J. A.
128-134).
The police saw Epps' disemboweled and mutilated body on Pregent's
kitchen floor clad only in a sweater and bra. The kitchen floor,
appliances, and cabinets were splattered with blood and tissue. /3/
Blood smears led from the kitchen to Pregent's blood-drenched sofa bed,
on which the police found Pregent wrapped in a blanket. The murder
weapon, a twelve-inch serrated kitchen knife with a seven and one-half
inch blade, and respondent's watch were recovered, respectively, from
Fleetwood's sink and from his counter top (IV Tr. 111-115, 122-123,
130-136, 140-141; V Tr. 8; VI Tr. 72). At the scene, respondent,
whose clothes and shoes were blood-splattered, explained to the police
that he had gotten covered with blood "trying to help the woman" (IV Tr.
84-87). Respondent also stated that he "didn't think it would go like
this" (VII Tr. 72). Both respondent and Pregent were arrested.
b. Later that morning at the police station, respondent made a
tape-recorded statement (J.A. 10-34). According to his statement,
respondent arrived at the party at Pregent's apartment in the late
afternoon and left at roughly 6:00 p.m. (J.A. 21-22). After "riding
around" with some friends, drinking "(q)uite a bit," and visiting
another friend's home outside of town, respondent returned to Pregent's
apartment near midnight (J.A. 10, 19, 21-22). He ate a couple of
sandwiches, talked with Pregent for a while, and then went to sleep on
cushions near Pregent's sofa bed; Pregent got into the sofa bed with
Epps (J.A. 12, 22-23, 28-30). Respondent soon felt sick and went into
the hall to get some air, noticing nothing unusual in Pregent's
apartment (J.A. 11, 14, 22, 30). Shortly thereafter, a woman (i.e.,
Meinier) stepped out of Fleetwood's apartment and stated that she had
heard a noise (J.A. 23, 31). Respondent followed Meinier into Pregent's
apartment and found Epps' body on the kitchen floor. Respondent became
soaked with Epps' blood when he tried to help Epps, and he washed the
blood off his hands in Fleetwood's kitchen sink (J.A. 13-14, 23-24,
26-27, 31). Respondent also claimed that he had lost his Timex watch
sometime after 10:30 p.m. on New Year's Eve (J.A. 11, 21); he denied
that he had ever held a knife that evening (J.A. 26) or that he had
sexual intercourse with anyone the night of the crime (J.A 18); and he
attributed scratches on his arm to his having played that night with
Pregent's cat (J.A. 15-16).
In another tape-recorded statement given to the police two days later
(J.A. 36-41), respondent retracted most of his earlier statement,
explaining that he had lied in order to "cover up for my buddy
(Pregent)" (J.A. 36). According to respondent's revised version of the
facts, he returned to Pregent's apartment at about 11:30 p.m. (J.A. 37).
Pregrent lay down on the sofa bed with Epps, and respondent lay down on
some adjacent cushions (J.A. 37). After respondent heard noises
indicating that Pregent and Epps were engaged in sexual intercourse,
respondent fell asleep (J.A. 39-40). Respondent was later awakened by
Pregent dragging Epps' body past his feet into the kitchen (J.A. 37,
40). When respondent arose to investigate, Pregent struck him, "daz(ing
him) for a while" (J.A. 37). Respondent then saw Pregent repeatedly
stab Epps in the kitchen; when respondent tried to pull Pregent away
from Epps, Pregent knocked him down (J.A. 37-38). After Pregent ended
his attack on Epps, he washed himself off and went back to bed (J.A.
38-39). Respondent then pulled the knife from Epps' body, walked across
the hall to Fleetwood's apartment, and told Meinier and Mood that he had
been in a fight (J.A. 39). /4/
c. Both of respondent's statements were introduced at trial (VII Tr.
84-85). The State also introduced several types of physical evidence
and the testimony of a forensic expert (e.g., VI Tr. 23-109). According
to this expert, the blood found on the knife and watch recovered from
Fleetwood's apartment, as well as the blood staining the clothes that
respondent was wearing when he was arrested, matched Epps' blood type
(VI Tr. 34-49, 56-60, 72, 107-108; IX Tr. 12-13, 16). /5/ In addition,
fibers matching the fibers of Epps' sweater and one "Negro" pubic hair,
like that of Epps, were found on respondent's jockey shorts (VI Tr.
4-42, 54). /6/
d. The prosecution called Robert Fleetwood as a witness. In his
direct testimony, Fleetwood recounted uncontroverted facts regarding the
party and respondent's presence at Pregent's apartment the afternoon of
New Year's Eve and again later that night an hour or so before Epps was
killed (III Tr. 40-54). In all significant respects, Fleetwood's
testimony was corroborated by other witnesses, including respondent,
whose own testimony placed him at Pregent's apartment at about the same
times that Fleetwood placed him there (II Tr. 94; X Tr. 29, 33; J.A.
140-142).
Near the end of Fleetwood's cross-examination, defense counsel sought
to impeach Fleetwood by questioning him about the dismissal of a
misdemeanor charge against him -- being drunk on a highway -- after he
had agreed to speak with the prosecutor about Epps' murder (III Tr. 69).
When the prosecutor objected to that inquiry on relevancy grounds, the
trial court allowed counsel to voir dire Fleetwood on the matter (id. at
70-88). Fleetwood acknowledged that the drunkenness charge was dropped
in exchange for his promise to speak with the prosecutor regarding Epps'
murder, but he denied that the agreement had an effect on his testimony
(id. at 75-76; J.A. 100-107). /7/ The trial court disallowed any
cross-examination about that agreement (III Tr. 82; J.A. 110) and also
refused to permit defense counsel to cross-examine Fleetwood about his
being questioned by the police in connection with an unrelated homicide
that had occurred after Epps' murder (III Tr. 83-88; J.A. 111-115).
e. Respondent was the only defense witness. As in his second
statement to the police, respondent attributed Epps' murder to Pregent
(X Tr. 17-90). Respondent admitted that he had visited the party twice
during the day and that he returned shortly before midnight. /9/ Once
there, respondent ate, talked with Pregent, and played with Pregent's
cat (X Tr. 43-47). After Pregent left the living room, respondent had
sexual intercourse with Epps on the sofa bed at her invitation. /10/
Afterwards, Pregent lay down on the bed with Epps, and respondent
reclined on some nearby cushions (id. at 47-49). Before he fell asleep,
respondent heard Pregent and Epps engaged in sexual intercourse (id. at
49-50). Respondent's sleep was interrupted by Pregent dragging Epps's
limp body past respondent's feet (id. at 50-51). When respondent asked
Pregent what was going on, Pregent struck him (id. at 51). Respondent
then saw "dark stuff" on the floor, put on his shoes because the floor
"was wet," and walked to the doorway of the kitchen (id. at 51-52).
Respondent saw Pregent squatting over Epps in the kitchen, stabbing and
cutting her (id. at 52-53). Respondent grabbed Pregent, but was knocked
down (id. at 53-54. /11/ The next thing respondent remembered was
seeing ppregent exit the utility room of the apartment (id. at 54).
When Pregent walked back to the living room, respondent pulled the knife
from Epps' body - - to protect himself from Pregent, respondent claimed
(id. at 56, 70-71) -- and went to Fleetwood's apartment to get help (id.
at 57-58). After washing the blood off his hands in Fleetwood's kitchen
sink (id. at 59), respondent and Meinier went to Pregent's apartment,
and respondent checked Epps's wrist for a pulse (id. at 60-61). Finding
none, they returned to Fleetwood's apartment and tried unsuccessfully to
wake up Fleetwood (id. at 61-62).
Defense counsel admitted in their opening and closing arguments to
the jury that respondent was present at Pregent's apartment when Epps
was killed (J.A. 62, 64-65, 181, 188-189, 192-194). /12/ In closing
argument, defense counsel also said that none of the five prosecution
witnesses who were present that night, including Fleetwood, "testified
to any fact suggesting anything other than that (respondent) was in that
apartment" (J.A. 189).
2. On appeal to the Delaware Supreme Court, respondent argued that
the trial court had erroneously denied him the right to establish
Fleetwood's bias by limiting his cross-examination. Relying in part
upon Davis v. Alaska, 415 U.S. 308 (1974), the court reversed
respondent's convictions on the ground that the trial judge's ruling
barring any cross-examination of Fleetwood regarding the dismissal of
the misdemeanor charge violated the Confrontation Clause by keeping from
the jury facts regarding bias that were central to assessing Fleetwood's
credibility. Pet. App. A5-A6. /13/ In so doing, it rejected the
State's argument that, since "Fleetwood's basic testimony was cumulative
in nature and unimportant," the Confrontation Clause error was harmless
beyond a resonable doubt (Pet. App. A6). The court held that "a blanket
prohibition against exploring potential bias throught cross-examination"
is "a per se error," that "the actual prejudicial impact of such an
error is not examined," and that "reversal is mandated" (id. at A7).
SUMMARY OF ARGUMENT
It is a basic principle of modern American law -- applicable equally
to criminal as to civil cases, and to both constitutional and
non-constitutional claims -- that an appellant is not entitled to a
reversal of the trial court's judgment unless he can show not only the
existence of an error, but also some likelihood that the error was
prejudicial to him (i.e., that it was not "harmless"). It is therefore
customary for appellate courts to reverse only upon finding a sufficient
probability that the result would have been different but for the error.
In this case, however, the Delaware Supreme Court held that this
principle may not be applied where there has been an erroneous
restriction of the defendant's right to show that a prosecution witness
is biased, on the ground that the Sixth Amendment prohibits affirmance
of a conviction in such circumstances even if it can be demonstrated
that the restriction on cross-examination could not have affected the
outcome of the trial.
It is of course true that certain errors will always require reversal
without separately considering whether a likely effect on the verdict
has been demonstrated. For example, a trial before a biased tribunal,
the denial of counsel, or the denial of a jury trial can never be
"harmless," despite indisputable proof of the defendant's guilt, because
the proceeding at which he was convicted lacked the fundamental
attributes of a trial as we know it. Or, to consider a somewhat
different class of cases, harmless error analysis would be pointless
where the standard for determining whether there was any error at all
already turns in part on a finding of prejudice, as in Brady claims.
Finally, there may be instance in which prejudice is presumed (and
harmless error inquiry thus foreclosed) because of the violation of an
important right designed to protect the defendent against an unjust
conviction in circumstances in which is it impossible to determine the
actual or probable effect of the error.
Denying a defendant the opportunity to impeach an adverse witness
does not fit into any of those limited categories of errors requiring
automatic reversal. Confrontation by means of impeachment is valued not
as an end in itself, but as a means to the end of enhancing the
reliability of the verdict. It is thus entirely appropriate, upon
finding an improper restriction on impeachment of an adverse witness, to
determine whether successful impeachment of the witness's credibility
might have affected the verdict. That inquiry is not inherently
impracticable. Whether an error of this type is prejudicial hinges upon
a variety of factors -- such as the nature of the testimony that the
witness has given, the nature and strength of the alleged bias, and the
defense offered at trial -- which will necessarily vary from case to
case. Moreover, because impeachment is simply one of several means to
an end, denying a defendant the opportunity to impeach a witness does
not invariably deprive the accused of the benefits of confrontation.
Finally, as was the case here, the testimony given by a particular
witness may be so insignificant or may be so indisputably accurate in
light of other, corroborating evidence that the denial of an opportunity
to impeach that witness cannot reasonably be said to have deprived the
defendant of a fair opportunity to establish his innocence.
ARGUMENT
THE ERRONEOUS RESTRICTION OF A DEFENDANT'S OPPORTUNITY TO DEMONSTRATE
BIAS ON THE PART OF A PROSECUTION WITNESS DOES NOT REQUIRE AUTOMATIC
REVERSAL WITHOUT REGARD TO PREJUDICE
We have no quarrel with the ruling by the court below that, once the
State chose to call Robert Fleetwood as a witness, respondent should
have been allowed to impeach him by bringing out the fact that
Fleetwood's cooperation may have been induced by the dismissal of his
pending misdemeanor charge. Nevertheless, we think it indisputable that
this error could not possibly have affected the jury's verdict in light
of the marginal significance of the facts to which Fleetwood testified
and their corroboration in all material respects by the testimony of
other witnesses, including the state's forensic expert, by physical
evidence, and by respondent's own statements to the police and his trial
testimony. /14/ See pages 4-8, supra. The Delaware Supreme Court did
not suggest otherwise, but instead declined to consider whether the
error may have been prejudicial. Relying upon Davis v. Alaska, 415 U.S.
308 (1974), it concluded that the Consitution requires automatic
reversal whenever a defendant is improperly precluded from bringing out
the possible bias of a prosecution witness. This holding, which
conflicts with the majority of federal and state court rulings on the
subject, is wrong. /15/
A. It is well settled that the Consitution does not guarantee a
defendant a perfect trial and that not every error that occurs prior to
or at trial requires that the defendant's conviction be set aside as the
remedy. That uncontroversial principle recognized that "(a)bsent (an
actual or threatened adverse) impact on the criminal proceeding, * * *
there is no basis for imposing a remedy in that proceeding" (United
States v. Morrison, 449 U.S. 361, 365 (1981)) as well as that "remedies
should be tailored to the injury suffered . . . and should not
unnecessarily infringe on competing interests" (Rushen v. Spain, 464
U.S. 114, 118 (1983) (quoting Morrison, 449 U.S. at 364)). The harmless
error doctrine, as it has come to be known, is merely one application of
this principle.
In Chapman v. California, 386 U.S. 18(1967), the Court rejected the
argument that all federal constitutional errors, regardless of their
nature or severity or the strength of the government's proof of the
defendant's guilt, must be deemed inherently prejudicial, requiring
reversal of a judgment of conviction. The Court reasoned that in the
context of a particular case a given error may have had little, if any,
likelihood of skewing the factfinding process at trial and that where a
reviewing court may confidently say that no such effect occurred, the
reversal of a conviction provides an unjustified windfall for the
defendant. 386 U.S at 21-24; see United States v. Hasting, 461 U.S.
499, 508-509 (1983). /16/ Since Chapman, the Court has repeatedly
reaffirmed the principle that an otherwise valid conviction should not
be set aside if the constitutional error that occurred before or during
trial was harmless. /17/ Indeed, in Hasting the Court made clear that
"it is the duty of a reviewing court to consider the trial record as a
whole and to ignore errors that are harmless." 461 U.S. at 509 (emphasis
added); see also 28 U.S.C. 2111; Fed. R. Crim. P.52(a).
The harmless error doctrine furthers several fundamental interests in
the administration of criminal justice. It gives effect to the
principle that the essential purpose of a criminal trial is to decide
the factual question of the defendant's guilt or innocence (United
States v. Nobles, 422 U.S. 225, 230 (1975)), rather than merely to
deliver a preliminary answer to allegedly more important abstract legal
questions that will ultimately be resolved on appeal (cf. Wainwright v.
Sykes, 433 U.S. 72, 90(1977)). It also ensures that the criminal
process is not treated as a "game" by removing incentives for defense
counsel to attempt to sow technical errors at trial for the sole purpose
of obtaining a reversal on appeal. See Kotteakos v. United States, 328
U.S. 750, 759(1946). It promotes public respect for the criminal
process by focusing on the underlying fairness of the trial, rather than
on immaterial technicalities. See Hasting, 461 U.S. at 509, Kotteakos,
328 U.S. at 759-760; R. Traynor, The Riddle of Harmless Error 14,
50(1970). The doctrine also contributes to the finality that is
essential of punishment is to serve its intended purpose. Cf. Sanders
v. United States, 373 U.S. 1, 24-25(1963) (Harlan, J., dissenting).
Finally, it conserves scarce trial resources by eliminating burdensome
retrials where correction of the error is not likely to improve the
reliability of the verdict, but may instead inject new errors. See
Hasting, 461 U.S. at 509. In essence, by eliminating needless retrials
where it can be confidently said that no materially prejudicial error
occurred at trial, the harmless error doctrine allows an appellate court
"to keep the balance true" (Snyder v. Massachusetts, 291 U.S. 97,
122(1934) between the government's interest in convicting the guilty and
a defendent's interest in avoiding an unjust conviction. See Hasting,
461 U.S. at 509; cf. Strickland v. Washington, No. 82-1554 (May 14,
1984), slip op. 24.
At the same time, the Court has recognized that some constitutional
rights are so essential to a fair trial that their violation warrants
reversal in every case. Chapman itself gave three examples of such
infractions. 386 U.S. at 23 n.8. Two of them -- compelling a defendant
to stand trial before a trier of fact with a financial interest in the
outcome (see Tumey v. Ohio, 273 U.S. 510(1927) and denying a defendant
the assistance of counsel at trial (see Gideon v. Wainwright, 372 U.S.
335(1963) -- strike at the heart of the modern concept of a criminal
trial. Their deprivation is properly viewed as functionally equivalent
to the denial of any trial at all. Cf. Drope v. Missouri, 420 U.S. 162,
171 (1975) (forcing a mentally-incompetent defendant to stand trial);
Moore v. Dempsey, 261 U.S. 86, 91(1923) (mob dominated trial). /18/ The
third error listed in Chapman -- introducing a coerced confession (see
Payne v. Arkansas, 356 U.S. 560 (1958) -- is likely to call into
question the reliability of the verdict in a manner that is not
susceptible to effective appellate review. /19/ An error like this
necessarily undermines an appellate court's confidence in the accuracy
of the verdict. Cf. Strickland, slip op. 24.
The final category of errors that, once found to have occurred, will
not be separately analyzed for harmlessness is defined in an entirely
different manner. This class consists of those matters as to which
prejudice is considered in the determination whether there has been an
error. See, e.g., United States v. Bagley, No. 84-48 (July 2, 1985),
slip op. 10-15 (Brady claims); Strickland slip op. 21-16 (ineffective
assistance of counsel); United States v. MacDonald, 435 U.S. 850, 858
(1978) (speedy trial claims); cf. United States v. Young, No. 83-469
(Feb. 20, 1985), slip op. 15 n.14 (plain error under Fed.R.
Crim.P.52(b)). The reason that such claims cannot be harmless, of
course, is that it is pointless to undertake a separate inquiry into
whether an error had an effect upon the outcome of the trial if the
defendant has already established that the error was prejudicial in
demonstrating its existence. /20/
B. The question in this case is how the error at respondent's trial
should be classified. In ruling that this error could not be harmless,
the Delaware Supreme Court did not suggest that it could be classified
among those constitutional violations that so far deny a defendant the
essentials of a trial as to preclude a valid conviction without regard
to the reliability of the verdict. That argument also could not be
seriously entertained; perhaps a blanket refusal to permit
cross-examination of all prosecution witnesses could rationally be
treated like a deprivation of counsel or trial before a biased tribunal,
but an erroneous restriction upon defense cross-examination of a
particular witness simply cannot be equated with denial of a meaningful
trial. Nor can the right to impeach a prosecution witness be compared
with something like the right to represent oneself at trial, which is a
right wholly unrelated to the reliability of the verdict. The purpose
of the Confrontation Clause is to enchance the reliability of the
factfinding process at trial, not to recognize the dignity and autonomy
of the accused. See, e.g., Tennessee v. Street, No. 83-2143 (May 13,
1985), slip op. 6; Ohio v. Roberts, 448 U.S. 56, 63-64(1980).
In determining that the Confrontation Cluse was violated, the
Delaware Supreme Court did not require respondent to demonstrate that
the restriction actually and materially prejudiced his defense. /21/
This case therefore also does not fit into the category in which any
error that occurs cannot be harmless because prejudice has already been
found in determining that an error existed.
Because the purpose of the Sixth Amendment is to ensure that a
defendant receives a fair trial, there is generally no reason to award a
defendant relief based upon a Sixth Amendment claim absent a showing of
a likely adverse effect upon the reliability of the trial process. See
United States v. Cronic, No. 82-660 (May 14, 1984), slip op. 9-10;
Morrison, 449 U.S. at 364-365. /22/ Ordinarily, this means that only
those errors that may have produced an inaccurate guilty verdict in the
case at hand require that a defendant receive a new trial. See, e.g.,
Strickland, slip op. 21-26. /23/ The decisions of this Court leave open
the possibility that a showing of actual prejudice from an erroneous
restriction on cross-examination is intrinsic to finding a Confrontation
Clause violation in the first place. /24/ Indeed, a substantial
argument, supported by the mode of analysis employed in Davis v. Alaska,
can be presented in favor of that rule. /25/
Respondent's conviction was reversed without regard to the existence
of case-specific prejudice arising from the restriction upon his
cross-examination of Fleetwood. There is accordingly no need to decide
in this case whether, if prejudice must in fact be shown in order to
justify reversal, that is because prejudice is an element of the finding
of error itself or because it is the criterion for deciding whether the
error requires reversal. /26/ Rather, the question here is whether a
non-prejudicial restriction on cross-examination compels reversal. The
affirmative answer of the Delaware Supreme Court would be correct only
if this type of error is so inherently prejudicial and so indeterminate
that it necessarily undermines an appellate court's confidence in the
reliability of the verdict and prevents the court from effectively
determining whether the error had an adverse effect on the trial. For
the following reasons, we believe that the denial of an opportunity to
impeach a prosecution witness regarding bias does not fit into this
category.
1. The Delaware Supreme Court relied in part on Davis v. Alaska,
supra, for its holding that this error was prejudicial per se. Pet.App.
A6. Properly read, however, Davis does not stand for the proposition
that an erroneous restriction on defense cross-examination or
impeachment invariably requires reversal and does not preclude
application of the harmless error doctrine in this case.
Davis is best understood in light of its facts. Davis was charged
with burglary and grand larceny for the theft of a safe from a bar. The
police found the safe that Davis had allegedly stolen abandoned near
Richard Green's home, and Green, the only eyewitness, testified that he
had seen Davis near this site on the day of the crime. The defense was
forbidden from eliciting on cross-examination of Green that he was on
juvenile probation for burglary both at the time of the offense charged
against Davis and at the time of the trial. The defense sought to
reveal this fact to impeach Green for bias by showing that he may have
slanted his testimony in the State's favor to shift suspicion away from
himself or to avoid revocation of his probation for not "cooperating"
with the prosecutor. 415 U.S. at 310-311. The trial judge, relying
upon state rules forbidding the disclosure of juvenile records in
judicial proceedings, prohibited questioning that would disclose Green's
juvenile record. Id. at 311.
This Court reversed Davis's conviction. It explained that the
opportunity to impeach a prosecution witness by showing that he is
biased against the defendant is an integral component of confrontation.
415 U.S. at 315-317. Emphasizing the importance of Green's testimony to
the prosecution's case and relying upon Alford v. United States, 282
U.S. 687 (1931), which had upheld a defendant's right to disclose that a
witness was being detained in custody at the time of trial and thus may
have been eager to "cooperate" with the government in order to have his
own charges dropped or reduced, the Court held that denying Davis any
opportunity to reveal that Green was on probation required that Davis's
conviction be set aside. 415 U.S. at 317-318, 320-321. As the Court
concluded, "(Davis) was thus denied the right of effective
cross-examination which 'would be constitutional error of the first
magnitude and no amount of showing of want of prejudice would cure it.'"
415 U.S. at 318 (quoting Smith v. Illinois, 390 U.S. 129, 131 (1968),
quoting in turn Brookhart v. Janis, 384 U.S. 1, 3, (1966)).
In so ruling, however, Davis did not except from the operation of the
harmless error rule all cases in which a defendant is denied an
opportunity to impeach a witness for bias. In the first place, Davis
did not cite, much less distinguish, Chapman or any of the Court's other
harmless error decisions. Nor did Davis explain why an erroneous
limitation upon the extent of defense cross-examination should be
treated differently from a complete denial of any opportunity to
cross-examine an adverse witness, which the Court had previously held
could be harmless in an appropriate case. See pages 22-24, infra. In
short, Davis can hardly be read to have rejected the application of a
well established doctrine that the Court never discussed. /27/
This is especially so since the dictum in Davis came only after the
Court had painstakingly demonstrated how Davis had in fact been
prejudiced. The Court repeatedly stressed the pivotal nature of Green's
testimony and the fact that any substantial blow to his credibility
would have been fatal to the state's case. Not only was "Green * * * a
crucial witness for the prosecution" (415 U.S. at 310), but also
"Green's testimony * * * provided 'a crucial link in the proof . . . of
(Davis') act'" (id. at 317 (citation omitted); see id. at 319).
Moreover, given the fact that the proof of Davis's guilt otherwise
rested entirely upon circumstantial evidence (id. at 310), "(t)he
accuracy and truthfulness of Green's testimony were key elements in the
State's case against (Davis)" (id. at 317). For that reason, the Court
observed that "(s)erious damage to the strength of the State's case
would have been a real possibility had petitioner been allowed to
(impeach Green)." Id. at 319. Finally, the Court observed that there
was a strong possibility that Green had chosen to rely upon the trial
court's ruling that his juvenile records could not be disclosed as a
means of giving testimony that "can be regarded as highly suspect at the
very least" (id. at 314; see id. at 313-314). /28/ Accordingly, even
read with liberality, Davis simply holds that the Confrontation Clause
entitles a defendant to disclose a known bias on the part of an adverse
witness and that denying Davis any such opportunity was prejudicial to
his defense.
2. The Court's decisions clearly support the proposition that a
Confrontation Clause violation like the one that occurred here can be
harmless in an appropriate case. Indeed, that principle was first
articulated more than 80 years ago in Motes v. United States, 178 U.S.
458 (1900). There, the government introduced at trial against all of
the codefendants a transcript of the testimony given at a preliminary
hearing by a witness whom the government had allowed to escape from
custody prior to trial. The admission of this statement, the Court
held, denied each defendant the opportunity to cross-examine the
declarant, in violation of the Confrontation Clause, requiring reversal
of the convictions of all but one co-defendant. 178 U.S. at 471-474.
Reversal of that defendant's conviction was unwarranted, however,
because he had testified at trial that he was solely responsible for the
crime. Id. at 474-475. Given his sworn admission of his guilt, the
Court held that "(i)t would be trifling with the administration of the
criminal law to award him a new trial" (id. at 476).
More recently, the Court has reaffirmed that principle in a series of
cases holding that the admission of a confession made by a nontestifying
co-defendant, in violation of Bruton v. United States, 391 U.S. 123
(1968), can be harmless. Bruton held that the receipt in evidence at a
joint trial of a confession made by a nontestifying co-defendent that
incriminated Bruton but was inadmissible as to him deprived Bruton of
the right to cross-examine an adverse witness. However, despite the
fact that admission of such potentially unreliable evidence could have a
"devastating" effect (391 U.S. at 136), constitutes a "'serious flaw()
in the fact-finding process at trial'" (Roberts v. Russell, 392 U.S.
293, 294 (1968) (citation omitted)), and poses "a serious risk that the
issue of guilt or innocence may not have been reliably determined" (id.
at 295), the Court has made clear that the admission of a co-defendant's
confession in violation of Bruton does not invariably call for reversal.
For instance, in Harrington v. California, 395 U.S. 250 (1969), the
Court found harmless the receipt of two confessions made by
nontestifying co-defendants because the proof of the defendant's guilt
was "so overwhelming" that the error was necessarily harmless unless
every constitutional error requires reversal, a proposition that the
Court had squarely rejected in Chapman. See 395 U.S. at 254. Schneble
v. Florida, 405 U.S. 427 (1972), and Brown v. United States, 411 U.S.
223 (1973), expressly reaffirmed the ruling in Harrington. See also
Parker v. Randolph, 442 U.S. 62, 77-81 (1979) (opinion of Blackmun, J.);
Dutton v. Evans, 400 U.S. 74, 91-93 (1970) (Blackmun, J., concurring);
cf. Rushen v. Spain, 464 U.S. at 117-118 & n.2 (right to presence at
trial). As a plurality of the Court summarized in Parker, "(i)n some
cases, the properly admitted evidence of guilt is so overwhelming, and
the prejudicial effect of the codefendant's admission so insignificant
by comparison, that it is clear beyond a reasonable doubt that
introduction of the admission at trial was harmless error." 442 U.S. at
70-71 (footnote omitted). Motes and Harrington thus make clear that
denial of the opportunity to cross-examine an adverse witness does not
fit within the limited category of constitutional errors that must be
deemed prejudicial in every case. /29/
3. There is a sound basis for this result. Whether denying a
defendant the opportunity to impeach an adverse witness will affect the
reliability of the fact-finding process at trial is dependent upon a
host of factors, such as the nature of the testimony that the witness
has given, the nature and strength of the basis for challenging his
credibility, the extent of cross-examination otherwise afforded the
defendent, the presence or absence of corroboration provided by the
testimony of other witnesses or by physical or documentary evidence, and
the defense offered at trial. /30/ Showing that a witness is under
indictment for an unrelated crime or is personally related to the victim
may be invaluable where the case is simply a "swearing match" between
the witness and the defendant; it will be of little or no benefit where
the government's proof rests largely upon documentary or physical
evidence or where the defense is insanity. Impeaching a witness's
credibility may be devastating to the state's case where that witness's
testimony is the only evidence establishing an essential element of the
crime, but discrediting that witness will have little effect on the jury
if his testimony is cumulative of the testimony given by several other
witnesses.
The government will also often have a substantial interest in keeping
secret certain facts that the defense wishes to elicit on
cross-examination, such as the identity of a government informant, and
it is well-settled that the strength of the government's countervailing
interest must be considered before disclosure can be ordered. /31/ The
Court has recognized, however, that there is no reason to set aside a
conviction if disclosure of the sought-after information could not have
contributed materially to the verdict. /32/ Indeed, even where the
government has no such interest in secrecy and simply fails to reveal
exculpatory information requested by the defense, including material
useful to impeach a witness, reversal is not required unless the
sought-after information is likely to have affected the outcome of the
trial. /33/ It necessarily follows from these cases that it cannot
rationally be presumed that the refusal to allow the defense to impeach
an adverse witness with such information is so prejudicial as to call
for reversal without any consideration of the weight of the state's
proof or the significance to the case of the witness's testimony.
To be sure, cross-examination is the primary Confrontation Clause
guarantee (see, e.g, Pointer v. Texas, 380 U.S. 400, 407 (1965)) and
functions, in Dean Wigmore's oft-quoted phrase, as the "greatest legal
engine ever invented for the discovery of truth" (5J. Wigmore, Evidence
in Trials at Common Law Section 1367, at 32 (1974)). Impeaching a
witness for bias is also a powerful means of discrediting his testimony.
See United States v. Abel, No. 83-935 (Dec. 10, 1984), slip op. 6-9;
Davis, 415 U.S. at 316; Alford, 282 U.S. at 692. Nonetheless,
impeachment is simply a means to an end and is not valued for its own
sake. /34/ A defendant who is denied the opportunity to impeach a
prosecution witness still has several other means at his disposal to
convince the jury that the witness should be disbelieved, /35/ and the
ability to use those alternatives before the jury must be given weight
under the Confrontation Clause. /36/ The evidence against a defendant
in a given case may also be so overwhelming or, as here, the testimony
offered by a particular witness so slight, uncontroversial, or amply
corroborated that an appellate court can safely say that the
restriction, even if erroneous, has not denied the defendant a fair
opportunity to establish his innocence. Accordingly, there is no basis
for concluding that an erroneous restriction upon defense
cross-examination is inherently prejudicial.
CONCLUSION
The judgment of the Supreme Court of Delaware should be reversed.
Respectfully submitted.
CHARLES FRIED
Acting Solicitor General
STEPHEN S. TROTT
Assistant Attorney General
ANDREW L. FREY
Deputy Solicitor General
PAUL J. LARKIN, JR.
Assistant to the Solicitor General
VINCENT L. GAMBALE
Attorney
SEPTEMBER 1985
/1/ December 31 was the final day of Fleetwood's lease and the
electricity had been turned off in his apartment (II Tr. 118, 130; III
Tr. 51; IV Tr. 43, 50). Epps had briefly visited Fleetwood's apartment
earlier that day to inquire about renting it (II Tr. 118, 123; III Tr.
45; J. A. 78-79).
/2/ Meinier did so because Fleetwood's apartment was without
electricity and she had no way of knowing how close it was to the New
Year (J.A. 125).
/3/ The medical examiner estimated that Epps died between midnight
and 1:00 a.m. (II Tr. 66, 80). Epps had been stabbed or cut 18 times in
various parts of her body, including the neck, heart, and vaginal area,
and one of the wounds was a two-foot long incision from the top of the
sternum to the pelvic area (id. at 58-65).
/4/ Pregent also made two tape-recorded statements following his
arrest, which were admitted at trial without objection. In essence,
Pregent stated that, after talking with respondent and after respondent
had "stretched out on some cushions," he (Pregent) fell asleep on the
sofa bed next to the fully-clothed Epps. Pregent could not sleep and
went to Fleetwood's apartment. When Pregent returned five minutes
later, respondent was still awake and "everything was still fine."
Pregent then got back into bed with Epps and "the next thing (he) knew,"
he was awakened and arrested by the police. Pregent claimed that he did
not see Epps' body until after he had been taken into custody, and he
denied having had sexual intercourse with Epps (J.A. 43-44, 46-47,
52-53, 56).
/5/ The pants Pregent was wearing when he was arrested were stained
with blood on the bottom of both legs, indicating that he had stepped
into a large pool of blood (VI Tr. 54-65, l09). A T-shirt recovered
from a water-filled trash can in Pregent's bathroom had a "very little
bloodstain" that was "smeared and diluted"; a towel and a pair of socks
also found in the trash can had no blood on them (VI Tr. 65-67; VII Tr.
43; IX 37-38).
/6/ Based on the physical evidence and photographs of the crime
scene, the forensic expert concluded that Epps had initially been
stabbed in Pregent's living room next to the sofa bed, that she then
fell or was placed on the sofa bed and later dragged by her shoulders
into Pregent's kitchen, where additional wounds were inflicted (VI Tr.
89-96). Epps' pants and panties had been removed before she was
assaulted (id. at 105).
/7/ Fleetwood stated that on the night of the crime he gave a
statement to the police that was basically identical to his trial
testimony (J.A. 105).
/8/ Additional voir dire was conducted regarding the unrelated
homicide; Fleetwood denied that he had been offered any favors,
inducement, promises, or deals with respect to that homicide in exchange
for his testimony at respondent's trial (III Tr. 85-86; J.A. 113-114).
/9/ Respondent testified that he had known Fleetwood and Pregent for
a few years and that Pregent had invited him to the party when they met
on the street on the afternoon of December 31, 1981 (X Tr. 26).
Respondent went to Pregent's apartment, met Epps for the first time, and
left after 30 minutes (id. at 28-31). Later that afternoon, respondent
returned to the party, but left a short time later (id. at 33-35).
After spending the remainder of the afternoon, and evening drinking with
friends, respondent returned to Pregent's apartment (id. at 35-41; J.A.
140-142).
/10/ No seminal fluid was found on any item of Epps' clothing or on
any other item that was analyzed by the government's forensic expert (VI
Tr. 42). The medical examiner did not state, however, whether any fluid
was found on Epps (II Tr. 52-88).
/11/ One of the arresting officers testified that he saw no bruises
on respondent (X Tr. 95).
/12/ For example, counsel told the jury in closing argument (XI Tr.
32, 42; J.A. 181, 188-189):
The defense does not dispute that (respondent) was in Daniel
Pregent's apartment and (respondent) was there when the murder
occurred. There is no dispute about that.
(Fleetwood's testimony) proves what (respondent) has never denied.
It proves what (respondent) has already testified to in this
trial. It proves that (respondent) was at Danny Pregent's
apartment before Doris Epps was murdered.
/13/ The court left open the question whether denying respondent an
opportunity to cross-examine Fleetwood about the unrelated homicide
investigation was also erroneous. Pet. App. A6 n3.
/14/ Given respondent's admissions to the police and defense
counsel's statements at trial (see pages 4-5, 8, supra), respondent
cannot claim that his trial testimony was in any way the "fruit" of the
restriction on cross-examination. Cf. Harrison v. United States, 392
U.S. 219 (1968).
/15/ Since Davis was decided, most federal and state courts have held
that an erroneous restriction on a defendant's opportunity to
cross-examine an adverse witness can be harmless. See, e.g., United
State v. Garza, 754 F.2d 1202, 1206-1208 (5th Cir. 1985); United States
v. Smith, 748 F.2d 1091, 1096 (6th Cir. 1984); Carrillo v. Perkins, 723
F.2d 1165, 1170-1173 (5th Cir. 1984); United States v. Whitt, 718 F.2d
1494, 1501-1502 (10th Cir. 1983); United States ex rel. Scarpelli v.
George, 687 F.2d 1012, 1013-1014 (7th Cir. 1982), cert. denied, 459 U.S.
1171 (1983); Kines v. Butterworth, 669 F.2d 6, 11-13 (1st Cir. 1981),
cert. denied, 456 U.S. 980 (1982); United States v. Gambler, 662 F.2d
834, 840-842 (D.C. Cir. 1981); United States v. Duhart, 511 F.2d 7,
9-10 (6th Cir.), cert. dismissed, 421 U.S. 1006 (1975); Snyder v.
Coiner, 510 F.2d 224, 227-229 (4th Cir. 1975); Ransey v. State, 680
P.2d 596, 597-598 (Nev. 1984); State v. Patterson, 656 P.2d 438, 439
(Utah 1982); State v. Pierce, 64 Ohio St. 2d 281, 414 N.E.2d 1038,
1043-1044 (1980); cf. Commonwealth v. Wilson, 381 Mass. 90, 407 N.E.2d
1229, 1247 (1980); Contra, State v. Parillo, 480 A.2d 1349, 1357-1358
(R.I. 1984). The Ninth Circuit is in disarray on this issue. Compare
United States v. Uramoto, 638 F.2d 84, 87 (9th Cir. 1980) (stating that
such errors cannot be harmless; witness in that case was crucial,
however), with United States v. Price, 577 F.2d 1356, 1362-1364 (9th
Cir. 1978), cert. denied 439 U.S. 1068 (1979)(such errors can be
harmless); Patterson v. McCarthy, 581 F.2d 220, 221-222 (9th Cir. 1978)
(finding that error was not harmless in that case). See also United
States v. Jackson, 756 F.2d 703, 706 (9th Cir. 1985)(per se rule of
reversal stated in Uramoto may be limited to denial of cross-examination
that is prejudicial); United States v. Williams, 668 F.2d 1064, 1070 &
n.14 (9th Cir. 1981)(noting "disharmony in this circuit surrounding that
issue"; reversing on ground that restriction was prejudicial); Chipman
v. Mercer, 628 F.2d 528, 533 (9th Cir. 1980)(erroneous restriction on
cross-examination cannot be harmless because defendant must show that
the verdict was adversely affected by the restriction to establish a
Confrontation Clause violation).
/16/ As the Court explained in Hasting, the harmless error doctrine
recognizes that, "given the myriad safeguards proved to assure a fair
trial, and taking into account the reality of the human fallibility of
the participants, there can be no such thing as an error-free, perfect
trial, and that the Constitution does not guarantee such a trial." 461
U.S. at 508-509.
/17/ See e.g., Rushen v. Spain, supra (right to be present at trial);
United States v. Hasting, supra (improper comment on defendant's
silence at trial, in violation of Self-Incrimination Clause); Hopper v.
Evans, 456 U.S. 605, 613-614 (1982)(statute improperly forbidding court
from giving a jury instruction on a lesser included offense in a capital
case, in violation of Due Process Clause); Moore v. Illinois, 434 U.S.
220, 232 (1977) (admission of identification in violation of Sixth
Amendment Counsel Clause); Brown v. United States, 411 U.S. 223,
231-232 (1973) (admission of out-of-court statement in violation of
Sixth Amendment Confrontation Clause); Milton v. Wainwright, 407 U.S.
371 (1972) (admission of confession in violation of Sixth Amendment
Counsel Clause); Chambers v. Maroney, 399 U.S. 42, 52-53 (1970)
(admission of evidence obtained in violation of Fourth Amendment);
Coleman v. Alabama, 399 U.S. 1 (1970) (denial of right to counsel at a
preliminary hearing in violation of Sixth Amendment Counsel Clause).
/18/ Analogous errors or deprivations that would appear to require
reversal without regard to their specific impact upon the defendant's
trial include denying the defendent a jury trial, trying a defendant in
absentia, or refusing to permit the defendant to testify. Whether a
jury charge that shifts the burden of proof on, or conclusively
presumes, the issue of the defendant's intent fits into this category
hinges upon whether such a charge amounts to a directed verdict of
guilty. Compare Connecticut v. Johnson, 460 U.S. 73, 84-88 (1983)
(plurality opinion), with id. at 94-102 (Powell, J., dissenting). The
right to represent oneself at trial (see Faretta v. California, 422 U.S.
806 (1975)) may also fit into this category, since it is an integral
component of the defendant's right to present a defense at trial (id. at
818-821), or it may be in a class by itself, because the right exists,
in part at least, "to affirm the dignity and autonomy of the accused"
(McKaskle v. Wiggins, No. 82-1135 (Jan. 23, 1984), slip op. 7; id. at
9; id. at 11 n.6 (White, J., dissenting)).
/19/ As Justice Harlan put it in Chapman, "particular types of errors
have an effect which is so devastating or inherently indeterminate that
as a matter of law they cannot resonably be found harmless." 386 U.S. at
52 n.7 (dissenting opinion). However, given the Court's subsequent
decisions in Harrington v. California, 395 U.S. 250 (1969), and Milton
v. Wainwright, supra, it is presently unclear whether the admission of a
coerced confession still fits into the class of errors that cannot be
harmless.
Justice Harlan also suggested that "certain types of official
misbehavior" should warrant reversal in every case to indicate society's
disapproval of "such intentional misconduct." 386 U.S. at 52 n.7
(dissenting opinion). However, the Court rejected that approach to the
harmless error doctrine in Hasting. See 461 U.S. at 507 ("the interests
preserved by the doctrine of harmless error cannot be so lightly and
casually ignored in order to chastise what the court viewed as
prosecutorial overreaching"); see also Mabry v. Johnson, No. 83-328
(June 11, 1984), slip op. 6; Smith v. Phillips, 455 U.S. 209, 219
(1982).
/20/ While a showing of prejudice is not an absolute prerequisite to
establishing a speedy trial violation (see Barker v. Wingo, 407 U.S.
514, 530-533 (1972)), because consideration of prejudice nonetheless
plays a central role in evaluating the claim, this too is a category of
cases in which it is fair to say that the harmless error principle has
in effect been folded into the decision whether there was an error.
/21/ Although the court stated that "(t)he question of bias was an
important issue before the (trial) court and the excluded evidence was
central to that issue" (Pet. App. A5-A6 (footnote omitted)), the court
did not suggest that the error was likely to have adversely affected the
verdict, which is an essential aspect of either a determination of
prejudice or harmless error analysis. Compare, e.g., Strickland, slip
op. 29-30, with Hasting, 461 U.S. at 512.
/22/ In several contexts, the Court has ruled that a showing of
prejudice is necessary before there is a Sixth Amendment violation or
before a defendant can obtain relief. See, e.g., Strickland, slip op.
21 (ineffective assistance of counsel); United States v.
Valenzuela-Bernal, 458 U.S. 858, 872-874 (1982) (claimed violation of
Compulsory Process Clause based upon deportation of potential witness);
United States v. MacDonald, 435 U.S. 850, 858-859 (1978) (prejudice
important in showing a violation of Speedy Trial Clause); Weatherford
v. Bursey, 429 U.S. 545 (1977) (prejudice necessary to establish a
violation of Counsel Clause by using co-defendant as government
informant); see also Morrison, 449 U.S. at 364-365 (general rule is
that an error does not warrant setting aside a conviction absent a
demonstrable effect upon the outcome of the trial); see generally
Cronic, slip op. 9-10 ("we begin by recognizing that the right to the
effective assistance of counsel is recognized not for its own sake, but
because of the effect it has on the ability of the accused to receive a
fair trial. Absent some effect of the challenged conduct on the
reliability of the trial process, the Sixth Amendment guarantee is
generally not implicated"). The same principle also applies in a
variety of other contexts. See, e.g. United States v. Lovasco, 431 U.S.
783 (1977) (proof of actual prejudice required to establish a due
process violation arising from pre-indictment delay).
/23/ That is not always the case, however, denial of a jury trial
would not be harmless despite irrefutable proof of a defendant's guilt.
The purposes of each particular Sixth Amendment guarantee must be
examined to make this determination. The severity of the deprivation
may also be pertinent. A complete denial of counsel requires automatic
reversal without any showing of particularized prejudice (Gideon v.
Wainwright, 372 U.S. 335 (1963)), whereas claims that particular acts or
omissions of counsel deprived the defendant of his Sixth Amendment
rights are evaluated with specific reference to potential prejudice
(Strickland, slip op. 21-23; Cronic, slip op. 11 n.26).
/24/ The Court's decisions involving a direct restriction on a
defendant's cross-examination have involved situations in which either
the witness gave particularly incriminating testimony or the defendant
was denied the opportunity to elicit especially favorable exculpatory
evidence. See, e.g., Davis v. Alaska, 415 U.S. at 310-314, 317-318;
Chambers v. Mississippi, 410 U.S. 284, 291-293 (1973); Smith v.
Illinois, 390 U.S. 129, 130 (1968); Brookhart v. Janis, 384 U.S. 1, 2-4
(1966). Similarly, the Court's decisions regarding the admission of
statements, such as a confession or prior testimony, made by a declarant
not subject to cross-examination at trial have also involved highly
damaging evidence. See, e.g., Barber v. Page, 390 U.S. 719, 720 (1968);
Brookhart v. Janis, 384 U.S. at 2, 4; Douglas v. Alabama, 380 U.S.
415, 416-417, 419 (1965); Pointer v. Texas, 380 U.S. 400, 403 (1965).
Some older cases not involving the Confrontation Clause either state or
suggest that the denial of an opportunity for effective
cross-examination will itself be treated as prejudicial. See, e.g.,
Alford v. United States, 282 U.S. 687, 692 (1931); Tla-Koo-Yel-Lee v.
United States, 167 U.S. 274, 277-278 (1897). But, in these cases as
well, the witness involved gave damaging testimony against the accused,
and the defense had a strong reason for pursing a particular line of
inquiry. In any event, even if these decisions are read to suggest that
a defendant need not show that the restriction might have affected the
verdict to establish a confrontation violation, that would not foreclose
application of the harmless error doctrine.
/25/ A parallel could be drawn to the different rules governing cases
like Gideon, Powell v. Alabama, 287 U.S. 45 (1932), and Strickland.
There, the question whether prejudice would be presumed hinged upon the
degree of the interference with the right to counsel; here, the
question would turn upon the degree of interference with a defendant's
opportunity to confront adverse witnesses. The complete denial of any
opportunity to cross-examine any adverse witness would be conclusively
presumed to result in prejudice. See Brookhart, 384 U.S. at 3.
Prejudice would be presumed where a defendant was denied the
opportunity, in truth or effect, to cross-examine a crucial adverse
witness. See Davis v. Alaska, supra; Smith v. Illinois, supra; Alford
v. United States, supra. Finally, the denial of any opportunity to
cross-examine a minor witness (e.g., one who gave cumulative or
undisputed testimony) or a particular restriction upon cross-examination
of a more important witness would call for a case-specific inquiry into
potential prejudice. This case would appear to fit into the third
category.
/26/ This point could be of significance if the outcome of the case
depended upon an allocation of the burden of proof, since the
prosecution would likely be allocated the burden of showing harmless
error. Here, the State's effort to assume that burden was rejected by
the court below on the ground that a showing of harmlessness would not
avoid reversal.
/27/ Neither of the cases cited by Davis ruled that an erroneous
restriction on defense cross-examination cannot be harmless. In
Brookhart, the defendant was denied the opportunity to cross-examine any
of the state's witnesses, and the state offered a confession by an
absent witness. The only question was whether the defendant had waived
his confrontation right. 384 U.S. at 4-8. In Smith, the defendant was
denied the right to ask the crucial witness for the state his name and
address; that witness's testimony was pivotal because the only issue at
trial was the relative credibility of the witness and the defendant.
390 U.S. at 130. In both cases, the error was presumed to be
prejudicial; in neither case did the Court discuss whether erroneous
restrictions on cross-examination could be harmless in other contexts.
/28/ During cross-examination of Green, the defense elicited the fact
that Green had been questioned by the police regarding the burglary of
the bar. 415 U.S. at 312-313. However, Green denied that he had ever
before been subject to a similar interrogation. Id. at 313. The trial
court cut off any further questioning along this line on the basis of
its pretrial ruling regarding the admissibility of Green's juvenile
records. Ibid. This Court noted that, given the likelihood that Green
had been questioned in some manner in connection with his arrest for the
burglaries on which Green had been adjudged a juvenile delinquent, "it
is doubtful whether the bold 'No' answer would have been given by Green
absent a belief that he was shielded from traditional cross-examination"
(id. at 314).
/29/ Other decisions make the same point in a different but analogous
context. It is firmly settled that the deliberate use of perjured
testimony -- a more egregious impropriety than anything that occurred
here -- does not require that a conviction be set aside unless "there is
a () reasonable likelihood that the false testimony could have affected
the judgment of the jury." United States v. Agurs, 427 U.S. 97, 103
(1976) (footnote omitted); see United States v. Bagley, slip op. 10-12,
& nn.8-9; Giglio v. United States, 405 U.S. 150, 154 (1972) (quoting
Napue v. Illinois, 360 U.S. 264, 271 (1959)). This standard applies
both to the introduction of perjured testimony and to the knowing
failure to correct such testimony when offered by a witness (Giglio v.
United States, supra; Napue v. Illinois, supra). It appears
functionally equivalent to the Chapman harmless error standard. See
Bagley, slip op. 10-12 & n.9. If the government's failure to disclose
that a witness has committed perjury does not automatically require
reversal, it logically must follow that denying the defense the
opportunity to impeach a witness for bias should not automatically
require reversal.
/30/ See, e.g., United States v. Abel, No. 83-935 (Dec. 10, 1984),
slip op. 8-9; see generally McCormick's Handbook of the Law of Evidence
Section 40 (E. Cleary 2d ed. 1972) (describing various types of bias).
/31/ See, e.g., Davis v. Alaska, 415 U.S. at 319-321 (juvenile
records privilege); Smith v. Illinois, 390 U.S. at 133-134 (White, J.,
concurring) (inquiries "which tend to endanger the personal safety of
the witness" may be foreclosed); Roviaro v. United States, 353 U.S. 53
(1957) (informant's identity); United States v. Harley, 682 F.2d 1018,
1020-1021 (D.C. Cir. 1982) (surveillance location).
/32/ See United States v. Valenzuela-Bernal, 458 U.S. at 867-871;
Rugendorf v. United States, 376 U.S. 528, 534-536 (1964); Roviaro v.
United States, 353 U.S. at 64-65.
/33/ See United States v. Bagley, slip op. 10, 14-15; see also
California v. Trombetta, No. 83-305 (June 11, 1984), slip op. 9.
/34/ Cf. Cronic, slip op. 9-10 (right to counsel). That is clear
from the Court's rulings that certain types of hearsay statements are so
reliable that they can be admitted at trial despite the fact that the
defendant has had no opportunity to cross-examine the declarant. Ohio
v. Roberts, 448 U.S. at 66; see also Dutton v. Evans, 400 U.S. 74, 89
(1970) (plurality opinion); Mattox v. United States, 156 U.S. 237,
243-244 (1895); Fed. R. Evid, 803(1)-(23), 804(b)(1)-(4).
/35/ There are five chief methods of attacking a witness's
credibility: (1) showing a defect in his capacity to observe or
remember the subject of his testimony; (2) demonstrating the witness's
bias against the defendant or in favor of the government; (3)
introducing prior inconsistent statements made by the witness; (4)
showing that the witness's character is generally untrustworthy; and
(5) proving that the substance of his testimony is false. See
McCormick's Handbook of the Law of Evidence, supra, Section 33, at 66.
Denying a defendant the opportunity to impeach a witness for bias still
allows the defendant several other alternatives. For instance,
respondent was able to elicit the fact that Fleetwood was heavily
intoxicated that evening, which would have certainly affected
Fleetwood's ability to see respondent in Pregent's apartment (III Tr.
60).
/36/ Confrontation at trial helps to assure the reliability of the
factfinding process in several related ways. By requiring a witness to
testify under oath, it reinforces the seriousness of the proceeding,
makes lying more difficult, given the presence of the defendant, and
guards against prejury through the threat of a subsequent prosecution.
By permitting the accused to cross-examine a witness, it allows the
defendant to challenge the witness's credibility. And by permitting the
jury to assess the witness's demeanor, it provides the jury with the
opportunity to determine whether the witness is credible. See Ohio v.
Roberts, 448 U.S. at 63-64 & n.6; California v. Green, 399 U.S. 149,
158 (1970). The trial judge's ruling here, of course, limited
respondent's opportunity to convince the jury that Fleetwood was lying.
Nonetheless, because Fleetwood was on the stand and was subject to
cross-examination in other respects, the purposes confrontation serves
were not wholly vitiated by the trial court's ruling.