In the
United States Court of Appeals
For the Seventh Circuit
____________
No. 03-2092
THEODORE W. OSWALD,
Petitioner-Appellee,
v.
DANIEL BERTRAND,
Respondent-Appellant.
____________
Appeal from the United States District Court
for the Eastern District of Wisconsin.
No. 01-C-0182—Lynn Adelman, Judge.
____________
ARGUED NOVEMBER 3, 2003—DECIDED JUNE 29, 2004
____________
Before POSNER, DIANE P. WOOD, and EVANS, Circuit Judges.
POSNER, Circuit Judge. Oswald, a Wisconsin state prisoner,
sought federal habeas corpus after exhausting his state
remedies in State v. Oswald, 606 N.W.2d 207 (Wis. App.
1999), review denied, 609 N.W.2d 473 (Wis. 2000). The
district court found that the state court of appeals had been
unreasonable in ruling that the judge who presided at
Oswald’s criminal trial had conducted a constitutionally
adequate inquiry into possible jury bias. 249 F. Supp. 2d
1078 (E.D. Wis. 2003). The state appeals.
2 No. 03-2092
Ordinarily it would be clear that the issue for the district
court and us would be whether in turning down Oswald’s
claim of constitutional error the state courts had made “an
unreasonable application of clearly established Federal law,
as determined by the Supreme Court of the United States.”
28 U.S.C. § 2254(d)(1). But this is only if the prisoner’s claim
was adjudicated by the state court “on the merits.”
§ 2254(d). If not, the special deference to a state court’s
determinations that is prescribed by section 2254(d)(1) goes
by the board. Braun v. Powell, 227 F.3d 908, 916-17 (7th Cir.
2000); Moore v. Parke, 148 F.3d 705, 708 (7th Cir. 1998);
Maples v. Stegall, 340 F.3d 433, 436-37 (6th Cir. 2003). The
state appellate court discussed and disposed of Oswald’s
claim that the jury selection procedure used in his case had
denied him an impartial tribunal, but it did not discuss the
claim with reference to federal law. No matter. So long as
the standard it applied was as demanding as the federal
standard, Mitchell v. Esparza, 124 S. Ct. 7, 10 (2003) (per
curiam); Early v. Packer, 537 U.S. 3, 8 (2002) (per curiam);
Reid v. True, 349 F.3d 788, 799-800 (4th Cir. 2003); Sellan v.
Kuhlman, 261 F.3d 303, 311-14 (2d Cir. 2001)—and there is
no suggestion that it was not, cf. Hammill v. State, 278
N.W.2d 821, 822 (Wis. 1979)—the federal claim is deemed
adjudicated on the merits and its rejection therefore entitled
in this habeas corpus proceeding to the deference prescribed
by section 2254(d)(1).
The nature of Oswald’s claim has now to be explained.
The due process clause of the Fourteenth Amendment
entitles a state criminal defendant to an impartial jury,
Morgan v. Illinois, 504 U.S. 719, 726 (1992), which is to say a
jury that determines guilt on the basis of the judge’s in-
structions and the evidence introduced at trial, as distinct
from preconceptions or other extraneous sources of deci-
sion. Patton v. Yount, 467 U.S. 1025, 1037 n. 12 (1984); Irvin
v. Dowd, 366 U.S. 717, 721-23 (1961); United States v.
No. 03-2092 3
McClinton, 135 F.3d 1178, 1185-86 (7th Cir. 1998); United
States v. Angiulo, 897 F.2d 1169, 1182-83 (1st Cir. 1990). In
addition—and this is critical—the clause requires the trial
judge, if he becomes aware of a possible source of bias, to
“determine the circumstances, the impact thereof upon the
juror, and whether or not it was prejudicial.” Remmer v.
United States, 347 U.S. 227, 230 (1954); see also United States
v. Thomas, 463 F.2d 1061, 1063-64 (7th Cir. 1972); United
States v. Humphrey, 208 F.3d 1190, 1198-99 (10th Cir. 2000);
United States v. Davis, 177 F.3d 552, 556-57 (6th Cir. 1999);
Howard v. Moore, 131 F.3d 399, 422 (4th Cir. 1997) (en banc).
In Smith v. Phillips, 455 U.S. 209, 217 (1982), the Supreme
Court put the two points together, saying that “due process
means a jury capable and willing to decide the case solely
on the evidence before it, and a trial judge ever watchful to
prevent prejudicial occurrences and to determine the effect
of such occurrences when they happen.” The Court also
made clear in Smith that while Remmer had been a federal
prosecution, the duty of inquiry is equally engaged when a
defendant is tried in a state court. 455 U.S. at 215, 218; see
also Whitehead v. Cowan, 263 F.3d 708, 724-26 (7th Cir. 2001);
Evans v. Young, 854 F.2d 1081, 1083-84 (7th Cir. 1988); United
States v. Bradshaw, 281 F.3d 278, 289-93 (1st Cir. 2002);
Howard v. Moore, supra, 131 F.3d at 422.
In 1994, Oswald, who was then 18 years old, robbed a
bank with his father. Fleeing toward Waukesha by car, they
were stopped by two policemen. The two Oswalds, both
armed with semi-automatic rifles, shot at the officers, killing
one, and continued their flight, in the course of which they
took a woman hostage and forced her to drive them in her
van. At a police roadblock there was another shootout; the
hostage and two officers were wounded, the hostage
escaped, and the Oswalds took off in the van; eventually it
crashed and they were arrested. The second shootout, the
hostage’s escape, and the crash of the getaway car were all
4 No. 03-2092
videotaped, and broadcast throughout the Waukesha area.
As the district judge explained, “the case generated an
enormous amount of publicity both in the immediate
aftermath of the crime and during the period leading up to
the trials of the Oswalds. The serious nature of the offenses,
the fact that a local police officer was killed, the existence of
the videotape (with its echoes of the O. J. Simpson case) and
the fact that the defendants were father and son combined
to make the case probably the most notorious in the history
of Waukesha County.” The widow of the police officer who
had been killed initiated a highly publicized petition for
reinstatement of the death penalty in Wisconsin. Hundreds
of T-shirts and sweatshirts were sold, many to police and
prison guards, depicting the crashed van, bearing the
legend “Oswald’s final mistake was coming to the Town of
Pewaukee,” and calling for the reinstatement of the death
penalty in Wisconsin.
Oswald (the son, the petitioner in our case) was tried
separately from his father nine and a half months after their
crime spree and was convicted and given the redundant
sentence of 565 years in prison consecutive to two life
sentences. His father was tried separately and received a
comparable sentence. Because of the avalanche of pretrial
publicity, Oswald could doubtless have obtained a change
of venue had he moved for it, but he did not. His only
defense against the criminal charges was that his father had
coerced or brainwashed him into participating in the rob-
bery and subsequent mayhem, and, again in the words of
the district judge, Oswald’s lawyer thought that “because
some of the publicity had portrayed him as a victim of his
abusive and manipulative father, a local jury might be more
receptive to his defense than a jury elsewhere” in Wisconsin.
The court sent jury questionnaires to 156 residents of
Waukesha County, more than 80 percent of whom re-
sponded that on the basis of the media coverage of the
No. 03-2092 5
crime they thought that Oswald was guilty. Fifty of the 156
were voir dired, and of those 50, 29 were picked to be the
jury pool. Since under Wisconsin law each side would have
seven peremptory challenges (the usual number for crimes
punishable by life in prison is six, but since the judge
decided to seat three alternates he gave each side one more
peremptory challenge, Wis. Stat. Ann. §§ 972.03, .04(1)), if
all 14 were used there would still be 15 jurors—12 regulars
and the three alternates. Hence the cut to 29.
It took four days to winnow the list down first from 156 to
50 and then from 50 to 29, and during this period the
prospective jurors spent all day in a room in the courthouse
from which the bailiff would lead them out one by one to be
questioned in the courtroom. On the last day, one of the
prospects, Roger Klitzka, in the course of being voir dired,
said, “I know I’ve learned more in the last three days here
sitting down there in that room about this case than I have
since the day that it happened . . . . [A]ccording to what I
hear, the young man is guilty of what he is being accused of
and things like that and everything and I just think it’s just
a waste of time.” The judge asked him whether he meant
“it’s a waste to have the trial at all,” and Klitzka confirmed
that that was indeed what he meant.
Apparently this was not just Klitzka’s personal opinion
(he was not selected for the jury). The implication of what
he said was that the entire jury pool had made up its mind
that Oswald was guilty. We will never know for sure,
however, what he meant. Rightly fearing that the prospec-
tive jurors had been conducting marathon bull sessions on
Oswald’s guilt, Oswald’s lawyer asked Klitzka to explain
what he had heard in these sessions. Klitzka refused to
answer, and at the prosecutor’s urging the judge refused to
direct him to answer. Oswald’s lawyer requested the judge
to ask the other prospective jurors what had been discussed
6 No. 03-2092
in the jury room. The prosecutor (who made every possible
effort to prevent the judge’s determining whether jurors
were biased) objected, saying it was “crazy” to think that
the jurors had not discussed the case when they had been
sitting “for 4 days in a crowded jury room downstairs” in “a
case that everybody other than one juror that we have
talked about has seen a videotape concerning probably the
highest profiled case in this county in 20 years and not to
even think or mention the case or mention how disgusted
they are with the process.” Apparently the “process” that
the prosecutor thought must have disgusted the prospective
jurors was the voir dire itself, for he mentioned a prospec-
tive juror being “interrogated” and feeling as if “she’s on
trial.”
Even though the circumstances strongly suggested that
the jury had made up its mind that Oswald was guilty, the
judge refused either to question Klitzka further or to recall
for further questioning any of the prospective jurors who
had already been voir dired. He did permit Oswald’s law-
yer to inquire about the discussions in the jury room from
the three members of the pool who had not yet been voir
dired. The first of the three confirmed that the jurors had
been discussing the case; the other two denied having heard
other jurors express an opinion—which, of course, doesn’t
mean that other jurors had not expressed an opinion; those
two might not have been listening.
That afternoon the judge, who must by this time have
been worried about whether the impartiality of the prospec-
tive jurors had been compromised, asked one of the bailiffs
responsible for shepherding the jurors what he had over-
heard in the jury room. The bailiff said he hadn’t heard the
jurors talking about the case, but added that he had been in
the jury room only intermittently. And the jurors might
have been hesitant to discuss the case—having been in-
structed not to—in the presence of a court official.
No. 03-2092 7
During the lunch recess that day the judge had gone to the
jury room and reminded the prospective jurors not to
discuss the case. On that occasion one of the prospective
jurors, William Schuenke, had told the judge: “I really want
to talk to you. I had written that letter. It’s really important.
I don’t want to—I’m part of the 29” (who had survived
the first stage of the voir dire). In his letter Schuenke had
written: “I will more than likely be so concerned about my
work that I’d not give in trial all the attention it should
receive to the point that I might just vote either way just to
end it.” Oswald’s lawyer, who had seen the letter, wanted
to question Schuenke further. The judge refused, though
aware that Schuenke’s remark to him in the jury room had
been heard by the other prospective jurors. The judge knew
this because he’d been at the front of the room and
Schuenke at the back, with the other jurors in between, yet
the judge had heard him, so the jurors must have as well.
The judge had received notes from three other jurors as well
saying they didn’t want to serve on the jury. An oral request
was relayed to him by a fourth.
When the 29 prospective jurors who had survived the first
cut returned to the courtroom, Schuenke had another
outburst, saying to the judge in the hearing of the rest of the
jury pool, “I asked to speak to you 2 days ago. You mean I
can’t get five minutes of your time?” The judge told him his
request had been addressed, but he replied, “Not to me they
weren’t you know.” Schuenke served on the jury because
the defense used up all its seven peremptory challenges on
other prospective jurors whom the judge had refused to
excuse for cause.
The trial lasted three weeks. The jury, after deliberating
for three hours (according to press reports—the information
isn’t in the record), returned a verdict of guilty on all 19
counts in the indictment.
8 No. 03-2092
The circumstances that we have narrated demonstrate
a high probability that some, maybe all, of the jurors who
tried Oswald were biased. It is not just what Klitzka said or
what Schuenke said; those were merely the most dramatic
indications that, as the judge obviously realized, the process
of jury selection was being poisoned. The question
is whether, given the indications of jury bias, the judge’s in-
quiry was adequate. From the case law we distill the prin-
ciple that adequacy is a function of the probability of bias;
the greater that probability, the more searching the inquiry
needed to make reasonably sure that an unbiased jury is
impaneled. United States v. Davis, 15 F.3d 1393, 1412-13 (7th
Cir. 1994); Tracey v. Palmateer, 341 F.3d 1037, 1044 (9th Cir.
2003); United States v. Bradshaw, 281 F.3d 278, 289-90 (1st Cir.
2002); United States v. Angulo, 4 F.3d 843, 847 (9th Cir. 1993).
The response to the jury questionnaires against the back-
ground of enormous publicity concerning the most sensa-
tional criminal episode in the county’s history, the fact that
Oswald seemed so obviously guilty as to make the necessity
for a trial questionable to a layperson, the tumult induced
by Schuenke’s vocal complaints, the flagrant disobedience
of the judge’s instructions that the prospective jurors not
discuss the case in advance of the trial, the likelihood that
Schuenke and perhaps other reluctant jurors would vote to
convict regardless of their actual views if that would make
the trial end quicker, the fact that, at least according to
Klitzka, the improper discussions had already produced a
consensus that Oswald was guilty as charged— all these
things, taken not separately but together, created a suffi-
ciently high probability of jury bias to require on the part of
the trial judge a diligent inquiry.
The inquiry he conducted did not satisfy that standard.
Reminding the jurors not to discuss the case after they had
been discussing it for four days was a case of closing the
barn door after the horse had escaped. Questioning one of
No. 03-2092 9
the bailiffs (and why just one?) was not likely to elicit much
useful information about the mood and mindset of the
prospective jurors. Allowing Oswald’s lawyer to question
just three of the prospective jurors about conversations in
the jury room was arbitrary, especially since by cutting off
the questioning of Klitzka the judge had made clear that
he wouldn’t permit follow-up questions. The bobtailed in-
quiry that the judge conducted flunked the constitutional
test that “the investigation be reasonably calculated to
resolve the doubts raised about the juror’s impartiality.”
Dyer v. Calderon, 151 F.3d 970, 974-75 (9th Cir. 1998) (en
banc); see also Mu’Min v. Virginia, 500 U.S. 415, 419-22, 430-
32 (1991); United States v. Beckner, 69 F.3d 1290, 1291-94 (5th
Cir. 1995); United States v. Boylan, 898 F.2d 230, 257-59 (1st
Cir. 1990); cf. Turner v. Murray, 476 U.S. 28, 35-36 (1986)
(plurality); Ham v. South Carolina, 409 U.S. 524, 525-27
(1973). To repeat, the greater the doubts, the more probing
the inquiry that is required.
It is true that by refusing to move for a change of venue,
the defense lawyer may have been trying to game the sys-
tem. Wisconsin law forbade the judge to order a change of
venue on his own or the prosecutor’s motion. Wis. Const.
art. 1, § 7; Wis. Stat. §§ 971.19(1), 971.22; State v. Bangert, 389
N.W.2d 12, 34-35 (Wis. 1986); State v. Mendoza, 258 N.W.2d
260, 266-69 (Wis. 1977). Oswald’s best hope may have been
that the trial might founder on inability to impanel an im-
partial jury in the county in which the crime had been com-
mitted. It is difficult to see how Oswald could prevail with
his Patty Hearst or Manchurian Candidate defense. Defenses
such as self-defense, public and private necessity, and
coercion are not intended for cases in which a person kills
an innocent to save his own hide, as when the shipwrecked
defendants in the famous case of Regina v. Dudley & Stevens,
14 Q.B.D. 273 (1884), ate the cabin boy rather than starve,
and were convicted of murder. See also United States v.
10 No. 03-2092
Buchanan, 529 F.2d 1148, 1153 (7th Cir. 1975); United States
v. LaFleur, 971 F.2d 200, 204-06 (9th Cir. 1991); Henry v. State,
586 So. 2d 1033, 1036 n. 6 (Fla. 1991) (per curiam), vacated
on other grounds, 505 U.S. 1216, on remand, 613 So. 2d 429
(Fla. 1992); Amin v. State, 811 P.2d 255, 260 (Wyo. 1991);
United States v. Holmes, 26 Fed. Cas. 360, 366-68 (C.C.E.D. Pa.
1842) (No. 15,383); State v. Cooper, 993 P.2d 745, 748 (N.M.
App. 1999); State v. Soine, 348 N.W.2d 824, 826 (Minn. App.
1984); State v. Banks, 539 P.2d 173, 175 (Ariz. App. 1975); 2
Wayne R. LaFave, Substantive Criminal Law §§ 9.7, 10.1(c),
10.4(g), pp. 72, 123-24, 157-58 and n. 82 (2d ed. 2003).
Wisconsin law permits a murderer to plead coercion as a
complete defense, provided he didn’t commit first-degree
murder, Wis. Stat. Ann. § 939.46(1)—and for reasons
unknown Oswald was not charged with first-degree murder
though he was tried for conspiracy to commit intentional
homicide. But there is an exception to the right to plead
coercion if the coercion was by the defendant’s
coconspirator, id., which is the only possibility here. In fact
Oswald’s theory is that he was brainwashed by his father
into committing the crimes. That is why we mentioned The
Manchurian Candidate, the movie based on a novel by
Richard Condon in which Laurence Harvey is brainwashed
by the Chinese Communists into becoming an assassin
narrowly prevented from killing a Presidential candidate by
Frank Sinatra.
If brainwashing is just a form, a particular instantiation,
of the defense of coercion, it is barred by the conspirator
exception; if brainwashing is a separate defense, it probably
is not recognized by the law, see Joshua Dressler, “Professor
Delgado’s ‘Brainwashing’ Defense: Courting a Determinist
Legal System,” 63 Minn. L. Rev. 335 (1979); and if it is
merely an effort to show that Oswald was somehow
incapable of forming an intent to kill, it would be highly
unlikely to persuade a jury. It is less surprising that the jury
No. 03-2092 11
rejected Oswald’s defense than that it was submitted to the
jury in the first place. The reason for the judge’s doing so, as
far as we can determine, is that Oswald denied that he had
conspired with his father—again something that it is
difficult to imagine a jury accepting.
But even if Oswald is certain to be convicted if he is
retried, this cannot justify our reversing the grant of his
petition for habeas corpus. Even a clearly guilty criminal is
entitled to be tried before an impartial tribunal, something
the jurors in this case may well have failed to understand.
Arizona v. Fulminante, 499 U.S. 279, 309-10 (1991); Irvin v.
Dowd, supra, 366 U.S. at 722; Tumey v. Ohio, 273 U.S. 510, 535
(1927); United States v. Spears, 558 F.2d 1296, 1297 (7th Cir.
1977); Coleman v. Kemp, 778 F.2d 1487, 1540-41 (11th Cir.
1985); United States v. Essex, 734 F.2d 832, 845-46 (D.C. Cir.
1984); State v. Baumruk, 85 S.W.3d 644, 650-51 (Mo. 2002). It
is one of the handful of rights of a criminal defendant that
is not subject to the doctrine of harmless error. For others
see Vasquez v. Hillery, 474 U.S. 254, 263-64 (1986) (plurality)
(racial discrimination in the selection of the grand jury);
Waller v. Georgia, 467 U.S. 39, 49-50 and n. 9 (1984) (right to
a public trial); McKaskle v. Wiggins, 465 U.S. 168, 177 n. 8
(1984) (right to self-representation); Gideon v. Wainwright,
372 U.S. 335, 343-45 (1963) (right to counsel).
We explained the thinking behind these exceptions to the
rule that harmless errors do not vitiate criminal convictions
in Walberg v. Israel, 766 F.2d 1071, 1074 (7th Cir. 1985): “If
the police, after arresting Walberg and obtaining an eye-
witness identification of him plus his confession, had taken
him directly to the penitentiary on the ground that a trial
would be a waste of time for someone so patently guilty, he
would be entitled to release on habeas corpus; he would
have been deprived of his liberty without due process
of law. The Constitution requires (unless the defendant
12 No. 03-2092
waives his rights) a certain modicum of adversary proce-
dure even if the outcome is a foregone conclusion because
the evidence of guilt is overwhelming.” Oswald was entitled
to be tried by a jury that undertook to decide his guilt on the
basis of the evidence rather than of what they had learned
from the pretrial publicity about the case and analyzed over
a four-day period while waiting for the trial to begin. It was
natural that having seen the videotape of Oswald’s crimes,
laypersons would assume his guilt was open and shut. But
a jury that decides guilt before the trial begins is little better
than a lynch mob.
The state appellate court’s condonation of the lack of a
minimally timely, minimally adequate, investigation was an
unreasonable application of Remmer v. United States and
Smith v. Phillips and is therefore subject to correction in this
habeas corpus proceeding. It is not as if the state appellate
court had given reasons for its decision that, even if we did
not find them persuasive, we could not pronounce unrea-
sonable. A state court can of course be wrong without being
unreasonable, and the reasonableness of a decision ordi-
narily cannot be assessed without considering the quality of
the court’s reasoning, cf. Middleton v. McNeil, 124 S. Ct. 1830,
1832-33 (2004) (per curiam); Price v. Vincent, 538 U.S. 634,
641-43 (2003); Woodford v. Visciotti, 537 U.S. 19, 24-27 (2002)
(per curiam), though the ultimate question, as we empha-
sized in Hennon v. Cooper, 109 F.3d 330, 334-35 (7th Cir.
1997), is not whether the state court gets a bad grade for the
quality of its analysis but, as the statute says, whether the
decision is an unreasonable application of federal law as
stated by the U.S. Supreme Court to the facts.
The appellate court’s discussion of the jury selection issue,
like the trial court’s investigation, was perfunctory. It is true
that the discussion takes up a number of pages in the
No. 03-2092 13
appellate court’s opinion. But most of them address two
issues that are peripheral to the question of the adequacy of
the trial court’s investigation. The first is the accuracy of the
testimony of one prospective juror about the discussions in
the jury room, a witness whom the district court found
credible though the state courts had not. The second is
whether Schuenke had lied in answering a question put to
him in the voir dire. We assume without having to decide
that the state courts resolved these issues correctly; neither
the testimony of the prospective juror in question, nor
Scheunke’s alleged lie, figured in our narrative of the
process by which the jury was selected. The appellate court
did say, in addition, that Oswald failed to prove that
Scheunke was prejudiced against him, and this may be true;
but Oswald was not given a chance to prove it—the inquiry
into Scheunke’s prejudice was too truncated. As far as the
narrative of the jury selection process is concerned, the
accuracy of which is not in question, the state appellate
court said nothing beyond the naked conclusion that the
judge had done enough. It mentioned Klitzka’s answers, but
did not address their bearing on the issue of jury prejudice.
The state does not argue that Oswald failed to present his
constitutional claim in his state appeal, and so the sparse-
ness of the state appellate court’s analysis cannot be excused
on that ground.
The fact that Oswald had a chance to present evidence at
a postconviction hearing does not alter our conclusion that
he was denied due process of law. The single sentence in its
opening brief that the state devoted to the adequacy of that
hearing is inadequate to preserve the issue in this court; it
has been waived. In any event we don’t think the
postconviction hearing would dissolve Oswald’s constitu-
tional claim even if the issue had been preserved. When the
issue of jury bias does not arise until after trial and convic-
tion, the trial judge cannot be faulted for having failed to
14 No. 03-2092
investigate it, and a postconviction hearing is an inevitable
and authorized substitute. Rushen v. Spain, 464 U.S. 114, 119
(1983) (per curiam); Smith v. Phillips, supra, 455 U.S. at 215-
18; id. at 221-24 (concurring opinion); Owen v. Duckworth,
727 F.2d 643, 646 (7th Cir. 1984) (per curiam); Fitzgerald v.
Greene, 150 F.3d 357, 364-65 (4th Cir. 1998); cf. United States
v. Bishawi, 272 F.3d 458, 462-63 (7th Cir. 2001). But as Dyer
v. Calderon, from which we quoted earlier, makes clear, if
the issue does surface during, or in this case before, trial, it
is the trial judge’s responsibility to conduct an adequate
investigation, given the unsatisfactory character of an
inquiry into jury bias after the trial is over and the defen-
dant convicted. Compare United States v. Bradshaw, supra,
281 F.3d at 289-93; United States v. Angiulo, supra, 897 F.2d at
1183-86. Having invested time in the trial and voted to
convict the defendant, jurors are unlikely to acknowledge
bias, or even to remember the particulars of the discussions
that they had had of the case before it was tried—their
recollections are likely to be overshadowed by what they
learned about the case during the trial. That is why cases
such as Bradshaw and Angiulo stress the importance of a
prompt determination, by the trial judge, at the jury-selec-
tion stage. Obviously if the problem arises or is discovered
later, it must be dealt with later; that is all that cases like
Rushen allow.
The order of the district court that Oswald be either
retried or released is therefore
AFFIRMED.
EVANS, Circuit Judge, dissenting. The majority, in bolster-
ing its decision, draws on an example we mentioned in
Walberg v. Israel, 766 F.2d 1071 (7th Cir. 1985), observing that
if police have an airtight case they nevertheless cannot take
an obviously guilty defendant “directly to the penitentiary
on the ground that a trial would be a waste of time for
someone so patently guilty.” But is this what happened
No. 03-2092 15
here? No. Is this even remotely close to what happened
here? No, again. Fifty prospective jurors were questioned
over 4 days; the state trial judge granted 21 out of 27 of
Oswald’s challenges for cause of potential jurors; the jury
heard from numerous witnesses and examined many
exhibits during the 3 weeks it took to complete the trial. Yet
today, the court orders new tickets to this old show for
reasons that, in my view, cannot support such a drastic step.
I agree with the majority’s announcement that AEDPA
requires us to grant deference, under 28 U.S.C. § 2254(d)(1),
to the 1999 decision of the Wisconsin Court of Appeals
rejecting Oswald’s direct state appeal. State v. Oswald, 606
N.W.2d 207. But after making that determination, the
majority proceeds first to offer its independent view of the
facts involving the jury selection process. It then concludes
that the Wisconsin Court of Appeals’ “condonation of the
lack of a minimally timely, minimally adequate, investiga-
tion was an unreasonable application of Remmer v. United
States and Smith v. Phillips, and is therefore subject to
correction in this habeas corpus proceeding.” Why is the
Wisconsin court’s conclusion unreasonable? The answer
seems to be that the majority considers the state court’s dis-
cussion of the issue to be “perfunctory.” This is where the
majority loses me. Giving the Wisconsin Court of Appeals’
decision the deference to which it is due under AEDPA, I
cannot agree that it was either perfunctory or unreasonable.
I would not order a do-over of Oswald’s trial.
The majority takes the drastic step of ordering a retrial
because it detects problems in the jury selection process
based on statements made by two potential jurors. First, it
is troubled by the statement of prospective juror Roger
Klitzka during voir dire. Klitzka, however, never made it
into the jury box for the trial. And his statement that
potential jurors were “discussing the case” (whatever that
means—it could cover things like how long the trial will
16 No. 03-2092
take and whether or not the jury would be locked up) found
no support, as the majority notes, from two of the three
potential jurors who were subsequently asked about it.
More importantly, his statement about waiting room
conversations does not, in my view, cast doubt on the
willingness of the jurors actually selected for service to give
Oswald a fair shake on his defense.
Second, the majority is concerned about juror William
Schuenke who, like a lot of people summoned for what will
be lengthy jury service, preferred to be elsewhere. He said
he was “concerned about my work” and might not be able
to give the “trial all the attention it should receive . . . .”
Schuenke never said he rejected out-of-hand Oswald’s
potential defense of coercion. He was asked, “Do you feel
like you have an open mind with respect to any possible
defenses that I might offer for my client?” His answer was
“Yeah.” The follow-up question was, “I guess that’s all I’m
trying to find out is whether based on what you’ve read or
heard you feel that there’s something you wouldn’t listen to
or wouldn’t even consider if it was evidence in the case?”
The answer: “No.”
From the minor problems with Klitzka and Schuenke,
mere snippets of events over a 4-day period, the majority
jumps to the unfounded conclusion that there is “a high
probability that some, maybe all, of the jurors who tried
Oswald were biased.” On this flimsy foundation, the ma-
jority strains to build a case for requiring a retrial.
From what the majority says, one might conclude that the
Wisconsin Court of Appeals brushed off the issue involving
juror bias with a paragraph or two. The fact is that the court
spent 18 pages on jury issues. I would be the last person to
say the length equals quality, but length certainly seems to
indicate a lack of perfunctoriness. Unlike my colleagues, the
No. 03-2092 17
Wisconsin court discussed in some detail the postconviction
hearing on Oswald’s motion for a new trial, which included
the jury issue.
The Wisconsin Court of Appeals set the matter in context,
noting that this was an unusual case “not only because of
the enormous media coverage of the charged criminal
activities, but also because of defense counsel’s plan to turn
the media coverage into a positive factor for his client.” The
media consistently portrayed Oswald as an “impressionable
teenager who was victimized by his abusive and manipulat-
ing father.” The defense strategy, the court said, was that
jurors who had heard and seen the extensive media cover-
age might be sympathetic to him. Given the facts of this
case, I can’t disagree with Oswald’s lawyer’s belief that
sympathy, although at best a Hail Mary defense, was
probably his client’s best chance for any sort of success.
Oswald’s focus during the jury selection process was not
on whether the prospective jurors thought that he was
“involved,” to use the carefully chosen words of Juror
Schuenke, or “guilty,” meaning much the same as “in-
volved” to lay jurors. In fact, a videotape showing much of
the drama was going to be presented as evidence at trial.
Oswald was not going to claim that he was sitting in the
bleachers at Wrigley Field at the time of the shoot-out. He
was not going to argue that a twin brother, not he, was
“involved.” Rather, he was planning to minimize his role by
placing responsibility for the crimes on his father. He was
looking for jurors who would be open-minded to his
defense that his father made him do it. For instance, a juror
was asked whether her feelings would prevent her “from
considering a defense of coercion if the Judge instructed
[her] to consider it?” Whether that strategy would work is
not an issue for us, yet I certainly agree with the majority’s
observation that any such “Manchurian Candidate” defense
would be “highly unlikely to persuade a jury.”
18 No. 03-2092
Given Oswald’s trial strategy, the decision not to request
a change of venue becomes explicable. One juror, no doubt,
gave the defense hope that the strategy might possibly
work. During voir dire, she said that she and her husband
were “talking about the possible defense and the fact that,
you know, the father having abused or having mentally
abused . . . .” If the media coverage had, in fact, been a
positive factor for Oswald, it would have been a strong one;
nearly all of the 156 prospective jurors who received
questionnaires had even seen a television videotape of the
shoot-out.
Given that Oswald’s strategic choice was to keep this very
high-profile trial in Waukesha County, it would have been
absurd for him to think he could find jurors who could
truthfully say they had not seen the videotape showing him
in the van during the shoot-out. It would also be absurd to
think that it would be possible to find honest potential
jurors who, based on their viewing of the tape, did not think
that Oswald was “guilty” in the sense that laypersons use the
term “guilty. On this issue, the reasoning of the Wisconsin
Court of Appeals bears repeating:
In his brief, Oswald claims that all the prospective
jurors expressed strong opinions that he was guilty.
These opinions, he argues, go “straight to the heart of
the issue: their ability to presume the defendant inno-
cent unless proven guilty of the charges.” However, the
facts and circumstances of this case reveal that this
statement is incorrect. For Oswald, the heart of the issue
during voir dire was securing a jury amendable to his
coercion defense strategy. As a part of this strategy,
Oswald was opting not to contest his participation in
the crimes. His tactical decision was not to protest his
guilt. This strategy involved searching for a particular
type of juror; this ideal juror would not presume that
No. 03-2092 19
Oswald was innocent but would be influenced by the
media coverage and also could be supportive of a
coercion defense. The particular type of juror that
Oswald sought is an important circumstance surround-
ing the voir dire.
In addition, prospective juror after prospective juror was
asked whether he or she could “set aside whatever it is that
you may have already heard or seen that you know is out
there someplace, could you set aside that and deal only with
what is presented during the trial?” The answers varied
from a forthright “yes” to more thoughtful answers indicat-
ing that the juror would try or would hope that he could.
Reading the transcript, one senses the struggle that the
prospective jurors were going through in their attempt, at
least in many cases, to give honest answers. To now look at
the cold record and say that they knew too much should not
be sanctioned by this court.
As to specific alleged juror misconduct, the defense was
given a full opportunity to explore the issue at a post-
conviction hearing. But all that was offered in the way
of evidence was the testimony of one prospective juror
(Jacqueline M.) who was stricken from the panel. The trial
court found that she was not credible for several reasons,
including the fact that she had given several false answers
on her juror questionnaire. The Wisconsin Court of Appeals
discussed the testimony at some length and set out the
reasons for the finding that it was incredible. It then upheld
the trial court’s conclusion that a new trial was not war-
ranted. Even if some judges on this court disagree with that
conclusion, it seems to me that it is hard to say it is “unrea-
sonable” under AEDPA. After all, “compelling institutional
considerations” require that we afford deference to the trial
court as it has the primary responsibility to evaluate
possible influence on the jury. Arizona v. Washington, 434
20 No. 03-2092
U.S. 497 (1978). The state trial court, backed up by the
Wisconsin Court of Appeals, did not make an unreasonable
call.
Supreme Court cases make clear that, even before
AEDPA, a state court’s findings are entitled to deference. In
Patton v. Yount, 467 U.S. 1025 (1984), during voir dire, a
juror and two alternates expressed that they would have
required evidence to overcome their beliefs that the de-
fendant was guilty. The trial judge asked if they could set
aside that opinion and accepted their answers that they
could. The Court applied the presumption of correctness
to the trial court’s resolution of the issue. Interestingly
enough, the Court also noted that potential jurors are people
who are unused to being questioned by lawyers in court
and “cannot be expected invariably to express themselves
carefully or even consistently. Every trial judge understands
this, and under our system it is that judge who is best
situated to determine competency to serve impartially.” At
1039. In this case, federal appellate judges sitting in Chicago
should be slow to condemn the factual determinations of a
state circuit judge in Waukesha County, Wisconsin—a judge
who presided over a lengthy 4-day jury selection process in
a high profile, high publicity case kept in the county as a
result of a strategic defense decision not to try and have it
moved.
Smith v. Phillips, 455 U.S. 209 (1982), involved a juror who
had applied to be an investigator in the prosecutor’s office,
a fact the prosecuting attorneys found out during trial but
withheld from the court. A hearing was held after trial, and
the judge found no evidence suggesting a sinister motive on
the part of the juror, which would have disqualified him.
When the case came to federal court, the district court
granted the petition for a writ of habeas corpus and the
No. 03-2092 21
Court of Appeals affirmed (but on different grounds). The
Supreme Court found the hearing held by the state trial
court to be sufficient:
[D]ue process does not require a new trial every time a
juror has been placed in a potentially compromising
situation. Were that the rule, few trials would be con-
stitutionally acceptable. The safeguards of juror impar-
tiality, such as voir dire and protective instructions from
the trial judge, are not infallible; it is virtually impossi-
ble to shield jurors from every contact or influence that
might theoretically affect their vote. Due process means
a jury capable and willing to decide the case solely on
the evidence before it, and a trial judge ever watchful to
prevent prejudicial occurrences and to determine the
effect of such occurrences when they happen.
At 217.
I do not minimize the duty of the trial judge to be vigilant
to juror misconduct. I do, however, believe that we must
pay more than lip service to the commands of AEDPA. The
Supreme Court has rather clearly said that alleged juror
misconduct and the response of the trial judge cannot be
evaluated in a vacuum. We need to be mindful of the con-
text in which the trial judge is operating and to keep our
eyes on the facts which are at issue. This trial judge was not
asleep at the wheel, and the Wisconsin Court of Appeals
was not so cavalier as the majority would have us believe.
As far as I can see, the panel of jurors that heard this case—
especially given the fact that young Oswald never contested
his physical participation in this insane crime spree—was
not poisoned. My colleagues’ conclusion to the contrary is
regrettable. I respectfully dissent.
22 No. 03-2092
A true Copy:
Teste:
_____________________________
Clerk of the United States Court of
Appeals for the Seventh Circuit
USCA-02-C-0072—6-29-04