In the

United States Court of Appeals
              For the Seventh Circuit

No. 09-2111

D ANIEL W. W ILSON,
                                           Petitioner-Appellant,
                               v.

D ONALD G AETZ,
                                           Respondent-Appellee.


           Appeal from the United States District Court
                 for the Central District of Illinois.
          No. 3:08-CV-03153-JES—Jeanne E. Scott, Judge.



       A RGUED A PRIL 12, 2010—D ECIDED JUNE 17, 2010




 Before C UDAHY, P OSNER, and E VANS, Circuit Judges.
  P OSNER, Circuit Judge. In 2004, Daniel Wilson was
convicted in an Illinois state court of murder while men-
tally ill, and sentenced to 55 years in prison. After ex-
hausting his state remedies in People v. Wilson, No. 4-07-
0359 (Ill. App. Mar. 3, 2008), petition to appeal denied,

(Ill. May 29, 2008), he sought federal
habeas corpus, was denied relief without an evidentiary
hearing (which he had requested), and appeals. He claims
2                                                 No. 09-2111

to have     received ineffective assistance of counsel at his
trial, in   violation of his federal constitutional rights. He
seeks a     new trial for murder, or at least an evidentiary
hearing     in the district court.
  As early as 1987 Wilson had begun having delusions
that “the Catholics” (unnamed, unspecified—his fear
was of some indefinite but immensely sinister Catholic
conspiracy, the sort of thing that Queen Elizabeth I
and other Protestant monarchs feared with greater justifi-
cation during the Reformation) were out to get him.
The delusions intensified in 1998. He believed that the
Catholics were planting cameras in his home to spy on
him and were trying to frame him for molesting his
adopted teenage daughter. He decided that she was part
of the frame up and ordered her from his house. He
stopped speaking to his wife after deciding she’d joined
the conspiracy, and later, after they divorced, he began
suspecting her of having had affairs with Catholics
during the marriage. He severed relations with his
parents because he thought the conspirators were in
touch with them too. He was convinced that the Catholics
were spreading lies and rumors about him and that
Catholic doctors were giving him false diagnoses of his
medical conditions.
   His boss, Jerome Fischer, had a starring role in Wilson’s
fantasy. Wilson believed that Fischer had hired a man
to install the secret cameras in Wilson’s home, spread
lies about Wilson, made fun of his health problems
before other employees, and was scheming to keep him
from home so that it would be easier for the Catholics
to install surveillance cameras.
No. 09-2111                                             3

  Wilson tried to stay at home as much as possible to
protect his home against penetrations by the Catholics
and he therefore refused to attend his employer’s
company-wide meetings, held in New Hampshire,
which would have kept him from his home (which was
in Quincy, Illinois) for a week—in his insane thinking, an
intolerably dangerous absence. He tried to enhance his
home security by buying first toy guns and then real
ones and leaving them where the hidden cameras
would see them, to frighten the Catholics. He bought a
bulletproof vest, installed additional locks, and nailed
the back porch door shut.
  On November 19, 2003, Fischer insisted that Wilson
accompany him the next day to a company meeting
in New Hampshire. Wilson brought up the conspiracy,
the cameras, and the efforts to lure him out of his house,
and said that for these reasons he didn’t want to go to
the meeting. Fischer told him that if he didn’t go he’d
be fired.
  The next morning Fischer called Wilson to tell him
he was on his way to pick him up at his house. Wilson
placed a loaded gun in his pocket, opened his front door
when Fischer rang the doorbell, and followed Fischer to
his car. As soon as Fischer got into the car Wilson shot
him dead. He dialed 911 immediately, and, “extremely
distraught” (as revealed by the police tape of the con-
versation), cried and sobbed, repeatedly saying “I can’t
believe it” and telling the dispatcher what he had done
and that it was a purposeful killing and not an accident.
Highly emotional when arrested, he confessed forthwith,
4                                              No. 09-2111

expressed regret, and declined to explain his reasons
for killing Fischer beyond saying that it was “over some-
thing that has been going on for a while.”
   He was charged with first-degree murder. The court
appointed a psychiatrist named Sadashiv Parwatikar to
evaluate his fitness to stand trial. Parwatikar concluded
that Wilson was unfit—that he would be unable to
assist his lawyer because the only details that he could
provide of the killing were details of the imaginary Catho-
lic conspiracy.
  That was in January 2004. Five months later, after
Wilson had spent most of the interim period in a mental
hospital receiving m edication (Olanzapine, an
antipsychotic drug) for what the hospital diagnosed as
a delusional disorder (Parwatikar had diagnosed Wilson
as schizophrenic), a psychiatrist at the hospital pro-
nounced Wilson fit to stand trial.
  In August, the month before the trial began, Andrew
Schnack, the lawyer whom Wilson’s mother had hired
to represent him, wrote and then phoned Parwatikar
asking him to testify at the trial about whether Wilson
had been insane when he killed Fischer. Parwatikar
replied that a fitness evaluation and a sanity evaluation
are not the same thing and reminded Schnack that he
had done only the former for Wilson. But, pressed by
Schnack, Parwatikar said he could render an opinion on
Wilson’s sanity at the time of the killing but added that
an effective insanity defense would require testimony by
a second expert as well, someone who would perform a
sanity evaluation of Wilson.
No. 09-2111                                             5

  Schnack didn’t retain a second expert and ignored
Parwatikar (beyond giving him treatment reports from
Wilson’s stay at the mental hospital and a tape of the 911
call) until a few hours before Parwatikar testified, when
at his urgent request the lawyer agreed to talk to him.
  Parwatikar testified that Wilson was a paranoid schizo-
phrenic who had killed Fischer under “the pressure of the
delusions.” Concerning the remorse that Wilson had
expressed moments after the killing, Parwatikar testified,
unhelpfully to Wilson, that it was like a mother whose
child runs into a busy street and she hits the child in
anger at the child’s recklessness and only later feels
sorry for having done so. Parwatikar acknowledged on
the stand having interviewed Wilson only to determine
his fitness to stand trial and not his mental state at the
time of the killing. Although Wilson’s mentation was
now much improved as a result of his treatment in
the hospital, Parwatikar had, he acknowledged, not
reinterviewed him.
  Parwatikar was taken apart in cross-examination by a
skillful prosecutor who forced him to concede that only
three paragraphs of his 14-page fitness report concerned
Wilson’s mental state when he had committed
the murder—the rest of the report was about his fitness
when Parwatikar had interviewed him. Parwatikar con-
ceded that his report had expressed no opinion on
whether Wilson had been sane when he killed Fischer. He
conceded that he had never spoken to the police officers
and jail personnel who had seen and talked to Wilson
immediately after the murder. He conceded that lawyer
6                                              No. 09-2111

Schnack had told him that he (Parwatikar) had all the
information he needed to form an opinion on Wilson’s
sanity when he had killed Fischer. He conceded that he
had formed his opinion of Wilson’s sanity before
listening for the first time, on the morning of his testi-
mony, to the tape of Wilson’s statement to police on the
day of the shooting. He even conceded that Wilson
had been legally sane when he called the police dispatcher.
  The prosecutor’s cross-examination of Parwatikar set
the stage for a powerful closing argument in which the
prosecutor dubbed the psychiatrist’s theory of Wilson’s
mental state “light-switch sanity”: “the notion that at one
moment in time, one instant in time a person is insane,
taking the life of a wonderful human being; the next
moment being sane, thereby, evidently, if I hear it right,
entitling this man to be found by you not guilty.”
  The prosecutor emphasized the limited nature of
Parwatikar’s investigation:
      Dr. Parwatikar, who’s expressed a point of view to
    you and who was retained by the court for another
    purpose, didn’t even see the defendant to perform
    that assessment [the sanity assessment]; didn’t ask
    him one question specific to that assessment; did not
    interview him for the purpose of doing a sanity assess-
    ment; didn’t write him; didn’t even pick up the phone.
    And the evidence tells you that “ain’t good enough.”
    That isn’t even close to “startin’ to be good enough.”
      Imagine somebody coming into the hospital uncon-
    scious, blue, and the medical staff looks at him, the
    doctor looks at him. Person is unconscious, turned
No. 09-2111                                              7

   blue. “All right. I’ve seen enough, examined him. He
   is not in any cardiac distress.” Leaves the room. Pretty
   soon the medical staff revives him or awakens him
   [the reference is to the restoration of Wilson’s fitness
   to stand trial by the mental hospital to which he was
   sent after the murder], nurse comes in, asks the
   doctor. “Hey, he is awakened. Do you want to examine
   him. Do you want to talk to him?” “Na, he’s awake.”
   Does that sound like good medical practice? Does
   that sound like a thorough approach?
     A man’s life—a man’s life has been taken. And
   against that backdrop, is that sufficient to persuade
   a reasonable person that a sufficient investigation
   into his sanity has been done? No.
The prosecutor pointed out that his expert witness,
Dr. Henry, who had opined that Wilson was sane when
he committed the murder, had interviewed Wilson
twice. The prosecutor asked the jury rhetorically “why
it’s okay not to talk to someone [i.e., Wilson] when . . .
they are even in better shape than they were before
when you tried to talk to them and you couldn’t get
any information from them about what happened on
the day of the offense. Why you would decide con-
sciously not to go back and talk to that person and try
to learn more heaven knows, but it does tell us some-
thing about the quality of that opinion [Parwatikar’s
opinion on Wilson’s sanity] when you folks [the jurors]
come into judgment.”
  Parwatikar attested in an affidavit submitted in the
state postconviction proceedings that “the prosecutor
8                                                  No. 09-2111

was very effective in his cross-examination by pointing
out the difference between a fitness examination and
a sanity examination. The credibility of my testi-
mony, in my opinion, was diminished by the fact that
I did not re-exam Mr. Wilson for an evaluation of his
sanity at the time of the crime . . . . I believe, if my opinion
had been bolstered by a second expert, there is a great
likelihood that the jury, presented with convincing evi-
dence of insanity at the time he committed the offense,
would have voted not guilty by reason of insanity.”
Parwatikar is not an expert on jurors’ reactions to
evidence, but his concerns echo People v. Nichols, 

, 989 (Ill. App. 1979), where we read that it’s
“crucial to [a] defendant’s insanity defense to have an
examination conducted with respect to his sanity at the
time of the offense,” and that a fitness evaluation is not
an adequate substitute because “insanity as a defense
differs markedly from fitness to stand trial.” See also
People v. Kegley, 

, 1123 (Ill. App. 1988).
The Supreme Court has said in like vein that “when
a defendant demonstrates to the trial judge that his
sanity at the time of the offense is to be a significant
factor at trial, the State must, at a minimum, assure
the defendant access to a competent psychiatrist who
will conduct an appropriate examination and assist in
evaluation, preparation, and presentation of the defense.”
Ake v. Oklahoma, 

, 83 (1985); see also Schultz v.
Page, 

, 1017-19 (7th Cir. 2002); Dando v. Yukins,

, 798-99 (6th Cir. 2006); Powell v. Collins, 

, 395-96 (6th Cir. 2003); Ford v. Gaither, 

(11th Cir. 1992); People v. 

, 529 N.E.2d
No. 09-2111                                                9

at 1120-23. Parwatikar neither conducted an appropriate
examination nor assisted meaningfully in evaluation,
preparation, and presentation of Wilson’s insanity
defense—the reason being the insouciance of Wilson’s
lawyer, Schnack.
  Schnack may have had reasons for his apparent care-
lessness, though there is no evidence that he did. Maybe
he thought that the more Parwatikar delved into the
question of Wilson’s sanity at the time of the murder the
more skeptical he might become that Wilson had been
insane. But that would not explain Schnack’s failure to
heed Parwatikar’s suggestion to hire a second psychiatric
expert, or his failure to meet with Parwatikar to discuss
the latter’s testimony until just hours before he testified.
  Another questionable feature of Schnack’s representa-
tion of Wilson was his failure to interview the members
of Wilson’s family, who had observed Wilson’s mental
deterioration over a period of years. The bare facts of
his bizarre behavior were adequately conveyed to the
jury by Parwatikar’s testimony, but as the Supreme Court
has pointed out, “making a case with testimony and
tangible things . . . tells a colorful story with descriptive
richness . . . . Evidence thus has [persuasive] force
beyond any linear scheme of reasoning, and as its
pieces come together a narrative gains momentum . . . . A
syllogism is not a story, and a naked proposition in
a courtroom may be no match for the robust evidence
that would be used to prove it.” Old Chief v. United
States, 

, 187-89 (1997). Testimony by Wilson’s
family would have made his insanity more palpable to
the jury.
10                                              No. 09-2111

  Especially because of the severity of the penalty that
Wilson was facing, Schnack should have done more:
should at least have interviewed the family members,
spent more time with Parwatikar discussing the
latter’s forthcoming testimony, and either have asked
Parwatikar to reinterview Wilson or retained a second
expert, or done both.
  A more difficult question is whether better representa-
tion would have been likely to change the verdict. The
prosecutor conceded that Wilson had been mentally ill
when he killed Fischer, and under Illinois law this
meant that the jury was being asked by the prosecution
for a verdict of “guilty [of first-degree murder] but men-
tally ill.” 720 ILCS 5/6-2(c). To obtain an acquittal on
grounds of insanity Wilson would have had to prove by
clear and convincing evidence that his mental illness had
prevented him from appreciating the criminality of his act.
720 ILCS 5/6-2(a), 2(e). That might seem a tough row to
hoe, especially in a murder case; one might expect jurors
to worry that an acquittal on grounds of insanity, even
though it would not preclude (and indeed would re-
quire) a civil commitment, would let the defendant walk
as soon as he could find a psychiatrist willing to declare
him cured. Wilson got a verdict of guilty but mentally ill
and maybe he had no realistic hope of a verdict of not
guilty by reason of insanity, no matter how good his
lawyer.
  But a substantial body of empirical research finds that
the enactment of “guilty but mentally ill” laws (laws
allowing the jury to enter a verdict of guilty together with
No. 09-2111                                               11

a finding that the defendant was mentally ill when he
committed the crime), contrary to intuition, does not
reduce the incidence of “not guilty by reason of insanity”
acquittals. Ingo Keilitz et al., The Guilty but Mentally Ill
Verdict: An Empirical Study pp. 14-15 (National Center for
State Courts 1985); Gare A. Smith & James A. Hall, “Evalu-
ating Michigan’s Guilty But Mentally Ill Verdict: An
Empirical Study,” 16 U. Mich. J. L. Reform 77, 92-93, 100-04
(1982); Ames Robey, “Guilty But Mentally Ill,” 6 Bulletin
of Am. Acad. of Psychiatry & L. 374, 380 (1978); but cf. R.D.
MacKay & Jerry Kopelman, “The Operation of the ‘Guilty
but Mentally Ill’ Verdict in Pennsylvania,” 16 J. Psychiatry
& L. 247, 248-50, 259-61 (1988). The advent of the “guilty
but mentally ill” verdict may actually have increased
the number of verdicts of not guilty by reason of
insanity by increasing the incentive of defendants to
plead insanity in states that make an insanity plea a
prerequisite to such a verdict. Christopher Slobogin, “The
Guilty but Mentally Ill Verdict: An Idea Whose Time
Should Not Have Come,” 53 Geo. Wash. L. Rev. 494, 507
(1985).
  The numerous critics of the guilty but mentally ill
verdict emphasize that the actual disposition of the
convicted person is usually the same as under a standard
verdict; the defendant will not be released earlier, and will
receive no more psychiatric treatment in prison, than a
prisoner convicted without any finding of mental illness.
Henry H. Fradella, “From Insanity to Beyond Diminished
Capacity: Mental Illness and Criminal Excuse in the Post-
Clark Era,” 18 U. Fla. J. L. & Pub. Policy 7, 30-31 (2007);
Jennifer S. Bard, “Re-Arranging Deck Chairs on the
12                                               No. 09-2111

Titanic: Why the Incarceration of Individuals with Serious
Mental Illness Violates Public Health, Ethical, and Consti-
tutional Principles and Therefore Cannot be Made Right
by Piecemeal Changes to the Insanity Defense,” 5 Houston
J. Health L. & Policy 1, 37-40 (2005); Maura Caffrey, Com-
ment, “A New Approach to Insanity Acquittee Recidivism:
Redefining the Class of Truly Responsible Recidivists,”
154 U. Pa. L. Rev. 399, 418-20 (2005); Robert D. Miller,
“The Continuum of Coercion: Constitutional and Clinical
Considerations in the Treatment of Mentally Disordered
Persons,” 74 Denver U. L. Rev. 1169, 1186-87 (1997); Ameri-
can Bar Association, Criminal Justice Mental Health
Standards § 7-6.10, p. 394 (1989); Lynn W. Blunt and
Harley V. Stock, “Guilty but Mentally Ill: An Alternative
Verdict,” 3 Behavioral Sciences & the Law 49, 63-64 (1985). In
contrast, acquittal by reason of insanity results in civil
commitment, and should the defendant recover his
sanity he would be entitled to be released. So there can be
a good deal at stake in the jury’s choice between a verdict
of guilty but mentally ill and an acquittal by reason of
insanity, especially in a case such as this, in which the
verdict of guilty but mentally ill resulted in a 55-year
prison sentence.
  A more telling reason to think that Wilson wasn’t
harmed by his lawyer’s lapses might seem to be that the
evidence that Wilson failed to appreciate the criminality
of his criminal act was weak. It has never been suggested
that he thought he was killing Fischer in self-defense
or thought he had any other legal justification for
the killing. The argument rather is that he was “out of
his mind” when he did it—that as Wilson’s opening
No. 09-2111                                             13

brief puts it, “Fischer unknowingly exacerbated Wilson’s
delusions by demanding that he leave the perceived
safety of his home and threatening to fire him if he
refused to do so. Wilson’s delusions, exacerbated by
conflict and fear, prevented him from understanding
reality until the removal of that exacerbating influence.”
The proposition that he was so insane that he could not
“understand reality” (as if for example he had thought
he was an elf shooting a bow and arrow at an orc) is
inconsistent with his behavior immediately before the
killing (when he placed a loaded gun in his pocket while
waiting for Fischer to ring the doorbell) and immediately
after (when he made the 911 call).
  The more plausible inference is that although Wilson
knew he was committing a crime under Illinois law he
could not prevent himself from doing so—he was acting
“under the pressure of the delusions,” as Dr. Parwatikar
put it. It used to be the law that a defendant was entitled
to an acquittal by reason of insanity if he “lack[ed] sub-
stantial capacity . . . to conform his conduct to the re-
quirements of law,” Ill. Rev. Stat. 1984, ch. 38, ¶ 6-2(a),
and so “if it be shown the [defendant’s] act was
the consequence of an insane delusion, and caused by it,
and by nothing else, justice and humanity alike demand
an acquittal.” Hopps v. People, 

(1863); see
also People v. Scott, 

, 246 (Ill. 1992). But
in 1995 Illinois struck from the insanity defense inability
to conform one’s conduct to the law’s requirements.
  Yet as then-Judge Cardozo had explained many years
earlier, an insane compulsion can negate a person’s
14                                                No. 09-2111

meaningful appreciation of the wrongfulness of his act.
“A mother kills her infant child to whom she has been
devotedly attached. She knows the nature and quality of
the act; she knows that the law condemns it; but she is
inspired by an insane delusion that God has appeared to
her and ordained the sacrifice. It seems a mockery to say
that, within the meaning of the statute, she knows that the
act is wrong. If the definition propounded by the trial
judge is right, it would be the duty of a jury to hold her
responsible for the crime. We find nothing either in the
history of the rule, or in its reason and purpose, or in
judicial exposition of its meaning, to justify a conclusion so
abhorrent . . . . We hold therefore that there are times and
circumstances in which the word ‘wrong,’ as used in the
statutory test of responsibility, ought not to be limited to
legal wrong . . . . Knowledge that an act is forbidden by law
will in most cases permit the inference of knowledge
that, according to the accepted standards of mankind, it
is also condemned as an offense against good morals.
Obedience to the law is itself a moral duty. If, however,
there is an insane delusion that God has appeared to
the defendant and ordained the commission of a crime,
we think it cannot be said of the offender that he knows
the act to be wrong.” People v. Schmidt, 

, 949
(N.Y. 1915). If God commands you to kill, you could
hardly be thought to “appreciate the criminality” of your
conduct even though divine command is not a defense
recognized in the criminal code. See also People v. Serravo,

, 139-40 (Colo. 1992); State v. Crenshaw, 

, 494 (Wash. 1983).
  And so what has been called the “deific decree” excep-
tion, e.g., People v. 

, 823 P.2d at 139; State v.
No. 09-2111                                              15

Potter, 

, 488 (Wash. App. 1992), entered
the law. It is not literally applicable to the present case.
Wilson’s delusions had a religious slant, but he didn’t
think he was acting on a direct command from God when
he killed Fischer, or even on an indirect ones. But to
distinguish between “deific” and all other delusions and
confine the insanity defense to the former would present
serious questions under the First Amendment’s estab-
lishment clause, and it is notable therefore that Judge
Cardozo placed his emphasis on a defendant’s inability
to appreciate his act as being morally wrong, whatever
the source of his moral beliefs. Convinced that he was
the victim of a vast conspiracy—that his persecutors,
including his boss, were infiltrating his home in order to
frame him for the crime of child molestation, and that if
he left the state he could well face catastrophe on his
return—Wilson may have thought, at the moment he
killed Fischer, that he was doing a morally justified deed.
   With the exception of People v. Kando, 

(Ill. App. 2009), the only Illinois “deific decree” cases
we’ve found were based on the now-abolished com-
pulsion component of the insanity defense, People v.
Wilhoite, 

, 55-58 (Ill. App. 1991); see also
People v. Baker, 

, 722-30 (Ill. App. 1993);
People v. Garcia, 

, 603-05 (Ill. App. 1987),
although Wilhoite cites Judge Cardozo’s opinion in the
Schmidt case approvingly. People v. Kando, however, was
decided after the change in the statute; and it cited the
Wilhoite and Baker opinions 

. It is a reasonable inference that the Supreme Court
of Illinois would approve an insanity defense along the
16                                              No. 09-2111

lines of Judge Cardozo’s opinion—which has lost none
of its intellectual power by the passage of years—and so
it would be available to Wilson.
  Baker indeed is rather similar to the present case. The
“defendant said that he got into an argument with his
father over a ‘method of religion’ and became upset. He
went to his bedroom where he retrieved a gun, which
belonged to his brother Andre, from a locked toolbox in
his room. He returned to his father and told him ‘the
father dies before the son.’ His father went for his throat
and he shot at his father. His father started to run and
he fired several more shots at his father who then fell in
a utility room. He turned to his mother whom he shot
and then stabbed. Afterwards, he returned to his father
and stabbed him. After he stood there and looked at
them for awhile, he got into his Camaro intending to go
to Las 

. The defendant was
diagnosed as schizophrenic, and the prosecution as in
this case used evidence of the defendant’s “normal” post-
killing behavior (the evidence of his driving to Nebraska,
presumably in an effort to escape apprehension, and
the testimony of the detectives who received his con-
fession two days after the killings that he appeared
normal during the interrogation) to argue that he was not
insane. 

Yet the court directed that the defendant
be acquitted by reason of insanity, and in so ruling
noted that the psychiatric witnesses had each inter-
viewed the defendant a number of times.
 Lawyer Schnack and Dr. Parwatikar were apparently
unaware of the “deific decree” cases and this drove
No. 09-2111                                            17

Parwatikar to embrace, to Wilson’s damage, the “light-
switch” theory, as the prosecutor called it: Wilson was
sane when he put the gun in his pocket, insane when
he killed Fischer, sane again when he dialed 911. A
further telling point that the prosecutor made to the jury
was that since it had taken the psychiatrists at the
mental hospital where Wilson was confined four months
to restore him to sanity, or at least sanity enough to
enable him to be tried, how could it have taken only
minutes for him to recover his sanity after shooting
Fischer if he had been insane when he shot him? Doubtless
the reality is that Wilson was insane throughout; but
insane persons have lucid intervals, as pointed out in our
recent case of Holmes v. Levenhagen, 

, 760-61
(7th Cir. 2010).
  For Parwatikar (or another expert) to have inter-
viewed Wilson after the prosecutor’s psychiatrist had
done so might have bolstered the latter’s testimony that
Wilson had been sane when he killed Fischer. Might—but
by not reinterviewing Wilson, Parwatikar had invited the
prosecutor’s analogy of Parwatikar to the doctor who
doesn’t bother re-examining the now-conscious patient,
and the prosecutor’s further argument that Parwatikar’s
failure to reinterview Wilson should persuade the jury
to reject all of his testimony—his failure to perform a
second examination, according to the prosecutor, showed
that his testimony was not medically sound.
  Schnack should have instructed a psychiatrist, whether
Parwatikar or another one, to interview Wilson after he
had been given antipsychotics and so could speak coher-
18                                              No. 09-2111

ently about the events surrounding the shooting, as he
had been unable to do when interviewed by Parwatikar.
  Remember too that Parwatikar was the witness
who narrated the history of Wilson’s delusions because
Schnack did not put any of Wilson’s family members on
the stand. With Parwatikar’s credibility grievously dam-
aged on cross-examination, the jury may have thought
that his narrative, though not challenged by the prosecutor,
should also be discounted, and therefore that Wilson
may never have been as crazy as the narrative suggested.
  Given the gravity of the charge against Wilson and the
ample evidence that he was driven to kill Fischer by an
insane delusion, we conclude that Schnack’s assistance
to Wilson fell below the minimum professional level
required (by interpretation of the Sixth Amendment) of
a lawyer representing a murder defendant; the Illinois
courts were unreasonable to think otherwise. Wilson’s
imperative need for better and more timely preparation
of Parwatikar (the attempt to prepare him hours before
he testified came too late), for a reinterview of Wilson by
Parwatikar, for acceptance of Parwatikar’s advice to
hire another expert, and for putting the lay witnesses to
Wilson’s mental deterioration on the stand, compels our
conclusion. The only reasons the state courts gave for
thinking Schnack’s representation adequate was that
Parwatikar was a distinguished psychiatrist and that
Schnack “questioned Parwatikar in a cogent manner,
enabling Parwatikar to come across in a favorable light.”
Unmentioned was that Parwatikar had told Schnack that
his testimony alone would not be adequate, that he was
No. 09-2111                                             19

correct, and that his credibility was demolished on cross-
examination.
  Whether Schnack’s failure to provide competent assis-
tance prejudiced Wilson—robbed him of a rea-
sonable chance of acquittal on grounds of insanity—is a
closer question, quite apart from the niceties of insanity
law that we have discussed. Remember that not only is
the burden of proving insanity on the defendant but it is
a heightened burden—proof by clear and convincing
evidence, rather than just by a preponderance of the evi-
dence. We are hesitant therefore to conclude that it was
unreasonable for the Illinois courts to conclude that
Wilson was not deprived of his constitutional right to
effective counsel. But unreasonableness—the normal
standard of review in federal habeas corpus pro-
ceedings under the Antiterrorism and Effective Death
Penalty Act of applications of federal law by state courts,
28 U.S.C. § 2254(d); see Bell v. Cone, 

, 698-99
(2002); Timberlake v. Davis, 

, 824-25 (7th Cir.
2005)—is not the standard applicable to this case,
because the Illinois courts have made no finding
on whether Wilson was prejudiced by the subpar rep-
resentation of him by his lawyer at trial. They found
only that his representation was adequate. And when
there is no state court finding on the issue of prejudice,
our review is plenary. Rompilla v. Beard, 

, 390
(2005) (“because the state courts found the representation
adequate, they never reached the issue of prejudice, and
so we examine this element of the Strickland claim [inef-
fective assistance of counsel] de novo”); Wiggins v. Smith,

, 534 (2003) (“in this case, our review is not
20                                              No. 09-2111

circumscribed by a state court conclusion with respect
to prejudice, as neither of the state courts below reached
this prong of the Strickland analysis”); Jones v. Ryan,

, 640-41 (9th Cir. 2009).
  But in its recent grant of certiorari in Harrington v.
Richter, No. 09-587 (U.S. Feb. 22, 2010), the Supreme
Court asked the parties to address the following ques-
tion: “Does AEDPA deference apply to a state court’s
summary disposition of a claim, including a claim
under Strickland v. Washington, 

(1984)?” The
lower court had held that because the state courts had
denied postconviction relief without any statement of
reasons, the federal courts should apply the standard of
reasonableness without giving any deference to the
state courts’ denial of relief. Richter v. Hickman, 

, 951 (9th Cir. 2009) (en banc). Unless and until the
Supreme Court rejects the no-deference approach, how-
ever, we are bound by it because it is the approach the
Court took in Rompilla and Wiggins. The dissenting
judges in Richter v. Hickman did not question the
majority’s decision to determine reasonableness de novo.

. Nor did the petition for certiorari.
The Supreme Court raised the question of the standard
of review on its own initiative.
  Wilson was entitled to an evidentiary hearing on the
issue of prejudice, and so the judgment of the district
court must be vacated and the case remanded. Should the
state have evidence that the lawyer’s representation
was adequate despite what we have said in this opin-
ion—Wilson’s lawyer might for example have had a
No. 09-2111                                              21

tactical reason for not asking Parwatikar to reinterview
Wilson—the district court should consider that evidence,
treating our determination of the inadequacy of the law-
yer’s representation as tentative.
                                V ACATED AND R EMANDED.




   E VANS, Circuit Judge, dissenting. When a poker player
looks at his hand and sees five different even-numbered
cards, only two of which are in the same suit, he knows
there’s no way he can win. His only option is to fold.
A defense lawyer in a murder case who is dealt an impos-
sibly bad hand can’t fold. He must do the best he can
even if the deck, and the odds, are stacked against him.
I think Daniel Wilson’s defense attorney (Andrew Schnack)
did about as well as could be expected given the
awful hand he was dealt. I would not send this case
back to the district court for more proceedings. For the
reasons that follow, I dissent.
   This indeed was a tragic crime. On the morning of
November 20, 2003, Jerome Fischer left the Quincy, Illinois,
home he shared with his wife and four children. He set
out to pick up Wilson, one of his employees, and drive the
two of them to the airport. As the Appellate Court of
Illinois put it, “The two men were to attend a company
22                                                   No. 09-2111

meeting in New Hampshire. While Fischer waited in his
car for Wilson, Wilson put a gun in his pocket, walked out
of his home, walked over to the car, and pulled the trig-
ger.” Fischer died on the spot. A photo admitted into
evidence at Wilson’s trial (Attorney Schnack objected to
it but its admissibility is not an issue at this time) showed
a gruesome scene of broken glass and blood splattered
on the steering wheel, console, floorboard, and passenger
side window.
  After killing Fischer, Wilson returned to his house and
called 9-1-1. He told the police dispatcher I “just killed
my employer.” The state appellate court noted that when
the dispatcher asked if it was an accident, Wilson replied
“No, I did it purposefully.” Shortly thereafter, he told
Quincy police officers who arrived at his home that he
killed Fischer “over an argument that had been
happening for a while.” The officers at the scene said
Wilson didn’t seem “confused or disorientated,” that he
was “pretty cognizant” and that he answered questions “in
a pretty straightforward manner.” Finally, after saying he
killed his boss, Wilson said he wished he had a chance “to
start over.”
  As the majority observes, the jury was given three
verdict options: guilty, guilty but mentally ill, and not
guilty by reason of insanity. It settled on the middle
option. 1 After a direct appeal (which raised many issues)


1
  It’s true, as the majority notes, that Wilson received a 55-year
sentence. But 30 were for the murder. The other 25 were
                                                     (continued...)
No. 09-2111                                              23

was denied by the Appellate Court of Illinois, Wilson, with
a new lawyer, filed a petition for postconviction relief,
alleging that he received ineffective assistance of counsel.
The trial court dismissed the petition. Wilson then filed a
motion to reopen proofs, supplement the record, and
reconsider the order denying relief. After a hearing, the
state trial court granted the motion to reopen proofs
and supplement the record, but it denied the motion to
reconsider the dismissal order. Wilson appealed and
the state appellate court affirmed the trial court’s order
in 2008. The Supreme Court of Illinois denied leave to
appeal. With his state remedies exhausted, Wilson filed
a petition for habeas corpus in the federal district court.
It was denied. The majority now sends the case back to
the district court where the state can present new
evidence regarding the adequacy of Schnack’s representa-
tion but if it has none, or if what it has comes up short,
the district court must schedule an evidentiary hearing
and determine if the attorney’s shortcomings prejudiced
Wilson. None of this, in my view, is necessary.
  Although the majority recites a litany of sins Attorney
Schnack allegedly committed, the primary shortcoming
seems to be his decision not to insist that Dr. Parwatikar
reexamine Wilson some five months later, after he
emerged from his stay in a mental hospital with a pro-
nouncement that he was “fit” to stand trial. The reexam-



1
  (...continued)
tacked on because, under Illinois law, the crime was com-
mitted with a firearm.
24                                              No. 09-2111

ination, this time a “sanity” examination, should have
obviously been undertaken. Or so says the majority.
Perhaps that’s right. But perhaps it isn’t.
   As I see it, Schnack had two less than ideal choices.
He could let Dr. Parwatikar testify that Wilson was insane
at the time of the murder without conducting a second
exam. That, of course, is something a skilled prosecutor,
like the one here, would explore on cross-examination
and drive home during closing arguments. On the
other hand, Schnack could have insisted on a second
exam by Dr. Parwatikar but that would be risky as well.
What if, after a second exam, Dr. Parwatikar came to the
same conclusion reached by the prosecution’s expert,
Dr. Henry? That would put the kibosh on any slight
hope, given the facts of this case, that the jury would give
Wilson a pass with a finding of not guilty by reason of
insanity. Whatever option Schnack picked was going to
have some downside so I cannot say the one he went with
was utterly unreasonable. In fact, the one he went with
might well have been the best. I say that because, after
reading Dr. Parwatikar’s August 18, 2004, report prepared
in response to a June 23, 2004 letter from Attorney
Schnack (both are attached as exhibits to this dissent),
I can’t say forgoing a reexamination was ill-advised.
  For Wilson to have any chance, he needed Dr. Parwatikar
in his corner. Why take a chance of losing him? Schnack
could present Dr. Parwatikar to the jury as a disinterested
expert. Schnack didn’t pick him out (as the prosecution
did for Dr. Henry), instead he was originally selected
to get involved in the case as a court-appointed expert
No. 09-2111                                             25

witness. Of all the psychiatrists to put on the case, would
a judge select an incompetent? That’s not a bad defense
argument. I recognize, of course, that a “fitness to stand
trial” examination and a “sanity at the time of the crime”
exam are not the same thing. But they are cousins. If
you take a car into a mechanic for a muffler job, he might
also say, “Oh, by the way, I’ve looked underneath your
car and the brakes should be replaced.” Would it be
unreasonable to think that a psychiatrist can’t, in a
manner of speaking, (“I thought he was unfit to stand
trial but I also concluded that he was insane when he
pulled the trigger”) do the same thing?
  Another reason not to take the chance of losing
Dr. Parwatikar concerns timing. He examined Wilson—
actually sat down and talked to him—rather soon after
Fischer was murdered. Because that chat took place
much closer in time to the commission of the crime, it’s
not unreasonable to think that a jury might give an
opinion growing out of that meeting more weight than
one based on an examination that took place so much
later. Like Dr. Henry’s.
  The majority also says that Schnack should have
spent more time prepping Dr. Parwatikar for his testi-
mony. The state trial court, while considering whether
Schnack spent adequate preparation time with
Dr. Parwatikar, observed that the direct examination
”. . . was cogent, it flowed.” Although the trial judge
thought that Schnack only spent a lunch hour preparing
Dr. Parwatikar for his testimony, he stated that Schnack
“sure got a lot out of that one-hour lunch period. Of
26                                              No. 09-2111

course, part of it is that I’m sure Dr. Parwatikar has a lot
of experience in that sort of testimony himself. It would
seem to me that they knew each other; this was not
some off-the-cuff sort of presentation.” Since the trial
court, who witnessed Dr. Parwatikar’s testimony, thought
the testimony was more than adequate, I cannot under-
stand why the majority thinks more prep time was needed.
  Another reason why Wilson needed Dr. Parwatikar
was that all the “Catholic Conspiracy” evidence, which the
majority recounts, came to the jury via Schnack’s direct
examination of the doctor who related what Wilson said
during the fitness examination. With that in the record,
much of it pretty looney, I see no reason why not
calling “family members” can be view as an incompetent
decision.
  The six affidavits of family members filed in the state
post-conviction proceedings (from an ex-wife, a step-
daughter, a step-father, an aunt, and Wilson’s mother
and the step-daughter of Wilson’s mother) do little to
shed light on Wilson’s mental state at the moment he
gunned down his boss. Plus, some of the “facts” asserted
in the affidavits would have, if put before the jury, cast
Wilson in a less than sympathetic light. For instances, the
step-father said, “Dan thought his step-daughter had been
coached by his ex-wife to seduce him.” The step-daughter,
in her affidavit, said she ran away from home and
had not even seen Dan Wilson during the two years
before the murder took place. And an interesting aside:
Wilson’s ex-wife was, according to her affidavit, “subpoe-
naed by the State’s Attorney to testify as a State witness,
but I was never called.”
No. 09-2111                                                   27

  The majority also suggests that Schnack should have
argued that Wilson was insane because he was acting
“under the pressure of delusions.” 2 This suggestion is
rather odd because as the majority notes, Illinois elimi-
nated the inability to conform one’s conduct to the law
prong from its insanity defense statute in 1995. Undeterred,
however, the majority extensively quotes then Judge
Cardozo from a New York case he penned almost two
years before the United States entered World War I and
17 years before President Hoover appointed him to a
seat on the Supreme Court. The majority says, “It is a
reasonable inference that the Supreme Court of Illinois
would approve an insanity defense along the lines of
Cardozo’s opinion and it would be available to Wilson.”
I don’t think so.
  As I see it, Attorney Schnack’s representation of Wilson
was not constitutionally ineffective. More importantly,
under ADEPA, the Appellate Court of Illinois did not
unreasonably apply Strickland v. Washington, 

(1984). I would affirm the district court’s judgment dis-
missing Wilson’s petition for habeas relief.




2
  On direct appeal, the Illinois Appellate Court stated that the
jury may not have found Dr. Parwatikar’s testimony to be
compelling because “[f]or example, Parwatikar’s continued
references to the ‘pressures’ of Wilson’s delusions often
spoke more to Wilson’s ability to conform his behavior to
the requirements of the law, than to Wilson’s ability to appreci-
ate the criminality of his conduct.”
28   No. 09-2111
No. 09-2111   29
30   No. 09-2111
No. 09-2111   31
32   No. 09-2111
No. 09-2111             33




              6-17-10