Wayne D. Kubsch v. Ron Neal

In the

United States Court of Appeals
For the Seventh Circuit
____________________
No. 14-1898
WAYNE KUBSCH,
Petitioner-Appellant,

v.

RON NEAL, Superintendent,
Indiana State Prison,1

Respondent-Appellee.
____________________

Appeal from the United States District Court for the
Northern District of Indiana, South Bend Division.
No. 3:11CV42-PPS — Philip P. Simon, Chief Judge.
____________________

ARGUED FEBRUARY 10, 2015 — DECIDED AUGUST 12, 2015
____________________

Before WOOD, Chief Judge, and TINDER and HAMILTON,
Circuit Judges.

1 We have substituted as respondent-appellee Ron Neal, the current
Superintendent of the Indiana State Prison, for Bill Wilson, the former
Superintendent. See Fed. R. App. Pro. 43(c)(2).
2 No. 14-1898

HAMILTON, Circuit Judge. Wayne Kubsch appeals the de-
nial of his habeas corpus petition. After being convicted of
murdering his wife, her son, and her ex-husband, Kubsch
was sentenced to death. Kubsch’s three principal arguments
on appeal are that his conviction and sentence are unconsti-
tutional because (a) the Indiana trial court excluded evidence
of a witness’s exculpatory but hearsay statement to police,
(b) he was denied effective assistance of counsel in seeking
admission of the witness’s hearsay statement, and (c) his
waiver of counsel and choice to represent himself at the sen-
tencing phase of his trial were not knowing and voluntary.
We reject all three claims. Kubsch argues for a constitu-
tional right to defend himself with otherwise inadmissible
hearsay, at least if the hearsay seems sufficiently reliable and
is sufficiently important to his defense. See Chambers v. Mis-
sissippi, 410 U.S. 284, 300–02 (1973). Kubsch’s evidence is not
sufficiently reliable to fit that narrow constitutional excep-
tion and to have required Indiana courts to disregard long-
established rules against using ex parte witness interviews as
substantive evidence at trial. His able trial counsel tried hard
to have the statement admitted; they were not successful but
also were not constitutionally ineffective.
As for the waiver of counsel claim, the Indiana Supreme
Court rejected the claim in a careful discussion tailored to
the facts of this case. Its rejection of the claim was not contra-
ry to or an unreasonable application of clearly established
federal law as determined by the Supreme Court of the
United States. See 28 U.S.C. § 2254(d)(1); Harrington v. Rich-
ter, 562 U.S. 86, 102–03 (2011).
In addition to the exculpatory hearsay claim, the related
ineffective assistance claim, and the waiver of counsel claim
No. 14-1898 3

that we address in detail, Kubsch raises a number of other
arguments on appeal, all of which are challenges to the effec-
tiveness of his counsel. We have considered all of these addi-
tional arguments, and we reject them for the reasons Chief
Judge Simon explained in his thorough opinion. See Kubsch
v. Superintendent, No. 3:11CV42–PPS, 2013 WL 6229136 (N.D.
Ind. Dec. 2, 2013). Accordingly, we affirm the denial of relief
as to both Kubsch’s convictions and the death sentence.
I. Factual and Procedural Background
A. Court Proceedings
The State of Indiana charged Kubsch with murdering
Beth Kubsch, Aaron Milewski, and Rick Milewski: his wife,
her son, and her ex-husband. The three were murdered in
Kubsch’s home on September 18, 1998. Kubsch was first tried
and found guilty in May 2000. The jury recommended and
the judge imposed the death penalty. On direct appeal the
Indiana Supreme Court held that the first trial violated
Kubsch’s constitutional rights when the prosecution used his
post-Miranda silence as evidence against him. Based on that
and other errors, the court vacated the convictions and or-
dered a new trial. See Kubsch v. State, 784 N.E.2d 905 (Ind.
2003).
Kubsch’s second trial in March 2005 is our focus. Once
more a jury convicted Kubsch of the three murders. There
were two big differences in the second trial, in addition to
avoiding the errors that had required the new trial. First,
Kubsch offered as evidence the videotaped interview of
Amanda Buck, a nine-year-old neighbor of Aaron and Rick
Milewski. Amanda told a police detective four days after the
murders that she had seen both Aaron and Rick alive and
4 No. 14-1898

well at their home on the day of the murders at a time for
which Kubsch has a solid alibi. The judge excluded her rec-
orded statement as hearsay and as having no impeachment
value. Second, unlike the first trial, Kubsch decided to waive
counsel and represent himself in the sentencing phase of the
trial. He also declined to present any mitigating evidence.
He told the jury he agreed with the State that no mitigating
factors outweighed the aggravating factors supporting a
death sentence, but he insisted on his innocence. He ended
his brief statement to the jury by saying he did not care what
penalty was imposed.
Again the jury’s verdict was for death and the judge im-
posed the death penalty. The state courts affirmed the con-
victions and sentence on direct appeal, Kubsch v. State, 866
N.E.2d 726 (Ind. 2007), and on post-conviction review,
Kubsch v. State, 934 N.E.2d 1138 (Ind. 2010).
Kubsch then petitioned for a writ of habeas corpus in
federal court, raising many more issues than we address in
this opinion. The district court denied relief on all claims,
Kubsch v. Superintendent, No. 3:11CV42–PPS, 2013 WL
6229136 (N.D. Ind. Dec. 2, 2013), and then denied Kubsch’s
Rule 59 motion, Kubsch v. Superintendent, No. 3:11CV42–PPS,
2014 WL 1260021 (N.D. Ind. March 24, 2014). Kubsch ap-
peals. We review the district court’s decision de novo. E.g.,
Harris v. Thompson, 698 F.3d 609, 622 (7th Cir. 2012).
B. The Case Against Kubsch
Chief Judge Simon aptly described the case against
Kubsch as a “slow-moving accumulation of a glacier of cir-
cumstantial evidence.” 2013 WL 6229136, at *3. A critical fac-
tor was that Kubsch’s account of his own actions changed
No. 14-1898 5

dramatically between the night of the murders and his trial
testimony, after he knew the constraints imposed by physical
and other evidence such as telephone records.
Kubsch lived with his wife Beth in Mishawaka, Indiana.
They shared the home with Beth’s twelve-year-old son, An-
thony Earley. September 18, 1998 was Beth’s birthday. She
had planned to meet Kubsch for lunch. Beth was supposed
to pick up Anthony late in the afternoon after a school
dance. When she did not appear, Anthony got a ride home
with a friend. At about 5:30, he found Beth’s car in the
driveway, along with a truck that her ex-husband Rick
Milewski was using. The house was locked. Only Wayne,
Beth, and Anthony had keys. No one seemed to be home.
There was no sign of forced entry.
As Anthony looked around the main floor of the house,
though, he saw bloodstains and signs of a struggle. He
opened the door to the basement. He saw Rick lying at the
foot of the stairs. The handle of a large kitchen knife was
sticking out of his chest. Anthony went down the stairs, real-
ized Rick was dead, and also found the body of his eleven-
year-old step-brother Aaron lying next to Rick.
Anthony ran for help. Mishawaka police officers arrived
about 5:45 p.m. Both Aaron and Rick had multiple stab
wounds. The police officers found no sign of gunshot
wounds. They also found no sign of Beth. After finding no
one else in the house, the police secured the scene until they
could obtain a search warrant.
That day Wayne Kubsch had finished work at an area
factory shortly before 2:00 p.m. Late in the afternoon, he was
returning to Mishawaka from picking up his son in Three
6 No. 14-1898

Rivers, Michigan. He dropped off his son at Kubsch’s
grandmother’s home. Kubsch arrived home about 6:45 and
found the house surrounded by police. Kubsch was told that
Aaron and Rick were dead and that no one knew where Beth
was.
Kubsch soon went with police officers to the South Bend
police department for questioning by detectives. That initial
interview was audio-and video-recorded. Kubsch appeared
preoccupied and careful, not distraught or frantic. He made
no reference to the search for his missing wife, though there
were obviously powerful reasons to be worried about her
safety. He showed little emotion.
In that first interview on the night of the murders,
Kubsch gave the police his first account of his movements
and activities that day. Kubsch said that he and Beth had
planned to meet for lunch to celebrate her birthday, but that
he had called her to cancel because he had been late for work
that morning. He also said that he had gotten permission to
leave work early for lunch so he could buy Beth a birthday
present (something he did not actually do until much later in
the day). He told the police that he had gone home at lunch
but could not get inside because he had forgotten his house
key. He also did not mention that he had gone home a sec-
ond time—shortly after work—before going to pick up his
son in Michigan.
Kubsch ended the interview. His friend Dave Nichols
and Nichols’ wife testified that Kubsch called them about
8:00 or 8:30 that evening and said two things known to the
killer but not yet known to the police. He told Nichols that
Beth was “gone,” which Nichols understood to mean that
No. 14-1898 7

she was dead, not missing.2 At that time, Beth’s body had not
yet been found. And while “gone” might be explained away
as ambiguous, Kubsch also told Nichols that Rick and Aaron
had been stabbed and shot. Not until autopsies were done the
next day did the police learn that Rick and Aaron, in addi-
tion to their multiple stab wounds, had each been shot in the
mouth.
At about 9:00 p.m., police officers on the scene discov-
ered Beth’s body. She was just a few feet from Rick and Aa-
ron, but she was hidden underneath the staircase behind
blankets that young Anthony had hung up as a sort of “fort”
or hiding place a few weeks earlier. She had been stabbed
eleven times. Her head was almost entirely covered in gray
duct tape. Her body was “hog-tied” with the same tape, her
wrists and ankles all bound together behind her back. (An
autopsy also showed a blow to the back of her head and de-
fensive wounds on her hands and wrists.) The officers quick-
ly told the detectives at the South Bend station that Beth had
been found murdered. The detectives then brought Kubsch
back for more questioning later that evening. He declined to
talk with them at that point, but he gave them permission to
search his car.
The investigation of physical evidence turned up no evi-
dence pointing conclusively to Kubsch. The only blood
found on the scene belonged to the victims. The police did
not find evidence of the victims’ blood on Kubsch or his

2 Nichols’ wife, Gina DiDonato, confirmed his account of the tele-
phone call and in response to a juror’s question made clear that Kubsch
told them that Beth was dead.
8 No. 14-1898

clothing. They also found no DNA or fingerprint evidence
that pointed to him or anyone else as the killer.
Various items of physical evidence were consistent with
Kubsch’s guilt. In isolation none is conclusive. Taken togeth-
er they point toward Kubsch as the killer, though not defini-
tively. In Kubsch’s car the police found the wrapper of a roll
of duct tape of the type used to bind Beth. A bloody roll of
duct tape at the top of the stairs matched the wrapper and
the tape on Beth’s body. A cloth fiber from the tape roll
matched a fiber from the carpet of Kubsch’s car. A receipt for
purchase of the duct tape, three days before the murders,
was found in Kubsch’s car.
The police also found in Kubsch’s car a wadded-up re-
ceipt from a deposit Beth had made the morning of the mur-
ders at the drive-through window of her credit union. The
presence of that receipt in Kubsch’s car contradicted the ac-
count he had given police the evening of the murders. (Even
Kubsch’s explanation at trial, that he found it next to the
home telephone on his first stop at home that day, was im-
probable if not physically impossible. That explanation
would have required Beth to do some improbable backtrack-
ing between two related errands.)
Of course, the locked house was also evidence that point-
ed toward Kubsch. The knife in Rick’s chest was from the set
of kitchen knives upstairs. A kitchen pan also had Beth’s
blood on it. As the prosecutor pointed out in closing argu-
ment, if the killer had been a stranger, it seems improbable
that he would have counted on tools found in the home—the
knife, the pan, and the duct tape—to carry out the murders.
No. 14-1898 9

Telephone records played an important role in the inves-
tigation and at trial. Recall that Kubsch had told police that
he returned home at lunch but could not get in without his
key. Home telephone records showed that was false. A call
had been placed from the home telephone while Beth was
running her errands that morning. Kubsch testified at trial
that he had in fact gotten into the house—through the gar-
age—where he said he made the call, smoked part of a mari-
juana cigarette, and then left to return to work around noon.3
Kubsch also made numerous calls with his cell phone on
the day of the murders. Records of those calls showed his
approximate locations at different times during the day. He
left work for the day just before 2:00. Though he told the po-
lice the night of the murders that he had then gone directly
to Michigan to pick up his son, he later admitted he had first
actually returned to his home. He claimed that he had
stopped at home for a few minutes between 2:30 and 2:45
and that no one else was home. At 2:51 Kubsch placed a cell
phone call from a cell sector near his home. Cell phone rec-
ords and other evidence showed that Kubsch then drove to
Michigan to pick up his son. The State’s theory has been that
Kubsch had an opportunity to commit the murders in the
time between approximately 2:00 and 3:00.
Another important discrepancy in Kubsch’s story was
that at 12:09 p.m. he called Rick Milewski and, according to
Rick’s brother, asked Rick to meet him at his house at 3:00

3 By the time Kubsch testified at trial, of course, he knew about the
telephone records and other evidence that contradicted in several key
respects the story he had first told the police in his interview the night of
the murders.
10 No. 14-1898

p.m. to help move a refrigerator. That request is hard to un-
derstand if Kubsch was planning to be on his way to Michi-
gan by then. (The prosecution’s theory was that Kubsch
planned to have Rick find Beth’s body but that Rick and Aa-
ron showed up too early, before Kubsch had left, so he killed
them too.)
Yet another discrepancy in Kubsch’s story came from
Beth’s mother, Diane Rasor. She testified that when she
talked with Kubsch on the afternoon of the murders, she
mentioned that she had not been able to get in touch with
Beth all day (Beth’s birthday, recall). Kubsch reassured her,
telling her that he had talked with Beth by phone and knew
Beth was running a number of errands and was not at home
to answer the phone. Several days after the murders, Kubsch
told Rasor that he had not talked to Beth the day she was
killed and he wished he had.
Kubsch also had a significant financial motive to murder
Beth. The prosecution showed that the couple was in deep
financial distress in 1998. Their cash flow was consistently
negative. Early that year Kubsch had refinanced eight of the
rental properties he owned, converting all available equity
into cash and substantially increasing the total debt to about
$424,000. Several credit cards or lines of credit were near
their maximum limits. About three months before the mur-
ders, Kubsch had bought a new insurance policy on Beth’s
life for $575,000, with himself as the sole beneficiary. Kubsch
claimed at trial that he had not realized they were in such
difficult financial straits, but he also testified that he took
care of the couple’s bills, as well as their credit cards and
lines of credit, and of course he had undertaken all the refi-
nancing earlier that year.
No. 14-1898 11

As Chief Judge Simon summarized:
The case against Kubsch was entirely circum-
stantial. There was no eyewitness, no DNA ev-
idence, no fingerprint testimony, indeed no fo-
rensic evidence at all that linked Kubsch to the
murders. There was, however, moderately
strong evidence of motive and opportunity. But
most damning to Kubsch was a series of lies,
inexplicable omissions, and inconsistencies in
what Kubsch told the police and later testified
on the witness stand, and these statements—in
conjunction with a few pieces of circumstantial
evidence—are what almost assuredly got
Kubsch convicted.
2013 WL 6229136, at *1.
II. Exclusion of Exculpatory Hearsay Evidence
Kubsch argues that he was convicted of the murders
through a violation of his federal due process right to pre-
sent a defense. The trial court did not allow him to introduce
as substantive evidence a witness’s videotaped interview
with a police detective four days after the murders. Nine-
year-old Amanda Buck and her mother Monica were inter-
viewed together by the detective. The Bucks lived across the
street from two of the victims, Rick and Aaron Milewski. In
the recorded twenty-minute interview, Amanda told the de-
tective that she had seen Rick and Aaron alive and well at
their home when she got home from school and daycare, be-
tween 3:30 and 3:45 p.m. on the day of the murders, Friday,
September 18, 1998.
12 No. 14-1898

The date and time are critical. Based on telephone rec-
ords and other evidence, the State argued at trial that
Kubsch murdered the three victims between approximately
2:00 and 3:00 p.m. Kubsch’s own testimony placed him at his
home between approximately 2:30 and 2:45, though he
claimed no one else was there. Cell phone records show that
by 3:30 p.m. that day, Kubsch was well on his way to the
town of Three Rivers, Michigan to pick up his son for the
weekend. He did not return to his home in Mishawaka, In-
diana until about 6:45, after the bodies of Rick and Aaron
had been discovered there.
The importance of the constitutional evidentiary issue
cannot be overstated. If the account given by Amanda in her
recorded interview is correct, then Kubsch could not have
committed the three murders for which he has been sen-
tenced to death. And apart from Kubsch’s own claims of in-
nocence—impeached as they are by his shifting accounts of
his movements that day—Amanda’s recorded interview is
the only support for Kubsch’s alibi defense.
Kubsch bases his due process claim on Chambers v. Mis-
sissippi, 410 U.S. 284 (1973), and its progeny. In Chambers the
Supreme Court reversed a murder conviction on direct ap-
peal. The Court held that the defendant was denied a fair
trial when the trial court prevented him from impeaching a
witness he had called and excluded hearsay evidence that
the same witness had confessed to three different acquaint-
ances that he was the killer. Kubsch relies on the hearsay
portion of the Chambers analysis and its often-quoted state-
ment that “the hearsay rule may not be applied mechanisti-
cally to defeat the ends of justice.” 410 U.S. at 302. The actual
holding of Chambers is considerably narrower, however, for
No. 14-1898 13

it depended on the combination of the trial court’s limits on
cross-examination and its exclusion of the multiple hearsay
confessions, and the particular facts and circumstances of the
case, which we describe in more detail below. See id. at 302–
03.
We address this issue in four steps. Part A explains the
details of Amanda’s statement and its treatment by the trial
court and the Indiana Supreme Court. Part B explains the
Chambers line of cases and the general constitutional stand-
ard for the right to present a defense, as well as its applica-
tion in cases involving hearsay. Part C considers the factors
indicating that Amanda’s recorded statement is or is not reli-
able for purposes of Chambers. Part D addresses the issue of
our standard of review, which turns out to be rather in-
volved, and explains our conclusion that Kubsch is not enti-
tled to relief.
A. The Statement in the State Courts
Four days after the murders, Sergeant Mark Reihl inter-
viewed nine-year-old Amanda Buck and her mother Monica
Buck together. The interview was in a police station and was
audio-and video-recorded. The Bucks lived across the street
from Rick and Aaron Milewski, and Sergeant Reihl asked
them what they remembered from the day of the murders.
Amanda answered most of the questions, but Monica added
her own recollections, including specific times. Amanda re-
called seeing both Aaron and Rick at their home across the
street after she got home from school and daycare, which
would have been between 3:30 and 3:45 on the afternoon of
the murders.
14 No. 14-1898

Amanda’s account was specific about many details, in-
cluding what she was doing and which truck Rick was driv-
ing. She specifically recalled seeing Rick go into his kitchen
and return with a glass. Her account was specific about the
time and date. She recalled that she and Aaron were plan-
ning to go on a school field-trip the next day, a Saturday, and
that Aaron had not shown up for the trip. Her mother Mon-
ica recalled having seen Aaron (but not Rick) when she got
home shortly after 4:00 p.m. after going to the bank to de-
posit her paycheck, which she usually did on Friday.
The interview was disclosed to the defense, but Kubsch
did not call Amanda or Monica as witnesses at his first trial,
which took place less than two years after they spoke to the
police. At the second trial in 2005, though, Kubsch called
then sixteen-year-old Amanda as a witness. She testified that
she did not remember whether she saw Rick and Aaron on
the afternoon of the murders. She also testified that she did
not even remember being interviewed by the police seven
years earlier. After her brief testimony, and outside the pres-
ence of the jury, Amanda reviewed the recording of her in-
terview. That apparently did not refresh her recollection be-
cause Kubsch offered no further testimony from her. Kubsch
never called Monica to testify.
The real purpose of calling the sixteen-year-old Amanda
was to put into evidence the video recording of the nine-
year-old Amanda. Kubsch first tried to introduce the record-
ing as substantive evidence. The recording was hearsay, of
course. It was an out-of-court statement offered to prove the
truth of its content. At trial, Kubsch argued that it should be
admitted as a recorded recollection. Indiana Rule of Evi-
dence 803(5), like its federal counterpart, recognizes an ex-
No. 14-1898 15

ception to the rule against hearsay for a “recorded recollec-
tion.” Recorded recollections are records of what a witness
once knew when her memory was fresh but now no longer
recalls. A recorded recollection also “accurately reflects the
witness’s knowledge.” Ind. R. Evid. 803(5)(C); see also Fed.
R. Evid. 803(5)(C). Examples might include a diary or journal
entry or a memorandum to file, as well as recorded inter-
views.
This recorded statement does not meet the last require-
ment of Rule 803(5). Amanda would have needed to “vouch
for the accuracy” of the statement for it to qualify as a rec-
orded recollection. Kubsch v. State (Kubsch II), 866 N.E.2d 726,
734 (Ind. 2007), quoting Gee v. State, 389 N.E.2d 303, 309 (Ind.
1979). As the trial court found and the Indiana Supreme
Court affirmed, “Buck could not vouch for the accuracy of a
recording that she could not even remember making.”
Kubsch II, 866 N.E.2d at 735. The videotaped statement did
not qualify as a recorded recollection under Indiana evi-
dence law. Id.4

4 The recording would also not be admissible under Federal Rule of
Evidence 803(5), which is substantially identical to its Indiana counter-
part and has the same requirement that the declarant endorse the accura-
cy of the prior recording. See, e.g., United States v. Green, 258 F.3d 683,
689 (7th Cir. 2001); United States v. Schoenborn, 4 F.3d 1424, 1427–28 (7th
Cir. 1993). In fact, neither Kubsch nor our dissenting colleague has iden-
tified any federal or state decision indicating that the recording of
Amanda’s interview would have been admissible under the law of any
American jurisdiction. See also, e.g., State v. Perry, 768 N.E.2d 1259, 1264–
65 (Ohio App. 2002) (under identical recorded recollection rule, affirm-
ing exclusion of video recording of interview with eight-year-old child
who, when testifying at trial two years later, did not remember the inter-
view and did not testify that the recording correctly reflected her
knowledge of events at the time it was made).
16 No. 14-1898

Kubsch next offered the videotaped statement to im-
peach Amanda’s trial testimony with extrinsic evidence of a
prior inconsistent statement. See Ind. R. Evid. 613(b). As not-
ed, Amanda testified that she simply did not remember talk-
ing to the police and did not remember whether she saw her
friend and neighbor Aaron between 3:30 and 3:45 p.m. the
day of the murders.
The trial court sustained the State’s objection to admitting
the statement as impeachment evidence because Amanda
“testified to no positive fact that is subject to impeachment.”
Tr. 3120. The Indiana Supreme Court agreed with respect to
Amanda’s trial testimony that she did not remember what
happened or whom she saw on the day of the murders.
Kubsch II, 866 N.E.2d at 735. However, Amanda also testified
at one point that she “probably didn’t see” Aaron at home
between 3:30 and 3:45 p.m. on the day of the murders. Tr.
2985. The Indiana Supreme Court held that this testimony
was properly subject to impeachment and that the trial court
had erred by not allowing the attempted impeachment.
Kubsch II, 866 N.E.2d at 735.
The Indiana Supreme Court also held, however, that the
error was harmless. Id. In the debate in the trial court about
the recording, the State said that if Kubsch were allowed to
use Amanda’s recorded statement to impeach her trial testi-
mony, the State would respond with additional evidence
impeaching the impeachment. The prosecutor asserted that
three days after the recorded interview, Lonnie Buck (Mon-
ica’s father and Amanda’s grandfather) had called Sergeant
Reihl and reported that both Amanda and Monica had been
mistaken about the day they recalled and that they had de-
scribed for him not the day of the murders but the day be-
No. 14-1898 17

fore. Monica had followed up with a later statement saying
that she and Amanda had not seen Aaron on the day of the
murders. At the time of the 2005 trial, the State was prepared
to call both Monica Buck and Sergeant Reihl to impeach the
proposed impeachment of Amanda.
The Indiana Supreme Court explained its finding of
harmless error:
Amanda’s testimony should have been im-
peached, but other testimony would have sup-
ported hers had she been impeached, and
therefore, her testimony likely did not contrib-
ute to the conviction. See Pavey v. State, 764
N.E.2d 692, 703 (Ind. Ct. App. 2002) (“An error
in the admission of evidence is not prejudicial
if the evidence is merely cumulative of other
evidence in the record.”).
866 N.E.2d at 735. Just before this passage, the court
dropped a footnote rejecting Kubsch’s federal constitutional
claim under Chambers:
The availability of this testimony is also the
reason why Kubsch’s claim that he was denied
his federal constitutional right to present a de-
fense fails. See Chambers v. Mississippi, 410 U.S.
284, 302 (1973) (protecting defendant’s due
process right by recognizing an exception to
application of evidence rules where evidence
found to be trustworthy).
Id. at 735 n.7.
Unless we keep in mind the difference between substan-
tive evidence and impeachment evidence, which may be
18 No. 14-1898

considered not for the truth of the matter asserted but only
to evaluate the credibility of other evidence, these terse pas-
sages finding harmless error may seem mistaken. After all, if
Amanda’s statement were admissible as substantive evi-
dence to prove that what she said in the interview was true,
then the mere fact that there was some contradictory evi-
dence would not justify its exclusion. (The State’s proffered
impeachment did not include any admission by Amanda
herself that she had been mistaken.) Conflicting evidence
would simply present an ordinary question for a jury to re-
solve, as the trial judge recognized, see Tr. 3015, though a
question of great importance because the statement would, if
believed, exonerate Kubsch.
When we focus, however, as the trial judge did on the
limited role of impeachment evidence, the harmless error
finding is clearly sound as a matter of state evidence law.
The only thing Amanda said in her trial testimony that was
subject to impeachment was that she “probably didn’t see”
Aaron on the afternoon of the murders. As the trial judge
pointed out, “She gave no substantive evidence in this case
whatsoever.” Tr. 3032. Amanda’s narrow substantive state-
ment that she “probably didn’t see” Aaron on the afternoon
of the murders was not inculpatory. It had essentially no
probative value for the jury, so there would have been no
point in impeaching her, and the exclusion of her statement
for impeachment purposes could not have contributed to
Kubsch’s convictions.
The Indiana Supreme Court’s rejection of the distinct
Chambers claim in footnote 7 is the focus of our scrutiny. In
the trial court, Kubsch had not asserted a distinct federal,
constitutional claim under Chambers. He made that federal
No. 14-1898 19

argument in his direct appeal, though, and the Indiana Su-
preme Court elected to decide the issue on its merits rather
than find a procedural default. Footnote 7 was quite sensible
to the extent that the recording was being offered only to
impeach the non-inculpatory “probably didn’t see him” por-
tion of Amanda’s trial testimony. The problem is that that
reasoning seems not to have actually engaged with Kubsch’s
argument under the federal Constitution that the recording
should have been admitted as substantive evidence. Again, the
mere fact that the State would have offered contradictory ev-
idence would have presented a jury question, not a basis for
excluding the evidence in the first place. We explore these
issues further in Part D on the standard of our review of the
state court’s decision.
B. The Right to Present a Defense
The exclusion of Amanda’s recorded statement was not
contrary to Indiana evidence law, as the Indiana Supreme
Court decided. That conclusion does not resolve the federal
constitutional question, though it informs our answer to that
question. In a series of decisions led by Chambers v. Mississip-
pi, 410 U.S. 284 (1973), the Supreme Court has held that the
accused in a criminal case has a federal constitutional right
to offer a defense. Both the accused and the state “must
comply with established rules of procedure and evidence
designed to assure both fairness and reliability in the ascer-
tainment of guilt and innocence.” Id. at 302. In some circum-
stances, however, the constitutional right to defend takes
precedence over rules of evidence. This can include the hear-
say rules, as Chambers itself showed.
Chambers is the closest Supreme Court case on its facts, so
to understand the scope of this right to defend with hearsay,
20 No. 14-1898

we consider that case in some detail. Leon Chambers was
accused of murdering a police officer in a chaotic disturb-
ance, essentially a small riot, as police were trying to arrest
another person. Another man named McDonald had con-
fessed to the murder: “McDonald had admitted responsibil-
ity for the murder on four separate occasions, once when he
gave the sworn statement to Chambers’ counsel and three
other times prior to that occasion in private conversations
with friends.” Id. at 289. McDonald was arrested after con-
fessing to Chambers’ counsel, but he was released when he
repudiated that confession at his own preliminary hearing.
Id. at 287–88.
Chambers called McDonald as a witness at trial. McDon-
ald’s written confession was admitted into evidence, but
McDonald again repudiated it. 410 U.S. at 291. Chambers
was not allowed to test McDonald’s memory or otherwise to
challenge his testimony. The state courts relied on the old
“voucher” rule under which a party who called a witness
was deemed to have vouched for his credibility and so was
not allowed to impeach him even if he was actually adverse.
The Supreme Court found, however, that the voucher rule
was no longer realistic and had been applied to limit unfair-
ly Chambers’ examination of a critical witness who was in
fact adverse. Id. at 295–98.
After his attempts to impeach McDonald were stymied,
Chambers then offered the testimony of three friends to
whom McDonald had confessed. Their testimony about
No. 14-1898 21

McDonald’s confessions was excluded as hearsay. Id. at 292–
93. The jury convicted Chambers of the murder.5
On direct appeal, the Supreme Court reversed based on
the combination of the voucher rule’s barring impeachment
of McDonald and the exclusion of the hearsay confessions.
Id. at 302–03. The Court noted that declarations against in-
terest have long been treated as sufficiently reliable to be ex-
cepted from rules against hearsay. Id. at 298–99. The Court
found that the excluded confessions “bore persuasive assur-
ances of trustworthiness” that brought them “well within the
basic rationale of the exception for declarations against in-
terest” and were “critical to Chambers’ defense.” Id. at 302.
The Court concluded: “In these circumstances, where consti-
tutional rights directly affecting the ascertainment of guilt
are implicated, the hearsay rule may not be applied mecha-
nistically to defeat the ends of justice.” Id. The combination
of the limits on impeachment and the exclusion of the con-
fessions led the Court to hold that “under the facts and cir-
cumstances of this case the rulings of the trial court deprived
Chambers of a fair trial.” Id. at 303.
Chambers does not stand alone. It is the key precedent in a
line of cases considering constitutional challenges to rules of
evidence that restrict the defense of an accused. See Washing-
ton v. Texas, 388 U.S. 14, 22 (1967) (rejecting state evidence

5 The Supreme Court’s account of the facts was deliberately terse. It
made no mention at all, for example, of the case’s racial dimensions and
the civil rights boycott at the heart of the events in a small town in rural
Mississippi in 1969. For a more complete account that emphasizes the
gap between local realities and formal legal recognition of civil rights,
see Emily Prifogle, Law and Local Activism: Uncovering the Civil Rights His-
tory of Chambers v. Mississippi, 101 Cal. L. Rev. 445 (2013).
22 No. 14-1898

rule that allowed accused accomplices to testify for prosecu-
tion but not for defense); Green v. Georgia, 442 U.S. 95, 97
(1979) (per curiam) (vacating death sentence where defend-
ant was barred from using same out-of-court confession that
prosecution used to obtain death penalty against declarant);
Crane v. Kentucky, 476 U.S. 683, 691 (1986) (rejecting state
court’s wholesale exclusion of testimony about circumstanc-
es of defendant’s confession); Rock v. Arkansas, 483 U.S. 44, 56
(1987) (rejecting state rule excluding all hypnotically re-
freshed testimony as applied to bar defendant’s own testi-
mony); Montana v. Egelhoff, 518 U.S. 37 (1996) (upholding
state rule barring consideration of evidence of voluntary in-
toxication in determining mens rea); United States v. Scheffer,
523 U.S. 303 (1998) (upholding military rule of evidence bar-
ring use of polygraph test showing “no deception” in denial
of drug use by defendant); Holmes v. South Carolina, 547 U.S.
319, 330 (2006) (rejecting state rule barring defendant from
introducing evidence of third-party guilt when prosecution
has introduced forensic evidence that, if credited, is strong
proof of defendant’s guilt).
In the Chambers line of cases, the Court has balanced
competing interests, weighing the interests in putting on a
full and fair defense against the interests in orderly proce-
dures for adjudication and use of reliable evidence that can
withstand adversarial scrutiny. In striking this balance, the
Court has recognized that “State and federal rulemakers
have broad latitude under the Constitution to establish rules
excluding evidence from criminal trials.” Holmes, 547 U.S. at
324 (brackets and internal quotation marks omitted), quoting
Scheffer, 523 U.S. at 308. Those rules are then put into practice
by trial judges “called upon to make dozens, sometimes
hundreds, of decisions concerning the admissibility of evi-
No. 14-1898 23

dence” in a criminal trial. Crane, 476 U.S. at 689. The latitude
exercised by rulemakers and the trial judges they empower
proves that the right to “present a complete defense” is not
absolute. Id. at 690, quoting California v. Trombetta, 467 U.S.
479, 485 (1984). Nevertheless, “to say that the right to intro-
duce relevant evidence is not absolute is not to say that the
Due Process Clause places no limits upon restriction of that
right.” Montana v. Egelhoff, 518 U.S. 37, 42–43 (1996) (plurality
opinion).
The general constitutional standard can now be stated
this way: rules of evidence restricting the right to present a
defense cannot be “arbitrary or disproportionate to the pur-
poses they are designed to serve.” Rock, 483 U.S. at 56. The
most recent in the Chambers line of cases explained that the
Court has struck down as “arbitrary” those restrictions that
“excluded important defense evidence but that did not serve
any legitimate interests.” Holmes, 547 U.S. at 325. We have
applied this constitutional standard to grant habeas relief in
strong cases. E.g., Harris v. Thompson, 698 F.3d 609 (7th Cir.
2012); Sussman v. Jenkins, 636 F.3d 329 (7th Cir. 2011). We
have also denied relief where there was room for reasonable
jurists to disagree. E.g., Dunlap v. Hepp, 436 F.3d 739 (7th Cir.
2006); Horton v. Litscher, 427 F.3d 498, 504 (7th Cir. 2005).
1. The Parity Principle
One way a state rule of evidence may be arbitrary is
where it restricts the defense but not the prosecution. Several
cases in the Chambers line have emphasized this “‘parity’
principle: a state rule that restricts the presentation of testi-
mony for the defense but not the prosecution will generally
be deemed arbitrary.” Harris, 698 F.3d at 632, citing Akhil
Reed Amar, Sixth Amendment First Principles, 84 Geo. L.J. 641,
24 No. 14-1898

699 (1996). For example, Washington v. Texas struck down a
state rule allowing alleged accomplices to testify against each
other but forbidding them from testifying for each other. 388
U.S. at 22. Green v. Georgia struck down another violation of
the parity principle. In that case state courts excluded hear-
say evidence that the defendant tried to introduce in his cap-
ital sentencing hearing after the state had used that same
hearsay evidence against his accomplice in the accomplice’s
trial. 442 U.S. at 96–97.
The parity approach to evaluating reliability enables “de-
fendants to benefit from the balance that the state tries to
strike when its own evidence-seeking self-interest is at
stake.” See Amar, 84 Geo. L.J. at 699. If the rule excluding
evidence is in fact the product of a genuine balancing of in-
terests by the state, that weighs in favor of respecting the
balance by regarding the evidence as unreliable no matter
which side it favors. See id.
Nothing in the record indicates that the State would have
been able to introduce Amanda’s recorded statement if it had
been inculpatory rather than exculpatory. Whether inculpa-
tory or exculpatory, Amanda “could not vouch for the accu-
racy of a recording that she could not even remember mak-
ing,” and her statement would not qualify as a recorded rec-
ollection regardless. Kubsch II, 866 N.E.2d at 735.
The State thus seems to have struck a genuine balance
that excludes hearsay evidence like this no matter whom it
benefits. But that is not the end of the matter. The Chambers
line of cases can also protect the accused from a restrictive
evidentiary rule that is disproportionate to its purposes.
That leads us to the question of reliability.
No. 14-1898 25

2. Reliability
Reliability is the core of the hearsay rule and its many ex-
ceptions. See Federal Rules of Evidence, Article VIII, Adviso-
ry Committee Notes (1972). Our adversarial system relies
first and foremost on in-court testimony. In court, a trier of
fact may watch and listen to a declarant whose testimony is
offered to prove the truth of its contents, and adverse parties
may further test such testimony through vigorous cross-
examination. “The principal justification for the hearsay rule
is that most hearsay statements, being made out of court, are
not subject to cross-examination.” Rice v. McCann, 339 F.3d
546, 551 (7th Cir. 2003) (Posner, J., dissenting); accord, Feder-
al Rules of Evidence, Article VIII, Advisory Committee
Notes; 30 Wright & Graham, Federal Practice and Procedure
§ 6325 (1997).
When deciding whether to fashion a hearsay exception,
the central question is whether the circumstances and con-
tent of an out-of-court statement give the court confidence
that the statement is sufficiently reliable to admit as evidence
despite the inability to test it directly in court. See, e.g.,
Chambers, 410 U.S. at 298–99 (“A number of exceptions have
developed over the years to allow admission of hearsay
statements made under circumstances that tend to assure
reliability and thereby compensate for the absence of the
oath and opportunity for cross-examination.”); Fed. R. Evid.
807(a)(1) (residual hearsay exception requires “equivalent
circumstantial guarantees of trustworthiness”).
The hearsay portion of Chambers thus turned on whether
McDonald’s hearsay confessions bore sufficient indications
of reliability that a mechanical application of the state hear-
say rule violated Chambers’ right to defend himself at trial.
26 No. 14-1898

The Chambers Court identified four factors that together pro-
vided “considerable assurance” of the reliability of the ex-
cluded confessions. First, each confession was made sponta-
neously to a close acquaintance of the declarant shortly after
the murder. Second, each statement was corroborated by
other evidence. Third, the statements were against the de-
clarant’s own interest. Fourth, the declarant was available at
trial for cross-examination. Id. at 300–01.
Green v. Georgia also addressed the exclusion of hearsay
testimony. Two men, Green and Moore, participated in a
rape and murder. Moore had been convicted and sentenced
to death. At his trial and sentencing, the state had used
against him his out-of-court confession to a friend that he
had fired the fatal shots. Yet when Green was being sen-
tenced and offered the same evidence to show that he was
less culpable than Moore, it was excluded as hearsay. 442
U.S. at 96–97. The Supreme Court reversed, emphasizing the
state’s use of the evidence against Moore as perhaps the
“most important” reason for trusting the reliability of the
testimony. Id. at 97. But the Court also made note of other
“substantial reasons” to treat the confession as reliable. The
confession was made spontaneously to a close friend, it was
against Moore’s penal interest, there was no reason to believe
Moore had any ulterior motive to make it, and there was
ample corroborating evidence. “In these unique circum-
stances,” the Court wrote, “the hearsay rule may not be ap-
plied mechanistically to defeat the ends of justice.” Id., quot-
ing Chambers, 410 U.S. at 302.
C. Amanda’s Statement—Reliable or Not?
Chambers and Green both reversed the exclusion of anoth-
er person’s hearsay confession against penal interest when
No. 14-1898 27

there were substantial indications that the confession was
reliable. The problem posed by Amanda Buck’s recorded in-
terview, and specifically by whether she saw Aaron and Rick
Milewski on the afternoon of the murders or on another day,
is quite different.
Weighing in favor of reliability, the interview was record-
ed, so there is no doubt about what was said, and the inter-
view took place just a few days after the events in question,
when memories were fresh. In addition, Amanda was quite
detailed and specific in her account. She had nothing to gain
by lying and there is no indication that she did so.
Other factors weigh against her statement’s reliability,
however. The extent of corroboration was central to the rea-
soning in Chambers. McDonald’s four independent confes-
sions corroborated each other. They were also corroborated
by the testimony of other witnesses: one who saw McDonald
shoot the officer, another who saw him with a gun immedi-
ately afterward, and another who knew he had owned a gun
like the murder weapon and later replaced it with another
similar gun. Chambers, 410 U.S. at 293 n.5, 300. Furthermore,
in Green the Court described the corroborating evidence
there as “ample,” and of course the state had treated the oth-
er man’s confession to firing the fatal shots as sufficiently re-
liable to use it to sentence him to death. 442 U.S. at 97.
In this case, by contrast, there simply is no corroboration
of Amanda’s statement on the critical point, which is wheth-
er Aaron and Rick were at their home alive and well between
3:30 and 3:45 on the day they were murdered.6 (No corrobo-

6 Kubsch points out that Rick Milewski was driving not his own
black truck but a white truck that he had borrowed from his brother. In
28 No. 14-1898

ration, that is, other than Monica’s initial statement that she
also saw Aaron at home that afternoon, a statement that
Monica later corrected, that was never offered as evidence,
and that could not have been admitted as substantive evi-
dence to corroborate Amanda’s statement.) The minimal cor-
roboration for Amanda’s recorded statement distinguishes
this case from Chambers and Green and their reasoning. See
Rice, 339 F.3d at 550 (affirming denial of habeas relief in part
because state court found hearsay statements in question
were not corroborated).
The availability of cross-examination was also central to
Chambers: “Finally, if there was any question about the truth-
fulness of the extrajudicial statements, McDonald was pre-
sent in the courtroom and was under oath. He could have
been cross-examined by the State, and his demeanor and re-
sponses weighed by the jury.” 410 U.S. at 301.
In this respect, as well, the evidence here is quite differ-
ent from the confessions in Chambers. Unlike the declarant in
Chambers, Amanda was essentially unavailable for cross-
examination. She took the stand at trial but testified that she
did not remember being interviewed by the police or what
she said to them. “A declarant is considered to be unavaila-
ble as a witness if the declarant … testifies to not remember-
ing the subject matter.” Ind. R. Evid. 804(a)(3); Fed. R. Evid.
804(a)(3).

her statement, Amanda said that Rick was driving a white truck that
day. But as Kubsch also acknowledges, Rick had borrowed that truck
from his brother a few weeks before the murders. The color of the truck
does not corroborate Amanda’s statement about which afternoon she
saw Rick and Aaron at home.
No. 14-1898 29

In addition, during the recorded interview, Amanda was
never pushed on the critical details—the date and time she
saw Aaron and Rick at their home. The interviewing officer
was simply taking her account as she spoke in an interview
in the early stages of the investigation. Amanda was not un-
der oath, and Sergeant Reihl did not test her story to see how
certain and accurate she might have been. Sergeant Reihl’s
gentle questioning, which was surely appropriate for his
purpose at the time, was not remotely like cross-examination
of the alibi witness in a murder trial where the stakes are life
and death. There was no cross-examination here; there was
not even a mild challenge.
By comparison, when a witness is unavailable, it is clear
that even former testimony is admissible under the rules of
evidence only if it is offered against a party who had both an
opportunity and a similar motive to develop that witness’s
testimony by direct, cross-, or redirect examination. Ind. R.
Evid. 804(b)(1); Fed. R. Evid. 804(b)(1).
Moreover, if the recorded statement had been admitted,
the State would have been unable to test its accuracy
through cross-examination. The prosecutor would have been
stuck questioning a witness who did not even remember
making the statement. See Fed. R. Evid. 804(a)(3) advisory
committee note (“the practical effect” of lack of memory “is
to put the testimony beyond reach”); 2 McCormick on Evi-
dence § 253 (7th ed.) (a declarant who does not remember
the subject matter of her testimony “is simply unavailable by
any realistic standard”).
In the adversarial system of Anglo-American law, we put
great trust in the power of cross-examination to test both the
honesty and the accuracy of testimony. It is virtually an arti-
30 No. 14-1898

cle of faith that cross-examination is the “greatest legal en-
gine ever invented for the discovery of truth.” California v.
Green, 399 U.S. 149, 158 (1970), quoting 5 Wigmore on Evi-
dence § 1367. Without cross-examination to test “any ques-
tion about the truthfulness” of Amanda’s recorded state-
ment, a powerful assurance of reliability present in Chambers
is absent here. Chambers, 410 U.S. at 301; see also Christian v.
Frank, 595 F.3d 1076, 1085 (9th Cir. 2010) (reversing grant of
habeas relief under Chambers; witness’s “unavailability con-
trasts sharply with the availability of McDonald in Chambers,
which the Supreme Court of the United States stressed great-
ly enhanced the reliability of the extrajudicial statements in
that case”).7

7Our dissenting colleague contends that this case is like Chambers
because Kubsch, like Chambers, tried to show that someone else commit-
ted the murders—Kubsch’s long-time friend Brad Hardy. Post at 95–96.
We disagree. In Chambers, the evidence against McDonald would have
exonerated Chambers; there was no evidence that they acted together.
Readers of the dissent might think there was a similar either-or dynamic
at work here. There was not. The prosecution argued that Hardy had
either helped Kubsch or had been set up by Kubsch as his fall guy.
Hardy testified in both of Kubsch’s trials, though at the time of the
first trial he was charged with conspiring with Kubsch to commit the
murders. (The charges were later dismissed.) Kubsch called Hardy on
the day of the murders at 9:11 a.m. Hardy and his mother, Constance
Hardy, each testified that Constance drove Hardy to Kubsch’s workplace
two hours later when Kubsch began his early lunch break. Hardy testi-
fied that Kubsch then drove him to a parking lot near the Kubsch house
and asked him to sneak up to the house from the rear to see if Beth was
home. Hardy also testified that the day after the murders Kubsch asked
him to lie about their activities the day before. (Kubsch denied Hardy’s
account.)
No. 14-1898 31

D. The Standards of Review and Their Application
To win a federal writ of habeas corpus, Kubsch must
show that he is in custody in violation of the Constitution or
laws or treaties of the United States. 28 U.S.C. § 2254(a).
Since the Antiterrorism and Effective Death Penalty Act
(AEDPA) amended § 2254 in 1996, though, if a state court
has adjudicated a federal claim on the merits, it is not
enough for the petitioner to show a violation of federal law.
The petitioner must also show that the state court adjudica-
tion of the claim “resulted in a decision that was contrary to,
or involved an unreasonable application of, clearly estab-
lished Federal law, as determined by the Supreme Court of
the United States,” 28 U.S.C. § 2254(d)(1), or “resulted in a
decision that was based on an unreasonable determination
of the facts in light of the evidence presented in the State
court proceeding.” 28 U.S.C. § 2254(d)(2). On Kubsch’s claim

Phone records showed that Kubsch again called Brad Hardy on the
day of the murders at 4:44 p.m. It is undisputed that Kubsch arrived at
Brad and Constance Hardy’s house 45 minutes later and stayed for an
hour before going to his home. The defense argued that this visit was for
the purpose of “invit[ing] [Brad] out to dinner that night.” Tr. at 3301. It
is curious that, on the evening of his wife’s birthday—when Kubsch
claims not to have seen Beth all day and after Beth’s mother called him to
say that she was concerned about not hearing from Beth—Kubsch would
take an hour-long detour to Hardy’s house just to extend a dinner invita-
tion, especially when he had spoken to Hardy just 45 minutes earlier. In
light of this curious detour, the fact that Beth’s credit cards were later
found in the woods near Hardy’s house could be viewed as implicating
Kubsch as much as Hardy. In short, the “significant evidence pointing to
Hardy” did not necessarily tend to exonerate Kubsch, as the dissent sug-
gests and in contrast to the evidence related to Gable McDonald in
Chambers.
32 No. 14-1898

under Chambers, our focus is on the state court’s legal analy-
sis under subsection (d)(1), not factual findings under (d)(2).
We agree with the district court that the Indiana Supreme
Court adjudicated on the merits Kubsch’s federal constitu-
tional claim under Chambers. Footnote 7 of the state court’s
opinion made that much clear, see Kubsch II, 866 N.E.2d at
735 n.7, so we must evaluate the decision under § 2254(d)(1).
Section 2254(d)(1) has two distinct prongs, the narrow “con-
trary to” prong and the broader “unreasonable application”
prong.
1. “Contrary to” Federal Law?
On the first prong, the Indiana Supreme Court’s adjudi-
cation of the Chambers claim was not “contrary to Y clearly
established Federal law, as determined by the Supreme
Court of the United States.” Because no Supreme Court cases
“confront ‘the specific question presented by this case,’ the
state court’s decision could not be ‘contrary to’ any holding
from” that Court. Woods v. Donald, 575 U.S. —, 135 S. Ct.
1372, 1377 (2015) (per curiam) (summarily reversing grant of
habeas petition), quoting Lopez v. Smith, 574 U.S. —, 135 S.
Ct. 1, 4 (2014) (per curiam). Under § 2254(d), clearly estab-
lished federal law includes only “the holdings, as opposed to
the dicta,” of Supreme Court decisions. White v. Woodall, 572
U.S. —, 134 S. Ct. 1697, 1702 (2014), quoting Howes v. Fields,
565 U.S. —, 132 S. Ct. 1181, 1187 (2012).
To note again just the most obvious differences between
this case and Chambers, Amanda did not make her statement
spontaneously to a close acquaintance, her statement was
not against interest, her statement was not corroborated, and
she was not subject to cross-examination about the state-
No. 14-1898 33

ment. Any of those distinctions would be enough to demon-
strate that the Indiana Supreme Court did not confront
“facts that are materially indistinguishable from a relevant
Supreme Court precedent” and arrive at the opposite result.
See Williams v. Taylor, 529 U.S. 362, 405 (2000).
2. “Unreasonable Application” of Federal Law?
The second and broader prong, whether the Indiana Su-
preme Court’s rejection of Kubsch’s claim under Chambers
was, also in the terms of § 2254(d)(1), an “unreasonable ap-
plication” of clearly established federal law as determined
by the Supreme Court of the United States, poses a more dif-
ficult question. The state court’s rejection of the Chambers
claim was at best incomplete and at worst wrong and unrea-
sonably so. That poses a methodological question on which
federal law is not settled. We explore that methodological
question below but ultimately conclude that Kubsch’s claim
under Chambers fails whether or not we apply deferential re-
view under AEDPA.
The narrow holding of Chambers, based on the combina-
tion of the restrictions on impeachment and the exclusion of
multiple reliable hearsay confessions by a declarant subject
to cross-examination, topped off by the “under the facts and
circumstances of this case” qualification, see 410 U.S. at 303,
means that state courts have considerable latitude in inter-
preting and applying Chambers. See Dunlap v. Hepp, 436 F.3d
739, 744 (7th Cir. 2006), quoting Yarborough v. Alvarado, 541
U.S. 652, 664 (2004). Nevertheless, the broader standard that
has emerged from Chambers and subsequent cases is that
courts cannot impose restrictions on defense evidence that
are arbitrary or disproportionate to the purposes they are
designed to serve. See Holmes, 547 U.S. at 325; Rock, 483 U.S.
34 No. 14-1898

at 56. The general standard requires a balance of competing
interests.
The open texture of that standard and the important fac-
tual differences between this case and Chambers—lack of cor-
roboration and lack of opportunity for meaningful cross-
examination—mean that the Indiana courts could have reject-
ed Kubsch’s claim under Chambers without unreasonably
applying clearly established federal law as determined by
the Supreme Court of the United States. See 28 U.S.C. §
2254(d)(1); see generally, e.g., Woods v. Donald, 135 S. Ct. at
1377 (“where the precise contours of a right remain unclear,
state courts enjoy broad discretion in their adjudication of a
prisoner’s claims”), quoting White v. Woodall, 572 U.S. —, 134
S. Ct. at 1705, quoting in turn Lockyer v. Andrade, 538 U.S. 63,
76 (2003). Only rarely has the Supreme Court “held that the
right to present a complete defense was violated by the ex-
clusion of defense evidence under a state rule of evidence.”
Nevada v. Jackson, 133 S. Ct. 1990, 1991–92 (2013) (per curiam)
(summarily reversing grant of habeas relief on Chambers
claim: “no prior decision of this Court clearly establishes that
the exclusion of this evidence violated respondent’s federal
constitutional rights”).
Thus, when habeas relief has been granted on a Chambers
claim, the facts were a much closer fit to the Supreme Court
precedents. In Cudjo v. Ayers, 698 F.3d 752 (9th Cir. 2012), for
example, the state court had found that the hearsay testimo-
ny was “trustworthy and material exculpatory evidence”
that should have been admitted under state law but still de-
clined to grant relief under Chambers. See id. at 763. Cudjo
thus held that its facts were “materially indistinguishable”
from Chambers. Id. at 767, quoting Williams, 529 U.S. at 405. In
No. 14-1898 35

discussing the rule that defendants have a constitutional
right to present a complete defense, Cudjo also commented
that “it would be extremely difficult to say that a state trial
court engaged in an ‘unreasonable application’ of this rule
when faced with new factual circumstances.” Id.; cf. Cudjo,
698 F.3d at 770–74 (O’Scannlain, J., dissenting).
Accordingly, if the Indiana Supreme Court had an-
nounced its rejection of Kubsch’s claim under Chambers
without any explanation at all, then we would affirm the de-
nial of habeas relief without further ado. See Harrington v.
Richter, 562 U.S. 86, 98 (2011) (“Where a state court’s decision
is unaccompanied by an explanation, the habeas petitioner’s
burden still must be met by showing there was no reason-
able basis for the state court to deny relief.”).
But the Indiana Supreme Court was not silent on the
point. It rejected Kubsch’s claim under Chambers in a foot-
note consisting of one sentence and one citation:
The availability of this testimony [from Monica
Buck and Sergeant Reihl to the effect that
Amanda had been mistaken] is also the reason
why Kubsch’s claim that he was denied his
federal constitutional right to present a defense
fails. See Chambers v. Mississippi, 410 U.S. 284,
302 (1973) (protecting defendant’s due process
right by recognizing an exception to applica-
tion of evidence rules where evidence found to
be trustworthy).
866 N.E.2d at 735 n.7.
This terse footnote shows that the state court was aware
of the federal constitutional claim and the governing Su-
36 No. 14-1898

preme Court precedent. It cited the page of the Chambers
opinion finding that the multiple hearsay confessions by
McDonald “bore persuasive assurances of trustworthiness”
and should have been admitted because they were so critical
to the defense. Keeping in mind the presumption that state
courts know and follow the law, see Woodford v. Visciotti, 537
U.S. 19, 24 (2002) (per curiam), we find it sufficiently clear
that the state court found that Amanda’s statement was not
sufficiently reliable to require its admission under Chambers.
The state court adjudicated the merits, so its decision re-
quires deference under AEDPA.
The problem is that the only reason actually given by the
Indiana Supreme Court—the availability of contradictory
testimony from Amanda’s mother and Sergeant Reihl—is the
weakest reason that might support that result. It was a good
reason to treat as harmless the exclusion of the recorded
statement as impeachment, but not as substantive evidence.
The mere existence of conflicting or impeaching evidence is
not a sufficient basis, or even a reasonable basis, for rejecting
the statement as substantive evidence. Conflicting evidence
would simply present a fact issue for the jury to weigh after
hearing all of that evidence. Perhaps the state court also had
in mind the stronger reasons for excluding Amanda’s rec-
orded statement, especially the lack of corroboration and the
lack of an opportunity for cross-examination, but if so it did
not mention them.
What is the role of the federal courts when a state court
offers such a weak reason for a result that could be a reason-
able application of federal law? See Brady v. Pfister, 711 F.3d
818, 824–27 (7th Cir. 2013) (identifying problem and discuss-
ing Supreme Court’s limited guidance). We must review the
No. 14-1898 37

actual reason deferentially. But if that reason was unreason-
able, do we proceed to de novo review? Or do we, instead of
doing de novo review, hypothesize reasons the court could
have used to see if they are reasonable under AEDPA? See
Stitts v. Wilson, 713 F.3d 887, 893 (7th Cir. 2013) (raising but
not answering this question).8
We have interpreted Richter as instructing federal courts
to consider what arguments “could have supported” a state
court decision when the state court “gave some reasons for
an outcome without necessarily displaying all of its reason-
ing.” Hanson v. Beth, 738 F.3d 158, 163–64 (7th Cir. 2013) (af-
firming denial of relief on Chambers claim based on exclusion
of evidence); see also Jardine v. Dittmann, 658 F.3d 772, 777
(7th Cir. 2011) (“This court must fill any gaps in the state
court’s discussion by asking what theories ‘could have sup-
ported’ the state court’s conclusion.”), quoting Richter, 562
U.S. at 102.9
The Indiana Supreme Court’s stated rationale for reject-
ing Kubsch’s claim can be described fairly as incomplete. So

8 In Stitts we considered whether to “look through” a state supreme
court’s ruling to a lower state court’s decision. In this case, we cannot
“look through” the Indiana Supreme Court’s ruling on the Chambers
claim. The claim was not presented to the trial court, and capital appeals
in Indiana go directly to the Indiana Supreme Court.
9 Makiel v. Butler, 782 F.3d 882, 905–06 (7th Cir. 2015), presented a re-
lated but distinct issue. In Makiel, the state court gave two reasons why
the exclusion of certain evidence did not violate the petitioner’s right to
present a complete defense. One reason was flawed but the second was
sound. The sound second reason was enough to call for AEDPA defer-
ence. Here, by contrast, the state court gave only one reason to reject the
constitutional claim, and that reason is flawed.
38 No. 14-1898

long as we have an obligation under § 2254(d)(1) to fill gaps
or to complete the state court’s reasoning, the result here is
not an unreasonable application of federal constitutional
law, and relief must be denied on this claim.10

10Most circuits endorse this approach that allows and even requires
federal courts to complete or fill the gaps in state courts’ reasoning in
support of results that are not unreasonable in light of Supreme Court
precedent. See Foxworth v. St. Amand, 570 F.3d 414, 429 (1st Cir. 2009)
(“on habeas review, the ultimate inquiry is not the degree to which the
state court’s decision is or is not smoothly reasoned; the ultimate inquiry
is whether the outcome is reasonable”); Rashad v. Walsh, 300 F.3d 27, 45
(1st Cir. 2002) (where federal courts were troubled by gaps in state
court’s rationale: “It is not our function, however, to grade a state court
opinion as if it were a law school examination.”); Cruz v. Miller, 255 F.3d
77, 86 (2d Cir. 2001) (“deficient reasoning will not preclude AEDPA def-
erence”); Collins v. Sec’y of Pennsylvania Dep’t of Corr., 742 F.3d 528, 548
(3d Cir. 2014) (while state court adjudication of Strickland claim consisted
of “admittedly cursory statements, AEDPA requires that we determine
what arguments or theories supported … or could have supported, the
state court’s decision”) (citation and internal quotation marks omitted);
Robinson v. Polk, 438 F.3d 350, 358 (4th Cir. 2006) (“In assessing the rea-
sonableness of the state court’s application of federal law, therefore, the
federal courts are to review the result that the state court reached, not
whether its decision was well reasoned.”) (brackets, citations, and inter-
nal quotation marks omitted); Higgins v. Cain, 720 F.3d 255, 261 (5th Cir.
2013) (“In considering whether the state court’s decision constituted an
unreasonable application of clearly established federal law, ‘a federal
habeas court is authorized by Section 2254(d) to review only a state
court’s “decision,” and not the written opinion explaining that deci-
sion.’”), quoting Neal v. Puckett, 286 F.3d 230, 246 (5th Cir. 2002) (en
banc); Holder v. Palmer, 588 F.3d 328, 341 (6th Cir. 2009) (“The law re-
quires such deference to be given even in cases, such as this one, where
the state court's reasoning is flawed or abbreviated.”); Williams v. Roper,
695 F.3d 825, 831 (8th Cir. 2012) (“In reviewing whether the state court’s
decision involved an unreasonable application of clearly established fed-
eral law, we examine the ultimate legal conclusion reached by the court,
No. 14-1898 39

3. De Novo Review
There is room to argue, however, that the state court’s
footnote 7 was not just incomplete but wrong, and unrea-
sonably so. And there is room to argue that where the state
court has provided a rationale for its decision, the federal
courts should focus their attention on the reasons actually
given rather than hypothesize a better set of reasons. See
Wiggins v. Smith, 539 U.S. 510, 528–29 (2003) (holding state
court’s rationale unreasonable without considering other
possibilities); Frantz v. Hazey, 533 F.3d 724, 737–38 & n.15 (9th
Cir. 2008) (en banc) (confining analysis to reasons actually
given by state court, without hypothesizing alternative ra-
tionales); Oswald v. Bertrand, 374 F.3d 475, 483 (7th Cir. 2004)
(“reasonableness of a decision ordinarily cannot be assessed
without considering the quality of the court’s reasoning,”
though “ultimate question Y is not whether the state court
gets a bad grade for the quality of its analysis but Y whether
the decision is an unreasonable application of federal law”).
As we explained in Brady v. Pfister, when evaluating a state
court’s reasoning in habeas cases, the Supreme Court has fo-

not merely the statement of reasons explaining the state court's deci-
sion.”) (citation omitted); Williams v. Trammell, 782 F.3d 1184, 1199–1200
(10th Cir. 2015) (“uncertainty” regarding rationale for a sparse state court
decision “does not change our deference;” federal court still must identi-
fy theories that could have supported the decision); Lee v. Comm’r, Ala-
bama Dep’t of Corr., 726 F.3d 1172, 1210–14 (11th Cir. 2013) (applying
AEDPA deference to incomplete state court opinion; state court need not
“show its work” by mentioning all circumstances relevant to Batson
claim); but see Frantz v. Hazey, 533 F.3d 724, 737–38 & n.15 (9th Cir. 2008)
(en banc) (confining evaluation of “unreasonable application” prong to
actual reasons given).
40 No. 14-1898

cused on the reasons actually given by state courts without
engaging in the exercise of trying to construct reasons that
could have supported the same result. See 711 F.3d at 826,
citing Rompilla v. Beard, 545 U.S. 374 (2005), and Wiggins v.
Smith, 539 U.S. 510 (2003). So AEDPA deference toward state
court decisions that reach defensible results for bad or in-
complete reasons is not necessarily settled law at this point.
This debate over methodology under § 2254(d) may be
ripening for a resolution. In Hittson v. Chatman, 576 U.S. —,
135 S. Ct. 2126 (2015), a short opinion concurring in denial of
certiorari reminded circuit and district judges of the Court’s
decision in Ylst v. Nunnemaker, 501 U.S. 797 (1991), in which
the Court instructed that when federal habeas corpus courts
review an unexplained order from a state appellate court,
they should “look through” that unexplained order and fo-
cus on the last reasoned rejection of the federal claim. See
501 U.S. at 803–04. In the Hittson concurring opinion, Justice
Ginsburg (joined by Justice Kagan) wrote that the Nunne-
maker “look through” presumption remains valid after Rich-
ter. See 135 S. Ct. at 2127, discussing Richter, 562 U.S. at 99–
100, citing Nunnemaker with approval; see also Brumfield v.
Cain, 576 U.S. —, 135 S. Ct. 2269, 2276 (2015) (applying Nun-
nemaker “look-through” approach to evaluate and reverse
lower state court’s factual findings supporting denial of evi-
dentiary hearing under § 2254(d)(2)); Johnson v. Williams, 568
U.S. —, — n.1, 133 S. Ct. 1088, 1094 n.1 (2013) (citing Nunne-
maker with approval); Hawthorne v. Schneiderman, 695 F.3d
192, 199–201 (2d Cir. 2012) (Calabresi, J., concurring) (argu-
ing that practice under Richter of inventing hypothetical rea-
sons for state court decision promotes neither comity nor ef-
ficiency).
No. 14-1898 41

Justice Ginsburg’s opinion in Hittson argued that the
Richter practice of hypothesizing rationales for state court
rejections of federal claims should be limited to cases where
no state court explained the rejection, and that where the
state court’s real reasons can be ascertained, the inquiry un-
der § 2254(d)(1) “can and should be based on the actual ‘ar-
guments or theories [that] supported … the state court’s de-
cision.” 135 S. Ct. at 2128–29, quoting Richter, 562 U.S. at 102.
This statement may imply that federal courts should shift to
de novo review as soon as they find that the reason actually
given by a state court was unreasonable, without trying to
hypothesize alternative rationales.
Because of this uncertainty in whether we may “com-
plete” the state court’s reasoning on this Chambers claim, it is
prudent for us also to consider Kubsch’s Chambers claim un-
der a de novo standard of review. Even if we conclude that
the state court’s footnote 7 was an unreasonable application
of Chambers to reject Kubsch’s claim, that would not neces-
sarily entitle Kubsch to habeas relief. He would still need to
show on the merits that his constitutional rights were in fact
violated, as § 2254(a) requires for a grant of actual relief. See
Brady, 711 F.3d at 827 (applying de novo review in the alterna-
tive); Mosley v. Atchison, 689 F.3d 838, 852–54 (7th Cir. 2012)
(where state court decision was unreasonable under
§ 2254(d)(1), remanding to district court to determine merits
de novo under § 2254(a)).
If de novo review applies, the issue is closer than under
§ 2254(d)(1), but we conclude that the exclusion of Amanda’s
recorded statement as substantive evidence did not violate
Kubsch’s federal constitutional right to put on a defense. As
explained above, Amanda’s statement is not corroborated on
42 No. 14-1898

the critical facts by any other evidence, and she was never
subjected to meaningful cross-examination. Even during the
recorded interview itself, she was never pushed by the inter-
viewer about the critical day and time, nor about the possi-
bility that her memory had confused events of two different
days.
Those facts distinguish this case from Chambers, which
was, on its face, a very narrow opinion. Recall that the hold-
ing in Chambers depended on the combination of the limits
the “voucher rule” placed on cross-examination and the ex-
clusion of the three hearsay confessions, which were directly
corroborated in many ways and had other indications of re-
liability. 410 U.S. at 302–03.
Even applying the more general principles from the
Chambers line of cases, we are not persuaded that the Consti-
tution requires the general rule against hearsay to give way
to Kubsch’s interest in offering as substantive evidence a
recorded, exculpatory interview of a witness who was in ef-
fect not available for cross-examination and whose account
does not have significant corroboration on the critical points.
A vast literature attempts to explain the complex edifice
of American hearsay law. A helpful and authoritative expla-
nation came from the Advisory Committee on the Federal
Rules of Evidence, published as an introductory note to the
hearsay article in the Rules. A helpful and more detailed
survey is available in 30 Wright & Graham, Federal Practice
and Procedure §§ 6321–6333 (1997). As noted above, issues
of reliability and trustworthiness are front and center in de-
ciding whether to relax the general prohibition on hearsay.
Our legal system relies primarily on in-person testimony
subject to meaningful cross-examination, the “greatest en-
No. 14-1898 43

gine ever invented for the discovery of truth,” to test evi-
dence. See California v. Green, 399 U.S. 149, 158 (1970), quot-
ing Wigmore on Evidence § 1367; see also Rice v. McCann,
339 F.3d 546, 551 (7th Cir. 2003) (Posner, J., dissenting) (“The
principal justification for the hearsay rule is that most hear-
say statements, being made out of court, are not subject to
cross-examination.”).
Lest this reasoning seem like reflexive devotion at the al-
tar of cross-examination, we draw help from Professors
Wright and Graham to explain why this is so important.
Their treatise identifies four dangers of hearsay: (1) defects
in the declarant’s perception; (2) defects in the declarant’s
memory; (3) defects in narration by both the declarant and
the witness; and (4) the declarant’s lack of sincerity or hones-
ty. 30 Wright & Graham, Federal Practice and Procedure
§§ 6324. Without an opportunity for cross-examination be-
fore the trier of fact, it can be difficult to test hearsay for
these defects. Most hearsay exceptions have evolved from
situations providing circumstantial guaranties of trustwor-
thiness that seem to be sufficient substitutes for that direct
scrutiny in a trial. See id., § 6333.
In the case of Amanda’s recorded statement, the third
and fourth dangers seem minimal. The recording eliminates
the risk that Amanda’s statement would be relayed inaccu-
rately, and she had no apparent difficulty describing what
she remembered. The nine-year-old Amanda in the inter-
view was also a disinterested witness, old enough to know
she should tell the truth and with no apparent reason to de-
ceive the police intentionally.
The first two dangers remain, however, with no meaning-
ful protections under these circumstances. There simply is
44 No. 14-1898

no way to test directly, by cross-examination or otherwise,
the accuracy of the nine-year-old Amanda’s memories of the
past several days, to test the possibility that she was misre-
membering what and whom she had seen where and on
which days. The accuracy of her memory was not tested or
even challenged during the recorded interview itself, nor
was the importance of being accurate about the time and
date brought to her attention in the interview. Nor is there
other evidence corroborating the recorded account as to the
critical date and time.
In light of these considerations, it was not arbitrary or
disproportionate to enforce the rules of evidence to exclude
Amanda’s recorded statement as substantive evidence. Ac-
cepting Kubsch’s theory, on the other hand, would upset a
good deal of the rules of evidence developed over genera-
tions to find the right balance so that trials can be decided
fairly and on the basis of reliable evidence. As the prosecutor
said in the trial court here, we could just show juries a series
of videotaped, ex parte witness interviews, but that is not
how we do trials in our legal system. There is no indication
in the narrow Chambers opinion that such a sweeping result
was intended then. Nor do the Supreme Court’s later cases in
the Chambers line endorse such a sweeping result.
Kubsch argues that he seeks only a narrow exception,
comparable to the narrow decision in Chambers. He tries to
limit the rule he seeks to hearsay witness statements that are
recorded (ensuring accuracy of transmission), about recent
events (fresh in the witness’s memory), detailed, and from
disinterested witnesses, at least where the evidence would
be critical to the defense. With inexpensive recording tech-
nology widely available, however, we can expect that such
No. 14-1898 45

evidence will often be available. Kubsch’s theory would thus
expand dramatically the availability, at least to the accused,
of hearsay evidence that cannot be subjected to meaningful
cross-examination. Considering the Chambers issue de novo,
we believe Kubsch is seeking a significant and unwarranted
expansion of existing doctrine, unmoored from the critical
assurances that corroboration and cross-examination pro-
vided in Chambers itself.
We do not doubt that hearsay rules sometimes exclude
evidence that is in fact accurate. They also exclude a good
deal of evidence that is unreliable. Those rules have evolved
based on experience to prevent the use of inaccurate and un-
reliable hearsay in trials. We also must recognize the risk of
error in our human and fallible criminal justice system, espe-
cially in a death-penalty case. That is why Chambers was de-
cided as it was, though the sentence there had been life in
prison rather than death. In that exceptional case, the famil-
iar rules of evidence worked arbitrarily to exclude reliable
evidence of innocence.
The risk of serious error is not enough, however, to open
the gates to all hearsay of this type, especially where it is not
corroborated as it was in Chambers and where it is not subject
to meaningful cross-examination. The unavoidable risk of
error may offer a strong argument against the death penalty
as a matter of policy, but that is not a choice available to us.
See, e.g., Glossip v. Gross, 576 U.S. —, 135 S. Ct. — (2015) (all
opinions).
Accordingly, we affirm the district court’s denial of relief
on the Chambers claim. The state court’s result on this ques-
tion was not an unreasonable application of federal law. And
even if the state court’s incomplete and unsatisfactory ra-
46 No. 14-1898

tionale had amounted to an unreasonable application of fed-
eral law, Kubsch’s claim does not prevail on the merits under
de novo review.
III. Ineffective Assistance of Counsel for Amanda’s Statement
Kubsch approaches Amanda’s statement from a different
angle by arguing that even if his stand-alone claim under
Chambers fails, his trial counsel provided ineffective assis-
tance by failing to do a better job in trying to have the re-
cording admitted into evidence. The Indiana Supreme Court
rejected this claim on appeal from the denial of post-
conviction relief, finding that it was barred by the doctrine of
res judicata. Kubsch III, 934 N.E.2d at 1143 n.2.
Under the controlling standard from Strickland v. Wash-
ington, 466 U.S. 668, 687 (1984), Kubsch must show (1) that
his trial lawyers’ performance was deficient, meaning that it
fell below an objective standard of reasonableness in light of
prevailing professional norms, id. at 690, and (2) that the de-
ficient performance prejudiced his case, meaning that there
is a reasonable probability that, but for the lawyers’ unpro-
fessional errors, the result of the proceeding would have
been different, id. at 694. Kubsch has not made either show-
ing.
The Indiana Supreme Court’s res judicata holding was
reasonable as far as it went. To the extent that Kubsch was
arguing that the recorded interview should have been ad-
mitted and would have made a difference in the trial, the
state court had already decided those questions against
Kubsch in the direct appeal. Kubsch II, 866 N.E.2d at 734–35.
A post-conviction petitioner cannot avoid claim preclusion
No. 14-1898 47

by merely repackaging an earlier claim. E.g., Reed v. State,
856 N.E.2d 1189, 1194 (Ind. 2006).
Kubsch’s post-conviction argument on this score was not,
however, merely a repackaging of the claim that the record-
ing should have been admitted as evidence. He also argued
and tried to offer evidence that if his trial lawyers had taken
some additional steps, the interview would have been ad-
mitted into evidence and was reasonably likely to change the
jury’s verdict. The state court’s res judicata holding did not
engage that evidence and argument.
Even if we review this claim de novo, however, Kubsch
has not shown that his trial lawyers were constitutionally
deficient. It is not as though the trial lawyers overlooked the
issue. Several months before the second trial, Amanda testi-
fied in a deposition where her mother was also present. See
Tr. 2983–84; 3013. We do not have that transcript, but the
lawyers obviously did. And they had the opportunity to talk
to Amanda’s mother Monica as well. The lawyers made clear
in their post-conviction testimony that they had no real in-
terest in anything Amanda or Monica might say from the
witness stand; they wanted the recording in evidence. PCR
Tr. 106; Tr. 3028.
The trial transcript shows they worked hard to convince
the trial court to admit the recording. See Tr. 2982–90; 3010–
35; 3112–23. They were not successful because they could not
lay a sufficient foundation to admit the recording under
Rule 803(5) as recorded recollection, and as explained above,
the inability to use it to impeach the non-inculpatory “prob-
ably didn’t see him” portion of Amanda’s brief trial testimo-
ny was harmless. To change this result, Kubsch needed to
come forward in the post-conviction proceedings with evi-
48 No. 14-1898

dence or new legal arguments that were available to his trial
lawyers, clearly should have been presented, and were rea-
sonably likely to turn the tide. As the district court ex-
plained, he failed to do so. Kubsch, 2013 WL 6229136, at *39–
40.
Kubsch criticizes his trial lawyers for having failed to
correct or challenge what he says is misinformation about
the reports that Amanda and her mother had been mistaken
in their interview with Sergeant Reihl, and argues that they
should have investigated in more detail her mother’s state-
ment of March 2000 asserting that they had mixed up
Thursday and Friday in the videotaped interview. Kubsch
has not shown what that further investigation would have
uncovered, let alone how it would have helped him.
Contrary to the dissent’s assertion, the trial judge did not
keep out Amanda’s videotaped statement because he
thought it would have been easily impeached. When the
prosecution and defense were debating the admissibility of
the statement before the trial court, the prosecutor argued
against admitting the statement “full well knowing that the
little girl was mistaken” and that her mother would testify to
that effect. Tr. 3015–16. The trial judge immediately respond-
ed: “The jury judges that. The jury judges if the girl is right
or the mother is right.” Tr. 3016. The judge kept the recorded
hearsay statement out as substantive evidence because it did
not qualify as a recorded recollection, and he kept it out as
impeachment because Amanda had said nothing worth im-
peaching.
In this appeal, the specific criticisms of counsel, by both
Kubsch and our dissenting colleague, are based on specula-
tion rather than the sort of evidence needed to support the
No. 14-1898 49

claim. Kubsch developed the factual record for this claim of
ineffective assistance of counsel in a three-day evidentiary
hearing in a state trial court in 2008. That is the record before
us on this question. See Cullen v. Pinholster, 563 U.S. —, 131 S.
Ct. 1388, 1398 (2011); 28 U.S.C. § 2254(d)(2) & (e).
Kubsch’s new lawyers called both of his trial lawyers as
witnesses in the post-conviction hearing. The transcript
shows that they were asked a few questions about Amanda’s
recorded interview and her mother’s statement from March
2000, but there simply was no inquiry into the lawyers’ sup-
posed “failures” on this score. Nor was there any effort to
show what would have happened if the trial lawyers had
done what Kubsch’s new lawyers argue should have been
done. They did not call Amanda or Monica or anyone else to
fill in the factual gaps. That proceeding and that hearing
were Kubsch’s opportunity to make a factual record showing
deficient performance that was harmful to his case. He simp-
ly did not make that showing.
Our dissenting colleague finds the trial lawyers deficient
in some additional ways: for not having asked Amanda if
her statements in the interview were accurate, if she was ac-
tually the girl shown in the video, and if she would have
told the police the truth; and for having failed to challenge
Lonnie Buck’s account of the correction on the date, to call
Monica to corroborate Amanda’s answers in the interview, to
track down bank records for Monica’s deposit of her
paycheck, and to pursue corroboration about the school field
trip. Post at 87–88. But again, there is no factual record to
support such speculation about what these efforts would
have shown. Kubsch’s post-conviction lawyers did not ques-
tion his trial lawyers on the witness stand about these mat-
50 No. 14-1898

ters, nor did they track down and offer the evidence that the
dissent says might have helped.
This is not to suggest that Kubsch’s post-conviction law-
yers were themselves anything other than highly competent
and diligent. Kubsch is now being represented by at least his
sixth team of capable and experienced capital defense law-
yers. See Ind. R. Crim. P. 24 (qualifications and compensation
for trial and appellate counsel in capital cases). The post-
conviction lawyers (the fifth team) no doubt investigated this
claim as thoroughly as possible. But when the time came to
offer actual evidence about the results of the investigation,
they simply did not have evidence that the dissent says
should have been “easily within reach.” We cannot grant re-
lief by filling in the gaps with our own speculation that fur-
ther investigation would have been sufficiently helpful to
Kubsch’s defense.
IV. Waiver of Counsel at the Penalty Phase
We turn now to Kubsch’s third principal claim on appeal.
At the penalty phase of the trial, Kubsch waived his right to
counsel and represented himself. He chose not to present
any mitigating evidence. He did make a statement to the ju-
ry in which he said the murders were a “horrific nightmare”
for which the death penalty would be appropriate, but he
also continued to assert his innocence. On direct appeal and
federal habeas review—though not in the intervening state
post-conviction proceeding—he has argued that his waiver
of counsel was not sufficiently knowing and intelligent be-
cause he was not “made aware of the dangers and disad-
vantages of self-representation.” See Faretta v. California, 422
U.S. 806, 835 (1975).
No. 14-1898 51

The Indiana Supreme Court considered and rejected the
claim. Kubsch II, 866 N.E.2d at 735–38. That decision was not
an unreasonable application of federal law under the cir-
cumstances of this case. See 28 U.S.C. § 2254(d)(1). Kubsch
made clear that he was waiving counsel because he did not
want to present evidence at the sentencing phase of the trial.
That decision simplified substantially the challenge of repre-
senting himself, so the trial judge’s colloquy was sufficient
under the circumstances. Neither Faretta nor any other Su-
preme Court decision required the judge to discourage
Kubsch from making his decision to waive counsel.
A. The Constitutional Standard
We first address the constitutional standard before turn-
ing to its application in this case. Faretta established that “a
defendant in a state criminal trial has a constitutional right
to proceed without counsel when he voluntarily and intelli-
gently elects to do so.” 422 U.S. at 807. Though “he may
conduct his own defense ultimately to his own detriment,
his choice must be honored out of ‘that respect for the indi-
vidual which is the lifeblood of the law.’” Id. at 834, quoting
Illinois v. Allen, 397 U.S. 337, 350–351 (1970) (Brennan, J., con-
curring). Faretta also cautioned that when “an accused man-
ages his own defense” he forgoes “many of the traditional
benefits associated with the right to counsel.” Id. at 835. Re-
spect for the value of these “relinquished benefits” is why
“the accused must knowingly and intelligently” waive the
right to counsel. Id. (internal quotation marks omitted), cit-
ing Johnson v. Zerbst, 304 U.S. 458, 464–65 (1938).
“The determination of whether there has been an intelli-
gent waiver of right to counsel must depend, in each case,
upon the particular facts and circumstances surrounding
52 No. 14-1898

that case, including the background, experience, and con-
duct of the accused.” Johnson, 304 U.S. at 464. Two other rel-
evant “case-specific factors” are “the complex or easily
grasped nature of the charge” and “the stage of the proceed-
ing.” Iowa v. Tovar, 541 U.S. 77, 88 (2004). To determine
whether a defendant has knowingly and intelligently
waived the right to counsel, “a judge must investigate as
long and as thoroughly as the circumstances of the case be-
fore him demand.” Von Moltke v. Gillies, 332 U.S. 708, 723–24
(1948).
Both the Indiana Supreme Court and this circuit consider
four factors in the waiver inquiry: “(1) the extent of the
court’s inquiry into the defendant’s decision, (2) other evi-
dence in the record that establishes whether the defendant
understood the dangers and disadvantages of self-
representation, (3) the background and experience of the de-
fendant, and (4) the context of the defendant’s decision to
proceed pro se.” Kubsch II, 866 N.E.2d at 736, quoting Poynter
v. State, 749 N.E.2d 1122, 1127–28 (Ind. 2001), quoting in turn
United States v. Hoskins, 243 F.3d 407, 410 (7th Cir. 2001).
The constitutional standard is flexible, and its application
must be adapted to the case. The Supreme Court has not
prescribed a list of admonitions that must be given to all de-
fendants who want to waive counsel. See Tovar, 541 U.S. at
92 (reversing state court’s finding that waiver was invalid:
“In prescribing scripted admonitions and holding them nec-
essary in every guilty plea instance … the Iowa high court
overlooked our observations that the information a defend-
ant must have to waive counsel intelligently will depend, in
each case, upon the particular facts and circumstances sur-
rounding that case.”) (citation and internal quotation marks
No. 14-1898 53

omitted); see also United States v. Moya-Gomez, 860 F.2d 706,
733 (7th Cir. 1988) (“Although we stress the need for a thor-
ough and formal inquiry as a matter of prudence and as a
means of deterring unfounded claims on appeal, we shall
not reverse the district court where the record as a whole
demonstrates that the defendant knowingly and intelligently
waived his right to counsel.”); United States v. Egwaoje, 335
F.3d 579, 585 (7th Cir. 2003) (reaffirming this holding of Mo-
ya-Gomez). The extent and formality of the waiver colloquy
are relevant, but it is the waiver itself, not the waiver collo-
quy, that is the proper focus of the inquiry.
This constitutional standard does not impose a separate
duty to discourage a defendant from representing himself. If
a defendant is not already “aware of the dangers and disad-
vantages of self-representation,” then the trial court must
educate him so that he is aware of those risks when he de-
cides. Faretta, 422 U.S. at 835. When a defendant wants to
take on the challenges of representing himself at trial, in-
cluding dealing with jury selection, presentation of evidence,
and jury instructions, the judge may and usually will try to
discourage that option as a means of forcing the defendant
to think carefully about unfamiliar risks.
We find no Supreme Court decision, however, requiring
a judge to discourage self-representation in all circumstanc-
es. If a judge believes, as the trial judge did here, that the de-
fendant is making a knowing and intelligent waiver, then
she would commit constitutional error by discouraging that
decision too strongly. Faretta clearly established the constitu-
tional right to self-representation. “That right is not honored
if judges must depict self-representation in such unremit-
tingly scary terms that any reasonable person would refuse.”
54 No. 14-1898

United States v. Oreye, 263 F.3d 669, 672 (7th Cir. 2001), quot-
ing United States v. Hill, 252 F.3d 919, 928–29 (7th Cir. 2001).
When a defendant raises the possibility of representing
himself, the trial court is placed “between the Scylla of
trammeling the defendant’s constitutional right to present
his own defense and the Charybdis of shirking its ‘constitu-
tional duty to ensure that the defendant only represents
himself with full awareness that the exercise of that right is
fraught with dangers.’” United States v. Sandles, 23 F.3d 1121,
1127 (7th Cir. 1994) (citation omitted), quoting Moya-Gomez,
860 F.2d at 732. Appellate courts have tried to keep the per-
missible middle ground between these opposing errors fairly
broad, allowing trial judges reasonable leeway to adapt the
inquiry to the circumstances of the case without requiring a
script or checklist. Trial judges seeking this middle way are
not constitutionally bound to discourage every defendant
from representing himself no matter the facts and circum-
stances of the case.
B. Kubsch’s Waiver of Counsel
With this constitutional standard in mind, we turn to the
facts of Kubsch’s waiver of his right to counsel for the sen-
tencing phase of his trial. The attorneys who represented
Kubsch at the guilt phase of his trial were a veteran team
who qualified as a capital defense team under Indiana Rule
of Criminal Procedure 24, which sets minimum qualifica-
tions for lead and co-counsel in capital cases. During the sen-
tencing phase they served as Kubsch’s legal advisors by
court appointment. Kubsch could ask them for advice, but
they could no longer speak for him in court.
No. 14-1898 55

Kubsch represented himself at the sentencing phase of
his trial because he did not want to present mitigating evi-
dence. Kubsch was advised by the court and counsel that if
his counsel had represented him in the sentencing phase, his
counsel would have made the final decision about which
witnesses to call. His attorneys planned to offer mitigating
evidence, and they named the witnesses they would have
called and provided Kubsch a written summary of that evi-
dence. The court asked Kubsch whether he wanted any of
those witnesses to be called. Kubsch confirmed that he did
not.
The court then told Kubsch what to expect in the sentenc-
ing phase of the trial if, as both sides planned, he and the
State presented no new evidence. Each side would address
the jury, and the court would instruct the jury on the appli-
cable sentencing law, including relevant aggravating and
mitigating factors. The court told Kubsch that as his own at-
torney he would have the right to address the jury directly.
Finally, the court considered the standard advice and
warnings given to defendants deciding whether to represent
themselves. The court noted that nearly all the advice and
warnings concern the challenges of trial, such as selecting
jurors and presenting evidence, which can be difficult with-
out legal training and experience. The court pointed out that
if the sentencing phase did not include additional evidence,
the most difficult obstacles for a pro se defendant would not
be present. The court then reiterated that Kubsch had the
right to make a statement to the jury and allowed his attor-
neys to withdraw their appearances.
Kubsch now argues that his waiver was not knowing and
intelligent because the court’s colloquy was insufficient and
56 No. 14-1898

because the judge did not attempt to discourage his choice.
The Indiana Supreme Court considered these arguments in
detail and in light of the circumstances of this case, particu-
larly Kubsch’s reasons for wanting to represent himself and
the stage of the proceeding, where he would only make a
statement to the jury about the appropriate penalty. Kubsch
II, 866 N.E.2d at 735–38.
The Indiana Supreme Court noted that Kubsch himself
“eliminated the need” for almost all of the standard advise-
ments given to defendants deciding whether to represent
themselves by confirming that he did not wish to present ev-
idence at the sentencing phase of his trial. Id. at 736. Accord-
ingly, the waiver colloquy was “sufficient to apprise the de-
fendant of the dangers he is facing in the particular matter at
hand.” See id. All that remained in the trial, as a practical
matter, was a closing argument on whether the death penal-
ty should be imposed.
The stakes were as high as they come in a trial, but they
were highest for the man who wanted to speak for himself.
The Faretta right of self-representation is founded upon re-
spect for the autonomy of the defendant:
The right to defend is personal. The defendant,
and not his lawyer or the State, will bear the
personal consequences of a conviction. It is the
defendant, therefore, who must be free person-
ally to decide whether in his particular case
counsel is to his advantage. And although he
may conduct his own defense ultimately to his
own detriment, his choice must be honored out
of “that respect for the individual which is the
lifeblood of the law.”
No. 14-1898 57

422 U.S. at 834.
The state court also noted the trial judge’s observation
about Kubsch’s competence at the end of this three-week tri-
al:
I want to state for the record, in this case, that
the Court observed Mr. Kubsch throughout tri-
al, that during trial he pretty much constantly
was able to confer with his attorneys, was able
to confer with his factual investigator that in-
terviewed witnesses in this case, that he testi-
fied in this case, that the Court found his testi-
mony to be coherent and relevant to the facts
of this case, and that the Court has no reason to
doubt Mr. Kubsch’s competency to represent
himself in this matter.
Tr. 3339–40, quoted in Kubsch II, 866 N.E.2d at 737. The state
court quoted this observation to help show that Kubsch was
capable of understanding, and did in fact understand, the
decision he was making. It also pointed out that “at the time
he chose to represent himself, Kubsch had already partici-
pated in two murder trials and one penalty phase.” Kubsch
II, 866 N.E.2d at 738. “In other words, he obviously knew
from his own experience of his right to call witnesses, pre-
sent other evidence, and propose mitigating factors.” Id.
Finally, the Indiana Supreme Court viewed Kubsch’s de-
cision to waive counsel as knowing because it was strategic,
intended to prevent his counsel from calling witnesses in the
penalty phase of the trial. Id., citing United States v. Todd, 424
F.3d 525, 533 (7th Cir. 2005). “Choosing to waive counsel be-
cause one does not agree with trial strategy is perhaps not
58 No. 14-1898

the best choice, or even a good choice, but it can be a rational
choice.” 866 N.E.2d at 738.
Citing John H. Blume, Killing the Willing: “Volunteers,” Su-
icide and Competency, 103 Mich. L. Rev. 939 (2005), Kubsch
argues now that his decision was not so much strategic as
suicidal, calculated to bring about his own execution and in-
dicating “a pre-existing mental illness.” That is indeed one
way to understand Kubsch’s behavior. Another way to un-
derstand Kubsch’s behavior, however, is to take at face value
his words at the sentencing phase of both trials. At both he
articulated a principled opposition to arguing that any miti-
gating evidence could outweigh the aggravating circum-
stances of the crimes a jury had convicted him of commit-
ting. Faretta was decided precisely to protect such principled
decisions. Kubsch now apparently regrets his decision to
proceed pro se. That does not mean his decision was any less
principled when he made it or that it was the product of
mental illness.
His strategy can also be understood in quite sensible
terms. Rather than begging for mercy from the jury that had
just convicted him of three brutal murders without any ap-
parent mitigating circumstances, Kubsch told the jury, “I
wouldn’t even dare try to insult your intelligence by wasting
your time by presenting mitigation.” Tr. 3372. He instead as-
serted several times that he is innocent. His approach can be
understood as a reminder that the jurors should consider the
possibility that they might have made a mistake, so that re-
sidual doubt should weigh against the death penalty. That
approach is entirely consistent with his defense at trial, even
though neither was successful. The state courts did not act
unreasonably in viewing the waiver as strategic and know-
No. 14-1898 59

ing. See United States v. Davis, 285 F.3d 378, 384–85 (5th Cir.
2002) (defendant chose to represent himself at sentencing
phase of capital trial for similar strategic reason; appellate
court issued writ of mandamus barring district court’s ap-
pointment of independent counsel to present mitigating evi-
dence over defendant’s objection).
Kubsch argues most strenuously that the trial judge had
a duty under “the spirit of Faretta” to discourage him from
waiving his right to counsel. That is not what Faretta said or
means. Faretta held that a defendant has a constitutional
right to waive counsel as long as the waiver is knowing, vol-
untary, and intelligent. The core of Faretta is respect for the
defendant’s autonomy even if he makes a foolish decision.
422 U.S. at 834; see also Davis, 285 F.3d at 384. There is no re-
quirement to discourage the defendant. As noted, we have
warned that excessive discouragement, even for a defendant
who wishes to handle the entire case, can violate Faretta. See
Hill, 252 F.3d at 929 (“A defendant bullied or frightened into
acquiescing in a lawyer that he would rather do without
would be in a much better position to say that the choice was
not made knowingly or intelligently.”).
The basic problem with Kubsch’s argument is that most
of the specific advice usually given to defendants was un-
necessary for him. He planned to present no mitigating evi-
dence and planned only to make a brief statement to the ju-
ry. Cf. Federal Judicial Center, Benchbook for U.S. District
Court Judges § 1.02 (6th ed.) (warnings focus on procedural
and evidentiary challenges before and during trial).
Kubsch responds that this view “shifts responsibility
from the trial court to the defendant, making the defendant
responsible to inform the court how he wished to proceed, to
60 No. 14-1898

determine the level of warning the court must give him.”
The Indiana Supreme Court did not make that mistake.
Kubsch’s counsel and then Kubsch himself explained his
plans to the trial judge. The judge was not required to ques-
tion Kubsch’s strategy, and he did not require Kubsch to
provide information. Kubsch volunteered it. The trial judge
adapted his approach to the waiver inquiry accordingly.
In a variation on this argument, Kubsch also argues that
the waiver colloquy was actually misleading. At one point,
the trial judge said, “In a way I’m saying, your representa-
tion would not be as complicated as if you were handling the
whole trial by yourself. Do you understand that?” Tr. 3342.
Taken in context, this statement was not misleading at all. It
was true. Making a statement to the jury was far simpler for
Kubsch than representing himself in the guilt phase of his
trial would have been. See Tovar, 541 U.S. at 88 (explaining
that the “information a defendant must possess in order to
make an intelligent” waiver depends in part on “the stage of
the proceeding”).
In sum, the federal Constitution required the trial judge
to determine whether Kubsch’s waiver of counsel for the last
phase of his trial was knowing, voluntary, and intelligent.
The Indiana Supreme Court did not apply that clearly estab-
lished federal law unreasonably by holding that Kubsch’s
waiver was valid in light of “the particular facts and circum-
stances surrounding that case, including the background,
experience, and conduct of the accused,” see Johnson, 304
U.S. at 464, and the stage of the proceeding, see United States
v. Hoskins, 243 F.3d at 410.
Accordingly, we AFFIRM the district court’s judgment
denying relief.
No. 14‐1898  61   

WOOD,  Chief  Judge,  dissenting.  My  colleagues  are  pre‐
pared  to  send  Wayne  Kubsch  to his  death  on  the  basis  of  a 
trial  at  which  the  jury  never  heard  critical  evidence  that,  if 
believed,  would  have  shown  that  Kubsch  was  not  the  man 
responsible  for  the  horrible  murders  of  his  wife  Beth,  her 
son,  Aaron Milewski, and her ex‐husband, Rick  Milewski. I 
am not. They concede that the evidence against Kubsch was 
entirely  circumstantial.  While  there  is  nothing  wrong  with 
circumstantial  evidence,  it  is  impossible  to  have  any  confi‐
dence in a verdict rendered by a jury that heard only part of 
the story. In my view, the state courts have reached a result 
that is inconsistent with, and an unreasonable application of, 
the  United  States  Supreme  Court’s  decision  in  Chambers  v. 
Mississippi,  410  U.S.  284  (1973).  Had  the  contested  evidence 
been  admitted  under  the  Chambers  exception  to  the  normal 
rules  of  evidence,  a  properly  instructed  jury  may  have  ac‐
quitted Kubsch. It also may have convicted him: I do not ar‐
gue  that  the  state  courts  wrongly  viewed  the  evidence  as 
sufficient  for  conviction.  But  that  is  not  the  question  before 
us. The  question is  whether Kubsch  was able to  present his 
entire  case  and  obtain  a  reliable  jury  verdict.  Because  I  be‐
lieve  that  he  was  deprived  of  this  essential  protection,  I 
would  grant  the  writ  and  give  the  State  of  Indiana  a  new 
opportunity to try him. 
I 
As  required  by  the  Antiterrorism  and  Effective  Death 
Penalty Act, I rely on the facts used by the Supreme Court of 
Indiana after Kubsch’s second trial, conviction, and sentenc‐
ing.  See  Kubsch  v.  State,  866  N.E.2d  726  (Ind.  2007)  (Kubsch 
II). That opinion summarized the facts that had been devel‐
oped  in  earlier  appeals.  See  Kubsch  v.  State,  784  N.E.2d  905 
62  No. 14‐1898 

(Ind.  2003)  (Kubsch  I);  see  also  Kubsch  v.  State,  934  N.E.2d 
1138  (Ind.  2010)  (Kubsch  III)  (opinion  at  post‐conviction 
stage).  
Wayne  and  Beth  Kubsch  were  married  in  November 
1997. It was a second marriage for both: Beth had two sons, 
Aaron  Milewski,  from  her  previous  marriage  to  Rick 
Milewski,  and  Anthony  Earley;  and  Kubsch  had  a  son, 
Jonathan,  who  lived  with  his  mother,  Tina  Temple.  Aaron 
lived with Rick in South Bend, Indiana, while Anthony lived 
with Kubsch and Beth in nearby Mishawaka. Kubsch owned 
the family home, and he also owned 11 rental properties in 
St.  Joseph  County.  They  were  encumbered  by  mortgages 
totaling approximately $456,000 as of mid‐1998. Kubsch also 
had credit‐card debt exceeding $16,000. He tried paying that 
off  by  refinancing  four  of  his  rental  properties,  but  by 
August  1998  the  credit‐card  debt  had  reached  $23,000,  and 
by  September  Kubsch  was  falling  behind  in  his  mortgage 
and  tax  payments.  At  about  that  time,  he  bought  a  life 
insurance  policy  on  Beth,  with  himself  as  the  sole 
beneficiary; the policy would pay $575,000 on her death.  
The  fateful  day was September 18, 1998. For  ease  of  ref‐
erence,  I  provide a  timeline  of  the events in Appendix  A to 
this dissent. Here I summarize what happened that day and 
the  evidence  that  pins  down  where  the  key  actors  were  lo‐
cated.  I rely  on the evidence that was  admitted  at  Kubsch’s 
second trial.  
That  morning,  both  Wayne  and  Beth  Kubsch  were  up 
early.  By  6:00  a.m.,  testimony  from  Beth’s  coworker  Archie 
Fobear  established  that  Beth  had  already  left  her  home  on 
Prism  Valley  Drive  in  Mishawaka  and  was  just  starting  to 
work at United Musical Instruments in Elkhart, Indiana, ap‐
No. 14‐1898  63 

proximately 11 miles away. Cellular telephone records indi‐
cated  that  Kubsch  made  a  call  at  that  time  from  the  sector 
just  adjacent  to  the  one  covering  the  home.  He  was driving 
to  his  place  of  employment  at  Skyline  Corporation,  also  in 
Elkhart;  he  punched  in  at  6:50  a.m.  Cell  records  show  that 
Kubsch  made  a  telephone  call  at  9:11  a.m.  somewhere  near 
his  workplace,  and  that  he  made  another  call  at  10:45  a.m. 
from Skyline’s break room. The latter call was to the home, 
presumably to Beth, who had finished her shift at 10:00 a.m., 
returned  home,  and  paged  him  twice  from  home  around 
10:30 a.m.  
At  10:48  a.m.,  a  five‐minute  call  was  placed  from  the 
Kubsch  home  to  the  home  of  Rick  Milewski.  At  that  point 
Beth left the house to run some errands. A security camera at 
the Teacher’s Credit Union shows Beth, along with her dog, 
in her car at a drive‐up window at 11:08 a.m. There is a cred‐
it union receipt stamped 11:14 a.m. confirming a completed 
transaction. A little while later, at 11:52 a.m., Beth was with 
credit counselor Edith Pipke at the Consumer Credit Coun‐
seling  Agency  in  South  Bend.  No  evidence  admitted  at  the 
second trial indicated where she was after she left the credit 
union and before she arrived for her appointment.  
In the meantime, Kubsch drove back to the Prism Valley 
house  after  punching  out  from  his  job  at  11:13  a.m.  Erin 
Honold, a neighbor, saw him and his car in the driveway be‐
tween 11:30 a.m. and noon, around the same time when Beth 
was  speaking  with  the  credit  counselor.  Telephone  records 
from the house indicate that a call was made at 11:37 a.m. to 
American  General  Finance;  Kevin  Putz,  an  employee  of  the 
company,  testified  that  he  spoke  to  Kubsch  that  morning. 
Between 12:09 and 12:11 p.m., Kubsch made three more calls 
64  No. 14‐1898 

using his cellphone, one to the house (implying that he was 
no  longer  there)  and  two  to  Rick  Milewski.  He  apparently 
interrupted  Rick  while  Rick  was  speaking  with  his  brother 
Dave  about  an  upcoming  hunting  trip.  Dave  testified  that 
Rick  said  that  Kubsch  was  calling  to  discuss  moving  a  re‐
frigerator at the Prism Valley house.  
Beth  paged  Kubsch  again  at  12:16  p.m.;  cell  records 
indicate that at 12:18 p.m., he called the house for 31 seconds 
from  the  vicinity  of  Osceola,  a  town  between  Mishawaka 
and  Elkhart.  Kubsch  returned  to  Skyline,  although  he  did 
not punch back in. He made two phone calls from the break 
room,  one  at  12:40  p.m.  and  the  other  at  1:17  p.m.  Between 
those  calls,  Rick  called  Beth  at  12:46  p.m.  Kubsch  punched 
out  of  work  again,  this  time  for  the  day,  at  1:53  p.m.  A 
minute  later,  he  called  the  house  from  Elkhart  and  was  on 
the  line  for  46  seconds.  The  next  call  from  Kubsch’s 
cellphone  came  at  2:51  p.m.;  it  was  from  a  sector  near  the 
house. The state’s theory was that these last two calls bracket 
the  time  when  he  committed  the  murders—between  1:53 
and 2:51 p.m.  
There are some problems with this theory, at least if it is 
meant  to  encompass  all  three  murders,  because  there  is  no 
evidence that Aaron left school early that day. To the contra‐
ry,  witnesses  testified  that  Aaron  was  waiting  outside  Lin‐
coln Elementary School in South Bend and that Rick picked 
him up there between 2:20 and 2:35 p.m. (The school is now 
called Lincoln Primary Center; its website indicates that the 
school  day  runs  from  8:15  a.m.  to  2:20  p.m.  See  LINCOLN 
PRIMARY  CENTER,  https://www.edline.net/pages/Lincoln_
Primary_Center (last visited Aug. 10, 2015).) In any event, by 
3:15  p.m.  or  so,  Kubsch  placed  numerous  calls  to  Beth’s 
No. 14‐1898  65 

mother,  Diane  Rasor;  he  eventually  connected  on  the  11th 
try.  Cellular  records  indicate  that  he  was  heading  north  at 
that point, toward the Michigan border.  
Between  4:42  and  4:47  p.m.  Indiana  time,  Kubsch  made 
some calls picked up by the cell tower in Schoolcraft, Michi‐
gan,  which  is  about  11  miles  north  of  Three  Rivers,  Michi‐
gan,  where  Kubsch’s  son  Jonathan  lived  with  his  mother. 
(For  the  sake  of  consistency,  I  use  Indiana  time  throughout 
this  account;  in  fact,  though  most  of  Indiana  and  most  of 
Michigan  are  in  the  Eastern  time  zone,  Indiana  in  1998  had 
not yet adopted Daylight Savings Time; thus Indiana was on 
Eastern  Standard  Time  in  September  1998,  while  most  of 
Michigan,  including  Three  Rivers  and  Schoolcraft,  was  an 
hour  ahead  on  Eastern  Daylight  Time.)  Around  5:00  p.m., 
Kubsch  picked  up  Jonathan;  he  also  said  hello  to  his  friend 
Wayne  Temple  around  5:30  or  5:45  p.m.  at  the  local  Kmart 
store.  He  then  headed  back  to Osceola  with  Jonathan,  stop‐
ping  for  ten  minutes  at  the  home  of  Constance  Hardy,  the 
mother of his friend Brad. At 5:56 p.m., he made a call from 
the cellular region close to the Prism Valley house. 
By this time, however, Anthony had come home and dis‐
covered the bodies of Rick and Aaron. This happened at 5:30 
p.m.  He  immediately  summoned  help,  and  so  by  the  time 
Kubsch  showed  up  at  the  house  at  6:45  p.m.,  police  were 
there and it was taped off as a crime scene. (Beth’s body had 
not yet been discovered.) The police took Kubsch to the sta‐
tion,  interviewed  him,  and  then  released  him.  Around  9:00 
p.m.,  they  discovered  Beth’s  body  concealed  in  the  base‐
ment. They brought Kubsch back in for a second interview. 
He did not appear surprised to learn of Beth’s death. Asked 
several  times  by  the  officers  to  tell  them  what  happened, 
66  No. 14‐1898 

Kubsch chose instead to invoke his right not to speak with‐
out an attorney. The police did not arrest him for the murder 
immediately. They did so three months later, when a person 
named  Tashana  Penn  Norman  told  them  that  she  and  her 
boyfriend  overheard  a  person  saying  that  he  had  “hurt[ ]  a 
little boy,” and she identified Kubsch as the speaker. He was 
arrested  on  December  22,  1998,  and  charged  with  all  three 
murders. 
II 
A 
Kubsch  was  tried  twice  in  this  case.  The  first  trial  took 
place in 2000. At  its conclusion,  the jury  convicted him and 
recommended  the  death  penalty,  and  the  court  sentenced 
him  accordingly.  The  Supreme  Court  of  Indiana  reversed 
that  judgment  in  Kubsch  I,  and  ordered  a  new  trial.  784 
N.E.2d  at  926.  The  second  trial  took  place  in  March  2005. 
Once again, the jury found Kubsch guilty and recommended 
the  death  penalty,  and  once  again,  the  trial  court  accepted 
the  recommendation  and  imposed  that  sentence.  In  Kubsch 
II, the Supreme Court of Indiana affirmed. 866 N.E.2d at 740. 
Kubsch  then  unsuccessfully  sought  post‐conviction  relief 
from  the  state  courts,  see  Kubsch  III,  934 N.E.2d  at  1154,  be‐
fore turning to the federal court with his current habeas cor‐
pus petition, see 28 U.S.C. § 2254.  
The  State’s  case,  as  my  colleagues  readily  admit,  was 
built  from  various  pieces  of  circumstantial  evidence.  It 
pointed to  Kubsch’s financial problems and the new life in‐
surance policy on Beth as plausible motives for the murders. 
It attempted to trace his movements through use of the cellu‐
lar  telephone  records  and  the  testimony  of  the  people  who 
No. 14‐1898  67 

interacted  with  Kubsch,  Beth,  Rick,  Aaron,  and  Anthony 
throughout that day. It found a fiber on the duct tape used to 
bind  Beth’s  body  that  matched  a  fiber  taken  from  Kubsch’s 
car,  and  it  also  noted  that  the  duct  tape  wrapper  in  the  car 
matched the brand of tape used on Beth. (It offered nothing 
to  show  how  common  this  brand  was.)  It  (as  have  my  col‐
leagues)  stressed  the  fact  that  Kubsch’s  account  of  his  own 
actions  during  the  day  was  not  consistent  on  key  matters, 
such  as  whether  he  went  home  during  the  lunch  hour, 
whether  he  was  alone  there,  and  when  he  headed  up  to 
Michigan.  These inconsistences, plus what the district  court 
called  a  “slow‐moving  accumulation  of  a  glacier  of  circum‐
stantial  evidence,”  satisfied  both  the  second  jury  and  all  of 
the  reviewing  courts  so  far  that  Kubsch  was  properly  con‐
victed and sentenced. 
B 
If  the  question  before  this  court  were  simply  about  the 
sufficiency of the evidence, I would agree with everyone that 
Kubsch’s  challenge  fails.  Indeed,  it  would  be  hard  to  find 
fault  with  the  extensive  discussion  my  colleagues  have  fur‐
nished. But that is not the question. It is instead whether the 
package  of  evidence  that  was  presented  to  the  jury  was 
complete,  and  if  not,  whether  the  excluded  evidence  was 
important and reliable enough to have made a difference.  
The  critical  evidence  that  was  kept  from  the  jury  was 
videotaped  testimony  by  a  girl  named  Amanda  (“Mandy”) 
Buck,  “who,  according  to  the  defense,  would  have  testified 
that  she  saw  Aaron  after  3:30  pm  on  the  day  of  the  mur‐
ders.”  Kubsch  II,  866  N.E.2d  at  730.  Mandy,  who  was  nine 
years old at the time, was interviewed immediately after the 
murders,  on  Tuesday,  September  22,  1998.  Because  of  the 
68  No. 14‐1898 

importance of what she said, I have included a full transcript 
of  the  interview  as  Appendix  B  to  this  dissent.  The  inter‐
viewer was Detective Mark Reihl; the interview took place in 
what  appears  to  be  a  room  in  the  police  station.  Mandy’s 
mother,  Monica,  was  present  throughout  and  volunteered 
information from time to time. 
After  establishing  some  basic  information,  Detective 
Reihl confirmed that Mandy was a fourth‐grader at Lincoln 
School, that she lived right across the street from Aaron and 
his  dad  Rick,  and  that  she  and  Aaron  were  “best  friends.” 
She  commented  that  Aaron  didn’t  like  Kubsch,  because  he 
would  get  rough  and  punch  too  hard  “and  stuff  like  that.” 
She saw Aaron frequently: “I always went over to his house. 
He always came over to my house and like we like used to 
study  for  the  same  spelling  words.  …  And  we  would  help 
each  other  on  homework  and  stuff.”  When  Reihl  asked  her 
when they got out of school, she replied “two twenty.” She 
lived  close  to  the  school,  she  said,  just  a  five‐minute  walk 
away.  
The  interview  then  turned  to  “last  Friday,”  which  was 
September 18, the day of the murders. On that day, as usual, 
Mandy was picked up from school by the Alphabet Acade‐
my;  from  there,  her  mother  typically  (and  that  day)  picked 
her  up  to  go  home  “[b]etween  three  thirty  and  quarter  to 
four.”  At  that  point  Monica  interjected  that  she  “waited  for 
[Monica’s]  mom  and  dad  to  get  home,  and  I  went  and 
cashed my check and came home.” Reihl then asked whether 
Monica noticed if Rick was across the street. Monica replied 
“I  didn’t  pay  no  attention.  All  I  saw  was  Aaron.”  Reihl  re‐
peated “You saw Aaron?,” and Monica said “[m]mm hmm.” 
She  did  not  remember  if  Rick’s  truck  was  there.  Turning 
No. 14‐1898  69 

back  to  Mandy,  Reihl  asked  again  what  time  she  got  home 
that day. Monica answered instead, repeating “3:30 or quar‐
ter to four.” Mandy confirmed that she saw Aaron then, and 
that  she  also  saw  “his  dad,”  who  “was  coming  from  their 
living room into the kitchen to get something to drink.” She 
explained that she was able to see this from her own house: 
“every  day  when  I  walk  home  I  always  see  Rick  walk  into 
the kitchen or walk into the restroom or walk into his room.” 
Asked  what  kind  of  car  Rick  drove,  Mandy  replied  “[a] 
Chevy? He used to drive a Chevy until it broke down.” She 
specified  that  it  was  a  black,  medium‐sized,  “kinda  short” 
truck. Because his truck had broken down, she added that he 
was  driving  a  white  truck  that  he  had  borrowed  from  his 
brother on Friday, and that the white truck was at the house 
when she got home from school.  
Reihl next asked whether she saw Rick and Aaron leave 
that  afternoon.  She  answered,  “Um,  yeah, like  I  was on  my 
porch and, and they let me blow bubbles and I was blowin’ 
my bubbles, and I seen Rick pull out and leave.” She was not 
sure  what  time  that  was,  because  she  left  her  watch  in  her 
gym  bag,  but  she  estimated  it  was  a  “medium”  time  after 
she  got  home,  and  she  commented  that  “it  takes  a  pretty 
long time to get to [Aaron’s] mom’s house.”  
She  then  went  into  some  detail  about  Aaron’s  plans  for 
the weekend. “He said that he was going to his mom’s house 
Friday, ‘cause he was gonna stay the night there to go to the 
field trip Saturday. … You know he was, he—he wanted to 
go  on  the  field  trip  bad.  …  But  by  the  time  Saturday  when 
we, when we were on the bus and stuff, he was gonna be in 
our  group,  and,  um,  he  never  showed  up.  He  wasn’t  there. 
And we didn’t know why.” She went camping after the field 
70  No. 14‐1898 

trip and told her grandmother that she had not seen Aaron. 
She  learned  about  the  murders  after  a  news  crew  came  to 
her  home  while  she  was  at  her  karate  lesson  the  following 
Monday, she said.  
Reihl then turned back to Monica and confirmed that she 
cashed her paycheck on Friday, shortly after she came home 
from  work  (around  3:50  p.m.).  She  said  again  that  she  had 
seen Aaron, but not Rick, and that she did not look to see if 
Rick’s  truck  was  there.  They  discussed  what  kind  of  truck 
Rick  drove;  interestingly,  Mandy  knew  more  about  it  than 
her  mother—she  liked  the  gold  printing  that  said  “Chevro‐
let”  across  the  back.  By  then,  the  interview  was  winding 
down.  Reihl  asked  Mandy  yet  again  whether  she  saw  both 
Aaron and his father, as well as the white truck, in the yard 
around 3:30 or 3:45 p.m., and she said yes. He asked whether 
“[t]hese times that you’ve given me today, uh, these are pret‐
ty accurate,” and Monica said, “Yeah, ‘cause I get off work at 
quarter  after  three.”  This  was  her  daily  routine.  With  that, 
the interview ended. 
A  few  days  after  Mandy’s  interview,  Reihl  called  Mon‐
ica’s place of employment and then her home, apparently in 
an attempt to see yet again whether both Mandy and Monica 
had  correctly  recounted  what  happened  and  when  it  hap‐
pened.  Reihl  spoke  to  Mandy’s  grandfather  (“Lonnie”)  and 
asked  him  to  find  out  if  Mandy  and  Monica  were  certain 
about their story. Lonnie called Reihl back and told him that 
the events that Mandy and Monica had described had taken 
place on Thursday, September 17, not on Friday. The prose‐
cutors recounted at Kubsch’s trial that Monica told the police 
that  “her  father was  at  her  house on  that  Thursday, and he 
later reminded her that it was Thursday instead of Friday.” 
No. 14‐1898  71 

She said that she—Monica—had confused the dates because 
she  was  so  busy;  she  offered  no  reason  why  Mandy  would 
have  confused  them.  Nor  was  there  any  effort  to  explain 
away  Mandy’s  detailed  comments  about  the  timing  of  the 
Saturday field trip and her subsequent camping trip, karate 
lesson,  and  so  on.  At  that  early  time,  not  a  week  after  the 
field trip, it would have been easy to confirm with the school 
whether  the  trip  took  place  on  Saturday,  September  19,  or 
Friday,  September  18.  (And  even  the  trial  evidence  shows 
Rick picking up Aaron at school between 2:20 and 2:35 p.m. 
on  Friday,  strongly  suggesting  that  there  was  no  field  trip 
that day.) In addition, it would have been relatively easy to 
confirm  when  Monica  was  paid  and  made  her  deposit,  just 
as evidence had shown when Beth visited her own bank. 
Mandy  was  called  to  testify  at  the  second  trial,  but  she 
had almost nothing to say. She claimed to have no memory 
of talking to the police or being interviewed by them in 1998. 
When Kubsch’s lawyer attempted to use the transcript of the 
interview  to  refresh  her  recollection  and  later  to  impeach 
her,  the  prosecution  objected  and  the  court  sustained  the 
objections.  The  court  also  refused  to  permit  the  use  of  the 
videotaped  interview  as  a  recorded  recollection,  despite 
Mandy’s  asserted  inability  to  recall  anything  about  the 
interview. 
C 
The  Supreme  Court  of  Indiana  upheld  the  trial  court’s 
rulings.  It  found  that  the  videotape  was  not  admissible  un‐
der Indiana’s evidentiary rule governing the use of recorded 
recollection, Ind. R. Evid. 803(5). In 2005 that rule covered: 
72  No. 14‐1898 

[a]  memorandum  or  record  concerning  a  mat‐
ter about which a witness once had knowledge 
but  now  has  insufficient  recollection  to  enable 
the  witness  to  testify  fully  and  accurately, 
shown  to  have  been  made  or  adopted  by  the 
witness  when  the  matter  was  fresh  in  the  wit‐
ness’s  memory  and  to  reflect  that  knowledge 
correctly … . 
(It essentially tracks Fed. R. Evid. 803(5), as it read before the 
2011  restyling  changes  were  made.)  The  court  was  con‐
cerned  about  the  final  element,  which  requires  that  the  re‐
cording  reflect  the  witness’s  knowledge  correctly.  It  found 
that Mandy’s inability to vouch for the accuracy of her prior 
statement precluded its use. The videotape was not admissi‐
ble  as  a  prior  inconsistent  statement,  the  court  added,  be‐
cause Mandy gave no substantive evidence at all in her tes‐
timony, and so there was (almost) no prior statement to im‐
peach.  
The  court  conceded,  however,  that  there  was  one  state‐
ment  that  was  subject  to  impeachment.  At  the  trial,  Mandy 
stated  that  “I  probably  didn’t  see  [Aaron],  because  I  go 
straight  [from]  home  to  the  day  care,  and  then  I  would  go 
home  afterwards.”  That  statement  directly  contradicts  her 
statement in the video that she saw Rick and Aaron that af‐
ternoon  from  her  porch,  and  the  court  acknowledged  that 
“Kubsch  should  have  been  allowed  to  impeach  her  on  this 
matter.” 866 N.E.2d at 735. It found the error harmless, how‐
ever, because it thought that Mandy’s account from the vid‐
eotape  would  have  been  impeached  by  the  call  from  her 
grandfather  suggesting  a  mistake  in  dates.  It  thought  that 
the prosecutor’s ability to put Detective Reihl and Monica on 
No. 14‐1898  73 

the stand, presumably to support the “mistake” theory, was 
“also the reason why Kubsch’s claim that he was denied his 
federal  constitutional  right  to  present  a  defense  fails.  See 
Chambers  v.  Mississippi,  410  U.S.  284  (1973)  (protecting  de‐
fendant’s  due  process  right  by  recognizing  an  exception  to 
application  of  evidence  rules  where  evidence  found  to  be 
trustworthy).”  866  N.E.2d  at  735  n.7.  At  a  minimum,  this 
passage conclusively shows that the Chambers argument was 
adequately presented to the state courts. 
Putting  to  one  side  for  the  moment  the  niceties  of  the 
rules of evidence, one thing is clear: if Mandy was correct in 
her videotaped interview that the events she was describing 
had  happened  on  Friday,  not  on  Thursday,  and  if  she  had 
seen  both  Aaron  and  Rick  as  late  as  3:45  or  4:00  p.m.  that 
day, then Wayne Kubsch could not have killed them. By that 
time,  he  was  headed  to  Michigan  to  pick  up  Jonathan.  The 
state has always pegged the time of the murders to midday, 
from  1:53  to  2:51  p.m.  It  has  never  argued  that  Kubsch  ar‐
ranged  for  someone  else  to  commit  the  murders  on  his  be‐
half, and it is obviously too late in the day to introduce such 
a  radically  different  theory.  And,  because  the  state’s  theory 
is that Kubsch killed Aaron and Rick because they stumbled 
on  him  as  he  was  murdering  Beth,  Mandy’s  testimony  un‐
dermines the conviction as it relates to Beth, too. 
No  evidence  could  be  more  critical  to  Kubsch’s  defense. 
And  the  possibility  that  the  state  might  have  been  able  to 
impeach  the  videotaped  account  cannot  cure  this  problem; 
that  impeachment  was  itself  subject  to  impeachment  from 
such details as the school’s records about the day of the field 
trip and the date when Monica cashed her paycheck. Under 
these circumstances, the Supreme Court’s decision in Cham‐
74  No. 14‐1898 

bers  overrides  the  state  evidentiary  rule  that  prevented  the 
jury  from  hearing  Mandy’s  statement.  This  was  evidence 
that, if believed, might have prompted the jury to acquit on 
one  or  more  of  the  counts.  As  I  explain  below,  the  Indiana 
Supreme  Court’s  decision  to  the  contrary  was,  in  my  view, 
contrary  to  and  an  unreasonable  application  of  Chambers, 
even under the strict standard of review that applies, which 
my  colleagues  discuss  in  such  detail  despite  our  agreement 
on that point. 
III 
Habeas corpus petitioners come to a federal court of ap‐
peals  with  at  least  two  strikes  against  them:  they  already 
have lost in the state courts (either on the merits or because 
of  one  of  many  procedural  hurdles  that  must  be  cleared); 
and  they  also  have  failed  to  convince  the  federal  district 
court  of  their  entitlement  to  relief.  They  face  the  daunting 
burden  of  satisfying  the  familiar  and  deliberately  demand‐
ing  standards  created  in  the  Antiterrorism  and  Effective 
Death  Penalty  Act  (AEDPA),  28  U.S.C.  §  2254(d),  under 
which  
An  application  for  a  writ  of  habeas  corpus  on 
behalf  of  a  person  in  custody  pursuant  to  the 
judgment  of  a  State  court  shall  not  be  granted 
with respect to any claim that was adjudicated 
on the merits in State court proceedings unless 
the adjudication of the claim— 
(1)  resulted  in  a  decision  that  was  contrary  to, 
or  involved  an  unreasonable  application  of, 
clearly  established  Federal  law,  as  determined 
by the Supreme Court of the United States; or 
No. 14‐1898  75 

(2) resulted in a decision that was based on an 
unreasonable determination of the facts in light 
of  the  evidence  presented  in  the  State  court 
proceeding. 
28 U.S.C. § 2254(d); see Harrington v. Richter, 562 U.S. 86, 102 
(2011) (“If this standard is difficult to meet, that is because it 
was meant to be.”).  
Kubsch therefore has the burden of showing that the last 
court in Indiana to speak to his case, see Ylst v. Nunnemaker, 
501 U.S. 797, 801 (1991), rendered a decision that was either 
contrary to, or an unreasonable application of, “clearly estab‐
lished  Federal  law, as determined  by the  Supreme Court  of 
the United States.” 28 U.S.C. § 2254(d)(1). (He has not sought 
to rely on 28 U.S.C. § 2254(d)(2), which deals with unreason‐
able determinations of fact, and so I do not discuss that op‐
tion here.)  As  we  observed in Lindh v. Murphy, 96  F.3d 856, 
873 (7th Cir. 1996) (en banc), reversed on other grounds, 521 
U.S.  320  (1997),  Congress  deliberately  restricted  the  juris‐
prudence to which a court faced with a habeas corpus peti‐
tion  may  resort:  only  federal  law  as  determined  by  the  Su‐
preme Court is available. This restriction acknowledges that 
the state supreme courts are equally responsible (along with 
the  lower  federal  courts)  for  applying  federal  law,  and  that 
the only federal court whose rulings bind them is the federal 
Supreme Court.  
With  that  in  mind,  I  turn  directly  to  the  Supreme  Court 
decision that controls Kubsch’s case: Chambers v. Mississippi. 
Chambers  and  the  line  of  cases  that  follow  it  “clearly  estab‐
lish” (to use AEDPA’s term) the fact that a state rule of evi‐
dence cannot be used in a way that denies an accused person 
his  right  under  the  Due  Process  Clause  to  a  fair  trial,  in 
76  No. 14‐1898 

which he has a fair opportunity to defend. My detailed look 
at  that  case  and  those  that  followed  it  demonstrates  why, 
contrary to the spin my colleagues have tried to place on it, 
the  position  I  take  is  not  opening  up  any  floodgate  for  the 
use  of  hearsay  evidence.  Only  evidence  that  satisfies  the 
strict criteria of Chambers will be admissible, and to see what 
that evidence must be like, it is necessary to recall the partic‐
ulars of the case. 
Petitioner  Leon  Chambers  was  tried  by  a  jury  in  Missis‐
sippi state court and found guilty of murdering a policeman; 
he was sentenced to life imprisonment. The story leading up 
to his conviction was sadly familiar. On a Saturday evening, 
Woodville  (Mississippi)  police  officers  Forman  and  Liberty 
went to a local bar to execute an arrest warrant for a young 
man  named  Jackson.  With  the  help  of  a  hostile  crowd  and 
some 20 to 25 men, Jackson resisted arrest. Forman then ra‐
dioed  for  assistance,  while  Liberty  retrieved  his  riot  gun 
from  the  squad  car.  Three  deputy  sheriffs  soon  arrived  in 
response to Forman’s call, but the situation was still not un‐
der  control.  Shooting  broke  out  while  Forman  was  looking 
away, but when he turned to check  on  Liberty, he saw that 
Liberty had been hit several times in the back. Before Liberty 
died,  he  turned  and  fired  toward  the  place where  the shots 
had  come  from.  His  second  shot  hit  a  man  in  the  crowd  in 
the back of the head and neck; the injured man turned out to 
be Chambers. 
Forman  saw  neither  who  shot  Liberty,  nor  whether  Lib‐
erty  managed  to  hit  anyone.  A  deputy  sheriff  later  testified 
that  he  saw  Chambers  shoot  Liberty,  and  another  deputy 
sheriff  testified  that  he  saw  Chambers  make  a  suspicious 
arm  movement  shortly  before  the  shots  were  fired.  At  the 
No. 14‐1898  77 

time, however, the remaining officers were trying to tend to 
Liberty. They put him in the police car and rushed him to a 
hospital,  but he  was  declared dead on arrival.  Chambers in 
the  meantime  was  lying  on  the  ground.  Returning  to  the 
scene,  some  of  his  friends  discovered  that  he  was  still  alive 
and took him to the same hospital, where he was treated and 
then arrested. Later he was charged with Liberty’s murder. 
Another  man,  Gable  McDonald,  was  also  in  the  rowdy 
group at the bar. A few days later, he left his wife in Wood‐
ville  and  moved  to  Louisiana,  where  he  found  work.  Five 
months  later,  he  returned  to  Woodville  to  see  an  acquaint‐
ance,  Reverend  Stokes.  After  talking  to  Stokes,  McDonald 
met with Chambers’s attorneys and gave them a sworn con‐
fession  that  he  was  the  one  who  shot  Liberty.  He  also  said 
that  he  had  told  a  friend,  James  Williams,  that  he  was  the 
killer.  He  admitted  that  he  used  a  nine‐shot,  .22‐caliber  re‐
volver,  which  according  to  the  autopsy  was  the  murder 
weapon.  McDonald  signed  the  confession,  surrendered  to 
the police, and was put in jail. 
A month later, at the preliminary hearing, McDonald re‐
canted. His new story was that Stokes had persuaded him to 
make  a  false  confession;  the  idea,  implausible  though  it 
sounded,  was  that  Stokes  promised  he  would  not  go  to  jail 
for  the  crime  and  that  he  would  share  in  the  proceeds  of  a 
lawsuit  Chambers  planned  to  bring  against  the  town.  The 
local  justice  of  the  peace  accepted  the  recantation  and  re‐
leased McDonald. 
Chambers’s  trial  took  place  the  next  year.  He  had  two 
theories of defense: first, he tried to show that there was no 
evidence indicating that he shot Liberty; second, he wanted 
to show that the real culprit was McDonald. He was stymied 
78  No. 14‐1898 

in the latter effort, however, by the confluence of two Missis‐
sippi  rules  of  trial  procedure.  First,  because  the  prosecutor 
refused to call McDonald as a witness, he was forced to call 
McDonald  himself.  This  triggered  Mississippi’s  voucher 
rule, under which the party who calls a witness is forbidden 
to  impeach  him.  Following  that  rule,  the  trial  court  refused 
to allow Chambers to treat McDonald as an adverse witness. 
Second,  his  effort  to  use  three  other  witnesses  to  whom 
McDonald  had  confessed  was  blocked  by  the  hearsay  rule. 
Chambers  was  prepared  to  show  that  each  of  those  three 
would testify that McDonald unequivocally said that he shot 
Liberty. Much of their testimony was corroborated.  
The Supreme Court found that the combination of these 
two rules of state procedure resulted in a fundamentally un‐
fair trial for Chambers. The rules rendered him utterly una‐
ble  to  subject  McDonald’s  repudiation  and  alibi  to  cross‐
examination,  and  they  prevented  him  from  putting  before 
the  jury  the  information  that  would  have  allowed  them  to 
decide whether to believe McDonald. The voucher rule, the 
Court  held,  “as  applied  in  this  case,  plainly  interfered  with 
Chambers’  right  to  defend  against  the  State’s  charges.”  410 
U.S. at 298. The Court found no need to decide whether that 
interference alone would have been enough, because it also 
found that when one added the effects of the hearsay rule to 
the  mix,  there  was  no  doubt  that  Chambers’s  constitutional 
rights  were  violated.  It  noted  that  the  hearsay  statements 
“were originally made and subsequently offered at trial un‐
der  circumstances  that  provided  considerable  assurance  of 
their reliability.” Id. at 300 (spontaneous, corroborated, inde‐
pendent, against McDonald’s penal interest). McDonald was 
present  in  the  courtroom,  under  oath,  and  subject  to  cross‐
examination.  The  Court  summarized  its  holding  with  these 
No. 14‐1898  79 

words:  “In  these  circumstances,  where  constitutional  rights 
directly  affecting  the  ascertainment  of  guilt  are  implicated, 
the  hearsay  rule  may  not  be  applied  mechanistically  to  de‐
feat the ends of justice.” Id. at 302.  
The  Court  did  not  abandon  Chambers  the  minute  it  was 
decided in 1973. To the contrary, as my colleagues concede, 
over  the  ensuing  years  the  Court  has  carefully  reviewed  a 
substantial  number  of  cases  in  which  Chambers  arguments 
have been made. Some decisions have found that state rules 
must  give  way  to  the  fundamental  dictates  of  due  process, 
while  others  have  concluded  either  that  the  evidence  is  not 
so  critical,  or  that  the  rule  as  applied  does  not  deprive  the 
defendant  of  a  fair  trial.  Even  in  the  latter  cases,  however, 
the  Court  has  confirmed  its  continued  adherence  to  Cham‐
bers. 
For  example,  in  Nevada  v.  Jackson,  133  S.  Ct.  1990  (2013), 
the defendant argued in a sexual assault case that a Nevada 
statute that precludes the admission of extrinsic evidence for 
impeachment purposes violated the Chambers principle. The 
Court rejected that argument and held that Nevada was enti‐
tled to apply its statute. Nevertheless, however, it said:  
[o]nly rarely have we held that the right to pre‐
sent  a  complete  defense  was  violated  by  the 
exclusion  of  defense  evidence  under  a  state 
rule of evidence. See [Holmes v. South Carolina,] 
547 U.S. [319], 331 [(2006)] (rule did not ration‐
ally serve any discernible purpose); Rock v. Ar‐
kansas,  483  U.S.  44,  61  (1987)  (rule  arbitrary); 
Chambers  v.  Mississippi,  410  U.S.  284,  302–303 
(1973)  (State  did  not  even  attempt  to  explain 
the reason for its rule); Washington v. Texas, 388 
80  No. 14‐1898 

U.S.  14,  22  (1967)  (rule  could  not  be  rationally 
defended). 
133 S. Ct. at 1992.  
Indeed,  only  three  years  before  Jackson  the  Court  found 
an  application  of  Chambers  to  be  so  uncontroversial  it  ad‐
dressed the matter in a per curiam opinion. Sears v. Upton, 561 
U.S.  945  (2010).  In  that  case,  evidence  of  petitioner  Sears’s 
cognitive impairments had not been brought to light in state 
court  during  his  capital  sentencing  hearing.  The  Court  first 
found that the state court had not applied the correct stand‐
ard  for  ascertaining  prejudice  for  purposes  of  a  Sixth 
Amendment claim of ineffective assistance of counsel. Id. at 
946.  It  then  said  that  “the  fact  that  some  of  such  evidence 
may have been ‘hearsay’ does not necessarily undermine its 
value—or its admissibility—for penalty phase purposes.” Id. 
at  950  (footnote  omitted).  In  the  accompanying  footnote,  it 
added this: “Like Georgia’s ‘necessity exception’ to its hear‐
say  rules,  …  we  have  also  recognized  that  reliable  hearsay 
evidence  that  is  relevant  to  a  capital  defendant’s  mitigation 
defense should not be excluded by rote application of a state 
hearsay rule.” Id. at 950 n.6.  
As the citation to Holmes in Jackson signals, the Court has 
not shrunk the Chambers principle to one that applies only to 
sentencing  proceedings,  in  which  the  normal  rules  of  evi‐
dence  do  not  strictly  apply.  In  Holmes,  the  question  was 
“whether a criminal defendant’s federal constitutional rights 
are violated by an evidence rule under which the defendant 
may  not  introduce  proof  of  third‐party  guilt  if  the  prosecu‐
tion  has  introduced  forensic  evidence  that,  if  believed, 
strongly supports a guilty verdict.” 547 U.S. at 321. Yes, the 
Court concluded, the defendant’s rights are violated by such 
No. 14‐1898  81 

an  evidence  rule,  despite  the  broad  latitude  that  state  and 
federal rulemakers enjoy. It continued as follows:  
Whether  rooted  directly  in  the  Due  Process 
Clause of the Fourteenth Amendment or in the 
Compulsory  Process  or  Confrontation  Clauses 
of  the  Sixth  Amendment,  the  Constitution 
guarantees  criminal  defendants  a  meaningful 
opportunity  to  present  a  complete  defense.  … 
This right is abridged by evidence rules that in‐
fring[e] upon a weighty interest of the accused 
and  are  arbitrary  or  disproportionate  to  the 
purposes they are designed to serve. 
Id. at 324 (quotation marks and citations omitted). One of the 
Court’s illustrations of this principle was Chambers. Id. at 325.  
Naturally, there are cases in which defendants have con‐
tended  that  they  should  be  entitled  to  the  benefits  of  the 
Chambers rule and the Court has turned them down. See, e.g., 
Fry v. Pliler, 551 U.S. 112 (2007) (cumulative evidence can be 
excluded); Clark v. Arizona, 548 U.S. 735 (2006) (state entitled 
to  limit  issues  for  which  evidence  of  mental  illness  and  ca‐
pacity may be used); Oregon v. Guzek, 546 U.S. 517 (2006) (no 
right to present evidence at sentencing phase that casts “re‐
sidual  doubt”  on  conviction);  United  States  v.  Scheffer,  523 
U.S. 303 (1998) (permissible to prohibit defendant in a court‐
martial  from  relying  on  polygraph  evidence).  But  it  is  no 
surprise that defendants have tried to test the outer limits of 
Chambers.  Sometimes  the  Court  has  acknowledged  the 
Chambers  rule  but  found  other  reasons  why  the  defendant 
could  not  prevail.  See  Taylor  v.  Illinois,  484  U.S.  400  (1988) 
(stressing  nevertheless  the  importance  of  ensuring  that  the 
jury  does  not  decide  based  on  a  distorted  record).  And,  in 
82  No. 14‐1898 

addition  to  the  cases  already  discussed,  there  are  others  in 
which defendants have prevailed. See, e.g., Rock v. Arkansas, 
483 U.S. 44 (1987) (refusing to allow Arkansas to use a per se 
rule excluding all hypnotically refreshed testimony); Crane v. 
Kentucky, 476 U.S. 683 (1986) (exclusion of evidence of physi‐
cal  and  psychological  circumstances  of  defendant’s  confes‐
sion  deprived  petitioner  of  fair  trial);  Green  v.  Georgia,  442 
U.S. 95 (1979)  (per curiam)  (application of  hearsay rule vio‐
lated due process even though correct as a matter of Georgia 
law).  
Chambers, in short, establishes a rule that binds state and 
federal courts alike. It ensures the fundamental fairness of a 
defendant’s trial. Its message is especially strong in our case, 
which,  like  Chambers  itself,  concerns  a  defendant’s  right  to 
demonstrate  his  innocence  on  capital  charges.  Just  as  in 
Chambers, in Kubsch’s case even though the videotaped evi‐
dence  of  Mandy’s  interview  was  technically  hearsay  (the 
very  same  rule  of  evidence  at  issue  in  both  Chambers  and 
Green), it was created in a way that provided substantial as‐
surances  of  its  accuracy.  It  missed  qualifying  for  the  “rec‐
orded recollection” exception to the hearsay rule by a hair. It 
included numerous details that were either undisputed (e.g., 
Mandy  was  a  friend  of  Aaron’s;  she  lived  across  the  street 
from him; they went to the same school) or easily subject to 
corroboration.  As  I  now  show,  these  are  precisely  the  cir‐
cumstances  in  which  the  Court  has  found  that  the  eviden‐
tiary rule must give way to the defendant’s due process right 
to a fair trial. 

No. 14‐1898  83 

IV 
A 
I begin with what may be the strongest reason for admit‐
ting  the  Mandy  videotape:  its  quality  as  a  de  facto  recorded 
recollection.  (I  say  “de  facto”  out  of  respect  for  the  Indiana 
Supreme Court’s ruling that it fell short, not because I would 
necessarily  have  come  to  the  same  conclusion.)  As  I  noted 
earlier, at the time of Kubsch’s second trial, Indiana Rule of 
Evidence 803(5) read as follows: 
The following are not excluded by the hearsay 
rule, even though the declarant is available as a 
witness: … (5) Recorded Recollection. A mem‐
orandum  or  record  concerning  a  matter  about 
which a witness once had knowledge but now 
has  insufficient  recollection  to  enable  the  wit‐
ness  to  testify  fully  and  accurately,  shown  to 
have  been  made  or  adopted  by  the  witness 
when  the  matter  was  fresh  in  the  witness’s 
memory and to reflect that knowledge correct‐
ly. 
This  rule, along  with  Indiana’s  other  rules  of  evidence,  had 
been  adopted  in  1994.  It  was  intended  to  codify  the  com‐
mon‐law  exception  to  the  prohibition  against  the  use  of 
hearsay evidence for records of past statements about which 
the  witness  has  no  present  memory.  By  requiring  only  “in‐
sufficient” recollection, the rule as adopted relaxed Indiana’s 
common‐law doctrine, which had required the complete ab‐
sence  of  any  memory  as  a  condition  of  admissibility. 
INDIANA  PROPOSED  RULES  OF  EVIDENCE  75  (1993);  see  also 
FED.  R.  EVID. 803(5) Committee Note (the model for the Indi‐
84  No. 14‐1898 

ana rule), (discussing “[t]he guarantee of trustworthiness … 
found  in  the  reliability  inherent  in  a  record  made  while 
events  were  still  fresh  in  mind  and  accurately  reflecting 
them”).  The  key  is  that  the  circumstances  surrounding  the 
preparation  of  the  record  make  it  particularly  reliable. 
INDIANA  PROPOSED  RULES  OF  EVIDENCE  75.  The  rule  itself 
does  not  specify  how  the  accuracy  of  the  recorded  version 
should be proved. The Indiana Supreme Court in Kubsch II, 
however,  took  the  position  that  the  witness  must  somehow 
vouch for its accuracy. See also 2 MCCORMICK  ON  EVIDENCE 
§ 283 (7th ed. 2013). That can be difficult, since by definition 
the  witness  does  not  recall  making  the  statement,  but  com‐
mon  practice,  conformity  with  other  things  the  witness 
knows,  or  even  a  statement  such  as  “I  would  not  have  lied 
about  that”  typically  satisfy  the  vouching  requirement.  See 
generally 30C MICHAEL  H.  GRAHAM, FEDERAL  PRACTICE  AND 
PROCEDURE § 7046 at 115–16 & n.4 (interim ed. 2011). 
In  applying  Rule  803(5),  Indiana  courts  both  before  and 
after  the  various  Kubsch  opinions  have  looked  to  see  if  the 
recorded recollection (1) relates to a matter about which the 
witness once had knowledge; (2) is one about which the wit‐
ness now has insufficient recollection to permit her to testify 
fully  and  accurately  at  trial;  (3)  is  one  that  the  witness  is 
nonetheless willing and able to adopt or vouch for; (4) is one 
made  when  the  matter  was  fresh  on  her  mind;  and  (5)  cor‐
rectly  reflects  the  witness’s  knowledge  at  the  time  of  the 
event. E.g., Impson v. State, 721 N.E.2d 1275, 1282–83 (Ind. Ct. 
App.  2000).  The  final  requirement  is  inevitably  awkward, 
because there is tension between the ability to vouch and the 
inability to recall. But Indiana courts have resolved that ten‐
sion by adopting a realistic approach to vouching; they have 
accepted even a simple statement that the report is accurate. 
No. 14‐1898  85 

E.g.,  A.R.M.  v.  State,  968  N.E.2d  820,  827  n.7  (Ind.  Ct.  App. 
2012);  see  also  Gee  v.  State,  389  N.E.2d  303,  309  (Ind.  1979) 
(“At the time  of his testimony he may have completely  for‐
gotten the event … but at that time he can vouch for the ac‐
curacy of the prior writing.”). In one case, the court was sat‐
isfied when a witness testified that she “told the truth in her 
videotaped statement.” Horton v. State, 936 N.E.2d 1277, 1283 
(Ind. Ct. App. 2010), vacated on other grounds, 949 N.E.2d 346 
(Ind.  2011).  And  at  times,  the  courts  have  simply  assumed 
that  the  report  in  question  accurately  reflects  the  witness’s 
knowledge  at the time of  the report.  See,  e.g., Small v.  State, 
736  N.E.2d  742,  745  (Ind.  2000)  (permitting  admission  of 
deposition answers because witness could not recall making 
specific  statements  in  the  deposition,  but  failing  to  address 
whether witness affirmed that she was truthful at the time of 
the deposition); Smith v. State, 719 N.E.2d 1289, 1291 (Ind. Ct. 
App.  1999)  (stating  only  that  “the  report  reflected  [the  wit‐
ness]’s knowledge correctly” without explaining why). 
It  is  easy  to  see  why  an  endorsement  from  the  witness 
would be important for many types of recorded recollection, 
such as diaries, letters, written reports, memoranda, or data 
compilations.  A  witness  might  be  able  to  authenticate  her 
signature, or her habit of writing every evening in a diary, or 
her acquaintance with the purpose and recipient of a memo‐
randum, without necessarily remembering what was said as 
a  matter  of  substance.  And  this  kind  of  vouching  serves  an 
important purpose for those kinds of records, because there 
is  nothing  otherwise to ensure  that it is this  witness’s recol‐
lections that were recorded.  
I recognize, however, that it is not up to this court to de‐
cide  whether  the  Supreme  Court  of  Indiana  correctly  inter‐
86  No. 14‐1898 

preted its own rule of evidence. This is so even though that 
court  barely  touched  on  the  reason  why  the  videotape  was 
inadmissible.  Here  is  the  entirety  of  its  explanation  for  the 
conclusion  that  the  final  element  of  Indiana’s  Rule  803(5) 
was not satisfied: 
Buck testified twice that she had no memory of 
being  interviewed  by  the  police  in  1998.  (Trial 
Tr. at 2985.) As a result, the trial court correctly 
denied Kubsch the opportunity to read Buck’s 
statement  into  evidence,  because  Buck  could 
not  vouch  for  the  accuracy  of  a  recording  that 
she could not even remember making. 
Kubsch  II,  866  N.E.2d  at  734–35.  This  merely  describes  the 
fact  that  this  was  a  matter  “about  which  [the]  witness  once 
had  knowledge  but  now  has  insufficient  recollection”  to 
permit full and accurate testimony. Indiana made clear at the 
time  it  adopted  Rule  803(5)  that  “insufficient”  recollection 
includes  no  recollection  at  all.  There  is  thus  no  reason  to 
think that the total absence of recollection precludes the use 
of the rule.  
The  Indiana  Supreme  Court  did  not  express  any  doubt 
that the other requirements of Rule 803(5) were satisfied. For 
purposes of Chambers, then, we have a situation in which the 
state hearsay rule was used to block critical evidence. There 
were, however, just as in Chambers, substantial assurances of 
reliability  of this  evidence, which I discuss below. This was 
therefore a situation in which the due process command ex‐
pressed in Chambers should have overridden the state’s evi‐
dentiary rule. 

No. 14‐1898  87 

B 
Putting Chambers temporarily to one side, the fact that the 
showing  at  trial  was  inadequate  to  satisfy  the  letter  of  Rule 
803(5) takes us to one of Kubsch’s other theories: that he re‐
ceived  ineffective  assistance  of  trial  counsel  in  a  number  of 
respects,  including  “in  their  attempt  to  admit  Amanda 
Buck’s videotaped statement.”1 Counsel failed to take any of 
a number of readily available steps to meet the requirements 
of  Rule  803(5)—steps  that  were  necessary,  under  Wiggins  v. 
Smith,  for  effective  assistance  of  counsel.  Indiana  courts  re‐
quire that the witness whose recollection has faded need on‐
ly tell the finder of fact that her statements in the recording 
were  accurate.  Kubsch’s  attorneys  never  asked  Mandy  that 
question. Instead, they dropped the subject after establishing 

1  My colleagues attempt to rehabilitate Kubsch’s lawyers in this re‐

spect,  but  they  are  forced  to  resort  to  speculation  about  what  a  proper 
investigation  would  have  revealed.  As  the  Supreme  Court  has  made 
clear, however, it is essential to evaluate the question whether counsel’s 
investigation  was  constitutionally  sufficient.  See  Wiggins  v.  Smith,  539 
U.S.  510  (2003).  There  the  Court  faced  a  case  in  which  the  petitioner’s 
claim “stem[med] from counsel’s decision to limit the scope of their in‐
vestigation into potential mitigating evidence.” Id. at 521. Quoting from 
Strickland, the Court reaffirmed that “counsel has a duty to make reason‐
able investigations or to make a reasonable decision that makes particu‐
lar  investigations  unnecessary.”  Id.  In  addition,  the  Court  squarely  rec‐
ognized that it is not enough to gather “some” information. Id. at 527. In 
language that applies with equal force to Kubsch’s case, it held that “[i]n 
assessing  the  reasonableness  of  an  attorney’s  investigation,  however,  a 
court must consider not only the quantum of evidence already known to 
counsel, but also whether the known evidence would lead a reasonable 
attorney to investigate further.” Id. Just so. Kubsch’s lawyers knew about 
Mandy’s videotaped statement, but that evidence would have led a rea‐
sonable  attorney  to  investigate  further.  Their  failure  to  take  that  step 
amounted to constitutionally ineffective assistance.  
88  No. 14‐1898 

the  fact  that  she  could  not  recall  speaking  to  the  police, 
which relates to a different requirement of the rule (one that 
was  easily  met).  They  should  have  asked  her  whether  she 
would have told the police the truth if such an interview had 
taken  place,  but  they  did  not.  They  could  have  shown  her 
the beginning of the videotape on the record—the trial tran‐
script  indicates  they  showed  Mandy  the  tape  off  the  record 
but  never  put  her  back  on  the  stand  afterward—and  asked 
her whether she was the girl depicted in the recording. They 
could have asked Monica or anyone else who knew Mandy 
well  about  her  reputation  for  truthfulness.  Any  of  these 
steps,  and  certainly  all  of  them  taken  together,  would  have 
met the requirements Indiana courts have set for compliance 
with Rule 803(5)’s requirement for evidence that shows that 
the recording reflects the witness’s knowledge correctly.  
Counsel also could have taken steps to counteract the tri‐
al court’s assumption that it would have been so easy to im‐
peach  Mandy’s  videotaped  account  that  any  error  in  refus‐
ing to allow it as a prior inconsistent statement would have 
been harmless. The state urged that this was the case based 
on  the  telephone  call  from  Mandy’s  grandfather,  Lonnie,  a 
few  days  after  the  interview  urging  the  police  to  disregard 
her statements because she was supposedly mistaken about 
the day she was talking about.  According to Lonnie,  every‐
thing  Mandy  recounted  had  happened  on  Thursday,  Sep‐
tember 17, not on Friday the 18th. But there is no reason to 
conclude,  without  any  adversarial  testing,  that  Lonnie  was 
correct. No  evidence at all  indicates how  reliable  his source 
of  information  for  that  statement  may  have  been.  He  may 
have  been  trying  to  extricate  his  granddaughter  from  in‐
volvement  in  the  murder  trial,  or  he  may  have  had  some 
other motive that no one ever explored.  
No. 14‐1898  89 

Had counsel for Kubsch been on their toes and complied 
with  their  duty  to  investigate  in  conformity  with  Wiggins, 
there are many ways in which they could have rehabilitated 
Mandy’s very clear testimony (see Appendix B) that she was 
recalling the events of Friday, just four days earlier than the 
interview.  Anyone  who  watches  the  video  can  only  be  im‐
pressed by how articulate, bright, and forthcoming Mandy is 
in  it.  If  there  were some  concern  about  the  fact  that Mandy 
was nine years old at the time, counsel could have put Man‐
dy’s mother, Monica, on the stand and asked on what day of 
the week she was paid and whether she possibly could have 
been depositing her paycheck on a Thursday. Records from 
Monica’s  bank  could  have  been  subpoenaed  to  see  when 
that deposit was made, and additional evidence such as se‐
curity  camera  footage  could  have  shown  the  day  on  which 
she  was  there.  The  school  district  could  have  been  subpoe‐
naed  for  records  confirming  on  what  day  the  field  trip  that 
Mandy  discussed  in  detail  actually  took  place.  Kubsch’s 
counsel did none of these things. 
My colleagues dismiss the video as unreliable, but saying 
so does not make it so. In fact, many factors support the reli‐
ability  of  this  video,  both  for  purposes  of  substantive  evi‐
dence and for purposes of impeachment: 
 It  was  created  only  four  days  after  the 
events about which both Mandy and Mon‐
ica were speaking. 
 Because the method of recording the  recol‐
lection  was  video,  rather  than  audio  or 
writing, there was no chance that the identi‐
ty  of  the  speakers  nor  the  content  of  their 
statements could be mistaken. 
90  No. 14‐1898 

 Mandy  provides  an  elaborate  timeline  and 
describes  small  details  from  her  direct  ob‐
servations of the victims at their home. 
 Mandy’s  mother,  Monica,  was  present 
throughout  the  interview  and  provided 
corroborating details at numerous points. 
 Neither  Mandy  nor  Monica  had  any  per‐
sonal interest in the case; there was thus no 
reason  to  fear  that  their  accounts  were 
slanted one way or the other. 
 Both  Mandy  and  Monica  were  available  at 
trial to testify after the video was shown, at 
which point the jury would have been able 
to  weigh  their  live  statements  at  trial 
against  their  recorded  statements  on  the 
video. 
The  failure  to  take  steps  that  would  have  allowed  the 
videotape  to  be  admitted  for  all  purposes  pursuant  to Indi‐
ana Rule 803(5), and that would also have permitted its use 
to  impeach  Mandy’s  statement  at  trial  that  she  “probably 
didn’t  see”  Aaron  that  afternoon,  amounted  to  insufficient 
performance  for  purposes  of  Strickland  v.  Washington,  466 
U.S. 668 (1984). It also severely prejudiced Kubsch. Mandy’s 
videotaped  testimony,  if  believed,  would  have  shown  that 
the  murders  of  at  least  Rick  and  Aaron,  and  probably  Beth 
(on  the  theory  that  Rick  and  Aaron  interrupted  the  assault 
on  Beth),  took  place  at  a  time  when  Kubsch  was  already  in 
or on his way to Michigan to pick up Jonathan. This was eas‐
ily  Kubsch’s  strongest  defense  to  the  charges,  and  it  was 
No. 14‐1898  91 

swept away by a combination of the trial court’s evidentiary 
rulings and counsel’s ineffectiveness. 
C 
The  majority  argues  that  despite  the  inherently  credible 
nature of the video and Mandy’s statements on it, there were 
three  other  primary  reasons  for  concluding  that  it  was  not 
reliable enough to meet the Chambers standard for use at tri‐
al: first, that Mandy’s statements were not corroborated; sec‐
ond,  that  she  was  “essentially  unavailable”  for  cross‐
examination; and third, that Detective Reihl “never pushed” 
Mandy on  “critical  details” during the 1998 interview, such 
as whether she had her dates and times correct. Ante at 27–
30.  I  begin  with  the  last  contention.  A  review  of  the  tran‐
script at Appendix B shows that this is simply not the case. 
The majority posits that Reihl “was simply taking [Mandy’s] 
account  as  she  spoke,”  but  Reihl  repeatedly  stops  and 
“pushes” Mandy to confirm what she is saying. He asks her 
over and over whether she is talking about Friday’s events. 
(E.g., “[D]o you remember last Friday?” “And did they pick 
you  up  Friday?”  “Was  that  white  truck  at  Rick’s  house  Fri‐
day?”  “Friday,  after  you  got  home,  they  left  just  a  little  bit 
after  when  you  got  home,  right?”)  At  the  end  of  the  inter‐
view, Reihl turns to her mother, Monica, and asks again for 
assurance:  “[t]hese  times  that  you’ve  given  me  today,  uh, 
these are pretty accurate?” Monica responds that they were, 
“pretty  well,”  because  “sometimes  I  have  to  stay  a  couple 
minutes after, so, I get home a little later. And that was just 
so happen [sic] to have been one of the days that was a little 
bit later.” It is also clear from the transcript that this was not 
the first time Monica and Mandy had spoken to Reihl about 
that  past  Friday’s  events.  At  various  points,  Reihl  indicates 
92  No. 14‐1898 

that he was following up on a conversation they had previ‐
ously “at the house.” Given these repeated assurances, there 
was  little  reason  for  Reihl  a  day  later  to  ask  the  two  inter‐
viewees  yet  again  “about  the  possibility  that  her  memory 
had confused events of two different days,” as  the majority 
suggests is necessary to meet the requirements of Chambers. 
Ante at 42. For all we know, Reihl did not like what he was 
hearing and was hoping that they would change their story. 
The majority also understates the degree of corroboration 
for Mandy’s account in the videotape (as I have said, corrob‐
oration that is just as good as that found in Chambers itself). 
Mandy’s  own  mother  interjects  corroborating  remarks  re‐
peatedly  during  the  interview.  My  colleagues  push  this  to 
one side because they believe that Monica’s subsequent off‐
the‐record, non‐testimonial statement to police that she (but 
not  Mandy)  had  the  wrong  day  effectively  erased  Monica’s 
own  consistent  corroboration  in  the  video.  The  transcript 
provides no support for this interpretation. To the contrary, 
Monica  is  an  active  participant  who  provides  her  own  de‐
tailed account of her afternoon on that Friday. Like Mandy, 
Monica herself saw Aaron after school, even though she did 
not see Rick. (No one thinks that Aaron and Rick took sepa‐
rate  cars  to  the  Kubsch  house;  Aaron  was  far  too  young  to 
drive.) And, as I already have pointed out, there was much 
more corroboration easily within reach.  
Last, some precision is necessary with respect to Mandy’s 
availability for cross‐examination. She was not “unavailable” 
in  the  sense  of  not  being  present  at  trial.  She  was  in  the 
courtroom  and  she  testified;  at  least  one  aspect  of  her  testi‐
mony, as the Indiana Supreme Court acknowledged, should 
have  been  impeached  by  her  statements  on  the  video.  She 
No. 14‐1898  93 

was “unavailable” only because her memory had failed. But 
that is true of every witness proffered under Rule 803(5). In‐
diana  courts,  like  others,  look  for  the  next‐best  assurances. 
Mandy never claimed that she was not the girl on the tape, 
nor has the state ever argued that the “Monica” on the tape 
was not Mandy’s mother. There was, in short, ample corrob‐
oration even on the record that exists to satisfy this aspect of 
the  Chambers  rule.  The  majority  sees  no  way  to  distinguish 
this  hearsay  from  the  ordinary  mine‐run  of  hearsay,  and  it 
accuses me of throwing the door open to admission of every 
recorded police interview. Not so. In many cases, the witness 
will have a good enough recollection of what happened that 
Rule  803(5)  will  never  come  into  play.  In  many  cases,  the 
proffered  hearsay  will  be  cumulative  or  relevant  only  to  a 
peripheral  matter. In the great majority of cases, the admis‐
sion of the hearsay statement will not have life‐or‐death con‐
sequences.  The  dissent  in  Chambers  worried  about  exactly 
the  same  things  the  majority  here  invokes.  But  the  dissent 
did not prevail, and the Supreme Court has continued to fol‐
low Chambers in the small group of cases to which it applies. 
This  court  should  not  be  second‐guessing  the  Supreme 
Court,  but  I  fear  that  is  what  the  majority  has  done.  Under 
its  view,  Chambers  will  never  apply  to  allow  a  defendant  to 
introduce pivotal evidence, if a state rule would block it. By 
so ruling, it is contravening the  Supreme Court’s command 
that “the hearsay rule may not be applied mechanistically to 
defeat the ends of justice.” Chambers, 410 U.S. at 302. 
In fact, this case is as close to Chambers as anyone is likely 
to find. My colleagues misapply the Supreme Court’s guid‐
ance in Williams v. Taylor, 529 U.S. 362 (2000) (O’Connor, J.), 
when  they  insist  on  a  precise  factual  match  between  Cham‐
bers  and  the  present  case.  The  Court  has  never  insisted  on 
94  No. 14‐1898 

factual identity between its earlier case and the new one. See 
id.  at  407  (“[A]  state‐court  decision  also  involves  an  unrea‐
sonable  application  of  this  Court’s  precedent  if  the  state 
court either unreasonably extends a legal principle from our 
precedent to a new context where it should not apply or un‐
reasonably refuses to extend that principle to a new context where 
it should apply.”) (emphasis added). Kubsch’s situation, while 
differing in some details from Chambers’s, is close enough to 
require application of the same principle. 
The majority fears that if Chambers requires admission of 
the videotape, then  state hearsay rules are out the window. 
But  their  gripe  is  with  the  Supreme  Court,  not  with  me.  I 
have shown why  and how  the facts cabin this  case. In  very 
few matters before the court will the price of insisting on ex‐
clusion of evidence that does not fit every technical require‐
ment of the state’s hearsay rule be death. That alone should 
lay to rest any fears that granting Kubsch relief under Cham‐
bers  will  produce  the  “sweeping”  result  the  majority  fears. 
Like defendant Chambers, Kubsch  was “thwarted in  his at‐
tempt to present this portion of his defense by the strict ap‐
plication  of  certain  [state]  rules  of  evidence.”  Chambers,  410 
U.S. at 289. In Kubsch’s case, the hearsay problem was com‐
pounded by the ineffectiveness of counsel’s efforts to get the 
tape admitted. 
In  Chambers  (also  a  murder  trial),  the  application  of  the 
state’s rules on vouching for witnesses and hearsay prevented 
the defendant from calling as an adverse witness the person 
who he said was the real murderer and three witnesses who 
would  have  supported  that  proposition.  The  state  excluded 
that  evidence  notwithstanding  the  fact  that  it  was  created 
“under  circumstances  that  provided  considerable  assurance 
No. 14‐1898  95 

of  [its]  reliability.”  Id.  at  300.  Those  circumstances  included 
the  fact  that  the  confessions  of  the  apparent  murderer  to 
which  each  excluded  witness  was  prepared  to  testify  were 
“made  spontaneously  to  a  close  acquaintance  shortly  after 
the  murder  had  occurred”;  each  was  corroborated  by  other 
evidence  in  the  case;  and  each  was  self‐incriminatory  and 
against the speaker’s interest. Id. at 300–01. The alleged true 
murderer “stood to benefit nothing by disclosing his role in 
the shooting,” and he was in the courtroom during the trial 
and  so  could  have  been  cross‐examined  by  the  state  and 
evaluated by the jury. Id. at 301.  
Mandy  and  Monica  Buck  were  not  potential  suspects  in 
this case, but their videotaped statements bore equally com‐
pelling  indicia  of  reliability.  The  majority  downplays  these 
facts,  but  they  overlook  the  significant  ways  in  which  the 
Supreme  Court  itself  has  confined  Chambers.  Granting  the 
writ  to  Kubsch  under  Chambers  would  not  abolish  the  rule 
against  hearsay,  any  more  than  Chambers  abolished  hearsay 
and vouching, the two rules at issue there. A set of very par‐
ticular  circumstances  must  arise  to  produce  a  case  like 
Kubsch’s, or like that in Chambers. As I already have pointed 
out, a result in Kubsch’s favor would not lead to the admis‐
sibility  as  substantive  evidence  of  “all  hearsay  of  this  type 
[videotapes?],” to use the majority’s words, ante at 45.  
In this case, the operation of Indiana’s hearsay rule, cou‐
pled  with  counsel’s  inadequate  efforts  with  regard  to  the 
tape,  prevented  Kubsch  from  showing  that  he  could  not 
have  been  the  murderer.  Like  Chambers,  Kubsch  also  tried 
to show that someone else was the guilty party—in Kubsch’s 
case, his sometime friend Brad Hardy. There appears to have 
been significant evidence pointing  to Hardy. Indeed, at one 
96  No. 14‐1898 

point  the  state  had  charged  him  with  conspiring  with 
Kubsch to commit the murders and with assisting a criminal 
(Kubsch). Kubsch II, 866 N.E.2d at 731. Hardy wound up tes‐
tifying  against  Kubsch  in  the  first  trial;  interestingly,  the 
state  did  not  drop  the  charges  against  him  until  two  years 
later.2  The  excluded  videotaped  evidence  in  Kubsch’s  case 
had even greater guarantees of reliability  than the evidence 
before the Supreme Court in Chambers. And the exclusion of 
the  videotape  drastically  undermined  Kubsch’s  ability  to 
demonstrate  that  someone  else  must  have  committed  the 
three  murders.  The  Chambers  exception  exists  for  just  this 
kind of case.  In  my  view,  the Indiana courts’  refusal  to rec‐
ognize  and  apply  it  amounts  to  constitutional  error  that 
must be recognized, even under the demanding standards of 
28 U.S.C. § 2254(d)(1). 
V 
Wayne  Kubsch  may  be  a  disagreeable  man,  as  Mandy 
said  in  her  videotaped  statement.  His  business  skills  may 
have been bad, and he may, as of September 1998, been flail‐
ing around for a way to solve his financial problems. And a 
jury  with  all  of  the  evidence  before  it  may  have  convicted 
him for the murders of Beth, Rick, and Aaron, if it had been 
persuaded that Mandy’s videotaped testimony was not wor‐
thy  of belief for  some  reason. But a jury with all of  the evi‐
dence  before  it  may  also  have  concluded  that  Kubsch,  no 
matter what his other flaws, could not have committed those 

2  As my colleagues point out, Hardy testified against Kubsch in the 

second  trial.  By  that  time  they  were  surely  adverse  to  one  another;  in‐
deed, it would not be surprising if Hardy’s charges were dropped in ex‐
change for that testimony.  
No. 14‐1898  97 

murders  because  Rick  and  Aaron,  and  perhaps  Beth,  were 
still alive at 3:45 p.m., when Kubsch was already far from the 
house driving to Michigan. We will never know, because my 
colleagues  are  unwilling  to  find  either  the  disregard  or  in‐
correct  application  of  Chambers  here,  nor  do  they  perceive 
ineffective  assistance  of  counsel.  I  cannot  subscribe  to  that 
result.  I  therefore  respectfully  dissent  from  the  decision  to 
affirm  the  district  court’s  denial  of  the  writ  and  the  conse‐
quent green light for Kubsch’s execution.  
98  No. 14‐1898 

APPENDIX A 
Timeline of events, September 18, 1998 

Time  Kubsch  Beth/Others 

6:00 am  Near Mishawaka home  Beth is at work in Elkhart 
(cell record).  (United Musical Instru‐
ments). 

6:50 am  At work in Elkhart   
(Skyline Corp.). 

9:11 am  Cellphone call near   
work. 

10:00 am    Beth finishes shift and goes 
home. 

10:30 am    Beth pages Kubsch twice 
from home. 

10:45 am  Call to Beth from Sky‐  
line break room. 

10:48 am    Beth makes a call from 
home to Rick’s house. 

~10:53 am    Beth goes out to run er‐
rands. 

11:08 am    Security camera at Teach‐
er’s Credit Union shows 
Beth with the dog in the 
car. 

11:13 am  Kubsch punches out of   
work. 
No. 14‐1898  99 

11:14 am    Beth’s credit union receipt 
shows transaction com‐
pleted. 

11:30 am to  Kubsch at home (seen   
noon  by Erin Honold). 

11:37 am  Call from home to   
American General Fi‐
nance. 

11:52 am    Beth meets with credit 
counselor Edith Pipke in 
South Bend. 

12:09 to  Kubsch makes 1 call to   
12:11 pm  house and 2 calls to 
Rick (cellphone). 

12:16 pm    Beth pages Kubsch again. 

12:18 pm  Kubsch calls the house   
(31 seconds) from Os‐
ceola (toward Elkhart). 

12:40 pm  Kubsch calls house   
from break room at 
Skyline. 

12:46 pm    Rick calls Beth at home. 

1:17 pm  Kubsch calls house   
from break room at 
Skyline. 

1:52 pm  Kubsch punches out   
again and does not re‐
100  No. 14‐1898 

turn. 

1:53 pm  Kubsch calls home   
from Elkhart area (46 
seconds). 

2:20 to 2:35    Rick picks up Aaron from 
pm  school in South Bend. 

2:51 pm  Kubsch makes call   
from near home (cell 
records). 

3:15 pm   Kubsch calls Beth’s   
mother from Elkhart 
(after 10 tries). Cell sec‐
tors indicate he is 
heading toward Mich‐
igan. 

3:45 to 4:15    Approximate time when 
pm  Mandy saw both Aaron 
  and Rick at their South 
  Bend home. 

4:42 to 4:47  Kubsch makes calls   
pm  near Schoolcraft, MI. 

5:00 pm  Kubsch picks up son   
Jonathan in Three Riv‐
ers, MI. 

5:30 to 5:45  Kubsch sees Wayne   
pm  Temple at Kmart in 
Three Rivers. 
No. 14‐1898  101 

5:30 to 6:30  Kubsch and Jonathan   
pm  stop in Osceola at 
home of Constance 
Hardy. 

5:30 pm    Anthony discovers the 
bodies of Rick and Aaron 
Milewski at the house. 

5:56 pm  Kubsch makes phone   
call on network close 
to the house. 

6:45 pm  Kubsch returns home;   
police are there; he 
goes to station for first 
interview. 

9:00 pm    Police discover Beth’s 
body in basement; they 
bring Kubsch back to the 
station. 

After 9:00  Kubsch interviewed   
pm  second time by police; 
he invokes Miranda 
rights. 

102  No. 14‐1898 

APPENDIX B 
Transcript of Police Interview with Monica and Mandy Buck 
September 22, 1998 

Det. Mark Reihl: [Inaudible] stepped out for a minute. I’ll go 
ahead  and  start  asking  you  a  couple  questions.  Okay,  and 
the  time  is  now  three  o’clock  PM.  And,  today  is  September 
the  twenty‐second,  nineteen  ninety‐nine—nineteen  ninety‐
eight. And Mandy, is it M‐a‐n‐d‐y? 
Mandy: Uh huh. 
Reihl: M‐a‐n‐d‐y. Buck. B‐u‐c‐k?  
Mandy: Uh huh. 
Reihl: And you’re how old? 
Mandy: Nine. 
Reihl: Your birthdate is? 
Mandy:  Ninety‐eight.  Nineteen  ninety‐eight.  Oh,  nineteen 
eighty‐nine. 
Reihl: This is nineteen ninety‐eight. 
Mandy: Nineteen eighty‐nine. 
Reihl: What month were you born? 
Mandy: February. 
Reihl: February. What day? 
Mandy: Eighth. 
Reihl: Nineteen eighty‐nine. 
Mandy: Yeah. 
No. 14‐1898  103 

Reihl: Alright. 
Mandy: But you can ask my mommy on that. I think so. 
Reihl: Oh, I’m pretty sure, all right? You’re pretty intelligent. 
I think you know.  
Mandy: Yeah, I think that, yeah yeah yeah. 
Reihl: Mandy was born February the eighth?  
Monica: Yeah. 
Reihl: Nineteen eighty‐nine? 
Monica: Mmm hmm. 
Reihl: Okay. 
Mandy: Cool, I got it right. 
Reihl: See, you got it right. Okay. And your mother’s name is 
Monica? 
Mandy: Uh huh. 
Reihl: M‐o‐n‐i‐c‐a? Correct me? 
Monica: Yeah. 
Reihl: Buck. And you live at thirteen twenty East Indiana in 
South Bend. 
Mandy: Uh huh. 
Reihl: And your home phone is two three three, seven seven 
three seven? 
Mandy: Two three three seven seven three seven. Yep. 
Reihl: Right. And you go to Lincoln School? 
Mandy: Yeah. 
Reihl: And you’re in which grade? Fourth? 
104  No. 14‐1898 

Mandy: Yeah. 
Reihl: Okay. How’s school this year? 
Mandy:  Umm,  good,  even  though  I  have  the  teacher  that, 
um, is the Wicked Witch of the West, she’s fine. She’s okay. 
Reihl: Well sometimes they gotta be like that so you kids will 
listen. 
Mandy: Yeah. 
Reihl:  Okay.  Well,  the  reason  you’re  here  is  that  you  live 
right across the street— 
Mandy: From Aaron? 
Reihl: From Aaron and his dad Rick. 
Mandy: Yeah. 
Reihl: Okay. And you and Aaron were pretty good friends, 
huh? 
Mandy: Best friends, yeah. 
Reihl: Best friends? 
Mandy: [Nods head] 
Reihl: How long have you known Aaron? 
Mandy: I don’t know. I think he moved there in like the be‐
ginning of May I think. Just beginning. I don’t know. I never 
kept track of it. I don’t know. ‘Cause he told me one day and 
then I just forgot. 
Reihl: Oh, that’s okay. 
Mandy: I can’t remember I think— 
No. 14‐1898  105 

Reihl: Time just goes by so fast, doesn’t it? And you said that 
Aaron  used  to  talk  sometimes  about  things  that  made  him 
sad? 
Mandy: Mmm hmm. [Nods head] 
Reihl: Made him upset? 
Mandy:  Right,  and  like  he,  he  he  wished  his  mom  didn’t 
break up with his dad and like go with Wayne. He was like, 
he didn’t like Wayne. 
Reihl: Aaron didn’t like Wayne? 
Mandy: No. 
Reihl: Well how come? 
Mandy: Um because, like, he would get rough with him and 
stuff and punch him too hard and stuff like that. 
Reihl:  Was  it  because—did  he  ever  say  was  it  because 
Wayne was mad at him or were they just playing? 
Mandy: He never said, he never said why he didn’t like him 
he just said like, he just said he just didn’t like him because 
Wayne was just like too rough and stuff. 
Reihl: Okay. Did he ever say if Wayne ever was rough with 
his mom? 
Mandy: No. 
Reihl: You didn’t talk about that? 
Mandy: No. 
Reihl: Okay. What else did you guys talk about? 
Mandy: Um, we talked about like, why he moved here and 
like what we wanted to be when we got older and, um, who 
106  No. 14‐1898 

are  our  friends  and  where  we  used  to  live  and,  like,  and  I 
introduced him to my parents; he introduced me to his dad. 
Then we just became best friends. 
Reihl: That’s great. 
Mandy:  I  always  went  over  to  his  house.  He  always  came 
over to my house and like we like used to study for the same 
spelling  words.  He’d  give  me  my  spelling  words  and  I 
would  give  him  his  spelling  words.  And  we  would  help 
each  other  on  homework  and  stuff.  We  were  pretty  good 
friends. 
Reihl: That’s, that’s wonderful. 
Mandy: We got along really good. 
Reihl: He’s a pretty good kid, huh? 
Mandy: Mmm hmm. [Nods head] 
Reihl: Smart? 
Mandy:  Uh  huh.  [Nods  head]  He  knew,  he  knew  his  times 
pretty  good.  He  could,  he  could  just  do  ‘em  in  a  flash.  He 
was pretty good at ‘em. He’s a lot better than me. 
Reihl: Did you, did you say you used to walk to school with 
him sometimes? 
Mandy: Uh no, I never walked. 
Reihl: Oh, you never did. 
Mandy: No. I see—I seen him walk to school. 
Reihl: Uh‐huh. 
Mandy:  I  never  walked  to—I  never  walked  to  school  or  to 
my house alone. 
No. 14‐1898  107 

Reihl: Okay, and how would he get home? 
Mandy:  Um,  usually  some,  if  he  wasn’t  grounded  from  his 
bike  would  ride  his  bike  home.  He  would  walk  home.  His 
dad  would  come  and  pick  him  up  when  he  had  his  truck. 
Um,  Rick  would  walk  to  school  and  pick  up  Aaron.  They 
would walk back home together. 
Reihl:  Mmm  hmm.  And,  and  you  guys  get  out  of  school  at 
what time? 
Mandy: Two twenty. 
Reihl:  Two  twenty.  And  how  long  does  it  take  him  to  get 
home do you think? 
Mandy:  Mmm  probably  like—we  don’t  live  too  far  from 
Lincoln. All you gotta do is go straight and turn and you’re 
there. 
Reihl: Oh. 
Mandy: Probably like five minutes to get there. 
Reihl: Uh‐huh. Okay. 
Mandy: If he was riding his bike it would only take him like 
two  minutes. But if he  was walking it would probably take 
him a pretty long time. 
Reihl: Mmm hmm. Now, do you remember last Friday? 
Mandy: Yeah. 
Reihl: Okay. And you told me earlier that you go to the Al‐
phabet Academy? 
Mandy: Uh huh. [Nods head] 
Reihl: And that they usually pick you up at school, right? 
108  No. 14‐1898 

Mandy: Uh huh. [Nods head] 
Reihl: Okay. And did they pick you up Friday? 
Mandy: Uh huh. [Nods head] 
Reihl: And you went straight to the Alphabet Academy? 
Mandy: Uh huh. [Nods head] 
Reihl: And say then you what, your mom picks you up from 
there? 
Mandy: Uh huh. [Nods head] 
Reihl:  Okay.  And  you  said  you  picked  her  up  about  what 
time? 
Monica: Between three thirty and quarter to four. 
Reihl: Okay. And you went straight home? Or where’d you 
go? 
Monica: I usually call down there and I watch her walk from 
there  down  to  our  house.  And  then  I  waited  for  my  mom 
and dad to get home, and I went and cashed my check and 
came home. 
Reihl: Okay, when you got home at three thirty, um, did you 
notice if Rick was at home across the street? 
Monica: I didn’t pay no attention. All I saw was Aaron. 
Reihl: You saw Aaron? 
Monica: Mmm hmm. 
Reihl: You don’t remember if Rick’s truck was there? 
Monica: No. 
Reihl:  Okay.  And,  then  Mandy  you  were  telling  me  that 
when you got home that was about what time? 
No. 14‐1898  109 

Monica: From day care? 
Reihl: Yeah. 
Monica: That was around three thirty, quarter to four. 
Reihl: Okay, and that’s when you saw Aaron? 
Mandy: Uh huh. [Nods head] 
Reihl: And you saw his dad? 
Mandy: Uh huh. [Nods head] His dad, he, his dad was com‐
ing from their living room into the kitchen to get something 
to drink. 
Reihl: Did you go over to Aaron’s house or you just saw him 
from your house? 
Mandy:  I,  I,  um,  when  I  walked,  when  I,  every  day  when  I 
walk home  I always see  Rick walk  into  the  kitchen or  walk 
into the restroom or walk into his room. 
Reihl:  I  mean,  did  you  see  him  from  outside  looking  in  or 
did you actually go into the house? 
Mandy: No, I um seen it from the outside ‘cause when ‘cause 
I seen him go into the kitchen. When he came back he had a 
drink  in  his—he  had,  um,  some  um—I  don’t  know  what  it 
was. He had a drink in his hand but it was in a cup. 
Reihl: Okay. 
Mandy: Like usually pop, ‘cause they like, they like Storm a 
lot. So, probably Storm. 
Reihl: What, uh, what does Rick drive? 
Mandy:  A  Chevy?  He  used  to  drive  a  Chevy  until  it  broke 
down. 
110  No. 14‐1898 

Reihl: A Chevy what?  
Mandy: [Eyes searching, no verbal response] 
Reihl: Is it a car or a truck? 
Mandy: Truck. 
Reihl: What color? 
Mandy: Black. 
Reihl: Okay. 
Mandy: It’s like, kinda short. I mean like it—did you see my 
mom’s  truck?  Um,  well,  uh  my  mom’s  truck,  my  mom’s 
truck’s  pretty  big.  His  is  probably  a  medium  truck,  you 
know. Kinda short. 
Reihl: What was he driving Friday? Did you see that? 
Mandy:  Um,  his  truck  broke  down  before  that.  He  was 
drive—driving  a  white  truck  which  was  his  brother’s.  And 
his brother had a car so his brother let Rick use the truck. 
Reihl: Okay. Was that white truck at Rick’s house Friday? 
Mandy: Yeah. 
Reihl: When you got home from school? 
Mandy: Yeah. 
Reihl: Okay. And this is about what time again? 
Monica: Three thirty, quarter to four. 
Reihl: Okay, so between three thirty and quarter to four— 
Mandy: Yeah. 
Reihl: You saw— 
Mandy: Aaron and Rick. 
No. 14‐1898  111 

Reihl: Okay, at the house. Did you ever see ‘em leave? 
Mandy: Um, yeah, like I was on my porch and, and they let 
me blow bubbles. And I was blowin’ my bubbles, and I seen 
Rick pull out and leave. 
Reihl: Okay. Now how long, how long after—and this might 
be  hard  to  guess  at—‘cause  you  probably  don’t  wear  a 
watch, do you? 
Mandy: Well, until my watch, well, yeah I did but my watch 
is in my bag and I—‘cause I had to take it off when we had 
gym. I just take it off. 
Reihl: So, about what time do you think they left their house, 
if you had to guess? 
Mandy: Um— 
Reihl: I know it’s gotta be a hard question. 
Mandy: Um— 
Reihl: Was it very long after you got home? 
Mandy:  Mmm,  medium.  Because  his  mom  lives  pretty  far 
away,  you  know.  And  you  know  but  I  think  it  was  like—I 
don’t know. 
Reihl: Okay. 
Mandy: It was probably in like medium because you know it 
takes a pretty long time to get to his mom’s house. 
Reihl: Well why was he going to his mom’s house. I think he 
told you, didn’t he? 
Mandy: Um, I guess to just visit her. 
Reihl: Okay, did he talk about going to his mom’s house? 
112  No. 14‐1898 

Mandy:  He  said  that  he  was  going  to  his  mom’s  house  Fri‐
day,  ‘cause  he  was  gonna  stay  the  night  there  to  go  to  the 
field trip Saturday. So it was probably why, and Rick proba‐
bly wanted to stay a little while to talk. You know, he was, 
he—he  wanted  to  go  on  the  field  trip  bad.  So,  they  were 
gonna  leave  pretty  early  to  get  to  the  school  on  time  to  go. 
But  by  the  time  Saturday  when  we,  when  we  were  on  the 
bus  and  stuff,  he  was  gonna  be  in  our  group,  and,  um,  he 
never  showed  up.  He  wasn’t  there.  And  we  didn’t  know 
why.  But  Saturday—Sunday  when  we  got  home  with  my 
cousins, um, ‘cause we go camp—we went camping after the 
field  trip,  we  just  went,  we  came  back  from  the  field  trip, 
and  my  mom  drove  her  truck  back  to  the,  back  up  to  our 
house  and  up  to  the  camper  and,  and  my  grandma  goes, 
“Did you see Aaron?” and I’m like, “No, he was supposed to 
be in our group, he wasn’t there.” And then Sunday, um, my 
um, my day care teacher said they showed it on TV but my 
grandpa  didn’t  get,  my  grandpa  didn’t  turn  it  on  there  be‐
cause  he,  he  didn’t  know  it  was  they  got  murdered  Friday 
night. So, I mean, and then Monday, um, Monday, Monday 
News  Center  16  came  to  my  house,  and  I  was  at  karate 
‘cause  I,  I  had  practice.  When  we  came  home  my  grandma 
said News Center 16 just, just came to our house like, proba‐
bly a while ago. 
Reihl: So you didn’t get a chance to talk to him then, huh? 
Mandy: No. 
Reihl: So, Friday, after you got home, they left just a little bit 
after when you got home, right? 
Mandy: Yeah.  
Reihl: And you saw ‘em leave? 
No. 14‐1898  113 

Mandy: Yeah. He pulled out. 
Reihl: And they were just together, Rick and Aaron, nobody 
else with ‘em? 
Mandy: No one else was with them, just Aaron and Rick. 
Reihl: Okay. 
Mandy: ‘Cause Rick, ‘cause Aaron’s mom—He didn’t know 
if  Aaron’s  mom  was  home  yet  so  Rick  was  thinking  if  his 
mom’s  not  there,  then  Wayne’s  probably  not  there.  So,  he 
said, “I’ll just drive you,” and they just  took off, pulled out 
and took off. 
Reihl: Okay. 
Mandy: And— 
Reihl: Monica, Monica, I’m sorry. 
Mandy: And Fri— and Thur—and when I was playing with 
them— 
Reihl: Mmm hmm. 
Mandy: There was, he had some clothes laying on his, laying 
on his on their swing on the front porch. Um, he had a whole 
bunch  of  clothes  laying  on  there  and  I,  I  didn’t  know  what 
they were for. You know, I thought he was gonna spend the 
night there Saturday and Sunday, come home Monday. Um, 
Sunday’s rolling around and he wasn’t there. Saturday, Sat‐
urday the field trip, he wasn’t there. 
Reihl: Monica, you said something back at your house when 
I was talking to you about um, you said you’d cashed your 
check. 
Monica: Yeah. 
114  No. 14‐1898 

Reihl: Friday? 
Monica: Yeah. 
Reihl:  And  that  was  about  what  time?  Was  that  after  you 
come home from work? 
Monica: Shortly after I came home from work. 
Reihl: Okay. And, what time do you think that was? 
Monica: Let’s see. Probably about ten minutes till four. 
Reihl:  Okay.  So  then  you  got  home  then  about—how  long 
were you gone to cash the check? 
Monica: Probably about fifteen minutes. 
Reihl: Okay, and when you got home, that would have put it 
a  little  after  four  o’clock?  And  was  Rick  still  at  the  house 
then? 
Monica: I  didn’t pay  no attention. Like I said, all I saw was 
Aaron. I really didn’t look to see if Rick’s truck was there. 
Reihl:  Well,  Aaron  was  still  there  when  you  got  back  after 
you cashed your check? 
Monica: Yeah. 
Reihl:  Okay.  And  you  don’t  remember  if  that  truck  was  in 
the— 
Monica: Nuh uh, I didn’t pay no attention. 
Reihl: Okay, um—You said something, too, didn’t you about 
you overheard something one time a couple months ago. 
Monica:  Yeah.  I  don’t,  like  I  said,  I  don’t  know  who  the 
woman  was.  But  he  was  standing,  they  were  standing  in 
their driveway. And, well he was standing in the driveway. 
No. 14‐1898  115 

She  was  sitting  in  the  truck.  And,  uh,  I  couldn’t  hear  what 
she was saying, but he was, you know, he was saying the F‐
word,  and  F  him,  he  don’t scare me,  and he was just  going 
on  and  on  and  on.  And  then  he,  then  she  left,  and  he  just 
went into the house. 
Reihl: This truck, what did it look like? 
Monica: It was a, it was a little black truck. 
Reihl:  Do  you  know,  do  you  know  your  vehicles?  Do  you 
know the difference between a— 
Monica:  Well,  the  lettering  on  the  back  was  kinda,  on  the 
back of it was kinda like, rusted like, and you couldn’t really 
tell what kind of car it was— 
Mandy: Um— 
Monica: —what kind of truck. 
Mandy:  Aaron’s  dad’s  truck  had  Chevy  right  there.  It  was 
just printed beautifully. It was gold and it was just right on 
there. You could just read it, so it couldn’t have been Aaron, 
Aaron’s  dad’s  truck,  ‘cause  Aaron’s dad’s truck was, but, it 
was  still  there  where  he,  it  broke  down.  I  mean  Aaron’s 
truck’s, dad’s truck was just beautiful. The Chevy was just— 
Reihl: But was this was this his ex‐wife? Was this— 
Monica: I don’t know. 
Reihl: —Elizabeth? 
Monica: I don’t know who she was. Like I said, all I saw, all 
I, I never seen the woman. You know, I, I just know that she 
had  blonde  hair.  Well,  I  seen  her  face,  but  she  had  blonde 
hair. 
Reihl: Was she a passenger in the truck? 
116  No. 14‐1898 

Monica: No. She was driving it.  
Reihl: Okay. 
Monica: And this was, then I saw her once a little while after 
that.  You  know,  like  a,  I  don’t  know,  a  couple  weeks  later. 
And that was the last time I seen her. 
Reihl: What was she driving then? 
Monica: Same thing. 
Reihl: This truck? 
Monica: Mmm hmm. I don’t know, I don’t, like I said I don’t 
know who she was. 
Mandy: Aaron’s mom’s, mom has um, blonde hair. 
Reihl:  Mmm  hmm.  I  was  just  trying  to  see  if  maybe  you 
could describe this truck. Was there anything, was it, was it 
a pickup truck where it has the open bed in the back or was 
it all closed up? 
Monica: Uh, let me think. I think it was open. See, ‘cause the 
one  that  that,  ah,  Aaron’s  dad  used  to  drive  had  the  little 
things that went down the side. 
Reihl: Mmm hmm. 
Monica:  But  it  wasn’t  all  closed  in.  It  just  had  like  little,  I 
don’t know what you’d call ‘em, it went from the top all the 
way  to  the  back  of  the  truck,  and  it  was  just  a  short  thing. 
This one was all open, I believe. I think it was. 
Reihl: It was just like a regular pickup truck. 
Monica: Yeah. 
Reihl: Okay. So it wasn’t like a little sport utility vehicle? 
No. 14‐1898  117 

Monica: No. 
Reihl:  Like  you  see  like  one  of  those  Suzuki  Samurais  or 
something like that? 
Monica: No. It was— 
Reihl: Kids drive a lot. 
Monica: It was pretty rusted. 
Reihl:  Okay.  All  right.  But  you  don’t  know  whether  or  not 
that was his— 
Monica: No I have no idea. 
Reihl: His ex‐wife Elizabeth or not? All right.  
Monica: I just know that he was highly upset that day. 
Reihl: Oh. 
Monica: And she didn’t look too happy, and she left and he 
went into the house.  
Reihl: Okay. 
Monica: Yeah, I don’t even, I don’t know who his ex‐wife is. 
I mean, it could have been her, but I, I don’t know. 
Reihl: Okay. Was there anything  else? I can’t  remember ex‐
actly what all we talked about at the house but, did you say 
that, uh, I was thinking that you said that Aaron had made 
some comments to you before, too, about— 
Monica: Oh, he just told me the once. 
Reihl: Oh. 
Monica:  He  just  told  me  one  time  that  he  doesn’t  like  his 
stepdad. But, I just figured he was just being a kid. 
Reihl: Yeah.  
118  No. 14‐1898 

Monica: You know, “My mom and dad’s divorced but I real‐
ly don’t like this guy. I don’t want Wayne really to be with 
my mom. I’d rather, you know, him and my mom be togeth‐
er—” 
Reihl: Mmm hmm. 
Monica:  “—than  my  stepdad,”  kinda  thing.  That’s  all  I 
thought it was. So I just really didn’t pay no attention to it. 
Reihl:  Okay.  Okay.  All  right.  Well,  just  so  I  got  this  right 
then, Mandy, you got home at about three thirty, quarter of 
four and you saw Aaron and his dad and that white truck at 
his house? 
Mandy: Yes. 
Reihl:  And  then,  Monica,  you  got  home  from  cashing  that 
check around four o’clock or a little after, and you saw them 
both at the house, or at least you saw Aaron? 
Monica: Yeah, I saw Aaron. 
Reihl: Okay. But you never saw ‘em leave. 
Monica: No. I was in the house by the time they left.  
Reihl:  Okay,  and  Mandy,  you  did  see  ‘em  leave,  but  you 
don’t know exactly when it was that they left? 
Mandy:  Yeah.  I  seen  ‘em  leave,  but,  you  know  I  didn’t  see 
no, I didn’t see no bags in the truck. And when, when they 
left, the clothes were still there. 
Reihl: Okay. On the swing? 
Mandy:  Um,  yeah.  ‘Cause  when  his  grandparents  were 
there,  they  picked  up  the  clothes  and  just  threw  ‘em  in  the 
box. 
No. 14‐1898  119 

Reihl: Okay. 
Mandy: And we thought that he was moving, like he didn’t 
like the neighborhood so he was moving. What we thought, 
and I don’t know if it, I didn’t know if Rick and Aaron Fri‐
day were gonna go look for a new house or go to his mom’s. 
I  didn’t know, I  thought  they were  going to look  for  a new 
house  and  then  come  back,  and  you  know,  and  go.  Like, 
then go to his mom’s. But, I didn’t, I didn’t know. 
Reihl:  Okay.  These  times  that  you’ve  given  me  today,  uh, 
these are pretty accurate? 
Monica:  Mmm  hmm.  Yeah,  ‘cause  I  get  off  work  at  quarter 
after three. And with the traffic and that, and sometimes the 
South Shore comes by and you gotta wait for that. 
Reihl: Mmm hmm. 
Monica: So, yeah, pretty well. 
Reihl: It’s pretty much a routine that you do every day? 
Monica: Yeah. 
Reihl: Every day that you work, that is? 
Monica:  Yeah.  Sometimes  on,  sometimes  I  have  to  stay  a 
couple  minutes  after,  so,  I  get  home  a  little  later.  And  that 
was just so happen to have been one of the days that was a 
little bit later. 
Reihl:  Okay.  All  right.  I,  I  don’t  have  any  more  questions 
that I can think of at the moment. Do you have anything else 
that you can think of? Maybe I overlooked, that I have over‐
looked? 
Monica: No. Do you? 
Mandy: [Shakes head] 
120  No. 14‐1898 

Reihl: I thank you very much for coming down. I’ll take you 
back home now. The time is, uh, three twenty PM. [Pause] I 
told you that would take you about fifteen, twenty minutes. 
Mandy: [Pointing to ceiling] Is that your camera? 
Reihl: It’s up there. 
Mandy:  Oh,  there  it  is.  I  thought  it  was—it’s  in  that  vent 
right there.