ATTORNEYS FOR APPELLANT ATTORNEYS FOR APPELLEE
Susan K. Carpenter Steve Carter
Public Defender of Indiana Attorney General of Indiana
Thomas C. Hinesley James B. Martin
Steven H. Schutte Deputy Attorney General
Deputy Public Defenders Indianapolis, Indiana
Indianapolis, Indiana
______________________________________________________________________________
In the
Indiana Supreme Court
_________________________________
No. 87S00-0106-PD-285
JOHN M. STEPHENSON,
Appellant (Petitioner below),
v.
STATE OF INDIANA,
Appellee (Respondent below).
_________________________________
Appeal from the Warrick Superior Court, No. 87D02-0210-PC-118
The Honorable Robert R. Aylsworth, Judge
_________________________________
On Petition for Post-Conviction Relief
_________________________________
April 26, 2007
Boehm, Justice.
After an eight-month trial, a jury found John Matthew Stephenson guilty of burglary,
theft, and the murders of three people. At trial, the defense contended that the murders of John
“Jay” Tyler; his wife, Kathy Tyler; and Brandy Southard were the result of a drug operation un-
related to Stephenson. The testimony of several defense witnesses implicated persons Stephen-
son contended were involved in the drug ring. Stephenson also presented alibi witnesses who
testified to his whereabouts at the time the State alleged the murders took place. The jury found
Stephenson guilty and found multiple murders as an aggravating circumstance supporting the
death penalty. The trial court, following the jury’s recommendation, sentenced Stephenson to
death. We affirmed both the convictions and death sentence. Stephenson v. State, 742 N.E.2d
463 (Ind. 2001), cert. denied, 534 U.S. 1105 (2002).
During both the guilt and sentencing phases of trial, Stephenson was forced to wear a
stun belt in the jury’s presence. No objection was made by Stephenson’s trial counsel to the stun
belt, and the trial record made no reference to the use of the belt. Stephenson sought post-
conviction relief, alleging that (1) the use of a stun belt was structural and fundamental error; (2)
trial counsel and appellate counsel were ineffective on a number of grounds, including trial
counsel’s failure to object to the belt; (3) new evidence undermined confidence in his convictions
and death sentence; (4) prejudicial outside influences biased the jury; and (5) the State withheld
material exculpatory evidence. Post-conviction relief was denied, and this appeal followed.
We affirm the denial of post-conviction relief. Specifically, we hold:
(1) Stephenson’s freestanding claims of error based on his wearing a stun belt at trial
were available on direct appeal and are therefore foreclosed in post-conviction pro-
ceedings;
(2) Because appearing in readily visible restraints is inherently prejudicial, if the issue
had been raised on appeal, reversal would have been required unless the State had
proved beyond a reasonable doubt that the error did not affect the result as to either
guilt or the penalty;
(3) Stephenson’s claim of ineffective assistance of counsel requires him to establish sub-
standard performance of counsel and a reasonable probability that the result would
have been different but for counsel’s errors and omissions;
(4) Even if Stephenson’s trial counsel’s failure to object to the belt or to the lack of
finding of need for any form of restraint fell below prevailing professional norms,
Stephenson has failed to establish a reasonable probability that any such objection
would have prevailed; he therefore has not established a reasonable probability that
the result of either the guilt or the penalty phases would have changed.
(5) In death penalty cases, we are to evaluate claims of newly discovered evidence under
the standard established in 2003 by Indiana Code section 35-30-2-9(k), which is
2
whether the previously undiscovered evidence undermines confidence in the convic-
tion or sentence;
(6) Because Stephenson’s claims of newly discovered evidence largely turn on the credi-
bility of various witnesses and were rejected by the post-conviction court, they do not
undermine confidence in Stephenson’s convictions or death sentence;
(7) Stephenson was not deprived of his right to a fair trial or due process because of the
jury’s exposure to various extraneous influences; and
(8) The post-conviction court’s conclusion that Stephenson failed to prove by a
preponderance of the evidence that the State suppressed evidence that was material to
his guilt or punishment is affirmed.
Standard of Review
Post-conviction proceedings are civil proceedings that provide defendants the opportunity
to raise issues not known or available at the time of the original trial or direct appeal. Conner v.
State, 711 N.E.2d 1238, 1244 (Ind. 1999). Thus, if an issue was known and available but not
raised on direct appeal, the issue is procedurally foreclosed. Timberlake v. State, 753 N.E.2d
591, 597 (Ind. 2001). If an issue was raised and decided on direct appeal, it is res judicata. Id.
If a claim of ineffective assistance of trial counsel was not raised on direct appeal, that claim is
properly raised at a post-conviction proceeding. Id. In post-conviction proceedings, the defen-
dant bears the burden of proof by a preponderance of the evidence. Wallace v. State, 553 N.E.2d
456, 458 (Ind. 1990).
We review the post-conviction court’s factual findings under a “clearly erroneous” stan-
dard but do not defer to the post-conviction court’s legal conclusions. Stevens v. State, 770
N.E.2d 739, 746 (Ind. 2002). We will not reweigh the evidence or judge the credibility of the
witnesses; we examine only the probative evidence and reasonable inferences that support the
decision of the post-conviction court. Conner, 711 N.E.2d at 1245.
I. Use of a Stun Belt at Trial
Stephenson contends that his appearance in a stun belt before the jury at his trial vio-
lated his federal constitutional rights under the Sixth and Fourteenth Amendments and also vio-
lated state law. This claim is asserted as both a freestanding claim of error and a ground for inef-
3
fective assistance of trial counsel. As an initial matter, this Court has ruled that the use of a stun
belt is not to be ordered in Indiana courts. Wrinkles v. State, 749 N.E.2d 1179 (Ind. 2001). We
agree with Stephenson that if this were an appeal from a trial conducted after Wrinkles was de-
cided and Stephenson had objected to the use of the belt at trial, he would be entitled to a new
trial as a matter of state law. In addition, Deck v. Missouri, 544 U.S. 622 (2005), decided after
Stephenson’s trial and appeal, clarified a number of relevant federal constitutional principles
governing the use of restraints. However, Stephenson’s trial and direct appeal occurred before
Wrinkles and Deck were decided, and there was no objection to the use of the belt at Stephen-
son’s trial. This appeal from the denial of post-conviction relief therefore presents the threshold
issue of which, if any, of the claims Stephenson now asserts have been procedurally defaulted
and how these issues relate to the claim of ineffective assistance of counsel.
A. Some Relevant Settled Principles
We think it useful to set out some settled principles of substantive law before addressing
these questions. The Supreme Court of the United States has not ruled on the use of a stun belt
as a violation of the Federal Constitution, but the Court has given guidance on a number of rele-
vant points. Requiring a defendant to appear in jail garb has long been held to deny due process.
Holbrook v. Flynn, 475 U.S. 560, 567 (1986); Estelle v. Williams, 425 U.S. 501, 503 (1976). An
objection to jail garb is required before the “compulsion” required for a due process violation is
found. Estelle, 425 U.S. at 512-13. In short, jail garb is categorically prohibited by the Fifth and
Fourteenth Amendments if the defendant objects.
Unlike jail garb, shackling may be imposed, but only if the trial court makes a particular-
ized finding of need in the specific case. This rule has long been in place under the common
law. Deck, 544 U.S. at 626-27; Coates v. State, 487 N.E.2d 167, 169 (Ind. Ct. App. 1985). For
many years courts have thought this to be a requirement of federal due process. Illinois v. Allen,
397 U.S. 337 (1970); see also Deck, 544 U.S. at 629. Most recently, this doctrine has been held
applicable to the penalty phase as well as the guilt phase of a death penalty trial. Deck, 544 U.S.
at 627. Deck also made clear, if there had been any doubt, that this rule has “constitutional di-
mensions” and unnecessary shackling constitutes a denial of due process. Id. at 629, 632. Jail
garb and unnecessary shackling are both “inherently prejudicial” and, if proper objection is
made, require reversal unless the State establishes “beyond a reasonable doubt that the [shack-
4
ling] error complained of did not contribute to the verdict.” Id. at 635 (alteration in original)
(quoting Chapman v. California, 386 U.S. 18, 24 (1967)).
As explained in Deck, three reasons underlie the prohibition on unnecessary shackling.
First, visible shackling “undermines the presumption of innocence and the related fairness of the
fact-finding process.” Id. at 630. Second, shackling can interfere with the defendant’s ability to
communicate with his lawyer and participate in the defense. Id. at 631. Third, shackles impair
the dignity of the judicial process. Id. at 631-32.
We have already noted that Indiana state law no longer permits the use of stun belts in
Indiana courts, but that rule had not been announced at the time of Stephenson’s trial. The pro-
hibition of stun belts is not based solely on the considerations that underlie the prohibition on jail
garb. It is also grounded in the perceived effect on the defendant of the threat of imminent high
voltage. It thus is not wholly dependent upon the jury’s awareness of the belt, and, like jail garb,
is “inherently prejudicial.” Wrinkles, 749 N.E.2d at 1194.
B. Freestanding Claim of Error
Stephenson contends that his failure to object to the belt at trial did not preclude him from
raising that issue on direct appeal because use of the belt constituted “fundamental error.” The
parties dispute whether the use of the belt met that standard. That issue is moot insofar as Ste-
phenson seeks to assert the belt as a freestanding claim in this post-conviction proceeding. Be-
cause no objection was raised at trial and the issue was not presented on direct appeal, a chal-
lenge to the use of the belt is foreclosed in this post-conviction proceeding as a freestanding
claim of error, either “fundamental” or otherwise. See e.g., Conner v. State, 829 N.E.2d 21, 25
(Ind. 2005); Stevens, 770 N.E.2d at 756-57; Sanders v. State, 765 N.E.2d 591, 592 (Ind. 2002).
Stephenson argues that the State conceded that the use of the belt was properly before the
post-conviction court as a freestanding issue. He bases this contention on the State’s proposed
findings and conclusions in which the State requested the trial court to find no prejudice from
failure to object to the belt. The State cited as its reason for lack of prejudice the fact that the
issue was before the post-conviction court. Assuming that a proposed finding can under some
circumstances preclude a party from contesting its own finding, a concession as to a conclusion
of law is not binding on this Court. See Myers v. State, 233 Ind. 66, 67, 116 N.E.2d 839, 839
(1954); Green v. State, 232 Ind. 596, 597, 115 N.E.2d 211, 212 (1953). Whether the issue of the
5
belt’s use was available as a freestanding issue is a question of law. The issue is therefore fore-
closed if, as here, it was available at trial and no objection was raised.
C. Structural Error
In a variant of the claim of fundamental error, Stephenson also contends that the use of
the belt constituted structural error that per se requires a new trial. He equates the use of the belt
with an impartial judge or wrongful denial of his right to a jury trial. He cites Deck v. Missouri,
544 U.S. 622, 633 (2005) for the proposition that unnecessary restraints “almost inevitably” af-
fect the jury’s perception of the defendant and place a “thumb on death’s side of the scale” of
justice. 1 These contentions were available at trial and on direct appeal and were not preserved.
They are therefore available only to the extent they support a claim of ineffective assistance of
counsel for failure to present them.
D. Ineffective Assistance of Trial Counsel
Stephenson raises the failure to object to the belt as establishing ineffective assistance of
trial counsel. Stephenson raised no claim of ineffective assistance in his direct appeal. We agree
that this claim is therefore properly presented in post-conviction proceedings as one of ineffec-
tive assistance of trial counsel for failure to object at trial. See Timberlake, 753 N.E.2d at 597.
Under Strickland v. Washington, a claim of ineffective assistance of counsel requires the defen-
dant to show by a preponderance of the evidence that (1) counsel’s performance was below the
1
Stephenson invokes Deck without addressing the point that Deck postdated his trial and direct appeal.
The State points out that Stephenson’s trial, appeal, and post-conviction proceeding all occurred before
Deck was decided in 2005. The State notes that it found no case holding Deck is to be applied retroac-
tively, i.e. applied in post-conviction proceedings to a trial that took place before Deck was handed down.
But retroactivity is not the issue here. Stephenson makes no claim that Deck applies retroactively under
the principles set forth in Teague v. Lane, 489 U.S. 288 (1989) and Daniels v. State, 561 N.E.2d 487 (Ind.
1990). Daniels adopted for Indiana the same retroactivity principles for a “new constitutional rule of
criminal procedure” that apply to federal constitutional rules under Teague. Daniels, 561 N.E.2d at 488-
89. Under Teague, rules that place “certain kinds of primary, private individual conduct beyond the
power of the criminal law-making authority to proscribe,” and those that require the observance of “pro-
cedures that . . . are ‘implicit in the concept of ordered liberty,’” and “without which the likelihood of an
accurate conviction is seriously diminished,” may be applied retroactively. 489 U.S. at 311, 313 (internal
quotation marks and citations omitted). The only authority addressing the point holds that Deck is not
retroactive under Teague. Marquard v. Sec’y for Dep’t of Corr., 429 F.3d 1278, 1312 (11th Cir. 2005).
The requirement of a finding of necessity before restraints may be used in the penalty phase of a trial does
not rise to the level of a procedure that is “implicit in the concept of ordered liberty.” Id. (quoting
Teague, 489 U.S. at 307); Whorton v. Bockting, 127 S. Ct. 1173, 1184 (2007); see also Schriro v. Sum-
merlin, 542 U.S. 348, 352 (2004) (“This class of rules is extremely narrow, and it is unlikely that any has
yet to emerge.”).
6
objective standard of reasonableness based on “prevailing” professional norms and (2) the de-
fendant was prejudiced by counsel’s substandard performance, i.e. there is a “reasonable prob-
ability” that, but for counsel’s errors or omissions, the outcome of the trial would have been dif-
ferent. 466 U.S. 668, 687, 694 (1984); French v. State, 778 N.E.2d 816, 824 (Ind. 2002); Ste-
vens, 770 N.E.2d at 746; Lowery v. State, 640 N.E.2d 1031, 1041 (Ind. 1994).
The issues raised by Stephenson’s claim of ineffective assistance, therefore, in broad
terms are:
1) Was it substandard performance to fail to object to the use of the belt?
2) Was it substandard to fail to object to the absence of an individualized hearing on the
need for any restraint?
3) Is a showing that the failure to object produced an “inherently prejudicial” condition of
trial sufficient in itself to establish the prejudice prong of Strickland?
4) If an inherently prejudicial condition is shown but is not a per se ground for reversal,
does the defendant bear the burden of showing prejudice by the condition or must the State es-
tablish lack of prejudice?
5) What standard of proof is required of the party with the burden as to prejudice, and
was that standard met?
1. Performance of Counsel
The trial record makes no reference to the belt or to the need for restraint. There is no
clear statement of the trial court’s policy requiring restraint. There is no evidence that Stephen-
son was obstreperous or disruptive. In short, the record shows nothing to support an individual-
ized determination that Stephenson required any form of restraint at trial, and there is no
explanation in the trial record for use of the stun belt or any other restraint. The belt was not
mentioned by the parties or the court in Stephenson’s direct appeal.
a. Failure To Object to the Belt as a Tactical Choice of Trial Counsel
The law is clear that counsel’s performance is presumed effective. “[T]he defendant
must overcome the strongest presumption of adequate assistance, and judicial scrutiny is highly
7
deferential.” Ben-Yisrayl v. State, 738 N.E.2d 253, 261 (Ind. 2000); see Conner, 711 N.E.2d at
1252. Moreover, as is frequently pointed out:
There is a strong presumption that counsel rendered adequate assistance and made
all significant decisions in the exercise of reasonable professional judgment.
Counsel is afforded considerable discretion in choosing strategy and tactics, and
these decisions are entitled to deferential review. Isolated mistakes, poor strategy,
inexperience, and instances of bad judgment do not necessarily render representa-
tion ineffective.
Stevens, 770 N.E.2d at 746 (citing Strickland, 466 U.S. at 689-690; Timberlake, 753 N.E.2d at
603; Perez v. State, 748 N.E.2d 853, 854 (Ind. 2001)).
The State contends that trial counsel made a tactical decision to allow Stephenson to wear
a stun belt. The record does not support this contention. At post-conviction, Anthony Long, one
of Stephenson’s trial counsel, explicitly stated that he made a conscious decision to allow Ste-
phenson to appear before the jury in a stun belt: “I understood our choices were either [the stun
belt] or shackles and that was certainly not an acceptable alternative.” Similarly, trial counsel
Dennis Vowels testified:
Q: Do you recall any discussions with Anthony Long about the benefits of a stun
belt as opposed to being shackled?
A: I don’t recall them, but we probably talked about it. I know we were not go-
ing to let him be shackled in front of a jury. I knew that.
Q: Would you agree that a stun belt worn under the clothes would be preferable
in front of a jury to the handcuffs, the waist and the leg irons?
A: Yes.
At the time of Stephenson’s trial in 1996 and 1997, no Indiana ruling had addressed the
use of stun belts. As in Wrinkles, counsel cannot be faulted for selecting the belt over more visi-
ble shackles, given that the case law addressing the issue had largely focused on the visibility of
the restraint, and not, as Wrinkles later pointed out, on the belt’s potential effect on the defen-
dant’s demeanor and ability to participate in the defense. Wrinkles, 749 N.E.2d at 1194. Al-
though counsel explained their preference for the belt over shackles, they did not explain why
they conceded that any restraint was appropriate and failed to require any finding on the record
as to the need for restraint. As explained above, at the time of Stephenson’s trial it was well-
settled as a matter of both state law and the requirements of federal due process that no form of
8
visible restraint was permissible without an individualized finding that the defendant presented a
risk of escape, violence, or disruption of the trial. Deck, 544 U.S. at 626-27; Coates, 487 N.E.2d
at 169. Counsel’s unquestioning acceptance of the need for any form of restraint cannot be justi-
fied as a tactical decision where the decision was made without awareness of the applicable law
favorable to the defendant. Dixon v. Snyder, 266 F.3d 693, 703 (7th Cir. 2001) (“If counsel was
unaware of the statute, then his decision not to cross-examine Carlisle cannot be accorded the
same presumption of reasonableness as is accorded most strategic decisions because it was not
based on strategy but rather on a ‘startling ignorance of the law.’” (quoting Kimmelman v. Mor-
rison, 477 U.S. 365, 385 (1986))). Finally, the State’s claim of a tactical decision does not enjoy
the same support it had in Wrinkles. In Wrinkles’s case, the penalty, not guilt or innocence, was
the only real issue. The decision to challenge the belt arguably fell into the tactical range, bal-
ancing the likelihood of success against the risk of alienating the judge by challenging an an-
nounced “policy.” In Stephenson’s case, unlike Wrinkles, guilt was vigorously disputed, so that
justification for counsel’s omission is weakened, and in any event no such tactical consideration
was advanced by counsel in post-conviction.
b. Failure To Object to a Visible Restraint as Substandard Performance
Even if failure to object to the belt was not justified as a tactical decision, the issue re-
mains whether it was below professional norms to fail to object to the use of any restraint or to
the absence of any finding of necessity for restraint. The law regarding use of stun belts was not
settled at the time of Stephenson’s trial, and counsel are not ordinarily found deficient for failure
to anticipate a change in the law. Smylie v. State, 823 N.E.2d 679, 690 (Ind. 2005); Fulmer v.
State, 523 N.E.2d 754, 757-58 (Ind. 1988).
Under the law at the time of Stephenson’s trial, an ineffective assistance claim based on
failure to object to restraints required the restraints to be visible. Failure to object to restraints is
not substandard performance where the jury is unaware of the restraints. 2 But the Seventh Cir-
2
See Evans v. Dir. of the Cal. Dept. of Corr., No. C-03-2498 MMC, 2005 U.S. Dist. LEXIS 9360, at *23
(N.D. Cal. 2005) (“No jurors were present because the charges were tried to the court.”); State v. DuPree,
820 N.E.2d 560, 566 (Ill. Ct. App. 2004) (“There [was] not a scintilla of evidence that the jury could have
been aware of the stun belt.”); Walker v. State, 602 S.E.2d 351, 355 (Ga. Ct. App. 2004) (finding the de-
fendant “failed to present any evidence showing the jury would have known he was in custody due to his
clothing” and “failed to identify any of the individuals he claims were in the hallway” when he was taken
from the elevator to the holding cell); Lytle v. Armontrout, No. 89-0263-CV-W-5-P, 1990 U.S. Dist.
9
cuit has held that failure to object to restraints that are “readily visible” is substandard perform-
ance of counsel. Roche v. Davis, 291 F.3d 473, 483 (7th Cir. 2002) (quoting Fountain v. United
States, 211 F.3d 429, 435 (7th Cir. 2000)). The Seventh Circuit found “unreasonable” 3 our con-
clusion that counsel was not ineffective for failure to object to shackling and failure to take steps
to prevent the jury from viewing the shackles. Id.
Although Roche addressed shackling, we think its reasoning is equally applicable to a
stun belt. The use of a stun belt, if perceived by the jury, produces all of the results that shack-
ling does. It sends a signal that the defendant may be dangerous and thereby impairs the pre-
sumption of innocence; it interferes with the defendant’s communication with his attorney; and it
has the same effect on the dignity of the process. Indeed, some courts have concluded that a stun
belt, if perceived by the jury, “may be even more prejudicial than handcuffs or leg irons because
it implies that unique force is necessary to control the defendant.” United States v. Durham, 287
F.3d 1297, 1305 (11th Cir. 2002) (quoting State v. Flieger, 955 P.2d 872, 874 (Wash. Ct. App.
1998)). Even if the jury is unaware of the belt, there remain the concerns that a stun belt “could
disrupt a different set of a defendant’s constitutionally guaranteed rights.” Id. First, “[a] stun
belt seemingly poses a far more substantial risk of interfering with a defendant’s Sixth Amend-
ment right to confer with counsel than do leg shackles.” Id. Second, the device poses a greater
threat to the defendant’s Sixth Amendment and due process rights to be present and participate in
his defense because “[i]t is reasonable to assume that much of a defendant’s focus and attention
when wearing one of these devices is occupied by anxiety over the possible triggering of the
belt.” Id. at 1305-06. These are in substance the same points emphasized by the majority in
Wrinkles in banning the belt under Indiana state law. 749 N.E.2d at 1194. In sum, in the ab-
sence of an explanation for counsel’s failure to object to a stun belt that is “readily visible,” that
failure is substandard performance.
LEXIS 2634, at *12 (W.D. Mo. 1990) (finding “no indication that the jury knew of or took into consid-
eration the fact” that the defendant was restrained or in jail garb). Most of these cases dealt with shackles,
not a stun belt.
3
The Antiterrorism and Effective Death Penalty Act of 1996 (AEDPA), 28 U.S.C. section 2254(d)
(2000), requires that a state court’s interpretation of a federal constitutional doctrine be “unreasonable”
before the federal court may entertain a petition for habeas corpus based on a challenge to that state
court’s ruling.
10
c. Was the Belt “Readily Visible?”
Whether the belt was “readily visible” is critical to this branch of the inquiry. Roche, 291
F.3d at 483 (quoting Fountain, 211 F.3d at 435). Some courts have used the term “visible” to
make the point that shackles or jail garb must be perceived by the jury. See, e.g., id.; Deck, 544
U.S. at 631-32. The broader notion is that it is prejudicial for jurors to be “aware” that a defen-
dant is being restrained or appearing in jail garb. Fountain, 211 F.3d at 435. We see little sig-
nificance to whether the jurors learned of the belt by seeing the restraint or by being informed of
it. In either case the defendant is branded a dangerous individual.
The post-conviction court made no specific finding as to the jurors’ awareness of the belt.
However, the post-conviction record demonstrates that several jurors knew that Stephenson wore
the belt during trial and recognized it for what it was. According to one juror’s affidavit:
During the trial, I became aware that John Stephenson was wearing a stun belt. I
could see he had what appeared to be a rectangle shaped box attached to his lower
back, underneath his shirt. I had seen a television show sometime previously to
being selected for this jury that described what a stun belt was and what its pur-
pose was. I have been hit by 220 volts and know what effect that amount of
power has on a person. I believed the stun belt was to control John Stephenson’s
behavior.
Some apparently learned of the restraint from other jurors. A second juror stated in her affidavit:
I was aware that John Stephenson was wearing a device that would prevent him
from running out of the courtroom. During the trial one of the male jurors men-
tioned to me that Mr. Stephenson was wearing this device. I was aware that he
was wearing something that controlled his behavior.
Similarly, a third juror affirmed, “During the trial, I recall that Mr. Stephenson had some type of
restraining device on him. I do not recall what the device looked like, or when or where I real-
ized he [was] wearing one.” A fourth juror was asked in deposition by Petitioner’s counsel, “Did
you know that John Stephenson had worn a stun belt through his trial?” The juror responded:
I would say, yes, that I did because he wasn’t handcuffed, and naturally I assumed
that he had that on because there was like a, you know, he wore like a loose shirt,
basically like what you have on, loose, and there was a bulge back there in the
back, so that’s how I knew that.
It was not clear that all jurors were aware of the belt. One testified in deposition offered in post-
conviction proceeding:
11
Q: Were you aware that John Stephenson was wearing a stun belt? Do you know
what I mean by that, first of all?
A: I was aware of that, and I’m not sure at what time, you know, what point I was
aware of that.
Q: But you know what I mean—
A: Yes.
Q: --by a stun belt?
A: Yes, I do.
Q: Okay, so go ahead if you hadn’t finished your answer.
A: Well, I’m not really sure exactly if it was after the trial that, that I heard about
that or – I’m thinking it was afterwards, but I’m not really sure.
Q: Did you notice it when you—
A: No.
Q: --walked by, walked past from the rear?
A: No, I never did notice it.
Even if at least one juror did not observe or recall the belt, on this record we think Stephenson
has established by a preponderance of the evidence that the belt was “readily visible” to the jury.
d. Summary of the Performance Prong of Strickland.
The record at post-conviction did not explore more fully counsel’s reasoning in accepting
the belt without question. For the reasons explained below in discussing the prejudice prong of
the ineffective assistance claim, it may have been that counsel recognized the futility of an objec-
tion and, as in Wrinkles, avoided confronting the trial judge on an issue they deemed ultimately
unsuccessful. On this record, however, failure to object to the belt cannot be justified as a tacti-
cal decision in Stephenson’s case because the explanation offered by counsel for their decision
boiled down to a failure to know the applicable law. Equally importantly, counsel failed to insist
upon a finding as to the need for restraint. Prevailing norms at the time of Stephenson’s trial re-
quired counsel to object to visible restraints where there is no evidence suggesting escape, vio-
lence, or disruptive behavior. We agree with Roche that failure to object to the belt without a
showing of justification fails to meet prevailing norms. No justification for the omission is
established in this record. We therefore conclude that Stephenson’s counsel’s failure to object to
the belt meets the first prong of Strickland.
12
2. Burden of Proof and Standard of Proof of Prejudice from Failure To Object to “In-
herently Prejudicial” Practices
We turn now to the issues of burden of proof and standard of proof of prejudice from fail-
ure to object to the belt.
a. The Strickland Formulation of the Prejudice Prong
Strickland established that prejudice from substandard performance of counsel requires a
showing by the petitioner that there was a “reasonable probability” of a different result if counsel
had met professional norms. We sometimes express the standard for prejudice from the failure
to object as requiring a reasonable probability that the objection would have been sustained. See,
e.g., Wrinkles, 749 N.E.2d at 1192 (citing Timberlake, 690 N.E.2d at 259). The standard is more
precisely stated as prejudicial failure to raise an objection that the trial court would have been
required to sustain. Otherwise stated, if the trial court overruled the objection, it would have
committed error, and the error would have had a prejudicial effect. See, e.g., Spinks v. McBride,
858 F.Supp. 865, 877 (N.D. Ind. 1994) (“In order to establish ineffective assistance of counsel
for failure to object, it must be shown that the trial court would have been required to sustain the
objection had it been made.” (citing Hill v. State, 442 N.E.2d 1049 (Ind. 1982))); Kimble v.
State, 451 N.E.2d 302, 306 (Ind. 1983) (“Before trial counsel’s failure to enter an objection may
be regarded as ineffective representation, the petitioner must show that had a proper objection
been made, the trial court would have had no choice but to sustain it.”). In most cases there is no
practical difference between these two formulations.
b. Wrinkles v. State
Wrinkles, like Stephenson, asserted that trial counsel were ineffective for not objecting to
the court’s ordering him to wear a stun belt at trial. Wrinkles contended that there was no reason
to require restraint. Wrinkles, 749 N.E.2d at 1192. We recognized that a defendant has the right
to appear before a jury unrestrained unless restraint is necessary for a trial without incident. Id.
at 1193 (citing Bivins v. State, 642 N.E.2d 928, 936 (Ind. 1994)). We also acknowledged that
this right springs from the “basic principle of American jurisprudence that a person accused of a
crime is presumed innocent until proven guilty beyond a reasonable doubt” and that in order for
the presumption to be effective, a defendant must appear unrestrained to avoid the appearance
that guilt was a “foregone conclusion.” Id. We reiterated that the reasons for requiring a defen-
13
dant to be restrained before a jury must be placed on the trial record. Id. (citing Roche v. State,
690 N.E.2d 1115, 1123 (Ind. 1997); Coates v. State, 487 N.E.2d 167, 169 (Ind. Ct. App. 1985),
overruled on other grounds by Hahn v. State, 533 N.E.2d 618 (Ind. Ct. App. 1989)). Deck has
now made clear that the Fourteenth Amendment imposes the same requirement. The majority in
Wrinkles acknowledged that a policy requiring all defendants to wear restraints would “not
likely withstand appellate scrutiny if the issue were presented.” 749 N.E.2d at 1195. We none-
theless held in Wrinkles that counsel’s failure to object to the stun belt’s use did not constitute
ineffective assistance in that case. We based that ruling on a lack of prejudice, without address-
ing whether counsel’s performance was substandard. Id. at 1195, 1196. The reason for that
holding was that the trial court’s “policy” dictated use of restraint and any objection to the belt
would not have prevailed. Id. The failure to require an individualized determination was not
asserted as a ground of ineffective assistance in Wrinkles.
We agree with the State that Stephenson’s case presents many similarities to Wrinkles.
The two are not identical, however, because this record shows no inflexible “policy” of the trial
court. Trial counsel Long testified that in his experience, the trial judge typically deferred to the
sheriff’s security decisions. The sheriff, in turn, cited concerns in transporting defendants from
jail to the courtroom as the basis for requiring restraint. These concerns did not seem to relate
directly to use of the belt at trial as opposed to its use in transit, but that issue was not explored at
the post-conviction hearing. 4 The officers in charge of security at the trial testified that they had
no knowledge of any incidents that would demonstrate a need for Stephenson to wear a stun belt.
4
At the post-conviction hearing, Sheriff Hargrove testified as follows:
Q: How was the decision made that John was going to wear this belt?
A: We just knew that he had to get him in and out of the building without the jury seeing
him in shackles or handcuffs and we elected to—it was fairly, in our opinion at the time,
it may not have been the new technology, but it was something that had just come to our
attention, and we decided we were going to utilize the belt for that reason.
Q: When you—just so I’m sure—I know when you say “we” and “it was our decision,”
you mean you and who else?
A: The chief deputies
Q: The chief deputies. You didn’t consult with the judge?
A: Well I, I don’t know whether—I don’t recall specifically speaking to the judge, and I
may have. But I know that we’d over the years, we’d had a lot of discussions with Judge
Campbell regarding the – getting the inmates in and out of the jail without the, without
the jury seeing them shackled or handcuffed.
14
Sheriff Bruce Hargrave; Charlie McCracken, the sergeant in charge of security at Stephenson’s
trial; Jerry Ash, the deputy of security; police officer Robert Irvin; and Jonetta Baker, jail com-
mander for the Warrick County Police Department, all testified at post-conviction that to their
knowledge Stephenson posed no security threat and had exhibited no behavior that would dem-
onstrate a specific need for a restraining device at trial. There was also extensive post-conviction
testimony from the sheriff’s office and others that Stephenson conducted himself as a “gentle-
man” throughout the arrest and trial. He had turned himself in in response to reports that law
enforcement was looking for him in connection with the murders and made no effort to escape
either before or during trial. From the sheriff’s testimony, it appears that no one gave careful
consideration to the need for any restraint while Stephenson was in the courtroom at his trial.
Rather, the need was assumed, and the only concern voiced by the sheriff was whether the jury
would see the restraint. Although the record shows that the trial judge followed the recommen-
dation of the sheriff, it does not indicate that either had an inflexible policy of requiring restraint,
or, if so, to what cases it applied.
c. The Effect of Deck
As already explained, Roche found prejudice from failure to object to restraints, and
Wrinkles rejected it. Both cases used the Strickland standard of a “reasonable probability” of a
different result. Deck now teaches that prejudice from shackling is governed by the “inherently
prejudicial” standard for constitutional error under Chapman v. California, 386 U.S. 18, 24
(1967). 544 U.S. at 635. Rather than placing the burden on the defendant-petitioner to show a
reasonable probability of a different result, an inherently prejudicial error requires the State to
establish, beyond a reasonable doubt, that the error has no effect on the ultimate resolution of the
trial. Deck, therefore, establishes a stronger presumption of reversible error from failure to ob-
ject to shackling than the Seventh Circuit applied in Roche or we applied in Wrinkles. We think,
however, that Deck does not relieve the petitioner in post-conviction proceedings from the bur-
den of establishing both the substandard performance and prejudice prongs of an ineffective as-
sistance claim.
Before Deck was decided, most claims of ineffective assistance of counsel for failure to
object to shackles or jail garb were rejected based on lack of prejudice, even though the court
explicitly or implicitly assumed substandard performance. We find only one published opinion
15
that has addressed this issue after Deck. 5 The Eleventh Circuit, in Marquard v. Secretary for
Department of Corrections, 429 F.3d 1278 (11th Cir. 2005), addressed a post-conviction claim of
ineffective assistance for failure to object to shackling at the penalty phase of a Florida death
penalty case. There was no shackling at the guilt phase, but the defendant appeared in shackles
at the penalty phase, and there was no inquiry establishing the necessity of shackles. The trial
and direct appeal had occurred before Deck was handed down. In 2002, the Florida Supreme
Court denied post-conviction relief. Like most pre-Deck decisions, that ruling addressed the
claim of ineffective assistance for failure to object to shackling solely by finding no prejudice.
Marquard v. State, 850 So.2d 417, 431 (Fla. 2002). The primary issue in federal habeas, there-
fore, was the retroactivity of Deck’s holdings that (1) unnecessary shackling at the penalty phase
violated the Fourteenth Amendment and (2) its harmlessness must be established beyond a rea-
sonable doubt. The Eleventh Circuit rejected Marquard’s claim of ineffective assistance for
three reasons. The first two grounds—Deck is not retroactive and counsel were not ineffective
for failure to anticipate Deck’s extension of the shackling ban to the penalty phase—do not re-
solve Stephenson’s claim because Stephenson appeared in the stun belt at the guilt phase of his
trial. The Eleventh Circuit also offered a third ground. The habeas court concluded that in a di-
rect appeal Deck shifted the burden to the state to prove harmlessness of shackling without a
specific-needs inquiry, but Deck “did not address, much less alter, the burden and different re-
quired prejudice showing on Marquard’s IAC shackling claim.” Marquard, 429 F.3d at 1313.
As a result, the defendant had the burden to establish prejudice under the Strickland standard.
5
In re Jasso, No. H029756, 2006 Cal. App. LEXIS 1382, at *25 (Cal. Ct. App. 2006), was initially pub-
lished but was ordered unpublished by the California Supreme Court. 2006 Cal. LEXIS 14200 (Nov. 29,
2006). Jasso was not a death penalty case and therefore addressed prejudice only as to the finding of
guilt. The defendant, without objection from his counsel, appeared in shackles wearing a white prison
jumpsuit with bold writing on the back identifying him as a prisoner. Id. at *10. In reviewing the denial
of habeas alleging ineffective assistance, the court concluded that failure to object to shackles and jail
garb fell “below an objective standard of reasonableness . . . under prevailing professional norms.” Id. at
25 (internal quotation marks and citation omitted). The court also found the omission prejudicial, reason-
ing that the effect of the jumpsuit and shackles on the jury’s perception of the defendant and the effect on
the defendant’s demeanor created a reasonable probability of a different result. Id. at *32. Relying on
Deck, the court concluded that “had counsel objected and the objection erroneously been denied, our re-
view would have governed by the stringent harmless-error standard for federal constitutional errors . . .
and the state would have had to prove beyond a reasonable doubt that the shackling and jumpsuit did not
contribute to the verdict.” Id. at *24-*25.
16
The law at the time of Marquard’s direct appeal would not have supported either his
claim of error for unnecessary shackling at the penalty phase or a claim of ineffective assistance
of counsel for failure to object to shackling at the penalty phase. It is clear that lack of retroac-
tivity does not bar Stephenson’s ineffective assistance claim at the guilt phase. The novel propo-
sition in Deck was that shackling during the death penalty phase violates the Fourteenth
Amendment. At the time of Stephenson’s trial, the points relevant to the guilt phase were al-
ready established law: unnecessarily shackling at the guilt phase of trial (or at trial in a non-
death case) violates the Fourteenth Amendment; jail garb is “inherently prejudicial”; and the
State bears the burden of proving beyond a reasonable doubt that an inherently prejudicial prac-
tice had no effect on the determination of guilt or the penalty. In short, the principles set forth in
Deck that are relevant to Stephenson’s case were all established by earlier precedent. Marquard
therefore did not directly address the situation we have before us, where Stephenson’s stun belt
at the guilt phase required a specific-needs inquiry under the law in place at the time of his trial
and direct appeal. We nevertheless think Marquard is correct that Strickland governs the preju-
dice prong of Stephenson’s claim of ineffective assistance for failure to object to shackling.
d. Prejudice
If the issue of shackling at the guilt phase had been preserved at trial and raised on direct
appeal, we would have been compelled to address that claim under the standard of Chapman v.
California, 386 U.S. 18 (1967), and the burden would have been on the State to establish lack of
prejudice beyond a reasonable doubt. But Strickland v. Washington places the burden of estab-
lishing prejudice on the petitioner in an ineffective assistance claim. The critical issue is there-
fore whether Stephenson established a reasonable probability that his counsel’s failure to object
to Stephenson’s stun belt affected the result in either the guilt or penalty phase.
The State principally relies on Stevens v. State, 770 N.E.2d 739 (Ind. 2002), as establish-
ing lack of prejudice from Stephenson wearing the belt at trial. That case, however, is inapplica-
ble here because at least several of Stephenson’s jurors were aware of the belt. In Stevens, we
found no ineffective assistance for counsel’s failure to object to the defendant’s wearing a stun
belt. The defendant had conceded that none of the jurors were aware that he was wearing the
stun belt but argued that their perception of him was affected by his behavior and appearance as
a result of the restraint. Id. at 757. The defendant argued that he appeared “withdrawn, subdued,
17
and unusually silent, which may have influenced [the jury] to recommend the death penalty.” Id.
Applying a pre-Deck standard of review, we affirmed the post-conviction court’s finding that
under the facts of that case the outcome would not have been different if there had been an objec-
tion. Id.
In French v. State, supra, we held that a defendant who wore handcuffs, shackles, and jail
clothing at his habitual offender proceeding could not show he was prejudiced by counsel’s fail-
ure to object because “it seem[ed] clear that the result of the proceeding was not affected by
counsel’s performance.” 778 N.E.2d 816, 826. The issue in the habitual offender proceeding
was whether the defendant had two prior felony convictions. He had already been convicted by
the same jury of the underlying offenses, so it was no surprise that he was incarcerated. More-
over, as to the only issue—his two prior convictions—the evidence was “clear.” Id. at 826.
Unlike French, Stephenson’s failure to prove a reasonable probability of a different result
does not rest on essentially indisputable evidence that establishes the merits of his conviction and
sentence. 6 Rather, Stephenson has shown no prejudice because he has not shown a reasonable
6
There was less than a conclusive amount of physical evidence connecting Stephenson to the crime. See
Stephenson v. State, 742 N.E.2d 463 (Ind. 2005). Much of the evidence against Stephenson came from
the testimony of Dale Funk, who claimed to be with Stephenson at the time of the killings, and Brian
Mossberger, who produced the murder weapon when the police questioned him about the murders. The
testimony of these two witnesses was largely consistent. It is clear that if Funk’s and Mossberger’s testi-
mony is credible, the guilty verdict is sustained. The following is their trial testimony in chronological
order of the events.
Funk testified that approximately two days before the killings he and Stephenson visited the mo-
bile home occupied by Southard and her fiancé Troy Napier. According to Funk, Stephenson entered the
unoccupied home and returned with ammunition. The three victims were killed on March 28, 1996.
Funk testified that he and Stephenson arrived at Mossberger’s house the evening of March 28. Stephen-
son went in to another room to talk with Mossberger while Funk remained at the kitchen table. Accord-
ing to Mossberger, he and Stephenson were talking when the Tylers and Southard briefly pulled into
Mossberger’s driveway in a pick-up truck. Mossberger testified that Stephenson said, “there goes Jay and
I’ve got to catch him.” Stephenson, 472 N.E.2d at 470. Funk testified that while he was sitting at the ta-
ble, Stephenson returned to the kitchen “walking pretty fast” and called to Funk, “if you’re coming, come
on.” According to Funk, Funk accompanied Stephenson in Stephenson’s car as Stephenson pursued the
Tyler truck through Warrick County rural roads until it stopped at an intersection. Id. Stephenson pulled
up behind the truck and Tyler opened his front door and leaned out, looking back at Stephenson. Id. At
that point Stephenson grabbed his SKS assault rifle, left the car, and began firing at the Tyler truck. Id.
Stephenson then returned to the car, drove around the corner, stopped the car, and proceeded on foot in
the direction of the Tyler truck. Stephenson returned a few minutes later and told Funk, “You breathe a
word of this and I’ll kill you.” Id. at 470-71. Mossberger testified that after Stephenson and Funk re-
turned to Mossberger’s house, Stephenson was holding a knife with “red smears” on the blade and told
Mossberger that “Jay, Kathy, and Brandy are no more.” Id. According to Mossberger, Stephenson
washed his knife in the kitchen sink and instructed Mossberger to “do something with the SKS; get rid of
18
probability that an objection, if made, would have been successful in the trial court, or would
have produced the basis for a successful appeal.
Stephenson contends that wearing the stun belt “interfered with his right to communicate
with counsel and participate in his own defense.” The arguments of the parties and the post-
conviction court’s findings with respect to the stun belt were focused on this Court’s opinion in
Wrinkles. The findings with respect to the belt in their entirety were:
At the time of the trial of this matter, it was not impermissible for the elec-
tronic restraint belt to be worn by Stephenson during the trial. The belt was never
activated while on Stephenson, he was not prevented from assisting counsel while
wearing the belt, and the same was a suitable and discreet alternative to other
methods of restraint for Stephenson while he was on trial on multiple counts, in-
cluding three counts of murder, in this case. The use of the belt was accepted by
Stephenson and his counsel without argument or objection on the record during
the trial proceedings.
Insofar as ineffective assistance claims were based on the belt, the post-conviction court found:
As previously included in other findings of fact, it was the Warrick
County Sheriff’s Department who obtained and placed the electronic restraint on
Stephenson and beneath his clothing. Judge Campbell did not order this to be
done, and trial counsel did not argue or object to this restraint, as compared to
other forms of restraint which may have been required otherwise. At the time the
Defendant was required to wear the electronic restraint, he was on trial for triple
murder, and facing a death penalty request by the State of Indiana. The law had
not yet ruled the use of such devices on a criminal Defendant to be improper or
violative of a Defendant’s rights. In view of all the circumstances surrounding
this case and this Defendant, including the extremely violent nature of the mur-
ders committed, the use of the electronic restraint belt was warranted, reasonable
and necessary under these circumstances.
it; make it gone.” Id. Funk also testified that he was present when Stephenson handed the weapon to
Mossberger and told him to get rid of it. Id. Mossberger testified he buried the rifle and ammunition in
the woods the next day but retrieved them two days later and placed them in his garage. Id. When the
police officers arrived at Mossberger’s house to question him later that day, he related his version of the
events of the day of the murders and showed the officers the rifle but not the ammunition. Id.
On the morning of Friday, March 29, police found the bodies of the Tylers and Southard dead
from gunshot wounds and stab wounds. Id. The forensic evidence was: (1) the fatal bullets were those
fired from the SKS assault rifle that Stephenson owned but was in Mossberger’s possession two days af-
ter the murders; (2) spent shell casings at the scene matched ammunition discovered in Southard’s and
Napier’s mobile home; and (3) Funk’s shoe prints, but not Stephenson’s, were found at the mobile home.
No other physical evidence connected Stephenson to the murders. The knife used in the murders was
never recovered. Id.
19
The post-conviction court concluded that “counsel for Defendant was not ineffective because of
failure to object to the use by the sheriff of the electronic restraint device on Stephenson during
trial.”
We take these findings and conclusions as a finding that the trial court would have over-
ruled any objection to the belt that counsel had raised. These three murders were contended by
both the defendant and the prosecution to have been related to organized drug activity. The
murders appeared to have been premeditated and had characteristics of an assassination. There
was testimony that the defendant had threatened to kill a critical witness. Under all these cir-
cumstances, the post-conviction court’s finding is not clearly erroneous. Indeed, given the state
of the law in 1996, we think it plain that the trial judge would have followed the sheriff’s rec-
ommendation and ordered that the belt be deployed at the guilt phase even if defendant’s counsel
had objected and required a hearing and findings as to the need for its use. Because at the time
shackling at the penalty phase was less controversial, the same result would have obtained at the
penalty phase. Given the fact-sensitive nature of that order, under the circumstances the issue
would not have presented a persuasive ground for appeal. The standard of review on appeal
would have been abuse of discretion. Bivins v. State, 642 N.E.2d 928, 936 (Ind. 1994); Evans v.
State, 571 N.E.2d 1231, 1238 (Ind. 1991). Accordingly, Stephenson has not established a rea-
sonable probability of a different result from counsel’s performance with respect to the belt at the
guilt phase.
C. Appellate Counsel’s Failure To Raise an Issue over the Belt
Although Stephenson asserts claims of ineffective assistance of appellate counsel, the
failure to raise any issue of the belt on direct appeal is not among them. Stephenson contends
that appellate counsel were unaware of the belt and correctly points out that there is no mention
of the belt in the trial record. He claims that had appellate counsel known of the belt’s use the
issue would have been raised on appeal. We think that this circumstance raises no issue of appel-
late ineffectiveness.
The performance of appellate counsel is not substandard for failure to assert an issue on
appeal if the issue is not revealed by the record and there is no evidence suggesting that appellate
counsel were otherwise made aware of it. We have observed that “the Supreme Court of the
United States has never suggested that [appellate] counsel must look outside the record for pos-
20
sible claims of error for the performance to be constitutionally effective.” Woods v. State, 701
N.E.2d 1208, 1221 (Ind. 1998). We agree with those courts that have refused to impose such a
burden on appellate counsel. Id. at 1221-22 (citing Kitt v. Clarke, 931 F.2d 1246, 1249-50 (8th
Cir. 1991)). Thus, we concluded in Woods that “[b]ecause there is no constitutional requirement
for appellate counsel to search outside the record for error, an ineffective assistance of appellate
counsel claim that is in substance a trial counsel claim requiring extrinsic evidence may be dead
on arrival.” Id. at 1222. A corollary of this conclusion is that it is incumbent upon trial counsel
to communicate to appellate counsel any matters outside the record that are appropriate for direct
appeal. If supplementation of the record is required, Indiana procedure allows for it, 7 but we
think the performance of trial counsel, not appellate counsel, is the only issue when nothing in
the record suggests the issue.
In sum, to the extent Stephenson presents a claim for post-conviction relief based on the
use of the stun belt at trial, that claim is one of ineffective assistance of trial counsel for failure to
object, failure to preserve the issue for appeal, and failure to communicate the circumstance to
appellate counsel. All of these omissions lead to the same result: they had no effect on the ulti-
mate result of the trial.
II. Other Ineffective Assistance of Counsel Claims
Stephenson raises a number of claims of ineffective assistance that are unrelated to the
stun belt.
A. Failure To Impeach Witness
At trial, Stephenson contended that three witnesses, his ex-wife, Dawn Krantz; Krantz’s
neighbor, Roxanne Lester; and Stephenson’s friend, Julie Girtman, established that he could not
have committed the murders in Warrick County between 9:50 and 10:00 pm on March 28 as al-
leged by the State. 8 The claim of ineffective assistance turns on failure to impeach Crystal
7
See Hatton v. State, 626 N.E.2d 442 (Ind. 1993); Davis v. State, 267 Ind. 152, 368 N.E.2d 1149 (Ind.
1977).
8
Krantz, testified that Stephenson had been with her the night of the murders. Specifically, she testified
that she had been at Lester’s house in Rockport when Stephenson arrived at approximately 10:45 or 10:50
pm that Thursday evening. Lester also testified that Stephenson “came over about 10:30.” Girtman testi-
fied that Stephenson had stopped by her house in Rockport for about twenty-five minutes shortly after 10
pm that night. The distance in miles or driving time between Rockport, in Spencer County, and the site
where the victims were found does not appear in the record.
21
Carter, a witness who contradicted Krantz on a point unrelated to Stephenson’s alibi. Stephen-
son argues this undermined Krantz’s credibility on the alibi. Krantz testified that Stephenson
took her to work the following Sunday, March 31, 1996, and that she did not see him again until
after he surrendered to the authorities on April 3, 1996. Whether Krantz and Stephenson saw
each other between noon on March 31 and April 3 appears to be insignificant in itself. However,
Carter, a friend and co-worker of Krantz, testified to having seen Krantz and Stephenson together
on the night of March 31. Stephenson contends that his attorneys failed to impeach Carter and,
as a result, rendered ineffective assistance. The post-conviction review court rejected this claim,
and so do we.
Carter’s initial testimony was consistent with Krantz’s. In her first appearance on the
stand, Carter testified that she had not seen Stephenson from the day of the killings through the
end of March. After her initial testimony, Carter called defense counsel Long and told him that
she had in fact seen Stephenson on March 31 in Owensboro, Kentucky, which is on the Ohio
River across from Spencer County, Indiana. Long notified the court, and Trooper Marvin Heil-
man immediately interviewed Carter. In a recorded statement, Carter said that she had not seen
Stephenson on March 31, which was consistent with her prior testimony. When recalled by the
State, however, Carter testified that she had called attorney Long to change her story because she
“thought the truth should come out.” Carter then testified that she and Krantz had seen Stephen-
son in Owensboro the night of March 31. Carter testified that Stephenson was driving Girtman’s
car and flagged Carter and Krantz down as they were driving home from Wal-Mart. 9 Carter
pulled over, and Stephenson and Krantz exited their vehicles and spoke while Carter remained in
her car. Carter could not hear what was said between the pair but testified that after they spoke
they both got into Carter’s car and told her to drive to a gas station in Owensboro. En route
Krantz and Stephenson talked, but Carter remembered, “[j]ust that [Stephenson] was upset about
something about a gun.” At the gas station, Krantz and Stephenson left the car, and Krantz re-
turned to the car about ten minutes later. According to Carter, Krantz told her “not to say noth-
ing.”
Stephenson contends that Krantz’s credibility was “crucial to his defense” and was un-
dercut by Carter’s testimony. Stephenson contends that trial counsel was ineffective for failure
9
Carter testified that she and Dawn Stephenson had worked the same shift on Sunday, March 31, 1996,
and upon getting off work around 10 pm, Carter drove the two of them to Wal-Mart in Owensboro.
22
to impeach Carter with the prior inconsistent statement given in her recorded interview with Of-
ficer Heilman. It is true that Carter’s final version contradicted Krantz on a point unrelated to
Stephenson’s alibi. But failure to impeach Carter with the taped statement she made to Trooper
Heilman did not render counsel ineffective. First, the jury had already heard Carter’s earlier de-
nial of any contact with Stephenson after the killings. The same statement to Heilman added
nothing to the inconsistency already in the record. Second, in an eight-month trial, some over-
sights are inevitable, and this episode in isolation does not render counsel’s overall performance
substandard. Smith v. State, 272 Ind. 216, 218, 396 N.E.2d 898, 900 (1979). Third, in Carter’s
second appearance on the stand, she elaborated on her recollection of the conversation between
Krantz and Stephenson in the car in Owensboro. She said she heard Stephenson refer to the gun
while talking to Krantz and say, “I loaned it to a buddy, and all hell’s broke loose.” This sup-
ported Stephenson’s basic defense that others had been the killers. For that reason, embracing,
rather than attacking, Carter’s account was a reasonable judgment on the part of the defense. For
all these reasons, failure to introduce the statement to Heilman was neither substandard nor
prejudicial.
B. Lack of Knowledge of Law Regarding Stephenson’s Statement to Police
Stephenson gave a voluntary statement to the police two days after the murders. Ste-
phenson argues that he was denied the effective assistance of counsel when counsel failed to at-
tack or mitigate the effect of this statement due to counsel’s mistaken belief that the statement
would not be admitted into evidence. The statement is in itself not particularly significant. The
defense’s theory of the case was that on the night of the killings Stephenson went first to Moss-
berger’s house in Warrick County, next saw the victims at a Circle S convenience store, and then
proceeded to his home in Rockport in adjacent Spencer County. In a police interview on March
30, 1996, Stephenson had given the officers a different sequence. He said that he saw the victims
at the Circle S convenience store, then drove by Mossberger’s house, and on to Rockport.
The defense theory obviously conflicted with Stephenson’s statement to police as to the
sequence of these events. As Stephenson sees it, the conflict was damaging to his credibility.
Stephenson contends that at the time of the statement he was intoxicated and for that reason mis-
taken about the chain of events. He contends that expert testimony could have bolstered this
claim. At post-conviction Stephenson produced Dr. Robert Smith, an expert in chemical de-
23
pendency, who testified that in his opinion Stephenson’s intoxication on March 30 impaired Ste-
phenson’s ability to recall the sequence of events on the night of the murders. Stephenson now
contends counsel mistakenly believed that the intoxication would preclude admitting his state-
ment to police as evidence at trial. He argues that if counsel had understood that the statement
was admissible, they should have prepared to mitigate the harm of the statement by challenging
its reliability through the use of an expert such as Dr. Smith.
At post-conviction Stephenson pointed to two legal errors in his counsel’s conclusion that
the statement to police was inadmissible. Long testified that he “should have” challenged the
statements based on intoxication. This seems unlikely to have succeeded. Generally, intoxica-
tion or mental impairment do not render a statement made by a defendant involuntary per se.
Crain v. State, 736 N.E.2d 1223, 1232 (Ind. 2000). These factors go to the weight of the state-
ment, not to its admissibility. Id. The evidence shows Stephenson was drinking beer, but no-
where near the stage of intoxication required to exclude the statements. 10
Trial counsel Vowels offered a second ground for his belief that the statement was inad-
missible. He testified at post-conviction review that counsel thought the tape of Stephenson’s
statement had been lost and that “under the Rules of Evidence, 8, 900 series, they couldn’t au-
thenticate the transcript. So up to that point, we had a shot of winning because they couldn’t ac-
count for Mr. Stephenson’s whereabouts and all they had was Mr. Mossberger and Mr. Funk’s
say so that John Stephenson was the shooter.” Under Boyd v. State, 430 N.E.2d 1146, 1148
(Ind. 1982), however, the interrogating officer’s testimony that the transcript accurately reflected
the conversation is sufficient for admissibility of the transcript even though the tape was lost.
10
Detective Gary L. Gilbert and Investigator Michael E. Hildebrand, both of the Indiana State Police,
took the statement. They testified at trial outside the presence of the jury and at post-conviction review
that Stephenson was drinking beer as he gave his statement to the officers on March 30, 1996. Both offi-
cers, however, testified that Stephenson did not appear intoxicated. Trooper Gilbert testified that Ste-
phenson “seemed to be aware of time and place. He seemed to understand our questions. And I didn’t
detect any kind of slurred speech or anything of the nature—any kind of impairment.” Likewise, Officer
Hildebrand testified that “he appeared to be coherent, understood our questions, and answered those in a
relevant way.” Furthermore, at post-conviction review, Trooper Gilbert testified that he had received
training over the years as a trooper concerning the recognition of an intoxicated individual, and that in his
opinion Stephenson “seemed to be in full control” and that as he understood it, Stephenson gave a truly
voluntary statement on his part. Gilbert also testified that, although it is unusual for a suspect to give an
interview while he is drinking, “in this particular case [Stephenson] indicated that it’d make him feel
more relaxed and I felt like he might be more open and honest with us if we’re not going to try and make
him defensive and tell him what he can and cannot do in his own home.”
24
We agree with Stephenson that his trial counsel’s faith in excluding the statement was
misplaced. But we do not agree that this error had any significant effect on the trial. The evi-
dence was admissible whether or not counsel correctly assessed its admissibility. The only issue
is whether it could have been effectively contested by expert testimony. The claim that beer can
cause memory to be somewhat impaired, particularly as to details that seem insignificant, is one
that jurors can assess. We think there is no reasonable probability that failure to produce an ex-
pert on this subject had any material effect on the trial.
C. Lack of Pretrial Investigations
Stephenson contends that his trial attorneys neglected to present any mitigating evidence
at the penalty phase and that this was both substandard performance and prejudicial. Specifi-
cally, Stephenson argues that trial counsel inadequately investigated Stephenson’s prior offenses
and failed to interview key witnesses who would have testified to either Stephenson’s character
or his past good deeds, including saving one person from drowning. The State responds that trial
counsel’s pre-trial investigation was sufficient; the decisions on what evidence to present were
strategic choices made after thorough consideration; and even if there were deficient perform-
ance, Stephenson cannot show any prejudice or that the result would have been different. The
post-conviction court found that mitigating evidence “could have been more thoroughly devel-
oped and investigated” but ultimately concluded that
under all the circumstances of this case, including counsel’s knowledge of Ste-
phenson’s prior criminal convictions, including but not limited to his shooting of
a loaded firearm into an occupied residence years before, counsel’s decision not
to open up Stephenson’s background to allow such evidence to be presented to the
jury by the prosecution during the death penalty phase of the trial cannot be said
to be unreasonable or irrational, or ineffective assistance of counsel.
The record supports this finding. Trial counsel testified persuasively that placing Ste-
phenson’s character in issue would open the door to rebuttal evidence of Stephenson’s signifi-
cant criminal history. Additionally, at the post-conviction hearing, Stephenson’s ex-wife testified
that Stephenson had physically abused her and on one occasion had beaten her so severely that
she was hospitalized. Vowels testified that he was also concerned that the jury would learn of
Stephenson’s conviction for shooting into an occupied dwelling in Virginia and a battery charge
alleging Stephenson struck a man with a shovel in Newburgh. Vowels testified, “I made a deci-
sion of strategy and I employed it.” The post-conviction court found that trial counsel’s decision
25
not to offer evidence of Stephenson’s character was not unreasonable in view of the considerable
negative evidence that Stephenson’s character evidence would have produced. We agree.
Stephenson cites Rompilla v. Beard, 545 U.S. 374 (2005), in support of his claim that
counsel’s investigation of his past was ineffective. Rompilla’s sole presentation in the penalty
phase under the Pennsylvania death penalty statute consisted of testimony from five members of
his family who begged for mercy. His son testified that he loved his father and would visit him
in prison. Stephenson argues that the effort of Rompilla’s attorneys, which was found ineffec-
tive, far exceeded Stephenson’s trial counsel’s performance.
We do not find Rompilla controlling here. In Rompilla, the aggravating circumstance
rendering the defendant eligible for the death penalty under Pennsylvania law was “a significant
history of felony convictions involving the use or threat of violence to the person.” 545 U.S. at
383. Rompilla’s counsel was found ineffective for failure to investigate the circumstances of
Rompilla’s prior convictions to determine whether they involved the use of violence. The Su-
preme Court of the United States found that it was clear that Rompilla’s trial counsel knew of the
Commonwealth’s plan to parade the details of the prior crime before the jury. Id. Rompilla’s
prior conviction file was a public document, “readily available for the asking at the very court-
house where Rompilla was to be tried.” Id. at 384. The Court reasoned
It is difficult to see how counsel could have failed to realize that without examin-
ing the readily available file they were seriously compromising their opportunity
to respond to a case for aggravation. The prosecution was going to use the dra-
matic facts of a similar prior offense, and Rompilla’s counsel had a duty to make
all reasonable efforts to learn what they could about the offense. Reasonable ef-
forts certainly included obtaining the Commonwealth’s own readily available file
on the prior conviction to learn what the Commonwealth knew about the crime, to
discover any mitigating evidence the Commonwealth would downplay and to an-
ticipate the details of the aggravating evidence the Commonwealth would empha-
size.
Id. at 385-86.
In Stephenson’s case, the omitted investigation and testimony all related to mitigation,
not to proof of an aggravating circumstance necessary for eligibility for the death penalty. As a
result, in Stephenson’s case the concern for opening the door to offsetting evidence is a proper,
indeed necessary, consideration. This presents an entirely different situation from Rompilla.
Defeating an eligibility aggravator avoids the death penalty. Establishing some mitigating char-
26
acter evidence does not close a door, and it can open one. That judgment is one in which coun-
sel’s choice is entitled to deference.
D. Ineffective Appellate Counsel
A claim of ineffective assistance of appellate counsel is analyzed under the same standard
of review as for trial counsel. Lowery v. State, 640 N.E.2d 1031, 1048 (Ind. 1994). Stephenson
contends that his appellate counsel provided ineffective assistance by failing to raise three prop-
erly preserved issues. Stephenson also contends that this Court rendered his appellate counsel
ineffective by limiting his brief to 28,000 words as provided by Appellate Rule 8.2.
Stephenson’s claim that counsel failed to raise his right to present a defense involves the
trial court’s denial of Stephenson’s request to produce notes of investigating officers. 11 Accord-
ing to Stephenson, one of the notes indicated that law enforcement officers knew that the “multi-
tude of vehicles” at Napier’s and Southard’s house was connected to Napier’s drug-dealing. The
defense contends that the three killings were related to Napier’s drug operation. Stephenson
submits that the officers’ notes were exculpatory of Stephenson and inculpatory of others. The
post-conviction court found that “the materiality of this information appears marginal or specula-
tive at best.” We agree. Stephenson contended that the deaths were drug related. Knowledge
that the officers suspected Napier of dealing does not exculpate Stephenson. These notes reflect
suspicions, not evidence. And even evidence that Napier was a dealer would not suggest that
Stephenson was innocent of these murders.
The trial court denied Stephenson’s request to question Herschel Siefert in front of the
jury. Stephenson argues that appellate counsel was ineffective for failure to raise this issue on
appeal. Near the end of trial, trial counsel learned of a witness who claimed to have been in jail
with Seifert and Jimmy Knight and to have overheard them conversing about the murders, spe-
cifically why Siefert ordered the killings. Other witnesses claimed to have overheard similar
statements from Siefert and Knight. Seifert had been arrested on federal drug charges and was
being detained in Vanderburgh County. Siefert was called by the defense outside the presence of
11
The post-conviction court points out that the trial court ordered “certain portions of the notes disclosed
to the defense” and that Stephenson alleges that one not disclosed to the defense indicated that a Troy
Napier was trading dope for cars. Stephenson contends that had the defense known this fact, trial counsel
would have investigated the vehicles located at the residence and such investigation could have led to the
existence of other suspects or other information that was exculpatory as to Stephenson.
27
the jury. When the defense asked Seifert if he ordered the three killings, Seifert invoked his Fifth
Amendment rights. The trial judge then denied trial counsel Long’s request to question Seifert in
front of the jury. The trial court held testimony of witnesses claiming to have heard incrimina-
tory statements was admissible as statements against his penal interest.
Stephenson claims that appellate counsel should have raised the issue that failure to allow
the defense to force Siefert to invoke his right before the jury was reversible error. This conten-
tion fails because defendants do not have a right to force a witness to invoke the Fifth Amend-
ment privilege before the jury. See United States v. Castorena-Jamie, 285 F.3d 916, 931 (10th
Cir. 2002) (“The district court did not abuse its discretion by refusing to allow the Defendants to
compel Castorena to appear before the jury simply to invoke his Fifth Amendment rights.”);
Bowles v. United States, 439 F.2d 536, 541 (D.C. Cir. 1970).
Stephenson contends that appellate counsel were ineffective for failure to raise the issue
that, “[o]ver defense objection, prospective jurors were excused because of their religious be-
liefs.” This is based on the State’s excusing for cause jurors who affirmed they would not con-
sider recommending the death penalty. According to Stephenson, Indiana Code section 35-37-1-
5(a)(3), 12 which allows such exclusion, “suppressed those prospective jurors’ rights to practice
their religion.” He also contends that exclusion of jurors on this ground violates the equal privi-
leges provision found in Article I, section 23 of the Indiana Constitution. We rejected this con-
tention in Ben-Yisrayl v. State, 753 N.E.2d 649, 655 (2001) and adhere to that view. Accord-
ingly, appellate counsel were not ineffective for their decision not to raise these issues on appeal.
Stephenson further claims that appellate counsel failed to challenge the reasonable doubt
instructions as erroneous. Specifically, Stephenson disagrees with the instructions in both the
guilt and penalty phase which define reasonable doubt as “a fair, actual, and logical doubt that
arises in your mind . . . . ” According to Stephenson “use of the verb ‘arise’ in the instruction
violated Stephenson’s rights under the United States Constitution” because “requiring the evi-
dence to generate reasonable doubt places an unconstitutional burden on the defendant to prove
his innocence or his claim for life . . . .” Stephenson also claims appellate counsel should have
challenged the instruction that “if the State failed to prove each of these elements beyond a rea-
12
Indiana Code section 35-37-1-5(a)(3) (2004) in pertinent part states that it is a good cause to challenge
a prospective juror “if the state is seeking a death sentence, that the person entertains such conscientious
opinions as would preclude the person from recommending that the death penalty be imposed.”
28
sonable doubt, you should find the Defendant not guilty . . . .” Stephenson argues that “should”
instead of “must” in this instruction allowed a jury to find him guilty on less than a reasonable
doubt. This court has previously held these same instructions proper. See Ben-Yisrayl, 738
N.E.2d at 265. We think it plain that many more issues would have presented a greater prospect
of success on appeal. None of these instruction errors was raised at trial and none approaches
“fundamental error” permitting it to be raised even if not preserved at trial.
Finally, Stephenson claims that this Court rendered appellate counsel ineffective and dis-
allowed meaningful appellate review by the briefing limitations and time limitations it placed on
appellate counsel. This claim is not supported by pointing to any argument that was omitted
from the one hundred page revised brief. Stephenson’s appellate counsel initially filed a 152
page brief without permission to exceed the fourteen thousand words or thirty pages allowed by
the appellate rules. We ordered re-briefing in compliance with the rule at that time which
granted this court discretion to permit a party to increase the number of pages to exceed the des-
ignated number. See Appellate Rule 8.2 (1998).
The post-conviction court rejected all of these claims and found that based on Strickland,
“Stephenson’s evidence, including but not limited to the testimony of [appellate counsel] at the
PCR hearing, does not prove by a preponderance of the evidence the ineffectiveness of appellate
counsel . . . .” The post-conviction court observed “appellate counsel raised all issues they be-
lieved were meritorious” and that according to Jones v. Barnes, 463 U.S. 745 (1983) the post-
conviction court held “appellate counsel need not raise every issue that presents itself for appeal,
and counsel is permitted to choose the strongest issues for argument and omit others.” We agree.
III. Newly Discovered Evidence at Post-Conviction Review Proceeding
Stephenson argues that he presented previously undiscovered evidence that requires the
vacation of his convictions and death sentence. The evidence he cites was presented at the post-
conviction review proceeding and also in a Verified Supplement to Motion to Correct Errors
filed after the denial of post-conviction relief. Stephenson challenges the standard used by the
post-conviction court to assess this evidence. He also contests the post-conviction court’s sum-
mary denial of his Verified Supplement to Motion to Correct Errors.
29
A. Standard for Claims of Newly Discovered Evidence
The familiar standard for determining whether new evidence mandates a new trial re-
quires that
(1) the evidence has been discovered since the trial; (2) it is material and relevant;
(3) it is not cumulative; (4) it is not merely impeaching; (5) it is not privileged or
incompetent; (6) due diligence was used to discover it in time for trial; (7) the
evidence is worthy of credit; (8) it can be produced upon a retrial of the case; and
(9) it will probably produce a different result at retrial.
See, e.g., Carter v. State, 738 N.E.2d 665, 671 (Ind. 2000). Applying this frequently cited nine-
factor test, the post-conviction review court determined that “[t]he newly discovered evidence
offered by Stephenson is largely cumulative, is not worthy of credit, and probably would not
produce a different result at a retrial of this case.” This finding was entered on May 12, 2003.
Stephenson contends that this nine-factor test is no longer appropriate in a death penalty
case after the addition of subsection (k) to Indiana’s death penalty statute. That subsection, Indi-
ana Code section 35-50-2-9(k), became effective on July 1, 2003. It provides:
A person who has been sentenced to death and who has completed state post-
conviction review proceedings may file a written petition with the supreme court
seeking to present new evidence challenging the person’s guilt or the appropriate-
ness of the death sentence if the person serves notice on the attorney general. The
supreme court shall determine, with or without a hearing, whether the person has
presented previously undiscovered evidence that undermines confidence in the
conviction or the death sentence. If necessary, the supreme court may remand the
case to the trial court for an evidentiary hearing to consider the new evidence and
its effect on the person’s conviction and death sentence. The supreme court may
not make a determination in the person’s favor nor make a decision to remand the
case to the trial court for an evidentiary hearing without first providing the attor-
ney general with an opportunity to be heard on the matter.
I.C. § 35-50-2-9(k) (emphasis added).
Stephenson argues that subsection (k) establishes a new, less onerous, standard for ana-
lyzing evidence discovered after the completion of post-conviction review proceedings in capital
cases. Stephenson also contends that although subsection (k) by its terms addresses new evi-
dence that arises after a capital defendant has pursued post-conviction relief, it would be unfair,
unreasonable, and inefficient to impose a higher burden on a capital defendant presenting new
evidence at a post-conviction review proceeding.
30
We agree that the legislature intended to establish a different standard for evidence dis-
covered after post-conviction review in capital cases. Specifically, the statute means what it
says: after the completion of post-conviction review proceedings in capital cases, this Court is to
determine whether previously undiscovered evidence “undermines confidence in the outcome.”
I.C. § 35-50-2-9(k). We also agree that this standard is less burdensome than the nine-factor
approach traditionally applied to claims of new evidence. The statute’s reference to evidence
that “undermines confidence” in the conviction uses the same language as the familiar standard
announced in Strickland v. Washington, 466 U.S. 668, 669 (1984) for the prejudice prong of
claims of ineffective assistance of counsel: “a reasonable probability . . . that the result of the
proceeding would have been different.” See Wrinkles, 749 N.E.2d at 1188-89. Under this stan-
dard, a “reasonable probability” is explained as one that “undermines confidence” and is estab-
lished if the petitioner shows that the new evidence might reasonably have produced a different
result. Strickland, 466 U.S. at 694. This standard requires less than a showing that the new evi-
dence “will probably” produce a different result. Id. Moreover, the nine-factor approach used to
evaluate newly-discovered evidence places a high value on finality of judicial resolutions. As a
result, the nine-factor approach “is a rigorous one.” Vacendak v. State, 264 Ind. 101, 108, 340
N.E.2d 352, 358 (1976). Subsection (k) applies only in death penalty cases. We assume it re-
flects the legislature’s view that in those cases normal concerns for finality must yield to the pub-
lic’s interest in assuring that we do indeed reach the proper result, even at the cost of added de-
lay, expense, and expenditure of judicial resources. See generally Gardner v. Fla., 430 U.S. 349,
357-58 (1997). The new statute undoubtedly reflects recognition that previously unavailable
technology, principally DNA analysis, can produce evidence that drastically affects one’s confi-
dence in earlier convictions. Confidence in the result necessarily turns on evaluation of the evi-
dence at trial in the light of the newly advanced material. The statutory standard of whether pre-
viously undiscovered evidence undermines confidence in the capital conviction therefore turns
on the totality of the evidence, i.e. the previously undiscovered evidence in light of the evidence
at trial. Cf. Strickland, 466 U.S. at 694 (stating that a court is to look at the “totality of the evi-
dence” in an ineffective assistance of counsel claim when deciding whether counsel’s errors
were sufficient to undermine confidence in the outcome of the proceeding).
Stephenson concedes that he does not meet the literal terms of subsection (k) because he
offers this evidence in the course of post-conviction review and not after the post-conviction pro-
31
ceeding has run its course. However, he argues that it would be arbitrary and inconsistent to im-
pose a less onerous standard of review on a capital defendant based solely upon the point in time
the capital defendant discovers new evidence. We agree with Stephenson that it would be unfair
and pointless to impose a higher burden on a capital defendant seeking to present previously un-
discovered evidence at a post-conviction review proceeding than would govern previously un-
discovered evidence after the completion of post-conviction review. In other words, to the extent
the evidence is truly “previously undiscovered,”—i.e. was not available at trial—it is of no con-
sequence what point the capital defendant unearths it. A contrary holding would simply prevent
consideration of the new evidence in conjunction with all other matters available in post-
conviction review but require considering it in a subsequent proceeding. The result would be
only delay and added expense. We conclude that the less burdensome standard set forth in Indi-
ana Code section 35-50-2-9(k) applies to newly discovered evidence claims in capital post-
conviction proceedings as well as to newly discovered evidence claims after the completion of
capital post-conviction review.
The statute provides that the defendant must “present” evidence that is “previously undis-
covered.” We take this to place on the defendant the burden of showing that the proffered evi-
dence was “previously undiscovered” and that it “undermines confidence” in either the guilt, the
penalty, or both. We think “previously undiscovered” is functionally equivalent to “newly dis-
covered” under Trial Rule 60(B). Thus, in order for Stephenson to meet the first and sixth re-
quirements of the judicially created test for “newly discovered evidence,” the evidence must be
“discovered since the trial,” and the defendant must have “used due diligence to discover it be-
fore trial.” Cf. Fed. R. Crim. P. 33. 13
B. Application of Indiana Code Section 35-50-2-9(k) to Stephenson’s Newly Discovered
Evidence Claims
The post-conviction review court applied the traditional nine-factor standard in evaluat-
ing Stephenson’s claims of newly discovered evidence and summarily denied Stephenson’s Veri-
13
Evidence is “newly discovered” if the defendant shows that (1) the evidence has been “discovered since
the trial” and demonstrates (2) “due diligence in discovering and presenting such evidence.” John A.
Glenn, What Constitutes “Newly Discovered Evidence” Within Meaning of Rule 33 of Federal Rules of
Criminal Procedure Relating to Motions for New Trial, 44 ALR Fed. 13 § 5 (1979) (citing numerous fed-
eral cases).
32
fied Supplement to Motion to Correct Errors. The statute directs this Court to reevaluate these
claims under the more generous standard of “previously undiscovered evidence that undermines
confidence in the conviction or death sentence.”
Stephenson contends that the post-conviction testimony of several persons and a state-
ment and a deposition taken after the denial of post-conviction relief are all “previously undis-
covered evidence” undermining confidence in his convictions and death sentence. Some of this
evidence implicates other persons in the murders, and some, if credited, is exculpatory of Ste-
phenson. For the reasons explained below, we conclude that the majority of the offered evidence
does not meet the standard of Indiana Code section 35-50-2-9(k) because Stephenson has not met
his burden of demonstrating that the evidence is “previously undiscovered.”
First, the findings by the post-conviction court are sparse and do not meet the require-
ments of Indiana Post Conviction Rule 1, section 6. That section requires that a post-conviction
court “make specific findings of fact, and conclusions of law on all issues presented.” In analyz-
ing this rule, this Court has made clear that “[i]t is the duty of the trial judge to relate the facts on
which he makes his determination (conclusion) that the petitioner is or is not entitled to the relief
he seeks.” Davis v. State, 263 Ind. 327, 331, 330 N.E.2d 738, 741 (1975). The Court analogized
this rule with Trial Rule 52(A), the purpose of which “is to have the record show the basis of the
trial court’s decision so that on review the appellate court may more readily understand the for-
mer’s view of the controversy.” Love v. State, 257 Ind. 57, 59, 272 N.E.2d 456, 458 (1971) (in-
ternal quotations marks omitted) (quoting Harvey, 3 Indiana Practice 426 (1970)). The standard
for reviewing a post-conviction court’s failure to enter specific findings and fact and conclusions
of law is a harmless error standard. See Davis, 263 Ind. at 331, 330 N.E.2d at 741.
In evaluating Stephenson’s claims of new evidence, the post-conviction court concluded
Stephenson did not present any new material evidence to undermine the
validity of his murder convictions. Stephenson has failed to prove by a prepon-
derance of the evidence that any newly discovered evidence might result in differ-
ent verdicts against him.
The newly discovered evidence offered by Stephenson is largely cumula-
tive, is not worthy of credit, and probably would not produce a different result at a
retrial of this case. As the Supreme Court has previously stated, the basis for
newly discovered evidence should be received with great caution and the alleged
new evidence carefully scrutinized. (citations omitted).
33
We find that the post-conviction court’s lack of specific findings of fact or conclusions of
law is harmless error in this case. Even if credited, the new evidence is basically consistent with
the State’s theory that Stephenson was at least one of the killers. Nothing in the new evidence
contradicts or calls into question the testimony of Funk and Mossberger, whose credibility was
the basis of the State’s case.
Stephenson offers the post-conviction testimony of ten individuals as previously undis-
covered evidence that implicates one Guy James “Jimmy” Knight. Stephenson also offers the
testimony of two others who implicated Brian Mossberger, one of the two principal witnesses
against Stephenson.
Only two of these twelve witnesses were asked in post-conviction review whether they
had disclosed their post-conviction evidence to Stephenson’s counsel. Danyel Renfro testified at
trial. At the post-conviction review proceeding, Danyel Renfro offered testimony that Knight
had threatened victim Brandy Southard shortly before she was murdered. 14 However, Renfro
also testified that she had given all of this information to Stephenson’s counsel before trial. Ste-
phenson’s brother, David, did not testify at trial. At post-conviction review, he testified that he
had a conversation with Stephenson shortly before the murders and that Stephenson told him that
Mossberger was in possession of the gun that was ultimately found to be the murder weapon. 15
David also stated that he disclosed this information to defense counsel at the time of Stephen-
son’s trial. 16 Because it was available to the defense at trial, neither Renfro’s testimony nor
David Stephenson’s testimony can be characterized as previously undiscovered. See Williams v.
State, 793 N.E.2d 1019, 1023-24 (Ind. 2003).
Four other post-conviction witnesses (Becky Beasley, Becky Francis, 17 Brandi Martin, 18
and Carl Bruner 19 ) testified at Stephenson’s trial. Although some of their post-conviction testi-
14
Renfro, who was “best friends” with victim Southard, testified at post-conviction review that Southard
told her that Knight threatened Southard and was demanding a sum of money from Southard. Renfro also
testified that Seifert indicated to her that he knew about the murders before they happened and that Knight
was “crazy.”
15
There was no hearsay objection to this testimony.
16
David Stephenson stated that he talked with Stephenson’s counsel about the subjects he discussed at
post-conviction review, but he thought he talked about the subjects in greater detail at post-conviction
review. He answered affirmatively when asked whether he would have revealed all of the information to
Stephenson’s counsel at the time of trial if questioned by Stephenson’s counsel.
17
Becky Francis testified that Seifert was upset about some “dope” that was missing for which victim
Southard was responsible and that Seifert told Knight to “take care of the situation.” Francis also testified
34
mony was not presented at trial, we cannot conclude that any of the post-conviction additions to
their trial testimony fall into the category of “previously undiscovered evidence.” Stephenson
has made no showing that the evidence was discovered since trial or that he or his counsel exer-
cised due diligence to elicit this testimony. The requirement of due diligence includes a showing
of why the evidence was unavailable if the witnesses were known to the defense. Without some
showing of intentional withholding by the witness, a person who testifies at trial can be assumed
to have been available for whatever exploration is appropriate. Similarly, if testimony was not
elicited, without more it is reasonable to assume counsel made a decision not to open the subject.
Stephenson makes no claim that any of these witnesses was prevented from presenting this evi-
dence at trial. Nor is there any claim of active concealment by any of these four.
Six other post-conviction witnesses (Terri Greenlee West, 20 Christina Baker Baren-
fanger, 21 David Kifer, 22 David Stephenson, Chad Adams, 23 and Carla Smith 24 ) did not testify at
that she expressed to Christina Barenfanger concerns that Knight was involved in the murders, Baren-
fanger relayed Francis’s concerns to Knight, and soon after Knight came to Francis’s house and con-
fronted her. Francis claimed that Knight said “[a]nybody ever tells on me, they won’t have to worry
about dying theirselves, they’ll watch every single person they care about die around them.” She testified
that she told her mother “everything” and that her mother contacted Stephenson’s counsel. Francis’s
mother did not testify in post-conviction, and counsel were not asked about Francis’s account of Knight’s
threats. It is unclear how much of Francis’s story was given to Stephenson’s counsel prior to trial. How-
ever, nothing from Francis’s testimony suggests that she withheld any information from Stephenson’s
counsel.
18
Brandi Martin testified that Southard told her that Knight had been threatening Southard. Martin also
testified that victim Kathy Tyler told her that she was afraid and thought she was being followed. There
is no indication from Martin’s post-conviction testimony or from Stephenson’s counsel that Martin with-
held any of this information from Stephenson’s counsel. Nor are we given any other explanation for its
unavailability at trial.
19
Carl Bruner stated that he gave a ride to Knight the morning of the trial and Knight stated something
similar to “they went too far” or “somebody went too far.” There is no indication from Bruner’s testi-
mony that he withheld any of this information from Stephenson’s counsel.
20
Terri Greenlee West testified that at the time of the murders, Becky Beasley, Jimmy Knight’s girlfriend,
was living with West. West claimed that the morning after the murders, Knight came to West’s house to
speak with Beasley outside and that Beasley came back into the house and was very upset about the mur-
ders. Beasley testified at PCR that she may have been living with West at the time of the murders, but
she was not completely sure. West was not asked why she did not come forward with this testimony ear-
lier, and there is nothing to indicate whether West was contacted by Stephenson’s counsel prior to or dur-
ing Stephenson’s trial.
21
Christina Barenfanger testified that she asked Knight if he was involved in the murders and he nodded
affirmatively and winked at her. When Barenfanger was asked if she ever thought “about going to the
police with what happened in the jail,” Barenfanger responded, “No. I knew—I’d seen him in there going
to court. I figured, I figured he was just messing with my head again which he had done, you know, a
hundred times before. But I seen him on TV going to court for it. I figured if they didn’t know then, then
35
Stephenson’s trial. Stephenson has made no showing that the information they provided at post-
conviction review was unavailable for trial. Thus, it cannot be considered “previously undiscov-
ered evidence.” In hindsight, one can argue that one or more of these witnesses should have
been called to testify at Stephenson’s trial. However, a great number of trial decisions or strate-
gies are flawed in some respect in hindsight, and there is no charge of ineffective assistance
based on failure to call any of them. Furthermore, even if counsel were not aware of these wit-
nesses before trial, the statute requires diligence in discovering these witnesses and places the
burden on Stephenson to show why their testimony was unavailable at trial. Stephenson offers
nothing to discharge that burden.
Stephenson offers the testimony of two post-conviction witnesses that is “previously un-
discovered evidence.” Richard Dwayne Williams’s testimony at trial implicated Knight and
Herschel Seifert in the murders. 25 During post-conviction review, Williams testified that on two
separate occasions after trial he was attacked and that he believed that the attacks were because
of his trial testimony implicating Knight and Seifert in the murders. This testimony is plainly
you know.” Barenfanger was not asked at PCR whether Stephenson’s counsel contacted her prior to or
during Stephenson’s trial.
22
David Kifer testified at PCR that he shared a jail cell with Knight and that after Knight testified at Ste-
phenson’s trial he came back to the jail and stated “Yeah, I know Stephenson didn’t do that. But he’s [a]
hit.” Kifer was not asked why he did not come forward with this testimony earlier, and there is nothing to
indicate whether Kifer was contacted by Stephenson’s counsel prior to or during Stephenson’s trial.
23
In a transcribed statement to the police in 2003, Chad Adams claimed that he was at a bonfire at Moss-
berger’s house the night the murders took place. He stated that everyone was outside when the victims
drove by the party, at which point Mossberger claimed “I’m gonna go catch that son of a bitch.” Adams
testified that after Mossberger returned Adams overhead Mossberger admitting to committing the murders
to Stephenson. He explained why he had not come forward with this earlier. “I don’t know, everybody
just disappeared and everything and all I know is if anybody says that anyone other than John didn’t do it
they were doing to die.”
24
Carla Smith, Chad Adams’s mother, testified in a deposition that Chad’s ex-wife’s cousin Donald
Goodman told her that Goodman was attending the party at Mossberger’s house and that Mossberger
came into the house with “blood on himself” the day of the murders. Goodman also told Smith that
Mossberger threatened Goodman and stated that Goodman “better keep his mouth shut” or he was “going
to be next.” This appears to be hearsay but was admitted without objection. Goodman did not testify.
Carla Smith explained why she had not come forward earlier: “Just didn’t want to get involved because I
knew what this guy had done and, you know, I was just like mainly everybody else, you know, scared
because I had a family too.”
25
Richard Dwayne Williams testified at trial that he was jailed in the same cellblock as Herschel Seifert
and Jimmy Knight and that he overheard Herschel and Knight conversing about the murders. Specifi-
cally, Williams heard Herschel getting angry with Knight because Knight left his clip in the murder
weapon and also because Herschel had advanced Knight a large amount of crank. Williams also heard
Knight respond to Herschel that Herschel had no right to be upset because Herschel was the one who or-
dered the hit.
36
previously undiscovered evidence because it recounts events that took place after trial. A second
post-conviction review witness, Robert Andrew Smith, had been called as a defense witness at
trial but had refused to testify and was held in contempt of court and placed in solitary confine-
ment. At post-conviction review, he testified for the first time that he heard Knight bragging
about the murders. This is also “previously undiscovered evidence” because it could not be pre-
sented at trial despite defense counsel’s diligent efforts.
Although this evidence is “previously undiscovered,” it does not undermine confidence in
Stephenson’s outcome. The attacks on Williams could have been related to Williams’s testi-
mony at trial, but there is no basis to conclude that they were motivated by revenge for Wil-
liams’s truthful trial testimony. It is equally plausible that the attacks were motivated by anger
because Williams falsely implicated Knight. At post-conviction review, Smith testified to events
that occurred seven years earlier and admitted that he “wasn’t trying to hear half of what
[Knight] was saying.” More importantly, the prosecutor freely conceded that Knight and Siefert
might have been involved in the murders but argued to the jury that only Stephenson’s guilt was
before them. Any pursuit of others was for the State to consider in the future. As we noted in
Stephenson’s direct appeal, the jury was presented with several witnesses’ testimony implicating
Knight and Seifert, and the jury chose to believe the State’s evidence against Stephenson. Ste-
phenson, 742 N.E.2d at 499. The additional testimonial evidence of Williams and Smith does
not undermine confidence in Stephenson’s convictions and death sentence.
IV. Jury Bias
Stephenson challenges the post-conviction court’s rejection of his claim that he was de-
nied his right to an impartial jury because the jury was improperly exposed to a variety of extra-
neous, prejudicial influences. Specifically, Stephenson argues that (1) the jury foreperson’s ac-
quaintance with the victim’s sister undermined the foreperson’s ability to serve as a fair and im-
partial juror; (2) the foreperson’s reading of pro-prosecution crime novels in the jury room dur-
ing trial created an unacceptable influence that contaminated the verdict; and (3) some jurors’
awareness of Stephenson’s prior involvement in a fight was prejudicial. Stephenson argues that
these singly and collectively deprived him of his Sixth Amendment right to a fair jury trial and
his Fourteenth Amendment right to due process.
37
A. Juror’s Acquaintance with Victim’s Sister
Stephenson contends that jury foreperson Michael Fox’s acquaintance with Kim Seibert,
the sister of victim Kathy Tyler, undermined Fox’s ability to remain fair and impartial. The right
to a jury trial includes “a fair trial by a panel of impartial, indifferent jurors.” Turner v. Louisi-
ana, 379 U.S. 466, 471 (1965). In certain circumstances, “[t]he failure of a juror to disclose a
relationship to one of the parties may entitle the prejudiced party to a new trial.” Godby v. State,
736 N.E.2d 252, 256 (Ind. 2000) (citing Haak v. State, 275 Ind. 415, 417, 417 N.E.2d 321, 326
(1981); Barnes v. State, 263 Ind. 320, 330 N.E2d 743 (1975); Block v. State, 100 Ind. 357, 1885
WL 4222)). To obtain a new trial based on a claim of juror misconduct, the defendant must
demonstrate that the misconduct was gross and likely harmed the defendant. Id. Furthermore,
the defendant must present “specific, substantial evidence” establishing that a juror was possibly
biased. Guyton v. State, 771 N.E.2d 1141, 1145 (Ind. 2002) (quoting Lopez v. State, 527 N.E2d
1119, 1130 (Ind. 1988)).
During voir dire, all prospective jurors were asked about any possible acquaintance with
the victims, the victims’ families, or any other trial witnesses. Fox did not disclose any relation-
ship with Seibert. Seibert testified in post-conviction that she had recognized Fox at Stephen-
son’s trial but did not know whether Fox recognized her. At some point during the trial, Fox was
informed by his wife that Seibert was Fox’s children’s Sunday school teacher, and Fox realized
that he had met Seibert. He did not inform the trial court of this fact. When he was asked about
his relationship to Seibert at post-conviction review, Fox stated
Oh, I just knew Miss Kim. You know, I didn’t know what her name was or any-
thing else, but, you know, when she told me that that is -- I put two and two to-
gether. You know, I just knew her name was Seibert or Seibert or however you
pronounce it.
Although Fox should have informed the court of his acquaintance with Seibert when he
discovered it, Stephenson has presented no specific evidence that Fox was biased and has made
no showing that Fox’s nondisclosure of this casual connection to Seibert had any effect on Fox’s
performance as a juror. In short, the post-conviction court’s conclusion that juror Fox’s relation-
ship with Seibert “does not show bias or a predisposition to convict” was not clearly erroneous.
Indeed, it seems clearly correct.
38
B. Juror’s Reading of Crime Novels
Stephenson contends that foreperson Fox’s reading of murder mysteries in the jury room
during trial was exposure to extraneous information that contaminated the verdict and amounted
to juror misconduct. As already noted, to warrant a new trial based on a claim of juror miscon-
duct, the defendant must demonstrate not only that the misconduct was gross but also that it
probably harmed the defendant. Carr v. State, 728 N.E.2d 125, 131 (Ind. 2000). Exposures to
extrinsic influences during trial such as newspapers or television programs recounting facts of
the trial are evaluated under this standard. See, e.g., id.
Juror Fox read crime novels at breaks throughout the eight months of trial. These novels
were completely unrelated to Stephenson’s trial. We are given no basis to conclude that they
would affect a juror’s impartiality. The post-conviction court concluded that “[t]he reading by
juror Michael Fox of murder mystery books by author Patricia Cornwell is not juror misconduct,
and does not indicate predisposition on the part of this juror to vote for conviction in this case.”
This finding was not clearly erroneous. Once again, it seems plainly correct.
C. Jurors’ Knowledge of Stephenson’s Prior Altercation
Stephenson argues that he was denied his right to due process when one juror mentioned
an unrelated altercation involving Stephenson. During post-conviction review, Stephenson sub-
mitted the affidavit of juror Merrily Reiff stating:
During the many months of the trial, the jury spent many hours in the small jury
room behind the courtroom. On one occasion I overheard a woman juror talking
with another juror about an incident involving John Stephenson at a bar in New-
burgh. The woman juror mentioned that there had been some sort of scuffle and
that John had used a shovel during the altercation. As the two jurors were talking,
a third juror spoke up and said they should stop talking about the incident imme-
diately, which they did.
The jurors never disclosed this information to the trial court. We have held that jurors’ consid-
eration of evidence not in the record amounts to the denial of a defendant’s right to confronta-
tion. See, e.g., Saperito v. State, 490 N.E.2d 274, 278 (Ind. 1986). However, not all constitu-
tional error requires reversal. See id.; Dyer v. State, 168 Ind. App. 278, 283, 342 N.E.2d 671,
674 (1976). If the error is harmless beyond a reasonable doubt, the verdict will stand. Dyer, 168
Ind. App. at 283, 342 N.E.2d at 674. The decision to overturn a verdict based on jury contact
39
with outside information turns on the “special facts” of the case. Grigsby v. State, 267 Ind. 465,
468, 371 N.E.2d 384, 386 (1978) (citing Marshall v. United States, 360 U.S. 310, 312 (1959)).
We look at the facts and determine whether the jury contact without outside information “has so
prejudiced the defendant that he was denied a fair trial.” Id. at 468, 371 N.E.2d at 386.
The State argues that there is no evidence to demonstrate “any predisposition by any juror
to convict or otherwise affect deliberations or considerations of the issues by any juror.” We
agree. It appears that the exchange among the jurors was short, and the juror that overheard the
other two did not mention the incident to the whole jury. Although it is unfortunate that any ex-
traneous information was injected into deliberations, this information is not sufficient to consti-
tute a per se contamination of the jury. Id. at 468, 371 N.E.2d at 386 (“[T]he Court has refused
to hold that jury contact with outside information is always cause for overthrowing a verdict . . .
.”). The post-conviction court’s conclusion that the jurors’ knowledge of the prior fight “does
not show a predisposition on the part of any juror to convict, and has not been shown to have
played any material part in deliberations, any individual juror’s consideration of the case, or oth-
erwise” was not clearly erroneous.
IV. Suppression of Material Evidence
Stephenson challenges the post-conviction court’s conclusion that he failed to “prove by
a preponderance of the evidence that that the State suppressed evidence that was material to Ste-
phenson’s guilt or to his punishment.” Relying on Brady v. Maryland, 373 U.S. 83 (1963), Ste-
phenson alleges that the State violated his right to due process when the State failed to provide
Stephenson with information that corroborated his statement and alibi witnesses’ testimony.
As we have repeatedly explained, under Brady, “the suppression by the prosecution of
evidence favorable to an accused upon request violates due process where the evidence is mate-
rial either to guilt or punishment, irrespective of the good faith or the bad faith of the prosecu-
tion.” 373 U.S. at 87. To establish a Brady violation, a defendant must show “(1) that the prose-
cution suppressed evidence; (2) that the evidence was favorable to the defense; and (3) that the
evidence was material to an issue at trial.” Conner v. State, 711 N.E.2d 1238, 1245-46 (Ind.
1999) (citing Minnick v. State, 698 N.E.2d 745, 755 (1998)). The evidence is material under
Brady if “the defendant . . . establish[es] a reasonable probability that the result of the proceeding
would be different if the State had disclosed [the] evidence.” Azania v. State, 730 N.E.2d 646,
40
655 (Ind. 2000) (citing Kyles v. Whitley, 514 U.S. 410, 433-34 (1995)). Finally, “the State will
not be found to have suppressed material information if that information was available to a de-
fendant through the exercise of reasonable diligence.” Conner, 711 N.E.2d at 1246 (citing
United States v. Morris, 80 F.3d 1151, 1170 (7th Cir. 1996)).
Stephenson alleges that the State withheld a surveillance tape that corroborated the de-
fense’s version of events. In his statement to the police, Stephenson claimed that he spoke to the
victims at Circle S, went to Mossberger’s house, and then went home. Stephenson, 742 N.E.2d
at 471. To investigate the truthfulness of Stephenson’s statement, Officer Michael Hildebrand
visited the Circle S. At trial Officer Hildebrand testified that the convenience store’s surveil-
lance tape already had been taped over when he went to review the videotape of the evening in
question. The State argued that Stephenson’s version of events as set forth in his statement to the
police had to be inaccurate. Later at trial Stephenson testified that he was mistaken about the
sequence of events and that he visited Mossberger first, went to Circle S, and then went home.
At post-conviction review, Stephenson offered the affidavit of Lisa Huddleston, who was a clerk
at the Circle S at all relevant times, to refute the testimony of Officer Hildebrand. The affidavit
stated that Huddleston viewed the surveillance tape with an officer and that she saw both Ste-
phenson and victim Southard on the tape. She did not offer testimony as to the time of the event
shown on the tape. The defense asserts that “[i]n the absence of any corroboration of Stephen-
son’s statement, the State was able to argue that Stephenson lied . . . . Being able to show that he
lied in his statement allowed the State to argue that Stephenson himself should not be believed . .
. .”
The State argues that the affidavit does not establish that the police suppressed evidence
and that, in any event, Stephenson has failed to show that the tape was material evidence. We
agree. Stephenson has made no showing that the tape was not available to the defense in the ex-
ercise of reasonable diligence. More importantly, there is no basis to conclude that the State
suppressed the evidence, as opposed to its being erased in the usual course of Circle S’s security
procedures. Huddleston’s affidavit, if fully credited, established only that Stephenson and
Southard were at the store at some point in time.
41
The post-conviction court’s conclusion that Stephenson failed to prove by a preponder-
ance of the evidence that the State suppressed evidence that was material to his guilt or punish-
ment was not clearly erroneous.
Conclusion
The post-conviction court’s denial of relief is affirmed.
Shepard, C.J., and Dickson, Sullivan, and Rucker, JJ., concur.
Shepard, C.J., concurs with separate opinion in which Dickson, J. joins.
42
Shepard, Chief Justice, concurring.
A few words about the rhetoric of modern death penalty litigation as regards the most
common single issue – effective assistance of trial counsel.
Stephenson’s two lawyers at trial were practitioners well known to the bench and bar.
Lead counsel Anthony Long had thirty-five years experience in civil and criminal trial work, in-
cluding four terms as Prosecuting Attorney for Warrick County, one of Indiana’s fastest-growing
jurisdictions and the site of the murders at issue. Co-counsel Dennis Vowels of Evansville had
more than a decade of criminal law experience at the time of Stephenson’s trial and had built a
respectable reputation in the field of criminal defense. Both had received specialized training in
the defense of capital cases.
The defense team went well beyond the lawyers, eventually consisting of six or seven
altogether, including a variety of experts, a fact investigator, a mitigation specialist, and parale-
gals. The year that this team spent defending Stephenson was an intensive one in which the de-
fense enjoyed essentially unlimited resources: a third of a million dollars in lawyer time,
$65,000 worth of expert time, and mitigation and paralegal efforts that brought the defense bill to
$558,000. The post-conviction record has provided the details of this collective effort at some
length.
The contention now before us is that the foregoing defense was “perfunctory.” It is de-
clared “woefully short,” “laughable,” a defense conducted by lawyers who were “willfully unin-
formed.”
The facts establish otherwise: a seasoned defense team of respected practitioners, aided
by a collection of experts and investigators, mounting a defense with the benefit of vast financial
resources. Well beyond any notion of what the Sixth Amendment guarantees.
Dickson, J., joins.