No. 01-1159

MICHAEL L. PIASKOWSKI,

Petitioner-Appellee,

v.

JOHN BETT,

Respondent-Appellant.

Appeal from the United States District Court
for the Eastern District of Wisconsin.
No. 99-C-1428--Myron L. Gordon, Judge.

Argued May 18, 2001--Decided July 10, 2001
 
 

  Before EASTERBROOK, MANION, and EVANS,
Circuit Judges.

  EVANS, Circuit Judge.  On November 10,
1992, the Green Bay, Wisconsin, police
department received an anonymous tip that
Keith Kutska, an employee of the James
River Paper Mill, was going to steal a
piece of electrical cord at the end of
his shift. The police passed the tip on
to the company. When Kutska left the mill
that evening, a security guard stopped
him and asked to search his duffel bag.
Kutska refused to permit the search and
left the premises, but he received a 5-
day, unpaid suspension from work for not
letting the guard peek into his bag.

  Kutska was not happy about this, a fact
that quickly became known around the
mill. Later, Kutska learned that the
police were tipped off by the anonymous
call, and he set out to determine who
blew the whistle on him. Fearing
repercussions, the mill worker who
provided the tip, Thomas Monfils, begged
the police not to give Kutska access to
the tape of his call. But, in a
bureaucratic screw-up, Kutska obtained a
copy of the tape with little difficulty.
How Kutska obtained the tape is detailed
in Monfils v. Taylor, 165 F.3d 511 (7th
Cir. 1998), cert. denied, 528 U.S. 810
(1999). We won't repeat here what we said
there. Suffice to say that once Kutska
got his hands on the tape, he immediately
identified Monfils as the tipster.

  Kutska brought the tape to the mill (on
November 21) a few days after he returned
from his suspension and played it a
number of times for various coworkers,
including once for Monfils, who admitted
making the call. Michael Piaskowski and
Randy LePak, who were with Monfils and
Kutska for this tape playing, reacted
strongly. On his way out the door after
hearing the tape, Piaskowski said, "Geez,
Tom, I just fuckin' don't believe you'd
do that." LePak was more expressive,
telling Monfils, "You can thank your
fuckin' lucky stars you didn't do it to
me, or I'd have killed you."

  News of these events spread quickly
throughout the mill. And many workers,
around 20 it would appear, eventually
heard the tape early that morning.
Shortly after 7 a.m., Kutska took the
tape to the control room for machine 9
(a/k/a "coop 9") and played it for a
group that included the five men with
whom he would later be convicted of
Monfils' murder: Piaskowski, Michael
Hirn, Reynold Moore, Dale Basten, and
Michael Johnson. Although Kutska had
stolen the electrical cord, he implied
that Monfils' accusation was false, and
he told Moore and Hirn to "give Monfils
some shit" for snitching on a fellow
union brother.

  Meanwhile, Monfils left his post at coop
7 to perform a periodic task known as
"turnover" scheduled for 7:34. A minute
later, an altercation involving Monfils
and a number of workers occurred near a
water fountain (or "bubbler" in Wisconsin
parlance) located between coops 7 and 9.
Monfils was attacked and seriously
injured. He was left, unconscious but
alive, lying in a ball on the mill floor.
Approximately 5 minutes later, mill
worker David Wiener observed Basten and
Johnson in an area which connects the
paper machines with the vat that supplies
pulp to the machines. Johnson was walking
backwards, 5 or 6 feet in front of
Basten, and they appeared to be carrying
something--and on the basis of this
evidence, that something was Monfils--
toward a pulp vat.

  At 7:45 Kutska and Moore entered coop 7,
followed within minutes by Piaskowski.
Kutska told Piaskowski to alert a
supervisor that Monfils was missing,
purportedly to get Monfils in trouble for
leaving his work station. Piaskowski did
so and added that there was "some shit
going down." A search was begun, and the
next day Monfils' body was discovered at
the bottom of the vat. A heavy weight,
usually kept near machine 7, was tied
around his neck. The coroner determined
that Monfils died by asphyxiation due to
the aspiration of paper pulp.

  No charges were immediately filed, and
for over 2 years the Green Bay police
investigated the crime. A break in the
case came in April 1995, when mill worker
Brian Kellner,/1 a friend of Kutska's,
told police that Kutska, the previous 4th
of July, admitted that the six defendants
(plus Jon Mineau, who was never charged)
confronted Monfils near the bubbler after
the 7:34 turnover. Kutska drew a diagram
of where each defendant stood and told
Kellner that someone slapped Monfils and
that Hirn shoved him. Kutska asked "what
if" some unidentified person hit Monfils
with a wrench or a board. At that point
in the assault, Kutska told Kellner that
he left the area. According to Kellner,
Kutska also stated that mill workers
Dennis Servais and Pete Delvoe were
present and witnessed the confrontation,
but both later denied this to the police.
In fact, Delvoe testified that he
completed the 7:34 turnover with Monfils
but stayed behind to clean the paper
machine afterward. Ten to 15 minutes
after the turnover, Piaskowski approached
Delvoe and asked him if he had seen
Monfils.

  Kellner relayed Kutska's statement at
trial, although at postconviction
proceedings he partially recanted
andtestified that Kutska identified only
himself, Hirn, and Moore as participants
in the confrontation. Only one other
witness testified about the
confrontation--James Gilliam, a jailhouse
informant and Moore's cell mate, who
repeated what Moore told him. Moore's
story, according to Gilliam, was that
Kutska, Moore, and unidentified others
gathered before the confrontation and
decided to scare Monfils. The men from
the meeting and unidentified others then
confronted Monfils after the 7:34
turnover. Kutska hit Monfils in the face,
and then "Mr. Monfils went in a cuddle
just like this and he [Moore] say he just
do it like everybody else and he was just
came from with his fist over the head .
. . ." All participants in and witnesses
to the attack, again according to Moore
via Gilliam, then went back to work.
Moore, according to Gilliam, was later
shocked to discover that Monfils was
found dead. Based essentially on this
evidence, a state court jury convicted
Piaskowski and the other five defendants
of first degree murder.

  In our civil suit opinion in Monfils v.
Taylor we affirmed a jury verdict in
favor of Monfils' estate against members
of the Green Bay police department for
their role in releasing the tape which
started the chain of events that led to
Monfils' death. Today we turn to the
criminal conviction of Piaskowski, the
defendant against whom, it appears, the
State presented the least evidence. After
exhausting his direct appeals, see State
v. Piaskowski, 1998 WL 644758 (Wis. Ct.
App.), review denied, 589 N.W.2d 629
(Wis.), cert. denied, 527 U.S. 1035
(1998), Piaskowski petitioned the federal
district court for a writ of habeas
corpus, arguing that the record contained
insufficient evidence to sustain his
conviction. The district court, Judge
Myron L. Gordon presiding, granted the
writ and forbade the State from retrying
him for Monfils' murder. The State
appeals from that order and we review it
de novo. See Washington v. Smith, 219
F.3d 620, 627 (7th Cir. 2000).

  Under 28 U.S.C. sec. 2254(d)(1), as
amended by the Antiterrorism and
Effective Death Penalty Act of 1996, a
prisoner may obtain habeas relief on a
claim that was adjudicated on the merits
by a state court if that court's decision
was "contrary to, or involved an
unreasonable application of, clearly
established Federal law, as determined by
the Supreme Court of the United States."
The clearly established federal law
applicable to this case is set out in
Jackson v. Virginia, 443 U.S. 307, 319
(1979), which holds that due process
requires reversal of a criminal
conviction if, viewing the evidence in
the light most favorable to the
prosecution, no "rational trier of fact
could have found the essential elements
of the crime beyond a reasonable doubt."
Because the murder in Jackson is
factually dissimilar to the murder of
Monfils, we will ignore the "contrary to"
clause of sec. 2254(d)(1), see Hennon v.
Cooper, 109 F.3d 330, 334-35 (7th Cir.
1997), and confine our inquiry to whether
the Wisconsin Court of Appeals' decision
(the Wisconsin Supreme Court declined to
review the case) that a rational jury
could have convicted Piaskowski based on
this record resulted from an objectively
unreasonable application of Jackson. See
Williams v. Taylor, 529 U.S. 362, 409
(2000).

  Under the State's theory of the case,
Monfils was still alive when he was
thrown into the pulp vat, so only Basten
and Johnson--who were seen carrying
something toward the vat after the
confrontation near the bubbler--directly
committed the murder. Piaskowski could be
found guilty as a party to the crime,
however, if he aided and abetted its
commission or conspired with others to
commit it. See Wis. Stat. sec. 939.05.
The State has abandoned its aiding and
abetting theory and now relies solely on
a conspiracy theory to uphold
Piaskowski's conviction.

  A conspiracy under Wisconsin law has two
elements:  (1) an agreement between two
or more persons to engage in conduct that
will further the commission of the
underlying crime, and (2) a conscious
intent to realize the criminal objective.
May v. State, 293 N.W.2d 478, 483 (Wis.
1980). Although the agreement need not be
express, O'Neil v. State, 296 N.W. 96,
102 (Wis. 1941) (tacit agreement
sufficient), the State must establish
more than Piaskowski's mere presence at
the scene of the crime in order to secure
a conviction. See State v. Charbarneau,
264 N.W.2d 227, 233 (Wis. 1978).

  The State's case against Piaskowski
relies almost exclusively on the
intersection of the accounts of Kutska
and Moore as related by Kellner and
Gilliam. Specifically, the State argues
that Kellner's trial testimony relaying
Kutska's story places Piaskowski at the
scene of the assault with the other five
defendants. Gilliam, in turn, testified
that Moore told him he (Moore) hit
Monfils "like everybody else." The State
argues that if "everybody" hit Monfils,
and Piaskowski was present at the
confrontation (as established by Kutska's
confession to Kellner), Piaskowski must
have participated in the beating. And
having just perpetrated a severe battery,
the State goes on, Piaskowski had the
same motive as did the others to cover up
the crime. The State infers that he did
so by agreeing with the others to dump
Monfils' unconscious body into the pulp
vat.

  Putting aside the weighty Sixth
Amendment Confrontation Clause problems
inherent in the use of the testimony of
Kellner and Gilliam against Piaskowski,
we are not convinced that their
respective stories implicate Piaskowski
in Monfils' murder to "a state of near
certitude." See Jackson, 443 U.S. at 315.
A strong suspicion that someone is
involved in criminal activity is no
substitute for proof of guilt beyond a
reasonable doubt.

  As an initial matter, although appellate
courts generally are disinclined to
evaluate the credibility of witnesses,
see Schlup v. Delo, 513 U.S. 298, 330
(1995), we agree with Judge Gordon in the
district court that Kellner's account of
Kutska's statement is a bit suspect. Both
Kutska and Kellner had been drinking on
the evening Kutska told his story
(Kutska, according to Kellner, had 40
beers!), and many of Kutska's details do
not mesh with the facts. For example,
Kutska claimed that Mineau participated
in the confrontation until Donald
Boulanger ordered him away to perform the
next turnover, but Boulanger did not
confirm this assertion (despite
investigators' best efforts to "refresh"
his recollection). Although not germane
to our review of this record, we note
that Kellner recanted a key portion of
his trial testimony (especially as it
relates to Piaskowski) during
postconviction proceedings, claiming that
Kutska did not identify any member of the
group that confronted Monfils other than
himself, Hirn, and Moore. The
inconsistencies in Kellner's trial
account of Kutska's story render his
credibility marginal at best. Indeed,
even the state trial judge observed that
"a conviction cannot be sustained on the
basis of [Kellner's] evidence," and
opined that if Kellner's testimony were
the only evidence in the case, "the Court
would never have allowed [it] to go to
the jury."

  Nevertheless, we are willing to accept--
as did Judge Gordon--that Kellner's
account places Piaskowski near the
bubbler at the time the confrontation
began. But Kellner's account says nothing
about what Piaskowski did or agreed with
the others to do. According to Kutska
(again, as relayed by Kellner's testimony
of his barroom conversation with Kutska 2
years after the murder), Hirn and one or
two unidentified members of the group
assaulted Monfils. At that point, Kutska
told Kellner he left the area, so he
offers no knowledge as to what happened
next. Thus, Kellner's testimony
establishes only that Piaskowski may have
been present at the bubbler when Monfils
was hit, but given these circumstances,
his mere presence is not sufficient, by
itself, to sustain Piaskowski's
conviction. The jury's conclusion that
Piaskowski participated in the beating
and/or conspired with the other
defendants to kill Monfils is
speculation.

  The State attempts to fill this gaping
hole in its case against Piaskowski with
the testimony of the jailhouse witness,
Gilliam. According to Gilliam's account
of Moore's statement (which on this
record is hearsay as to Piaskowski),
Kutska first punched Monfils, and then
Moore punched and kicked Monfils "like
everybody else." Moore did not identify
Piaskowski as part of the group that
constituted "everybody," however. Indeed,
there is no mention of Piaskowski in
Gilliam's testimony. Moreover, the
statement that Moore beat Monfils "like
everybody else" is ambiguous to say the
least. One possibility is that, as the
State contends, Moore meant that all
those present at the confrontation
participated in the beating. Equally
possible, however, is that Moore meant he
beat Monfils in the same manner as
everyone else who took part in the
beating, which may or may not have
included all those present at the scene.
The State did not attempt to clarify what
Moore meant. Gilliam, of course, could
not have shed any light on the intended
meaning of Moore's words. Even accepting
as fact that Piaskowski was at the scene
of the beating, Gilliam's testimony about
what Moore told him does not constitute
proof beyond a reasonable doubt that
Piaskowski played a role in the incident,
either as a direct participant or as a
conspirator. No reasonable jury, we
think, could have come to a contrary
conclusion.

  The State's meager circumstantial
evidence against Piaskowski is also
innocuous. The fact that Piaskowski was
present in coop 9 prior to the beating
and entered coop 7 after the beating, 2
or 3 minutes after Kutska and Moore,
proves little because Piaskowski spent
much of his workday in those areas. The
State also makes much of the fact that
Piaskowski complied with Kutska's
direction to report Monfils missing and
added on his own accord that there was
"some shit going down." According to the
State, the fact that Piaskowski made the
report without asking what had happened
and added his personal assessment of the
situation proves that he was part of a
conspiracy to commit the murder. It does
no such thing. Monfils was, in fact, mis
sing, and as Delvoe confirmed, Piaskowski
had been looking for him prior to
returning to coop 7. True, a quick-
thinking Piaskowski could have been
trying to make Delvoe into an alibi
witness by asking if he had seen Monfils
(when he really knew Monfils was in the
vat). This, however, like so much else in
this case, is conjecture camouflaged as
evidence.

  Neither is Piaskowski's vague reference
to "shit going down" illuminating. In
fact, it had been an extraordinarily
eventful morning, and Piaskowski's
statement very plausibly could have been
addressing the fact that Kutska was
playing the tape of Monfils' call to the
police for anyone who would listen, and
that many employees who heard the tape--
most of whom were never charged with
murder-- were very angry at Monfils.
Moreover, even if Piaskowski's call
proves he knew about Monfils' fate, it
does not prove he was involved in his
murder; perhaps he merely witnessed the
beating or heard about it secondhand from
one of the assailants. In short, the two-
stage inference that the "shit going
down" was murder, and that Piaskowski's
knowledge of the murder necessarily
constitutes his participation in it,
requires a leap of logic that no
reasonable jury should have been
permitted to take. Although a jury may
infer facts from other facts that are
established by inference, each link in
the chain of inferences must be
sufficiently strong to avoid a lapse into
speculation. See United States v. An
Article of Device, 731 F.2d 1253, 1262
(7th Cir. 1984); Yelk v. Seefeldt, 151
N.W.2d 4, 9 (Wis. 1967). In this case,
the chain of inferences the State
attempts to forge fails in multiple
places. Piaskowski may have been involved
in the attack on Monfils and his murder,
but under our system of law, that must be
proven beyond a reasonable doubt. The
scant evidence here falls short of
meeting that burden.

  Having determined that no rational jury
could convict Piaskowski, little further
analysis is required to confirm Judge
Gordon's conclusion that the Wisconsin
Court of Appeals' decision was an
unreasonable application of the law as
determined by the Supreme Court in
Jackson. The most serious error the court
of appeals made is that it found--without
reference to any supporting evidence--
that Piaskowski "kicked and beat Monfils"
during the confrontation near the
bubbler. If the evidence supported this
assertion, we would agree that a rational
jury could have convicted Piaskowski and
that an appellate court reasonably could
have affirmed his conviction. As we have
discussed, however, the record is devoid
of any direct evidence that Piaskowski
participated in the beating of Monfils,
and the available circumstantial evidence
at most casts suspicion on him. This is a
far cry from guilt beyond a reasonable
doubt. Accordingly, we affirm the
district court's decision to issue the
writ of habeas corpus.

  Anticipating the possibility of our
ruling today, the State requests
permission to retry Piaskowski under the
theory that he conspired with the other
defendants to beat Monfils and Monfils'
death was a natural and probable
consequence of the beating. Under
Wisconsin law, a conspirator is
responsible for any crime committed as a
natural and probable consequence of the
intended crime. Wis. Stat. sec. 939.05
(2)(c). Piaskowski, of course, objects to
any subsequent trial on the ground that
the Fifth Amendment's Double Jeopardy
Clause bars retrial for the same offense
after an appellate court has reversed a
conviction based on insufficiency of the
evidence. See Burks v. United States, 437
U.S. 1, 18 (1978).

  This court and others have carved out a
narrow exception to the rule established
in Burks, however. In United States v.
Lanzotti, 90 F.3d 1217 (7th Cir. 1996), a
number of defendants who leased video
poker and video slot machines to taverns
were charged with participating in an
illegal gambling business. One element of
the federal offense was violation of a
state law, which in Lanzotti was an
Illinois statute that outlawed
"[p]lay[ing] a game of chance or skill
for money" and "bargain[ing] for the sale
or lease of . . . any gambling device."
720 ILCS 5/28-1(a)(1), (3). The
government initially proceeded under the
theory that the defendants violated both
subsections of the Illinois statute, but
during trial, prosecutors inexplicably
abandoned any argument that the
defendants had leased gambling devices.
In accordance with the government's
strategy, the jury was instructed only on
the "playing a game of chance" subsection
of the statute. After the jury returned a
guilty verdict, the district court
granted a new trial, finding--as the
government then conceded--that the
evidence did not support the conclusion
that the defendants had "played a game of
chance" within the meaning of the first
subsection of the statute.

  Lanzotti and his codefendants were
reindicted for the same offense, this
time on the theory that they had aided
and abetted others (the tavern patrons)
in "playing a game of chance." On appeal
from the denial of their motion to
dismiss on double jeopardy grounds, we
held that the trial court's grant of a
new trial was not the functional
equivalent of an acquittal. Id. at 1222.
A new trial was granted because the jury
had been instructed on a theory of the
offense inapplicable to the defendants'
conduct, a so-called "trial error," not
because the defendants were factually
innocent. Id. Because the government had
introduced overwhelming evidence of the
defendants' guilt under the aiding and
abetting theory (as well as under the
second subsection of the Illinois
statute), we held that the Double
Jeopardy Clause would not prohibit a
second trial under the correct legal
theory. Id.

  Unlike the defendants in Lanzotti, there
is scant evidence in the record in this
case that supports Piaskowski's guilt
under the State's new legal theory. In
fact, our analysis today of the evidence
allegedly establishing Piaskowski's
participation in a murder conspiracy
applies equally to his alleged
participation in a beating conspiracy.
Thus, unlike the Lanzotti trial court's
grant of a new trial, our decision today
is the functional equivalent of an
acquittal, at least with respect to the
State's belated "natural and probable
consequences" theory. Under Burks,
therefore, the State may not retry
Piaskowski for Monfils' murder without
offending the Double Jeopardy Clause.

  For these reasons, the district court's
order issuing a writ of habeas corpus is
AFFIRMED.

FOOTNOTES

/1 Kellner was deer hunting, and not at work, when
Monfils was killed.