In the
 United States Court of Appeals
               For the Seventh Circuit
                          ____________

Nos. 02-1920, 02-2260, 02-2356 & 02-2357
JAMES NEWSOME,
                                                 Plaintiff-Appellee,
                                 v.

HELEN MCCABE (as personal representative of
the estate of JOHN MCCABE), RAYMOND MCNALLY,
and CITY OF CHICAGO,
                                Defendants-Appellants.
                     ____________
           Appeals from the United States District Court
       for the Northern District of Illinois, Eastern Division.
             No. 96 C 7680—Paul E. Plunkett, Judge.
                          ____________
  ARGUED JANUARY 8, 2003—DECIDED FEBRUARY 10, 2003
                    ____________


 Before FLAUM, Chief Judge, and EASTERBROOK and
DIANE P. WOOD, Circuit Judges.
  EASTERBROOK, Circuit Judge. Fifteen years after his
conviction for killing Edward Cohen, James Newsome was
pardoned on the ground of innocence: fingerprints and
other information strongly imply that Dennis Emerson
committed the crime. Newsome filed this suit under 42
U.S.C. §1983, seeking damages from police officers who, he
contends, induced three witnesses to identify him as the
killer. Two years ago we held that officers McCabe and
2                Nos. 02-1920, 02-2260, 02-2356 & 02-2357

McNally are not entitled to qualified immunity if, as
Newsome alleges, they not only induced witnesses to accuse
him falsely but also concealed their improper activities.
Newsome v. McCabe, 

, rehearing denied, 

(7th Cir. 2001). On remand the City of Chicago,
which has a financial stake in the outcome as a potential
indemnitor of the officers, intervened to protect its inter-
ests. A jury found that, by concealing evidence favorable to
the defense, McCabe and McNally had violated Newsome’s
constitutional right to due process of law and awarded him
$15 million in damages, to which the district judge (after
denying all post-trial motions) added some $850,000 in
attorneys’ fees and costs. See 

(Apr. 2, 2002), 

(May 15, 2002).
In this appeal Chicago does not dispute the legal accuracy
of the jury charge, does not contend that the evidence was
inadequate to support the verdict, and does not argue that
the financial awards are excessive. Instead it asserts
official immunity, and it also contests some of the district
judge’s evidentiary decisions at trial.
  Many of Chicago’s contentions are variations (or recapitu-
lations) of arguments that we found unconvincing the last
time around. Chicago thus has preserved them for presenta-
tion to a higher court, but in this tribunal they are barred
by the law of the case. Chicago’s presentation effectively
asks us to use a claim of immunity to resolve the case de
novo on appeal, discarding the rule that a jury’s verdict
must be sustained if the evidence (and reasonable infer-
ences), when viewed in the light most favorable to the
prevailing party, would permit a reasonable juror to find in
that party’s favor. According to Chicago, two legal proposi-
tions call for independent appellate review: first, a claim of
immunity presents a question of law for the court, see
Rakovich v. Wade, 

, 1204 (7th Cir. 1988) (en
banc); second, as part of the immunity inquiry, a court
Nos. 02-1920, 02-2260, 02-2356 & 02-2357                    3

considers whether a constitutional violation occurred. See
Saucier v. Katz, 

, 201 (2001).
  Anthony Rounds, Josie Nash, and John Williams supplied
the principal evidence at Newsome’s criminal trial. Rounds
and Nash, who had been in Cohen’s grocery store when the
murder occurred, positively identified Newsome as the
killer; Williams, who had been outside, testified that he saw
Newsome flee. By the time of the civil trial more than 20
years later, Nash had died and Williams could not be found,
but Rounds denounced his earlier testimony as the result
of insistence by McCabe and McNally that he select No. 3
(Newsome) in a lineup; and Newsome testified that he saw
the officers coach the witnesses during the lineups. The
officers admitted (as Williams had testified during a sup-
pression hearing) that they improperly displayed photos
before the lineup occurred to improve the chance that
Williams would pick Newsome. Chicago wants us to
disbelieve the testimony Rounds gave at the civil trial, or at
least give an innocent reading to his testimony that the
officers threatened him with imprisonment if he told the
prosecutors what actually happened at the lineup. Accord-
ing to the City, all McCabe and McNally meant is that
Rounds faced criminal investigation unless he told the
whole truth—a standard warning to witnesses who spin out
inconsistent tales.
  Claims of qualified immunity neither require nor autho-
rize de novo appellate review of the evidence. The Supreme
Court made this clear: “A court required to rule upon the
qualified immunity issue must consider . . . this threshold
question: Taken in the light most favorable to the party
asserting the injury, do the facts alleged show the officer’s
conduct violated a constitutional right?” 

(emphasis added). This question is for bench rather
than jury, and the court of appeals does not defer to the
district judge’s resolution of it—but just as with motions for
summary judgment under Fed. R. Civ. P. 56, or for judg-
4                Nos. 02-1920, 02-2260, 02-2356 & 02-2357

ment as a matter of law under Fed. R. Civ. P. 50, what the
court decides is whether the record could support a finding
of unconstitutional conduct. See Johnson v. Jones, 

, 313 (1995). To say that a motion for judgment as a
matter of law under Rule 50 is addressed to the court does
not imply that the court resolves factual disputes; just so
with claims of official immunity. Saucier made the exis-
tence of a constitutional tort the initial inquiry because it’s
pointless to decide whether the impropriety of a public
actor’s conduct was clearly established at the time the con-
duct occurred, if the conduct was not improper at all. In
other words, unless the plaintiff’s complaint rests on a good
legal theory, and the record presents a triable issue under
that theory, the defendant is entitled to prevail expedi-
tiously. When the legal theory is sound, however, and there
is a material dispute about the strength of the evidence,
then the case must be tried, and the jury’s resolution is
conclusive. If an interlocutory review tests only the suffi-
ciency of the complaint, a later review may be required to
test the sufficiency of the evidence, see Behrens v. Pelletier,

(1996)—but, as with any other appellate re-
view under Rules 50 or 56, the prevailing party receives the
benefit of all reasonable credibility determinations and
inferences.
  This jury heard Rounds. It knew that he was contradict-
ing testimony given at Newsome’s criminal trial; it knew
that Rounds is not the most savory character. But just as
many a criminal defendant goes to prison on the testimony
of former partners in crime who say that they have at last
gone straight, so a jury could believe these witnesses when
they decided to sing in a new key. And Rounds’ testimony
about the warning, taken in the light most favorable to the
verdict, permitted the jury to find that McCabe and
McNally not only manipulated the identifications (some-
thing that would not by itself support an award of damages,
as our opinion denying rehearing in 2001 explained) but
Nos. 02-1920, 02-2260, 02-2356 & 02-2357                   5

also obstructed the ability of the prosecutors and defense
counsel to get at the truth in the criminal trial—which does
support the jury’s verdict.
  Seeking a way around our decision that the officers are
not entitled to qualified immunity, Chicago now contends
that they should have received absolute immunity. The
theory is that Newsome effectively accuses the officers of
suborning perjury. Witnesses enjoy absolute immunity from
civil liability on account of their testimony, see Briscoe v.
LaHue, 

(1983), and that immunity also covers
preparation. “It would be a hollow immunity if the ag-
grieved party could turn around and say, in effect: ‘True,
your delivery of bad testimony is immunized, but preparing
to deliver that testimony is not, so I can litigate the sub-
stance of your testimony.’ Substance is exactly what Briscoe
puts off limits.” Buckley v. Fitzsimmons, 

,
1245 (7th Cir. 1990) (emphasis in original), reversed in part
on other grounds (after an intermediate remand), 

(1993). Now Chicago contends that testimonial immu-
nity should be extended to non-witnesses who assisted in
the testimony’s preparation. We rejected that extension in
Ienco v. Chicago, 

, 1000 (7th Cir. 2002), and
see no reason to revisit that issue because Newsome’s case
does not present it. McCabe and McNally were not held
liable for conspiring with the eyewitnesses to commit per-
jury; their liability is under the due process clause because
they concealed exculpatory evidence—the details of how
they induced the witnesses to finger Newsome. By the time
of trial, Rounds, Nash, and Williams may have sincerely
believed that Newsome was the murderer. They did not
need immunity; instead they (and Newsome) needed
protection from steps that took advantage of memory’s
frailty, and Newsome’s lawyers needed (but did not receive)
information vital to probe whether manipulation occurred.
  Most persons have difficulty remembering or describing
the features of strangers. A person who sees a criminal for
6                Nos. 02-1920, 02-2260, 02-2356 & 02-2357

only a brief time takes away a vague sense of appearance
and behavior—and that sense may be focused by a sketch,
photograph, showup, or lineup after the events. Sometimes
the witness zeroes in on the correct person, sometimes not;
there is an element of chance and an opportunity for ma-
nipulation. Once the witness decides that “X is it” the view
may be unshakable. Psychological research has established
that the witness’s faith is equally strong whether or not the
identification is correct. We described these findings in
Krist v. Eli Lilly & Co., 

(7th Cir. 1990): “An
important body of psychological research undermines the
lay intuition that confident memories of salient experi-
ences . . . are accurate and do not fade with time unless a
person’s memory has some pathological impairment. . . .
The basic problem about testimony from memory is that
most of our recollections are not verifiable. The only war-
rant for them is our certitude, and certitude is not a re-
liable test of certainty. . . . [T]he mere fact that we remem-
ber something with great confidence is not a powerful
warrant for thinking it 

(citations
to the scholarly literature omitted). See Elizabeth F. Loftus
& James M. Doyle, Eyewitness Testimony: Civil and
Criminal (3d ed. 1997); Elizabeth F. Loftus, Eyewitness
Testimony (1979; rev. ed. 1996); Daniel L. Schacter, The
Seven Sins of Memory: How the Mind Forgets and Remem-
bers 112-37 (2001). See also United States v. Hall, 

, 1118-20 (7th Cir. 1999) (concurring opinion). Jurors,
however, tend to think that witnesses’ memories are reli-
able (because jurors are confident of their own), and this
gap between the actual error rate and the jurors’ heavy
reliance on eyewitness testimony sets the stage for errone-
ous convictions when (as in Newsome’s prosecution) every-
thing depends on uncorroborated eyewitness testimony by
people who do not know the accused. This is why it is vital
that evidence about how photo spreads, showups, and
lineups are conducted be provided to defense counsel and
Nos. 02-1920, 02-2260, 02-2356 & 02-2357                     7

the court. It is also why the constitutional violation justify-
ing an award of damages is not the conduct of the lineups
but the concealment of evidence about them. Secreting
evidence is not covered by absolute immunity.
  Because recollection is suggestible, it was important in
this civil case to explore the question whether the testi-
mony of Rounds, Nash, and Williams identifying Newsome
at the criminal trial was attributable to deliberate manipu-
lation or instead to chance. For if chance errors are to
blame, and the witnesses would have identified Newsome
no matter how the officers prompted them during the
lineups, then defendants’ conduct did not cause the wrong-
ful conviction and an award of damages would be improper.
To explore this issue Newsome presented the testimony of
Gary Wells, a professor of psychology who has performed
experiments and written scholarly works in this field. See,
e.g., Gary L. Wells & Elizabeth A. Olson, Eyewitness
Identification, 54 Ann. Rev. Psych. 277 (2003); Gary L.
Wells, Eyewitness Identification: A System Handbook
(1988). Wells conducted an experiment to determine the
likelihood that three persons who saw Emerson nonetheless
would identify Newsome. He showed two panels of subjects
different pictures of Emerson for 15 seconds then, after
some time had passed, showed them pictures of the men in
the lineup and asked them to choose the one they had seen
in the initial photograph. Of 50 members on the first panel,
none selected Newsome’s photo; of 500 members on the
second panel (which was shown a different photo of Emer-
son), 15 chose Newsome’s photo. Performing a chi-square
test, Wells calculated that the probability of all three
eyewitnesses independently picking Newsome out of a line-
up by chance error was substantially less than one in 1,000,
implying that the officers must have manipulated their
identifications.
 Chicago asked the district judge to exclude Wells’ testi-
mony under Fed. R. Evid. 702, which as amended in 2000
8                Nos. 02-1920, 02-2260, 02-2356 & 02-2357

codifies (with some variation) the holding of Daubert v.
Merrell Dow Pharmaceuticals, Inc., 

(1993).
The district judge concluded that Wells is an expert on the
subject of identification, that his testimony was based on
sufficient data, that his methods were reliable by the
standards of the field, and that he applied these methods
reliably to the facts of Newsome’s case. Experiments of the
kind that Wells performed are the norm in this branch of
science and have met the standard for scholarly publication
and acceptance. There were of course potential problems.
For example, Wells assumed that Emerson is the killer, so
that the witnesses saw him; if anyone other than Emerson
committed the murder, the test is invalid. Wells was candid
about this vital assumption, which was open to probing and
argument by the defendants. Wells also assumed that two-
dimensional images (pictures) yield the same effects on
memory as three-dimensional views (live action in the vic-
tim’s grocery store; lineups in the police station; identifica-
tions in open court). This may or may not hold, but the
claim of equivalence was open to exploration at trial, and it
is hard to see what else Wells could have done. Even if he
could have conscripted Emerson and the lineup participants
for an experiment, time has so altered their appearance
since the events of October 1979 that the results would
have been unreliable. Chicago does not contend that there
was a better way to find out whether Rounds, Nash, and
Williams would have identified Newsome without the
coaching. Instead it insists that Wells’ testimony was irrel-
evant because he did not determine how the witnesses had
been induced to believe that they saw Newsome commit the
murder. Yet testimony need not prove everything in order
to be useful. As we have said, the jury had to consider the
possibility that unhappy chance rather than malfeasance
led to the mistaken conviction. Wells provided information
valuable in this endeavor. Appellate review of the district
judge’s decision is deferential, General Electric Corp. v.
Nos. 02-1920, 02-2260, 02-2356 & 02-2357                   9

Joiner, 

(1997), and there was no abuse here;
indeed, we would have acted precisely as did the district
judge. As Hall explains, it may be prudent to avoid compli-
cating criminal trials with general scientific evidence about
the psychology of identification—though scientific evidence
that a given person deviates from the norm (for example, is
exceptionally suggestible) may be invaluable. See United
States v. Hall, 

(7th Cir. 1996). No matter how
criminal trials should be managed to keep the jurors’ minds
on the main event, however, Wells’ testimony was not a
distraction in this civil proceeding but went to an important
ingredient of the plaintiff’s claim.
  Chicago presents several other objections to the district
judge’s handling of the trial. These do not require sepa-
rate discussion; all were treated thoughtfully in Judge
Plunkett’s post-trial opinion, and we substantially agree
with his disposition of them. He did not separately discuss
the City’s request for monetary sanctions on account of
Newsome’s failure to reveal his arrest on heroin charges
two weeks before the civil trial, his loss of employment as
a paralegal, and thus his perjury at trial when he described
himself as gainfully employed (which enabled counsel to
argue that Newsome is a model citizen who rose above a
disordered youth). Yet the district judge did conclude that
Newsome’s deceit does not require a new trial—a decision
that Chicago no longer contests—and it was not necessary
to discuss separately the possibility that a financial sanc-
tion might have been appropriate. It is not clear to us that
Chicago adequately alerted the district judge to this
possibility (the record does not contain a formal motion
seeking monetary sanctions, see Fed. R. Civ. P. 37(c)(1));
but whether it did or not, there was no abuse of discretion
in the judge’s resolution, even if we might have handled
this issue differently.
                                                  AFFIRMED
10             Nos. 02-1920, 02-2260, 02-2356 & 02-2357

A true Copy:
      Teste:

                      ________________________________
                      Clerk of the United States Court of
                        Appeals for the Seventh Circuit




                 USCA-02-C-0072—2-10-03