PUBLISHED
                                                Filed: June 4, 2004

UNITED STATES COURT OF APPEALS
                 FOR THE FOURTH CIRCUIT


TIMOTHY LANIER ALLEN,                  
              Petitioner-Appellant,
                 v.
                                                  No. 02-5
R. C. LEE, Warden, Central Prison,
Raleigh, North Carolina,
               Respondent-Appellee.
                                       

                              ORDER


   On the Respondent’s Petition for Rehearing by the en banc panel,
Judges Wilkinson, Niemeyer, Luttig, and Williams voted to grant the
petition, and Chief Judge Wilkins and Judges Michael, Motz, Traxler,
King, Gregory, and Shedd voted to deny the petition. The Petition for
Rehearing is therefore denied. Judge Luttig wrote an opinion dissent-
ing from the denial of the Petition for Rehearing.

  Judges Widener and Duncan did not participate in the decision.

                                               For the Court,

                                           /s/ Patricia S. Connor
                                                    Clerk

LUTTIG, Circuit Judge, dissenting from the denial of rehearing:

   Timothy Lanier Allen murdered North Carolina State Trooper Ray-
mond E. Worley in cold blood, by shooting him three times at point-
blank range — once through the head — and leaving him slumped
over alongside the road, literally to drown in his own blood. For this
2                           ALLEN v. LEE
murder, a jury of North Carolina citizens sentenced Allen to death.
Now, almost nineteen years after Allen’s murder of Trooper Worley
and almost eighteen years after the jury of Allen’s peers sentenced
Allen for his crime, this court has invalidated Allen’s sentence on
nothing more than its unexplained speculation that some juror might
have voted to sentence Allen differently had the jury not been
instructed that it had to unanimously find the following inconsequen-
tial (if not affirmatively damaging) "mitigating" factors in order to
consider them in sentencing Allen:

    Timothy Allen was only 30 years old when he murdered
    Trooper Worley.

    Timothy Allen’s multiple previous convictions were only
    for shoplifting, breaking and entering, and burglary.

    Timothy Allen was raised in a single-parent home, by only
    his mother and grandmother.

    Timothy Allen is the father of three sons.

    Timothy Allen supported one of his three sons.

    Timothy Allen "remained employed" when he was not in
    prison.

    Timothy Allen’s parents had a fight at Allen’s third birthday
    party.

This, in the face of the fact that the jury did find unanimously the
three contrastingly nonfrivolous mitigating factors that Allen had gen-
uine remorse for the murder; that he had achieved while incarcerated;
and that he had no history of crime involving deadly weapons, and
nonetheless decided to sentence Allen to death, given the heinousness
of his murder of Trooper Worley.

   Not only does the majority’s decision defy common sense and
insult the men and women who served on the jury in this case, it so
distorts the harmless error inquiry required of federal courts conduct-
                              ALLEN v. LEE                              3
ing habeas review under Brecht v. Abrahamson, 

(1993),
as to render that inquiry all but meaningless in future cases. It is for
these reasons that I dissented from the court’s judgment on the
McKoy issue when the en banc decision issued (in abbreviated form
because of the court’s extraordinary delay in deciding the case). See
Slip Op. at 53-55 (4th Cir. 2004). It is for these reasons that I dissent
now, with full opinion, from the denial of rehearing with respect to
that judgment.

                                    I.

   Despite the fact that en banc reargument took place in this case
almost a year ago and the deadline for deciding this case had long
since passed by the time it issued its opinions, the court failed to pro-
duce an opinion clearly explaining to the State of North Carolina its
reasons for setting aside the death sentence that the jury of North Car-
olina citizens imposed eighteen years ago. Instead, the court produced
a fractured array of opinions including, most notably, a plurality opin-
ion authored by Judge Gregory on the issue presented under McKoy
v. North Carolina, 

(1990), which does not even attempt
the harmless error analysis required of an appellate court by Brecht.

   The plurality does not even purport to undertake the carefully-
reasoned, "individualized" reweighing of the aggravating and mitigat-
ing circumstances, against the backdrop of the specific record facts of
the crime committed, which is required under Brecht. Instead, in a
few pages that appear as afterthought in attempted justification of a
result mistakenly reached before the court even understood that it was
required to undertake the Brecht inquiry prior to granting the writ of
habeas corpus, the plurality lists the aggravating circumstances in
sterilized form, admittedly divorced from the actual circumstances of
Allen’s crime; lists the above-recited mitigating circumstances;
recites only that evidence from the record that supports these mitigat-
ing circumstances, while ignoring any contrary evidence; strings
together quotations from the applicable precedents; and then asserts
in classic ipse dixit that a juror could find "any one" of the listed fac-
tors (aside from two which are so frivolous even the majority cannot
bring itself to rely on them) and on the strength of that single factor
decide against imposition of the death penalty. Thereafter, in empty
incantation of the language from the authorities it string cites, it sim-
4                            ALLEN v. LEE
ply states that it cannot say with fair assurance that the jury would not
have sentenced Allen differently had it been properly instructed that
it could consider the mitigating circumstances listed.

   With that, and nothing more, the plurality sweeps away the sen-
tence imposed on Allen by the citizens of North Carolina almost two
decades ago. Such a substantively meaningless harmless error review
betrays that the court has yet to give any more thought to the question
of the harmfulness of the instructional error in this case under Brecht
than the original panel had given the question at the time that it filed
its opinion granting the writ of habeas corpus without so much as a
mention of Brecht or, for that matter, even of the requirement for
harmless error review under Brecht before a writ of habeas corpus
may be granted in cases such as this. See Allen v. Lee, 

,
656-58 (4th Cir. 2003).

                                   A.

   In order to hold, as the plurality does here, that a faulty jury
instruction had a "substantial and injurious effect or influence on the
jury’s verdict" sentencing a defendant to death, "the court must deter-
mine what the sentencer would have done absent" the error. Stringer
v. Black, 

, 230-31 (1992). Thus, the court must compare
(1) the aggravating and (2) mitigating evidence which the jury did
consider, (3) to the particular evidence which the jury was prevented
by the faulty jury instruction from considering and (4) then analyze
whether, and to what extent, that evidence would have affected or
influenced the verdict eventually reached by the jury. See, e.g., Boyd
v. French, 

, 327-28 (4th Cir. 1998); Horsley v. Alabama,

, 1492-93 (11th Cir. 1995). The Supreme Court could
not have been any clearer that, in conducting this analysis, "[w]hat is
important . . . is an individualized determination on the basis of the
character of the individual and the circumstances of the crime." See
Barclay v. Florida, 

, 958 (1983) (emphasis in original)
(instructing state courts conducting harmless error analysis at the sen-
tencing stage of capital trials).

   Despite the simplicity of this analytical framework and the clarity
of the Supreme Court’s directive, the plurality does not even purport
                              ALLEN v. LEE                              5
to make "individualized" findings of the kind mandated by the
Supreme Court — none.

   Nowhere in its opinion does the plurality even describe the actual
circumstances of Allen’s murder of Trooper Worley, without which
it is not even possible to begin to conduct Brecht’s harmless error
inquiry.1

   Nowhere in its opinion does the plurality discuss even a single one
of the five mitigating factors that were foreclosed from the jury’s con-
sideration, each one of which the plurality deems to be "colorable" in
its reasonable potential to affect the jury’s sentencing decision. And
it only selectively recounts the evidence from the record that would
support these factors, omitting altogether the substantial record evi-
dence that undermines the mitigating quality of the little evidence that
it does recount.

   Needless to say, nowhere in its opinion does the plurality even pur-
port to assess the effect that any single one of the foreclosed mitigat-
ing factors would have had, when balanced against the true weight of
the aggravating factors that supported the jury’s sentence of death; the
plurality simply states, without even a sentence of reasoning aside
from conclusory statements about the ways in which these factors
highlighted Allen’s "redeeming qualities," that "any one" of the five
mitigating factors may have swayed the jury’s verdict, Slip Op. at 45
— necessarily including, for example, the factors that the defendant
"remained employed when not incarcerated" and that Allen’s parents
had a fight at his third birthday party, twenty-seven years prior to his
murder of Trooper Worley.2
  1
     Even as to the statutory aggravators, the plurality says no more than
that the crime was "terrible," and that these aggravators were "exacerbat-
ing." Slip Op. at 44-45.
   2
     Chief Judge Wilkins, in his separate concurring opinion, does not,
himself, repeat the sweeping statement that "any one" of the foreclosed
mitigators might have supported a vote in favor of a life sentence. In his
opinion, Judge Wilkins rests instead on the argument that "[a] reasonable
juror could have concluded that Allen should be spared from the death
penalty so that his children would not lose their father." Slip Op. at 19.
Although this may be thought a narrower argument and opinion for pur-
6                             ALLEN v. LEE
   And, although it would perhaps go without saying, given that the
plurality does not even describe the crime for which Allen was con-
victed and sentenced, nowhere in its opinion does the plurality even
attempt to balance any of the aforementioned mitigating and aggra-
vating factors together with the particulars of the crime committed by
Allen.

   Confirmation of the fact that the plurality undertakes no balancing
of the mitigating circumstances that it holds could support a different
verdict, against the aggravating factors, is found in the completely
opposite treatment that it accords two mitigating factors that it holds
would not have affected the jury’s verdict. With respect to the submit-
ted mitigating factor regarding Allen’s age, for example, the court
undertakes a full and analytically sound discussion, ultimately con-
cluding that "a person of thirty cannot reasonably be characterized as
too young to appreciate the seriousness of his crime," and therefore
that no reasonable juror would have voted to spare Allen’s life on the
basis of Allen’s age. See Slip Op. at 43. Absent is any such discussion
of even one of the mitigating factors that the plurality holds could
have substantially swayed the jury’s verdict.

   That the plurality does not meaningfully discuss even a single one
of these relevant considerations — a detailed discussion of each of
which is essential to the Brecht analysis — itself confirms that its dis-
position not only is not, but could not possibly be, based on a rea-
soned reweighing of the aggravating and mitigating factors in light of
the entirety of the record facts, as required under Brecht. Rather, the
plurality ultimately conducts its harmlessness analysis as if in a fun-
house mirror, with the aggravating factors minimized to a fraction of
their true size due to the plurality’s failure even as much as to
describe Allen’s crime, and the mitigating factors magnified to a mul-
tiple of their size due to the plurality’s selective recitation of the
record evidence. These distortions alone, which are hopelessly
unfaithful to "the individualized treatment" that is required, render the

poses of this case, than the plurality’s argument and opinion, which
Judge Wilkins inconsistently joins, this argument is in actuality the most
far-reaching and detrimental argument for our court’s future harmless
error jurisprudence of any made by any member of the court in this case.
                            ALLEN v. LEE                              7
plurality opinion unmistakably in conflict with Supreme Court prece-
dent.

                                  B.

   But the plurality’s conclusory holding of harm under the peculiar
facts of this case produces an opinion even more fundamentally
flawed than one that simply fails to make the required "individual-
ized" weighing of the respective aggravating and mitigating factors in
light of the totality of the record facts. The core of the plurality’s
explanation that the faulty jury instruction resulted in "actual preju-
dice" to Allen, such as it is, reads in full as follows:

    [E]ven if eleven jurors agreed that five non-unanimous miti-
    gating factors were present, under the unconstitutional jury
    instruction they could have found no additional mitigating
    circumstances. Thus, instead of those eleven jurors weigh-
    ing eight mitigating circumstances against the two aggravat-
    ing factors, they were only permitted to weigh the three
    mitigating circumstances on which they were unanimous.
    Recognizing that all aggravating and mitigating factors do
    not warrant the same weight, and that jury deliberations are
    an inexact science at best, this means that a possible fifty-
    five additional votes (eleven jurors times five factors) could
    have been cast in favor of mitigation. Any one of these fifty-
    five possible votes could, in turn, have formed the basis for
    a decision against the imposition of the death penalty. When
    the substantial evidence presented is evaluated in light of the
    broad discretion conferred on jurors in capital sentencing
    proceedings, we find ourselves unable to say, as we must to
    uphold a sentence of death, that none of the jurors would
    have been persuaded to vote for life imprisonment instead.
    After all, it only takes one hold-out juror to prevent the
    imposition of the death penalty, and, in this case, we cannot
    say with ‘fair assurance’ that no juror would have been
    swayed by the mitigating factors that the jurors were unlaw-
    fully precluded from individually considering, including the
    highly discretionary catch-all factor, particularly when com-
    bined with the unanimously found mitigating factors.
8                             ALLEN v. LEE
Slip Op. at 45-46 (emphases added).

   It is clear from these paragraphs that the plurality does not merely
fail to rest its conclusion of harm on "an individualized determination
on the basis of the character of the individual and the circumstances
of the crime." It rests its conclusion exclusively on the polar opposite
of the individualized determination required — a generalized view
that any mitigating factor foreclosed from jury consideration, if not
obviously frivolous (like Allen’s age), is per se substantially injurious
to the jury’s verdict under Brecht, because any one juror could con-
ceivably find any one such mitigating factor and, on the basis of that
factor, refuse to sentence the defendant to death. In other words, the
plurality rests its conclusion on the premise that jury deliberation is
a game of random chance, that jurors are entitled to base their votes
on considerations other than the evidence — even irrational consider-
ations, in this case such as that Allen’s parents fought at his third
birthday party, twenty-seven years ago — and thus that any factor that
is foreclosed from the jury is necessarily prejudicial. That this is the
underlying rationale of the plurality’s opinion is clear from its con-
cluding observation that "[i]t only takes one hold out juror to prevent
the imposition of the death penalty," coupled with its earlier pro-
nouncement that "any one of the[ ] fifty-five [potentially foreclosed]
votes could, in turn, have formed the basis for a decision against the
imposition of the death penalty." Slip Op. at 45-46 (emphases added).3
    3
    So as to leave no doubt as to its holding in this regard, the plurality
even showcases three cases in which North Carolina juries sentenced
defendants to life imprisonment even though their crimes were especially
heinous and shocking to the public conscience, including (i) one in which
a defendant had been convicted of first-degree murder after killing a vic-
tim during the course of a rape, see State v. McKinnon, 

,
475 (N.C. 1991); (ii) one in which a defendant had been convicted of
first-degree murder after he enticed a victim away from a shopping mall,
forced her at knife-point to perform oral sex on him, and then stabbed
her repeatedly, see State v. Franklin, 

, 580-82 (N.C.
1983); and (iii) one in which a defendant had been convicted of first-
degree murder after stabbing his wife to death in front of the couple’s
children, see State v. Wilds, 

, 472 (N.C. App. 1999).
  In fact, of course, the plurality actually has no idea whether the sen-
tences imposed in the cited cases are examples of irrational jury behav-
                               ALLEN v. LEE                                9
   This is rejection of the reasoned decision required of an appellate
court under Brecht when reviewing for the harmfulness of an instruc-
tional error such as that in this case. This approach cannot possibly
be squared with the Supreme Court’s admonition to federal courts
conducting habeas review that, "[t]he State is not to be put to [the]
arduous task" of resentencing a defendant, "based on mere specula-
tion that the defendant was prejudiced by trial error." Calderon v.
Coleman, 

, 146 (1998) (per curiam). Neither can it be
defended in light of the explicit instruction in Kotteakos v. United
States, 

(1946), the very precedent from which the
Brecht standard was derived, that, when assessing the probable effect
of an error, the court must not assume generally that the jury would
act without reason. 

.

                                     II.

   Faithful application of the Brecht harmlessness standard in this
case can only yield the conclusion that the instructional error here was
harmless. Weighing the specific aggravating factors of Allen’s brutal
murder of Trooper Worley with the overall circumstances of the
crime, against the inconsequential, and only thinly-supported mitigat-
ing factors that were foreclosed from the jury’s consideration, and

ior, as opposed to rational jury behavior. The published opinions to
which the plurality cites report only the circumstances of the crime com-
mitted and the additional fact that a life sentence was imposed; they
report nothing about the mitigating circumstances presented to the jury,
and, therefore, nothing about the weighing process in which the jury ulti-
mately engaged in order to reach the decision to impose only a life sen-
tence. Thus, it is impossible to assess whether the juries in these cases
were acting rationally or irrationally.
   The plurality’s citation to these cases is telling, however, for this rea-
son. The plurality actually provides, through its parenthetical descrip-
tions, more detail about the aggravating circumstances in these irrelevant
cases than it provides about the aggravating circumstances in the one
case that is relevant — the case presently before the court — a curiosity
that is symbolic of the fundamental flaw in the plurality’s opinion in this
case, that it never even attempts the individualized reweighing of the
aggravating and mitigating factors required under Brecht.
10                            ALLEN v. LEE
weighing these together with the mitigating factors that were unani-
mously found to exist but unanimously found not to outweigh the
aggravating factors, there is a remote possibility (if any at all) that
Allen’s jury would have voted to spare Allen’s life had they not been
foreclosed from consideration of the remaining mitigating factors at
issue. Such a remote possibility does not begin to satisfy Brecht’s
requirement that the error have had a "substantial and injurious effect
on the jury’s verdict." See 

.

                                    A.

   As I have noted, the plurality does not even believe that the facts
surrounding Trooper Worley’s murder are relevant to the Brecht anal-
ysis, a belief which, by itself, stands in testament to the fact that it has
not undertaken the analysis required of us. But those facts are as fol-
lows.

   On May 11, 1985, in the midst of an interstate crime spree, Allen
pulled up behind the patrol car of North Carolina State Highway
Trooper Raymond E. Worley in a stolen van. Trooper Worley had,
only moments earlier, stopped a different stolen van — a vehicle that
Allen himself had helped to hot-wire and steal — driven by Antonio
Worrell, one of Allen’s partners in the crime spree. After escorting
Worrell into his patrol car, Worley motioned for Allen to exit from
the stolen van that he was driving and approach the patrol car, which
Allen did. Unbeknownst to Trooper Worley, however, Allen was
armed, and, when Trooper Worley reached to unlock the patrol car
door for him, Allen fired three shots at Worley at point blank range,
once through the head. Allen and Worrell then left Worley on the side
of the highway, slumped over in the patrol car. Literally drowning to
death in his own blood, Worley died some four minutes later, leaving
his wife without a husband and his two children without a father.

   It was with this particular crime in mind, which the North Carolina
Supreme Court described as "coldblooded, unprovoked, and unjusti-
fied," State v. Allen, 

, 871 (N.C. 1988), that the jury
unanimously found that the two aggravating factors of the murder of
a law enforcement officer while he was engaged in the performance
of his official duties and murder committed for the purpose of avoid-
ing lawful arrest, had been proven beyond reasonable doubt. J.A. 102-
                              ALLEN v. LEE                             11
03. It was with this particular crime in mind that the jury unanimously
found that those mitigating factors that it had unanimously found —
that Allen had expressed remorse for the murder, that Allen had
earned a GED while incarcerated, and that Allen had no history of
crime involving deadly weapons — were insufficient to outweigh the
vileness of the murder Allen had committed, and accordingly imposed
the death penalty. And it is with the details of this crime in mind that
we must, in conducting harmlessness review under Brecht, assess the
likelihood that any of the jurors would have voted to spare Allen’s
life, had they not been instructed that they could only weigh mitigat-
ing circumstances that they had unanimously found to exist.

                                   B.

    The plurality believes that, because "the jury’s process was care-
fully circumscribed by statutory dictates, limiting deliberation to two
aggravating factors, that the murder was committed for the purpose
of avoiding a lawful arrest and that the murder was of a law enforce-
ment officer while engaged in the performance of his official duties
. . . there is no legal basis to weigh the overall circumstances of the
crime outside these two factors." Slip Op. at 42 (emphasis added).

   First, even if it were true that it is impermissible as a matter of law
to consider the details of Allen’s brutal crime "outside" the two aggra-
vating factors that were submitted and unanimously found, jurors
obviously would have considered the details of Allen’s crime when
assessing the strength of these two factors. Indeed, they were affirma-
tively required to do this, because they could not assess whether the
aggravating factors in this case were "sufficiently substantial to call
for the imposition of the death penalty," see N.C. Gen. Stat. § 15A-
2000(c)(2) (2003) (emphasis added), or whether the aggravating fac-
tors "outweigh[ed] any mitigating circumstance in a sufficiently sub-
stantial manner so as to call for the death penalty," see State v.
Prevatte, 

, 486 (N.C. 2002) (emphasis added), without
assessing the evidence supporting the particular aggravating factors
submitted.

   Understanding that which the plurality does not, the judge who pre-
sided over Allen’s sentencing instructed the jurors that they were to
decide "from all the evidence what value to give to each circumstance
12                            ALLEN v. LEE
and then weigh the aggravating circumstances so valued, against the
mitigating circumstances, so valued." J.A. 95 (emphasis added). Just
as it would be impossible to assess how much mitigating weight
should be accorded a factor such as Allen’s "rearing in a single-parent
home" without considering evidence of the role Allen’s father played
in Allen’s life even after Allen’s parents divorced, so also would it be
impossible to assess how much aggravating weight should be
accorded the factors "that the murder was of a law enforcement offi-
cer while engaged in the performance of his official duties" and "that
the murder was committed for the purpose of avoiding a lawful
arrest" without any consideration at all of (i) the reckless crime spree
Allen and his friends had been enjoying before Trooper Worley
pulled them over, which explains why Allen had a motive to kill in
order to avoid the lawful arrest he feared, (ii) that Trooper Worley
had pulled the van driven by Allen’s friend Worrell over to the side
of the road and had just escorted Worrell into his patrol car when
Allen approached, which explains why it can be said that Trooper
Worley was doing no more no less than performing his "official
duties" when he reached, unsuspecting, to let Allen into his car, and
(iii) that Allen thus took advantage of Trooper Worley’s defenseless
position to shoot him three times, once through the head, and then left
him literally to drown in his own blood, which explains what "the
murder . . . of a law enforcement officer" would have meant to any
reasonable juror who had already voted to find Allen guilty as
charged of murdering Trooper Worley. Only with these details firmly
in mind, details of which the plurality holds are irrelevant as a matter
of law, could a juror possibly have begun to engage in any reasoned
assessment of the strength of the aggravating factors that the state had
submitted.

   But, of course, it is not true that "there is no legal basis to weigh
the overall circumstances of the crime outside the[ ] two [aggravating]
factors." See Slip Op. at 42.4 In fact, as North Carolina statute makes
  4
    There is one respect in which the plurality opinion can actually be
considered an improvement of the panel opinion with respect to the
aggravating circumstances. While Judge Gregory’s plurality opinion now
at least recites the literal language of the aggravating factors that were
submitted to the jurors, Judge Gregory’s panel opinion originally issued
a writ of habeas corpus requiring Allen’s resentencing without even recit-
                              ALLEN v. LEE                               13
perfectly clear, "[i]n the [sentencing] proceeding there shall not be
any requirement to resubmit evidence presented during the guilt
determination phase of the case, unless a new jury is impaneled, but
all such evidence is competent for the jury’s consideration in passing
on punishment." N.C. Gen. Stat. § 15A-2000(a)(3) (2003). And all of
the evidence that the jurors heard in the guilt determination phase of
the case, including the medical examiner’s testimony that, as a result
of the wounds which Allen inflicted, "Worley’s lungs were hyperin-
flated due to blood rushing into the airways, essentially drowning him
in his own blood," State v. Allen, 

, 858 (N.C. 1988),
and the testimony that Worley actually had a wife and two children
whom Allen had deprived of a husband and father, was relevant to the
assessment the jurors were required to make of the strength or weak-
ness of any particular mitigating factor which they found to exist. It
is for this reason, of course, that the trial court instructed the jury that
it "must decide from all the evidence what value to give to each cir-
cumstance." J.A. 95 (emphasis added); see also State v. Goodman,

, 590 (N.C. 1979) ("The jury is charged with the heavy
responsibility of subjectively, within the parameters set out by the
statute, assessing the appropriateness of imposing the death penalty
upon a particular defendant for a particular crime.") (emphasis
added).

   Ultimately, the inquiry under Brecht requires that we assess the
effect that the error had on the reasonable juror during the sentencing
proceedings in this case. Such a juror would not have either forgotten
or ignored the details of the heinous crime Allen committed, for the
conviction of which that very juror had himself just voted during the
guilt phase of Allen’s trial. And as the foregoing paragraphs demon-
strate, neither would such a juror have in any way been legally fore-
closed from considering the details of this crime when assessing the
weight to give each of the submitted aggravating and mitigating fac-
tors in order to determine whether Allen should live or die for the

ing the literal language of the aggravating factors that were submitted
to the jury and unanimously found. See Allen v. Lee, 

, 656-
58 (4th Cir. 2003). But, of course, as noted previously, the panel opinion
did not even appreciate that the writ could not issue except upon a find-
ing that the instructional error was harmless under Brecht.
14                            ALLEN v. LEE
particular crime he had committed. To the contrary, the jury was
charged by the judge to make precisely such an assessment "from all
the evidence." J.A. 95 (emphasis added). The jury having been so
charged, it is no more justifiable now for the plurality to ignore the
details of Allen’s crime while deliberating over whether Allen’s sen-
tence should be vacated, than it would have been for a juror to ignore
the details of Allen’s crime, while deliberating over the sentence
Allen deserved.

                                   C.

   While, alone, the plurality’s refusal to consider the particulars of
Allen’s crime in the course of reversing Allen’s sentence is fatal to
its finding of non-harmlessness under Brecht, the individual weakness
of each of the foreclosed mitigating factors underscores even further
the magnitude of the plurality’s error in this case.

   As, the plurality itself reluctantly is forced to acknowledge, neither
the first nor second foreclosed mitigating factors, that Allen "ha[d] no
significant history of prior criminal activity" and that Allen was thirty
years old at the time that he murdered Worley, could reasonably have
been considered sufficiently mitigating by a juror as to warrant life
imprisonment rather than death, in light of the aggravated nature of
Allen’s murder of Worley. J.A. 100.

   As to the first of these mitigating factors, Allen’s criminal history
record is not one that would prompt thought of mitigation. The record
discloses that Allen had multiple prior convictions, for shoplifting,
breaking and entering, and two counts of second degree burglary, R.
3362.

   Neither would the second mitigating factor, Allen’s age, lead any
juror to vote for a life sentence for Allen. Indeed, at the time of the
murder, Allen was neither juvenile, nor elderly; rather, born to Peters-
burg, Virginia parents in 1955, Allen was, as noted, thirty years old
at the time that he murdered Trooper Worley. It would, of course, be
entirely irrational for a juror to consider this fact one in mitigation of
the brutal murder of a state trooper while he was engaged in the dis-
charge of his official duties.
                             ALLEN v. LEE                            15
   The third foreclosed factor was that "Timothy Lanier Allen was
reared in a single-parent home." J.A. 100. No more than the first two,
would this mitigating factor ultimately be considered mitigating by a
reasonable juror. Again, the plurality only recites that evidence from
the record that would arguably support its finding that "substantial
evidence" exists in the record on the basis of which a juror could find
several of the mitigating factors; it ignores all of the contrary record
evidence.

   As to this mitigating factor, the plurality relates that Allen "only"
saw his father three or four times a year after his parents divorced and
until Allen entered sixth grade. Slip Op. at 44. The plurality omits to
mention, however, that, although Allen’s parents had divorced when
Allen was three, Allen’s father remained financially supportive and in
close contact with his children after the divorce. R. 3130-31. In fact,
Allen’s mother testified, and the jurors in the sentencing proceeding
heard, that after Allen’s mother moved with her children from Peters-
burg, Virginia, to Washington, D.C. (where Allen’s father had relo-
cated after the divorce) shortly before the start of Allen’s sixth-grade
year, Allen’s father "was over to the house each and every other day
to see the children and to see that they was minding, that they were
in school, and were they getting their lessons, and if anything went
wrong, he always chastised them." R. 3131. And, when Allen came
of age to work, his father gave Allen and his first wife a job at the
grocery store that Allen’s father had established and owned. R. 3075.
Thus, although the plurality would like to leave one with the impres-
sion that Allen did not have a strong male role model, such is any-
thing but the case.

   Additionally, Allen’s mother testified that Allen had received a
church education and had "plenty" while growing up. R. 3143-44.
And as for the difficulties of "single-parent rearing," Allen’s mother
further stated that she did not raise her children alone; the children’s
grandmother lived with them at all times, and looked after them when
Allen’s mother was at work. R. 3127, 3129. The plurality mentions
none of this obviously relevant evidence, even as it claims that "any
one" of the five factors it deems "colorable" — necessarily including
Allen’s "single-parent rearing" — could have led a juror to vote for
a life sentence.
16                           ALLEN v. LEE
   In light of all of this evidence, and Allen’s admission that he had
suffered no physical abuse or neglect from his parents while growing
up, R. 3087, it is unsurprising that the jury did not unanimously vote
to consider Allen’s rearing in a single-parent home to be a mitigating
factor. It would have been affirmatively surprising for any juror to
find that Allen’s rearing should be accorded any mitigating weight at
all in light of the specific crime that Allen had committed.

   The fourth and sixth foreclosed mitigating factors were that "Timo-
thy Lanier Allen is the father of three sons" and that "Timothy Lanier
Allen was a supporting parent of at least one of his children." J.A.
100. While it is true as a factual matter that Allen fathered three sons,
Allen lived with exactly none of his offspring. And in the very miti-
gating factor submitted to the jury, he only claimed before the court
to be a supporting parent of one of them. The juxtaposition of these
two mitigating factors — that Allen had three sons, but supported
only one of them — would more likely have evoked distaste than
sympathy.

   And even accepting for the sake of argument that the evidence of
Allen’s close relationship with his eldest son is the most relevant miti-
gating evidence the jury was precluded from considering, the possibil-
ity that a juror would have spared Allen’s life on the strength of this
relationship is at best remote, in light of the fact that Allen himself
took the life of not only a husband, but a father of two children, when
he left Trooper Worley mortally wounded, all in a vain attempt to
escape responsibility for the senseless crime spree on which Allen had
embarked. R. 1259.

   The fifth foreclosed mitigating factor was that "Timothy Lanier
Allen remained employed during the times he was not incarcerated."
J.A. 100. Obviously this provides an exceedingly uncomfortable basis
for mitigation. The very statement of this factor is an unsettling
reminder to the jury that there were other periods in Allen’s life dur-
ing which he was doing time in prison for other crimes, precisely the
kind of evidence that would, if anything, reinforce the jury’s consid-
ered sentencing conclusion. It is for this reason that "the dissent [did]
not contest" whether or not Allen truly was able to "remain[ ]
employed when he was not incarcerated" see Slip Op. at 44, not for
any reason from which the plurality should draw encouragement.
                              ALLEN v. LEE                             17
   The seventh factor not considered by the jury was the catch-all,
"[a]ny other circumstance or circumstances arising from the evidence
which you, the jury, deem to have mitigating value." J.A. 101. As
with every other of the mitigating factors, the majority refrains from
stating that any juror would have been likely to rely on this factor in
voting for a life sentence. But it recites the following as evidence that
might have been considered under this catch-all:

      Allen presented substantial evidence that his parents
      engaged in physical fights in his presence, which petrified
      him. In fact, their ultimate separation and divorce resulted
      from one such fight that occurred during Allen’s third birth-
      day party in which the police were called. Similarly, the jury
      heard evidence that Allen was a shy, timid person, bullied
      by others, who did not use firearms, and was unwilling or
      unable to fight back. And although Allen was never physi-
      cally abused as a child, from an early age he lost himself in
      alcohol and drugs.

Slip Op. at 44.

   None of this evidence would have remotely influenced a reason-
able juror to spare Allen’s life. It denigrates the seriousness of Allen’s
offense for the majority even to intimate that, for example, a juror
would have found that a fight that occurred between Allen’s parents
at Allen’s third birthday party, twenty-seven years before he mur-
dered Trooper Worley, or that Allen was "shy" would be relevant to
Allen’s culpability in taking the life of Raymond Worley, much less
sufficiently dispositive to warrant a life sentence.5 Obviously, Allen’s
  5
   Suffice it to say, the plurality also dramatically overstates the value
of the testimony that Allen was a "shy, timid person . . . , [who] was
unwilling or unable to fight back." Slip Op. at 44. This testimony was not
uncontroverted. The state expressly rejected the characterization of Allen
as a "shy, timid person" at sentencing, and with good reason. See R.
3284-85. The very circumstances of Allen’s murder of Trooper Worley
— and the crime spree that precipitated it — serve to refute a description
of the defendant as "shy and timid," as do his numerous, prior criminal
convictions for breaking and entering and for burglary. And this very
example shows yet another way in which evidence of the circumstances
of Allen’s crime, which the plurality refuses even to acknowledge, is
essential to any candid assessment of the weight of the mitigating evi-
dence submitted.
18                           ALLEN v. LEE
does not even begin to approach the type of "nightmarish" childhood
that might have led a juror to believe Allen had no option but to lead
a life of violence and self-destruction that ultimately culminated in the
murder of which Allen was convicted. Williams v. Taylor, 

, 395 (2000). It bears repeating that Allen testified that he had suf-
fered no physical abuse or neglect at his parents’ hands, and that his
parents got along most of the time when they were together. Compare
Wiggins v. Smith, 

, 2542 (2003) ("Wiggins experi-
enced severe privation and abuse in the first six years of his life while
in the custody of his alcoholic, absentee mother. He suffered physical
torment, sexual molestation, and repeated rape during his subsequent
years in foster care. The time Wiggins spent homeless, along with his
diminished mental capacities, further augment his mitigation case.");

(revealing that "Williams’ parents had been
imprisoned for the criminal neglect of Williams and his siblings," and
that "Williams had been severely and repeatedly beaten by his
father").

   In fact, although I am sure she hoped, as would any mother, that
the jurors would spare her son’s life, I doubt even Allen’s own mother
would have agreed with the plurality’s claim that Allen’s childhood
ought be considered mitigating with respect to the terrible crime Allen
had committed. Such, indeed, is the only inference one can reason-
ably draw from the following colloquy which occurred between one
of the prosecutors and Allen’s mother at the sentencing proceedings,
and which the jurors charged with sentencing Allen heard:

     Q. Did you have occasion to talk with [Allen] about this
     stealing people’s things —

     A. Yes.

     Q. Anytime?

     A. Yes.

     Q. What did you tell him about stealing people’s things?

     A. What doesn’t belong to him, he was taught never to put
     his hand on nothing that did not belong to him.
                             ALLEN v. LEE                            19
    Q. How about—

    A. Because he had no reason to do any of that if he done
    these things he had no reason to do it. We had plenty and
    he had no reason to have to go to any of that.

R. 3142-43 (emphasis added).

                                  D.

   Aware of the weakness of each mitigating factor individually, the
plurality attempts to buttress its reasoning by emphasizing "the possi-
ble cumulative impact of the additional mitigating factors." Slip Op.
at 46 (emphasis added). Thus, the plurality hedges its reliance on the
bold statement that "any one" of the five foreclosed factors might
have supported a vote for a life sentence with the fall-back position
that "even if a juror might not have found that each of those additional
factors independently outweighed the aggravators, a reasonable juror
well could have concluded to the contrary when considering those
factors collectively, and in addition to the three unanimously found
mitigators." 

nothing more than unabashed speculation of the kind that
the Supreme Court has admonished cannot justify putting the state to
the arduous task of resentencing at any time, let alone eighteen years
after a jury has rendered its verdict. See 

.
Having never even attempted to assess the weight of the aggravating
factors, and having never candidly assessed the strength of the indi-
vidual mitigating factors, the plurality actually has no idea what
cumulative impact the foreclosed mitigators would have had, if not
for the impermissible unanimity instruction. But the nature of Allen’s
crime, and the strength of the aggravators that the state submitted and
the jurors did unanimously find, make it highly improbable that a
juror would have been persuaded to spare Allen’s life even by the
cumulative impact of all the foreclosed mitigators.

                                  E.

  In sum, on this record, actually balancing the aggravating and miti-
gating factors against the backdrop of the record facts of Allen’s cold-
20                           ALLEN v. LEE
blooded murder of Trooper Worley, which the plurality never even
attempts to do, no judge could claim that the instructional error in this
case had a "substantial and injurious effect" on the jury that sentenced
Allen to death.

                                  III.

    With good reason unwilling to join in the plurality’s untenable
finding of a "substantial and injurious effect," Judge Traxler rests
decision on the narrower ground of "grave doubt," concluding that the
evidence is "in ‘virtual equipoise as to the harmlessness of the
[McKoy] error.’" Slip Op. at 26-27. While even Judge Traxler’s con-
clusion is ultimately unpersuasive in light of the palpable weakness
of both the mitigating factors themselves and the evidentiary support
for these factors, and the contrasting strength of the aggravating cir-
cumstances, one can hardly help but be struck by the many ways in
which Judge Traxler’s opinion confirms the indefensibility of the plu-
rality’s opinion, and thus the precariousness of the majority’s judg-
ment. Among other things, clearly understanding the plurality’s error
in refusing even to describe the circumstances of Allen’s crime, Judge
Traxler describes those circumstances in detail, up to and including
the medical examiner’s statement that because of the mortal wounds
Allen had inflicted, "Worley’s lungs were hyperinflated due to blood
rushing into the airways, essentially drowning him in his own blood."
Compare Slip Op. at 34-49 (plurality opinion) (never once describing
the circumstances of Allen’s crime) with Slip Op. at 20-22 (Traxler,
J., concurring in the judgment) (setting forth a detailed description of
the evidence of Allen’s crime which the jurors heard during the guilt
phase of the trial). Second, Judge Traxler notes and incorporates into
his analysis the fact that Allen was raised by his mother and his
grandmother, who lived with the family at all times, a fact that the
plurality entirely omits in holding that Allen’s "single-parent rearing"
could be relied upon to sentence Allen to life rather than death. Com-
pare Slip Op. at 43-44 (plurality opinion) (explaining that "Allen
offered evidence that from the time he was a young boy he was raised
in a single-parent home," but nowhere acknowledging that Allen’s
grandmother helped raise him) with Slip Op. at 27 (Traxler, J., con-
curring in the judgment) (acknowledging that Allen was reared "by
his mother and grandmother"). Third, Judge Traxler acknowledges
and incorporates into his analysis Allen’s own mother’s testimony at
                             ALLEN v. LEE                             21
sentencing that Allen had "plenty," testimony that the plurality refuses
even so much as to cite. Compare Slip Op. at 43-45 (plurality opin-
ion) (stating that Allen had "exhibited some redeeming qualities —
including efforts to overcome a less than ideal childhood," while
omitting to recite the testimony of Allen’s mother that Allen "had
plenty and he had no reason to have to go to any of that") with Slip
Op. at 27 (admitting that "Allen’s mother testified that Allen received
a church education and had ‘plenty’"). And finally, flatly rejecting the
plurality’s holding that "any one" of five mitigating factors might
have influenced a juror to spare Allen’s life, Judge Traxler concludes
only that the evidence as to the harmlessness of the error is in "equi-
poise." Compare Slip Op. at 45-46 (plurality opinion) ("Any one of
the[ ] fifty-five possible [but foreclosed] votes could, in turn, have
formed the basis for a decision against the imposition of the death
penalty.") with Slip Op. at 26-27 (Traxler, J., concurring)
("[A]lthough I cannot say with certainty that the unanimity instruction
had a substantial and injurious effect or influence on the verdict of
death, the nature of the error in this case unquestionably leaves me in
grave doubt as to whether it was in fact harmless.").

   In effect, the juxtaposition of Judge Traxler’s analysis with the plu-
rality’s conclusory findings all but confirms to any reader the untena-
bility of the conclusion that the McKoy error in this case was harmful.

   Indeed, himself uncertain even about his tentative conclusion that
the mitigating evidence is enough to produce "grave doubt" about the
harmlessness of the error, Judge Traxler adds an alternative rationale
for his finding of "grave doubt," a prediction that had the North Caro-
lina Supreme Court itself recognized the error in its reliance on the
jury poll, it likely would not have engaged in any reweighing of the
evidence proffered during the sentencing phase. Rather, Judge Traxler
concludes, that court would have presumed that because the existence
of at least some of the mitigating factors submitted by Allen were
supported by evidence, the McKoy error was not harmless. From this
prediction, Judge Traxler contends, "[i]t seems perverse that we . . .
consider denying habeas relief from a North Carolina death sentence
imposed under North Carolina’s death sentencing scheme because we
believe, based upon our ‘reweighing’ of the evidence under Brecht,
that it should be upheld in the face of knowing that the North Carolina
Supreme Court did not find the McKoy error harmless for that reason
22                            ALLEN v. LEE
under the Chapman [v. California, 

(1967),] standard and
would not have performed such a reweighing itself." Slip Op. at 29
(Traxler, J., concurring in the judgment) (emphases in original).

   Though Judge Traxler purports to be aware of the difference
between the North Carolina Supreme Court’s application of Chapman
(to decide on direct review whether a constitutional error was harm-
less) and our application of Brecht (to decide on collateral review
whether a constitutional error was harmless and therefore warrants
the issuance of a writ of habeas corpus), see Slip Op. at 30, he fails
to recognize that this difference is fatal to his argument.

   What Judge Traxler alludes to, but nowhere explicitly states, is that
the Chapman standard, which is the standard applied by the North
Carolina Supreme Court in all of the cases to which Judge Traxler
cites, is a much higher standard for harmlessness than Brecht. While
Chapman requires proof "beyond a reasonable doubt that the error
complained of did not contribute to the verdict 

, for the error to be declared harmless, Brecht only requires the
court to assure itself that the error did not have "substantial and injuri-
ous effect or influence in determining the jury’s verdict," 

. The difference between these harmlessness standards is
no mere accident. In fact, the Supreme Court in Brecht explicitly
adopted a less stringent standard for harmlessness (and, thus, a corre-
sponding more stringent standard for harmfulness) in the context of
collateral review, explaining that "application of a less onerous
harmless-error standard on habeas promotes the considerations under-
lying our habeas jurisprudence." 

of a less oner-
ous harmlessness standard on collateral review, and not the
application (which Judge Traxler in this case alternatively advocates)
of the same harmlessness methodology the state court would likely
have applied, is the mechanism through which the Supreme Court has
instructed federal courts to show "comity and respect for the state
courts." Slip Op. at 46 (Traxler, J., concurring in the judgment).

   The reason why the difference between Brecht and Chapman is
fatal to Judge Traxler’s claim is this: Because of that difference, as
the Court emphasized in Brecht, "‘an error that may justify reversal
on direct appeal will not necessarily support a collateral attack on a
                              ALLEN v. LEE                             23
final judgment.’" 

(quoting United States v. Frady, 

, 165 (1982)).

   In this case, thus, it can be assumed arguendo that the North Caro-
lina Supreme Court would have ordered resentencing had it both rec-
ognized that the jury poll does not reveal anything about how the
individual jurors would have voted and then reached a conclusion that
the McKoy error was harmful under Chapman after eschewing any
reweighing of the evidence. But we, while conducting collateral
review, are still constrained to deny relief, because the error by any
reasonable standard must be considered harmless under Brecht.

   Ultimately, therefore, what Judge Traxler views as "perverse" is
nothing more than a logical consequence of the fact that the Brecht
standard allows more errors to be declared harmless than does the
more stringent standard under Chapman v. California.

                                   IV.

   Under Brecht v. Abrahamson, a federal court setting aside a sen-
tence of death owes the state court system and the jurors that reached
that determination a detailed explanation of the reasons why a reason-
able juror would have been swayed by the evidence she was pre-
cluded from considering, no less than the court owes the same to a
defendant when it concludes that a constitutional error is harmless and
denies to him the opportunity to be re-sentenced by a jury, see Cle-

.

   Instead of the detailed, reasoned explanation required under Brecht,
the plurality has offered only conclusory statements bereft of any
analysis or even discussion, in justification of the court’s order releas-
ing Allen from the death sentence imposed for his cold blooded mur-
der of Trooper Worley. In place of the reasoned weighing of
mitigating factors and aggravating circumstances that is required by
Brecht, and that until today has been required by this Circuit, it has
attempted to substitute a glib standard premised impermissibly on the
fact of occasional random or irrational juror conduct. And in place of
studied attention to the individualized circumstances of a petitioner’s
life and the particular circumstances of his crime, it has substituted
attention to such frivolous details as that a fight might have occurred
24                             ALLEN v. LEE
between parents at a birthday party as many as twenty-seven years
before the crime in question was even committed.

  An opinion that flouts more conspicuously the Supreme Court’s
decision in Brecht v. Abrahamson one would be hard-pressed to
imagine.

     For these reasons, I dissent.