Elawyers Elawyers
Washington| Change

United States v. Jeffrey MacDonald, 15-7136 (2018)

Court: Court of Appeals for the Fourth Circuit Number: 15-7136 Visitors: 45
Filed: Dec. 21, 2018
Latest Update: Mar. 03, 2020
Summary: PUBLISHED UNITED STATES COURT OF APPEALS FOR THE FOURTH CIRCUIT No. 15-7136 UNITED STATES OF AMERICA, Plaintiff – Appellee, v. JEFFREY R. MACDONALD, Defendant – Appellant. Appeal from the United States District Court for the Eastern District of North Carolina, at Raleigh. James C. Fox, Senior District Judge. (3:75-cr-00026-F-1; 5:06-cv-00024-F) Argued: January 26, 2017 Decided: December 21, 2018 Before MOTZ, KING, and HARRIS, Circuit Judges. Affirmed by published opinion. Judge King wrote the op
More
                                      PUBLISHED

                       UNITED STATES COURT OF APPEALS
                           FOR THE FOURTH CIRCUIT


                                       No. 15-7136


UNITED STATES OF AMERICA,

                     Plaintiff – Appellee,

              v.

JEFFREY R. MACDONALD,

                     Defendant – Appellant.


Appeal from the United States District Court for the Eastern District of North Carolina, at
Raleigh. James C. Fox, Senior District Judge. (3:75-cr-00026-F-1; 5:06-cv-00024-F)


Argued: January 26, 2017                                    Decided: December 21, 2018


Before MOTZ, KING, and HARRIS, Circuit Judges.


Affirmed by published opinion. Judge King wrote the opinion, in which Judge Motz and
Judge Harris joined.


ARGUED: Joseph Edward Zeszotarski, Jr., GAMMON, HOWARD & ZESZOTARSKI,
PLLC, Raleigh, North Carolina, for Appellant. John Stuart Bruce, OFFICE OF THE
UNITED STATES ATTORNEY, Raleigh, North Carolina, for Appellee. ON BRIEF:
Leslie K. Cooley, Assistant United States Attorney, Jennifer P. May-Parker, Assistant
United States Attorney, OFFICE OF THE UNITED STATES ATTORNEY, Raleigh,
North Carolina, for Appellee.
KING, Circuit Judge:

       Over the last four decades, we have repeatedly been called upon to review Jeffrey

R. MacDonald’s trial and convictions in the Eastern District of North Carolina for the

murders in 1970 of his pregnant wife and their two young daughters at Fort Bragg. In

April 2011, we remanded for further proceedings on two habeas corpus claims being

pursued by way of a successive 28 U.S.C. § 2255 motion — a prosecutorial misconduct

claim based on the newly discovered evidence of former Deputy U.S. Marshal Jimmy

Britt (the “Britt claim”), plus a freestanding actual innocence claim premised on the

results of DNA testing (the “DNA claim”). See United States v. MacDonald, 
641 F.3d 596
(4th Cir. 2011). Thereafter, in July 2014, the district court ruled that MacDonald has

failed to make the evidentiary showing necessary to pursue the Britt and DNA claims by

successive § 2255 motion, and that he has also failed, in any event, to establish the merits

of the Britt claim or the DNA claim. See United States v. MacDonald, 
32 F. Supp. 3d 608
(E.D.N.C. 2014). We granted MacDonald a certificate of appealability under 28

U.S.C. § 2253 and, as explained herein, now affirm the district court’s judgment denying

the § 2255 motion. 1



       1
          Between our 2011 decision and today, in March 2016, we also affirmed the
district court’s denial of MacDonald’s motion for additional DNA testing pursuant to the
Innocence Protection Act of 2004, 18 U.S.C. §§ 3600-3600A (the “IPA”). See United
States v. MacDonald, 642 F. App’x 264 (4th Cir. 2016). MacDonald had filed his IPA
motion in September 2011, early in the remand proceedings. The district court denied the
IPA motion in August 2014, after denying MacDonald’s § 2255 motion. See United
States v. MacDonald, 
37 F. Supp. 3d 782
(E.D.N.C. 2014).


                                             2
                                           I.

                                           A.

       As detailed in our 2011 decision, see 
MacDonald, 641 F.3d at 599-603
, the brutal

murders of MacDonald’s wife and two daughters occurred in the family’s Fort Bragg

apartment on February 17, 1970. At the time, MacDonald was a physician and Captain in

the U.S. Army Medical Corps whose training and work included emergency medicine

and surgery. He was initially charged with the murders by the Army, but those charges

were eventually dismissed. In January 1975, MacDonald was indicted by the federal

grand jury in the Eastern District of North Carolina. As a result of a series of pretrial

motions and interlocutory appeals, MacDonald’s trial did not begin until July 1979. At

the conclusion of the seven-week trial, the jury found MacDonald guilty of one count of

first-degree murder and two counts of second-degree murder. His convictions resulted in

three consecutive life terms of imprisonment and were ultimately affirmed on direct

appeal. Between 1984 and 1997, MacDonald filed a series of motions for postconviction

relief, including § 2255 motions. MacDonald did not succeed on any of those motions or

in the related appeals, except that, in 1997, we granted his request for DNA testing and

remanded for that limited purpose. 2



       2
         On direct appeal following the 1979 trial, we reversed MacDonald’s convictions
on the ground that his Sixth Amendment guarantee of a speedy trial had been
contravened by the government’s delay in obtaining the federal indictment. See United
States v. MacDonald, 
632 F.2d 258
(4th Cir. 1980). The Supreme Court reversed our
judgment and remanded for further proceedings. See United States v. MacDonald, 
456 U.S. 1
(1982). On remand, we assessed MacDonald’s remaining appellate contentions
(Continued)
                                           3
       The ensuing proceedings have involved the claims now before us, i.e., the Britt

and DNA claims. See 
MacDonald, 641 F.3d at 603-07
. In January 2006, before the

DNA testing was completed, we granted MacDonald prefiling authorization for the

successive § 2255 motion raising the Britt claim. See 28 U.S.C. § 2244(b)(3). That is,

we determined that the § 2255 motion makes a prima facie showing that it satisfies

§ 2255(h). See United States v. Winestock, 
340 F.3d 200
, 205 (4th Cir. 2003) (“The court

of appeals must examine the application to determine whether it contains any claim that

satisfies § 2244(b)(2) (for state prisoners) or § 2255[(h)] (for federal prisoners).”). In

pertinent part, a successive motion can be sustained under § 2255(h) if it contains a claim



and affirmed his convictions. See United States v. MacDonald, 
688 F.2d 224
(4th Cir.
1982).

        In 1984, MacDonald filed his first motion for postconviction relief, seeking a writ
of habeas corpus under 28 U.S.C. § 2255 and a new trial pursuant to Federal Rule of
Criminal Procedure 33. The district court denied relief, see United States v. MacDonald,
640 F. Supp. 286
(E.D.N.C. 1985), and we affirmed, see United States v. MacDonald,
779 F.2d 962
(4th Cir. 1985). Thereafter, in 1990, MacDonald filed his second motion
for postconviction relief, again asserting claims under § 2255. The district court rejected
those claims as lacking merit and being procedurally defaulted. See United States v.
MacDonald, 
778 F. Supp. 1342
(E.D.N.C. 1991). Without reaching the merits of the
claims, we affirmed the procedural default ruling. See United States v. MacDonald, 
966 F.2d 854
(4th Cir. 1992). In 1997, MacDonald moved under Federal Rule of Civil
Procedure 60(b) to reopen the proceedings on his second postconviction motion, and he
requested DNA testing. The district court construed the Rule 60(b) motion to partially
constitute an unauthorized successive § 2255 motion, otherwise refused to reopen the
proceedings under Rule 60(b), and withheld DNA testing. See United States v.
MacDonald, 
979 F. Supp. 1057
(E.D.N.C. 1997). By a 1997 order and a 1998 opinion,
we denied authorization for a successive § 2255 motion and affirmed the denial of Rule
60(b) relief, but we granted the request for DNA testing and remanded that matter to the
district court. See United States v. MacDonald, 
161 F.3d 4
(4th Cir. 1998) (unpublished)
(explaining 1997 order remanding for DNA testing).


                                            4
based on “newly discovered evidence that, if proven and viewed in light of the evidence

as a whole, would be sufficient to establish by clear and convincing evidence that no

reasonable factfinder would have found the movant guilty of the offense.” See 28 U.S.C.

§ 2255(h)(1). Pursuant to § 2244(b)(4), however, we left it to the district court to conduct

a more searching assessment of whether the Britt claim satisfies § 2255(h)(1).          See

Winestock, 340 F.3d at 205
(“When the application is thereafter submitted to the district

court, that court must examine each claim and dismiss those that are barred under

[§ 2244(b)(2) or § 2255(h)].”).

       In March 2006, shortly after the § 2255 motion was submitted to the district court,

the results of the DNA testing finally became available. MacDonald promptly moved to

add the DNA claim as a predicate to the pending § 2255 motion, as well as to have the

DNA test results considered as part of the “evidence as a whole” in the court’s

assessment of the Britt claim.      Separately, MacDonald submitted a mass of other

evidence — including evidence excluded at trial, evidence submitted with prior

unsuccessful motions for postconviction relief, and evidence more recently discovered —

that he contended was also part of the “evidence as a whole.”

       By its decision of November 2008, the district court ruled that the DNA claim’s

absence from the § 2255 motion at the time of prefiling authorization deprived the court

of jurisdiction over that claim. See United States v. MacDonald, No. 3:75-cr-00026

(E.D.N.C. Nov. 4, 2008), ECF No. 150 (unpublished). The court also excluded the DNA

test results and the other evidence submitted by MacDonald from its assessment of the

Britt claim. The court then decided, focusing on the facts alleged in support of the Britt

                                             5
claim and on the evidence admitted at trial, that MacDonald had not made the evidentiary

showing necessary to sustain the § 2255 motion.

      On appeal, we concluded that the district court had taken an “overly restrictive

view of what constitutes the ‘evidence as a whole,’” erroneously omitting the DNA test

results and other non-trial evidence from its assessment of the Britt claim.          See

MacDonald, 641 F.3d at 614
. We thus remanded “for a fresh analysis of whether the

Britt claim satisfies the applicable standard of § 2255(h)(1).” 
Id. 3 Additionally,
we

determined that the court had erred in deeming itself to be without jurisdiction over the

DNA claim, and that the court instead should have considered adding the DNA claim to

the § 2255 motion under Rule 15(a) of the Federal Rules of Civil Procedure. 
Id. at 615-
16. Rather than instructing the court to conduct a belated Rule 15(a) evaluation on

remand, we simply granted MacDonald prefiling authorization for the DNA claim.


      3
          With respect to the district court’s assessment of the Britt claim, our 2011
decision also recognized that the court had erred by applying the § 2244(b)(2)(B)(ii)
standard for state prisoners, rather than the § 2255(h)(1) standard for federal prisoners.
See 
MacDonald, 641 F.3d at 609-10
. We deemed that error to be “probably harmless”
because of the similarities between § 2244(b)(2)(B)(ii) and § 2255(h)(1). See 
id. at 610;
see also § 2244(b)(2)(B)(ii) (requiring dismissal of a claim raised in a successive 28
U.S.C. § 2254 application unless “the [newly discovered] facts underlying the claim, if
proven and viewed in light of the evidence as a whole, would be sufficient to establish by
clear and convincing evidence that, but for constitutional error, no reasonable factfinder
would have found the applicant guilty of the underlying offense”); In re Dean, 
341 F.3d 1247
, 1249 n.4 (11th Cir. 2003) (observing that § 2244(b)(2)(B)(ii) and § 2255(h)(1) are
“materially identical”). But see Case v. Hatch, 
731 F.3d 1015
, 1035 (10th Cir. 2013)
(concluding that § 2255(h)(1) contains crucial linguistic differences that render it more
lenient than § 2244(b)(2)(B)(ii)). In the end, we did not concern ourselves with the
harmlessness question, because it was necessary to remand for a fresh analysis of the
Britt claim in any event.


                                            6
                                           B.

       At bottom, by our 2011 decision, we remanded for the district court to assess the

Britt and DNA claims under § 2255(h)(1), and for such other and further proceedings as

may be appropriate. In so doing, we examined various requirements for successive

§ 2255 motions engendered by the Antiterrorism and Effective Death Penalty Act of

1996 (“AEDPA”).      Most importantly, we recognized that § 2255(h)(1) obliges the

movant “to proffer some new evidence in support of his habeas corpus claim.” See

MacDonald, 641 F.3d at 612
(citing § 2255(h)(1)’s requirement for “newly discovered

evidence”). Section 2255(h)(1) then requires the court to determine whether that new

evidence, “if proven and viewed in light of the evidence as a whole, would be sufficient

to establish by clear and convincing evidence that no reasonable factfinder would have

found the movant guilty of the offense.”

       To define what constitutes the “evidence as a whole” for purposes of a

§ 2255(h)(1) assessment, we relied on pre-AEDPA precedent from which § 2255(h)(1)

derived, including the Supreme Court’s decision in Schlup v. Delo, 
513 U.S. 298
(1995).

We discerned that “the ‘evidence as a whole’ is exactly that: all the evidence put before

the court at the time of its . . . § 2255(h)(1) evaluation.” See 
MacDonald, 641 F.3d at 610
.   We elaborated that the court must make its “§ 2255(h)(1) determination —

unbounded ‘by the rules of admissibility that would govern at trial’ — based on ‘all the

evidence, including that alleged to have been illegally admitted [and that] tenably

claimed to have been wrongly excluded or to have become available only after the trial.’”

Id. at 612
(alteration in original) (quoting 
Schlup, 513 U.S. at 327-28
). In other words,

                                           7
“the ‘court must consider all the evidence, old and new, incriminating and exculpatory,

without regard to whether it would necessarily be admitted under [evidentiary rules].’”

Id. (alteration in
original) (quoting House v. Bell, 
547 U.S. 518
, 538 (2006) (explaining

Schlup)). We cautioned, however, that the movant is not entitled to “the benefit of every

doubt.” 
Id. Rather, “the
court must give ‘due regard to any unreliability of’ the evidence

and ‘may have to make some credibility assessments.’” 
Id. at 612
-13 (quoting 
Schlup, 513 U.S. at 328
, 330).

       The § 2255(h)(1) determination, as we explained, requires the “court to assess how

reasonable jurors would react to the overall, newly supplemented record.”             See

MacDonald, 641 F.3d at 613
(quoting 
House, 547 U.S. at 538
). If the court rules in the

movant’s favor under § 2255(h)(1), the movant merely passes a procedural bar to having

his habeas corpus claim considered on its merits. That is, the movant remains “obliged to

prove . . . that claim before obtaining any § 2255 relief thereon.” 
Id. at 614.


                                             II.

       To inform its procedural and merits assessments of the Britt and DNA claims on

remand, the district court conducted a seven-day evidentiary hearing in September 2012

and obtained post-hearing memoranda from MacDonald and the government. In its

decision of July 2014, the court provided a lengthy and detailed chronicle of the evidence

underlying the Britt and DNA claims and the other 28 U.S.C. § 2255(h)(1) “evidence as a

whole.” See 
MacDonald, 32 F. Supp. 3d at 619-86
. At the outset, the court explained

that — rather than attempting to identify every piece of evidence it had reviewed — it

                                              8
endeavored to highlight the evidence deemed most significant by the parties. The court’s

decision described evidence admitted, and excluded, at the 1979 trial, as well as evidence

submitted between 1984 and 1997 with MacDonald’s various unsuccessful motions for

postconviction relief. The decision also detailed later-discovered evidence, including, but

not limited to, the evidence underlying the Britt and DNA claims.           We also have

conducted a comprehensive review of the record and now present our own summary of

key evidence. 4

                                            A.

       The 1979 murder trial, at the federal courthouse in Raleigh, focused on competing

accounts of the murders of MacDonald’s pregnant wife, Colette, and their daughters,

five-year-old Kimberly and two-year-old Kristen. 5 As MacDonald has described it, the

day preceding the murders — February 16, 1970 — was an ordinary one. Around 6:00

a.m., MacDonald completed a twenty-four-hour shift at a civilian hospital where he

“moonlighted” for extra income, and then he made the sixty-mile drive back to his

family’s Fort Bragg apartment. MacDonald ate breakfast at home, worked on official

military duties throughout the day on Fort Bragg, and, shortly after 5:00 p.m., took

Kimberly and Kristen to feed their pony. That evening, the family had dinner together in

       4
          We summarize the trial evidence in section II.A; the previously-submitted and
later-discovered postconviction evidence, excluding the evidence principally underlying
the Britt and DNA claims, in section II.B; the Britt claim evidence in section II.C; and the
DNA claim evidence in section II.D.
       5
        We use the indictment’s spelling of “Kimberly.” The correct spelling, however,
may be “Kimberley.”


                                             9
the apartment, and then MacDonald stayed there with Kimberly and Kristen while

Colette attended a class at the Fort Bragg campus of North Carolina State University.

Colette returned home at about 9:40 p.m. after stopping at a convenience store for milk,

and she and MacDonald watched television together in the living room while the girls

slept in their bedrooms.

       After Colette retired to the master bedroom, according to MacDonald, he finished

a television show and washed dishes in the kitchen. At some point, MacDonald heard

Kristen cry in her bedroom, and he took her a bottle of milk to get her back to sleep in her

bed. He also listened to the radio while lounging on the living room couch and reading a

crime novel into the first hours of February 17, 1970.        Around 2:00 or 2:15 a.m.,

MacDonald went to the master bedroom, only to discover that Kristen had crawled into

bed with Colette and wet MacDonald’s side of the bed. MacDonald carried Kristen to

her bedroom, took a blanket to the living room, and, within a few minutes, fell asleep on

the couch. As was usual, MacDonald had left two lights on — a light in the kitchen that

gave some illumination to the nearby living room, and a light in the family’s shared

bathroom between their three bedrooms.

       It is what transpired next in the MacDonalds’ apartment in the early morning

hours of February 17, 1970, that has been most vigorously disputed by MacDonald, on

the one hand, and by the government, on the other. In MacDonald’s account, his wife

and daughters were slaughtered by a group of intruders who also assaulted him.

MacDonald became aware of the intruders when he was awakened suddenly in the living

room by Colette’s and Kimberly’s screams and cries for help from the bedroom area at

                                            10
the other end of the apartment’s hallway. As MacDonald arose from the couch in

response, he saw three men with short, military-style haircuts and a woman with blond

hair and a white floppy hat who appeared to be carrying a lighted candle and chanting,

“Acid is groovy; kill the pigs.” MacDonald engaged in a battle with the male intruders in

the living room for less than a minute. The trio assaulted MacDonald with a baseball-bat-

like club, as well as a knife and an ice pick. During the encounter, MacDonald’s blue

pajama top was either ripped from around his back or pulled over his head. The blue

pajama top then became wrapped around his wrists and hands. Once that happened,

MacDonald used the pajama top to protect his chest from a series of punches and stabs.

He sustained one significant laceration before being knocked unconscious.          When

MacDonald awoke later lying facedown over the step between the floors of the living

room and the hallway leading to the bedrooms, the intruders were gone, and his family

was dead.

      On the word of MacDonald, his subsequent actions included the following, though

not entirely in this order: removing his puncture-hole-filled blue pajama top from his

wrists and hands; rotating between his lifeless wife and daughters in their respective

bedrooms while unsuccessfully attempting to revive them; pulling a knife from Colette’s

chest and throwing the knife away from her within the master bedroom; laying his blue

pajama top over Colette; covering Colette with a second item, evidently a white Hilton

hotel bathmat, that he took from a nearby green vinyl-upholstered chair; checking his

own wounds in the bathroom; and using the master bedroom and kitchen telephones to



                                           11
request help from medics and military police. Finally, MacDonald lost consciousness

once again, ending up on the master bedroom floor with Colette.

      MacDonald subsequently identified the alleged female intruder as a young local

drug user named Helena Stoeckley who had emerged as a suspect due to, inter alia, the

blond wig and floppy hat she was known to wear. Under the government’s theory,

however, MacDonald concocted the attack by intruders and staged the crime scene to

conceal that he was the perpetrator.    At trial, the government sought to prove that

MacDonald proffered a false exculpatory statement — a bogus and convoluted story —

evidencing consciousness of guilt. The prosecutors relied on physical evidence tending

to implicate MacDonald in the murders, as well as a dearth of evidence substantiating

MacDonald’s account of any homicidal home invaders, including Stoeckley.

                                          1.

      The government’s trial evidence reflected that MacDonald summoned help by

telephone at approximately 3:45 a.m. on February 17, 1970. The military police officers

dispatched to the MacDonalds’ apartment found an apparently unconscious MacDonald

next to and partly covering his wife Colette’s lifeless and bloodied body on the master

bedroom floor. The similarly lifeless and bloodied bodies of daughters Kimberly and

Kristen were on their beds in their bedrooms. Kristen had been stabbed more than thirty

times, and Colette and Kimberly had been both clubbed and stabbed repeatedly.

MacDonald also had multiple wounds, but just the single laceration of any significance,

in his right chest.   He was hospitalized for a resulting pneumothorax, or partially

collapsed lung.

                                          12
                                            a.

       As described by the first responders, it had been a cold, rainy, windy night, and

there had been little activity on Fort Bragg. When the military police officers arrived at

the MacDonalds’ ground-level apartment, the front door was locked, but an officer was

able to enter the apartment’s utility room through a pair of rear exterior doors — a closed

but unlocked screen door and a wood door that was wide open. In the master bedroom

adjacent to the utility room, the officers spotted MacDonald lying with Colette’s body on

the floor. During an immediate search of the remainder of the apartment, the officers saw

the bodies of Kimberly and Kristen on their beds.

       When found by the military police officers, MacDonald was clad only in blue

pajama pants. The officers initially thought MacDonald was dead, but then saw him stir

and gave him mouth-to-mouth resuscitation.          A struggle occurred as MacDonald

attempted to get up and the officers restrained him. MacDonald sought information about

the well-being of his wife and daughters. He also told the officers that he had been

stabbed and that four intruders had come into the apartment. According to the officers,

MacDonald indicated that the intruders were drug-crazed “hippies” and generally

described the three men and blond-haired woman in the floppy hat. Medics and officers

soon removed MacDonald from the apartment and transported him by ambulance to Fort

Bragg’s Womack Army Hospital.

       Around that same time, special agents and other investigators with the Army’s

Criminal Investigation Division (the “CID”) arrived at the apartment and began

processing the crime scene.      Among other investigatory activities, the CID team

                                            13
confirmed the deaths of Colette, Kimberly, and Kristen; drew outlines of their bodies;

removed the bodies from the apartment for autopsies; and took a multitude of

photographs, both with and without the bodies present. The investigators also collected

potential physical evidence, including four suspected murder weapons, a variety of

bloodstained household items, blood scrapings from immovable objects, and numerous

hairs, fibers, and latent palm and fingerprints.

       One of the suspected murder weapons seized by the CID team was a Geneva

Forge-brand paring knife with a bent blade that had been found on the master bedroom

floor. Others were an Old Hickory-brand paring knife and an ice pick discovered outside

the rear entrance to the apartment under a large bush. The last suspected murder weapon

was a virtually rectangular piece of lumber, approximately three feet long and just less

than two inches wide and two inches deep, located on the ground near the bush. Little or

no blood was visible on either knife or the ice pick, but blood was evident on the piece of

lumber. Additionally, the piece of lumber bore various fibers — including two purple

sewing threads — as well as some dried white- and cream-colored paint. Concerned

about the bloodstained piece of lumber’s exposure to continuing rain, a CID agent

marked its location on the ground, placed it in a clean cardboard box, and, without re-

entering the apartment, took it to his police car and secured it in the trunk.

       As observed by the CID, lumber had been used on projects within the apartment,

including homemade closet shelves.        There were scrap pieces of lumber inside the

apartment, in the utility room, and outside the rear entrance to the apartment, in an open

well and in a locked storage shed. Both a piece of lumber and a pair of used latex

                                              14
surgeon’s gloves in the storage shed were stained with dried paint that looked like the

paint on the bloodstained piece of lumber. Similar paint covered the headboard and

footboard of Kimberly’s bed and a wood slat supporting that bed.

       In the master bedroom, the CID team documented that Colette’s body was lying

on a shag area rug that covered much of the wood floor. Her head was close to, but not

touching, the green vinyl-upholstered chair.    Colette was clothed in pink two-piece

pajamas. There were thirty puncture holes and eighteen cuts in her pajama top, and no

puncture holes or cuts in her pajama pants. Blood was all over the pink pajamas,

including heavy stains below the knees of the pants.

       The pullover-style top matching MacDonald’s blue pajama pants was covering

Colette’s chest, and the white Hilton hotel bathmat was over her abdomen. Both the blue

pajama top and the bathmat were bloodstained. The blue pajama top was haphazardly

folded, with its right arm inside-out, and almost completely atop Colette. The exception

was the blue pajama top’s left arm, which trailed out over the floor. There were forty-

eight puncture holes in the blue pajama top, along with two cuts. The blue pajama top

also had a tear down its front panel from its V-neck and additional tears at seams in its

left side, left shoulder, and left sleeve and cuff. A bloodstained pocket that apparently

had been detached from the blue pajama top was resting on a bloodstained throw rug near

Colette’s feet.

       The word “PIG” was written in blood, in capital letters that were each about eight

inches tall, on the headboard of the master bed. The bed’s bottom sheet had a large urine

stain on one side and was partially loose from the mattress. The top sheet and bedspread,

                                           15
both bloodstained, were in a pile on the master bedroom floor near the door to the

hallway. Blood was spattered on the walls, ceiling, and some fixtures of the master

bedroom, such as radiators. There was also blood on various portions of that room’s area

rug, including under and around Colette’s body.         The master bedroom telephone’s

handset was off the hook and hanging by its cord over the side of the dresser holding the

telephone’s base.

       The CID team found wood splinters throughout the master bedroom.                  For

example, one splinter, approximately three inches long, was in a blood clot beneath

Colette’s head. Another splinter of similar length was near the largest bloodstain in the

master bedroom’s area rug. Splinters were on the portion of the area rug within the

outline drawn by investigators of Colette’s trunk and limbs, as well as the portion of the

area rug adjacent to Colette’s left hand and arm. A splinter was among debris collected

from Colette’s hands, which also included loose hairs and various fibers.

       Additionally, the CID team’s search of the master bedroom yielded sixty purple

sewing threads — similar in appearance to the threads stuck to the bloodstained piece of

lumber found outside the apartment — plus seventeen blue fabric yarns and a single blue-

black sewing thread. In many places where there were splinters, there were also threads

and yarns. Three of the purple threads were with the splinter in the blood clot beneath

Colette’s head, and three more purple threads were with the splinter near the largest

bloodstain in the area rug. Fifteen purple threads and one blue yarn were on the portion

of the area rug within the outline of Colette’s trunk and limbs, largely in the area that had



                                             16
been below her buttocks. Twelve purple threads and the single blue-black thread were on

the portion of the area rug adjacent to Colette’s left hand and arm.

       The locations of the other threads and yarns collected from the master bedroom

included the following: one of Colette’s forearms (two blue yarns); the throw rug near

Colette’s feet (three purple threads and four blue yarns); the bottom sheet and a

pillowcase on the master bed (nineteen purple threads and nine blue yarns); the portion of

the area rug near a corner of the master bed’s footboard (two purple threads); the floor

between the wall and the master bed’s headboard, in the area where “PIG” was written on

the headboard in blood (one purple thread); and the bedspread in the pile of bedding on

the floor (two purple threads and one blue yarn). The CID team retrieved a bloodstained

hair twisted around one of the purple threads from the bedspread, as well as two hairs

from the bottom sheet on the master bed.

       In the top sheet in the pile of bedding on the master bedroom floor, there was a

bloodstained piece of latex that appeared to be a finger section of a glove. There was

another bloodstained piece of latex on the area rug near Colette’s left elbow. Additional

pieces of latex were also found within the master bedroom.

       Elsewhere in the apartment, Kimberly’s body was tucked into the bed in her

bedroom, with the covers pulled up to her chest area. She was clothed in a bloodstained

nightgown that had no puncture holes or cuts. Kimberly was lying with the right side of

her head exposed and the left side of her head against a pillow. The CID team collected

fourteen purple sewing threads and five blue fabric yarns from Kimberly’s bed, including



                                             17
several that were found beneath the covers once they were pulled back. There also were

splinters on the bed and bloodstains and spatter on the nearby wall.

       Across the hallway, Kristen’s body was on the bed in her bedroom, with the

covers pulled up to her waist. Kristen was wearing two-piece pajamas and an undershirt,

all bloodstained. Nine cuts and eleven puncture holes were in her undershirt; twenty-five

cuts but no puncture holes were in her pajama top; and no cuts or puncture holes were in

her pajama pants. Among the debris collected by the investigators from Kristen’s bed

were one purple sewing thread and one blue fabric yarn found on the bedspread, and

several splinters discovered on the bottom sheet. There were bloodstains throughout

Kristen’s bedding, including a large soaking stain on her top sheet. There was blood

spatter on the wall above the bed, and a large pool of blood with other smaller

bloodstains on the wood floor next to the bed. A bloodstained and broken acrylic hair tie

was nearby on a bloodstained throw rug. On the wood floor at the entrance to the

hallway, there were two footprints made in blood — one of a bare left foot and the other

of a bare right foot — positioned as if their maker had been exiting Kristen’s bedroom.

       There was blood throughout the portion of the hallway between the bedrooms and

the family’s shared bathroom, including a large bloodstain on a rug covering the wood

hallway floor just outside the master bedroom. In the bathroom, in and around the sink,

there were blood drops, other bloodstains, and hairs, some bloodstained. There were also

bloodstains on a stepladder and a wicker stool. A mirror was hanging on the wall above

the bathroom sink. On the pair of sliding doors to the linen closet adjacent to the

bathroom, three small stripes of blood were located about six feet above the floor at the

                                            18
doors’ juncture. The linen closet held a large quantity of medical supplies, including

disposable scalpel blades and a variety of prescription drugs and other medications.

       At the other end of the hallway — in the area where MacDonald said he awoke

after battling intruders in the living room and being knocked unconscious — there was a

single bloodstain. That bloodstain was on the wood hallway floor just before the step

between the hallway (on the upper side) and the living room (on the lower side). Several

blue fibers or threads, clustered in a bunch or ball, were on the hallway floor in the

vicinity of the bloodstain.

       In the living room, the rectangular coffee table that sat in front of the couch was

overturned and resting on the long side that had been farthest from the couch. On the

living room floor beneath that side of the coffee table was the March 1970 issue of

Esquire magazine, covered by a boxed children’s game. There was a finger-size smear of

blood above the “q” and “u” in “Esquire” on the magazine’s cover page, which featured

the then-recent murders committed in California by Charles Manson and his followers.

Additional items on the living room floor included an overturned flower pot and

MacDonald’s eyeglasses, which had a pink fiber caught in a hinge and a small spot of

blood on one lens. The only other suspected bloodstains in the living room were three

stains on the wall above the couch. The CID team searched the living room and its area

rug for splinters and for fibers considered to be of evidentiary value, including purple and

blue-black sewing threads and blue fabric yarns like those found in the bedrooms and on

the bloodstained piece of lumber. The investigators’ search of the living room, however,

yielded no such fibers and no splinters.

                                            19
       There were two rooms off the living room, i.e., the dining room and the separate

kitchen. Nothing seemed to be askew in the dining room, where several Valentine’s Day

cards were on display atop a buffet. In the kitchen, however, drops of blood were on the

floor in front of the sink. The cabinet beneath the sink held a box of Perry-brand latex

surgeon’s gloves, and the underside of the sink within that cabinet was bloodstained.

There were also a pair of latex surgeon’s gloves and a pair of yellow dish-washing gloves

sitting beside the sink. The handset of the kitchen telephone was off the hook.

       Despite the CID team’s search for evidence consistent with MacDonald’s account

of murderous intruders, the investigators did not find anything matching MacDonald’s

description of the baseball-bat-like club used to attack him. There were no signs of

forced entry into the apartment or of the premises being rummaged for items to steal.

Indeed, many items that likely would have been irresistible to drug-abusing burglars —

such as jewelry, silver serving pieces, a rifle, and the prescription drugs and other

medications — were within easy view and unstolen. Although it was susceptible to

footprints, a sandy strip of soil that ran between a paved walkway and the apartment’s

exterior contained no footprints, including around the rear entrance into the utility room.

The only wet spots and grass on the floors inside the apartment were attributed to the first

responders.

       Nevertheless, the CID team did collect three samples of dried wax from the

apartment — found on the coffee table in the living room, a chair in Kimberly’s

bedroom, and Kimberly’s bedspread — that might corroborate the presence of the female

intruder described by MacDonald as perhaps carrying a candle. The investigators took

                                            20
fourteen candles from the apartment for comparison to the wax samples. They also

recorded dozens of latent palm and fingerprints.

      Most of the physical evidence amassed from the crime scene was examined at two

facilities, the CID laboratory at Fort Gordon in Georgia and the FBI laboratory in

Washington, D.C. The autopsies of Colette, Kimberly, and Kristen were performed

locally at Womack Army Hospital. Their clothing was removed prior to the autopsies,

and loose hairs, fibers, and other debris were collected from their bodies, all to be

examined with the other physical evidence.

                                             b.

      According to the trial evidence of the Womack Army Hospital pathologist who

conducted Colette’s autopsy, his examination revealed that Colette suffered numerous

blunt force injuries — that is, clubbing wounds — mainly to her head, arms, and hands.

Multiple blows to Colette’s head resulted in five separate ragged tears in the skin on her

forehead and scalp. Some of those lacerations were four or five inches long, and the skin

was gaping, so that Colette’s skull was easy to see. There was extensive bruising and

hemorrhaging in the soft tissue surrounding the lacerations, and there were additional

bruises across Colette’s chin and in her right temple area. Colette’s skull was fractured,

and her brain was somewhat damaged. The head wounds were consistent with a direct

frontal assault by someone wielding the bloodstained piece of lumber found on the

ground outside the MacDonalds’ apartment. Colette also had a sharp-angled pattern

bruise on her chest consistent with the same piece of lumber.



                                             21
       The pathologist determined that blunt force injuries to Colette’s arms and hands

appeared to be defensive wounds, incurred while she attempted to fend off blows to other

parts of her body. Both bones in her right arm were broken, a bone in her left arm was

broken in two places, and the skin of each arm was lacerated, scraped, and bruised. In

addition to blunt force injuries, there was a stab wound on Colette’s left hand that likely

was made by a sharp instrument.

       Colette’s autopsy revealed that she sustained additional, more significant stab

wounds in her neck and chest — wounds that were consistent with a knife. There were

nine knife-like wounds in Colette’s neck, going through the skin and into soft tissue.

Colette’s trachea was penetrated in two places, and her thyroid was lacerated twice.

There were seven knife-like wounds in Colette’s chest, going through the skin and chest

wall into the chest cavity. Both of Colette’s lungs and her pulmonary artery were

lacerated, and the chest wounds caused substantial bleeding. Each of the chest wounds

was fairly perpendicular, indicating that a knife blade went straight into Colette, at about

a ninety-degree angle to her body, while she was lying flat on the floor. Only one chest

wound was indicative of a twisting knife; each of the other knife-like wounds was

apparently made by a knife going straight in and coming straight out.

       The pathologist and an FBI laboratory analyst deemed Colette’s knife-like

wounds, as well as the cuts in the pink pajama top that Colette was wearing, to be

consistent with the Old Hickory knife found under a bush outside the apartment. Tests

performed by the laboratory analyst showed that the Old Hickory knife’s blade, being

sharp, produced “clean” cuts in fabric like those in Colette’s pink pajama top. The

                                            22
laboratory analyst also tested the Geneva Forge knife found on the master bedroom floor;

he observed that the bent blade of that knife was dull and thus produced “tearing” cuts

inconsistent with the pink pajama top’s clean cuts.

       Further autopsy findings were that Colette had twenty-one puncture wounds in the

front area of her chest and three puncture wounds in her upper left arm, each consistent

with an ice pick such as that found with the Old Hickory knife. Many of the ice-pick-like

wounds were superficial, and the five or so that passed through the chest wall did not

appear to involve serious damage to any organ. Like the knife-like wounds, the ice-pick-

like wounds were perpendicular and thus likely inflicted upon Colette while she was

lying flat on the floor. Furthermore, the ice-pick-like wounds were symmetrical, without

any tearing, suggesting that Colette’s body was motionless when the wounds were made

— a conclusion corroborated by the cleanness of the puncture holes in Colette’s pink

pajama top.

       The pathologist did not attribute Colette’s death to her ice-pick-like wounds or to

her extensive blunt force injuries. He explained that those injuries were survivable,

though clubbing wounds to the center of Colette’s head and to her chin may have

rendered her unconscious. In the pathologist’s view, Colette bled to death from some of

her knife-like wounds, specifically those in her chest.

       Another Womack Army Hospital pathologist performed the autopsies of five-year-

old Kimberly and two-year-old Kristen. According to that pathologist’s trial evidence,

Kimberly’s autopsy revealed that she sustained two major clubbing wounds — one to

each side of her head — consistent with the bloodstained piece of lumber. Those blunt

                                             23
force injuries involved external bruising and abrasions, the fracture of Kimberly’s nose,

and the protrusion of a small portion of bone through the left side of her face. The more

extensive external bruising occurred on the right side of Kimberly’s head, across her right

cheekbone and ear.     Beneath those bruises, the right side of Kimberly’s skull was

fractured. There were additional fractures at the base of her skull and associated bruising

of her brain. Kimberly also had eight to ten stab wounds in the right front side of her

neck that caused some bleeding and transected her windpipe.          Those wounds were

consistent with the Old Hickory knife and made at close to a ninety-degree angle.

      The pathologist opined that Kimberly’s clubbing wounds occurred first and that

the associated brain injury was the most prominent factor in her death. The brain injury

may have killed Kimberly immediately; if not, it likely left her near death and

unconscious.   According to the pathologist, the knife-like wounds were only then

inflicted upon Kimberly and could have contributed to her demise.

      The pathologist’s separate examination of Kristen showed that she had seventeen

gaping stab wounds — four in the center of her chest, one in her neck, two above one of

her shoulder blades, and ten in the middle of her back — apparently made by a knife at

about a ninety-degree angle. Related to those knife-like wounds, Kristen’s thyroid gland

and windpipe were lacerated, and her heart was penetrated four times, twice from the

front and twice from the back.        The pathologist and the FBI laboratory analyst

determined that the knife-like-wounds and the clean cuts in Kristen’s pajama top and

undershirt were consistent with the Old Hickory knife.



                                            24
       There were several additional cuts on Kristen’s hands and fingers, including a

deep and significant cut to the ring finger of her right hand, that appeared to be defensive

wounds or wounds inflicted ancillary to more serious injuries. Kristen also had multiple

puncture wounds, including five in the center of her chest that penetrated deeply and ten

in an “S” pattern on the right side of her chest that were mostly superficial. In the view

of the pathologist and the laboratory analyst, the puncture wounds and the puncture holes

in Kristen’s undershirt were consistent with an ice pick. The absence of puncture holes in

Kristen’s pajama top indicated that her assailant lifted the pajama top before inflicting the

ice-pick-like wounds. Ultimately, the pathologist concluded that Kristen died from the

knife-like wounds to her heart, which caused a critical loss of blood and may also have

stopped the heart from beating.

                                             c.

       The trial evidence established that, while the autopsies of Colette, Kimberly, and

Kristen were being performed, MacDonald was being treated at Womack Army Hospital.

He had arrived at the emergency room still wearing his blue pajama pants and with blood

smeared on his face and hands. The clinical technician who first examined MacDonald

noted his vital signs were stable and just one of his wounds — the laceration in his right

chest that resulted in the pneumothorax — required bandaging.            MacDonald’s blue

pajama pants were removed for examination of his lower extremities and left on the floor

of the treatment area. During that period, the technician observed that the blue pajama

pants were bloodstained and torn. After the CID agents had visited the hospital without



                                             25
taking the blue pajama pants as evidence, the technician discarded the pajama pants by

throwing them into a garbage can.

       A physician who examined MacDonald in the emergency room said that

MacDonald exhibited emotional distress and may have been delirious. In that regard, the

physician pointed to a statement MacDonald made acknowledging that he had found his

wife and daughters pulseless and lifeless, which was quickly followed by a contradictory

inquiry from MacDonald as to when his family would arrive at the hospital for treatment.

Otherwise, MacDonald seemed to have organized thought processes and to be alert as to

time and place.

       MacDonald was hospitalized for about a week for treatment of his pneumothorax,

which involved two separate insertions of chest tubes. A surgeon concluded, based on

the smooth edges of the related laceration in MacDonald’s right chest, that the laceration

was made by a sharp instrument.        According to MacDonald’s medical records and

hospital personnel who examined him, MacDonald also had a slightly raised contusion on

his left forehead with some abrasion of the skin, four punctate wounds in his left chest, a

laceration in his upper left abdomen through the fat layer to the rectus muscle, a shallow

cut in the skin on his left arm, and a superficial cut between two fingers of one hand. The

medical records and hospital personnel did not note or observe that MacDonald had any

other injuries, including any additional head wounds, any ice-pick-like wounds, or any

injuries to his back.




                                            26
                                            d.

       As explained at the trial, the continuing investigation conducted by CID and FBI

agents and laboratory analysts revealed that each member of the MacDonald family had a

different blood type in the ABO blood group system. Specifically, Colette had type A

blood, Kimberly had type AB, Kristen had type O, and MacDonald had type B. CID

laboratory analysts tested bloodstains and scrapings from the MacDonalds’ apartment to

determine blood type and thus the possible source of the blood among the four family

members. Notably, the laboratory analysts could not declare, for example, that any type

A blood collected from the apartment came from Colette, rather than from one of the

billions of other persons in the world with type A blood. The laboratory analysts could

state only that the type A blood was consistent with Colette.

       In examining the suspected murder weapons, the CID laboratory analysts observed

that the Geneva Forge knife found on the master bedroom floor had just a small amount

of blood on it, on the tip of its bent blade. No blood was detected on the blades of the

Old Hickory knife and the ice pick found outside the apartment, but when the blades were

removed, small amounts of blood were discovered on the handles. The blood on the

Geneva Forge knife gave a weak indication of being type A, and the blood on the Old

Hickory knife was type A, consistent with Colette. The blood on the ice pick could not

be typed. The visibly bloodstained piece of lumber found on the ground outside the

apartment was coated in both type A and type AB blood, consistent with Colette and

Kimberly.



                                            27
      Bloodstains on the white Hilton hotel bathmat that had been covering Colette’s

abdomen were also type A and type AB, consistent with Colette and Kimberly. The

same FBI laboratory analyst who examined Colette’s and Kristen’s clothing — and who

testified as a prosecution expert in hairs, fibers, and fabric damage, impressions, and

stains — examined the bathmat. He concluded that, given the configurations of two of

the bathmat’s bloodstains, they could have been made by the Old Hickory knife and the

ice pick, then wet with blood, being wiped clean on the bathmat.

      Based on interviews with persons who had previously visited the apartment, the

investigators determined that the Geneva Forge knife, the Old Hickory knife, and the ice

pick could have belonged to the MacDonalds. Colette’s mother said that she had used an

ice pick that had been kept in the MacDonalds’ kitchen, and a former babysitter for the

MacDonalds later corroborated that the family owned an ice pick.

      The investigators more confidently pinpointed the apartment as the source of the

bloodstained piece of lumber.     Their conclusion stemmed from evidence that the

bloodstained piece of lumber had previously been split from a “two-by-four,” and that

another fragment of the same two-by-four was being used as the wood slat supporting

Kimberly’s bed. A CID laboratory analyst who testified as a prosecution expert in wood,

paint, and fibers confirmed the link between the bloodstained piece of lumber and the bed

slat by comparing, inter alia, their grain patterns and annual growth rings. He also

matched the dried paint on the bloodstained piece of lumber to the paint covering the bed

slat and Kimberly’s headboard and footboard, and to the paint on the paint-stained piece



                                           28
of lumber and latex surgeon’s gloves in the locked storage shed outside the rear entrance

to the apartment.

      The CID laboratory analyst further observed that the splinter found near the largest

bloodstain in the master bedroom’s area rug — one of the splinters that was about three

inches in length — fit perfectly into the bloodstained piece of lumber. Moreover, the

laboratory analyst opined that the bloodstained piece of lumber could have been the

origin of additional splinters collected from the master bedroom, the bedrooms of

Kimberly and Kristen, and one of Colette’s hands, because the piece of lumber and the

splinters were similar in color, texture, and cell structure. Another laboratory analyst,

with the FBI, determined that some of the fibers on the bloodstained piece of lumber

were consistent with the throw rug in the master bedroom.

      Other analyses showed that the two purple sewing threads on the bloodstained

piece of lumber were consistent with the puncture-hole-filled and torn blue pajama top

that MacDonald had laid over Colette. Specifically, the CID and FBI laboratory analysts

who both specialized in fibers concluded that the purple sewing threads on the

bloodstained piece of lumber matched threads used in most of the seams of MacDonald’s

blue pajama top. Likewise, the seventy-five purple sewing threads littered throughout the

bedrooms were consistent with the blue pajama top’s seam threads. There were also

some blue-black sewing threads used in the blue pajama top, in the cuffs of the sleeves;

those threads matched the single blue-black sewing thread in the master bedroom. The

twenty-three blue fabric yarns in the bedrooms were consistent with yarns used in the



                                           29
construction of the blue pajama top’s fabric. The laboratory analysts noted that the blue

pajama top was well-worn and faded from being laundered many times.

       The FBI laboratory analyst determined that the blue pajama top was probably torn

by someone grabbing the left section of the V-neck and pulling down while the top’s

wearer was stationary, or by someone grasping the V-neck while the wearer was spinning

to the right. The investigators deduced that threads and yarns dropped from the blue

pajama top when it was torn and continued to fall in the tearing’s aftermath. Several

other threads and yarns remained barely attached to the blue pajama top.

       On the blue pajama top, there were more than twenty stains of type A blood,

consistent with Colette. Some of those bloodstains appeared to have once been part of

larger bloodstains that were subsequently intersected by tears, suggesting to the FBI

laboratory analyst that the staining occurred before the blue pajama top was torn. The

blue pajama top also had one stain (on the lower front panel) of type AB blood,

consistent with Kimberly, and one stain (on the left sleeve) of type B blood, consistent

with MacDonald.

       The detached pocket that had been resting on the throw rug in the master bedroom

was stained with blood of Colette’s type, as was the throw rug itself. The FBI laboratory

analyst confirmed that the pocket had been separated from the blue pajama top, and one

of the CID blood examiners deduced that the detachment occurred before the

corresponding area of the blue pajama top became bloodstained. In that regard, it was

significant to the CID laboratory analyst that the pocket was stained only on its front side,



                                             30
and not on its back side, though the corresponding area of the blue pajama top was

soaked in blood.

       The two cuts in the blue pajama top were in its lower right front panel and its

upper left back shoulder area. The FBI laboratory analyst concluded that, because the

cuts in the blue pajama top were “tearing” cuts, they were more likely made by the dull

Geneva Forge knife than by the sharp Old Hickory knife.

       From crime scene photographs, the FBI laboratory analyst determined that the

blue pajama top’s forty-eight puncture holes were in portions of the blue pajama top that

had covered Colette’s chest. There were seventeen puncture holes in the blue pajama

top’s back panel, nine in the right front panel, eight in the shoulder area of the right

sleeve, thirteen in the right sleeve, and one in the shoulder area of the left sleeve. The

only part of the blue pajama top that was not atop Colette — the left sleeve that trailed

out over the floor — contained no puncture holes.

       The FBI laboratory analyst observed that all forty-eight puncture holes in the blue

pajama top were consistent with the ice pick and “clean,” i.e., perfectly symmetrical with

no tearing. Consequently, the laboratory analyst concluded that the puncture holes had

been made in the blue pajama top while it was stationary. He explained that, had the blue

pajama top been in motion when the puncture holes were made, there would have been

tearing of the fabric yarns around the puncture holes.

       Based on the condition of the blue pajama top, the investigators came to suspect

that the puncture holes were made while Colette, then motionless, was being stabbed

twenty-one times in her chest with the ice pick.         In other words, the investigators

                                            31
suspected that the blue pajama top was already covering Colette when her ice-pick-like

wounds were inflicted. To test that theory, the FBI laboratory analyst and a colleague re-

folded the blue pajama top to match the crime scene photographs as nearly as possible,

including by turning the right sleeve inside-out and by leaving the left sleeve trailing. By

then lining up puncture holes in the blue pajama top and placing probes through them in

the pattern of Colette’s ice-pick-like chest wounds, the laboratory analysts ascertained

that the twenty-one thrusts of a weapon into Colette’s chest could have made the forty-

eight puncture holes in the blue pajama top as the weapon passed through overlapping

portions of the fabric.

       Other examinations conducted by CID laboratory analysts indicated that the word

“PIG” was written on the master bed’s headboard in type A blood, consistent with

Colette. It appeared that the writer used two fingers, and that those fingers were not bare

and may have been covered by a latex glove. The finger-shaped piece of latex in the pile

of bedding and another piece of latex near Colette on the master bedroom floor were

stained with type A blood. A forensic chemist concluded that the bloodstained pieces of

latex and others collected from the master bedroom were consistent with Perry-brand

latex surgeon’s gloves kept beneath the kitchen sink. The forensic chemist detected some

variations between the latex samples — including differences between even the known

Perry-brand gloves — and he deemed those variations to be insignificant.

       As for the pile of bedding on the master bedroom floor, there were a few drops or

spatter of type AB blood on the top sheet, consistent with Kimberly. Otherwise, the

stains on the top sheet and the bedspread were of type A blood, consistent with Colette.

                                            32
The FBI laboratory analyst with knowledge of fabric impressions and stains opined that

then-wet type A blood had made several impressions on the top sheet: two consistent

with bloodstains on the left and right cuffs of Colette’s pink pajama top; one consistent

with a bloodstain on the left cuff of MacDonald’s blue pajama top; two consistent with

bloodstains on the right cuff of the blue pajama top; one consistent with a bare and

bloody left shoulder; and two consistent with bloody left and right hands. The laboratory

analyst also concluded that the heavy stains below the knees of Colette’s pink pajama

pants, also of type A blood, were made by direct contact with bleeding wounds.

       The same FBI laboratory analyst, as a prosecution expert in hair comparison,

determined that the bloodstained hair twisted around the purple sewing thread in the

bedspread could have come from Colette. The laboratory analyst further concluded that

the two hairs on the bottom sheet of the master bed were consistent with Colette’s known

hair samples and had been forcibly pulled out by their roots. Regarding the hairs in

Colette’s hands, the laboratory analyst matched a hair in Colette’s right hand to Colette

and deemed a hair in her left hand to be unidentifiable.

       Spatter on the master bedroom walls, ceiling, and fixtures was of type A blood,

consistent with Colette, and blood on the master bedroom area rug was both type A and

type AB, consistent with Colette and Kimberly. The large stain on the hallway rug just

outside the master bedroom was of type AB blood, consistent with Kimberly.

       Most of the blood in Kimberly’s bedroom was her type AB, including the blood

spattered on the wall above her bed. High on that wall, however, there was a stain of

Colette’s type A blood.

                                            33
      The spatter on Kristen’s bedroom wall was of Colette’s type A blood, as was the

large soaking stain on Kristen’s top sheet and some of the other bloodstains on Kristen’s

bottom sheet and bedspread. Otherwise, Kristen’s bedding was stained with her type O

blood, except for a couple stains of Kimberly’s type AB blood. The broken acrylic hair

tie found on the throw rug in Kristen’s bedroom was stained with Kristen’s type O blood.

That hair tie matched others known to be worn by Colette and kept on her nightstand in

the master bedroom. The CID fiber examiner determined that a fiber in Colette’s left

hand corresponded with Kristen’s throw rug.

      All blood on Kristen’s bedroom floor and throw rug was type O, consistent with

Kristen — except the blood that made the prints of the bare left and right feet of someone

exiting the bedroom into the hallway. The blood that made the footprints was type A,

consistent with Colette. A CID laboratory analyst determined that the bloody right

footprint was unidentifiable, but he matched the bloody left footprint to MacDonald.

      The blood in the bathroom — in and around the sink and on the stepladder and

wicker stool — was type B, consistent with MacDonald. Several hairs removed from the

sink, including bloodstained hairs, were matched to Colette and Kimberly. One hair,

which was coated in tar or a similar substance, was determined to be unidentifiable. At

the end of the hallway between the bedrooms and bathroom, the three small stripes of

blood at the juncture of the linen closet doors were MacDonald’s type B.

      The bloodstain at the other end of the hallway, on the floor near the step down into

the living room, was of either Kristen’s type O or MacDonald’s type B. There was no



                                           34
discussion at MacDonald’s trial of a laboratory analysis of the clustered blue fibers or

threads found on the hallway floor near the bloodstain.

         In the living room, the stains on the wall above the couch — though initially

suspected to be bloodstains — proved not to be bloodstains after all. Moreover, none of

the actual blood collected from the living room was MacDonald’s type B. The smear of

blood on the March 1970 issue of Esquire magazine appeared to be a mix of type A and

type AB, consistent with Colette and Kimberly.          The spot of blood on the lens of

MacDonald’s eyeglasses gave a weak indication of being type O, consistent with Kristen.

The pink fiber caught in the eyeglasses’ hinge was not matched to anything in the

apartment.

         The drops of blood on the kitchen floor in front of the sink were type B, consistent

with MacDonald. Although CID laboratory analysts conducted tests confirming that the

substance on the underside of the sink was blood, the laboratory analysts did not type that

blood.

         A CID laboratory analyst determined that the three samples of dried wax collected

from the apartment did not match any of the MacDonald’s fourteen candles.                The

laboratory analyst further concluded that the three wax samples were not consistent with

each other, i.e., that each of the samples had a different source. Of the forty-three latent

fingerprints and twenty-five latent palm prints recorded by the CID team, examiners

identified twenty-six fingerprints and eleven palm prints as belonging to MacDonald or

another person known to have been in the apartment. The other seventeen fingerprints

and fourteen palm prints were identifiable but not matched to any person.             Those

                                              35
included prints on a jewelry box and window blinds, all in the master bedroom, and on a

drinking glass in the living room. The examiners deemed an additional twenty-four

fingerprints and seven palm prints to be insufficient for identification. Of note, there

were no prints on the Geneva Forge knife, the Old Hickory knife, the ice pick, or the

bloodstained piece of lumber. Several of MacDonald’s latent fingerprints had been left,

however, on the inside pages of the Esquire magazine.

       The investigators were struck by similarities between MacDonald’s account of

murderous intruders and the Esquire magazine’s articles about the Manson murders and

other cult-like activities.   For example, the magazine contained many references to

blonds, burning candles, and use of the words “acid” and “groovy.” In the Manson

murders, the victims were mutilated and the word “PIG” was allegedly smeared on a wall

in the victims’ blood. Those and other pertinent passages from Esquire were read into

the record during MacDonald’s trial.

                                           e.

       The trial evidence also included MacDonald’s prior statements and testimony,

presented through witnesses and written transcripts. MacDonald had been interviewed at

Womack Army Hospital by CID and FBI agents in the hours after the murders on

February 17, 1970, and again on February 18 and 19. About seven weeks later, on April

6, MacDonald submitted to further questioning by CID agents. The Army charged

MacDonald with the murders on May 1, and he thereafter voluntarily testified in his

Army proceedings, which began in June and ended in October 1970. MacDonald also



                                           36
testified before the federal grand jury, on five days in August 1974 and one day in

January 1975.

       The CID and FBI agents who interviewed MacDonald at Womack Army Hospital

acknowledged that he was at times emotional and unable to coherently describe what had

just occurred in his family’s apartment. During those interviews, however, MacDonald

committed to several details of his account of murderous intruders, including the

following: that the four intruders he saw consisted of a black man, two white men, and

the blond-haired woman, who was also white; that all of the intruders were wearing some

sort of shoes; that the female intruder was wearing boots that appeared to be black in

color, but were actually dark brown and wet; that the black male intruder assaulted him

with the club that he believed to be a baseball bat; and that, after the attack, he pulled a

knife from Colette’s chest and threw it away from her within the master bedroom. When

informed that there were bloody prints of bare feet in Kristen’s bedroom, MacDonald

acknowledged that he probably stepped in blood and made the prints as he rotated

between the bedrooms in his fruitless quest to revive his wife and daughters.

       By their subsequent questioning of MacDonald in April 1970, CID agents sought

to develop and clarify details of MacDonald’s account. MacDonald described being

awakened on the living room couch by Colette’s and Kimberly’s screams from the

bedroom area and continuing to hear their voices during his brief encounter with the

intruders. According to MacDonald, Colette pleaded for help and cried, “Jeff, Jeff, why

are they doing this to me?” Kimberly yelled, “Daddy,” over and over.



                                            37
       MacDonald said that, in addition to having short hair, each of the three male

intruders was of moderate height. He noted that one of the white men was slightly

shorter than the other, and the black man was wearing an Army field jacket with Sergeant

stripes.   As for the blond-haired woman who seemed to be carrying a candle and

chanting, “Acid is groovy; kill the pigs,” MacDonald said she was wearing a mini skirt or

shorts with her floppy hat and her boots, which ended just below her bare knees.

MacDonald stood by his earlier statements that the woman’s boots were brown, and he

denounced then-recent newspaper reports that the boots were white.

       Furthermore, MacDonald reiterated that the black male intruder assaulted him in

the living room with the baseball-bat-like club. Having recounted that he grabbed and

held onto the club during the melee, MacDonald conveyed some certainty that the club

was a baseball bat. He also recalled grabbing one of the white male intruder’s hands and

seeing a blade in that hand. MacDonald said that, after his blue pajama top became

wrapped around his wrists and hands, he used the pajama top to blunt the torrent of blows

from the two white men. Although MacDonald was unsure whether the pajama top had

been ripped from around his back or pulled over his head, he suggested it was more likely

that the top was torn off, as he had no memory of backing his head through it.

       According to his April 1970 interview, once he awoke after the attack, MacDonald

got up and walked straight down the hallway to the master bedroom, removing the blue

pajama top from his wrists and hands as he went.           MacDonald specified that he

immediately pulled the knife from Colette’s chest and covered her with the blue pajama

top in an effort, however illogical, to treat shock and keep her warm. Additionally,

                                           38
MacDonald recalled covering Colette with the item he took from the nearby green vinyl-

upholstered chair, and he acknowledged that item must have been the white Hilton hotel

bathmat. MacDonald stated that he could not remember moving Colette, but that it was

possible she had been leaning against the green chair and he moved her slightly — no

more than an inch or two — to lay her flat on the floor and attempt mouth-to-mouth

resuscitation.

       Without purporting to be sure of the exact order of his subsequent movements,

MacDonald recalled that he next went to the bedrooms of Kimberly and Kristen, and that

he then made a second rotation between the three bedrooms. MacDonald believed it was

during his second visit to the master bedroom that he used the telephone to call for help.

Probably after his second rotation between the bedrooms, he checked his own wounds in

the bathroom mirror, and he also washed his hands in the bathroom sink. Thereafter, he

again requested help, now by way of the kitchen telephone. The next thing he knew, he

was lying next to Colette on the master bedroom floor and being given mouth-to-mouth

resuscitation by a military police officer.

       MacDonald theorized during the April 1970 interview that the intruders had

entered the apartment through the rear doorway into the utility room. He explained that,

on one of his visits to the master bedroom, he stepped into the utility room and saw that

the wood exterior door was open. He also said that Colette usually locked the apartment,

but that she may have forgotten that night and he had not checked. Assuming the

intruders came through the utility room and launched their attack in the master bedroom,

however, MacDonald acknowledged that it was puzzling why he was not awakened by

                                              39
screams before the intruders reached the living room, and why the intruders left Colette

and Kimberly crying for help. MacDonald speculated that perhaps the intruders had

wounded Colette but thought they had killed her, and then Colette began screaming after

the intruders exited the master bedroom. Alternatively, MacDonald surmised that there

could have been two more intruders with Colette and Kimberly while the four other

intruders confronted MacDonald in the living room.

      When the CID agents posited that four or six intruders should have left more

evidence of their presence — more disorder and damage than just the “living day-to-day

mess” found in the apartment — MacDonald responded that the intruders he encountered

were not hyper and had dull affects. When the agents later stated that a multi-day search

had yielded no evidence of any stranger being in the bedrooms, MacDonald expressed

surprise and explained that, aside from his family and the intruders, there had been many

people in those rooms.

      Shown photographs in April 1970 of the four suspected murder weapons,

MacDonald asserted that he did not recognize the Geneva Forge knife, the Old Hickory

knife, the ice pick, or the bloodstained piece of lumber. MacDonald insisted that he had

never used an ice pick in the apartment, and that he did not believe an ice pick was kept

there. With respect to the bloodstained piece of lumber, MacDonald established with the

CID agents that it was essentially a “two-by-two,” and then he disclaimed possessing any

two-by-twos at the time of the murders. Even after being confronted with evidence that

paint on the bloodstained piece of lumber matched paint on other lumber and latex



                                           40
surgeon’s gloves in the locked storage shed, MacDonald persisted in denying that he

recognized the bloodstained piece of lumber.

      Throughout the April 1970 interview, the CID agents revealed physical evidence

to MacDonald and accorded him the opportunity to address it. For example, the agents

informed MacDonald that threads and yarns matching his blue pajama top were found

beneath Colette’s body. MacDonald could not explain how the threads and yarns got

there. When told that similar threads and yarns were also in Kimberly’s and Kristen’s

beds, MacDonald immediately responded, “Holy Christ,” and then suggested that the

threads and yarns may have been stuck to his bare arms and fallen off as he visited the

bedrooms and attempted to revive his daughters. As another possibility, MacDonald said

that the threads and yarns could have adhered to the intruders during their battle with

MacDonald in the living room and later dropped from the intruders after they returned to

the bedrooms to complete the murders of Colette, Kimberly, and Kristen. In proffering

his theories, MacDonald confirmed his recollection that he covered Colette with the blue

pajama top when he first went to the master bedroom, and that the pajama top was not

with him during his visits to his daughters’ bedrooms.

      In his subsequent testimony in the 1970 Army proceedings and before the federal

grand jury in 1974 and 1975, MacDonald stuck with his assertion that the blue pajama

top was either ripped from around his back or pulled over his head during his battle with

the intruders in the living room. Furthermore, MacDonald adhered to his account of

covering Colette with the blue pajama top on his first visit to the master bedroom after

the attack. MacDonald elaborated that he either dropped or threw the blue pajama top as

                                           41
he entered the master bedroom, and then he pulled the knife from Colette’s chest and

retrieved the pajama top from the master bedroom floor to lay over her. MacDonald

specified that he shook the blue pajama top to open it like a towel before covering Colette

with it. He also expressed certainty that Colette had been leaning against the green chair

and that he moved her to lay her flat on the floor. Whereas he had previously said that he

moved Colette no more than an inch or two, he now estimated that he moved her six

inches. Such details were fodder for the government at the 1979 trial, wherein it sought

to establish that MacDonald had made changes to his story to make it more consistent

with the physical evidence, such as the threads and yarns beneath Colette’s body.

                                            2.

       The government’s theory at trial was that, in the early morning hours of February

17, 1970, MacDonald got into an argument with Colette in the master bedroom —

perhaps because Kristen had wet MacDonald’s side of the bed, or perhaps because of

something else.   Whatever the reason for the argument, it quickly escalated into a

physical altercation. Colette may have used the Geneva Forge knife to protect herself

from MacDonald, leaving the two tearing cuts in the blue pajama top that he was then

wearing. Additionally, Colette may have used another object to hit MacDonald in the

head. Likely enraged and exhausted from his work and home activities, MacDonald

wielded the soon-bloodstained piece of lumber against Colette, poking it into her chest

and repeatedly swinging it at her as Colette raised her arms and hands in self-defense.

Kimberly was awakened by the fracas and afoot in the master bedroom. MacDonald then



                                            42
intentionally struck Kimberly with the piece of lumber on the left side of her head, or

accidentally did so as he swung at Colette.

       Under the government’s theory, all type A blood in the apartment was Colette’s,

all type AB was Kimberly’s, all type O was Kristen’s, and all type B was MacDonald’s.

The government deduced that when MacDonald struck Kimberly with the piece of

lumber, her blood sprayed onto the master bed’s top sheet, which was then on the bed,

and onto the front panel of MacDonald’s blue pajama top. The strikes that connected

with Colette spattered her blood on the master bedroom walls, ceiling, and fixtures. As

MacDonald and Colette brawled, MacDonald bled on his blue pajama top’s sleeve, and

Colette got blood all over the blue pajama top, in spatters and by direct contact. Its

pocket, probably already stained with Colette’s blood, became detached from the blue

pajama top, and then the pajama top was stained with more of Colette’s blood and

violently torn. Colette likely grabbed the left section of the blue pajama top’s V-neck and

caused the tearing, either by pulling down as MacDonald stood still or by holding on

while he spun to the right. Threads and yarns dropped from the blue pajama top when it

was torn and continued to fall in the master bedroom thereafter, landing on the master

bed and on the area rug and throw rug covering the floor. The blue pajama top shed

many of the threads and yarns when MacDonald — still wearing the pajama top —

swung the piece of lumber. At the same time, the piece of lumber shed splinters, leaving

numerous threads, yarns, and splinters interspersed.

       According to the government, MacDonald may have incurred injuries in the

master bedroom, including the contusion on his left forehead (possibly from being hit in

                                              43
the head by Colette); the four punctate wounds in his left chest (possibly made by

Colette’s fingernails as she grasped the blue pajama top); and the minor lacerations and

cuts in his upper left abdomen, on his left arm, and between two of his fingers (possibly

inflicted by Colette with the Geneva Forge knife). Colette and Kimberly sustained

serious clubbing wounds in the master bedroom, but they were not yet dead.

      By that time, the government surmised, MacDonald perceived that things had

gone too far, his family would never live together normally again, and his personal and

professional future was at stake. MacDonald thus resolved to kill his wife and daughters.

Inspired by the Manson murders, MacDonald employed a variety of weapons and staged

the crime scene to mimic a cult slaying and portray himself as a fellow victim and lucky

lone survivor.

      The government concluded that MacDonald moved Kimberly from the master

bedroom to her bedroom, as indicated by her bloodstains in the master bedroom and on

the hallway rug just outside the master bedroom. MacDonald tucked Kimberly into her

bed with the battered left side of her head against a pillow, leaving threads and yarns

from the blue pajama top beneath Kimberly’s covers. MacDonald then struck Kimberly

again with the piece of lumber, this time on the exposed right side of her head. When

MacDonald raised it to strike Kimberly, the piece of lumber — now wet with blood from

the attack in the master bedroom — hit the wall above Kimberly’s bed and transferred

some of Colette’s blood there. During the strike, Kimberly’s blood spattered on the wall,

and splinters and blue pajama top threads and yarns showered onto the bed. In addition



                                           44
to clubbing Kimberly in her bedroom, MacDonald also stabbed her there with the Old

Hickory knife eight to ten times in the neck.

       As for Kristen, the government characterized her killing — in her own bed, where

she had apparently slept through the preceding horror — as a particularly cold-blooded

murder. MacDonald stabbed Kristen seventeen times with the Old Hickory knife through

her pajama top and undershirt, and, having lifted Kristen’s pajama top, at least fifteen

times with the ice pick through her undershirt alone. During the attack, Kristen’s hands

and fingers also were cut.      A spot of her blood may have been propelled onto

MacDonald’s eyeglasses, which MacDonald may have been wearing at the time. Kristen

bled in her bed and on the floor nearby.

       The government theorized that, while MacDonald was in Kristen’s bedroom, the

badly injured Colette somehow got there to try to help her younger daughter. From the

evidence, the government inferred an encounter between MacDonald and Colette in

which Colette’s acrylic hair tie was ripped from her head, stained with Kristen’s blood,

and left broken on the floor. MacDonald struck Colette again with the piece of lumber,

spraying Colette’s blood on Kristen’s bedroom wall and scattering splinters on her bed.

Colette fell onto the bed — likely unconscious — and made the large soaking bloodstain

on Kristen’s top sheet. The piece of lumber also transferred Kimberly’s blood to some of

the bedding, and a blue pajama top thread and yarn fell on the bedspread.

       In the government’s account, MacDonald went to the master bedroom for the top

sheet and bedspread from the master bed, which he brought to Kristen’s room and laid on

the floor. MacDonald moved Colette onto that bedding, where she continued to bleed.

                                            45
As he bundled Colette to carry her back to the master bedroom, MacDonald stepped in

Colette’s blood on the bedding. MacDonald then made bloody prints of his bare feet on

the wood floor as he carried Colette out of Kristen’s bedroom and into the hallway —

explaining why the footprints were made in Colette’s type A blood, but the only other

blood on Kristen’s bedroom floor was her type O. Meanwhile, MacDonald and Colette

made impressions on the master bed’s top sheet, in Colette’s blood, with their bloody

blue and pink pajama top cuffs, a bare and bloody shoulder, and bloody hands. Colette

made heavy bloodstains below the knees of her pink pajama pants when she was slumped

over on Kristen’s bed or when she was folded up and being carried to the master

bedroom.

       The government deduced that, back in the master bedroom, MacDonald placed

Colette on the area-rug-covered floor, atop and amongst an array of splinters and blue

pajama top threads and yarns. MacDonald stabbed the motionless Colette sixteen times

with the Old Hickory knife, in her neck and chest. Perhaps seeking an explanation for the

immense amount of blood on his blue pajama top, and perhaps wanting an excuse for his

own shortage of stab wounds, MacDonald laid the blue pajama top over Colette’s chest

and stabbed her through it twenty-one times with the ice pick. He also used the ice pick

to stab Colette three times in her upper left arm. MacDonald wiped the Old Hickory

knife and the ice pick on the white Hilton hotel bathmat, leaving only a little of Colette’s

blood on the Old Hickory knife and a trace of blood on the ice pick so small it could not

be typed. He also laid the bathmat over Colette’s abdomen. By then, the bloodstained

piece of lumber bore two threads from the blue pajama top and fibers from the master

                                            46
bedroom throw rug, apparently acquired when the piece of lumber was resting on the

throw rug sometime during the slaughter.

      Thereafter, according to the government, MacDonald discarded the weapons he

had used — the Old Hickory knife, the ice pick, and the bloodstained piece of lumber —

by throwing them from the utility room’s exterior doorway into the apartment’s rainy and

wet backyard. MacDonald managed to conceal the knife and ice pick beneath a large

bush, while the piece of lumber landed on the ground nearby. To support his story of

intruders, MacDonald left the rear exterior doors unlocked and open. He consulted the

March 1970 issue of Esquire magazine for details of Manson-style murders, inadvertently

staining it with blood, probably a mix of Colette’s and Kimberly’s. MacDonald also left

latent fingerprints on the inside pages of the Esquire magazine, or he had already made

those prints while previously perusing that issue. So it would appear that a struggle had

occurred in the living room, MacDonald turned the coffee table on its side, pinning the

Esquire magazine beneath the boxed children’s game and toppling the flower pot and his

eyeglasses onto the living room floor. Because no skirmish had actually occurred there,

however, the living room did not contain blood, splinters, and blue pajama top threads

and yarns like that recovered from the bedrooms.

      Although the government acknowledged it was possible that Colette inflicted

MacDonald’s only serious wound — the laceration in his right chest resulting in the

pneumothorax — the government deemed it much more likely that MacDonald, utilizing

his knowledge of surgery, inflicted that wound himself with a scalpel or other sharp

instrument. The government theorized that MacDonald did so in the family’s shared

                                           47
bathroom, while looking at his reflection in the mirror hanging above the sink. In that

regard, the government emphasized the drops of blood and bloodstains, all MacDonald’s

type B, in and around the bathroom sink and on the stepladder and wicker stool. The

government also pointed to the three small stripes of MacDonald’s type B blood at the

juncture of the sliding doors to the linen closet housing numerous medical supplies,

including disposable scalpel blades.

      The government noted that MacDonald may have deposited Colette’s and

Kimberly’s hairs, some bloodstained, in the bathroom sink as he washed up after the

murders. The government further surmised that MacDonald retrieved a latex surgeon’s

glove from the cabinet beneath the kitchen sink, dripping his blood on the adjacent floor

and perhaps smearing more blood on the sink’s underside. MacDonald donned the latex

glove and, to copy the Manson murders, wrote “PIG” in Colette’s blood on the master

bed’s headboard. As evidenced by the blue pajama top thread on the floor between the

wall and the headboard, MacDonald was in that vicinity. He also left pieces of the latex

glove throughout the master bedroom, including the bedding that he piled on the floor

after using it to transport Colette from Kristen’s bedroom.       Further substantiating

MacDonald’s violence against Colette were the blue pajama top threads and yarn

contained within that bedding; Colette’s bloodstained hair twisted around one of those

threads; the two hairs that had been forcibly removed from Colette and deposited onto the

master bed; and the debris in Colette’s hands, including another of her own hairs and a

fiber from Kristen’s throw rug.



                                           48
       The government’s account had MacDonald summoning help by telephone and

then purposely engendering chaos with the first responders to contaminate the evidence

and foster confusion. Having realized that he failed to clean and dispose of the Geneva

Forge knife, and not knowing what evidence it might contain (as it turned out, just a

small amount of blood on the tip of its bent blade), MacDonald lied and said he pulled

that knife from Colette’s chest. MacDonald also fabricated the baseball-bat-like club

before learning that the investigators had found the bloodstained piece of lumber. When

confronted with a photograph of that distinctive piece of lumber, MacDonald falsely

denied that he recognized it. He also falsely denied that he recognized the Geneva Forge

knife, the Old Hickory knife, and the ice pick, and that his family had kept an ice pick in

the apartment’s kitchen.

       Pointing out that MacDonald read crime novels and had been trained by the Army

to withstand interrogation, the government argued that he did his best in the frenzied

circumstances to stage the crime scene and concoct a story as close to the real events as

possible. Thus, for example, the government deemed it plausible that Kimberly yelled,

“Daddy,” over and over. Rather than crying “Jeff, Jeff, why are they doing this to me,”

however, Colette very well may have cried, “Jeff, Jeff, why are you doing this to me?”

The government underscored that ample evidence belied MacDonald’s account, and that,

in any event, his story made little sense. Why, the government posed, would a band of

intruders massacre a pregnant woman and two little girls but leave the strongest person in

the apartment — the one most able to fight back — alive, relatively unscathed, and able

to serve as an eyewitness? The government also stressed that MacDonald had repeatedly

                                            49
changed details of his account, seemingly to match the emerging evidence up to and even

during the trial.

                                             3.

       At trial, MacDonald took the stand in his own defense and continued to insist that

he and his family were attacked by intruders. MacDonald qualified, however, that he had

never been — and never claimed to be — certain of every detail of what happened in the

family’s apartment in the early morning hours of February 17, 1970. Many of his

thoughts and memories, MacDonald explained, were hazy, vague, and confusing.

       Echoing his prior statements, MacDonald testified that he went to sleep on the

living room couch sometime after 2:00 a.m. and was subsequently awakened by the

sound of Colette’s screams. He then heard Colette crying, “Jeff, Jeff, help me. Why are

they doing this to me?”        He also heard Kimberly repeatedly yelling, “Daddy.”

MacDonald saw the three short-haired male intruders of moderate height standing at the

end of the couch — the black man wearing the Army field jacket with Sergeant stripes,

plus the two white men, one slightly shorter than the other. MacDonald also saw the

blond-haired female intruder in the floppy hat, above-the-knee skirt or shorts, and boots.

He perceived that the light flickering in the woman’s face came from a candle, and he

heard the words, in a monotone female voice, “Acid is groovy; kill the pigs.”

MacDonald recalled at trial that he may have also heard the words, “Acid and rain,” and

he denied that he had ever referred to the intruders as “hippies.”

       MacDonald testified that he struggled with the three men for approximately thirty

to forty-five seconds in the area of the couch and the nearby step up to the hallway.

                                             50
MacDonald said that the melee began when he attempted to sit up on the couch but was

knocked back by the baseball-bat-like club wielded by the black man. The club struck

MacDonald in the left arm and head in a single blow, causing him to “see stars.”

MacDonald expressed particular confusion about the order and details of subsequent

events in the brief encounter. Those events included getting up from the couch and

attempting to fend off further blows from the club by holding onto an arm of the black

man and then onto the club itself. As the black man tried to jerk away from MacDonald,

MacDonald was pulled forward toward the end of the couch. Eventually, MacDonald

lost his hold on the club and was struck by it several more times. Upon losing hold of the

club, MacDonald contemplated that he might be killed. Meanwhile, the two white men

repeatedly punched and stabbed MacDonald. At the time, MacDonald did not perceive

the punches to be very effectual and only once sensed he might have been stabbed, in the

right chest. While grabbing the hand of one of the white men, MacDonald had seen in

that hand a glint of metal that he thought was the blade of a knife.

       In familiar testimony, MacDonald said that his blue pajama top was pulled over

his head or ripped from around his back during the battle, and that the pajama top became

wrapped around his wrists and hands. Although he had previously suggested that it was

more likely the pajama top was torn off, MacDonald testified that he did not recall a

ripping sound and simply did not know whether the top was pulled over his head or

ripped from around his back. Once the pajama top became wrapped around his wrists

and hands, MacDonald used it to fend off blows aimed at his chest, while simultaneously

and unsuccessfully attempting to free his hands from it. He fought back against the three

                                             51
men until being knocked unconscious. MacDonald only later learned that he had been

assaulted with an ice pick, in addition to the knife and the baseball-bat-like club.

       MacDonald testified that, after hearing Colette’s and Kimberly’s voices during his

encounter with the intruders, he awoke to silence. MacDonald’s body was stretched

facedown over the step between the living room and hallway, with his head and chest on

the hallway floor and his feet on the living room floor. He did not know how much time

had passed since the attack.

       According to MacDonald, he got up and went to the master bedroom, where he

found Colette on the floor, leaning against the green vinyl-upholstered chair. MacDonald

recalled taking his blue pajama top off his wrists and hands, pulling a knife out of

Colette’s chest, throwing the knife away from Colette, giving her mouth-to-mouth

resuscitation but feeling the air coming out of stab wounds in her chest, and checking for

Colette’s pulse but finding no signs of life in her. Contrary to his prior statements — first

that he may have moved Colette an inch or two, and then that he definitely moved her

about six inches — MacDonald now specified that he moved Colette six to twelve inches

in the course of trying to revive her. MacDonald said he next went to his Kimberly’s and

Kristen’s bedrooms. Then or later, he fruitlessly gave his daughters mouth-to-mouth

resuscitation and ascertained that they had no pulses. MacDonald estimated that he spent

seven to ten minutes on efforts to revive his wife and daughters, rotating between the

three bedrooms and going into each twice.

       Although he had previously stated that he covered Colette with his blue pajama

top and the white Hilton hotel bathmat on his initial visit to the master bedroom,

                                             52
MacDonald testified at trial on direct examination that he covered Colette with the blue

pajama top and the bathmat when he was in the master bedroom for the second time. He

said he could not specify the location of the blue pajama top between removing it from

his wrists and hands, and using it to cover Colette — suggesting, contrary to his prior

statements, that the blue pajama top may have been with him when he checked on

Kimberly and Kristen during his first rotation between the bedrooms.

         In the midst of going from bedroom to bedroom, as MacDonald recounted, he

went to the hallway bathroom to check his own head because it was hurting and he was

unable to think clearly. In the bathroom mirror, he saw a bruise and a smear of blood on

his forehead, as well as blood around his mouth that apparently came from his wife and

daughters when he attempted mouth-to-mouth resuscitation. He used the bathroom sink

to rinse additional blood — also thought to be his wife’s and daughters’ — from his

hands.    There or in the hallway or master bedroom, MacDonald noticed the small,

“bubbling” wound in his right chest.       His trial testimony raised the possibility that

MacDonald also looked for medical supplies to use on his family, evoking an explanation

for the stripes of his type B blood on the linen closet doors.

         As for his call for help, MacDonald said he believed that he initially used the

telephone in the master bedroom on his second stop there. MacDonald dialed “0,” gave

the operator his name and address, advised that there had been stabbings and people were

dying, and requested medics and military police.          When the operator asked if the

apartment was “on-Post or off-Post,” MacDonald became frustrated and dropped the

telephone’s handset without hanging up. To the best of his recollection, he then peered

                                             53
through the utility room and saw that the apartment’s wood rear exterior door was open.

MacDonald went to the doorway and looked outside, where he saw no persons and heard

no sounds. He next made his second trip to his daughters’ bedrooms, before going to the

kitchen and picking up the handset of the kitchen telephone. The operator was still on the

line and connected MacDonald with a Sergeant, who continued to collect information

from MacDonald while concurrently ordering assistance from Womack Army Hospital.

MacDonald recalled thinking of the Sergeant, “He’s getting help”; ending the phone

conversation; and resolving to return to Colette.

       MacDonald’s next memory was of being administered mouth-to-mouth

resuscitation by a military police officer while lying next to Colette on the master

bedroom floor, and then struggling with the officer while trying to get up. According to

MacDonald, the officer and his colleagues were pushing MacDonald into and pulling him

away from Colette while shouting things like, “Don’t touch that,” “Put that down,” and

“Tell Womack to get their ass here.” MacDonald sought help for his wife and daughters,

and upon being questioned by an officer, responded that he and his family were stabbed

and beaten by “a black male and two white males” accompanied by “a girl with a floppy

hat” and “a light on her face.” There was more yelling, running, and pushing by the first

responders as MacDonald was forced onto a gurney and taken out of the apartment to the

waiting ambulance.

                                             4.

       On cross-examination, MacDonald testified that he still did not recognize the

Geneva Forge knife, the Old Hickory knife, the ice pick, or even the bloodstained piece

                                            54
of lumber, with its distinguishing paint stains. He acknowledged, however, that it was

possible his family owned an ice pick, and that there were many pieces of lumber in the

apartment’s utility room, outdoor well, and storage shed. MacDonald also admitted that

the bloodstained piece of lumber, being “rough,” could not have been the baseball-bat-

like club that he said struck him.

       Addressing the forty-eight puncture holes in his blue pajama top, MacDonald

insisted the holes were made by the ice pick while he used the pajama top, wrapped

around his wrists and hands, to fend off the intruders’ blows. Asked why the puncture

holes were symmetrical and clean, with no tearing, MacDonald responded that the blows

were coming straight at him while he pushed out against them. He acknowledged that he

did not sustain any ice pick wounds on his hands or arms, and he could not explain how

that was so.

       Confronted with the discrepancy between his trial testimony that he covered

Colette with the blue pajama top during his second visit to the master bedroom, and his

earlier statements that the covering occurred when he was first in the master bedroom

after the attack, MacDonald claimed confusion and uncertainty.             Pressed by the

prosecutor, MacDonald said his best recollection was that he covered Colette with the

blue pajama top during his first visit to the master bedroom, after he pulled the knife from

her chest. MacDonald conceded that the knife — which had to be the Geneva Forge

knife — should have been bloody, and he could not explain why it had only a small

amount of blood on it, on the tip of its blade.



                                              55
      The prosecutor next cross-examined MacDonald about the changes in his account

of moving Colette as he attempted to revive her, pointing out that MacDonald first said

he may have moved Colette no more than an inch or two, then stated that he definitely

moved her about six inches, and now testified that he moved her as much as six to twelve

inches. MacDonald reasoned that, to get Colette away from the green vinyl-upholstered

chair, he would have had to move her six inches or more. He denied changing his

account to explain how threads and yarns from his blue pajama top ended up beneath

Colette’s body.

      Other particularly damning threads and yarns, as emphasized by the prosecutor,

were the two threads on the bloodstained piece of lumber discovered on the ground

outside the rear entrance to the apartment. The evidence being that the bloodstained

piece of lumber stayed outside the apartment after it was found, and MacDonald’s own

testimony being that he did not go outside in the aftermath of the intruders’ attack,

MacDonald could not explain how the threads got on the piece of lumber.

      During MacDonald’s cross-examination, the prosecutor drew attention to the

immensity of the number of blue pajama top threads and yarns — seventy-eight —

collected from the master bedroom. The prosecutor also specifically pointed to the thread

wrapped around the bloodstained hair, matching Colette, in the pile of bedding on the

master bedroom floor; the threads and yarns in Kimberly’s and Kristen’s beds, including

beneath Kimberly’s covers; and the fact that many of the threads and yarns were

interspersed with splinters. MacDonald suggested that threads and yarns may have

originated not from his blue pajama top, but from the matching blue pajama pants that

                                           56
were discarded while he was being treated at Womack Army Hospital. He also reiterated

his theory, introduced in prior statements, that threads and fibers may have been stuck to

his bare arms and fallen off as he rotated between the three bedrooms to render aid to his

wife and daughters.

       Confronted with the lack of blue pajama top threads and yarns in the living room,

where the intruders allegedly attacked him, MacDonald responded that the blue pajama

top must have been pulled over his head during the encounter, rather than torn off.

MacDonald reiterated that he did not know how or when the blue pajama top was torn,

and he testified that he did not recall hearing any ripping sounds, either during the battle

in the living room or in the master bedroom thereafter.

       Assuming that Colette’s blood got on the blue pajama top only after it was laid

over her chest, MacDonald could not explain why some of the bloodstains appeared to

have been made before the pajama top was torn. He also could not explain how Colette’s

blood got on the blue pajama top’s pocket. As for how the pocket ended up on the throw

rug near Colette’s feet, MacDonald posited that the pocket may have fallen from the blue

pajama top when he removed it from his wrists and hands, and that the detached pocket

may have been moved about in the chaos caused by the first responders. Asked how

Kimberly’s blood got on the blue pajama top, MacDonald responded that he could offer

only conjecture.

       MacDonald testified that he could not recall whether, in the aftermath of the

attack, he handled the master bed’s top sheet and bedspread that were in the pile of

bedding on the master bedroom floor. If impressions on the top sheet were made by the

                                            57
bloody left and right cuffs of his blue pajama top, MacDonald had no theory as to how

that occurred.    He also expressed disbelief that he made any bloody shoulder and

handprints on the top sheet.

       As he had in his earliest interviews with CID and FBI agents, MacDonald

conceded that the bloody left footprint found on Kristen’s bedroom floor and matched to

him was probably, in fact, his. MacDonald further testified that — assuming the CID

blood analysts correctly determined the footprint was made in Colette’s type A blood, but

all other blood on Kristen’s bedroom floor was her type O — he could not explain how

he made solely that left footprint, with the corresponding right footprint, while exiting

Kristen’s bedroom.

       When the cross-examination turned to MacDonald’s own wounds and the

bloodstains of his type B, MacDonald denounced the government’s theory that he used a

scalpel or some other instrument to inflict the laceration in his right chest. As for why

there was no type B blood in the living room where he said he was clubbed and stabbed,

MacDonald offered only that his wounds did not bleed much. Asked why, then, there

was type B blood in the family’s shared bathroom, MacDonald explained that he

examined his bubbling chest wound there and must have gotten blood on his hands that

he then transferred to the bathroom sink. He had no explanation for why his blue pajama

top had just a single type B bloodstain, on its left sleeve.

                                              5.

       Rather than trying to account for all of the government’s evidence, MacDonald

and his defense lawyers disputed aspects of that evidence and the conclusions the

                                              58
prosecutors drew therefrom. In so doing, the defense highlighted problems with the

investigation, beginning with the military police officers and medics — numbering more

than a dozen — who had tramped through the apartment. Although the defense did not

fault the first responders for seeking to aid any survivors and find any perpetrators who

might have been present, the defense maintained that the officers and medics inevitably

altered the crime scene. Moreover, the defense pilloried some for remaining in the

apartment beyond their usefulness and for making avoidable and potentially critical

mistakes.

       The evidence reflected, for example, that a military police officer picked up the

master bedroom telephone. In the living room, a medic moved the flower pot that had

been overturned on the floor, setting it upright. After being told not to touch anything

else, the medic sat on the couch. Crime scene photographs showed the same throw

pillows in varying positions on the couch; a pile of clothes alternately on the couch and

on the floor at the living room end of the hallway; and MacDonald’s eyeglasses in two

different positions on the living room floor.

       Additionally, the military police officer who gave MacDonald mouth-to-mouth

resuscitation could not rule out that he shifted Colette’s nearby body while doing so. To

transport MacDonald to Womack Army Hospital, medics rolled a gurney in and out of

the apartment, through the living room, hallway, and master bedroom. The defense

argued that important evidence — particularly the blue pajama top threads and yarns that

were found in the master bedroom but not the living room — may have been moved



                                                59
around by the gurney, as well as by the undoubtedly wet boots of the first responders and

the presumably rain-soaked clothes of the medic who sat on the living room couch.

       According to the defense, the conduct and work of the CID investigatory team was

similarly flawed. The defense pointed to, inter alia, the testimony of a Womack Army

Hospital physician who had come to the apartment to confirm the three deaths under the

direction of the CID. The physician examined the bodies of Colette, Kimberly, and

Kristen after waiting for photographs to be taken, but perhaps before the outlines of the

bodies were drawn. He testified that he moved the bodies and clothing during the

examination, without making any effort to return them to the same positions.

Additionally, the physician said he may have gotten blood on his hands that he

transferred from one body to the next. He also recalled removing the blue pajama top

from Colette’s chest and then randomly laying it back down. In conflict with that

testimony, a CID agent recalled that the physician substantially moved only the body of

Kristen, and that Kristen’s body was then rephotographed to document the changes in

position.

       In any event, it was undisputed that Colette was taken out of the apartment on a

gurney that, like MacDonald’s gurney, was rolled through the living room, hallway, and

master bedroom. Kimberly and Kristen were placed together on a canvas stretcher,

where their bodies may have touched. The defense contended that, as the three bodies

were examined and then removed from the apartment, there were yet more opportunities

for evidence like the blue pajama top threads and yarns to be redistributed and otherwise

disturbed.

                                           60
       As further highlighted by the defense, the CID agents failed to order the extraction

of hair samples from the bodies of Colette, Kimberly, and Kristen while their autopsies

were being conducted at Womack Army Hospital, and the FBI later had to have the

bodies exhumed to obtain known hairs for comparison to the hairs in evidence.

Furthermore, the CID agents ordered the fingerprinting of Kimberly, but not Kristen. By

the time that was attempted, Kimberly’s body had been embalmed, and thus her prints

could not be obtained. Although the CID got prints from Colette, those prints were

incomplete. Meanwhile, the CID agents overlooked MacDonald’s blue pajama pants as

potential evidence, leaving them to be discarded at the hospital.

       Back at the apartment, a CID agent removed the bloodstained piece of lumber

from the ground without first having it photographed. Rather than being preserved and

examined, the contents of trashcans outside the apartment were picked up during the

regular route of the local garbage truck. Inside the apartment, members of the CID team

used at least one of the apartment’s two toilets, located in the MacDonalds’ shared

bathroom and the utility room, before the sewer lines were dug up and searched for

evidence that might have been flushed.

       The investigators acknowledged that they did not collect every fiber spotted in the

apartment, including every blue pajama top thread and yarn. They picked up threads,

yarns, and other fibers by hand, rather than with a vacuum specially designed for

evidence collection. Although they said there were no blue pajama top threads or yarns

in the living room, only one investigator among them used a magnifying glass, and not

just the naked eye, to examine the living room and its area rug.

                                            61
       An expert for the defense on fingerprint processing testified that the crime scene

was grossly underprocessed for latent prints. Focusing on areas of obvious potential

significance, the expert specified that a light switch on the utility room wall was not

processed; a swinging interior door between the utility room and the master bedroom was

processed on just one part of one side; the glass tops of dressers in the master bedroom

were not fully processed, even though one held the jewelry box with the unsourced print;

the master bed’s headboard and the adjacent master bedroom wall were processed in only

small areas, near where “PIG” was written in blood; a second master bedroom wall was

not processed at all; other walls in the three bedrooms and hallway received limited

processing that missed doorjambs and light switches; there was little or no processing of

the living room wall behind the couch or of the kitchen walls; and only some of the

kitchen cabinet doors and drawers were processed.

       Rather than “lifting” latent prints, the CID team dusted the prints and took

photographs for the purpose of identification, but botched several of those photographs

and had to retake them. Two apparent prints on the apartment’s wood rear exterior door

went unexamined because every photograph of the prints failed to include identifying

ridge details. The blood on the March 1970 issue of Esquire magazine may have been a

bloody fingerprint, but investigators handling the magazine apparently smeared the blood

before realizing it was there. Two or three of the latent prints in the apartment that were

identified, including at least one print within the Esquire magazine, were matched to CID

and FBI agents.



                                            62
      Although MacDonald conceded that the bloody left footprint on the wood floor of

Kristen’s bedroom was probably his, the defense questioned the CID laboratory analyst’s

identification of that footprint.   As the defense established, the laboratory analyst

matched the footprint to MacDonald at the apartment.          His conclusion was never

reviewed and confirmed, however, because the footprint was ineffectively photographed,

and that section of the wood floor broke apart after being removed from the apartment

and transported to the CID laboratory.

      The defense also pointed to evidence of the underprocessing of blood and hairs,

and it launched multiple attacks on the training, methods, and conclusions of the CID and

FBI laboratory analysts and other prosecution experts. Among the defense experts was a

chemist who opined, contrary to the chemist for the government, that it was highly

unlikely that the pieces of latex recovered from the master bedroom were made by the

same manufacturer, at least within the same time period, that produced the Perry-brand

latex surgeon’s gloves beneath the kitchen sink. The defense chemist premised his

conclusion on both the government chemist’s test results and independent testing with

more state-of-the-art equipment.

      Two defense experts on fabric damage, impressions, and stains shared opinions

that were variously favorable — and not so favorable — to MacDonald. The defense

experts disagreed with the FBI analyst’s conclusion that bloodstains on the master bed’s

top sheet were impressions consistent with the right cuff of Colette’s pink pajama top, as

well as a bare left shoulder and left and right hands.      Rather, the defense experts

concluded that those particular stains on the top sheet were probably made by a transfer

                                           63
of wet blood from one part of the sheet to another while the sheet was being folded or

bunched up. The defense sought to blame the investigators, arguing that blood on the

sheet may have still been wet when a CID agent picked up the sheet and placed it in an

evidence bag. Even so, one of the defense experts agreed with the FBI analyst that other

bloodstains on the master bed’s top sheet could have been impressions made by the left

cuff of Colette’s pink pajama top and the right cuff of MacDonald’s blue pajama top.

That defense expert specified that the impressions consistent with the blue pajama top’s

right cuff were apparently made by gentle contact, such as by brushing away the sheet.

       The same defense expert examined the blue pajama top and noted the absence of a

very fine, nearly invisible spray of blood droplets on the pajama top.       The expert

explained that he detected such a spray of blood droplets on the master bed’s bottom

sheet, which he attributed to the clubbing attack with the bloodstained piece of lumber.

According to the expert, if the assailant was wearing the blue pajama top as the

government alleged, there likely would have been similar blood droplets on the pajama

top. The expert elaborated that, although the blue pajama top was heavily bloodstained,

there were yet areas of the pajama top where he would have expected to find a fine spray

of blood droplets.

       The defense expert also tested fabric similar to that of the blue pajama top and

found that the fabric released no more than a few yarns when torn, possibly explaining

why there were no blue pajama top yarns found in the living room even though the

pajama top may have been torn there. The expert further observed, however, that when

the fabric was torn at seams, it released scores of sewing threads. More violent tearing

                                           64
released more sewing threads, and additional threads were in position to drop during

subsequent forceful movement. The expert acknowledged that MacDonald’s blue pajama

top was torn both through its front panel and at several of its seams, and he did not

dispute that the threads and yarns collected from the bedrooms were consistent with the

blue pajama top.

       Turning to the puncture holes in the blue pajama top, the defense expert lambasted

the FBI laboratory analysts’ purported reconstruction of how twenty-one thrusts of the ice

pick into Colette’s chest could have made the blue pajama top’s forty-eight puncture

holes. That reconstruction was wholly invalid, the defense expert testified, because it did

not take into account the FBI laboratory analysts’ own prior determinations of whether

each puncture hole was an “entrance” or an “exit” hole. The defense expert specified

numerous instances in which the reconstruction had a previously designated entrance

hole as an exit hole, and vice versa. The defense also established that the FBI laboratory

analysts did not attempt to line up the puncture holes in the blue pajama top with those in

Colette’s pink pajama top, or consider the different sizes of the blue pajama top’s

puncture holes, which evidently varied by how far the ice pick was inserted. Further

undermining the reconstruction, the FBI laboratory analysts presumed that the puncture

holes were made in the blue pajama top when the pajama top was positioned on Colette’s

chest as photographed at the scene. The evidence reflected, however, that before being

photographed, the blue pajama top was likely repositioned by first responders, CID team

members, and others.



                                            65
         In addition to criticizing the FBI’s reconstruction of the blue pajama top’s

puncture holes, the defense expert disputed the conclusion of one of the FBI laboratory

analysts that those holes were necessarily made while the pajama top was stationary. The

defense expert relied on his testing of fabric similar to the blue pajama top’s and his

observations that an ice pick made circular holes, without any tearing, while the fabric

was in motion. The defense expert acknowledged, however, that the fabric he used was

brand new, and that the same fabric would probably be more susceptible to tearing if, like

the blue pajama top, it had been laundered many times. Furthermore, the defense expert

conceded that the circumstances of his testing — he stabbed an ice pick through a single

layer of fabric, stretched fairly tightly over a plastic-wrapped ham, that was moving only

laterally — did not and were not meant to simulate MacDonald’s account of using the

pajama top as a shield from a series of blows while it was wrapped around his wrists and

hands.

         Indeed, the defense expert testified that MacDonald’s account was too vague and

uncertain for any scientifically reliable testing. That testimony prompted the government

to stage a demonstration in the courtroom, in which one prosecutor used a facsimile of

the blue pajama top, wrapped around his wrists and hands, to fend off stabs from a

second prosecutor wielding an ice pick. The first prosecutor was wounded by the ice

pick during the demonstration, and the ice pick left a few circular holes, but mostly tears,

in the pajama top.       The defense expert called the government’s demonstration

“interesting” but not scientifically reliable due to the uncertainty surrounding

MacDonald’s account.

                                            66
       The government otherwise responded to the defense’s attacks on the investigation

by downplaying the significance of mistakes that were made, asserting that the

prosecution did not rise or fall on whether the crime scene was wholly unadulterated or

whether the investigation was perfect. In the government’s view, blunders like picking

up the flower pot and sitting on the couch were not significant enough to engender

reasonable doubt about MacDonald’s guilt. Moreover, although the first responders or

CID team may have unknowingly caused some movement of evidence such as the blue

pajama top threads and yarns, it was implausible that there had been threads and yarns in

the living room that were all relocated to the bedrooms, or that other potentially

exculpatory evidence was rendered inculpatory by random redistribution. It also was

improbable that the investigators shifted any significant amount of evidence around the

apartment without leaving a bit of that evidence in the living room.

       Ultimately, the government urged the jury to convict MacDonald based upon the

evidence found and analyzed, and not to speculate about what the investigators might

have overlooked. In so doing, the government lauded much of the work of the CID and

FBI, including analyses and conclusions that the defense experts either corroborated or

did not dispute, e.g., that the threads and yarns matched MacDonald’s blue pajama top

and likely dropped therefrom in some quantity once the pajama top was torn, and that

impressions on the master bed’s top sheet were consistent with MacDonald’s and

Colette’s bloodstained pajama top cuffs.      The government also emphasized pivotal

evidence that the defense failed to explain or refute, including the two blue pajama top

threads on the bloodstained piece of lumber and the bloody footprint in Kristen’s

                                            67
bedroom. As the government argued, the jury could question or even reject ninety

percent of the prosecution’s case but believe a crucial ten percent and convict

MacDonald.

       By contrast, the defense suggested that the first responders’ and investigators’

“bungling” and “ineptitude” tainted the entire prosecution and necessitated MacDonald’s

acquittal. The defense argued that the investigation need not have been perfect, but that it

was fatally flawed for lack of basic competency.

                                             6.

       Another significant aspect of the defense was that MacDonald was wholly

incapable of the monstrous acts that had been perpetrated against his wife and daughters.

He depicted himself as a loving husband and father, reading from the Valentine’s Day

cards that had been displayed on the family’s dining room buffet — cards that he,

Colette, Kimberly, and Kristen had exchanged less than three days before the murders.

MacDonald also elaborated on happy times depicted in a series of family photographs.

To corroborate MacDonald, the defense called more than a dozen character witnesses.

But MacDonald also admitted on direct and cross-examination that he had committed

adultery multiple times during his marriage to Colette and resumed sexual activities with

other women within months of the slayings. Unsurprisingly, the defense insisted that

evidence of MacDonald’s philandering did not prove he was his family’s murderer, and

the government largely concurred. The government argued, however, that the physical

evidence established that MacDonald slaughtered his wife and daughters, nullifying the

character evidence that he was not capable of such brutality and horror.

                                            68
                                             7.

       For its part, the defense stressed the physical evidence tending to support

MacDonald’s account of murderous intruders, along with other potentially exculpatory

evidence. Of great importance to the defense was its chemist’s opinion that the pieces of

latex recovered from the master bedroom were inconsistent with the Perry-brand latex

surgeon’s gloves beneath the kitchen sink. The defense characterized that conclusion as

devastating to the prosecution, both because it undermined the government’s theory that

MacDonald retrieved a latex glove from the kitchen to write “PIG” on the master bed’s

headboard, and because it evidenced intruders who brought their own latex gloves to the

apartment.

       As further physical evidence of intruders, the defense pointed to the three samples

of dried wax collected from the coffee table in the living room, the chair in Kimberly’s

bedroom, and Kimberly’s bedspread. According to the defense, because the three wax

samples did not match any of the MacDonalds’ fourteen candles, the wax samples were

evidence of the blond-haired female intruder who appeared to be carrying a lighted

candle while her three male cohorts attacked MacDonald in the living room. The defense

also invoked physical evidence that was not matched to any person, alluding to items

such as the hair in Colette’s left hand, the apparently tar-coated hair in the bathroom sink,

and the many identifiable but unsourced latent palm and fingerprints, including those on

the jewelry box in the master bedroom, the master bedroom window blinds, and the

drinking glass in the living room.



                                             69
       As proof that MacDonald was assaulted by intruders in the living room, the

defense cited the unsourced pink fiber caught in the hinge of his eyeglasses, which also

bore the spot of blood that gave a weak indication of being Kristen’s type O.

Additionally, the defense specified the clustered blue fibers or threads found on the floor

at the living room end of the hallway near the bloodstain of Kristen’s type O or

MacDonald’s type B. The defense suggested that those blue fibers or threads may have

come from MacDonald’s blue pajama top, thereby corroborating his account of being

knocked unconscious in the living room and waking sometime later lying facedown over

the step between the living room and hallway.

       Through the testimony of other physicians, including friends of MacDonald who

visited him at Womack Army Hospital, the defense sought to show that the

pneumothorax was a life-threatening injury that MacDonald would have been reckless

and foolish to inflict on himself, and that MacDonald sustained additional wounds not

documented at the hospital or otherwise noted by the medical providers there. The

additional wounds included black eyes, bruises on MacDonald’s forehead and left arm,

and, most significantly, puncture wounds in his chest consistent with the ice pick. The

defense acknowledged that the purported ice-pick-like wounds did not leave any scars,

but argued that the lack of scars was not anomalous or suspicious.

       With respect to the weapons found at the crime scene, the defense emphasized that

no witness specifically identified the Geneva Forge knife, the Old Hickory knife, or the

ice pick as belonging to the MacDonalds. And, while conceding that the bloodstained

piece of lumber evidently came from the apartment, the defense insisted that the piece of

                                            70
lumber would have been easily acquired by the intruders — from the storage shed or

open well in the backyard or from the utility room — as they invaded through the

apartment’s rear entrance.

      To cast doubt on the CID team’s observations that there were no obvious signs of

robbery in the apartment, the defense established that the government had compensated

MacDonald for several pieces of missing jewelry. In the summer of 1970, MacDonald

had requested that jewelry from the apartment, which then remained a sealed crime

scene. MacDonald’s request was not immediately granted, however, and the jewelry did

not appear in an inventory of the apartment’s contents later that autumn. The jewelry

included two of Colette’s rings that MacDonald said the intruders may have stolen from

the jewelry box in the master bedroom with the unsourced latent print.

      That is not to say that the defense proffered robbery as the intruders’ principal

motive. Rather, the defense theorized that the intruders set out to commit violence,

targeting MacDonald and his family either by chance or because of a grudge against

MacDonald arising from his professional dealings with drug abusers.          MacDonald

explained that he had treated and counselled drug-abusing soldiers, and that, because he

sometimes had to report those soldiers to their commanding officers, he came to be

known by some as a “fink.” MacDonald also said he had treated drug abusers at the local

Cape Fear Valley Hospital and other civilian facilities, where patients included soldiers

seeking to hide their drug abuse from the Army.

      According to the defense, two witnesses may have seen the intruders outside the

MacDonalds’ apartment, before and after the murders. One of those witnesses resided in

                                           71
the same Fort Bragg apartment complex and had a view of the MacDonalds’ apartment

from his. Around midnight, while that neighbor was constructing a model boat in his

apartment’s workshop with the window open, he heard voices and looked outside to

investigate. The neighbor testified to spotting a white female, flanked by two white

males, walking on a paved walkway in the direction of the MacDonalds’ apartment.

Each member of the trio was wearing a white sheet or choir robe and carrying a lighted

candle, and the female had blond or light brown hair and was not wearing a hat or other

head covering. The sheets or robes were long and concealed the trio’s footwear. The trio

was about forty yards from the MacDonalds’ apartment when last sighted by the

neighbor, who subsequently went to bed at about 12:30 a.m.

       Shortly before 4:00 a.m., the second witness — a military police officer

responding to MacDonald’s call for help — observed a female standing near a street

corner approximately one-half mile from the MacDonalds’ apartment.          The officer

testified that the female was wearing what appeared to be a dark-colored, wide-brimmed

rain hat and matching rain coat. The officer perceived that the female was a woman in

her twenties, and he did not notice her hair color or whether she was wearing boots. But

for the emergency call, the officer might have stopped and checked out the woman, in

light of the inclement weather, early hour, and fact that a nearby commissary and gas

station were closed at the time. Upon later hearing the descriptions MacDonald gave of

the intruders, the officer advised his supervisor of the woman he had seen and suggested

that a patrol be sent to find her.



                                          72
      The defense argued that the trio spotted by the neighbor could have been

surveilling the apartment complex before returning a couple hours later to attack the

MacDonalds, and that the woman observed by the military police officer may have been

en route from the crime scene. The government countered, however, that neither the trio

spotted by the neighbor nor the woman observed by the officer matched MacDonald’s

descriptions of the alleged intruders. That is, unlike the neighbor’s trio, MacDonald’s

intruders were not wearing white sheets or choir robes, and only one of them, the blond-

haired female intruder, was possibly carrying a candle. Moreover, whereas the neighbor

saw a hatless female and the officer observed a woman in a dark-colored hat,

MacDonald’s female intruder was wearing a white hat. Further disputing the notion that

witnesses saw MacDonald’s intruders, the government emphasized that it took the

neighbor many years to share his account with the authorities or anyone else, even though

he learned of the murders of Colette, Kimberly, and Kristen within hours.

      The government also sought to diminish the defense’s other evidence of intruders.

According to the government, whether the pieces of latex in the master bedroom matched

the Perry-brand gloves beneath the kitchen sink was not determinative of MacDonald’s

guilt or innocence, because MacDonald could have used other latex gloves.             As

demonstrated by the latex gloves beside the kitchen sink and in the locked storage shed,

latex gloves were commonplace in the MacDonald household. The government asserted

that evidence like the wax samples and unsourced hairs and latent palm and fingerprints

— far from being proof of intruders — was easily explained by the apartment’s “living

day-to-day mess” and the many innocent visitors who, as MacDonald himself told

                                           73
investigators, must have left traces of themselves behind.            Additionally, as the

government had to acknowledge, Colette, Kimberly, and Kristen could not be eliminated

as the source of some of the unmatched prints, because the CID team failed to obtain all

their prints for comparison. Addressing the unsourced pink fiber caught in the hinge of

MacDonalds’ eyeglasses, the government suggested that the fiber could have come from

the hospital where MacDonald had just worked. The government also argued that the

pink fiber, as well as the spot of blood on the eyeglasses’ lens, could not support

MacDonald’s account of intruders, because MacDonald had disclaimed that he was

wearing his eyeglasses during the alleged attack, and because the spot of blood was not

his type B.

       The government insisted that — rather than being bolstered by evidence such as

the pink fiber in his eyeglasses and the clustered blue fibers or threads on the floor at the

living room end of the hallway — MacDonald’s account was discredited by the lack of

his blood and blue pajama top threads and yarns in the living room itself, along with the

relative mildness of his injuries. The government underscored that, notwithstanding the

defense’s protestations that MacDonald’s pneumothorax was life-threatening and that he

was wounded more than documented at the hospital, MacDonald’s injuries paled in

comparison to his family’s, and his claim of undocumented ice-pick-like wounds was

especially unworthy of belief. The government also questioned whether any jewelry

really was missing from the apartment and, in any event, noted the lack of proof that it

was stolen by the alleged intruders. Other details of MacDonald’s account that the

government deemed dubious included the following: the intruders brought their own

                                             74
knives, ice pick, and baseball-bat-like club for the attack, but providentially ran across the

MacDonalds’ piece of lumber, added it to their arsenal, and opted to wield it against

Colette and Kimberly; although possibly motivated by revenge against MacDonald, the

intruders made not a single mark on MacDonald’s bare back after they knocked him

unconscious, lying facedown; the intruders perhaps stole a few pieces of jewelry from the

apartment, but nothing else; and the intruders left the bloodstained piece of lumber

behind, together with their knives and ice pick, but absconded with their baseball-bat-like

club. In the circumstances, the government contended, MacDonald’s account simply was

not plausible or sufficiently corroborated by physical and other evidence.

                                              8.

       As its ultimate shot at the prosecution, however, the defense identified the alleged

female intruder as Helena Stoeckley and called her as a witness at the trial. At the time

of the murders, Stoeckley was a nineteen-year-old resident of Fayetteville, the city

bordering Fort Bragg. Authorities and acquaintances quickly viewed her as a suspect,

because she generally fit MacDonald’s description of the blond-haired woman who

accompanied the three male intruders. Specifically, although Stoeckley had dark-colored

hair, she occasionally wore a blond wig. She regularly wore a floppy hat and was a

heavy user of drugs including heroin, opium, barbiturates, cocaine, marijuana, hashish,

and the hallucinogens PCP, LSD, and mescaline.

                                              a.

       Helena Stoeckley testified at trial that, between 1969 and 1971, she resided in a

Fayetteville neighborhood popular with “hippies” and other drug users, both civilians and

                                             75
soldiers. She and other residents of the neighborhood commonly used words like “pig”

to refer to the police and “acid” for drugs like LSD. Stoeckley and her ilk called things

they liked “cool” and “far out,” while younger kids used the term “groovy.” Stoeckley

was familiar with the Manson murders from news reports, though she did not know of

Charles Manson in connection with any personal cult-like activities. Nevertheless, she

had an interest in witchcraft, which she studied by reading books and talking to a witch

friend. Through her studies, Stoeckley learned that high priestesses of witchcraft often

wore vestments or other robes of white (for benevolent purposes) or black (for doing

harm), and that their ceremonies involved lighted candles. Stoeckley said she never wore

such robes or became a witch herself, and she lost interest in witchcraft around 1972 or

1973 after “a bad experience with some people in Florida.”

      Focusing on the hours leading up to the MacDonald murders of February 17,

1970, Stoeckley recalled talking to her friend Greg Mitchell, a soldier, in the driveway of

her Fayetteville house sometime before midnight. Earlier, Stoeckley had injected heroin

and opium six or seven times, and she had smoked marijuana throughout the day.

Mitchell gave Stoeckley a mescaline tablet that she swallowed while standing in her

driveway. Immediately thereafter, right at midnight, Mitchell left to go on duty at Fort

Bragg, and Stoeckley went inside her house. Stoeckley testified that she was certain she

was not wearing her blond wig at the time, because Mitchell did not like the wig. As she

always did, Stoeckley wore black or purple clothing and one of her three pairs of boots

— black ankle-high boots, brown knee-high boots, or white thigh-high boots. Because

she frequently did, she also may have been wearing her floppy hat. Stoeckley did not

                                            76
testify to the color of her hat, but she was shown a depiction of a white floppy hat that she

said was similar to hers.

       According to Stoeckley, her next memory of February 17, 1970, was of returning

to her house around 4:30 or 5:00 a.m. as a passenger in a blue car with two or three

soldiers. The group came from a Dunkin’ Donuts restaurant. Stoeckley said she could

not identify the soldiers or account for her whereabouts between midnight and 4:30 a.m.

It was not unusual for Stoeckley to be out and about at those hours or to have memory

loss related to her drug abuse.

       Shortly after arriving home that morning, Stoeckley learned of the MacDonald

murders by way of a radio bulletin. The next day, February 18, 1970, Stoeckley was

located by Fayetteville police and questioned about the murders. The police briefly held

her for further questioning by CID agents, but the CID agents did not show up, and

Stoeckley had to be released. Sometime later, the CID did interview Stoeckley. She also

was repeatedly questioned and harassed about her possible involvement in the murders by

local police, acquaintances, and others in the community, and she was identified as a

suspect in news reports.

       The attention on Stoeckley and her lack of an alibi made her question whether she

was involved in the murders and fear that she would be implicated. Stoeckley testified

that, consequently, she burned her blond wig within a couple days of the murders.

Around the same time, she disposed of her white thigh-high boots. She kept her floppy

hat, but wore it much less often, until it was stolen some months later.



                                             77
      Several days after the murders, prior to the funerals of Colette, Kimberly, and

Kristen, Stoeckley hung several funeral wreaths on a fence outside her house. Stoeckley

explained that she had retrieved the funeral wreaths from a nearby alley where they had

been discarded by a neighborhood florist. She said she and her friends regularly retrieved

discarded flowers from the alley, and she did not relate the funeral wreaths to the

MacDonalds. Stoeckley acknowledged, however, that she had discussed the possibility

of attending the MacDonald funerals with others. She also drove by the MacDonalds’

apartment about a year later to see if it would jolt a memory of being involved in the

murders.

      In the latter half of 1970, Stoeckley moved to Nashville, Tennessee, to escape

harassment from the local police and to avoid further embarrassment to her family in

Fayetteville. Stoeckley continued to be approached about the murders, however, and she

frequently relocated — bouncing between North Carolina, South Carolina, Florida, and

Ohio, and moving three times for the specific purpose of evading additional interviews

sought by the CID. During that period, she unsuccessfully attempted rehabilitation from

drug addiction, and she battled health problems, including the hepatitis that had afflicted

her since about 1969.

                                            b.

      On the request of the defense, the trial court had issued a material witness warrant

for Helena Stoeckley’s arrest, and the FBI had located her in South Carolina. After being

transported to the federal courthouse in Raleigh, Stoeckley had met for several hours with



                                            78
the defense lawyers and then the prosecutors during a break in the trial. Her trial

testimony occurred the next day.

       For Stoeckley’s direct examination, the defense lawyers were prepared with a

series of crime scene photographs that they had shown Stoeckley the previous day and

questioned her about. When presented during the trial with the first of the photographs

— of the MacDonalds’ living room, including the couch, overturned coffee table, and

step up to the hallway — Stoeckley testified that she did not recognize the room and had

no reason to believe she had been there before. That testimony prompted defense lawyer

Bernard Segal, in a bench conference, to request leave of the trial court to treat Stoeckley

as a hostile witness. Pleading surprise, Segal asserted that when Stoeckley was shown

the same photographs the previous day, she said she recalled standing over a body while

holding a candle and seeing a man’s body on the floor.           Segal also proffered, in

anticipation of changes in Stoeckley’s responses to other photographs, that Stoeckley

stated she remembered the following: riding a toy horse in Kristen’s bedroom; standing

outside the MacDonalds’ apartment looking at her hands and saying, “My God, the

blood; oh my God, the blood”; and standing near the street corner where the military

police officer en route to the MacDonalds’ apartment saw the woman in the dark-colored

floppy hat. According to Segal, when Stoeckley was shown photographs of the bodies of

the murder victims, she said Colette and Kristen looked familiar to her, though Colette’s

chin injuries made identification difficult.

       In response, prosecutor James Blackburn advised the trial court that Stoeckley had

made no such statements during the prosecutors’ meeting with her the previous day.

                                               79
Blackburn asserted that Stoeckley told the prosecutors she had been shown photographs

by the defense lawyers, and that Stoeckley then insisted to the prosecutors that she did

not recognize any of the scenes in those photographs, had never been in the MacDonalds’

apartment, was not involved in the murders, and believed MacDonald was guilty.

Blackburn also said that he relayed to defense lawyer Wade Smith what Stoeckley said in

her meeting with the prosecutors and was under the impression that she had made similar

statements to the defense lawyers.

       Smith then went on the record at the bench conference that Stoeckley generally

told the defense lawyers, “I don’t remember.” In several instances, however, Stoeckley

made statements that Smith characterized as giving an interesting insight into Stoeckley’s

mind. Before allowing the defense’s examination of Stoeckley to resume, presiding

Judge Franklin T. Dupree, Jr., asked Stoeckley whether she had told both sides the same

story, and Stoeckley answered that, as far as she knew, she had done so. Rather than

ruling on the defense’s request to treat Stoeckley as a hostile witness, Judge Dupree then

indicated that he would revisit the matter as appropriate.

       Back on the witness stand, Stoeckley testified that she did not recognize any of the

scenes in the various photographs.         She acknowledged, however, that — in a

conversation the previous day without the defense lawyers present — she had told a

prospective defense witness, her former Nashville neighbor and co-worker Jane Zillioux,

that Kristen looked like a little girl she knew, and that she may have seen Colette on Fort

Bragg sometime before the murders. Stoeckley also said she commented to Zillioux that

the toy horse in Kristen’s bedroom appeared in a photograph to be broken. Presented

                                             80
with photographs of the toy horse, which bounced on four springs attached to a stationary

frame, Stoeckley pointed to one showing a broken-looking “part going to the spring.” 6

Stoeckley denied telling anyone the previous day that she recalled riding the toy horse.

She similarly denied saying that she recalled standing outside the MacDonalds’

apartment and looking at her hands, and then standing near the street corner where the

military police officer may have spotted her. As for her previous day’s response to

photographs of the MacDonalds’ living room, Stoeckley testified that when she said she

recalled seeing a body and holding a candle in that room, it “was only like in a dream or

something like that.”

                                              c.

                                              (1)

       The defense sought to call several witnesses (sometimes called the “Stoeckley

witnesses”) to impeach Helena Stoeckley and relay self-incriminating statements that she

had made over the years since the MacDonald murders. To determine whether he would

permit the Stoeckley witnesses to testify, Judge Dupree had the parties question the

witnesses outside the presence of the jury.

       One of the Stoeckley witnesses was William Posey, who had been Stoeckley’s

neighbor in Fayetteville at the time of the murders. Posey had contacted MacDonald’s

lawyers during the 1970 Army proceedings and become a defense witness therein. In

       6
       The record reflects that, although a spring may appear broken or detached in one
of the crime scene photographs, other photographs clarify that the toy horse was
unbroken, with all four of its springs attached.


                                              81
Posey’s account, he had observed Stoeckley arriving home between approximately 4:00

and 4:30 a.m. on February 17, 1970, in a blue Ford Mustang. A few days later, after

learning that Stoeckley had been questioned about the murders by police, Posey ran into

Stoeckley and commented that he could provide an alibi for her, but Stoeckley “backed

off” from Posey without further discussion of the matter. A couple days before Posey’s

testimony in the Army proceedings, he sought out Stoeckley in Fayetteville and struck up

a conversation about the murders. In that conversation, Stoeckley said that she would not

commit murder because she was not a hostile person, and that “holding the light would be

the only thing she would do.” Stoeckley also mentioned “a kid’s horse thing,” saying

that “it wouldn’t roll.” She seemed to be referring to the toy horse in Kristen’s bedroom,

which appeared not only in the crime scene photographs presented to Stoeckley at trial,

but also in a photograph that had been taken through Kristen’s bedroom window and

published by a local newspaper immediately after the murders.

      Another Stoeckley witness was Stoeckley’s former Nashville neighbor and co-

worker Jane Zillioux.    According to Zillioux, when she asked Stoeckley, sick with

hepatitis in the autumn of 1970, why she did not return home to obtain care from her

family, Stoeckley responded that her family did not want her around because she had

been involved in “some murders.” Shocked, Zillioux asked Stoeckley, “Well, did you do

it?” Stoeckley said something like, “I don’t know whether I did or not,” elaborating that

“I’ve been a heavy drug user and when you are on drugs, you do funny things.” Pressed

by Zillioux for more information, Stoeckley recollected, “When I came to myself, I was

in the rain,” and “I was terrified.” Flipping her hands, Stoeckley added, “So much blood.

                                           82
I couldn’t see or think of anything except blood.” Stoeckley told Zillioux that she “was

with three boys,” that she did not know them, that she had been with them to obtain

drugs, and that the sight of the blood on her hands prompted her to ask them to take her

home. Stoeckley also said she soon disposed of her blond wig and white boots, as “they

were the same clothes that the woman in the case was supposed to have been wearing.”

During her conversation with Zillioux, Stoeckley was intermittently sobbing, incoherent,

and hysterical.

       A fellow Stoeckley witness and Nashville acquaintance, Charles “Red” Underhill,

said that he found Stoeckley sobbing hysterically in her apartment in December 1970.

When Underhill asked her what was wrong, Stoeckley responded, “They killed her and

the two children,” and “They killed the two children and her.”           In a subsequent

conversation, Stoeckley indicated that the killings had occurred in North Carolina.

       Three more Stoeckley witnesses were law enforcement officers.          Fayetteville

police detective Prince Beasley — the officer who located Stoeckley and questioned her

on February 18, 1970 — said he had immediately thought of Stoeckley, whom he knew

as a reliable informant, upon learning of the intruders described by MacDonald. When

confronted by Beasley about her possible involvement in the murders, Stoeckley told

him, “In my mind, it seems that I saw this thing happen,” but “I was heavy on

mescaline.”       Beasley’s written report of the interview reflected that Stoeckley

remembered nothing of the relevant time frame except perhaps getting into a blue Ford

Mustang that belonged to a man named Bruce Fowler.



                                            83
      Nashville police officer James Gaddis related that Stoeckley sought him out in

early 1971 and asked him to find out if she was still wanted for questioning in the

MacDonald murders. At that time, Stoeckley disclosed that she thought she was present

when the MacDonald murders took place, and that her clothing fit the description of the

alleged female perpetrator. Stoeckley gave Gaddis the names of two persons she had

been with that night, a “Greg” (apparently Greg Mitchell) and the blue Mustang owner

(Bruce Fowler).   On other occasions between late 1970 and early 1971, however,

Stoeckley made conflicting statements to Gaddis, sometimes while obviously under the

influence of drugs and often while exhibiting depression.     Stoeckley variously told

Gaddis, inter alia, that she remembered nothing of the early morning hours of February

17, 1970, except arriving home in a blue car; that she was not involved in the MacDonald

murders but knew who was; that she had only suspicions about who the killers were; and

that she was certain MacDonald was responsible for the murders.

      Army investigator Robert Brisentine, Jr., interviewed Stoeckley in Nashville for

more than ten hours over two consecutive days in April 1971. Stoeckley appeared to be

under the influence of drugs and somewhat depressed on the first day, but in a better

mental condition on the second day. She acknowledged that, because of her heavy drug

use, she had not always been oriented as to time and place, and she had suffered

nightmares since the murders that caused fear and insomnia. As described by Brisentine

and recorded in his written notes, Stoeckley made a series of conflicting statements —

just as she did with Gaddis. Stoeckley said that she could identify the murderers and

would do so if granted immunity from prosecution; that she did not know who the

                                          84
murderers were, but suspected five people (blue Mustang owner Bruce Fowler, his wife

Janice Fowler, a white male named Don Harris, and two black males named Joe Kelley

and “Eddie”); and that, because four hippies could not have invaded the MacDonalds’

apartment without being noticed by neighbors, she was certain it was MacDonald who

murdered his family. At one point, Stoeckley stated that she had heard prior to the

murders that local hippies were angry with MacDonald because he would not prescribe

methadone for their drug addictions, but Stoeckley later retracted that statement and said

she only thought she had heard of MacDonald before the murders.

      Following his interview with Stoeckley, Brisentine personally investigated the

Fowlers and eliminated them as suspects in the MacDonald murders. Brisentine was not

involved in and could not testify to investigations of Don Harris, Joe Kelley, and

“Eddie.”    Both Brisentine and Gaddis indicated that Stoeckley demonstrated no

“insider’s” knowledge of the crime scene. In what could have been an incriminating

statement, Stoeckley told Brisentine that, in her nightmares, she saw “PIG” written in

blood on the headboard of the MacDonalds’ master bed. When Stoeckley demonstrated

how “PIG” was written, however, she wrote it horizontally, rather than vertically as it

actually was inscribed.     Moreover, as the government emphasized in its cross-

examination of Brisentine, the fact that “PIG” was written in blood on the master bed’s

headboard had been included in news reports of the murders.

      Upon hearing from the foregoing Stoeckley witnesses, Judge Dupree denied the

defense’s request to have those six witnesses testify before the jury and relay the

statements Stoeckley made to them about the murders. The judge explained that, because

                                           85
they were uncorroborated and untrustworthy, Stoeckley’s statements constituted

inadmissible hearsay. Additionally, Judge Dupree ruled that the statements were not

admissible for the purpose of impeachment and should be excluded because of a danger

of unfair prejudice and confusing the issues.

                                            (2)

       Some of the Stoeckley witnesses nevertheless testified before the jury about

matters other than Helena Stoeckley’s statements. Her former Fayetteville neighbor

William Posey described seeing Stoeckley arrive home between 4:00 and 4:30 a.m. on

February 17, 1970, in the blue Ford Mustang. Posey could not see the Mustang’s other

passengers, but based on male voices and laughter coming from the car, he deduced that

there were at least two men inside the Mustang. Posey recalled that Stoeckley, in a

purple blouse and with an object in her hand, hurried into her house. According to Posey,

Stoeckley often wore the same purple blouse and a white floppy hat, and she sometimes

wore a blond wig and boots. Posey did not take note of whether Stoeckley was wearing

her wig, hat, or boots when she exited the Mustang. Thereafter, on the day of the

MacDonald funerals, Posey noticed that Stoeckley had donned clothing he had never

seen her wear before, i.e., a long black dress and a black veil over her face. As Posey

described it, Stoeckley quietly sat outside among the funeral wreaths she had hung as if

she was in mourning. Posey said the funeral wreaths were a first for Stoeckley, but that

she regularly had flowers in her apartment that had been discarded by the neighborhood

florist. Posey also testified that Stoeckley and her friends had once sacrificed a cat while

engaged in witchcraft, and that Posey had seen her a few times in the company of a black

                                            86
male who wore a gray Confederate-style military jacket that was different in appearance

from a typical green Army jacket.

       Fayetteville police detective Prince Beasley testified to immediately thinking of

Stoeckley on February 17, 1970, when he learned of MacDonald’s description of the

band of intruders that included two white males, a black male wearing an Army jacket

with Sergeant stripes, and a blond-haired white female in a floppy hat. Beasley explained

that, over the previous two months, he had seen Stoeckley numerous times wearing her

blond wig and one of several floppy hats, hanging out with groups of people that often

included a black male in an Army jacket with Sergeant stripes. Beasley recalled that the

black male always wore the Army jacket, but that such jackets were commonly worn by

men in Fayetteville, as were floppy hats by women. A few days after Beasley’s February

18, 1970 questioning of Stoeckley, she gave Beasley her blond wig and a black floppy

hat, which Beasley held for a short period and then returned to Stoeckley. Beasley noted

in his testimony that Stoeckley usually wore purple or black clothing, and that she had

once admitted to him being part of a “Black Cult” whose members took LSD and

smeared themselves with the blood of cats after slitting their throats. Having reviewed

the March 1970 issue of Esquire magazine reporting on the Manson murders and other

cult-like activities, Beasley said the magazine’s description of California was very much

like parts of Fayetteville in the late 1960s and early 1970s.

                                             (3)

       In the midst of the Stoeckley witnesses’ restricted appearances before the jury, the

defense sought to introduce new self-incriminating statements that Helena Stoeckley

                                             87
made while staying in Raleigh and waiting, pursuant to a subpoena, to possibly be

recalled as a witness. Before ruling on whether he would allow the new statements to be

relayed to the jury, Judge Dupree heard from the relevant witnesses outside the jury’s

presence.

      As developed in the trial court, Stoeckley’s new statements were made to one of

the six existing Stoeckley witnesses, her Nashville acquaintance Red Underhill, and to a

seventh and final Stoeckley witness, MacDonald defense lawyer Wendy Rouder.

Underhill had run into Stoeckley over a weekend trial break at her Raleigh motel, the

Journey’s End, and discovered that she may have just been physically abused by her

companion and fiancé, Ernest Davis. Concerned about Stoeckley’s safety and well-

being, Underhill contacted defense lawyer Bernard Segal, who asked Rouder to check on

Stoeckley. Rouder then went with Underhill to the Journey’s End, where, at Stoeckley’s

request, they convinced Davis to immediately return to South Carolina. After Davis left,

Stoeckley asked Rouder to stay with her, and Rouder spent approximately two hours at

the Journey’s End with Stoeckley.

      According to Rouder, she engaged in mostly small talk with Stoeckley during their

time together at the Journey’s End. At one point, however, Stoeckley paused and brought

up the MacDonald murders, saying, “I still think I could have been there that night.”

Rouder asked her, “What makes you think so?” Stoeckley responded, “I don’t know,”

paused, and then said, “That rocking horse.” After another pause, Stoeckley continued,

“You know, Kristen . . . . Those pictures, when I looked at those pictures, I knew I had



                                          88
seen her somewhere before.” There was an additional pause and Stoeckley said, “And

that driveway, I remember being in that driveway.”

      Shortly thereafter, according to Rouder, she asked Stoeckley if the guilt had ever

left her over the years. Stoeckley answered, “No, what do you think I have taken all

these damn drugs for?” When Rouder then asked Stoeckley if the drugs helped relieve

“this memory,” Stoeckley responded, “No, because you always have to come down.”

Rouder commented that it must be awful trying to live with the guilt, and Stoeckley said

simply, “Yes.”     At another point, Rouder asked Stoeckley, “If MacDonald were

convicted, could you live with that guilt too?” Stoeckley answered, “I don’t think so.”

Rouder persisted, “Isn’t there anything you think you can do to help get rid of the guilt?”

Stoeckley offered, “I just want to take sodium pentothal or hypnosis or something.”

      Because of the earlier disturbance caused by Stoeckley and her fiancé, the

manager of the Journey’s End asked Stoeckley to leave the motel, and the defense

arranged for Stoeckley to transfer to the Hilton near the courthouse.         Rouder and

Underhill helped Stoeckley with the move, and then the trio went together to Underhill’s

motel, the Downtowner, to gather his belongings for his own move to the Hilton.

      In Rouder’s account, while she and Stoeckley were alone together waiting for

Underhill at the Downtowner, Stoeckley again raised the MacDonald murders.

Specifically, Stoeckley commented, “I still think I was there in that house that night.”

Rouder asked, “Helena, is it a feeling you are having or a memory?”             Stoeckley

responded, “It’s a memory. I remember standing at the couch, holding a candle, only —

you know — it wasn’t dripping wax. It was dripping blood.” Rouder followed with,

                                            89
“Helena, why don’t you just go and say that in court?” Stoeckley answered, “I can’t with

those damn prosecutors sitting there.”         Rouder memorialized all of Stoeckley’s

statements in contemporaneous handwritten notes that were read aloud in court.

       As Underhill explained to the trial court, he moved to a Hilton room adjoining

Stoeckley’s at her request. Underhill said that Stoeckley was frightened and seeking

protection, and that she repeatedly said at the Hilton something along the lines of “her life

would not be worth five cents out on the street.”          Stoeckley also told Underhill,

apparently in connection with the MacDonald murders, that she could “name three

people,” but she would not do so in court because, in Underhill’s words, “it would mean

automatically death for her if she did.”

       Having heard from Underhill and Rouder, Judge Dupree barred them from

testifying before the jury about Stoeckley’s new statements. In so ruling, the judge

referred to Stoeckley as “one of the most tragic figures that I have ever had to appear in

court,” and as “a person whose mind is so far impaired and distorted by this drug

addiction that she has become and remains in an almost constant state of hallucination.”

Citing Stoeckley’s obvious paranoia about being involved in the MacDonald murders,

Judge Dupree observed that “what she [says] here in court and what she tells witnesses,

lawyers in a motel room, simply cannot have attached to it any credibility at all.” In sum,

the judge deemed Stoeckley’s various out-of-court statements to be “perhaps the most

clearly untrustworthy evidence that I have had put before me.” Judge Dupree also

disclosed to the parties that Stoeckley had twice contacted him over the weekend break in

the trial, claiming to be “in mortal dread of physical harm” from defense lawyer Segal

                                             90
and requesting counsel to represent her. Thereafter, Stoeckley had a court-appointed

attorney for the trial, and the judge ordered the prosecutors, defense lawyers, and other

witnesses not to have contact with her out of concern that she was “very susceptible to

suggestion.” As it turned out, Stoeckley was not recalled to the witness stand.

                                              d.

      Lastly with respect to Helena Stoeckley, MacDonald revealed in his trial

testimony on direct examination that he had arranged for two sets of police sketches of

the intruders who allegedly murdered his family, including the blond-haired woman in

the floppy hat. MacDonald said that the first sketch of the female intruder, prepared

during his 1970 Army proceedings, did not actually look like the woman. He similarly

criticized the 1970 sketch of the taller white man, but he commended the 1970 sketches

of the shorter white man and the black man.

      The second set of sketches was made in 1979 in anticipation of the trial.

MacDonald asserted that those sketches, prepared while he was under hypnosis, were

much more accurate depictions of the intruders. The 1979 sketch of the female intruder

looked remarkably like Stoeckley — as Stoeckley herself acknowledged in her trial

testimony. The 1979 sketch of the taller white male intruder included a cross necklace

that MacDonald had not previously mentioned.         That sketch, according to multiple

defense witnesses, looked like a soldier named Allen Mazerolle known to have worn a

cross necklace and to have been a part of Stoeckley’s social circle in early 1970. The

defense elicited from Stoeckley that she had dealt drugs for Mazerolle and had a personal

relationship with him, and that Mazerolle had physically abused her.

                                            91
      On cross-examination, MacDonald conceded that, by the time the 1979 sketches

were prepared, he had seen photographs of Stoeckley. Indeed, one of those photographs

had been presented to MacDonald during the 1970 Army proceedings. Asked then if he

recognized Stoeckley, MacDonald had denied recognizing her.              Also on cross-

examination at trial, MacDonald acknowledged that he had not mentioned the cross

necklace on the taller man prior to making the 1979 sketches. MacDonald explained that

was because he had not been prompted to mention the cross necklace when making the

1970 sketches, and because the hypnosis he underwent in 1979 made the cross necklace

come “sharply back into focus.”

                                            e.

      In its closing arguments, the defense noted that it was not obliged to prove who

murdered MacDonald’s wife and daughters, but named Helena Stoeckley and Allen

Mazerolle as likely perpetrators. The government countered that MacDonald’s failure to

identify Stoeckley during the 1970 Army proceedings eliminated her as a suspect.

Furthermore, the government asserted that MacDonald obviously made the second set of

police sketches in 1979 to falsely incriminate Stoeckley and Mazerolle. The government

also emphasized differences between Stoeckley and the female intruder described by

MacDonald and allegedly seen by witnesses.         For example, Stoeckley testified to

discarding her white boots within days of the MacDonald murders, but MacDonald had

insisted to investigators that the female intruder’s boots were brown. The military police

officer responding to MacDonald’s February 17, 1970 call for help saw a woman in a

dark-colored hat, and Fayetteville police detective Prince Beasley recalled Stoeckley

                                           92
giving him a black floppy hat shortly after the murders, but MacDonald said the female

intruder’s hat was white. Although Stoeckley reportedly gave Beasley her blond wig and

then, upon the wig’s return, disposed of it for fear that it implicated her, neither

Stoeckley’s Fayetteville neighbor William Posey nor any other witness put Stoeckley in

the wig at the time of the murders.

       Finally, the government underscored the lack of physical evidence incriminating

Stoeckley, including the absence of her fingerprints at the crime scene. 7 Concurrently,

the government pointed to what it characterized as compelling and unrefuted physical

evidence of MacDonald’s guilt, particularly his blue pajama top, with all the implications

of its bloodstains, tears, and puncture holes; the blue pajama top threads and yarns on the

bloodstained piece of lumber and in other critical locations; the bloody impressions that

the defense expert agreed were consistent with MacDonald’s blue pajama top and

Colette’s pink pajama top on the master bed’s top sheet; and the bloody left footprint in

Kristen’s bedroom that MacDonald had conceded was probably his.

                                          ***

       Following less than seven hours of deliberations, the jury returned verdicts against

MacDonald, finding him guilty of the first-degree murder of Kristen and the second-

degree murders of Colette and Kimberly.



       7
         The trial evidence reflected that the CID team had compared the latent palm and
fingerprints recorded in the MacDonalds’ apartment with the known prints of Stoeckley
without finding any matches.


                                            93
                                            B.

       Through his motions for postconviction relief, MacDonald has sought for decades

to prove that it was Helena Stoeckley and her cohorts — not he — who murdered

Colette, Kimberly, and Kristen.       Much of MacDonald’s focus has been on self-

incriminating statements that Stoeckley continued to make after the 1979 trial, up to her

death in early 1983 from pneumonia and hepatitis-related cirrhosis of the liver. In some

of those varying statements, Stoeckley implicated Allen Mazerolle, i.e., the cross-

necklace-wearing Stoeckley associate depicted in MacDonald’s 1979 police sketch and

named by the defense at trial as a likely perpetrator. Significantly, however, by the time

the district court denied MacDonald’s first postconviction motion in 1985, it had been

established that Mazerolle was in jail at the time of the murders and could not have been

one of MacDonald’s alleged intruders. MacDonald then settled on a different theory of

the murders — an account that is premised on other selected details of Stoeckley’s post-

trial statements and that he continues to endorse today.

                                             1.

                                             a.

       As set forth in the June 2013 memorandum submitted to the district court by

MacDonald following the 2012 evidentiary hearing on his latest 28 U.S.C. § 2255

motion, MacDonald promotes a Stoeckley-statement-based account of the murders that

had Helena Stoeckley and her cohorts angry with MacDonald and other military

physicians for refusing to assist them with their drug problems. See United States v.

MacDonald, No. 3:75-cr-00026, at 50-53 (E.D.N.C. June 10, 2013), ECF No. 343 (the

                                            94
“2013 MacDonald Memorandum”).            Stoeckley’s group included three white men

previously mentioned herein — Greg Mitchell, Bruce Fowler, and Don Harris — plus a

black man known as “Smitty,” whose real name was Dwight Edwin Smith and who also

may have been the “Eddie” once named by Stoeckley. The group members resolved to

persuade MacDonald to provide them drugs and to agree not to report drug users. By

watching MacDonald and his wife Colette, the group members learned where the

MacDonalds lived and that Colette was a student at the Fort Bragg branch of North

Carolina State University.

      Under MacDonald’s theory, some members of Stoeckley’s group went to the Fort

Bragg campus of North Carolina State on the evening of February 16, 1970, to speak to

Colette and seek her intervention with MacDonald. Then, at about 10:30 p.m., Stoeckley,

Mitchell, Harris, Smith, and others met at Stoeckley’s house to discuss going to the

MacDonalds’ apartment and confronting MacDonald. Around 11:00 p.m., Stoeckley,

Mitchell, and Smith went to the Village Shoppe restaurant in Fayetteville and stayed until

closing, when they moved on to the Fayetteville Dunkin’ Donuts. At approximately 2:00

a.m., Stoeckley, Mitchell, and Smith, along with Harris and Fowler, left the Dunkin’

Donuts for the MacDonalds’ apartment.

      As MacDonald would have it, Stoeckley and her cohorts entered the apartment

through the rear exterior doors into the utility room and proceeded straight to the living

room, where they saw MacDonald asleep on the couch with a book across his chest and a

Valentine’s Day card on the couch next to him. The television was on, but because of the

lack of programming at that hour, there was “snow” on the screen. Stoeckley lit a candle

                                           95
to add light to the dimly illuminated room. Stoeckley’s group jostled MacDonald to

awaken him. When MacDonald awoke, he became excited and began to fight the group

members. Smith, dressed in the Army jacket with Sergeant stripes, struck at MacDonald.

Stoeckley, wearing her blond wig, a floppy hat, and boots, said something like, “Hit the

pig,” or “Kill the pig,” and “Acid is groovy.” Around that time, Stoeckley heard Colette

calling for help from another room and gurgling as if clearing her throat. Stoeckley went

to that room and saw Colette being assaulted by Mitchell and a second man. One of the

MacDonald children was in the same room and appeared to be sleeping. At some point,

Stoeckley went into one of the children’s bedrooms and spotted items that included the

toy horse, with a broken or unattached spring. Stoeckley also heard the sound of running

water in the bathroom and saw Mitchell washing his hands in the bathroom sink.

      In the account promoted by MacDonald, Stoeckley answered a ringing telephone

in the apartment when directed to do so by others in her group. A person with a soft

voice asked for “Dr. MacDonald,” and Stoeckley began to laugh. A group member then

ordered Stoeckley to hang up the telephone.         Stoeckley and her cohorts became

frightened and departed the apartment in a hurry, leaving all of their weapons behind,

except for a pair of scissors. One of the group members had stolen items from a jewelry

box. The group returned to Dunkin’ Donuts, where Stoeckley washed blood from her

hands in the restroom sink. Later that morning, around 4:30 a.m., Stoeckley went home.

Several days later, she gave her bloodstained clothing and boots to her friend Cathy Perry

with a request to dispose of them. Meanwhile, Stoeckley discarded her blond wig and

floppy hat. When called as a witness at MacDonald’s 1979 trial, Stoeckley intentionally

                                           96
acted confused to deceive the trial judge for fear of being prosecuted. Pursuant to

MacDonald’s theory, Stoeckley later falsely implicated Allen Mazerolle, knowing he had

been in jail at the time of the murders, as a way to cast doubt on the various self-

incriminating statements she had made. 8

                                             b.

       In support of his Stoeckley-statement-based account of the murders, MacDonald

points to additional self-incriminating statements that came to light only after he filed his

latest 28 U.S.C. § 2255 motion. Helena Stoeckley made the first of those statements to

Jerry Leonard, the lawyer appointed to represent Stoeckley following her testimony at

       8
         The account of the murders promoted by MacDonald is an edited version of the
district court’s 1985 recitation of Stoeckley’s then-known post-trial statements. See
United States v. MacDonald, 
640 F. Supp. 286
, 321-22 (E.D.N.C. 1985). As the court
detailed, many of the statements were given under dubious and likely coercive
circumstances. 
Id. at 318-21.
Furthermore, the court observed that Stoeckley’s post-trial
“statements contain numerous inconsistencies rendering it almost impossible to reconcile
them into one cohesive statement of events.” 
Id. at 321
n.22. The court endeavored,
however, to accord “MacDonald the benefit of all doubts” and “recite in large part what
MacDonald claims Stoeckley’s statements prove occurred on the night of the murders
and thereafter.” 
Id. Details in
the district court’s 1985 recitation omitted from the 2013 MacDonald
Memorandum include that, after subduing MacDonald in the living room, Stoeckley’s
group asked MacDonald to obtain drugs for the group members. MacDonald agreed to
call a friend for drugs from the kitchen telephone, but instead called military police to
report the intruders. The group members overheard MacDonald talking to the military
police and again assaulted him, this time knocking him unconscious. In some of
Stoeckley’s post-trial statements, as additionally noted by the district court in 1985, see
MacDonald, 640 F. Supp. at 322-23
, Stoeckley said that the group members had engaged
in an eight-minute discussion with MacDonald in the living room. Stoeckley also
confessed that her group had broken into the MacDonalds’ apartment on several
occasions prior to the murders, including one in which Stoeckley stole a bracelet with a
blue stone setting.


                                             97
MacDonald’s 1979 trial, while she remained in Raleigh subject to being recalled as a

witness. As a result of Stoeckley’s self-incriminating statements to him, Leonard had

prepared Stoeckley to invoke her Fifth Amendment rights if she returned to the witness

stand. Leonard did not share Stoeckley’s statements with anyone until 2012, when the

district court reviewed the statements in camera and ruled that Leonard could disclose

them at MacDonald’s evidentiary hearing, because exceptional circumstances warranted

breaching the attorney-client privilege.

       According to Leonard, in speaking with Stoeckley in a private conference room at

the federal courthouse during MacDonald’s 1979 trial, he explained that their

conversations were protected by the attorney-client privilege. They also discussed legal

issues such as the statute of limitations for murder charges. Stoeckley initially asserted

that she could recall nothing of the early morning hours of February 17, 1970. Later,

Stoeckley asked Leonard what he would do if Stoeckley actually had been present during

the MacDonald murders. Once Leonard advised that he still would help Stoeckley and

needed to know the truth, Stoeckley commented that she had been afraid to tell Leonard

the truth, but that the truth was “not as bad as everybody thought.” Stoeckley then said

that she had been at the MacDonalds’ apartment, and that she did not hurt anyone there

and had not anticipated that anyone would be hurt.

       In line with the account endorsed by MacDonald, Stoeckley told Leonard that the

idea to go to the MacDonalds’ apartment came up one night while Stoeckley was doing

drugs with her friends.     At least one member of the group had a grudge against

MacDonald because of his perceived discrimination against hard drug users. That group

                                           98
member persuaded the others to go to the apartment to confront MacDonald. The end

result was that things got out of hand and the murders occurred. During the violence,

Stoeckley answered a ringing telephone in the apartment and then quickly hung up at the

direction of another group member. Stoeckley also noticed the toy horse and that one of

its springs was not attached.

       The other self-incriminating statements on which MacDonald relies were made by

Stoeckley to her mother, who was also named Helena Stoeckley, on two occasions — the

first sometime after the 1979 trial, and the second in approximately October 1982 when

the younger Stoeckley knew her death was imminent. The elder Stoeckley memorialized

those statements in an affidavit in March 2007, shortly before her own death.       As

reflected in that affidavit, the younger Stoeckley said on both occasions that she was

present at the MacDonald murders. In her second conversation with her mother, the

younger Stoeckley provided details somewhat consistent with MacDonald’s present

theory of the murders. That is, the younger Stoeckley said she and her cohorts went to

the MacDonalds’ apartment to “intimidate” MacDonald because of his harsh treatment of

local drug users. The group was high on drugs and did not set out to harm anyone. Once

inside the apartment, the younger Stoeckley saw the toy horse in one of the children’s

bedrooms. In this version of the murders, however, the younger Stoeckley was with three

men, and only Greg Mitchell was identified by name. The younger Stoeckley witnessed

the stabbing of MacDonald by an unspecified perpetrator. Mitchell and a second man

went “out of their minds” and killed MacDonald’s family. When the younger Stoeckley

and the third man realized what was happening, they fled the apartment.

                                           99
       The elder Stoeckley asserted in her affidavit that her daughter had tried to tell the

truth to the FBI and other law enforcement officials, but their response to the younger

Stoeckley was to “keep quiet.” Furthermore, the elder Stoeckley said that she and her

since-deceased husband had received several requests from the FBI after the murders for

the younger Stoeckley “to stop contacting them.” The elder Stoeckley also recounted

that, in their second conversation, the younger Stoeckley revealed that she had lied at

MacDonald’s trial because she “was afraid of the prosecutor.” At the close of her

affidavit, the elder Stoeckley declared her belief that “Helena was telling me the full

truth” in her self-incriminating statements, and that “she wanted [to] ‘set things straight’

before she died.”

       During the 2012 evidentiary hearing, Eugene Stoeckley — the elder Helena

Stoeckley’s son and the younger Helena Stoeckley’s brother — testified that the 2007

affidavit arose from conversations with his mother about the MacDonald murders, which

had occurred when he was eleven years old and had long been a forbidden topic among

his family members. The crux of Eugene Stoeckley’s conversations with his mother had

been that she believed the younger Helena Stoeckley was present during the murders and

MacDonald was not guilty of the crimes.

                                             c.

       In addition to the self-incriminating statements of the younger Helena Stoeckley,

MacDonald relies on confessions allegedly made by Greg Mitchell of his own

involvement in the murders. For example, an employee of a Fayetteville ministry that

provided counselling to young people with drug and alcohol problems executed a

                                            100
declaration that, in early 1971, a man who looked like Mitchell broke into the ministry’s

farmhouse and wrote the words “I killed MacDonald’s wife and children” in red paint on

one of the walls. In a series of sworn statements and FBI interviews that became more

detailed over time, a husband and wife who had worked and socialized with Mitchell in

Charlotte, North Carolina, from 1972 until Mitchell’s alcoholism-related death in 1982

recounted that a depressed and heavy-drinking Mitchell had confessed generally to a

“serious crime” at Fort Bragg and specifically to the MacDonald murders. The sole

detail that Mitchell shared in any of his alleged confessions was the motive for going to

the MacDonalds’ apartment. In one confession, Mitchell and his friends went to the

apartment to “teach [MacDonald] a lesson” for failing to help Mitchell with his heroin

addiction — not to kill anyone — but “things got bad” because Mitchell’s group was

high on mescaline, angel dust, PCP, and a fourth drug. In another confession, Mitchell

and his friends had been shipping heroin from Vietnam to the United States in body bags,

believed that MacDonald was intercepting the heroin, and went to the apartment to

demand money or “dope.”

                                           d.

      MacDonald also relies on self-incriminating statements made by Helena

Stoeckley’s friend Cathy Perry. Although the Stoeckley-statement-based account had

Perry involved only after the murders — when Stoeckley allegedly gave her bloodstained

clothing and boots to Perry for disposal — Perry placed herself in the MacDonalds’

residence when the murders were committed. Specifically, Perry told an FBI agent in

1984 that she went to the residence, which she described as a two-story house, with two

                                          101
other white females and five or six white males. In Perry’s version of events, the group

broke into the house through the front door, saw a man lying on the couch, and injected

him with drugs, causing him to collapse. The group proceeded upstairs, where the

members pounded on a baby boy wrapped in a blanket. After trying to protect the baby

but being forced to hit him, Perry spotted another male child and hid him in a closet.

That boy was subsequently taken to a bathroom and killed there.         Meanwhile, the

children’s mother woke up once but went back to sleep. Perry then woke the mother

again and unsuccessfully attempted to persuade her to jump out a window with Perry. A

little while later, when one of the group members ordered Perry to tie up the mother and

kill her, Perry stabbed the mother several times in the leg and abdomen. After murdering

the mother, Perry wrote in blood on the wall, “Fuck you pigs from all of us to you,” with

the year “1970.” Perry’s statement to the FBI was that her group was in the house from

about 11:00 p.m. until 4:00 or 5:00 a.m. She described the weather as warm and clear.

       Around the time of the MacDonald murders, Perry was known to have stabbed

two people and her pet puppy. When she made her statement to the FBI, Perry was under

a physician’s care for schizophrenia.

                                           e.

       MacDonald seeks to bolster his Stoeckley-statement-based account of the murders

with additional witnesses. Most notably, MacDonald contends that Helena Stoeckley’s

claim that she answered a ringing telephone is corroborated by a witness who said he

called the MacDonalds’ apartment at about 2:00 a.m. on February 17, 1970.           That

witness, Jimmy Friar, executed a declaration in 1983 while imprisoned in South Carolina.

                                          102
In his declaration, Friar explained that he was a patient at Fort Bragg’s Womack Army

Hospital in February 1970. Prior to that time, he had been hospitalized at the Walter

Reed Army Medical Center in Washington, D.C., where he became friendly with a Dr.

Richard MacDonald. On a couple of occasions, Friar had left Walter Reed, become

intoxicated, and called Dr. Richard MacDonald for help returning to the facility. On

February 16, 1970, Friar left Womack to drink and shoot pool in Fayetteville; became

intoxicated and disoriented; and, thinking he was in Washington, attempted to call Dr.

Richard MacDonald.

       According to Friar, he called the Fort Bragg operator from a Fayetteville hotel and

represented himself as a physician friend of “Dr. MacDonald,” without specifying a first

name. The operator gave Friar a number, which he called, asking for “Dr. MacDonald.”

The woman who answered was laughing, and someone in the background said, “Hang up

the goddamned phone.” The call was then disconnected. Friar expressed certainty that

he made the call around 2:00 a.m., as he “had decided to leave for Fort Bragg sometime a

little before 2:00 a.m.”

       To corroborate Stoeckley’s claim that, hours before the murders, members of her

group confronted Colette at the Fort Bragg campus of North Carolina State University,

MacDonald points to the affidavit of Edith Boushey, who taught English there. Boushey

said she witnessed a man she identified as Greg Mitchell talking to Colette at about 9:40

p.m. on February 16, 1970, near a stairwell of a campus building. As Boushey walked

by, she heard the man say, “If you go along, I think it will be alright.” Boushey heard



                                           103
only the first part of Colette’s response, “I dread . . . .” The man talking to Colette was

accompanied by at least three women, including two wearing floppy hats.

       Other witnesses confirmed that Stoeckley associated with men she implicated in

the murders —Mitchell, Don Harris, and Dwight Edwin Smith — in addition to men

including a black man nicknamed “Moses” who, like Smith (or “Smitty”), was known to

wear an Army jacket with Sergeant stripes. Witnesses also said they saw persons around

Fayetteville who looked like Stoeckley and her cohorts, in assorted configurations, both

in the days and hours before and after the murders.

       For example, two witnesses saw a white woman in white knee-high boots and a

white floppy hat in the Fayetteville Dunkin’ Donuts between midnight and 1:30 a.m. on

February 17, 1970. The first of those witnesses said the woman was with three men, and

the second witness said she was with two. Both witnesses included a black man in an

Army-type jacket in the group. In the second witness’s account, she looked out the

restaurant window after the group left the Dunkin’ Donuts at about 1:30 a.m. and spotted

a black or blue van stopped outside with the black man driving and the white woman in

the passenger seat. A white man on a black motorcycle pulled up near the van, and the

black man said to the motorcycle rider, “We’ll see you there.” The van then headed

toward Fort Bragg.

      Those and other witnesses variously said persons they saw resembled Stoeckley,

Mitchell, Harris, Smith, and some of MacDonald’s police sketches, and they reported

vehicles including not only a dark-colored van and a motorcycle, but also a white van and

a dark-colored sedan. The men in the group were described sometimes as having short

                                           104
hair and sometimes as having long hair. When the color of the Stoeckley-like woman’s

boots was specified, the boots were white or light-colored, as was her clothing in some

descriptions.

                                              f.

       Lastly in advancing his Stoeckley-statement-based account of the murders,

MacDonald relies on physical evidence and expert opinions he discovered and obtained

after his trial, including the following:

       ●        Three blond synthetic hairs, of up to twenty-two inches in length,
                that were removed from a hairbrush found next to Colette’s purse on
                the buffet in the MacDonalds’ dining room, and that were not
                matched to any item in the apartment;

       ●        Five black wool fibers collected from the crime scene — two from
                the bloodstained piece of lumber located on the ground outside the
                apartment’s rear entrance, one from the right biceps area of Colette’s
                pink pajama top, and two from the area around Colette’s mouth —
                that also were not matched to any item in the apartment;

       ●        An expert’s opinion that some of Colette’s and Kristen’s stab
                wounds were consistent with scissors; and

       ●        The same expert’s separate opinion that a blow that fractured
                Colette’s skull was consistent with a frontal clubbing by a left-
                handed assailant, along with evidence that Greg Mitchell was left-
                handed and MacDonald is right-handed.

       MacDonald’s position is that the three blond synthetic hairs are corroborative of

the blond-wig-wearing Helena Stoeckley’s presence during the murders, and that the five

black wool fibers are proof of intruders and very well may have come from the black

clothing routinely worn by Stoeckley. Moreover, the expert’s first opinion substantiates

Stoeckley’s claim that she and her cohorts used scissors in the murders that they took


                                             105
with them when they fled the crime scene, and the expert’s second opinion tends to

simultaneously implicate Greg Mitchell in the clubbing attack on Colette and to

exonerate MacDonald.

       In MacDonald’s view, the foregoing evidence augments other physical evidence

of intruders introduced during his trial, such as the dried wax in the living room and

Kimberly’s bedroom, and the identifiable but unsourced latent palm and fingerprints in

various areas of the apartment. MacDonald also emphasizes the potentially exculpatory

“lost” evidence identified at trial — particularly his blue pajama pants and the section of

the wood floor in Kristen’s bedroom containing the bloody left footprint — as well as

such evidence he only learned about postconviction. That unavailable and potentially

exculpatory evidence includes a pair of beige-colored boots, and perhaps some clothing,

that may have been bloodstained and may have belonged to Stoeckley or her friend Cathy

Perry. It also includes what MacDonald refers to as the “bloody syringe,” i.e., a syringe

described in investigation notes as “a half-filled syringe that contained an as-yet unknown

fluid . . . located in a hall closet, which also contained some evidence of blood.”

MacDonald commends the lost items as evidence that could have proved the culpability

of Stoeckley and her cohorts had it been preserved.

                                            2.

       For its part, the government roundly discredits MacDonald’s Stoeckley-statement-

based account of the murders.      To start, many details of that account are patently

incompatible with the physical evidence and with MacDonald’s own pretrial statements

and trial testimony. For example, whereas Helena Stoeckley had her group entering the

                                           106
MacDonalds’ apartment, heading straight to the living room, and jostling MacDonald to

awaken him, MacDonald said he was roused by screams and cries for help from Colette

and Kimberly, who apparently had already been attacked. In Stoeckley’s account, there

was a Valentine’s Day card on the couch next to MacDonald and a toy horse with a

detached spring in one of the children’s bedrooms.         But in reality, there was no

Valentine’s Day card on the couch or elsewhere in the living room, and the toy horse, in

Kristen’s bedroom, was unbroken. 9

       The government contends that the statements MacDonald deems to be the most

credible — those allegedly made by Stoeckley in confidence during the 1979 trial to her

lawyer and in anticipation of death in late 1982 to her mother — very well may have

been fabricated or misremembered.        Even accepting that Stoeckley did make the

statements to her lawyer and mother, the government maintains that they are no more

trustworthy than any of Stoeckley’s other conflicting and uncorroborated stories. The

government describes the statements as little more than “I was there” admissions.

Moreover, the scant details contained in Stoeckley’s confession to her mother are only

partially consistent with MacDonald’s Stoeckley-based-statement account of the murders,

diverging over, e.g., Stoeckley and her cohorts’ departure from the MacDonalds’


      9
         Further at odds with MacDonald’s testimony and other evidence, Stoeckley had
MacDonald engaging in an eight-minute discussion with Stoeckley and her cohorts in the
living room, agreeing to call a friend from the kitchen telephone to obtain drugs for them,
calling and talking to military police instead, and being knocked unconscious in the
kitchen. Stoeckley also confessed to several prior break-ins and the theft of a bracelet
that MacDonald never reported.


                                           107
apartment. As MacDonald would have it, the group members left the apartment together,

having become frightened after Stoeckley answered the ringing telephone. Stoeckley told

her mother, however, that she fled the apartment with just one male cohort when the pair

realized the other group members were murdering MacDonald’s family. 10

      The government emphasizes that Stoeckley’s lawyer, Jerry Leonard, testified at

the 2012 evidentiary hearing that he did not feel compelled to disclose the self-

incriminating statements that Stoeckley made to him, as she provided few details, she

also told him she had no memory of the early morning hours surrounding the murders,

and her contradictory declarations were unexceptional, in that it was well-known that she

uttered many conflicting statements to a multitude of persons. The government further

underscores that Stoeckley’s mother, the elder Helena Stoeckley, expressed confidence in

her daughter’s innocence on several occasions — including in an interview with

MacDonald’s defense team in August 1979 (during the trial prior to the younger

Stoeckley’s testimony), an FBI interview in July 1984 (following the younger

Stoeckley’s October 1982 confession and January 1983 death), and a second FBI

interview in April 2007 (soon after the elder Stoeckley signed the March 2007 affidavit

attesting to the October 1982 confession). In those interviews, the elder Stoeckley said

that the younger Stoeckley had soberly and clearly denied involvement immediately after



      10
          In yet another confession — to Sara McMann, a woman with whom Stoeckley
and her then-infant son were living in late 1982, just prior to her death — Stoeckley said
that she “ran out screaming” from the MacDonalds’ apartment.


                                           108
the murders, but that she was “a physical and mental wreck” by 1979 and easily

influenced to “talk all kinds of nonsense.” 11

       Turning to the confessions of Stoeckley’s alleged accomplices, Greg Mitchell and

Cathy Perry, the government argues that those confessions also lack credibility.

Mitchell’s confessions, assuming they actually occurred, contained no details of the

murders other than varying statements of motive. By contrast, Perry’s self-incriminating

statements to an FBI agent were full of details — but details wholly incompatible with

the known facts and evidence. Indeed, Perry was not even implicated in the murders by

Stoeckley. As for Mitchell and the other three men who were named in the Stoeckley-

statement-based account of the murders — Bruce Fowler, Don Harris, and Dwight Edwin

Smith — the FBI investigated them without uncovering any evidence of their

involvement.    Additionally, none of the identifiable but unsourced latent palm and

fingerprints recorded in the MacDonalds’ apartment was matched to Mitchell.

       With respect to the additional witnesses invoked by MacDonald to corroborate his

Stoeckley-statement-based account of the murders, the government especially attacks the

credibility of Jimmy Friar, who said he called the MacDonalds’ apartment around 2:00

       11
         In her statements of confidence in her daughter’s innocence, the elder Stoeckley
blamed Fayetteville police detective Prince Beasley — a father figure (“Daddy Beasley”)
to the younger Stoeckley — for convincing her that she was involved in the murders. It
was Beasley who identified the younger Stoeckley as a suspect, questioned her the day
after the murders, and thereafter helped secure many of the post-trial confessions
underlying MacDonald’s Stoeckley-statement-based account. The government has
produced voluminous evidence casting doubt on Beasley’s reliability as an investigator
and his motives in pursuing the younger Stoeckley, such as a financial stake in a
proposed book about the MacDonald case.


                                             109
a.m. on February 17, 1970, while a fugitive Womack Army Hospital patient. The record

reflects that Friar emerged as a potential witness during MacDonald’s 1979 trial, and that

he then and thereafter gave a number of conflicting statements about the purported

telephone call. Although MacDonald’s defense team opted not to call Friar as a witness

at trial, Friar’s story circulated around the federal courthouse and was published by the

media around the very time that Stoeckley was in Raleigh testifying and awaiting

possible recall to the witness stand. Strikingly, that is when Stoeckley allegedly first

claimed — to her lawyer, Jerry Leonard — that she had answered a ringing telephone in

the MacDonalds’ apartment.        Stoeckley then inserted the same claim in post-trial

confessions, and Friar executed the 1983 declaration on which MacDonald now relies.

The 1983 declaration, however, is neither consistent with Friar’s prior statements (which

were not, in any event, consistent with each other), nor facially plausible in isolation. For

example, the 1983 declaration insists Friar made the call around 2:00 a.m., before

MacDonald said he went to sleep on the living room couch and thus before the alleged

invasion of the apartment by Stoeckley and her cohorts. Moreover, Friar was supposedly

so intoxicated and disoriented that he thought he was in Washington, D.C., yet he had the

presence of mind to call the Fort Bragg operator, obtain the phone number for “Dr.

MacDonald,” place the call to the MacDonalds’ apartment, and years later recollect in

detail those actions and his exchange with the woman who answered.

       The government next points to evidence contradicting Edith Boushey, the English

instructor who said she saw a man who looked like Greg Mitchell, accompanied by

women in floppy hats, confront MacDonald’s wife Colette at the Fort Bragg campus of

                                            110
North Carolina State University around 9:40 p.m. on February 16, 1970.                 The

government counters Boushey’s affidavit with an affidavit from Colette’s friend

Elizabeth Ramage, who carpooled and attended class with Colette that evening. Ramage

said she was with Colette at all relevant times, did not witness any unusual event such as

the encounter described by Boushey, and left the campus with Colette between 9:20 and

9:30 p.m. MacDonald himself said that Colette arrived home at about 9:40 p.m. after

stopping at a convenience store for milk, and she apparently did not tell MacDonald

about any confrontation with drug users asking her to intervene with him on their behalf.

       At best, according to the government, other witnesses placed Stoeckley and her

associates in Fayetteville and the general Fort Bragg area around the time of the murders.

Of course, some of the witnesses’ descriptions — e.g., of a woman in white boots and

men with long hair — do not match MacDonald’s descriptions of the brown-boots-

wearing female intruder and her short-haired male accomplices.

       The government further asserts that the physical evidence and expert opinions that

MacDonald discovered and obtained following his trial do not tend to prove the

Stoeckley-statement-based account of the murders. More specifically, as the government

articulates:

       ●       It has never been established whether the three blond synthetic hairs
               removed from the hairbrush in the MacDonalds’ dining room came
               from a doll (as the government deems likely) or from a wig (as
               MacDonald says is possible);

       ●       Although MacDonald speculates that the five black wool fibers
               collected from the crime scene came from Stoeckley’s clothing,
               laboratory analyses showed that no two of those fibers appeared to
               have originated from the same source, thereby suggesting that the

                                            111
             fibers were just more of the unsourced debris found throughout the
             apartment and acknowledged by the investigators at trial;

      ●      MacDonald’s own expert qualified his opinions that some of
             Colette’s and Kristen’s stab wounds were consistent with scissors
             and that Colette was clubbed by a left-handed assailant, explaining
             that the stab wounds were also consistent with an ice pick and that
             an enraged or intoxicated right-handed assailant could deliver a left-
             handed-type blow; and

      ●      The Stoeckley-statement-based account, that Stoeckley’s group left
             all the murder weapons behind except the alleged scissors,
             conspicuously omits the baseball-bat-like club that MacDonald
             claimed was used to attack him but that was not found at the crime
             scene.

      Contesting the importance of the physical evidence of intruders highlighted by

MacDonald, as well as the “lost” evidence that he argues could incriminate Stoeckley and

her cohorts, the government makes these additional points:

      ●      The three samples of wax, each originating from a different source,
             were more indicative of the MacDonalds’ fondness for candles —
             they had fourteen in their apartment — than of the single candle-
             carrying intruder described by MacDonald;

      ●      Although there were a number of identifiable but unsourced latent
             palm and fingerprints in the apartment, none was matched to
             Stoeckley or Mitchell, and none was recorded in a particularly
             significant place, such as one of the murder weapons;

      ●      MacDonald’s discarded blue pajama pants were an unlikely source
             of the distinctive threads and yarns found throughout the three
             bedrooms, as there were no such threads and fibers in other rooms,
             including the living room, bathroom, and kitchen, where MacDonald
             said he wore the pajama pants during and after the attack;

      ●      Even if the threads and yarns came from MacDonald’s blue pajama
             pants, their presence in places like the bloodstained piece of lumber
             remains inculpatory;



                                          112
       ●     MacDonald has admitted that the bloody left footprint in Kristen’s
             bedroom was likely his and has identified no other conceivable
             source;

       ●     The beige boots that MacDonald asserts may have been bloodstained
             and may have belonged to Stoeckley or Cathy Perry were
             temporarily in the custody of investigators and determined not to
             contain any blood; and

       ●     With respect to the so-called “bloody syringe” referenced in the
             investigation notes, it appears that the note taker was mistaken about
             the existence of “a half-filled syringe that contained an as-yet
             unknown fluid,” and that, in any event, the “evidence of blood” was
             the blood of MacDonald’s type B at the juncture of the linen closet
             doors, and not blood on any syringe in that closet.

       Finally, aside from disputing MacDonald’s Stoeckley-statement-based account of

the murders as implausible and insufficiently corroborated, the government points out

that evidence has materialized since the trial that is unhelpful to MacDonald. Under that

evidence, MacDonald had lost twelve to fifteen pounds over the three to four weeks prior

to the murders with the help of diet pills, or “speed,” used with another medication “to

counteract the excitability of the speed.” MacDonald intentionally withheld from the

investigators that he may have taken a diet pill with dinner just hours before the murders

— thereby concealing a possible explanation for why he lost control and brutalized his

family. 12


       12
          It has also been clarified in the postconviction proceedings that the urine stain
on the bottom sheet of the MacDonalds’ master bed was matched to Kimberly (contrary
to MacDonald’s claim that Kristen had wet the bed), and that at least one of the blue
fibers collected from the hallway floor near the living room did not match MacDonald’s
blue pajama top (at least somewhat undermining his assertion that the fibers corroborate
his account of being knocked unconscious in the living room and waking sometime later
lying facedown over the step between the living room and hallway).


                                           113
                                           3.

      Rather than acknowledging and grappling with much of the government’s multi-

pronged attack on his Stoeckley-statement-based account, MacDonald simply declares

that no reasonable factfinder hearing the account could convict him of murdering his wife

and daughters. As MacDonald would have it, Helena Stoeckley’s account constitutes

definitive proof “that she was in the MacDonald house and saw the murders.” See 2013

MacDonald Memorandum 3.          In particular, the circumstances of Stoeckley’s self-

incriminating statements to her lawyer and mother indicate “substantial reliability and

trustworthiness.” 
Id. Far from
providing too few details of the murders, Stoeckley said

enough to convince the jury of her guilt:       “‘I was there.’”    
Id. at 4.
  One of the

accomplices named by Stoeckley, Greg Mitchell, also “directly and unambiguously

confessed to murdering the MacDonald family.”          
Id. at 61.
   And even the friend

Stoeckley did not implicate, Cathy Perry, should be viewed as a viable suspect

notwithstanding her own preposterous confession, for Perry’s other “stabbings at the time

of the [MacDonald murders] and her complete psychological disintegration almost

immediately after the [murders] demonstrate she was the type of person capable of

committing such crimes.” 
Id. at 64.
      In MacDonald’s view, the Stoeckley-statement-based account is thoroughly

corroborated by Jimmy Friar, who said he called the MacDonalds’ apartment around 2:00

a.m. on February 17, 1970; Edith Boushey, who said she witnessed a confrontation

earlier that night between MacDonald’s wife Colette and a Mitchell-like man at the Fort

Bragg campus of North Carolina State University; and all the other witnesses who said

                                          114
they saw persons around Fayetteville who looked like Stoeckley and her cohorts in the

days and hours before and after the murders. MacDonald makes no effort to address

questions about the credibility of those witnesses and discrepancies within and between

their various statements.

       Furthermore, MacDonald insists that the three blond synthetic hairs and the five

black wool fibers, because they could have come from Stoeckley’s wig and clothing,

corroborate the presence of intruders and the Stoeckley-statement-based account. For

additional corroboration, MacDonald invokes his expert’s opinions about wounds made

by scissors and by a left-handed, club-wielding assailant, without acknowledging the

expert’s qualifications of those opinions. MacDonald also relies on allegations of the lost

“bloodstained” beige boots and the “bloody syringe,” without acknowledging evidence

that the beige boots were analyzed and determined to be blood-free, and that the syringe,

bloody or not, never existed.

       In other instances, MacDonald has altered his own prior statements and testimony,

apparently to better meet the evidence. For example, whereas MacDonald previously

said the female intruder appeared to be carrying a candle, and three different samples of

dried wax were found in his family’s apartment, he now elaborates that the female

intruder “appeared to be carrying one or more lit and dripping candles.” See 2013

MacDonald Memorandum 47. Whereas MacDonald previously said he was attacked

with a baseball-bat-like club that could not have been the bloodstained piece of lumber

found outside the apartment, and he could not explain how two threads matched to his

blue pajama top came to be on that piece of lumber, he now asserts that he was

                                           115
“repeatedly struck by a club or clubs,” which may have included the bloodstained piece

of lumber, and that “his pajama fibers could have stuck to the [piece of lumber] while he

was being attacked.” 
Id. at 43.
       MacDonald disregards and dismisses other unfavorable evidence — such as the

government’s evidence that he may have been taking “speed” at the time of the murders

— while lauding every piece of evidence that could possibly corroborate the Stoeckley-

statement-based account.       In MacDonald’s words, his evidence is “dramatic,”

“remarkable,” and “compelling,” and it “would have turned the tables at the trial.” See

2013 MacDonald Memorandum 101, 103, 121.

                                              C.

       Against the backdrop of this and other evidence at trial and in the postconviction

proceedings, MacDonald now seeks relief on his latest 28 U.S.C. § 2255 motion. The

first claim raised therein is the Britt claim, i.e., the prosecutorial misconduct claim based

on the newly discovered evidence of former Deputy U.S. Marshal Jimmy Britt that a

threat from prosecutor James Blackburn stopped Helena Stoeckley from confessing on

the witness stand during MacDonald’s trial.

                                              1.

       In a series of sworn statements executed between February 2005 and February

2006, Jimmy Britt said that, while assigned to MacDonald’s 1979 trial as a Deputy U.S.

Marshal, he had been tasked with picking up Helena Stoeckley in South Carolina and

bringing her to Raleigh to testify pursuant to the material witness warrant. According to

Britt, during the drive from South Carolina to Raleigh, Stoeckley brought up the

                                            116
MacDonald murders, mentioned the toy horse in the MacDonalds’ apartment, and

indicated she and others were in the apartment when the murders occurred. Stoeckley’s

statements were allegedly witnessed by Geraldine Holden, a U.S. Marshals Service

employee identified by Britt as his official travel companion.

       The day after arriving in Raleigh, in Britt’s account, he was responsible for getting

Stoeckley to the federal courthouse. He escorted Stoeckley first to the office being used

by MacDonald’s defense team, and then to the United States Attorney’s office. Britt did

not enter the defense team’s office with Stoeckley, but he did go into the United States

Attorney’s office with her, on the invitation of prosecutor James Blackburn, then an

Assistant U.S. Attorney. During Blackburn’s subsequent interview of Stoeckley, other

prosecutors may have been in and out of the room. While Blackburn and Britt were

alone with Stoeckley, however, she made statements similar to those made during the

previous day’s car trip.

       As recounted by Britt, Stoeckley mentioned the toy horse “and various other

things” to Blackburn, and she said she and other persons were in the MacDonalds’

apartment, where she had gone to acquire drugs. Britt alleged that Blackburn’s response

was an improper threat to prosecute Stoeckley, in these or very similar words: “If you

testify before the jury as to what you have told me or said to me in this office, I will

indict you for murder.” Britt also suggested unethical conduct between Blackburn and

the trial judge, asserting, inter alia, that he escorted Stoeckley to the courtroom

immediately after the interview and waited there with MacDonald’s defense lawyers for



                                            117
the court proceedings to resume while Blackburn had an ex parte meeting with Judge

Dupree in a private office.

       Britt’s sworn statements varyingly specified Charleston and Greenville as the

place in South Carolina where he first picked up Stoeckley, which would have been on a

Wednesday. The statements also varyingly specified the local Wake County jail and a

Holiday Inn as the spot in Raleigh where Britt retrieved Stoeckley that Thursday morning

to bring her to the federal courthouse for her interviews. One statement had Britt taking

Stoeckley from the courthouse to the Journey’s End motel following her testimony that

Friday; moving Stoeckley to the Holiday Inn over the weekend trial break; and, on

Monday, buying a one-way ticket and putting Stoeckley on a bus back to South Carolina,

after Judge Dupree purportedly instructed the jury to disregard Stoeckley’s Friday

testimony and ruled that she could not retake the witness stand “because her brain was

scrambled like an egg.”

       In early 2005, Britt contacted MacDonald defense lawyer Wade Smith to tell him

about witnessing Stoeckley’s self-incriminating statements and Blackburn’s threat.

Within the previous few years, Britt had shared his account with two friends and former

U.S. Marshals Service colleagues. Britt said he had been deeply troubled by what he

witnessed, but he waited more than two decades to report it out of respect for Judge

Dupree, who died in 1995, and other court employees. Britt explained in a November

2005 affidavit: “Working on the side of law enforcement in the courthouse was my

career. I did not want to betray, or appear to be betraying, the people I worked with and

respected.”

                                          118
      Jimmy Britt himself died in 2008 and thus did not provide testimony at the 2012

evidentiary hearing.   No statement was obtained from Geraldine Holden, who was

seriously ill when Britt came forward with his account in 2005 and subsequently passed

away. James Blackburn testified during the 2012 evidentiary hearing and disputed Britt’s

account of Blackburn’s interview of Helena Stoeckley.            Specifically, Blackburn

recounted that three other prosecutors were with him for the Stoeckley interview, and that

Britt was not in the room.       Blackburn denied hearing Stoeckley make any self-

incriminating statements or threatening her with a murder prosecution. Additionally,

Blackburn recalled that there were no court proceedings that day following the

interviews, contradicting Britt’s account of waiting with Stoeckley and the defense team

in the courtroom while Blackburn had an ex parte meeting with Judge Dupree.

                                            2.

      MacDonald argues that James Blackburn is not credible, in that he was disbarred

as a lawyer in 1993 after being convicted of crimes including embezzlement, obstruction

of justice, and forgery for his post-MacDonald-trial conduct in private practice. To

corroborate Jimmy Britt, MacDonald relies on a newspaper photograph of Britt escorting

Helena Stoeckley during the 1979 trial, as well as a polygraph examination that indicated

Britt was truthful about witnessing Stoeckley’s confessions and Blackburn’s threat to

prosecute her. Specifically, the polygraph examiner detected no deception on Britt’s part

when he answered “yes” to the questions, “Did you hear Helena Stoeckley tell Jim

Blackburn she had seen a broken hobby horse while she was inside the MacDonald

home,” and “Did you hear Jim Blackburn tell Helena Stoeckley he would have her

                                           119
indicted for murder if she testified she had been inside the MacDonald home?” The

examiner also detected no deception in Britt’s response of “no” when asked, “Are you

now lying about the conversation between Jim Blackburn and Helena Stoeckley?”

       For further corroboration, MacDonald points to Mary Britt, who was married to

Jimmy Britt at the time of MacDonald’s trial. Mary Britt testified at the 2012 evidentiary

hearing that she remembered Jimmy Britt coming home one day during the trial and

saying he had to travel to South Carolina the following day to pick up a witness. When

he next returned home, Jimmy Britt “was very excited” and told his wife that the witness

he had transported “talked in the car coming back about her involvement” and “described

the inside of the apartment where the MacDonalds lived . . . to a T even to the fact of a

child’s hobby horse that was broken.” The following evening, Mary Britt asked about the

witness’s trial testimony that day, and Jimmy Britt responded that “they say they can’t

use her testimony because her brain is fried from the use of drugs.”

       MacDonald also relies on Lee Tart, Jimmy Britt’s friend and former U.S. Marshals

Service colleague, who confirmed under oath in a 2005 interview that Britt told him in

September 2002 about Stoeckley’s self-incriminating statements and Blackburn’s threat.

Tart said Britt’s demeanor indicated that he was troubled by what he allegedly witnessed,

and they later discussed it on a second occasion.

       According to MacDonald, Britt’s account is generally corroborated by Stoeckley’s

similar self-incriminating statements, particularly her confessions to her lawyer Jerry

Leonard during the 1979 trial and just before her death to her mother in late 1982.

MacDonald claims more specific substantiation from the elder Helena Stoeckley’s

                                           120
recollection of her daughter saying she had lied in her trial testimony because she “was

afraid of the prosecutor.”     Additionally, MacDonald argues that, like the younger

Stoeckley’s confessions to Leonard and the elder Helena Stoeckley, her confession to

Britt possesses substantial reliability and trustworthiness because of the circumstances in

which it was made.

       MacDonald defense lawyer Wendy Rouder, one of the “Stoeckley witnesses” who

was not permitted to testify at trial to Stoeckley’s self-incriminating statements, took the

stand at the 2012 evidentiary hearing and said Britt’s account “rang a bell” for her.

Rouder had been prepared to testify in 1979 that, over the weekend break following

Stoeckley’s trial testimony, Stoeckley admitted her presence during the MacDonald

murders but said she could not confess in court “with those damn prosecutors sitting

there.” In her 2012 evidentiary hearing testimony, Rouder explained that when she heard

Britt’s account, she thought “ah-ha, that’s why [Stoeckley] said she can’t testify with

those damn prosecutors sitting there.” Rouder also testified in 2012 that Stoeckley’s

“damn prosecutors” comment had actually been more expansive than Rouder had

described it in 1979. That is, Rouder’s 2012 testimony included the detail — omitted

from her 1979 description of Stoeckley’s “damn prosecutors” comment — that Stoeckley

had finished the comment by saying the prosecutors would “‘burn me, fry me, hang me’

or words to that effect.”

                                             3.

       In response to MacDonald’s Britt claim evidence, the government acknowledges

proof that Jimmy Britt escorted Helena Stoeckley to the federal courthouse for her

                                            121
interviews with the defense and the prosecution during the 1979 trial. Under that proof,

Britt and Geraldine Holden transported Stoeckley from the Wake County jail (not the

Holiday Inn) to the courthouse on the Thursday morning of the interviews, and Britt

returned Stoeckley to the jail later that day. It was then, the government explains, that the

newspaper photograph of Britt and Stoeckley — which appeared in the newspaper’s

Friday edition — was apparently taken.

       The government has further demonstrated, however, that the Wake County jail and

the federal courthouse were only about six blocks apart, meaning that Britt’s Thursday

morning transport of Stoeckley would have taken mere minutes in which it was unlikely

that Stoeckley had the time or compulsion to make any self-incriminating statements to

Britt about the MacDonald murders. Additionally, the government has produced a bevy

of evidence — including eyewitness testimony and official written records — showing

that neither Britt nor Geraldine Holden had been tasked with picking up Stoeckley in

South Carolina and bringing her to Raleigh the previous day. Rather, Stoeckley was

retrieved that Wednesday from a jail in Pickens County, South Carolina (not Greenville

or Charleston), by Deputy U.S. Marshal Vernoy Kennedy. Kennedy and a female guard

drove Stoeckley to a designated spot at the intersection of Interstates 77 and 85 in

Charlotte, North Carolina, where custody of Stoeckley was transferred to Deputy U.S.

Marshal Dennis Meehan and his wife Janice Meehan.               The Meehans then drove

Stoeckley to the Wake County jail in Raleigh.

       Records of the court proceedings also contradict Britt. For example, contrary to

Britt’s account of waiting with Stoeckley and the defense team in the courtroom for the

                                            122
proceedings to resume immediately after the Thursday interviews, Judge Dupree had

adjourned court for the day in the midst of the interviews. Contrary to Britt’s account of

taking Stoeckley to the Journey’s End following her Friday testimony and moving her to

the Holiday Inn over that weekend, the trial transcripts demonstrate that Stoeckley stayed

in a different hotel on Friday, checked into the Journey’s End on Saturday, and moved to

the Hilton on Sunday with the help of MacDonald lawyer Wendy Rouder and Rouder’s

fellow Stoeckley witness Red Underhill. Contrary to Britt’s account of putting Stoeckley

on a bus back to South Carolina that Monday because the judge had ruled she could not

retake the witness stand, Stoeckley remained in Raleigh for the duration of the trial,

subject to being recalled as a witness and with the representation of court-appointed

counsel Jerry Leonard. In further contradiction of Britt, Judge Dupree never instructed

the jury to disregard Stoeckley’s Friday testimony and never said on the record that

Stoeckley’s “brain was scrambled like an egg.”

      The government argues that the patent inaccuracy of so many details of Britt’s

account renders his entire story untrustworthy and unreliable.         In particular, the

government contends that — because Britt had no long car ride with Stoeckley like the

one in which she allegedly confessed to him her involvement in the MacDonald murders

— it is implausible both that Stoeckley actually made the confession to Britt and that she

then repeated the confession to prosecutor James Blackburn, prompting him to threaten

her with murder charges. It is also unsubstantiated that, following the prosecution’s

interview of Stoeckley, Blackburn had an ex parte meeting with Judge Dupree while



                                           123
Britt, Stoeckley, and the defense team waited in the courtroom for the proceedings to

resume.

       In the government’s view, that Britt reportedly passed a polygraph examination

does not bolster his credibility, because the results of that examination were not

validated. Britt’s account also is not proven by his former wife Mary Britt and his friend

Lee Tart, who simply repeated what Britt told them. Nor is Britt’s account substantiated

by Stoeckley’s confessions to her mother and lawyer, or by her statements to her mother

and MacDonald lawyer Wendy Rouder that she falsely testified at trial because she “was

afraid of the prosecutor” and could not confess “with those damn prosecutors sitting

there.” As the government would have it, Rouder called her own credibility into question

by testifying in 2012 that Stoeckley said the prosecutors would “burn me,” “fry me,” or

“hang me,” in that such a detail was not included in Rouder’s 1979 description of

Stoeckley’s “damn prosecutors” comment — or in the contemporaneous handwritten

notes that Rouder had used to refresh her memory at trial at to exactly what Stoeckley

had said to her.

       In challenging Britt’s allegations of Stoeckley’s confession to prosecutor

Blackburn and Blackburn’s threat in response, the government underscores that

Stoeckley did not tell anyone, including her mother and Wendy Rouder, that she feared

the prosecutors because Blackburn or any other prosecutor had actually threatened her

with murder charges. Moreover, that Stoeckley was cognizant of her potential criminal

liability was evidenced as early as April 1971, when Stoeckley told another of the



                                           124
Stoeckley witnesses, Army investigator Robert Brisentine, Jr., that she would identify the

murderers if granted immunity from prosecution.

       While contesting that Stoeckley ever confessed to Britt and Blackburn, the

government also takes the opportunity to highlight new evidence contradicting

MacDonald defense lawyer Bernard Segal’s representations to the trial court in 1979, in

his quest to treat Stoeckley as a hostile witness, that Stoeckley had made self-

incriminating statements during her interview with the defense team.           Stoeckley’s

purported statements referred to, inter alia, standing over a body in the MacDonalds’

apartment, riding the toy horse in Kristen’s bedroom, and seeing blood on her hands. At

the time Segal represented to the trial court that Stoeckley had made the statements,

fellow MacDonald lawyer Wade Smith advised Judge Dupree that Stoeckley generally

told the defense team, “I don’t remember,” but in several instances said things that gave

an interesting insight into her mind. During the 2012 evidentiary hearing, Smith, who by

then was no longer representing MacDonald, clarified that he did not hear Stoeckley

make the specific statements alleged by Segal or say other “useful things for us.” Smith

allowed that it was possible that Stoeckley made self-incriminating statements while he

was out of the interview room, but he reiterated that “when I was present she did not say

things that helped us.”

                                            D.

       The second claim raised in MacDonald’s latest 28 U.S.C. § 2255 motion is the

DNA claim, i.e., the freestanding actual innocence claim premised on DNA testing that

failed to identify the source of three naturally shed hairs collected from the crime scene.

                                           125
No matches were found for the hairs when they were compared to MacDonald, his

murdered wife Colette, their murdered daughters Kimberly and Kristen, and alleged

perpetrators Helena Stoeckley and Greg Mitchell. According to MacDonald, the three

unsourced hairs constitute powerful evidence of homicidal intruders — albeit not

Stoeckley or Mitchell — that proves his innocence of the murders.

       MacDonald emphasizes the hair designated as “91A,” which he asserts was among

the fingernail scrapings from Kristen’s left hand at the time of her autopsy. Relying on

the observation of the Army pathologist who conducted the autopsy that Kristen may

have had defensive wounds on her hands, MacDonald argues that it is highly likely that

specimen 91A came from Kristen’s murderer and became lodged beneath her fingernail

as she defended herself from the attack. MacDonald insists that, because specimen 91A

was not matched to him, he could not have been the perpetrator. Describing them as

likewise exculpatory, MacDonald also relies on the hairs designated as “58A(1),” found

on Kristen’s bedspread, and “75A,” found on the master bedroom’s area rug within the

outline of Colette’s body.

       The government counters with evidence suggesting that specimen 91A was not, in

fact, part of Kristen’s fingernail scrapings — including evidence that Kristen’s hands

were bloody, but specimen 91A was not. Moreover, the government underscores that

specimens 58A(1) and 75A were similarly blood-free, and that each of the three naturally

shed hairs came from a different source and was not matched to Stoeckley or Mitchell.

Invoking the absence of any match to a known person who could have been one of

MacDonald’s alleged intruders, the government argues that the three hairs should simply

                                          126
be treated as more of the unsourced and immaterial debris that was discovered throughout

the crime scene.

      Additionally, the government points to results of the DNA testing that tend to

incriminate MacDonald, particularly with respect to the hair designated as “51A(2).” At

trial, the defense team had singled out that previously unidentified hair as compelling

evidence of intruders, in that it was found in Colette’s left hand and appeared to be

bloody and broken. The DNA testing determined that specimen 51A(2) came from

MacDonald.

      As MacDonald would now have it, however, specimen 51A(2) is not inculpatory,

because that hair could have gotten into Colette’s hand while he was trying to revive her.

And MacDonald is adamant that the DNA testing’s failure to identify the source of

specimens 58A(1), 75A, and 95A, considered with all his other evidence of intruders

accumulated over the last four decades, entitles him to the vacatur of his convictions and

the award of a new trial.     See 2013 MacDonald Memorandum 121-22 (“The new

evidence amassed since the trial, and augmented by the compelling testimony at the

[2012] evidentiary hearing, is remarkable. . . . [A] jury hearing the newly discovered

evidence [of Helena Stoeckley’s post-trial confessions and prosecutor James Blackburn’s

threat of murder charges], as well as the DNA evidence of the unsourced hairs, would

undoubtedly acquit MacDonald based on this evidence alone.”).




                                           127
                                           III.

      Having considered the entire record in these proceedings, the district court

determined in its 2014 decision that neither the Britt claim nor the DNA claim passes the

procedural bar of 28 U.S.C. § 2255(h)(1). See United States v. MacDonald, 
32 F. Supp. 3d
608, 687-700 (E.D.N.C. 2014). That is, the court ruled that MacDonald has failed to

show that the Britt and DNA claims are based on “newly discovered evidence that, if

proven and viewed in light of the evidence as a whole, would be sufficient to establish by

clear and convincing evidence that no reasonable factfinder would have found

[MacDonald] guilty of [the murders of his wife and daughters].”           See 28 U.S.C.

§ 2255(h)(1). Alternatively, reaching the merits of the Britt and DNA claims, the court

concluded that MacDonald has failed to prove the constitutional violations alleged in

those claims. See MacDonald, 
32 F. Supp. 3d
at 700-03, 705-09. We review the court’s

procedural and merits rulings in turn, utilizing and applying our own in-depth

examination of the record.

                                           A.

      As recognized in our 2011 decision, we review de novo the district court’s

conclusion that the Britt and DNA claims fail to satisfy § 2255(h)(1)’s requirements. See

United States v. MacDonald, 
641 F.3d 596
, 609 (4th Cir. 2011) (citing Reyes-Requena v.

United States, 
243 F.3d 893
, 900 (5th Cir. 2001) (“A district court’s denial of a second

§ 2255 motion on the ground that the motion fails to meet AEDPA’s conditions is a legal

conclusion, which we review under a de novo standard of review.”)); see also O’Dell v.

Netherland, 
95 F.3d 1214
, 1250 (4th Cir. 1996) (en banc) (explaining that the analysis set

                                           128
forth in Schlup v. Delo, 
513 U.S. 298
(1995), from which § 2255(h)(1) derived, involves

“‘mixed’ questions of law and fact” that “are reviewable de novo”). Nevertheless, to the

extent that the district court made credibility determinations that are pertinent to our

§ 2255(h)(1) analysis and concern witnesses heard by the court during its 2012

evidentiary hearing, we must accept those findings unless clearly erroneous. See 
O’Dell, 95 F.3d at 1250
. In sum, our job is to independently examine the cold record of the trial

and the past and present postconviction proceedings; to assess the import and reliability

of all the evidence, according deference only to the district court’s relevant witness

credibility determinations; and to reach our own judgment as to whether there is clear and

convincing evidence that no reasonable factfinder would have convicted MacDonald of

the murders. See 
id. Even so,
we find ourselves in agreement with much of the district court’s 2014

decision.   In particular, the court jointly assessed the Britt and DNA claims under

§ 2255(h)(1) — a sensible approach considering that the evidence underlying the DNA

claim is part of the “evidence as a whole” with respect to the Britt claim, and vice versa.

See MacDonald, 
32 F. Supp. 3d
at 688-89.           Moreover, although the court made

credibility determinations as to the witnesses at the 2012 evidentiary hearing, the court

also recognized that such determinations were not necessary to its § 2255(h)(1) analysis.

Specifically, the court ruled that — even accepting that Helena Stoeckley made the

various self-incriminating statements attributed to her over the decades, including the

alleged confessions to Deputy U.S. Marshal Jimmy Britt, her court-appointed counsel

Jerry Leonard, and her mother, the elder Helena Stoeckley — those statements, together

                                           129
with the DNA evidence of the three unsourced hairs and all the other ostensibly

exculpatory evidence, are insufficient to satisfy the § 2255(h)(1) standard. See 
id. at 695-
98. That is so, as the court explained, because the younger Helena Stoeckley herself was

simply never reliable or credible. 
Id. at 696
(“The court does not see how the fact that

Stoeckley also allegedly confessed to a marshal, her attorney, and her mother . . . changes

the utter unreliability of Stoeckley herself.” (footnotes omitted)). Like the district court,

we jointly assess the Britt and DNA claims under § 2255(h)(1), and we accept that

Stoeckley made the statements attributed to her, rendering it unnecessary to our analysis

to defer to the court’s credibility determinations.

                                              1.

       We begin our § 2255(h)(1) analysis by focusing on MacDonald’s 1979 trial. To

briefly recap, the government’s theory was that, during a physical altercation between

MacDonald and his pregnant wife Colette in the master bedroom of their Fort Bragg

apartment in the early morning hours of February 17, 1970, MacDonald seriously

wounded Colette and five-year-old daughter Kimberly by striking each of them with a

piece of lumber. Things having gone too far, MacDonald resolved to kill his family and

stage the crime scene to mimic a cult slaying in the manner of the then-recent Charles

Manson murders. MacDonald moved Kimberly to the bed in her bedroom, struck her

again with the now-bloodstained piece of lumber, and repeatedly stabbed her with an Old

Hickory knife. He also went to the bedroom of two-year-old daughter Kristen, where

Kristen had been asleep in her bed, and stabbed her multiple times with the Old Hickory

knife and an ice pick.       The seriously-wounded Colette confronted MacDonald in

                                             130
Kristen’s bedroom, where MacDonald struck Colette again with the bloodstained piece of

lumber. MacDonald then bundled Colette in the top sheet and bedspread from the master

bed, carried her from Kristen’s bedroom to the master bedroom, laid her on the master

bedroom floor, and stabbed her with the Old Hickory knife. He also covered Colette’s

chest with the bloodstained and torn blue pajama top he had been wearing and stabbed

her through it twenty-one times with the ice pick, creating forty-eight puncture holes in

the blue pajama top. To strengthen the cover-up, MacDonald added to minor injuries

possibly inflicted on him by Colette (perhaps with an Old Hickory knife left on the

master bedroom floor), by lacerating his own chest in the family’s shared bathroom

(perhaps with a disposable scalpel blade from the nearby linen closet). He also staged the

living room to make it appear that he had been attacked there by intruders; wrote the

word “PIG” in Colette’s blood on the master bed’s headboard while wearing a latex

glove, which he may have taken from a box of gloves beneath the kitchen sink; and

discarded the bloodstained piece of lumber, Old Hickory knife, and ice pick by throwing

them from the apartment’s rear exterior doorway into the backyard.

      In MacDonald’s account, by contrast, he had been sleeping on the couch in the

living room and was awakened sometime after 2:00 a.m. by the sound of Colette’s and

Kimberly’s screams and cries for help from the bedroom area. MacDonald was then

attacked in the living room by three male intruders — a black man wearing an Army field

jacket with Sergeant stripes, plus two white men —accompanied by a blond-haired white

woman in a floppy hat and boots who appeared to be carrying a lighted candle and saying

words that included, “Acid is groovy; kill the pigs.”         The black man assaulted

                                           131
MacDonald with a baseball-bat-like club that could not have been the bloodstained piece

of lumber used against Colette and Kimberly, and the white men punched and stabbed at

MacDonald with a knife and ice pick. During the brief encounter in the living room,

MacDonald’s blue pajama top was ripped from around his back or pulled over his head,

and then the pajama top became wrapped around his wrists and hands.            Once that

happened, MacDonald used the blue pajama top to fend off the white men’s punches and

stabs, resulting in the pajama top’s forty-eight puncture holes. The intruders knocked

MacDonald unconscious in the living room, and he awoke sometime later lying facedown

over the step between the living room and hallway, with his head and chest on the

hallway floor. MacDonald went straight to the master bedroom, where he pulled the

Geneva Forge knife from Colette’s chest, attempted to revive her, and covered her with

the blue pajama top. He next went to each of his daughters’ bedrooms, and then, in this

or some other order, made a second rotation between the three bedrooms, checked on his

own wounds in the bathroom, and called for help from the master bedroom and the

kitchen. At some point, he noticed that the apartment’s rear exterior wood door was

open, went to the doorway, and looked outside, but he stayed inside the apartment.

      Although the government certainly had evidence to corroborate its theory of the

murders — including the murder weapons, blood evidence enhanced by the family

members’ four different blood types, threads and yarns from the torn blue pajama top, an

expert reconstruction of how the twenty-one thrusts of the ice pick into Colette’s chest

could have made the blue pajama top’s forty-eight puncture holes, and a separate expert

opinion that the “clean” puncture holes were necessarily made while the blue pajama top

                                          132
was stationary — the trial did not turn on whether the jury believed every aspect of the

government’s case. Rather, the trial turned on whether the jury disbelieved MacDonald’s

account. In the words of presiding Judge Franklin Dupree in the midst of the seven-week

trial: “I think this case is going to rise or fall on one thing and one thing alone and that is

whether or not the jury buys the Defendant’s story as to what happened.” Detrimentally

to MacDonald, the government was able to satisfy its burden of proof beyond a

reasonable doubt by showing that MacDonald’s account was unworthy of belief.

                                              a.

       Of course, the jury must have been troubled by the fundamental question posed by

the government in its closing arguments: Why would a band of intruders massacre a

pregnant woman and two little girls but leave the strongest person in the apartment — the

one most able to fight back — alive, relatively unscathed, and able to serve as an

eyewitness? The government relied on much more than the comparative mildness of

MacDonald’s injuries, however, to disprove his story.

       Prominently, the government pointed to inconsistencies between the blood

evidence and MacDonald’s account. If MacDonald was clubbed and stabbed by intruders

in the living room, how could it be that none of his type B blood was found within that

room? And why was there just one type B bloodstain on the blue pajama top that he had

been wearing and then supposedly used as a shield? If MacDonald was left facedown

and unconscious over the step between the living room and hallway, why was there no

more than a single possible type B bloodstain on that area of the hallway floor?

Moreover, if MacDonald first went to the three bedrooms after regaining consciousness

                                             133
and only later went to the bathroom and kitchen, why were there numerous type B

bloodstains in the bathroom, along with type B blood on the nearby linen closet doors and

the kitchen floor? If MacDonald pulled the Geneva Forge knife from Colette’s chest in

the master bedroom, why was there only a scant amount of blood on the tip of that knife’s

blade? And if the blue pajama top only then came into contact with a bloody Colette,

how did the pajama top become stained with Colette’s type A blood before being torn?

How did its pocket become detached from the blue pajama top before the corresponding

area of the pajama top was soaked in type A blood? How did Kimberly’s type AB blood

end up on the blue pajama top at all? If not from stepping in Colette’s type A blood on

the master bed’s top sheet and bedspread while bundling Colette to carry her to the

master bedroom, how did MacDonald make footprints in type A blood while exiting

Kristen’s bedroom, when all the other blood on Kristen’s bedroom floor was her type O?

And how were impressions of the blue pajama top’s right cuff made in type A blood on

the master bed’s top sheet?

      In addition to the blood evidence, the blue pajama top’s threads and yarns were

central to the government’s trial presentation against MacDonald. How could it be that

investigators found no blue pajama top threads or yarns in the living room and just a

cluster of fibers or threads possibly matching the blue pajama top on the floor at the

living room end of the hallway, but collected seventy-eight blue pajama top threads and

yarns from the master bedroom, nineteen from Kimberly’s bedroom, and two from

Kristen’s bedroom? How did blue pajama top threads and yarns end up interspersed with

splinters from the bloodstained piece of lumber? Beneath Colette’s body? On the floor

                                          134
between the master bedroom wall and the master bed’s headboard, in the area where

“PIG” was written on the headboard in blood? On the bloodstained bedspread from the

master bed? On and beneath Kimberly’s and Kristen’s bed covers?

      To be sure, MacDonald offered some answers to those questions, beginning with

the broad proposition that the first responders’ and investigators’ “bungling” and

“ineptitude” discredited all of the government’s physical evidence. MacDonald more

specifically suggested that the CID may have mistyped the blood evidence. He also

posited that the first responders and investigators inadvertently redistributed the blue

pajama top threads and yarns. Or maybe he transferred the blue pajama top threads and

yarns to the intruders as they attacked him in the living room, and then the intruders

dispersed the threads and yarns while committing the murders. And perhaps threads and

yarns also became stuck to his arms and fell in places like Kimberly’s and Kristen’s beds

while he checked on them. Maybe he placed Colette’s body atop blue pajama top threads

and yarns when he moved her while trying to revive her. Perhaps many or all of the

threads and yarns actually came from his matching blue pajama pants.

      Notwithstanding that the first responders and investigators clearly made mistakes

and conducted an imperfect investigation, the jury rejected MacDonald’s broad

ineptitude-based challenge to the government’s physical evidence. Furthermore, the jury

could have easily found that MacDonald not only failed to convincingly answer the blood

and thread-and-yarn evidence, but also raised new questions about his own account.

Where was the proof, and not just speculation, that the CID had mistyped any blood?

Was it feasible that there had been blue pajama top threads and yarns in the living room

                                          135
that were all relocated to the bedrooms, or that any significant number of threads and

yarns were shifted around the apartment without one being moved to the living room, or

that threads and yarns were randomly redistributed to inculpatory locations? How could

threads and yarns have transferred to the intruders and become stuck to MacDonald’s

arms during the battle in the living room, without any being left behind there?

Considering his initial statement to investigators that he had barely moved Colette, if at

all, was it believable that MacDonald could have placed Colette atop threads and yarns

when he moved her while trying to revive her? And if the threads and yarns actually

originated from MacDonald’s blue pajama pants, why were there no threads and yarns in

the living room, bathroom, and kitchen where MacDonald said he wore the pajama pants

during and after the intruders’ attack?

       Whatever the jury’s view of the foregoing blood and thread-and-yarn evidence, it

seems likely that the guilty verdicts were clinched with other unanswered evidence

inconsistent with MacDonald’s account — namely, the two additional blue pajama top

threads on the bloodstained piece of lumber found on the ground outside the apartment,

and the forty-eight puncture holes apparently made by the ice pick in the blue pajama top.

If, as MacDonald testified at trial, the black male intruder assaulted him with a baseball-

bat-like club that could not have been the bloodstained piece of lumber, and if

MacDonald did not go outside the apartment in the aftermath of the attack, how did the

blue pajama top threads get on the piece of lumber? And if the puncture holes were made

in the blue pajama top while MacDonald used it, wrapped around his wrists and hands, to

shield his chest from an onslaught of stabs and punches from the two white male

                                           136
intruders, how could it be that each of the puncture holes was symmetrical and clean,

with no tearing? Indeed, how did MacDonald avoid sustaining any ice pick wounds to

his hands or arms?

      MacDonald acknowledged in his trial testimony that he had no answer for the blue

pajama top threads on the bloodstained piece of lumber. As for the blue pajama top’s

forty-eight puncture holes, MacDonald lodged a compelling challenge to the validity of

the government’s expert reconstruction of how the twenty-one thrusts of the ice pick into

Colette’s chest could have made the puncture holes. Nonetheless, the government also

had its expert’s opinion that, because of the cleanness of the puncture holes, they were

necessarily made while the blue pajama top was stationary.        MacDonald sought to

counter that opinion with his own expert’s opinion that clean puncture holes were

possible with the pajama top in motion. But the circumstances of the defense expert’s

testing were far different from MacDonald’s alleged use of the pajama top as a shield, as

the defense expert stabbed an ice pick through a single layer of new fabric that was

stretched tightly over a plastic-wrapped ham, moving only laterally, and admittedly less

susceptible to tearing than the well-laundered blue pajama top. Moreover, the defense

expert conceded that he could not simulate MacDonald’s account. The jury was left with

no evidence, and surely substantial doubt, that the forty-eight puncture holes could have

been made while MacDonald was using the blue pajama top to fend off stabs and

punches from any ice-pick-wielding assailant and his cohort.




                                          137
                                           b.

      In the absence of convincing answers to the critical questions raised by the

government’s physical evidence, MacDonald’s account depended at trial on compelling

proof of the alleged intruders. Instead, MacDonald merely had, inter alia, a chemist’s

opinion that pieces of latex found in the master bedroom and thought to have been part of

the glove used to write “PIG,” were inconsistent with the latex gloves kept beneath the

kitchen sink. MacDonald argued that the defense chemist’s opinion evidenced intruders

who brought their own latex gloves to the apartment. MacDonald also asserted that three

samples of dried wax from the living room and Kimberly’s bedroom were evocative of

the candle-carrying female intruder, because the wax samples did not match any of the

fourteen candles in the apartment. As further proof of intruders, MacDonald pointed to

unsourced hairs; unsourced latent palm and fingerprints, including one on a jewelry box

from which two of Colette’s rings may have been stolen; and an unsourced pink fiber

caught in the hinge of his eyeglasses, which had been found after the murders on the

living room floor.

      The government responded by pointing out the weakness of MacDonald’s

evidence as proof of intruders. That is, MacDonald himself could have used latex gloves

other than those beneath the kitchen sink. The wax samples, which did not even match

each other, were probably just more of the apartment’s “living day-to-day mess”

encountered by the investigators.    The unsourced latent palm and fingerprints were

probably left by some of the many innocent visitors that MacDonald told investigators

had been in the apartment, or may have belonged to Colette, Kimberly, and Kristen, as

                                          138
the investigators had failed to obtain all of their prints for comparison. Colette’s rings

could have gone missing sometime after the murders, as no one looked for those

particular items or noted their absence in the murders’ immediate aftermath. And the

pink fiber was immaterial because MacDonald had disclaimed wearing his eyeglasses

during the intruders’ attack.

       Notably, MacDonald also relied at trial on purported eyewitnesses to prove his

account of intruders. The first of those eyewitnesses was a neighbor who said he saw a

hatless white female and two white males — all carrying lighted candles and wearing

white sheets or robes — walking toward the MacDonalds’ apartment between midnight

and 12:30 a.m. The second was a military police officer who, while responding to

MacDonald’s call for help shortly before 4:00 a.m., saw a female in a dark-colored rain

hat standing near a street corner about one-half mile from the MacDonald’s apartment.

As the government emphasized, however, MacDonald did not describe his intruders as

wearing white sheets or robes, only his blond-haired female intruder was carrying a

candle, and she was wearing a white floppy hat, not a dark-colored hat or no hat at all.

                                             c.

       Finally, MacDonald also had Helena Stoeckley to accuse of being the blond-haired

female intruder. It is no wonder that the accusation failed to persuade the jury, as it was

revealed that MacDonald had made a police sketch of the female intruder during his 1970

Army proceedings that did not look like Stoeckley, and that he had denied recognizing

Stoeckley in a photograph presented to him at those proceedings. After learning that

Stoeckley was a suspect and seeing multiple photographs of her, however, MacDonald

                                            139
created a 1979 police sketch of the female intruder that looked remarkably like Stoeckley,

along with a new sketch of one of the white male intruders that looked like a known

Stoeckley associate. Meanwhile, the investigators had compared the latent palm and

fingerprints recorded in the MacDonalds’ apartment with Stoeckley’s prints, but found no

matches.

      Otherwise, the jury heard that Stoeckley was someone who could have been

MacDonald’s blond-haired female intruder. Like a lot of local “hippies” at the time,

Stoeckley regularly wore a floppy hat and boots. She sometimes wore a blond wig. She

was a heavy drug user. She had many male associates, including at least one black man

who, like numerous men in the Fort Bragg area, wore an Army fatigue jacket with

Sergeant stripes. She admittedly used words like “pig” and “acid.” She had an interest in

witchcraft. She hung funeral wreaths around the time of the MacDonald funerals. And,

as Stoeckley conceded on the witness stand, she could not account for her whereabouts

between about 12:30 and 4:30 a.m. on February 17, 1970.

      Even if the jury had fully heard from the seven “Stoeckley witnesses” — the

witnesses that the trial court would not allow to relay self-incriminating statements that

Stoeckley had made to them — the jury would have merely learned that Stoeckley had

uttered many conflicting statements that contained few details of the murders, and then

only publicized details, such as the toy horse in Kristen’s bedroom (which had appeared

in a newspaper photograph) and the word “PIG” written in blood on the master bed’s

headboard (which had been included in news reports). The jury also would have heard

that Stoeckley said things like, “I remember standing at the couch, holding a candle, only

                                           140
— you know — it wasn’t dripping wax. It was dripping blood.” Indeed, the testimony of

the Stoeckley witnesses would have comported with other evidence before the jury that

Stoeckley had demonstrated no “insider’s” knowledge of the crimes. For example, there

was evidence that Stoeckley discarded her blond wig and a pair of thigh-high white boots

she thought might implicate her in the murders, but that MacDonald meanwhile insisted

to investigators that his female intruder was wearing knee-high brown boots, criticizing

the inaccuracy of then-recent media reports that the boots were white.          In these

circumstances, there was ample reason to disbelieve that Stoeckley actually was involved

in the murders.

                                           2.

      We turn in our § 2255(h)(1) analysis to the past and present postconviction

proceedings, wherein it has been incumbent upon MacDonald — now bearing the burden

of proof — to overcome procedural hurdles and prove his claims. At this point, to move

forward with his Britt and DNA claims, MacDonald must demonstrate his actual

innocence by clear and convincing evidence. The surest way to do that would be to

refute the physical evidence tending to disprove his account and to at last make a

compelling showing of the alleged murderous intruders.

                                           a.

      MacDonald instead focuses his efforts on advancing a theory of the murders that is

premised on selected details of post-trial confessions made by Helena Stoeckley —

details that vary from his own story and the evidence in several significant ways. In the

Stoeckley-statement-based account of the murders, members of Stoeckley’s group

                                          141
confronted Colette just hours before the murders while she was at the Fort Bragg campus

of North Carolina State University to take a class.       The group members wanted

MacDonald to provide them drugs and to agree not to report drug users, and they urged

Colette to intervene. The plan thereafter changed to confronting MacDonald directly, and

so the group, now consisting of Stoeckley, three white men (Greg Mitchell, Bruce

Fowler, and Don Harris), and a black man in an Army jacket with Sergeant stripes

(Dwight Edwin Smith), set off from a local Dunkin’ Donuts at about 2:00 a.m. for the

MacDonalds’ apartment.      Having entered the apartment through its rear exterior

doorway, the group passed through the master bedroom and proceeded straight to the

living room, where MacDonald was asleep on the couch with a Valentine’s Day card next

to him. Stoeckley lit a candle to add light to the dimly illuminated living room, and

group members jostled MacDonald to awaken him. When MacDonald awoke, he became

excited and began to fight the group members.        Smith struck at MacDonald, and

Stoeckley said something like, “Hit the pig,” or “Kill the pig,” and “Acid is groovy.”

Around that time, Stoeckley heard Colette calling for help from another room. When

Stoeckley went to investigate, she saw Colette being attacked by Mitchell and a second

man, as a child in the same room slept. While the group was in the apartment, Stoeckley

saw the toy horse, with an unattached or broken spring, in one of the children’s

bedrooms.    She also answered a ringing telephone.        The caller asked for “Dr.

MacDonald,” Stoeckley began to laugh, and then another group member ordered

Stoeckley to hang up the telephone. Having become frightened, Stoeckley and her

cohorts hurriedly departed the apartment, with some items stolen from a jewelry box and

                                          142
one of their weapons, a pair of scissors. The group members left all their other weapons

behind. They returned to the Dunkin’ Donuts, where Stoeckley washed blood from her

hands in the restroom sink. Stoeckley soon got rid of everything she had been wearing,

giving her bloodstained clothing and boots to her friend Cathy Perry with a request to

dispose of them, and discarding her blond wig and floppy hat herself.

      Obviously, the Stoeckley-statement-based account of the intruders going straight

to the living room and jostling MacDonald to awaken him, conflicts with MacDonald’s

own account of being roused by screams and cries for help from Colette and Kimberly as

they were apparently being attacked in the bedroom area. Moreover, there was no

evidence of a Valentine’s Day card on the couch next to MacDonald or anywhere else in

the living room. According to other evidence, the toy horse was unbroken, with all of its

springs attached. And the proposition that the intruders took all of their weapons with

them except a pair of scissors, conspicuously omits the baseball-bat-like club that

MacDonald claimed was used to attack him but that was not found at the crime scene. It

is also striking that MacDonald never reported, as Colette surely would have told him

when she returned from class that night, that she had been confronted at the Fort Bragg

campus of North Carolina State by a group of drug users urging her to intervene with him

on their behalf.   Yet those are details from Stoeckley’s post-trial confessions that

MacDonald has chosen to rely on. 13


      13
       For troublesome details from Stoeckley’s self-incriminating statements that
MacDonald now disregards, see supra notes 8 & 9.


                                          143
      Another dubious aspect of Stoeckley’s post-trial statements is that they initially

implicated Allen Mazerolle — the same man depicted in the new police sketch of one of

the white male intruders that MacDonald produced in 1979 for his trial. The defense had

gone so far as to name Mazerolle, in closing arguments, as a likely perpetrator. Evidence

later emerged that Mazerolle was in jail at the time of the murders, causing MacDonald to

abandon his Mazerolle accusation and Stoeckley to change her story from, in our words,

“Mazerolle was involved,” to “I falsely implicated Mazerolle, knowing he had been in

jail, as a way to cast doubt on my own confessions.” Since then, MacDonald has pointed

the finger only at the latest four men named by Stoeckley as the perpetrators, including

Colette’s alleged attacker Greg Mitchell. He has also suggested that Stoeckley’s friend

Cathy Perry may have participated in the murders, even though Stoeckley said Perry

played only a post-murder role, helping Stoeckley dispose of her bloodstained clothing

and boots.

                                           b.

      Despite its inconsistencies with his own story, MacDonald insists that the

Stoeckley-statement-based account of the murder is credible, citing corroboration from

several sources. First and foremost, he relies on the confessions that Helena Stoeckley

allegedly made during the trial to Deputy U.S. Marshal Jimmy Britt and to her court-

appointed counsel Jerry Leonard, and in anticipation of death to her mother, the elder

Helena Stoeckley.     Those confessions, however, are similar in their limited and

questionable substance to many of the younger Stoeckley’s various other self-

incriminating statements. Indeed, those confessions contained even fewer details than the

                                          144
confessions from which MacDonald has drawn the Stoeckley-statement-based account of

the murders. And, in at least one respect, the younger Stoeckley’s confession to her

mother contradicts the Stoeckley-statement-based account.        That is, the younger

Stoeckley told her mother that she fled the MacDonalds’ apartment with just one male

cohort when the pair realized the other group members were murdering MacDonald’s

family. But in the Stoeckley-statement-based account, all of the group members left the

apartment together, having become frightened after Stoeckley answered the ringing

telephone. The only detail consistently included in the younger Stoeckley’s confessions

to Britt, Leonard, and the elder Stoeckley was the oft-repeated claim that the younger

Stoeckley saw the toy horse in the MacDonalds’ apartment.          Consequently, as the

government suggests, those confessions constitute — at best — thinly-detailed and

uncorroborated “I was there” admissions.        Still, MacDonald argues that the three

confessions possess substantial reliability and trustworthiness because of the

circumstances in which they were made.

      Additionally, MacDonald invokes self-incriminating statements allegedly made by

Greg Mitchell and Cathy Perry.      Unhelpfully to MacDonald, no physical evidence,

including latent palm and fingerprints, linked Mitchell to the crime scene, and the only

details contained in Mitchell’s confessions were varying statements of motive.

Furthermore, the younger Stoeckley not only failed to implicate Perry in the murders, but

Perry’s confession was wholly incompatible with the known facts and evidence.

Nevertheless, MacDonald contends that Mitchell’s confessions were direct and

unambiguous, and that we should believe Perry may have helped kill his family because

                                          145
Perry’s personal history of stabbings and schizophrenia rendered her the type of person

capable of committing murder.

       MacDonald also claims corroboration of the Stoeckley-statement-based account

from a witness who allegedly saw the confrontation between Colette and Stoeckley’s

group at the Fort Bragg campus of North Carolina State, and a witness who allegedly

made the telephone call to the MacDonalds’ apartment that Stoeckley answered.

Additional witnesses said they may have seen persons around the area who looked like

Stoeckley, men she accused of the murders, and some of MacDonald’s various police

sketches of the alleged intruders.   The government has countered with compelling

evidence aimed at disproving the campus confrontation and the telephone call, as well as

showing that Stoeckley’s first claim of answering the ringing telephone coincided with

publication by the media of the purported caller’s story.     Addressing MacDonald’s

additional witnesses, the government asserts that those witnesses did no more than place

Stoeckley and her associates in the general Fort Bragg area around the time of the

murders. Tellingly, MacDonald ignores the questions raised by the government about his

witnesses’ credibility and import.

                                           c.

       MacDonald’s final effort to corroborate the Stoeckley-statement-based account of

the murders is dependent upon new physical evidence — not only the DNA evidence of

the three unsourced hairs, but also three blond synthetic hairs removed from a hairbrush

in the MacDonalds’ dining room; five unsourced black fibers removed from the

bloodstained piece of lumber, Colette’s pink pajama top, and the area around Colette’s

                                          146
mouth; an expert’s opinion that some of Colette’s and Kristen’s stab wounds were

consistent with scissors; and the same expert’s opinion that a blow that fractured

Colette’s skull was consistent with a frontal clubbing by a left-handed assailant, along

with evidence that Greg Mitchell was left-handed and MacDonald is right-handed.

MacDonald argues that the new evidence — together with the physical evidence of

intruders highlighted during his trial, and potentially some “lost” evidence as well —

strongly implicates Helena Stoeckley and her cohorts.

       Beginning with the three unsourced hairs, MacDonald emphasizes specimen 91A,

a hair that he asserts was among Kristen’s fingernail scrapings at the time of her autopsy.

Because Kristen may have also had defensive wounds on her hands, MacDonald deems it

to be highly likely that specimen 91A came from Kristen’s murderer, who thus could not

have been him. MacDonald further argues that specimens 58A(1) and 75A, hairs found

on Kristen’s bedspread and on the master bedroom’s area rug within the outline of

Colette’s body, are likewise exculpatory. The government counters with evidence that all

three of those unsourced hairs were naturally-shed and blood-free, suggesting that

specimen 91A was not actually found beneath a fingernail of Kristen’s bloody hands, and

that each of the hairs was simple household debris.        That is particularly true, the

government argues, because each of the hairs came from a different source and was not

matched to Stoeckley or Mitchell. On the other hand, the government asserts that there

was a hair whose DNA findings were inculpatory: specimen 51A(2), a broken, bloody,

and previously unidentified hair that had been found in Colette’s hand and singled out by

the defense at trial as compelling evidence of intruders. That hair was matched through

                                           147
DNA testing to MacDonald, who counters that specimen 51A(2) is not inculpatory

because it could have gotten into Colette’s hand while he was trying to revive her.

       MacDonald next argues that the three blond synthetic hairs and five unsourced

black wool fibers are corroborative of Stoeckley’s presence during the murders, as

Stoeckley was allegedly wearing her blond wig and routinely wore black clothing.

MacDonald also relies on his expert’s opinions to substantiate the Stoeckley-statement-

based account of the intruders’ use of scissors, and to implicate Mitchell and exonerate

himself in the clubbing attack on Colette. As the government points out, however, it has

never been established whether the three blond synthetic hairs came from a doll (as the

government deems likely) or from a wig (as MacDonald says is possible). The black

wool fibers were likely just more household debris, as no two of the fibers matched each

other. And the expert qualified his opinions in ways unhelpful to MacDonald, explaining

that Colette’s and Kristen’s stab wounds were also consistent with an ice pick, and that an

enraged right-handed assailant could deliver a left-handed-type blow.

       Undeterred, MacDonald insists that the DNA and other new physical evidence

augments his trial evidence (e.g., the three samples of dried wax and the many

identifiable but unsourced latent palm and fingerprints), as well as potentially-

exculpatory lost evidence (including a “bloody syringe” supposedly found in the linen

closet).   All of the evidence, according to MacDonald, substantiates the Stoeckley-

statement based account and proves his innocence by clear and convincing evidence,

satisfying the requirements of § 2255(h)(1). By contrast, the government views the new

evidence as just as weak and speculative as MacDonald’s trial evidence and any lost

                                           148
evidence. In all these decades, the government maintains, MacDonald has failed to make

a convincing showing of murderous intruders or to refute the powerful evidence that

belies his account, thereby preventing him from now overcoming § 2255(h)(1)’s

procedural bar.

                                            3.

      Our comprehensive review of the trial and postconviction evidence convinces us

that MacDonald has not met the rigorous requirements of § 2255(h)(1). As we cautioned

in our 2011 decision, § 2255(h)(1) was “designed to ensure that [it] could be satisfied

only in the rare and extraordinary case.” See 
MacDonald, 641 F.3d at 614
-15 (internal

quotation marks omitted). Though we have given MacDonald the opportunity to do so,

he has not demonstrated that his is one of the rare and extraordinary cases justifying

pursuit of a claim premised on newly discovered evidence by way of a successive § 2255

motion. Simply put, we cannot say that the new evidence underlying MacDonald’s Britt

and DNA claims, considered with all the other evidence, would be sufficient to establish

by clear and convincing evidence that no reasonable factfinder would have found him

guilty of the murders of his wife and daughters.

      In so ruling, several points stand out to us. First, MacDonald has made little

effort, and definitely not succeeded, in refuting the cogent evidence proffered by the

government to disprove his account of murderous intruders. Notable examples of that

evidence, as heretofore discussed, include MacDonald’s bloody footprints in daughter

Kristen’s bedroom, the forty-eight “clean” puncture holes in MacDonald’s blue pajama



                                           149
top, and the two blue pajama top threads on the bloodstained piece of lumber used in the

clubbing attack on wife Colette and daughter Kimberly.

       Moreover, MacDonald has failed to otherwise make a sufficiently compelling

showing of the alleged intruders, instead advancing a theory of the murders based on

details of alleged intruder Helena Stoeckley’s post-trial statements that materially conflict

with the evidence and his own story. Nothing MacDonald offers for corroboration — not

other confessions of Stoeckley and her alleged cohorts, or purported eyewitnesses, or old

or new physical evidence — comes close to clearly and convincingly substantiating the

Stoeckley-statement-based account.

       And finally, although MacDonald invokes past comments of the trial court and

this Court to argue that Stoeckley’s post-trial “I was there” confessions are enough to

prove his innocence, we are constrained to disagree. Specifically, MacDonald points to a

September 1979 letter from presiding Judge Franklin Dupree to MacDonald defense

lawyer Wendy Rouder that was produced at MacDonald’s 2012 evidentiary hearing. The

letter opened by acknowledging that Rouder had expressed interest during the trial in

applying for a clerkship, and that Judge Dupree had encouraged Rouder’s application.

Addressing a change of circumstances, however, the letter continued: “At that time I

confidently expected that the jury would return a not guilty verdict in the case, but since

the case will now be appealed and you will doubtless keep working on it, it will not be

possible for me to appoint you to a clerkship position.”

       Thereafter, in the first of this Court’s decisions on direct appeal, our Judge

Murnaghan wrote for the panel majority as part of the later-reversed ruling for

                                            150
MacDonald on his Sixth Amendment speedy trial claim: “Had Stoeckley testified as it

was reasonable to expect she might have testified” — if Stoeckley had told the jury, “I

was there” — “the injury to the government’s case would have been incalculably great.”

See United States v. MacDonald, 
632 F.2d 258
, 264 (4th Cir. 1980). By this Court’s

second decision on direct appeal, the panel ruled that the trial court did not commit an

abuse of discretion by excluding the testimony of the seven “Stoeckley witnesses” as to

the self-incriminating statements that Stoeckley had made to them.         In a separate

concurring opinion, Judge Murnaghan noted that “the case provokes a strong uneasiness

in me,” and that “I believe MacDonald would have had a fairer trial if the Stoeckley

related testimony had been admitted.” See United States v. MacDonald, 
688 F.2d 224
,

236 (4th Cir. 1982) (Murnaghan, J., concurring).

      MacDonald argues that the comments of Judge Dupree and Judge Murnaghan

demonstrate that the government’s trial evidence was weak and that he thus would have

been acquitted if Stoeckley had simply told the jury, “I was there.” By extension,

according to MacDonald, Stoeckley’s post-trial “I was there” confessions are sufficient to

prove his innocence for purposes of the § 2255(h)(1) inquiry. We disagree because —

whatever Judge Dupree and Judge Murnaghan meant by their comments (and we are far

from certain that MacDonald’s interpretations are correct) — we now know much about

Stoeckley’s untrustworthiness and lack of credibility that Judge Dupree and Judge

Murnaghan did not know then.




                                           151
                                            B.

       Although we conclude that MacDonald has not overcome the procedural bar of 28

U.S.C. § 2255(h)(1), we find it prudent to reach and assess the district court’s alternative

rulings in its 2014 decision on the respective merits of the Britt claim and the DNA

claim. In evaluating the court’s merits rulings, “we review its legal conclusions de novo

and its findings of fact for clear error.” See United States v. Roane, 
378 F.3d 382
, 395

(4th Cir. 2004).

                                             1.

       We start with the Britt claim, which MacDonald describes in this appeal as being

premised on allegations that prosecutor James Blackburn contravened his constitutional

rights in three separate ways:    (1) by concealing exculpatory evidence, i.e., Helena

Stoeckley’s alleged confession to Blackburn in the presence of Deputy U.S. Marshal

Jimmy Britt, see Brady v. Maryland, 
373 U.S. 83
(1963); (2) by threatening to charge

Stoeckley with murder and thereby causing her to change her trial testimony, see United

States v. Golding, 
168 F.3d 700
(4th Cir. 1999); and (3) by misleading the trial court

about what Stoeckley had said and failing to correct Stoeckley’s false testimony, see

Napue v. Illinois, 
360 U.S. 264
(1959); Alcorta v. Texas, 
355 U.S. 28
(1957). The district

court ruled that MacDonald has failed to prove any constitutional violation, in that he has

not demonstrated by a preponderance of the evidence that Stoeckley confessed to

Blackburn or that Blackburn threatened her in response.

       More specifically, the district court found the evidence crucial to the Britt claim —

the sworn statements of Jimmy Britt — “to be unreliable and incredible.”                See

                                            152
MacDonald, 
32 F. Supp. 3d
at 702. The court also deemed Britt’s “assertions as to

Stoeckley’s confession to Blackburn and Blackburn’s threat to be untrue.”                
Id. Addressing other
evidence of Blackburn’s alleged threat, including Stoeckley’s

statements to her mother and MacDonald defense lawyer Wendy Rouder that she falsely

testified at trial because she “was afraid of the prosecutor” and could not confess “with

those damn prosecutors sitting there,” the court found the other evidence to be deficient

even if true. The court explained that Stoeckley’s statements to her mother and Rouder

“only allow for speculation that Stoeckley falsely testified because Blackburn threatened

her.” 
Id. To be
sure, the district court’s factual findings are well-supported by the evidence,

and not clearly erroneous. Consequently, we affirm the court’s ruling that the Britt claim

fails on its merits.

                                            2.

       Turning to the DNA claim, MacDonald seeks relief under Herrera v. Collins, 
506 U.S. 390
(1993) and its progeny, wherein the Supreme Court assumed “that in a capital

case a truly persuasive demonstration of ‘actual innocence’ made after trial would render

the execution of a defendant unconstitutional,” see 
Herrera, 506 U.S. at 417
, without yet

“resolv[ing] whether a prisoner may be entitled to habeas relief based on a freestanding

claim of actual innocence,” see McQuiggin v. Perkins, 
569 U.S. 383
, 392 (2013). As we

explained in our 2011 decision, MacDonald faced a “daunting burden” in establishing his

eligibility for relief because of his actual innocence, in that the Supreme Court has never

“come across any prisoner who could make the ‘extraordinarily high’ threshold showing

                                            153
for such an assumed right.” See 
MacDonald, 641 F.3d at 616
(quoting 
Herrera, 506 U.S. at 417
).

       Reasoning that a freestanding actual innocence claim would require a showing at

least as high as that required by § 2255(h)(1), the district court ruled against MacDonald

on the merits of his DNA claim. See MacDonald, 
32 F. Supp. 3d
at 707 (“[E]ven if the

court assumes that the standard for evaluating a freestanding actual innocence claim is

identical to the standard employed in the court’s gatekeeping duties under § 2255(h)(1),

for the reasons the court already has discussed, MacDonald has failed to meet this

‘extraordinarily high’ burden.”). We agree with the court’s analysis and affirm. 14



                                            IV.

       Pursuant to the foregoing, we affirm the judgment of the district court denying

MacDonald’s successive 28 U.S.C. § 2255 motion.

                                                                              AFFIRMED




       14
          In addition to the Britt and DNA claims, the district court considered and
rejected in its 2014 decision what it called the “footnote claim,” a Brady claim that had
been suggested in a footnote of one of MacDonald’s memoranda following the 2012
evidentiary hearing. Following the 2014 decision, MacDonald filed a motion to alter or
amend the judgment, which the court denied. MacDonald does not pursue the footnote
claim in this appeal. To the extent that he challenges the denial of his motion to alter or
amend the judgment, we affirm the denial of that motion.


                                           154

Source:  CourtListener

Can't find what you're looking for?

Post a free question on our public forum.
Ask a Question
Search for lawyers by practice areas.
Find a Lawyer