Court Records

May 31, 1985

U.S. Court of Appeals for the Fourth Circuit

On Appeal from the United States District Court for the Eastern District of North Carolina

Brief of Appellant

Related Files

VS. : NO. 85-6208
: [D.C. No. 75-26-CR-3 E.D.N.C.]



The defendant/appellant, Dr. Jeffrey R. MacDonald (hereinafter "MacDonald") is before this Court on appeal from the District Court's denial of his post conviction motions for relief and motion to recuse the trial judge.

MacDonald's motions had their genesis in a post-trial investigation which uncovered evidence provided by 35 witnesses that a group of drug abusers, identical to the group MacDonald had described at the time of the murders, committed the MacDonald murders. Three members of the group confessed to the murders and two made admissions of their participation.

The post-trial investigation also led to the startling discovery that the trial judge was the father-in-law of the prosecutor who directed the government's early investigation of MacDonald, concluded MacDonald was guilty, interviewed and discredited the key defense witness Helena Stoeckley, and threatened resignation if MacDonald were not indicted.

Finally, the investigation disclosed intentional governmental misconduct designed to deprive MacDonald of a fair trial, including the intentional intrusion into his right to counsel, and the deliberate suppression of exculpatory evidence critical to his defense.


1. Whether the trial judge erred as a matter of law in denying MacDonald's motion to recuse himself despite the disclosure that his son-in-law originally directed the Justice Department's investigation of MacDonald, prepared the case for trial and threatened to resign unless MacDonald were prosecuted.

2. Whether the court erred in denying MacDonald a new trial where the newly discovered evidence would probably have changed the results of the trial.

3. Whether the court misapplied the law applicable to cases of suppressed exculpatory evidence in denying MacDonald a new trial.

4. Whether de novo review of the facts is required in light of the trial court's gross mischaracterization of the facts where the trial court's findings were based almost exclusively upon documentary evidence.

5. Whether the trial court erred in holding that MacDonald's Sixth Amendment and Fifth Amendment rights were not violated by the government's forensic criminalist's examination of MacDonald in the guise of a psychiatric examination.


The standard of review to be applied to an appeal from a denial of the motions herein is whether the trial court abused its discretion.

The trial court abused its discretion by consistently misinterpreting, misstating and mischaracterizing the facts before it and by either applying the wrong legal standard or creating mythical legal standards for MacDonald to satisfy.

As is more specifically described below, the court's mischaracterization of the facts was of such proportions as to require a de novo review of the evidence.

Moreover, the unusual circumstances of this case, which saw government conduct designed to abridge the appellant's right to counsel and right to present evidence and the impropriety apparent in the case being heard by the father-in-law of the investigative prosecutor, compel this Court in the exercise of its supervisory powers to reverse the judgment of the District Court as a matter of sound judicial administration. United States v. Reed, 647 F.2d 678, 687 (6th Cir. 1981).


A summary of the facts presented at trial is provided in order to give the Court a perspective against which to assess the significance and weight of the newly discovered evidence upon which MacDonald based his post-conviction motions.[1]

In the early morning of February 17, 1970, military police were summoned to MacDonald's home at Ft. Bragg, North Carolina where they found MacDonald's wife and two children clubbed and stabbed to death and MacDonald unconscious and lying partially across his wife's body in the master bedroom. United States v. MacDonald, 531 F.2d 196, 200 (4th Cir. 1976).

MacDonald told the MP's and investigators that his family had been attacked by drug-crazed intruders, four of whom he saw and described. (TR 6580-84). MacDonald, asleep on the couch in his living room, was awakened by the cry of his wife Colette and 5 year old daughter Kim; as he awoke was attacked by four intruders whom he identified as a woman, two white men and a black man. (TR 6574-6579-81). The woman was blond and wore a floppy hat, boots, a short skirt and carried a flickering light or candle. (TR 6588-6592). The black man was stocky and wore an Army jacket with sergeant stripes. One of the white men wore a cross on a chain around his neck. (TR 6590, 6627-28).

MacDonald was knocked unconscious and when he awoke he was on the hall steps in the living room. (TR 6594-95). He arose and went into the master bedroom where he found his wife dead. (TR 6595-96, 6599). In efforts to administer aid to his family, MacDonald went to the children's rooms and returned to the master bedroom. (TR 6599-6605).

MacDonald received multiple stab wounds, one of which narrowly missed his heart and punctured his lung. United States v. MacDonald, 456 U.S.[1] (1982). He also received two head wounds. (TR 6586). MacDonald described how his pajama top had somehow been wrapped around his hands during his struggle with the intruders and further provided his recollection of how and when he removed the pajama top to place it over his wife's body. (TR 6586, 6605).

The government's case at trial was a theoretical reconstruction of the crime which was based entirely upon physical evidence discovered at the crime scene, primarily blood spatter and fiber evidence, which was inconsistent with MacDonald's account of the crime.

Notwithstanding its attempt to reconstruct the crime through circumstantial evidence, the government was unable to explain various items of evidence. For example, torn and bloody pieces of rubber gloves were found in the master bedroom. The government argued that MacDonald wore the gloves during the crime despite expert testimony that the rubber gloves used in the murders were of a different composition than those stored in the MacDonald residence. (TR 4891, 4909-10). The government did not explain the presence of wax drippings in MacDonald's house from candles not found there (TR 3841) nor did the government explain why the hair clutched in Colette's hand did not match any MacDonald family member. (TR 2298, 3858, 3852-53).

MacDonald was the only witness to testify about the circumstances of the murders. At trial a woman named Helena Stoeckley was called as a witness. She testified that at the time of the murders she had a floppy hat, (TR 5599) wore a shoulder-length blond wig, owned a pair of boots and that her appearance was similar to MacDonald's description of the female assailant. (TR 5588-90). She said she had no memory of the events between midnight and 4:00 a.m. on February 17, 1970 but denied involvement in the murders. (TR 5553, 5584-5604). The defense attempted to call six persons to whom Stoeckley had made statements suggesting her involvement in the MacDonald murders. The trial court found Stoeckley unreliable and excluded her statements from evidence.

MacDonald attempted to offer psychiatric testimony as to the inconsistency of his personality with crimes of this type. The court required MacDonald to submit to examination by Dr. James Brussel, a psychiatrist retained by the prosecution, as a precondition to such testimony. After MacDonald submitted himself to Brussel's examination and the court reviewed Brussel's report in camera,[2] the court ruled that it would not admit psychiatric testimony by either side, holding that the competing opinions of Brussel and MacDonald's psychiatric experts would confuse rather than assist the jury. Id. at 1099.

1/  The trial evidence is recited in some detail in MacDonald's Motion to Vacate Sentence and Motion for New Trial, J.A., pp. 88-288, 398-477 and incorporated herein by reference.

2/  United States v. MacDonald 485 F.Supp. 1087, 1094-97 (E.D.N.C. 1979).

2. The Newly Discovered Evidence

The new evidence established the participation of Stoeckley and several colleagues in the murders and demonstrated that the government manipulated the physical evidence to preclude the presentation of physical evidence supporting MacDonald's innocence.[3]

3/  The evidence is recited in detail in MacDonald's Motion for New Trial (J.A. pp. 9-29 and at pages 5-20 of MacDonald's Motion to Set Aside Judgment of Conviction (J.A. pp. 101-102, 298-313) and is incorporated by reference herein.

A. Stoeckley's post-trial confessions:

After trial, Stoeckley was persuaded to describe in detail her group's commission of the MacDonald murders to Ted Gunderson, formerly the Special Agent in Charge of the FBI offices in Los Angeles, Dallas and Memphis. (J.A. pp. 150-52, 203-08; Exhibits 2, 3, 8, 10, T2).[4]

Stoeckley also confessed to the murders in a videotaped interview for the "60 Minutes" television program and in an interview with Fred Bost, a news reporter. A videotape of the "60 Minutes" interview and an audio tape of the Bost interview are part of the record herein.(Exhibits 27, 29 and 30).

Stoeckley told Gunderson that in 1970 she associated with a group of drug abusing social dropouts, military personnel and their girlfriends. The group included Greg Mitchell, Don Harris, a black male who wore an Army jacket with sergeant stripes known as "Smitty" or "Zig-Zag",[5] Cathy Perry Williams (hereinafter Cathy Perry) and others. (J.A. pp. 150-15, 204-205, 2676-77).

Because of their anger with MacDonald's refusal to assist military personnel with their problems, Stoeckley's group decided to try to persuade MacDonald to assist soldiers by providing drugs and agreeing not to report users. (J.A. pp. 2749-50~ 2816-18, 2824-25, 2858-59, 2869-70). The group surveilled MacDonald, learning where he lived and that his wife Colette was a student at a university extension school at Ft. Bragg. (J.A. pp. 204, 2677, 2692-93, 2750). On the evening of February 16, 1970, some members of Stoeckley's group went to the university and spoke to Colette to seek her intervention with MacDonald. (J.A. pp. 2837-38). At 10:30 p.m. that evening, Stoeckley, Mitchell, Harris, "Smitty" and others met at Stoeckley's apartment and discussed going to MacDonald's home. (J.A. pp. 2750, 2824-25, 2888-89). The group imbibed in narcotics, including marijuana, mescaline and LSD. (J.A. pp. 2671-78, 2759, 2816, 2825-26). Stoeckley, Mitchell and "Smitty" went to the Apple House Restaurant around 11:00 p.m. and left at closing time to go to the Dunkin' Donuts Restaurant. "Smitty" was wearing a fatigue jacket with E-6 stripes. (J.A. pp. 151, 205, 2683, 2826).

At 2:00 a.m., Stoeckley, Mitchell, Harris and "Smitty" left Dunkin' Donuts and drove to MacDonald's residence as Mitchell urged them to "get on with it." Stoeckley stated that they got out of a car a few doors from MacDonald's house, walked to the back of the residence and entered through an unlocked door. (J.A. pp. 206, 2678, 2694, 2746, 2752, 2804, 2806, 2892-94, 2911, 3030-32). They went through the house into the living room where MacDonald slept on the couch. (J.A. pp. 26, 2679, 2695, 2752, 2808, 2820, 2896, 3032). Stoeckley lit a candle to provide light additional to the dim light in the house. (J.A. pp. 2679, 2695, 2753, 2762).

As MacDonald was jostled awake he began to fight with the intruders. "Smitty" struck at MacDonald and Stoeckley said something like "hit the pig" and "acid is groovy." (J.A. pp. 2679-80, 2695, 2755, 2761, 2820, 2848, 2901-02, 2932, 2981, 3033). After originally struggling, MacDonald was subdued and told they wanted to obtain drugs. (J.A. pp. 2679, 2695, 2755,2821, 2897, 2900, 2911-12, 3035). MacDonald agreed to call a doctor friend to see if he could obtain drugs at that hour. (J.A. pp. 2679, 2696, 2755, 2821-22, 2901, 2912). When MacDonald went to the wall phone in the kitchen he asked for the MP's and then was assaulted and knocked unconscious. (J.A. pp. 2679-80, 2697, 2757, 2821, 2902-03, 2912).

Stoeckley heard Colette call for help in a gurgling voice. (J.A. pp. 2679, 2696, 2756, 2981). Stoeckley went to a back room where she saw Colette being assaulted by Mitchell and another. (J.A. pp. 2681-82, 2697, 2756-57, 2681-82, 2851, 2907). Stoeckley said she saw a record player and some children's books in one of the children's rooms and a broken hobby horse in the other. (J.A. pp. 26825 26985 27565 2758) 2765, 2808-09, 2848, 2913, 2935, 3088-89). Stoeckley also saw Mitchell washing his hands in the bathroom sink. (J.A. p. 2809).

While they were in MacDonald's house the phone rang and Stoeckley picked it up. (J.A., pp. 2680, 2700, 2760-615 2914, 2981, 3035). A man asked for Dr. MacDonald at which point Stoeckley laughed and a member of the group told her to "hang up the God damn phone." (J.A. pp. 2680, 2700, 2761, 2914, 2981).

Stoeckley did not recall whether the group left the crime scene in more than one car; however, she went with others in a car to Dunkin' Donuts and thereafter the group went to the Hickory Trailer Park where a friend had a trailer. (J.A. pp. 2683, 2699, 2700, 2766, 2826, 2918-20, 3039-40). The only murder weapon not left behind was a scissors which Stoeckley took with her. (J.A. pp. 28, 198). Stoeckley stated that approximately 24-hours after the murders, she, Mitchell, Harris and "Smitty" went to her apartment. (J.A. pp. 29245 2982, 3057-58). Stoeckley stated that she feigned confusion at MacDonald's trial because of her fear of prosecution. (J.A. p. 2928).

4/  All references to the Joint Appendix are referred to herein as "J.A." followed by the page number.

5/  The Court found that "Zig Zag" or "Smitty" was a man named Dwight Smith, but there was no evidence to support this finding.

B. Mitchell and Perry's Confessions

Greg Mitchell, whom Stoeckley saw clubbing Colette, confessed to four people, including his two best friends, that he killed MacDonald's family. (J.A. pp. 189-91, 270-71, 474-82). Cathy Perry, who had a history of stabbings and who hid bloodstained clothes and boots with a friend a few days after the murders, confessed to participating in the February 1970 murders of the pregnant wife and two children of a Fort Bragg doctor who had given drug abusers a hard time. (J.A. pp. 1802-10).

C. Harris and "Smitty's" admissions

The week after the murders, Stoeckley, a stocky black male wearing a fatigue jacket with E-6 striped, identical in appearance to "Smitty", and a man named Don, wearing a cross on a chain around his neck and similar in appearance to Harris, discussed a "ritual." "Smitty" mentioned the ritual had "avenged" some grievance. In that connection, Stoeckley mentioned candles and Don said "shed blood cleanses" and nodded his assent to Stoeckley's statement to friends that he and Stoeckley took part in the MacDonald murders. (J.A. pp. 214-17).

Several months after the murders, Stoeckley and a man named "Don" identical in appearance to Harris told a friend they and others, including a black man, had killed MacDonald's family. (J.A. pp. 212-13).

3. Corroborating evidence

The court discredited the 33 defense witnesses whose testimony was presented by way of affidavit and the 5 witnesses who testified at the evidentiary hearing. The extraordinary detail of information provided by the newly discovered witnesses is important to an understanding and assessment of their credibility. While the great detail of the corroborating affidavits precludes their recitation here, a detailed summary of the newly discovered evidence appears at pages 9 through 29 of the Motion for New Trial, J.A. pp. 101-121, and pp. 3-6, Exhibits A - J of the Motion to Set Aside Judgment of Conviction, J.A. pp. 1812-52, and is incorporated by reference herein. In summary, that evidence included:

(a) Corroboration of Stoeckley's association with Mitchell, Don Harris, a stocky black man who wore an Army fatigue jacket with sergeant stripes and Cathy Perry (Declarations of Bowen, Mitchell and Harmon, J.A. pp. 160-161; 236-37; and 214- 217, respectively);

(b) Harris' admission of associating with Stoeckley and "Smitty" and frequenting Dunkin' Donuts and the Apple House Restaurant. (Declaration of Shedlick, J.A. pp. 256-25).

(c) Corroboration by a cook at Dunkin' Donuts that a black male called "Smitty" who wore a fatigue jacket frequented the restaurant and left the area shortly after the murders.
(Declaration of Shedlick (#2), J.A. pp. 258, 259).

(d) Corroboration by Mitchell's wife that he was at the drug rehabilitation center where he first admitted murdering the MacDonald family. (Declaration of Mitchell, J.A. pp. 239-40).

Evidence from a medical examiner that the blow that killed Colette was inflicted by a club swung in a left-handed swing and and was consistent with a blow inflicted by a left-handed person. (Declaration of Wright, M.D., J.A. pp. 279-86).

(e) Evidence that Mitchell was left-handed and that Mitchell's blood was found on Colette's hand while MacDonald' was not. (Declaration of Mitchell, J.A. pp. 239-40). (Motion to Expand the Record).

(f) Forensic evidence corroborating Stoeckley's statement that scissors were used in the murders. (Declaration of Wright, M.D., J.A. pp. 279-86).

(g) Evidence that a black man, a white male identified as Mitchell, and a woman resembling Stoeckley wearing a floppy and boots were together at a drive-in at Fort Bragg on February 15 or 16, 1970. (Declaration of Campbell, J.A. pp. 172-173).

(h) Evidence that a white female with a floppy hat, a black male and two white males were together in a van in Fayetteville, a few hours before the murders. (Declaration of Humphries, J. A. pp. 223-25).

(i) Evidence that a professor at North Carolina State University extension saw persons, including a man matching Mitchell's description and a woman matching Stoeckley's description speaking with Colette at the school at Fort Bragg at approximately 9:40 p.m. on February 16, 1970, corroborating Stoeckley's statement about a confrontation with Colette at North Carolina State. (Declaration of Boushey, J.A. pp. 154-157).

(j) Evidence corroborating the Stoeckley group's presence in Dunkin' Donuts restaurant from approximately midnight until approximately 1:30 a.m. on February 17, 1970 until leaving in a dark-colored van going toward Fort Bragg. (Declaration of Marian L. Campbell attached to Motion for New Trial, J.A. pp. 175-80; and Declaration of Bushey, J.A. pp. 164-66).

(k) Evidence that Jimmy Friar called the MacDonald residence around 2:00 a.m. on February 17, 1970, asked for Dr. MacDonald and spoke with a hysterical woman who was told by a voice in the background to hang up the phone, corroborating Stoeckley's statement that a man called looking for Dr. MacDonald during the murders. (Declaration of Jimmy Friar, J.A. pp. 200-01).

(l) Evidence from a psychological examination that Stoeckley was capable of both recollection and cognition and understood the difference between truth and falsity at the time of her confession. (Declaration of Dr. Beaber, J.A. pp. 140-41).

(m) Evidence from a military policeman that a group of three men ran out of a wooded area approximately a block and half behind the MacDonald residence at 2:15 a.m. on the morning of the murders and entered a dark van similar to the one which was seen earlier leaving Dunkin' Donuts going towards Ft. Bragg. (Declaration of Torres, J.A. pp. 260-63, 2422-28).

(n) Evidence that Stoeckley, whose boots had dark stains on them, and others, including a black male wearing an Army fatigue jacket, were at a drive-in restaurant near Fort Bragg at approximately 8:00 a.m. after the murders and that Stoeckley asked the waitress if she had heard that the MacDonalds had been murdered the previous evening. (Declaration of Sonderson, J.A. pp. 274-75).

(o) Evidence that at approximately 9:00 a.m. that same day Stoeckley, a black male in a fatigue jacket, and two white males were at a convenience store across from the Hickory Trailer Park. Stoeckley's boots were darkly stained and she smelled "like a hog slaughter." (Testimony and Declaration of Averitt, J.A. pp. 2382-2402).

The new evidence also included evidence which had been suppressed by the government at trial despite MacDonald's particularized demands for it and its obvious materiality. The suppressed evidence is corroborative of the Stoeckley group's involvement and MacDonald's account of the crimes and included:

(a) Evidence that Stoeckley was considered by six police officers on two police forces to be a credible and reliable informant whose information had led to many arrests and convictions, including highly sensitive internal affairs cases. (J.A. pp. 1707-25).

(b) A bloody syringe half-filled with an unknown fluid which was found in a hall closet in the MacDonald house by a CID investigator during the crime scene investigation (J.A. p. 301).

(c) Evidence that Cathy Perry left bags containing bloody clothing and boots with a friend shortly after the MacDonald murders, evidence that Perry had been involved in numerous stabbings while under the influence of drugs, and evidence that Perry had a psychotic break-down shortly after the MacDonald murders. (Declaration of Shedlick, J.A. pp. 267-69).

(d) The existence of a small piece of skin with an oily texture taken from under Colette's fingernail. The skin was lost or destroyed and its loss covered up by the government. (J.A. pp. 304-07).

(e) Photographs of the letter "G" written on the wall in Stoeckley's apartment in Nashville in 1970, taken by a CID investigator, and the observation of the CID investigator that the letter was similar to that found in the word "PIG" which was written in blood on the headboard of the bed in the MacDonald residence. (J.A. pp. 307-09).

(f) Seven fingerprints taken from the hallway in the MacDonald residence which could have been used to establish the presence of intruders. The government mischaracterized what had happened to the fingerprints to hide the fact that they had been intentionally destroyed. (J.A. pp. 312-13).



6/  Dr. MacDonald's arguments for recusal are contained in his Motion to Recuse, Reply to Government's Response to Motion to Recuse, and Motion for Reconsideration of Recusal Motion. These motions are incorporated by reference herein at J.A., pp. 36-87, 1601-17, 1853-60.

A. Factual Background

The Army investigation of the MacDonald murders was monitored from its inception by Assistant U. S. Attorney Proctor who was then the son-in-law of the judge who tried MacDonald and later ruled upon MacDonald's post-trial motions. (J.A. p. 46). When the Army proceedings against MacDonald concluded, Proctor analyzed the Army's investigative materials to determine whether there was sufficient evidence to charge MacDonald, and checked leads pointing to Stoeckley. (J.A. p. 42). Proctor interviewed Stoeckley, concluded she and her associates were not involved in the murders, and that MacDonald had killed his wife in the heat of passion and killed his children to disguise the murder as the acts of drug crazed hippies. Id. He further concluded that MacDonald had gotten the idea for the drug-crazed hippies story from an article in Esquire Magazine. Id. Proctor prepared extensive documentation in anticipation of trying MacDonald and planned to present the MacDonald case to the grand jury in November 1970. (J.A. p. 43). Proctor went to Washington on numerous occasions to persuade the Justice Department to prosecute MacDonald. (J.A. p. 1854). Proctor "begged, pleaded, even threatened to resign" to obtain approval to prosecute. Id. When the Justice Department refused to reopen the case, Proctor sought the intervention of the Army, telling the Army that he had the case prepared for trial. (J.A. p. 43). Proctor's interest was so great he volunteered to give the Army the Justice Department files, even if the United States Attorney himself would not give the materials to the Army. (J.A. pp. 77-78).

Those circumstances created at the very least an appearance of partiality by the trial Judge. SCA Services, Inc. v. Morgan, 557 F.2d 110, 116 (7th Cir. 1977); see, also, Rice v. McKenzie, 581 F.2d 1114, 1116 (4th Cir. 1978), quoting H.R. Rep. No. 93-1453, 93rd Cong. & Admin. News 6351, 6354-55. However, the judge declined to recuse himself and then denied MacDonald's motions. (J.A. 1765).

B. Recusal was required under Title
28 U.S.C. Section

Title 28 U.S.C. Section 455(a) requires disqualification of a federal judge when there is even the "appearance of partiality on the part of the judge." Rice v. McKenzie, supra.

The court below held that recusal was not required, relying upon S. J. Groves & Sons v. International Brotherhood of Teamsters, 581 F.2d 1241 (7th Cir. 1978) and incorrectly holding that in Groves the Seventh Circuit had modified SCA Services, supra, a case urged by MacDonald to require recusal. The court's reliance was both misplaced and deceptive.[7]

The Seventh Circuit did not modify SCA Services in Groves. Rather the court distinguished the facts of the case before it from those of SCA Services, to uphold the District Court's decision denying a motion to recuse. Moreover, the Groves court expressed strong reservations in denying the recusal motion., and specifically limited its Groves holding to the precise facts of Groves.

In the instant case, until the time of Proctor's resignation as Assistant U. S. Attorney, he actively pursued the investigation of MacDonald and recommended his prosecution. Thus, unlike Groves, the appearance of partiality existed from the moment Judge Dupree was assigned the case by virtue of the existence of a familial relationship between the Judge and Proctor, the father of the judge's grandchild, who had so vehemently pushed for the defendant's prosecution, and who had reached conclusions as to Stoeckley's non-participation "[W]here the appearance of impropriety is already established, a taint on the Judicial system remains as long as [the judge] presides over the case." 581, F.2d 1231, 1247, quoting Fredonia Broadcasting Corp. v. RCA Corp., 569 F.2d 251 (5th Cir. 1978), cert. denied 439 U.S. 859.

The trial court also erred finding that Proctor had a limited role in the case and was unsuccessful in having MacDonald prosecuted during his tenure as U.S. Attorney. The court's factual finding flies in the face of the uncontradicted evidence that Proctor's role was substantial, ongoing, and vigorous. As the factual underpinnings of the court's legal conclusion were erroneous, the conclusion is erroneous.

Additionally the trial judge misconstrued 28 U.S.C. 455(a) in relying upon his recollection of never having discussed the case with Proctor and the defendant's failure to claim on the record that he did not receive a fair trial (J.A. p. 1776). The fairness of the trial, or whether the Judge and his son-in-law conversed about the case have no relevance whatsoever to recusal under section 455(a).

Recusal is mandated in any situation giving rise to an objective appearance of partiality and a Judge must exercise his discretion in favor of recusal if there is any question of partiality. H.R.Rep. No. 93-1453, 93rd Cong., 2d reprinted in [19741 U.S. Cong. & Admin. News 6351, 6455; Postashnick v. Port City Construction Co., 609 F.2d 1101, 112 (5th Cir. 1980), cert. denied, 449 U.S. 820 (1980). As the the relationship of Judge Dupree and Proctor give rise to an objective appearance of partiality Judge Dupree abused his discretion in not granting recusal. Indeed, the appearance of partiality is so strong that this Court should grant MacDonald a new trial to avoid any questions about the partiality's effect upon the outcome of the trial. Id.

7/  Memorandum of Order Denying Recusal (J.A. p. 175).



A. Stoeckley as a recanting witness

The court found that Stoeckley was a recanting witness despite the fact that at trial, Stoeckley testified only that she resembled the woman MacDonald described, was similarly garbed at the time of the murders and that she had no memory of events between midnight and 4:00 a.m. In finding Stoeckley a recanting witness the court held that the new trial motion would be governed by the standards applicable to a recanting witness, at least with respect to Stoeckley's statements. (J.A. p. 2006). Under that standard, if the court concludes the recanting witness testified truthfully at trial the motion for new trial must fail. United States v. Wallace, 528 F.2d 863, 866 (4th Cir. 1976).

The court held that Stoeckley's post-trial statements failed to satisfy the recanting witness standard, finding: "that Stoeckley was telling the truth at trial. . ., " and even assuming Stoeckley's post-trial statements were available at trial, "the jury might have reached a different conclusion" in any event because the government would have been able to develop the inconsistencies in Stoeckley's statements. (J.A. p. 2007).

The disingenuousness of the court's result oriented finding that Stoeckley testified truthfully at trial is evident from the court's numerous prior holdings as to Stoeckley's incredibility at trial and the court's March 1985 opinion in which the court found Stoeckley's trial testimony unreliable and false. (J.A. 2007). See Section 4(A) infra.

As the court's factual finding that Stoeckley was truthful at trial is so at odds with the record evidence and contradictory to the court's other findings respecting Stoeckley's credibility, it is respectfully urged that the court's finding should be reversed. Under those circumstances, a new trial is warranted under the recanting witness standard. Stoeckley's post-trial confession would have provided the jury amazingly detailed and corroborated direct evidence of MacDonald's account of the crimes which the jury could weigh against the government's hypothetical case. Beaber's testimony as to Stoeckley's unimpaired cognitive facilities and the corroborating confessions and admissions of Stoeckley's colleagues would add weight and substance to her confessions.[8]

Under these circumstances, the trial would have changed dramatically from an evidentiary clash between the government's circumstantial case and Dr. MacDonald's weakly corroborated testimony to one where there were three corroborated confessions and two admissions by members of the Stoeckley group, a group identical to that described by MacDonald. Thus the jury might have reached a different conclusion. Accordingly, Stoeckley's post-trial confessions require a new trial under the recanting witness standard.

8/  MacDonald's argument and facts regarding his Motion for New Trial are fully set forth at pp. 101-39 of the Joint Appendix and are incorporated by reference herein.

B. The newly discovered evidence would
probably have produced an acquittal
in the event of re-trial

9/  MacDonald has urged in both his motions for new trial and to set aside judgment that because the case involved both suppressed evidence and newly discovered evidence the court should apply the standards set forth in Brady V. Maryland and United States v. Aurs, (J.A. pp. 121, 123, 126, 288-348) and incorporated herein by reference. In the event this Court does not adopt the recanting witness standard, the tests articulated under Brady/Agurs is the appropriate standard herein.

The traditional standard of review by which newly discovered evidence is evaluated is whether it would probably produce an acquittal in a retrial of this case. United States v. Cravero, 536 F.2d 637 (5th Cir. 1976). It is evident from a review of the facts herein that the court erred in failing to order a new trial.

In the aggregate, the newly discovered evidence established the presence of assailants identical in description to the murderers described by MacDonald at the time of the murders. It established that five members of that group admitted their participation in the murders. This new evidence was supported by newly discovered physical evidence consistent with and corroborating the participation of the Stoeckley group in the murders.

At trial, the Government produced no evidence of motive, no admissions, and no eye witnesses. All the prosecution "proved" was that MacDonald's account of the murders was inconsistent with the government's hypothetical reconstruction of the crime. The premise of the government's case was severely impeached by the newly discovered evidence which proved that the government edited the crime scene evidence to eliminate evidence pointing to MacDonald's innocence and to the involvement of others.

The new evidence proved that Stoeckley's group assembled on the evening of the crime, planned the visit to the MacDonald residence and while there committed the murders. Their flight from the crime scene was almost certainly triggered by Friar's call. It appears that some members of the group split up to go to the van which had been parked on Bragg Boulevard where Torres observed them. Stoeckley, "Smitty" and Mitchell apparently left in one vehicle and parked behind a drive-in restaurant at Fort Bragg where they slept off the drugs they had taken and where Stoeckley asked a waitress if she had heard the MacDonalds had been murdered. Later Stoeckley and Smitty went to the convenience store by the Hickory Trailer Park where Averitt saw Stoeckley with brownish stains on her boots, smelling like a hog slaughter.

A jury hearing the detail of the new evidence and the corroboration provided by the various witnesses would probably acquit MacDonald based on this evidence alone. The newly discovered testimony of the witnesses does not stand alone however. It is corroborated by the confessions and admissions of Mitchell, Perry, Harris and "Smitty" that they participated in the murders. Mitchell's admissions are corroborated by Stoeckley's statement that it was Mitchell who actually murdered Colette. Both Stoeckley's statement and Mitchell's admission are corroborated by forensic evidence that a left-hander probably inflicted the blows which killed Colette as Mitchell was left-handed.

Perry's hiding her bloodstained boots and clothing several days after the murders, her history of stabbings and the fact of Perry's psychotic condition corroborate her admissions.

Rule 33 was promulgated to serve the interests of justice in those cases where an injustice was done. Here, the combination of government misconduct, judicial impropriety and strong evidence pointing to persons other than the defendant, 5 of whom have admitted their involvement, calls out for a new trial in the interests of justice.



10/  MacDonald's argument and factual documentation regarding government suppression is located in the Motion to Set Aside Judgment of Conviction ("Brady Motion") and Reply to Government's Response to Brady motion located at J.A. 301-11, 1618-36 and incorporated herein by reference.

A. The court erred in its application of
Brady/Agurs to the facts herein

Brady v. Maryland, 373 U.S. 83 (1963) and United States v. Agurs, 427 U.S. 97, 96 S.Ct. 2392,2397 (1976) require the court in the case herein to apply one of the following standards for reviewing the materiality of suppressed evidence and determining whether a due process violation has occurred:

(1) Whether there was "any reasonable likelihood that the false testimony could have affect ed the judgment of the jury" (required when the prosecution creates a false impression of material fact). Hameric v. Bailey, 386 F.2d 390 (4th Cit. 1967), Napue v. Illinois, 360 U.s. 264 (1958); and (2) whether "the suppressed evidence might have affected the outcome at trial." (required when the evidence was specifically requested).[11]

MacDonald demonstrated that (1) he had specifically requested the evidence prior to trial[12] and (2) in suppressing the evidence, the government knowingly created a false impression of material fact in presenting and arguing its case. Additionally, as MacDonald showed that the suppressed evidence was material under any standard and that the government's conduct was intentional, an application of an even lower standard of materiality is compelled. United States v. Disston, 582 F.2d 1108 (7th Cit. 1978). United States v. Keogh, 391 F.2d 138 (2nd Cit. 1968.)[13]

11/  MacDonald's discussion of the appropriate standards is fully set forth in Brady motion (J.A. pp. 314-48) and incorporated by reference herein.

12/  A discussion of MacDonald's specific Brady requests is fully set forth at pages 25-28, J.A. pp. 319-21, 1812-52) and incorporated by reference herein.

13/  A discussion of the government's intentional conduct is fully set forth located at J.A. pp. 331-36 and incorporated by reference herein.

B. In the instant case the government presented
a false impression of material fact.

14/  The prosecution had in its possession but did not disclose CID Agent Medlin's statement that he had seen a half-filled syringe in the hall closet and that it appeared as though someone with a bloody hand had reached into the closet for something, evidence suggesting an intruder tampering with MacDonald's medical supplies. At trial, the prosecution allowed Ivory to testify that nothing out of the ordinary existed in the hall closet, that the hypodermic syrin ges remained undisturbed and no evidence suggesting intruders in MacDonald's home. (TR 1750).

The government failed to disclose its loss of skin found under Colette's fingernail and the fact that the loss was covered up. Similarly, the destruction of seven fingerprints was covered up. Thus, in an entirely circumstantial case, the prosecution presented a tidy hypothetical reconstruction of the crime scene while disguising the loss of critical evidence which would have destroyed the premise of this prosecution.

Despite its having withheld evidence of the half-filled bloody syringe and Perry's boots, the government argued that MacDonald's testimony about intruders lacked credibility, stressing that the medication in the MacDonald residence remained undisturbed (TR 7112) and that there was no evidence of any boots. (TR 7112).

Moreover, the government successfully opposed the admission of Stoeckley's self-incriminating statements by arguing their lack of corroboration without disclosing its knowledge of the "G" on Stoeckley's wall, Perry's bloody boots and clothes, and the information that Stoeckley was evaluated by six police officers on two police forces to be a credible, reliable and productive informant. (TR p. 5778, 5791, 5792). The prosecution thus created a false impression of material fact.

Despite the substantial evidence that the government knowingly presented a false impression of material fact, the court held that it need not address the applicable standard of materiality, concluding summarily that: (1) the government did not suppress evidence, and (2) there had been an insufficient showing that the suppressed items would have been favorable to the defense.

In so concluding, the court misapplied the rule of Brady/Agurs and erred as a matter of law.[15]

Under Agurs, the Court used the notions of favorability and materiality interchangeably and thus the concepts become one. Moreover, even under the pre-Agurs standard "favorable" means anything "potentially useful" to the defendant and the burden of demonstrating that evidence is favorable is slight. Moore v. Illinois, 408 U.S. 786, 92 S.Ct. 2562, 2568 (1972). Thus, the court's failure to analyze whether any of the items were potentially useful was erroneous.[16]

15/  MacDonald's discussion of the favorability of the evidence is fully set forth in Brady motion (J.A. p. 8-18, Declaration of forensic criminalist Richard Fox (J.A. p. 349-365), Defendant's Reply to Government's Opposition to Brady motion (J.A. pp. 1625-1633).

16/  For example, the court ignored the potential utility of the half-filled syringe noting only that counsel for MacDonald conceded at oral argument that the evidence might have been favorable to the government. There is an equally good chance that the syringe could have pointed to drug-seeking intruders. The court also speculated that the "G", the "boots" and the "skin" might have been unfavorable to MacDonald, erroneously ignoring the possibility that they might have been favorable.


A. De Novo Review is Required

MacDonald presented affidavit evidence from 33 witnesses, videotape and audio tape confessions, and three testimonial witnesses in support his motion for new trial. The government's opposition to MacDonald's motions was based primarily upon the affidavits of FBI agents summarizing witness interviews. Thus, the court's findings of fact, including its credibility findings, were based almost exclusively upon documentary evidence. (J.A. pp. 1986, 1887).

The court discredited three confessions by speculating as to non-record facts, and then ignored two admissions even though they were supported by the record. The court also mischaracterized the corroborative evidence by ignoring the particulars of that evidence and then finding it not sufficiently particular. The court's mischaracterization and misrepresent- ation of evidence was so pervasive and so extensive that the court abused its discretion in making its factual findings.

Rule 52(a) F.R.Civ.P. and Rule 23(c) F.R.Cr.P. permit this Court to disregard the trial court's findings of fact where factual issues are decided on documentary evidence. Orvis v. Higgins, 180 F.2d 537, 539 (2nd Cir. 1950), cert. denied, 340 U.S. 810, 71 S.Ct. 37 (1950); Tidewater Patent Development Co. v. Kitchen, 371 F.2d 1004 (4th Cit. 1967)[17], cert. denied 389 U.S. 821, 88 S.Ct. 46 (1967).

The de novo review standard "is a natural and proper concomitant of appellate power and should commend itself to . . ." this case because the customary opportunity of the trial court to observe the demeanor of the witnesses is minimal where only a few of the many witnesses appeared in person. United States v. General Motors, 384 U.S. 127, 86 S.Ct 1321 (1966). Moore, Federal Practice and Procedure, Para. 52-04, pg. 52-08.

As more particularly set forth below, the case herein calls for de novo review because: (1) the documentary evidence, which is the subject of the court's credibility determinations, is equally available to this court, and (2) an independent examination of the evidence is required to undo the appearance of partiality created by the relationship of the judge to the investigative prosecutor and reinforced by the court's consistent mischaracterization of facts, substitution of its speculation for evidence and contradiction of its own findings in this case. Moreover, as the court's opinion is replete with factual determinations which, upon review, will leave this Court with a firm and definite conviction that a mistake has been committed, its findings must be overturned as clearly erroneous. Jones v. Pitt County Board of Education, 528 F.2d 414, 418 (4th Cir. 1975).

17/  This Circuit has not applied this standard in some instances, for example, Bowman v. Curt G. Joa, Inc., 361 F.2d 706 (4th Cir. 1966). However, this Circuit has adopted this standard in Harris v. United States, 370 F.2d 887 (4th Cir. 1966) and Tidewater Patent Development Co. v. Kitchen, 371 F.2d 1004, cert. denied, 399 U.S. 821, 88 S.Ct. 46 (1967).

1. New Trial Findings

A. The court finding that Stoeckley
was a credible witness at trial

The court held it was certain that Stoeckley was telling the truth at trial when she testified that she could not recall her whereabouts on the night of the murders (J.A. pp. 2006-07) and thus concluded post-trial confessions did not satisfy the prerequisites for a new trial under the recanting witness standard.

The court's post-trial finding directly contradicts the court's trial finding that Stoeckley's "testimony, I think, has no trustworthiness at all" (TR 5808); its "unmistakable impression . . . that [Stoeckley] was suffering from drug-induced mental distortion and that she could be of no help to either side of the case." 485 F.Supp. 1087 (E.D. N.C. 1979); and its post-trial finding that Stoeckley "was a totally unreliable, untrustworthy witness" (J.A. p. 2007).

B. The court's findings regarding Greg Mitchell

MacDonald introduced evidence that Mitchell confessed to the murders on four separate occasions, and provided corroborating Mitchell's confessions.

The court held "little can be said for sure about the evidence [of Mitchell's involvement] except that it is at best speculative and circumstantial." (J.A. p. 1995). The court made specific findings regarding Mitchell's confessions[18] which blatantly mischaracterized the evidence, and ignored the corroborating evidence offered by the defense, finding, for example:

". . . [Mitchell] voluntarily appeared at the Charlotte, North Carolina, office of the FBI in late 1981 where he denied any knowledge of the murders."

(J.A., p. 1995).

Thus, the court ignored the unrefuted evidence that shortly after being summoned to the FBI office to be questioned regarding the murders, Mitchell told his closest friends that he needed some money to leave the country because the FBI "was after him and [was] hot on his trail" for a serious crime that he had committed a long time ago at Fort. Bragg, North Carolina. (J.A. pp. 379-482). The court also discounted evidence from three disinterested witnesses as to Mitchell's confessions at The Manor in 1971 and because Mitchell was left-handed, the court discounted the testimony of Dr. Wright that Colette's murderer was probably left-handed. (J.A. p. 279).

Although the court conceded that neither Stoeckley not Mitchell could recall their where- abouts on the night of the murders, and Mitchell denied "being with Stoeckley that night" (J.A. p. 1985) the court speculated that it was likely that Mitchell and Stoeckley went some- where alone during the time the murders were committed. (J.A. p. 1985). That is, the court purported to buttress its blatant speculation by crediting some parts of Mitchell's statement and discrediting other parts of the same statement.

18/  Mitchell confessed on four separate occasions. He confessed at a prayer session in a drug rehabilitation ministry called "The Manor." The next morning. Mitchell was seen running from The Manor's farmhouse after the words "I killed MacDonald's wife and children" had been written in paint on the walls of the farmhouse; (Declarations of Phillips, Cannady, and Sisneros, J.A. pp. 189, 191, 251-53, 270-271, respectively); and Mitchell confessed to his friends the Lanes on two separate occasions.

C. The court's findings regarding Cathy Perry

In November 1984, Cathy Perry telephoned the FBI and said she wanted to make a statement regarding the MacDonald murders. She later changed her mind, but an FBI agent went to her home and persuaded her to talk. (J.A. p. 1801).

The court discounted Perry's confession as yet "another example of the bizarre behavior that the case has evoked from people who for some reason find it fascinating and see themselves as having played a part in this gruesome story", (J.A. p. 1999) implying that Perry confessed to gain attention.

The court's finding as to Perry's confessions flies in the face of record evidence that Perry did not seek publicity and refused to be interviewed. Perry consistently declined to give interviews in the years since the crime and even after her initial call to the FBI in November 1984 became reluctant to speak to the FBI. After she gave her statement to the FBI, she refused to speak to defense investigators. (J.A. p. 1802).

The court held: "the facts retold by [Perry] are completely at odds with the known facts and those . . . confessed to by Stoeckley." (J.A. p. 1999). In so holding, the court ignored the facts in Perry's statement consistent with Stoeckley's and MacDonald's account of the murders, including: (1) the woman victim was pregnant; (2) there were two children; (3) a white male was lying on the couch; (4) someone wrote in blood on the walls; (5) a body was placed in a blanket or sheet; (6) she pulled a child into a closet to hide. (a tuft of Kris' hair was found in the closet). J.A. 1871); and, (7) someone ordered her to tie the woman victim with a rope. (Some of Colette's injuries appeared to be rope burns. J.A. p. 1871).

Perry's stabbings at the time of the crimes and her complete psychological disintegration almost immediately after the crimes demonstrate that she was the type of person capable of committing such crimes. The court, in an ironic twist, pointed to these facts to discredit Perry's confessions.[19] (J.A. p. 1999).

19/  Perry's statement, J.A., pp. 1803-05.

D. The court's findings regarding the statements
of other witnesses corroborating Stoeckley's

In his motion for a new trial, MacDonald submitted affidavits of 33 witnesses corroborating the confessions of Stoeckley, Perry and Mitchell. The court summarily described the nature of these affidavits concluding:

"If the statements of these witnesses are true, the net effect of them all would only be to place Stoeckley and her friends in Fayetteville at locations close to where she and other members of the group lived in 1970 . . . . [T]his is only very weak circumstantial evidence proving that Stoeckley and her group were in the general Fort Bragg area at about the time of the murders."

(J.A. p. 1988).

The court appears to have purposefully mischaracterized the corroborative affidavits by eliminating their critical and often startling details and ignoring tacit admissions. For example, the court dismissed the combined affidavits of Bowen, Harmon, Mitchell and Harris as merely demonstrating Stoeckley's association with Mitchell, Harris and others. In fact, Harmon recalls that a week after the murders she saw Stoeckley, "Smitty," Harris, Mitchell and Perry in Rowan Park, continued to see them together for two weeks, and overheard a conversation which "Smitty" said something about the "ritual" and things being "avenged." Stoeckley said something about candles and Harris said "shed blood cleanses." (J.A. pp. 214- 16).

Richard Comisky fully corroborated Harmon, yet the court ignored his declaration. Comisky stated that in the Fall of 1970 he was in Rowan Park with Stoeckley and Harris when Stoeckley said to him "we did the MacDonald thing." When Comisky asked her what she meant she replied "we did the killings." Harris told her to "be cool" and not to "put their business on the street." Stoeckley asked Comisky whether fingerprints could be obtained from wax, and she asked Don Harris if he thought she should have brought the "spade" with them. (J.A. 212-213).

Similarly, the court trivialized the statements of Sonderson and Averitt by finding they said only that they saw members of the Stoeckley group together in the hours following the murders. In fact, Averitt had said that at approximately 9:00 a.m. she saw Stoeckley, "Smitty" and two white males in a dark sedan at a convenience store across from the Hickory Trailer Park. She stated that Stoeckley was wearing a blond wig which was falling off exposing her dark hair, a wide-brimmed weather hat, a light cream-colored plastic coat, a dark skirt and three-quarter length boots covered with a dark brown substance and that Stoeckley "smelled like a hog slaughter." (Averitt testimony, pp. 28-31, J.A. pp. 2393, 2400; Averitt Declaration, J.A. p. 1901). Sonderson saw the same group an hour earlier and Stoeckley asked Sonderson whether she had heard the MacDonalds had been murdered.

By ignoring such crucial detail, the court was able to conclude that the net effect of the witness statements was only to place Stoeckley "in the general Fort Bragg area at about the time of the murders." (J.A. p. 1988).

Additionally, Jimmy Friar's statement (J.A. p.200) that he telephoned the MacDonald residence at approximately 2:00 a.m. on the morning of the crimes and spoke to a woman who answered and was told to hang up directly corroborated a very peculiar detail of Stoeckley's statement, yet the court ignored it in assessing Stoeckley's credibility.[20]

20/  Similary, the court dismissed the testimony of Carlos Torress holding that descriptions of three men running from a wooded area a block and a half from the MacDonald residence at about 2:15 a.m. on February 17, 1970 to a dark van was not credible because he had just returned from Vietnam, was in the process of a divorce, and "wasn't in a condition to reveal this and get any more nervous and attention." (J.A. p. 1993). Moreover, the court held that Torres' statement was inconsistent with Stoeckley's statement that the group drove in a blue car, ignoring Stoeckley's statements that she was not sure how many of her group were involved, or how others traveled to MacDonald's.

A. The Court's finding that the half-filled
syringe did not exist

In denying MacDonald's motion to vacate sentence on the basis of government suppression of material evidence the court mischaracterized the evidence holding that the government had not suppressed any evidence.

2. Brady v. Maryland Findings

21/  Discussion of the half-filled syringe is fully set forth in the Brady motion (J.A. pp. 318-19).

Three days after the murders FBI Agent Tool interviewed Medlin of the Army's crime laboratory concerning the crime scene processing at the MacDonald residence. Medlin informed Tool that "a half-filled syringe that contained an as yet unknown fluid was located in a hall closet, which also contained some evidence of blood."[22] The government failed to disclose Medlin's statement to MacDonald despite its exculpatory character. In denying MacDonald's Brady motion, the court found, despite Medlin's statement, that a "half filled bloody syringe" never existed (J.A. pp. 1931, 1933). The court based its finding upon: (1) affidavits of four govern- ment witnesses that they had not observed a bloody half-filled syringe in their processing of the crime scene; and (2) Medlin's 1984 affidavit that when he originally gave his statement to the FBI, he was only summarizing the information provided him by other members of the crime scene processing team and that he had no first hand knowledge of the contents of the closet. Agent Tool was not produced by the government.

The court speculated that "the implication of Mr. Medlin's statement to the FBI immediately after the murders is that Mr. Medlin misunderstood what the other investigators told him about the contents of the closet." (J.A. pp. 1929-33). Thus, the court ignored Medlin's unrefuted statement at the time of the crime and instead relied on Medlin's statement years after the crime which did not refute the original statement but merely denied Medlin's personal knowledge of the syringe. Moreover, the court ignored the fact that the suppression of Medlin's statement was prejudicial irrespective of the syringe's existence.[23]

22/  Medlin's statement to FBI Agent Tool, J.A. pp. 1218-19.

23/  Interestingly Medlin was the only government witness to testify at trial as to the existence of a bloody footprint - the print having been destroyed. Thus at trial, the court accepted Medlin's recollections when they supported the government while in the post-trial motions the court rejected Medlin's recollections when they supported MacDonald.

B. The court's findings regarding
Cathy Perry's boots and clothing

MacDonald presented evidence that Cathy Perry hid bloody clothing and boots with a friend shortly after the murders. The friend in turn gave the items to an attorney indicating that they were possibly linked to the MacDonald murders. Although the attorney delivered these items to the CID informing the CID of the possible connection with the MacDonald murders, and the CID signed a receipt for those items it kept for testing, MacDonald was never informed of this potentially useful evidence despite requests for all exculpatory evidence, including recent reports by citizens of possible suspects, police and other investigatory files of possible suspects. (J.A. p. 1815).

The court found that MacDonald should have known about the existence of the boots and their receipt by the CID because attorneys Douthat and Nance learned of the boots and each had represented MacDonald at an earlier time. (J.A. 1939). This finding was made in the face of Nance's and Douthat's sworn statements that as they were not MacDonald's lawyers at the time and they told neither MacDonald nor his lawyers about the boots[24] (Declarations of Nance and Douthat, J.A. pp. 1689-1694) and despite the court's conclusion that it was "unwilling to go so far as to find that Nance and Douthat told MacDonald's lawyers about the boots in the face of the affidavits by these two members of the bar." Id.[25]

The court further found that the government failed to perceive any connection between the boots and murders (J.A. p. 1937), and thus returned the boots and other articles to Mrs. Garcia, ignoring the uncontradicted declaration of attorney Nance that when he turned Perry's boots over to the CID he informed the CID that the items were possibly linked to the MacDonald murders. (Declaration of James R. Nance, J.A. p. 1693).

24/  The Court also stated "the Court has difficulty believing that these two lawyers, particularly Douthat, were so removed from the defense team in January of 1971 that they would have failed to bring the boots to the attention of MacDonald's lawyers, especially when it was well known that MacDonald had claimed a woman wearing boots had participated in the crimes. J.A.P. 1939)."
The court further characterized the evidence when it stated that "Nance went to the office of Captain James Douthat, MacDonald's appointed military counsel at the Article 32 proceeding, the afternoon of January 6, 1971 and released belongings to CID agents William Ivory and Peter Curen [Kearns]," suggesting that Nance chose to go to Douthat because he knew Douthat represented MacDonald. J.A. P. 1936). In fact, when Nance called the CID regarding Perry's garments, MacDonald had been discharged from the Army, Douthat no longer represented MacDonald and was not in contact with him. Douthat signed the document acknowledging items given to the CID at the insistence of the investigating agents to assist in the ongoing investigation of MacDonald. (Douthat, J.A. p. 1691).

25/  The court relied upon Stoeckley's trial testimony that she threw her white boots in a trash can because they might have connected her to the murders . . . and held "this testimony contradicts MacDonald's claim that the boots given by Nance to CID agents belonged to Stoeckley and further undermines his argument that the government suppressed exculpatory evidence. (J.A. p. 1970)."
The court thus ignored the fact that (a) the boots could have been Cathy Perry's and that Perry admitted her participation in the murders and gave the boots to Garcia, and (b) that Stoeckley said she had two pairs of boots, a beige pair and a white pair.

C. The court's findings regarding skin found
under Colette MacDonald's fingernail

MacDonald presented evidence that a small piece of skin, which had been found under Colette's fingernail was lost or destroyed and its loss covered up by the government.[26] The court found that the "'chances are very low that [the piece of skin) would have been exculpatory"' and speculated that the skin taken from Colette's fingernail was MacDonald's skin because the evidence at trial suggested there were scratches on MacDonald's chest. (J.A. p. 1945). Such gratuitous speculation puts at issue evidence not in the record and thus invites a reopening of the record to contradict the court's speculation. The evidence discloses that Colette had bloodstains of Greg Mitchell's blood type on her left hand but not MacDonald's. (See Motion to Expand Record).

MacDonald presented evidence that the government possessed statements by CID Agent Ivory regarding the loss of the skin. While the government disclosed numerous of Ivory's statements it did not disclose Ivory's statements regarding his observations of the skin or its subsequent disappearance or other CID references to lost skin, nor did the government disclose the various memoranda discussing the government's concern about Dr. MacDonald's factual documentation and arguments regarding the lost or destroyed skin are fully set forth in his Motion to Set Aside Judgment (Brady] (J.A. and attached affidavits, pp. 288-398); Defendant's Reply to Government's Opposition to Brady and attached affidavits (J.A. pp. 1642-1655); Addendum to Reply to Government's Opposition to Brady (J.A. pp. 1683-1707); Addendum to Brady motion (J.A. pp. 1812-1853) and incorporated herein by reference lost skin. (J.A. 375-77).

The court ignored the substantial evidence that the government hid the loss of the skin by providing all of Ivory's statements except those regarding the skin and failing to divulge other documentation disclosing the loss of the skin and held that:

"In short, while the court cannot condone the government's loss of this piece of evidence, neither can it conclude that there has been a suppression of evidence contrary to the requirements of Brady and Agurs."

(J.A. p. 1947).

26/  Dr. MacDonald's factual documentation and arguments regarding the lost or destroyed skin are fully set forth in his Motion to Set Aside Judgment [Brady] (J.A. and attached affidavits, pp. 288-398); Defendant's Reply to Government's Opposition to Brady and attached affidavits (J.A. pp. 1642-1655); Addendum to Reply to Government's Opposition to Brady (J.A. pp. 1683-1707); Addendum to Brady Motion (J.A. pp. 1812-1853) and incorporated herein by reference.

D. The court's finding regarding Proctor's
involvement in the case

In denying MacDonald's recusal motion the court ignored evidence of Proctor's substantial involvement in the case, and found:

" . . . that Proctor's role in the case was limited to representing agent Caverly at the Article 32 hearing and assisting FBI and CID investigators in this case . . . ."
J.A, p. 1775).

In seeking reconsideration of the recusal motion MacDonald provided further evidence of Proctor's active ongoing and vigorous role in the case, and the court responded by mischaracterizing its own earlier findings:

". . . [the court's] decision was not related so much to the feeling that Mr. Proctor's relationship to the case was minimal, but to my certain knowledge that whatever his interest in the case was, it was not known to the Court and would not have made any difference whatever. In other word words, it was more related to a minimal relationship between Mr. Proctor and myself before I came on the bench, . . . . "

(J.A. p. 2661).

That is, the court, in order to avoid the impact of the compelling evidence of Proctor's substantial involvement mischaracterized its own earlier finding in order to sustain its denial of the recusal motion.

Thus, throughout its opinion the court mischaracterized the facts: by ignoring corroborating details and then holding evidence was too general to be probative; by refusing to find intentional misconduct when intentionality is the only explanation for governmental failure to disclose exculpatory evidence; by speculating that facts existed despite record evidence to the contrary; and by finding the same witness credible for some purposes but incredible for other purposes. In each instance, the court's mischaracterization supports a factual interpretation and theory of the case identical to that arrived at by Assistant U. S. Attorney Proctor when he investigated the case. Under the circumstances, a de novo review of the facts is required to avoid the appearance of partiality inherent in the court's review of a case investigated by his son-in-law.


27/  A detailed recitation of Brussel's influential and pervasive role in constructing the government's case against MacDonald is contained in MacDonald s Motion to Vacate Sentence, J.A. pp 398-442, and incorporated herein by reference.

The court conditioned MacDonald's introduction of psychiatric evidence on the government's having a psychiatrist of its own choosing examine MacDonald. The government designated Dr. Brussel to conduct the exam without disclosing Brussel's pre-existing role as a government investigative agent in the MacDonald investigation, Brussel's detailed findings of fact and recommendations as to how the government's case should be amplified to convict MacDonald,[28] or Brussel's pre-existing opinion as to MacDonald's guilt and psychosis.

Outside the presence of counsel, Brussel examined MacDonald almost exclusively about the physical circumstances of the crime with few, if any, questions as to the psychiatric issues. Brussel spoke with the prosecutor about his findings.

Brussel's report was furnished to the court in camera. Largely on the basis of Brussel's exam-ination the court excluded the testimony of defense psychiatric experts because "we had shaping up a battle of the experts," which would confuse rather than assist the jury. United States v. MacDonald, 485 F. Supp. 1087, 1094-97 at 1099 (E.D.N.C. 1979), aff'd 688 F. 2d at 227-228.

In denying MacDonald's motion to vacate his conviction under 28 U.S.C. Section 2255, the court ignored MacDonald's contention that the government was deceptive in not disclosing Brussel's prior investigative role and earlier factual findings, and held that the government's failure to disclose those facts was not deceptive because MacDonald knew that Brussel had been consulted by the government as a psychiatrist prior to trial.

Despite testimony from the former president of the American Board of Forensic Psychiatry that Brussel's interview was not calculated to achieve the goals of a psychiatric exam, was unethical, and suggested Brussel was acting as government agent, the court found that Brussel was not an investigative agent; that the prosecution had not given Brussel questions to ask MacDonald; and, that the prosecutor had neither debriefed Brussel nor used any Brussel information against MacDonald. The Court further held as a matter of law that Estelle v. Smith, 451 U.S. 454, 461-74 (1981); and Massiah v. United States, 377 U.S. 201, 204-207 (1964) were inapposite, as the court found no evidence the prosecutor used information from Brussel's examination. (J.A. p. 1919).

The court erred: (a) in holding that the prosecution did not use the information obtained through Brussel's examination and, (b) in ignoring the intentionality of the government's conduct.

28/  MacDonald's argument regarding the Sixth Amendment violation is fully set forth in his Motion to Vacate Sentence J.A., pp. 417-32, and incorporated herein by reference.

A. The government's interference with MacDonald's
right to counsel was prejudicial

Although Brussel's examination almost exclusively concerned the physical evidence in the case, the court found the government had not relied upon Brussel as an investigative agent because the government already "knew that MacDonald was unable to explain certain physical evidence at the crime scene." (J.A. p. 1922).

Critically, the court ignored the government's prejudicial misuse of Brussel's examination. At trial, the government submitted, and the court relied upon, Brussel's report to preclude MacDonald from putting on expert psychiatric testimony. (J.A. p. 1910, 1914). Thus Brussel's sham examination enabled the government both to to prepare for cross-examination and to create a competing "psychiatric opinion" which the court relied upon in denying MacDonald's psychiatric evidence. Thus, the prosecutor's circumvention of MacDonald's right to counsel was prejudicial to MacDonald as the prosecutor used the fruits of the Sixth Amendment vio- lation to adversarial advantage. Massiah v. United States, 377 U.S. 204-207 (1964).

B. The government's interference with MacDonald's
right to counsel was intentional and thus no
prejudice need be shown

Where the government intentionally violates a defendant's Sixth Amendment rights, prejudice need not be shown. Assuming, arguendo, this Court were to find no prejudice to MacDonald from the government's violation of his Sixth Amendment rights, the deliberately deceptive character of the government's conduct would nonetheless require that MacDonald's conviction be set aside.

In United States v. Levy, 577 F.2d 200 (3d Cir. 1971) the Third Circuit held that a knowing violation of the defendant's Sixth Amendment rights resulting in the disclosure to the government of confidential information required reversal of the defendant's conviction.[29] The court held that under such circumstances, the defendant need not show prejudice resulting from the disclosures. See also Briggs v. Goodwin, 698 F.2d 486 (D.C. Cir. 1983), vacated on other grounds, 712 F.2d 1444 (D.C. Cir. 1983), cert. denied, 104 S.Ct. 704 (1984). Where the Court held: "A deliberate attempt by the government to obtain defense strategy information or to otherwise interfere with the attorney-client relationship through the use of an undercover agent may constitute a per se violation of the Sixth Amendment. Id. at 493 n. 22. The court distinguished Weatherford v. Bursey, 429 U.S. 545, 97 S.Ct. 837 (1977) where the Sixth Amendment violation was not deliberate.

The intentional Sixth Amendment violation herein requires either that the indictment be dismissed or that MacDonald's conviction be set aside and a new trial ordered. United States v. Morrison, 449 U.S. 364, 101 S. Ct. 665 (1981).

29/  The government's deceptive use of the Brussel interview of MacDonald was grossly unethical just as conduct in Levy. The prosecutors violated disciplinary Rule 7-104(A)(1) of the Rules of Professional Conduct.


MacDonald was tried upon a hypothetical reconstruction of the crime from evidence found at the crime scene. That crime scene has been shown by new evidence to have been edited by the government to eliminate evidence pointing to intruders such as those described by MacDonald. At the same time, MacDonald has presented substantial newly discovered evidence that the murders were committed by the very group which he identified at the outset of the case. Five members of the group have made admissions of their culpability for the murders.

Newly discovered evidence has revealed that the prosecutor who began the Justice Department's investigation of MacDonald and who vigorously advocated his prosecution on the theory upon which MacDonald was ultimately convicted was the son-in-law of the judge who tried MacDonald and who ruled upon his post-trial motions. The judge's post-trial rulings consistently sustained the positions first posited by his son-in-law.

Under all the circumstances a new trial is required in the interests of justice.

Date: May 31, 1985

Respectfully submitted,

/Brian O'Neill/

/Myrna K. Greenberg/
Brian O'Neill
A Professional Corporation

By: /Brian O'Neill/
Attorney for Appellant