UNITED STATES COURT OF APPEALS FOR THE DISTRICT OF COLUMBIA CIRCUIT
April 10, 1973
UNITED STATES OF AMERICA
JAMES A. LEWIS, APPELLANT APPEAL FROM THE UNITED STATES DISTRICT COURT FOR THE DISTRICT OF COLUMBIA. 1973.CDC.99
UNITED STATES COURT OF APPEALS FOR THE DISTRICT OF COLUMBIA CIRCUIT
Fahy, Senior Circuit Judge, McGowan and Robinson, Circuit Judges. Fahy, Senior Circuit Judge, concurring specially.
DECISION OF THE COURT DELIVERED BY THE HONORABLE JUDGE ROBINSON
This appeal follows appellant's second trial on an indictment charging single counts of armed robbery, *fn1 unarmed robbery *fn2 and carrying a dangerous weapon, *fn3 and two counts of assault with a dangerous weapon. *fn4 The first trial ended when the jury was unable to agree; the second led to appellant's conviction on all of the charges but unarmed robbery. The sole issue on appeal is the validity of a ruling by the trial judge granting the prosecutor leave to cross-examine appellant's proposed character witnesses, if called, as to their awareness of a recent arrest of appellant on a charge of violating the federal narcotic laws. We find error, but we further find that in the circumstances of the case it does not warrant reversal. We accordingly affirm the conviction. I
The indictment in suit emanated from a daytime holdup of two employees of a Safeway store as they walked therefrom toward a bank to make a deposit of store receipts. At the second trial, four eyewitnesses identified appellant as one of two men who staged the holdup, and other evidence pointed strongly to his participation in the affair. *fn5 Appellant asserted an alibi and presented eyewitnesses who in some respects corroborated his version of events transpiring before the robbery. *fn6
On the first trial, appellant presented a witness who testified that his character for honesty, peace and good order was good, and the prosecuting attorney was permitted, on cross-examination of the character witness, to ask whether she had heard of a prior arrest of appellant on a charge of unauthorized use of a motor vehicle. *fn7 Two weeks before commencement of the second trial, appellant was arrested for an alleged violation of the federal narcotic laws. *fn8 The issue confronting us traces its origin to that arrest.
At the second trial, as the Government's case-in-chief neared completion, the prosecutor, in the absence of the jury from the courtroom, inquired as to whether character evidence would again be offered in appellant's behalf. Defense counsel *fn9 stated that he had requested two character witnesses to be on hand but that he was uncertain as to whether he would call them. Counsel explained that only after arranging for the presence of the witnesses had he learned of appellant's narcotic arrest, and that he "would just like to talk to my character witnesses to see if they know about it." During the colloquy, the prosecutor reminded the trial judge that at the first trial, at which the same judge presided, the Government had been allowed to query appellant's character witness on her awareness of appellant's arrest for unauthorized use of a vehicle. The prosecutor made it clear that he wished a ruling permitting the same range of cross-examination as to that arrest, and as well the narcotic arrest in the event that the character witnesses were used.
Later during the trial, as the presentation of the case for the defense approached an end, the subject was again discussed at a bench conference, and the prosecutor renewed his request for leave to interrogate the character witnesses as to their knowledge of the two arrests. Defense counsel advised that he had not had an opportunity to see the character witnesses; "I want to talk to [them,] I may not use them," he reminded the prosecutor and the trial judge. The judge then offered a short recess for that purpose, but ruled that if character witnesses were called the prosecutor could inquire on cross-examination as to whether they had heard of the arrests. The recess was taken, and when the trial resumed the defense rested without presenting character testimony.
Since, following the judge's ruling, the character witnesses were not used, the record does not establish the aspects of appellant's community reputation to which they would have testified. On appellant's first trial, however, his good reputation for peace, good order and honesty was attested, and there is nothing to suggest that, aside from the presentation of an additional witness, the endeavor at the second trial would have differed. The offenses being tried were all crimes of violence, and before the judge ruled appellant had taken the witness stand in his own behalf. The good-character evidence would naturally have pertained to traits both of peacefulness and veracity, and we have no cause to believe that appellant would not have sought from his character witnesses all the help he could have gotten. Both the prosecutor and the trial judge appear to have assumed that the testimonial content of appellant's second effort would have duplicated the first. In assessing appellant's challenge to the ruling, then, we assume that had the character witnesses been called they would have testified that appellant bore in the community the reputation of a truthful, law-abiding citizen. *fn10
We also recognize that the decision not to use the character witnesses was tactical, and for very good reason, but, as we shall later see, *fn11 that consideration does not shield the judge's ruling from attack. As defense counsel emphasized during the two colloquies preceding the ruling, he had just learned of appellant's narcotic arrest and had not had a chance to ascertain whether the witnesses knew of it. On the first trial, the judge had permitted inquiry of appellant's character witness as to her knowledge of his arrest for unauthorized use of a motor vehicle. Particularly with the same judge presiding at the second trial, the expectation of a similar indulgence -- this time extending to the narcotic arrest also -- loomed large to defense counsel, who had represented appellant at the first.
Very understandably, then, counsel made plain his intention to interview the character witnesses on hand for the second trial further before undertaking to offer them. It was evident that the prosecutor would be allowed to ask about the unauthorized-use arrest, but the recent narcotic arrest posed a dilemma of greater magnitude. If the witnesses already knew of it, their avowal of good reputation was more vulnerable than before, and if they had been previously unaware of it their familiarity with appellant's community reputation was wider open to debate. *fn12 As the legal profession well knows, the accused may pay dearly for any use of second-rate character witnesses; *fn13 and in the instant case cross-examination as to the narcotic arrest would itself have brought the fact of that arrest to the attention of the jury. We have no difficulty in recognizing that inquiry about an arrest for unauthorized use of a vehicle is one thing, but that inquiry disclosing an arrest for a narcotic violation is something else again. On the other side of the ledger, from counsel's viewpoint, there would be no showing of good character at all if the witnesses were not called.
Since, after the recess taken for counsel's interview with the character witnesses, the defense rested without further ado, it is obvious that the showing on character was abandoned in order to avoid the prosecutor's questions respecting the narcotic arrest. It is clear, however, that the decision not to use the character witnesses does not immunize the action of which appellant complains. The prosecutor solicited and the trial judge rendered a ruling that appellant's character witnesses could be questioned as to whether they had heard of his arrest on the narcotic charge. The decision to forego character evidence was reached only after the ruling was made. There was no reason whatever to yield on the character presentation which defense counsel had planned save to keep away from mention of that arrest. On the contrary, it would be ridiculous to assume that the presentation would not have gone forward had the judge ruled the other way. It was the judge's ruling which opened the door to the unwanted inquiry -- in the context of the prosecutor's position, and inquiry which was a virtual certainty. It would blink reality to say the ruling did not influence the decision, if indeed it did not actually dictate it. We must, then, scrutinize the ruling to ascertain whether it was correct, and whether it was prejudicial if incorrect. *fn14 II
The principles governing the use of character evidence in criminal trials in this circuit have been delineated in past decisions, and need only brief summarization here. The accused may elect to advance one or more of his character traits as evidence of his innocence. *fn15 If he does, his proof is confined to evidence of his reputation in the community for those traits. *fn16 His presentation, in terms of number of such traits, may be as narrow or as broad as he chooses so long as it remains germane to issues on trial. *fn17 Good-character evidence may of itself generate a reasonable doubt as to guilt, *fn18 and the accused is entitled to have the jury so instructed. *fn19
The Government's options as to character-proof depend upon what the accused undertakes to do. Conviction of crime must rest upon proof beyond a reasonable doubt of the elements of a crime specifically charged, and evidence of a general predisposition toward crime is taboo. *fn20 The prosecutor cannot offer bad-character evidence unless the accused first introduces evidence of good character; *fn21 even then, the prosecutor's proof is restricted to community reputation, *fn22 and to the trait or traits to which the accused's own character evidence related. *fn23
When, however, a character witness, either for the accused or for the prosecution, is offered, he becomes subject to cross-examination as to his testimonial qualifications just like any other witness. The probe on cross-examination may extend to those matters, among others, which legitimately affect the witness' knowledge of the accused's community reputation for the character trait or traits which he confirms. Accordingly, it is well settled, both here and elsewhere, that it may become appropriate on cross-examination to ask a good-character witness whether he has heard reports of particular events, including prior convictions or arrests of the accused, which are inconsistent with the reputation to which he has testified. *fn24 Questions of this sort are permitted as a test of the credibility of the witness, for the good-reputation testimony may be doubted if the witness has heard the report, and the witness' acquaintance with the accused's community reputation may be disbelieved if he has not heard. *fn25 The inquiry is indulged solely for that purpose, and the jury must be instructed to limit consideration of the interrogation to an assessment of the worth of the witness' testimony. *fn26
This court has frequently sustained inquiries of this type, *fn27 but has also recognized the danger inherent in the practice. Merely to ask a character witness about his knowledge of a report is to get the facts reported before the jury; once there, notwithstanding the judge's limiting instruction, they may influence the jury's determination on the issue of guilt. *fn28 Not surprisingly, then, the practice has drawn criticism; *fn29 but it remains well nigh universal in the state courts, *fn30 and in the federal courts it might be here to stay. *fn31
The courts have, however, developed a series of restrictions designed to mitigate the potential hazard. The matters the witness is to be asked about should first be established to the trial judge's satisfaction as actual events. *fn32 The questions put to the witness should be carefully and narrowly framed. *fn33 The questions, of course, must be restricted to events affecting the character trait or traits the accused has placed in issue; *fn34 their propriety is to be determined "by comparison with the reputation asserted." *fn35 The process demands close supervision; "wide discretion is accompanied by heavy responsibility on trial courts to protect the practice from any misuse." *fn36 And obedient to the principle governing any use of evidence indicative of other criminality, *fn37 the inquiry should be permitted only when "the probative value of the information which might be elicited outweighs the prejudice to the defendant." *fn38 III
In the case at bar, as already noted, the prosecutor proposed to ask appellant's character witnesses, if called, whether they had heard of either of his two previous arrests. One of the arrests occurred on a charge of unauthorized use of a motor vehicle, and the propriety of inquiry as to that arrest is not in issue in this court. On the first trial, appellant's character witness was examined as to her awareness of the arrest, and it is not contended on this appeal that similar leeway at the second trial would have been improper. While the relationship between the unauthorized-use arrest and appellant's reputation for veracity is dubious, *fn39 inquiry concerning a character witness' knowledge of it might be an appropriate effort to test any representation of a good community reputation for peace and good order. *fn40 As to repute for that trait of character, appellant's arrest for unauthorized use of a vehicle bore a substantial degree of relevance, *fn41 particularly in view of the temporal proximity of that arrest to the offense on trial. *fn42 It is not apparent in the circumstances of this case that the balance of probative value and prejudice would so strongly favor exclusion of the inquiry as to require an overruling of the trial judge's discretion to permit it. *fn43
So much of the trial judge's ruling as authorized questioning as to the character witnesses' knowledge of the narcotic arrest -- the complaint on appeal -- stands on somewhat different ground, however. The questioning could not be justified by appellant's introduction of evidence of a good reputation for truth and veracity. *fn44 To be sure, this court has indulged impeachment of a testifying accused by a showing of his previous conviction of narcotic offenses, *fn45 but that rests "on the theory that a demonstrated willingness to commit some kinds of crime may well indicate a willingness to lie on the witness stand." *fn46 The same cannot be said for an arrest, for the guilt of the arrestee remains to be proved. "Arrest without more does not, in law any more than in reason, impeach the integrity or impair the credibility of a witness. It happens to the innocent as well as the guilty." *fn47 Consequently, as only recently we said, "evidence of arrests, without more, . . . may not be introduced for the purpose of testing the credibility of the defendant directly or indirectly through his character witness," *fn48 and that theme has pervaded our decisions over most of the past decade. *fn49 Not the least of our concern has been that any probative contribution the arrest might have made to credibility -- either of the accused or the witness -- was greatly outweighed by the prejudice it inflicted on the issue of guilt. *fn50
But justification, at least to some extent, might be found, however, in appellant's introduction of a good community reputation for peace and good order. *fn51 In Michelson v. United States, *fn52 character witnesses for the accused, on trial for bribery of a federal officer, testified that he had a good reputation for "honesty and truthfulness" and for "being a law-abiding citizen." On cross-examination, the prosecutor was permitted to ask whether they had heard that the accused had previously been arrested for receiving stolen goods. The Supreme Court upheld the propriety of the questioning. "It is not only by comparison with the crime on trial," said the Court, "but by comparison with the reputation asserted that a court may judge whether the prior arrest should be made subject of inquiry." *fn53 So, the Court added, "by this test the inquiry was permissible. It was proper cross-examination because reports of his arrest for receiving stolen goods, if admitted, would tend to weaken the assertion that he was known as an honest and law-abiding citizen. The cross-examination may take in as much ground as the testimony it is designed to verify." *fn54 The cases utilizing the same approach are legion, *fn55 and it is evident that the principles they espouse portend controlling significance here.
There is, however, one feature of the instant case which has no counterpart in Michelson. Appellant's narcotic arrest took place two weeks before the commencement of his second trial, but ten months after occurrence of the offense for which he was being tried. The point in time at which any character trait for peace and good order was relevant to the issues was the date of the offenses on trial, *fn56 and appellant's effort to establish it favorably would have been limited to a showing of his community reputation therefor at that time. *fn57 Since proof of a reputation at a given time may tend to indicate what the reputation at a later time is, *fn58 his character witnesses might have been allowed to testify as to a reputation existent during a period prior to and not remote from the offense date. *fn59 On the other hand, since the community's view of the accused's character could well be affected by the gossip which frequently follows on the heels of a criminal charge, his reputation in the community after the charge became publicized might not be a trustworthy index to his actual character. *fn60 For this reason, the courts have generally held that a reputation subsequent to publication of the charge on trial is not admissible in evidence. *fn61
A number of courts have also stated that cross-examination of good-character witnesses on their knowledge of events indicating misconduct on the accussed's part, like direct presentations on reputation themselves, should be limited to awareness of events which transpired before the time in issue. *fn62 That, in our view, is a wholesome admonition, and ordinarily should be observed. We realize that testimony tending to prove a reputation is substantive evidence, while character-witness cross-examination is a probe into credibility, but this difference in function does not make for a difference in admissibility. The cross-examination unavoidably brings the disparaging event to the jury's attention, *fn63 and the accused's best or only response may be his own good reputation. Since the time in issue is the cutoff point for the accused's presentation of good reputation evidence, *fn64 questions to character witnesses which unveil subsequent events may well embarrass the accused's ability to protect himself. The accused's predicament may be accentuated by the fact that even if his presentation is not so restricted, his reputation after occurrence of the offense on trial may not truly reflect his real character. *fn65 Should these disadvantages be heaped on the accused, the prejudice he suffers may overbalance the probative value to credibility of the character witness which the cross-examination otherwise could have.
While advertent to these dangers, we deem the problem under discussion one to be dealt with primarily by the trial judge. Some discretion in the matter is more in keeping with the broad latitude which judges have as to the admission of character testimony, *fn66 and which traditionally they have exercised over the scope of cross-examination, than is any inexorable rule on the subject. A measure of discretion imparts to the proceedings a desirable degree of flexibility to shape the judge's ruling to the "numerous and subtle considerations" *fn67 appearing. Not every situation calls for exclusion of questions exploring knowledge of events occurring after the time in issue. *fn68 Not every subsequent event is an unacceptable topic, *fn69 nor a topic so prejudicial as to outweigh its probative significance; *fn70 some events otherwise objectionable perhaps could be made unobjectionable. *fn71 A decision to permit inquiry respecting subsequent events should, of course, be reached cautiously, and only for the best of reasons. But in the final analysis the matter should be left to careful handling by the trial judge, subject to appellate correction only where mishandling is clear. *fn72
We are thus brought to the ultimate question in the case before us. Did the trial judge err in ruling that appellant's character witnesses, if produced, could be asked by the prosecutor about their knowledge of the narcotic arrest? While, as we have said, an issue over appellant's reputation for truth and veracity would not justify such an inquiry, *fn73 an issue over his reputation for peace and good order possibly could. *fn74 Thus the propriety of the cross-examination depended largely upon the particular approach which appellant's character presentation would take. If confined to a reputation for truth and veracity, the inquiry proposed by the prosecutor would be out of order. *fn75 So also it would be if the presentation were limited to a reputation for peace and good order existing at or prior to the date of the offenses *fn76 unless, after a weighing of all competing factors, there was a heavy overbalance in the Government's favor. *fn77 And even if appellant undertook to develop a reputation as a law-abiding citizen over a period embracing his narcotic arrest, the prosecutor's question could properly be allowed only if in the sound judgment of the trial judge the value of the answer solicited plainly outweighed the possibility of prejudice to the accused. *fn78
The discretion which trial judges are thus summoned to exercise is an informed discretion -- one which takes into account all relevant factors deserving of consideration. *fn79 The discretionary function is aborted if the judge acts without the information essential to comprehension and treatment of those factors. That, we find, is what happened here. When the judge ruled that appellant's good character witnesses could be questioned as to their awareness of the narcotic arrest, he was in the dark on pivotal aspects of the testimony which the witnesses would give. He did not know whether the witnesses would speak to appellant's reputation for peace and good order, or instead to his reputation for truth and veracity, or whether they would vouch for the quality of both. *fn80 He did not know -- if peace and good order were the testimonial topic -- whether the reputation would be attested as of the offense date or as of some later time. These facets of the expected testimony bore heavily on the limits of permissible cross-examination, *fn81 and a clear understanding of them was crucial to the ruling to be made, but the record makes it obvious that the judge ruled without that understanding. *fn82 Moreover, as we have said, the ruling necessitated a careful balance of probative value and prejudice *fn83 and there was no indication of such a process here.
We are mindful of the fact that at appellant's first trial his character witness testified to his good reputation for honesty, peace and good order, not simply as of the time of the offenses, but up to the time of trial. It may be that the trial judge, when he made the ruling under review, traded on recollection of that fact. But it was an evaluation of the current situation, and not a memory of the old, which was called for but not supplied here. *fn84 Appellant's character evidence at the second trial might have differed materially from that adduced at the first, and unless the judge knew just what it would be, there was no sound basis for measuring the permissibility of the prosecutor's question on the narcotic arrest. To paraphrase Michelson, *fn85 unless the judge has a grasp of how much ground has been -- or, in this case, would have been -- traversed by the offering on good character, he cannot define the ground which the cross-examiner may cover in attempting to discredit that testimony. *fn86 Here, with no information of the testimony anticipated from character witnesses for the defense, the trial judge announced that he would allow the prosecutor to ask about appellant's narcotic arrest. That was not an exercise of an informed discretion but, we hold, was error. IV
There is yet another aspect of this case demanding attention. Trial error does not occasion reversal unless it is harmful, and so a further question is whether an annulment of appellant's conviction is warranted. With the Government's case against appellant strong, it may be that, despite the erroneous handling of the matter of character witness cross-examination here, affirmance of the conviction is in order. To this facet of the case we now turn.
At about 2:30 on a September afternoon, Vernon E. Helvey and Ahmad Ardroudi were walking from the Safeway store at which they were employed to a bank to make an after-hours deposit. Helvey carried a canvas bag containing $2,540. As the couple reached a point directly in front of a restaurant, two men approached from the rear and staged the holdup from which this litigation emanated. One of the men pointed a gun at Helvey and demanded the money bag, and when Helvey handed it over to the gunman both bandits fled.
Two police officers, George W. Pickeral and James G. Croson, were seated inside the restaurant at the time, and through a picture window they both saw the incident as it unfolded. The officers dashed out of the restaurant and pursued the fleeing robbers, closing in for a time to a distance of 25 yards. During the chase both of the robbers disappeared from sight, one permanently and the other -- the gunman -- temporarily as he ran into an alley. The officers then separated, Officer Pickeral stopping at the entrance to the alley and Officer Croson getting into a taxicab whose driver said he could point out the gunman. *fn87 Very shortly thereafter, the officers rejoined, proceeded to a parked car about a block from the alley, and there they arrested appellant. The period elapsing from the robbery to the arrest was only "a few minutes."
The testimony of the four eyewitnesses to the holdup consistently described the bandit with the gun as wearing dark trousers, a dark jacket, a dark hat and square-rimmed sunglasses. When arrested, appellant wore dark trousers, with both knees torn out, and a white shirt, but neither a hat nor sunglasses. Immediately after the arrest, Officer Croson retraced the route which the flight of the robbers apparently had taken and in the process recovered the canvas money bag taken from Helvey, the sunglasses and hat worn by the gunman, and a pistol.
At the time of his arrest, appellant, by the officers' descriptions, was on his knees on the floor in the back of the car in which he had been seen to enter. He was partially concealed by clothing hanging in the rear of the car, and was out of breath and perspiring heavily. The car was occupied by Richard Rooks and his sister, visitors from New York who had lost their way. Rooks had stopped the vehicle and perceiving appellant running on the side of the street, called him over and asked for directions to Fourteenth Street. Appellant replied that he was going there and asked for a lift, so Rooks let him come in. The police officers arrived almost immediately and made the arrest.
The theses advanced by the defense were alibi and mistaken identification and what now follows is an outline of the events testified to by appellant. About noon on the date of the robbery he left home to get his car, which he lent to George Stokes on the day before. Informed by Stokes that he had turned the car over to Isaac Harris and Nathaniel Taylor, appellant set out to find it. At a poolroom he was told that the car was to be used in a "hustle" *fn88 and that he should retrieve it in a hurry. He resumed his trek and eventually arrived in the vicinity of the robbery, and spent fifteen minutes at a playground. Leaving the playground, he met Harris and Taylor and inquired about the car. They were in a hurry and told appellant to come with them, which for a time he did. While running along with them, appellant tripped and fell, tearing his trousers at the knees. When, however, appellant saw that they had a gun, he left them, and later came upon Rooks' car. Asked by Rooks for directions to Fourteenth Street, appellant got into the car and was promptly arrested.
A few of the details to which appellant testified were corroborated by other witnesses. Appellant's mother stated that he had left home that day wearing a light shirt but neither a hat nor sunglasses. Stokes confirmed the prior borrowing of appellant's car and its turnover to Harris and Taylor. Another defense witness recalled that appellant had been at the poolroom, and still another saw appellant at least as late as sometime between 1:30 and 2:00 p.m.
Undeniably, the Government's evidence at trial pointed convincingly to appellant's guilt. Appellant was identified as the gunman by each of the four eyewitnesses to the holdup. The two victims confronted the gunman face-to-face in broad daylight for an estimated half-minute and, at an eight-man lineup prior to trial, each picked appellant as the bandit with the gun. The two police officers saw the gunman on the sidewalk directly in front of the restaurant as they peered through the picture window from their seats 18 to 20 feet away. The in-trial identifications of appellant were positive; notwithstanding vigorous cross-examination of the four by defense counsel, none was shaken in his identification.
Appellant was arrested within a few minutes of the holup a very short distance away, and after a chase during which the gunman was out of the officers' view only briefly. The items incriminating the gunman -- the bag of money, the dark hat, the sunglasses and the gun itself -- had been discarded along the route which the chase had taken and within the compact area in which appellant admittedly had been. Appellant was seen running from the area by Richard Rooks as he sat in his stopped car, and when Rooks asked for directions to Fourteenth Street, appellant stated that he was going there and begged a ride. Both police officers, to some extent corroborated by Rooks, *fn89 testified that after they spotted appellant climbing into the car, they closed in and found him breathing heavily, perspiring freely, and crouched in the back of the car in an apparent attempt to hide. The entire episode -- from the holdup to appellant's apprehension -- consumed but a very few minutes.
The towering strength of the Government's case stood in sharp contrast with the relative feebleness of the evidence for the defense. Although in a few inconclusive details appellant's story was backed by other witnesses, it remained uncorroborated in its most vital aspects. Although appellant admittedly was in the vicinity of the robbery at the time it transpired, there was no witness to his activities within 30 to 60 minutes thereof. There were, moreover, unexplained oddities in appellant's version which could hardly escape notice. He professed to be in a great hurry to locate his car but he paused for a period of 10 to 15 minutes at a playground while on the way. Unquestionably he was coatless when arrested, although it had rained heavily somewhat earlier on that September day. *fn90 Notwithstanding the fact that he had already left Isaac Harris and Nathaniel Taylor, he was still running when Rooks spotted him a full block away.
The question at hand is whether the error in the trial judge's ruling that appellant's character witnesses might be asked about their knowledge of his narcotic arrest fatally tainted appellant's conviction. The answer depends upon "what effect the error had or reasonably may be taken to have had upon the jury's decision." *fn91 In dealing with that problem, we "must take account of what the error meant to them, not singled out or standing alone, but in relation to all else that happened." *fn92 And "if, when all is said and done, the conviction is sure that the error did not influence the jury, or had but very slight effect, the verdict and the judgment should stand. . . ." *fn93 Reversal of the conviction is warranted only "if one cannot say, with fair assurance, after pondering all that happened without stripping the erroneous action from the whole, that the judgment was not substantially swayed by the error." *fn94 In the milieu of this case -- the devastating strength of the Government's case and the serious weaknesses in the defense -- we feel that appellant's character testimony would likely have had little or no effect on the verdict. *fn95 In sum, we are with "fair assurance" led to the conclusion that the testimony would not have "substantially swayed" the jury. We hold, then, that the trial judge's error does not warrant reversal of the conviction.
Affirmed. IN AGREEMENT
FAHY, Senior Circuit Judge, concurring specially:
I agree there was error in the trial court's ruling as to the permissible scope of the cross-examination of character witnesses should they be called by the defense. I also agree that the error was harmless for the reasons set forth by Judge Robinson. It is unnecessary for me to go beyond this, though I express my appreciation of Judge Robinson's scholarly opinion.