Frank Guglielmini and John Testa appeal their convictions by a jury after trial in the Eastern District of New York in November 1966 on three counts for concealing assets of a bankrupt, and of Guglielmini on an additional count for concealing records, in violation of 18 U.S.C. 152, for which the district court imposed concurrent five-year prison terms and five years' probation to follow service of sentence.

The appellants allege that they were denied a fair trial by the conduct of the trial judge and the prosecutor, and they complain of errors in the court's charge. Our study of the record shows substantial support for their claims. As we find that the cumulative effect of these errors was such as to render it highly doubtful that the appellants enjoyed the fair trial to which they were entitled, we reverse the convictions.

Simply stated, the government's case was that Frank Guglielmini and John Testa ran the business of the Miracle Supermarket, owned by Frank Guglielmini's brother, John Guglielmini, for the purpose of defrauding creditors and concealing the assets and books of the business from the receiver in bankruptcy.

The indictment charged that on April 1, 1960, an involuntary petition in bankruptcy was filed against Frank Guglielmini's brother, John-- also known as John Papa and doing business as the Miracle Supermarket in Brooklyn, New York,-- who thereafter on June 6, 1960 filed a voluntary petition in bankruptcy under Chapter XI of the Bankruptcy Act and was adjudicated a bankrupt on September 28, 1960. It was charged that commencing about February 1, 1960, Frank Guglielmini and John Testa were engaged in the management of the business, but that only a part of the assets were made available to the receiver, although the bankrupt had substantial property, largely foodstuffs and cash receipts from sales, and that Frank Guglielmini and John Testa knowingly and fraudulently concealed from the receiver property and cash receipts of approximately $39,000.

A second count charged knowing and fraudulent concealment of the same property from the creditors of the bankrupt, and a third count charged that in contemplation of a bankruptcy proceeding, and with intent to defeat the bankruptcy laws, the defendants knowingly and fraudulently transferred and concealed the same property. A fourth count charged that the two defendants, after the commencement of the bankruptcy proceeding, knowingly and fraudulently concealed the books, records, and documents of the bankrupt from the receiver.

The proof amply supported the verdicts of guilty. In summary, the government showed that the defendants ordered approximately $48,000 worth of groceries, meat and fowl in March 1960, and that less than half of it was in the bankrupt's inventory on April 2, 1960 or accounted for by bank deposits in March. Meanwhile merchandise was sold, below cost, for cash which was not deposited in the business. Cf. United States v. Olweiss, 138 F.2d 798 (2 Cir. 1943), cert. denied, 321 U.S. 744, 64 S. Ct. 483, 88 L. Ed. 1047 (1944). In addition, there was evidence that the books of account were given to Frank Guglielmini prior to the bankruptcy and that they were never produced.

The appellants complain that during the ten days of trial the trial judge by his 'reproachful, acrimonious and apparently baiting criticism' of Mr. Lewis, counsel for Frank Guglielmini, denied the defendants the effective assistance of counsel and conveyed to the jury the impression that he, the judge, 'entertained feelings of suspicion and hostility towards the defense.'

The record contains numerous instances of repartee between the judge and defense counsel. Most of this was wholly unnecessary and much of it could only have served to demean counsel and cast an unfavorable light on the defense. For example, the court, on the third and fourth days of the trial, brouhgt out in the presence of the jury that the Assistant United States Attorney was trying his first criminal case and that he was a 'young boy' compared to Mr. Lewis, to whom the court said (R. 502-03):

'The Court: You must remember that you have tried criminal cases for years and years and years, is that right? And both of us have some experience in trying cases for many years, and we are much older than this young boy who is trying this case.

'Mr. Lewis: But he is very competent young boy who may be better than a decrepit old man.

'The Court: For this first criminal case he is trying--

'Mr. Lewis: He's tried many civil cases and has been here many years.

'The Court: I said for the first criminal case he is trying we can't be too critical, but when it comes to some of the jargon, he may not be as adept as those with more experience * * *.'

Later, during a conference at which the jury was not present, counsel moved for a mistrial because of the court's remarks.

The court's remarks were wholly unnecessary. While the trial judge undoubtedly thought he was merely redressing the balance between youth and experience, the obvious effect of such colloquies was to place the defense at a disadvantage in the eyes of the jury, as it cast the prosecutor in the role of a young neophyte David contesting against a practiced Goliath.

These were not the only unnecessary and disparaging remarks by the court. For example, after a question asked by Mr. Lewis had been objected to as argumentative, the trial judge said, 'Your time to sum up, Mr. Lewis, will come later on' (R. 111). This remark was repeated on numerous occasions during the trial. Again, these remarks can only have served to discredit the defense in the eyes of the jury. To the same effect was the judge's discussion, in the presence of the jury, of the Jencks Act, 18 U.S.C. 3500, and the type of material that is producible thereunder (R. 285-290, 508-510). This discussion also discredited the defense and may have led the jury to believe that the defense was making demands which improperly imposed upon the government.

The trial judge's frequent participation in the trial, by questions and comments, also tended to give the jury the impression that he credited the prosecution and disbelieved the defense. This was particularly noticeable in his frequent and sometimes lengthy interruption of Frank Guglielmini's testimony. There are few pages of this defendant's testimony, which runs from page 971 to page 1146 of the record, which are free of some question by the court, and there are numerous instances where the court took over the cross-examination from the prosecutor for extended periods. While the trial judge may, and indeed should, take an active part in the trial where necessary to clarigy evidence and assist the jury, the persistent questioning the trial judge conducted in this case, together with his comments to defense counsel, conveyed to the jury far too strong an impression of his belief in the defendants' guilt. See, e.g., United States v. Persico, 305 F.2d 534 (2 Cir. 1962); United States v. DeSisto, 289 F.2d 833 (2 Cir. 1961).

Few claims are more difficult to resolve than the claim that the trial judge, presiding over a jury trial, has thrown his weight in favor of one side to such an extent that it cannot be said that the trial has been a fair one. Where there is any substance to such a claim the reviewing court must examine the entire record and attempt to determine whether the conduct of the trial has been such that the jurors have been impressed with the trial judge's partiality to one side to the point that this became a factor in the determination of the jury. This we have done. A reading of the record leaves us with the firm impression that the defendants did not receive the fair trial to which our law entitles them.

Frank Guglielmini also complains of the prosecutor's references to his connection with another bankruptcy fraud as unnecessary and prejudicial. We agree. Guglielmini, having taken the stand in his own behalf, was cross-examined regarding his prior employment. When he could not recall where he worked in 1958, the prosecutor asked him: 'Well, maybe I can refresh your recollection. Did you testify in this court in a bankruptcy fraud trial--' Counsel's immediate objection was overruled by the court, and the prosecutor brought out the name of the defendant at that trial, one Sal Bilello, that it involved the Top Sail Supermarket, that the defendant had testified at the prior trial, and that as meat manager in that market he had done the 'same things that you are doing now' (R. 1089-90).

It was error for the prosecutor to link the defendant Guglielmini to a prior bankruptcy fraud trial when its only relevance was to bring out what he had been doint in 1958. This evidence was not, of course, admissible under the rule that prior criminal acts of a defendant may, if sufficiently similar, be proved as some evidence of his guilty knowledge or intent in committing the act charged. Compare, e.g., United States v. Klein, 340 F.2d 547 (2 Cir.), cert. denied, 382 U.S. 850, 86 S. Ct. 97, 15 L. Ed. 2d 89 (1965); United States v. Ross, 321 F.2d 61, 67 (2 Cir.), cert. denied, 375 U.S. 894, 84 S. Ct. 170, 11 L. Ed. 2d 123 (1963). The purpose in putting the question could only have been to point out to the jury that the defendant was well acquainted with bankruptcy frauds. The trial judge erred in overruling counsel's prompt objection.

Moreover, the prosecutor later compounded his improper questioning by referring to the incident in his summation, in these words (R. 1291-92):

'Now, he (Frank Guglielmini) testified he was the meat manager for a Sal Bilello in March of 1958. Once again before Easter. He testified in that trial as a government witness. He was the meat manager. Is that where he learned this scheme? Is that where he learned about building up the business and using the bankruptcy laws?'

Again counsel's timely objection and motion for mistrial were overruled. The trial judge erred in not sustaining the objection and directing the jury to disregard the prosecutor's improper and unfair argument. As the principal issue in the case was the intent of the defendants, it is clear that permitting the intrusion of this irrelevant matter, which involved nothing more than Frank Guglielmini's employment by someone who was later tried for bankruptcy fraud, was a substantial error. The strong inference of guilty knowledge which the prosecutor's argument communicated to the jury could well have been a factor in their finding Guglielmini guilty. Compare, e.g., United States v. Tomaiolo, 249 F.2d 683, 688-690 (2 Cir. 1957); Nigro v. United States, 117 F.2d 624, 631-632, 133 A.L.R. 1128 (8 Cir. 1941).

The appellants complain of the court's instructions to the jury regarding reasonable doubt. The court pointed out the presumption of innocence and the fact that it persists until 'overcome by evidence which satisfies you beyond a reasonable doubt as to the defendant's guilt.' The court further stated that the defendant did not have to prove his innocence, and that the burden was on the government 'to prove beyond a reasonable doubt each essential element necessary to constitute the crime charged.'

'Now, what is reasonable doubt? To justify you in returning a verdict of guilty against either one of these defendants on any one of these counts, the evidence that you believe must be of such a character as to satisfy your judgment to the exclusion of every reasonable doubt. I underline reasonable doubt so that if you can reconcile the evidence with any reasonable hypothesis consistent with a defendant's innocence, then it is your duty to do so, and in that case find that defendant not guilty under that count of the indictment. In fact, if you find a probability of innocence, reasonably, then you have a reasonable doubt. I didn't say positively. I said probability. So if after weighing all of the proofs and looking only to the evidence, the proofs you impartially and honestly entertain the belief that the defendants or either of them may be innocent of any one of the offenses charged in this indictment, then as to that indictment, that defendant is entitled to the benefit of that doubt, if you have that kind of a doubt, and then as to that count, that defendant should be acquitted.'

While it may seem that a juror listening to the whole of the judge's instructions would understand the government's burden and the factors to be considered in assessing the reasonableness or unreasonableness of any doubt, we think the judge's statement that a 'probability of innocence' constitutes reasonable doubt may have confused one or more jurors. Putting the test in terms of a 'probability of innocence' implies that a reasonable doubt exists only when the jury finds it more likely than not that the defendants are innocent. This would appear to put the burden of proving their innocence on the defendants or at least to reduce the prosecution's burden from 'beyond a reasonable doubt' to merely a preponderance of the evidence. It is sufficient for a finding of 'reasonable doubt' that there exist an honest, substantial misgiving founded on the evidence and not on mere speculation; that misgiving need not be so strong as to lead to the belief that innocence is more likely than not.

It may well be that appellate courts have come to place too much reliance on the use of certain words in instructing juries about reasonable doubt. But we do not know how experienced and sophisticated each of the jurors may be in these matters; and we can never know just what words or thoughts are remembered and become the guides which weigh most heavily in the jury's deliberations. It is quite possible that one or more jurors may have carried with them into the jury room the idea that the test is the probability of innocence and that they concluded that, in the absence of such probability, there was no reasonable doubt and it followed that guilt was established. As we cannot know to what extent this confusing element of 'probability of innocence' may or may not have been corrected by the remainder of the judge's charge on reasonable doubt, we hold that such instruction was error. Cf. United States v. Persico, 349 F.2d 6 (2 Cir. 1965).

We have examined the numerous other claims of error, but find that they do not merit discussion.

We do not hold that any one of the errors committed during the trial would have required reversal of the convictions. We do hold that, occurring at the same trial, the total effect of the errors we have found was to cast such a serious doubt on the fairness of the trial that the convictions must be reversed.

The majority concede that 'The proof amply supported the verdicts of guilty.' Thus, in theory at least, the defendants were convicted by a jury of their peers upon ample proof-- the accepted American system for meting out justice for the commission of crimes. Why then the reversal? It is because the majority decide that in appellate retrospect the trial game has not been played with the niceties which they prefer should have been observed during the contest. In my opinion, the result reached and the reasons here stated have the effect of giving judicial sanction to trial tactics of skillful defense counsel-- additions to the bag of legal tricks already possessed-- which remove the trial of criminal cases further and further from the search for the truth (as proclaimed from the platform) and reduce it (in the arena where the contest is actually held) to an endeavor to strive for, and even create, reversible error. These tactics have been the 'hallmark' of many of successful defense advocate for generations. Why we now bestow upon such tactics reversal status, I cannot understand.

In this specific situation, as gleaned from the pages of a printed record and from our own knowledge of the participants, which cannot and should not be erased so long as we are aware of personalities in our own circuit, for two and one-half weeks the jury, the judge, the prosecutor and defense counsel had more than an ample opportunity to measure the mannerisms and characteristics each of the other. First, the jury: observing the entire scene from its jury box, entrusted with the ultimate resolution of the facts based upon their individual conclusions and the impressions made (never to be revealed) by every happening in the courtroom. The judge: every judge has his own individual manner of conducting a trial and, of necessity, must have great leeway in so doing. Even appellate decisions cannot reduce the judicial personality to a common colorless denominator. In this particular case, the judge, as has been his wont during his tenure in office, followed the case closely and took an active part therein. However, whatever he did or said in the presence of the jury was undoubtedly weighed by the jury along with all other factors in evaluating his personality. And as long as the jury system endures, it is premised on their strict adherence to the admonition, here properly given in the charge, that: 'You are not to construe that (questions by the judge) at all as indicating that the Judge has any ideas about who should prevail in this case at all. It is done for the purpose where the Court thinks it appropriate to make something clear, and that is all.' The prosecutor: it scarcely required characterization of the prosecutor as a 'young boy' to reveal this fact to the jury. Judges may well take judicial notice of the fact that juries can distinguish between young and old. The majority stress the fact that 'The court's remarks (youth and first criminal trial) were wholly unnecessary.' The majority then step (figuratively) into the jury box and conclude, with more assurance than I can muster, that 'the obvious effect of such colloquies was to place the defense at a disadvantage in the eyes of the jury, as it cast the prosecutor in the role of a young neophyte David, contesting against a practiced Goliath.' I would doubt that even expert psychologists would say that the jury's collective mind was so diverted from the ample proof relevant to guilt that this comment by the trial judge tipped the scales against the defendants. If a principle of law is to evolve that any case tried by a young man as his first criminal case calls for reversal, young prosecutors will indeed have difficulty breaking into the field. Furthermore, it must be remembered that David had certain assistance beyond the powers of his stripling body1 even as the prosecutor here had such assistance as the righteousness of the facts-- at least in the eyes of the jury-- afforded. The defense counsel: experienced, resourceful, rarely second-best in courtroom repartee, a Goliath not even stunned by the prosecutor David's slingshot stone, capitalized well on his contretemps with the trial court-- witness his success with the majority here who apparently believe that a trial judge's efforts to restrain counsel from making jury speeches during questioning or while making an objection should be discouraged. Yet the practice varies widely: many a trial judge will not permit counsel any leeway whatsoever in their forensic attempts to reach the jury by these means while others prefer to have grounds stated.

The incident involving Frank Guglielmini's former employment might well have been omitted, but to say that it 'could well have been a factor in their (the jury's) finding Guglielmini guilty' scarcely is consistent with the majority's opinion of ample proof to sustan the conviction. If their conclusion had been that the facts establishing guilt were so weak or inconclusive that such testimony might have tilted the scale, the argument would have greater applicability-- but their own conclusion is to the contrary.

This leaves as the only point of possible substance the 'reasonable doubt' charge. Some day the courts may recognize the utter impossibility of conveying metaphysical concepts to a jury and approve a simple charge, if charge be needed at all, such as 'reasonable doubt is a doubt based on reason.' Such simplicity, of course, might well be anathema to the Law-- hence, clarity (or probably greater confusion) is supplied by such cliches as 'not a whimsical doubt,' 'not a capricious doubt,' 'the kind of doubt a juror would have in his everyday affairs' and all the other valiant efforts trial judges make to accomplish the impossible. Appellate courts also take refuge in cliches in saying that reading the charge as a whole, the jury obtained a general idea of what they were supposed to do or, if they disagree with the jury, by holding that the jury was obviously misled. Even reading the portions of charge quoted by the majority (and the charge as a whole), there can be no 'reasonable doubt' that the jury obtained a fair idea of their responsibility-- in fact, the 'probability of innocence' comment may well have added to, rather than subtracted from, the prosecution's burden.

However, in reading the charge as a whole, the emphasis placed by the trial judge on reasonable doubt was more than protective of appellants' interests. First, the presumption of innocence 'goes with the defendant from the beginning to the end of the trial and prevails until, if that happens, unless and until that presumption is overcome by evidence which satisfies you beyond a reasonable doubt as to the defendant's guilt' and, further 'you don't convict unless you believe that either one of these defendants on any count is guilty beyond a reasonable doubt.' The judge then elaborated on 'reasonable doubt' as is custmoary in practivally all 'reasonable doubt' charges, for example, 'reasonable doubt does not mean speculative notions or possibilities resting upon mere conjecture, mere suspicion, mere guesswork, mere surmise, not arising or deducible from the proof or that want of proof'; it is 'an honest, substantial misgiving generated by the evidence, or want of evidence' and then the court gave the rather standard admonition that their doubt should be based upon 'an abiding conviction as you would be willing to act upon as to the most weighty and important affairs of your own life,' and that then (i.e., when free from reasonable doubt) and 'only then' would the jury be justified in convicting. In my opinion, the charge conveyed the substance of reasonable doubt as well as is humanly possible. The addition of the hundreds of synonyms to be found in Roget's Thesaurus under 'reasonable' and 'doubt' would not have added to the definition which the trial judge was endeavoring to convey.

Finally, the reading of this 1,414-page record leads me to the conclusion that it would be quite presumptuous to hold that the jury came to its determination of guilt because it was 'impressed with the trial judge's partiality to one side * * *.' Nor do I believe that appellate courts should speculate as to the basis for the jury's conclusions and then act on the assumption that they know. If courts on appeal inject this unknown element into their decisions, then trial judges will have to become mere figureheads. They will participate in the trial at their peril. Knowing this, astute defense counsel will have still another avenue, along with possible reversible error can be found because that which succeeds here today should succeed in the future.

I would affirm.