UNITED STATES
v.
WARD et al.
Nos. 9623, 9624.
Circuit Court of Appeals Third Circuit.
Argued June 21, 1948. Decided August 13, 1948.John M. Smith, Jr., and Henry Weiss, both of Philadelphia, Pa., for appellants.
Asher W. Schwartz, Sp. Asst. to Atty. Gen. (Gerald A. Gleeson, U. S. Atty., and Edward A. Kallick, Asst. U. S. Atty., both of Philadelphia, Pa., on the brief), for appellee.
Before BIGGS, McLAUGHLIN, and KALODNER, Circuit Judges.
KALODNER, Circuit Judge.
The defendants were found guilty in the court below upon a single indictment charging, in one count, conspiracy to defraud the United States, 18 U.S.C.A. § 88, and, in seventeen counts, presenting false claims to an agency of the United States, 18 U.S. C.A. § 80. A verdict was directed on two of the seventeen counts. The motions of the defendants for directed verdicts on all other counts, and later for a new trial, were denied. This appeal followed.
*461 The indictment included others, who, together with the defendants now appealing, formed the A. C. Welding Company (hereinafter referred to as "A. C."), a partnership, for the purpose of engaging in the welding business, and more particularly for the repair and maintenance of ships. The charges against them centered about the conduct of A. C. in its performance as a sub-contractor of the Pioneer Engineering Company (hereinafter referred to as "Pioneer"), a prime contractor with the War Shipping Administration on a "cost-plus" contract basis. The government sought to prove in this case that the defendants padded the payrolls of A. C. with the result that it received more than it was entitled to for the work actually done, to the detriment of the United States and in violation of the statutes already referred to.
One of the government's witnesses was one Leonard Johnson, a former agent of the Federal Bureau of Investigation, who had audited A. C.'s books. Whether Johnson was properly utilized as an expert, and whether his testimony was essentially a summary of the government's evidence or was erroneously permitted to the prejudice of the defendants, are, in our opinion, the primary questions on this appeal.
The payroll books of A. C. contained the name of each employee, his address and social security number, and the wages paid on a weekly basis. They did not list the particular days worked in a week, or the hours, or the names of the vessels on which the work was performed.
Johnson testified he disregarded the payroll records because he found them "consistently" inaccurate and unreliable. The latter conclusion, he said, was based upon (1) a comparison of the payroll records with invoices which had been submitted by A. C. to Pioneer, (2) the "general knowledge in the trade" that no ships were "in" at certain times for which work was recorded in the payrolls, and (3) information obtained from interviewing approximately 110 A. C. employees. Johnson, however, was not preliminarily qualified as such an expert who might properly give evidence based on a "general knowledge in the trade", and the records of the interviews with A. C.'s employees were not introduced in evidence.
The invoices which A. C. had submitted to Pioneer, who paid them and was later reimbursed under its "cost-plus" contract with the War Shipping Administration, contained the name of the vessel on which work covered by the invoice was performed, the number of men employed, the type of work, the aggregate amount of time spent, the rate of pay, and the total amount of the invoice. Attached to the invoice was a list of names of men who allegedly performed the work, classified according to their particular jobs.
Having disregarded the payroll records, Johnson proceeded to "summarize", in numbers of men and amounts of money, the invoices which were "unsupported". It is asserted by the government that this testimony was grounded upon a comparison of the invoices and attached lists with the testimony of the laborers named therein as given during the trial.
The defendants at the time did not dispute that the witness Johnson could properly compare the invoices with the testimony or other records in evidence in order to calculate for the jury the difference in dollars and cents. Cf. United States v. Johnson, 1943, 319 U.S. 503, 519, 63 S.Ct. 1233, 87 L.Ed. 1546. Indeed, the learned trial judge permitted the testimony on that reasoning and at the time Johnson was on the stand cautioned the jury as follows:
"Members of the jury, I think I should say to you now that the testimony of this witness, of course, as has been indicated, is predicated largely upon the testimony of these employees who were on the stand before. These calculations are made on that basis. It will be the determination in this last analysis as to whether or not those witnesses told the truth; and that is a matter for you to decide." (N. T., page 717)
However, it was contended in the court below, as it is here, that Johnson did not so limit his testimony and that in formulating his conclusions he relied on material dehors *462 the record so that he was, in effect, a "thirteenth juror" to the prejudice of the defendants.
We subscribe to the defendants' contention and are of the opinion that they are entitled to a new trial.
Johnson did more than merely disregard the payroll records. He went further. In comparing the invoices and attached lists with the testimony adduced in court, he made use of other materials not in evidence, by means of which he first determined the "reliability" of the witnesses. Sworn statements had been taken by the Federal Bureau of Investigation prior to the trial from many of A. C. employees, including those who appeared as witnesses. These statements were not part of the record of the case. Johnson sat throughout the trial with prepared cards on which had been summarized the previous statements, and as each witness testified his testimony was compared with the cards to ascertain his "reliability". Just how the differences were treated was not comprehensively explained, but "in the event that the testimony of the witness differed" from the sworn statement which had been summarized on the cards, Johnson "treated each case individually" and could not state "in broad terms" his method. After having thus selected those witnesses whom he considered "reliable", Johnson prepared a "schedule" which constituted the foundation for his "summarizing" testimony that for a given vessel A. C. had rendered invoices to Pioneer which were "unsupported" both as to a specific number of men and an ascertained amount of money.
It is readily apparent that Johnson, in his attempt to strengthen the government's case, did more than "summarize" the evidence. From materials and information extraneous to the record he confessedly determined the "accuracy" of evidence already in the case, and "eliminated" the "inaccurate" evidence as well as the testimony of witnesses regarded by him as "uncertain". This, of course, was an erroneous method of proof. For while it can scarcely be gainsaid that voluminous records may be summarized, or that an expert may interpret and explain certain evidence in the case and on the basis of such evidence perform tasks beyond the training and experience of the layman juror, it is nevertheless axiomatic that the "expert" may not go so far as to usurp the exclusive function of the jury to weigh the evidence and determine credibility. Nor may expertness be the medium for injecting into the record material and information not a part of the expert's qualifications and otherwise inadmissible through his lips. Apropos is the statement we made in United States v. Michener, 3 Cir., 1945, 152 F.2d 880, 883: "The admission into the evidence of the auditors' conclusions and opinions as distinguished from mere statements of fact summarizing what was observed by them in the records was erroneous. * * *" And it was in aggravation of the error for Johnson to base his determination of the value of the evidence in the case upon information which had been collected, and which could have been directly submitted to the jury for its appraisal.
The government maintains, nonetheless, that the defendants are not in a position to complain, since the net result of Johnson's calculations was to exclude "some government testimony from the record". That answer is demonstrably inadequate, and particularly so in view of the government's statement that "This practice was obviously designed to give the Court and the defense the benefit of the check which had previously been made by the Federal Bureau of Investigation in determining the reliability of the witness." (Government brief, page 6; emphasis supplied.)
The effect of Johnson's testimony was not simply to exclude "some government testimony". Rather, it affirmatively declared the unreliability of records and witnesses on a basis neither accessible to the jury, nor assessable by it. Moreover, when Johnson testified that the witnesses whose testimony he had used in his calculations were considered by him to be "reliable" because their testimony coincided with the sworn statements they had previously given *463 to the Federal Bureau of Investigation, he was once again performing a jury function while endowing those witnesses with an aura of cerdibility impenetrable by the jury. If the jury believed Johnson's testimony that such statements had been taken and that they corroborated the witnesses used by him in his schedule, then manifestly the jury was furnished an additional basis, incompetently rendered and unwarranted in the record, for determining the credibility of the evidence it was required to consider. Johnson may have affected the government's case by attributing unreliability to some of its own witnesses. But that was certainly accomplished at the expense of the defendants by attributing to other government witnesses a definite reliability premised upon alleged prior consistent statements not in evidence, and upon which was superimposed Johnson's "individual treatment" not explainable "in broad terms".
We are of the opinion, therefore, that Johnson's testimony was erroneously permitted and that it was prejudicial to the defendants. We think, also, that the words of caution to the jury, quoted above, were inadequate to the situation. Purportedly reserving to the jury its right to determine the issue of credibility, the learned trial judge failed to take into consideration the illegitimate premise for resolving that issue which Johnson had supplied. In this connection, it may be noted that while in the charge to the jury Johnson was characterized as an expert and general instructions were given on expert testimony, nothing was said, or indeed could have been said, in correction of the prejudicial error committed.
Other assignments of error by the defendants are, in our opinion, without merit, and in view of the result here reached need not be discussed.
For the reasons stated, the judgments of the court below will be reversed and the cause remanded with instructions to grant a new trial.