3.102), but he referred to the person later identified as "Edwards" by an alias in the investigative reports he submitted in April, 1980. (N.T. 3.103). Between the time of the critical events and the trial, Lee had participated in some twenty-six or twenty-seven other investigations (N.T. 3.63-3.64), so a potential for confusion did exist. Even if these facts go to the weight of Lee's identification testimony and not its admissibility, Lee did impede defense counsel in testing that testimony by either returning the photographs to the file without record or destroying them.
But the Supreme Court has held that the corrupting effect of a suggestive identification is to be weighed against: (1) the opportunity of the witness to view the criminal at the time of the crime; (2) the witness' degree of attention; (3) the accuracy of his prior descriptions of the criminal; (4) the level of certainty demonstrated at the confrontation; and (5) the time between the trial and the confrontation. Manson v. Brathwaite, 432 U.S. 98, 114, 97 S. Ct. 2243, 2253, 53 L. Ed. 2d 140 (1977) (undercover narcotics agent's view of a single police photograph and subsequent incourt identification of defendant held to comport with due process).
As in Manson, we find that various indicators of Lee's ability and reliability ultimately outweigh any corrupting effect of the challenged identification. Lee encountered both defendants face-to-face in daylight for several hours over the course of three days. (N.T. 1.7-1.14). This exceeded the observation time of the undercover agent in Manson. Lee examined the police photographs within four days of these encounters. (N.T. 1.30, 1.46). Lee was certain that the photographs were those of the persons with whom he dealt. (N.T. 3.94, 1.31-1.32). Lee was not a lay eyewitness but a veteran of the police department for eleven years, approximately eight of which had been spent as a member of a drug enforcement task force. (N.T. 2.36-2.37). Lee viewed the photographs at his leisure and alone. (N.T. 1.17 et seq.). There was little pressure on Lee to "acquiesce in the suggestion that such a display entails." Manson, supra at 116, 97 S. Ct. at 2253. Under these circumstances, and despite his very confused testimony about the use of the photographs, Lee's identification testimony was admissible for consideration by the jury.
C. Jencks Act
This ground for a new trial is raised by Butts. At the time of the alleged offenses, Officer Lee was employed by the Philadelphia Police Department and assigned to the DEA. Lee testified that he made the "undercover" drug deals and Butts and Edwards were the men with whom he dealt. Officer Henry Cunningham ("Cunningham") was a member of the DEA surveillance team which monitored the illicit transactions during the days of March 25, 26, and 27, 1980. Lee prepared two handwritten investigative reports describing the transactions and Cunningham prepared one such report. (Ex. GS-1, 2, 3; N.T. 3.26, 4.4). These reports were prepared sometime after the events of March 25, 26, and 27, 1980 and submitted to a typist on April 11 and 15, 1980. (N.T. 3.29-3.31; 4.10). The officers destroyed their handwritten drafts after comparing them with the typed investigative reports ("DEA-6"). (N.T. 3.34, 4.5). This was their routine practice. (N.T. 3.34, 4.9). The typed reports were then submitted to a supervisor (N.T. 3.30, 4.5) but the supervisor did not (and could not) compare the rough drafts with the typed reports. (N.T. 4.9). The officers testified that their typed reports, which were made available to defense counsel, were identical in content to their handwritten drafts. (N.T. 3.27, 4.5). Lee stated that the only changes he made on the rough draft itself were spelling corrections and the like. (N.T. 3.35). Cunningham did not recall correcting his rough draft. (N.T. 4.16). The supervisor did not testify.
The Jencks Act, 18 U.S.C. § 3500,
requires the government to produce, on motion of defendant, any statement of a government witness which relates to the subject matter of that witness' testimony. In this Circuit it is clear that a handwritten draft of an investigative report by a DEA agent might be Jencks Act material and should at least be available to the court for inspection. United States v. Walden, 578 F.2d 966 (3d Cir. 1978). As a matter of judicial discretion, a rough draft need not be disclosed to the defense if it is substantially identical to the final report. Walden, supra at 971.
In Walden, rough drafts of DEA investigative reports were available at trial but defendant's motion for production was denied. On remand from the Court of Appeals to determine whether or not the handwritten drafts were statements within the meaning of the Jencks Act, the district court then found that the rough drafts had since been destroyed. The court held the drafts were statements, approved or adopted by the agent and thus subject to the requirements of the Jencks Act, because he knew when he sent them to his supervisor for review that they would thereafter be forwarded to a typist. United States v. Walden, 465 F. Supp. 255, 261 (E.D.Pa.1978), aff'd, 590 F.2d 85 (3d Cir.) (per curiam), cert. denied, 444 U.S. 849, 100 S. Ct. 99, 62 L. Ed. 2d 64 (1979). But the agent in Walden had the opportunity to make substantive revisions to the typed report, Walden, supra 578 F.2d at 970, so that his adoption and approval of the rough draft could not have been inferred from the finality of the report in the form sent to his supervisor. But see United States v. Bagnariol, 665 F.2d 877 (9th Cir. 1981) (citing United States v. Kaiser, 660 F.2d 724 (9th Cir. 1981)). Rather, the inference of adoption might have been drawn from the fact that the draft report "was at least in a form sufficiently acceptable to the agent that he allowed it to be reviewed by his superior." Walden, supra 578 F.2d at 970 (citing United States v. Carrasco, 537 F.2d 372 (9th Cir. 1976)). In Carrasco, the court held that a government witness' diary turned over to a DEA agent was a Jencks Act "statement" because a statement, unlike a diary remaining in the hands of its author, "seeks to transmit information from declarant to the reader." Carrasco, supra at 375.
In this case, Lee and Cunningham gave their drafts to a typist, compared the typed reports with their drafts, signed the typed reports, and then submitted the typed reports to their supervisor. Unlike Walden, their supervisor did not receive the rough drafts.
However, Lee gave Cunningham the rough draft of Lee's first report. (N.T. 4.12). Of the three reports written by Lee and Cunningham, it was only this first report that completely described Butts and Edwards (Ex. GS-1).
Cunningham admitted receiving that report but said that he thought Lee gave it to him to submit to the typist and that he did not read it; this cannot be the whole story. (N.T. 4.13; 3.140-3.141). Cunningham's report was submitted to the typist on April 11, 1980 (Ex. GS-3). That report refers to the descriptions of the defendants in Lee's first report (Ex. GS-1) which was also submitted to the typist on April 11, 1980. As Cunningham testified that his own handwritten and typed reports were identical in content, he must have read Lee's handwritten report because it was specifically incorporated in his own report. Lee failed to state why he gave his report to Cunningham. However, given their close and friendly relationship to which Lee did testify, it is permissible to infer and we find that Lee's handwritten draft of his April 11, 1980 report (Ex. GS-1) was at least in a form sufficiently acceptable to him that he allowed it to be reviewed by his surveillance team member who subsequently incorporated it by reference in his own report about the same events. It follows that this draft was adopted or approved by Lee and was a statement within the meaning of the Jencks Act. Indeed, the government "does not dispute" that all of the drafts were "statements" within the meaning of the Jencks Act.
Therefore, the issue is whether the destruction of this undisputed Jencks material constitutes mere harmless error. Butts contends that the officers' destruction of those rough drafts impaired his defense because he was deprived of potential impeachment material in a case where the testimony of Lee was critical. Pursuant to the statute, which provides for striking that witness' testimony or declaring a mistrial as a sanction for non-production, 18 U.S.C. § 3500(d), Butts moves for a new trial.
These sanctions are not mandatory. Courts of this Circuit have held Jencks Act violations to be harmless error in some circumstances. See, United States v. Walden, 465 F. Supp. 255 (E.D.Pa.1978), aff'd, 590 F.2d 85 (3d Cir.) (per curiam), cert. denied, 444 U.S. 849, 100 S. Ct. 99, 62 L. Ed. 2d 64 (1979); United States v. Vella, 562 F.2d 275 (3d Cir. 1977) (per curiam ), cert. denied, 434 U.S. 1074, 98 S. Ct. 1262, 55 L. Ed. 2d 779 (1978); United States v. Niederberger, 580 F.2d 63 (3d Cir.), cert. denied, 439 U.S. 980, 99 S. Ct. 567, 58 L. Ed. 2d 651 (1978); United States v. Harris, 560 F.2d 148 (3d Cir. 1977), cert. denied, 434 U.S. 986, 98 S. Ct. 614, 54 L. Ed. 2d 480 (1978); United States v. Deerfield Specialty Papers, Inc., 501 F. Supp. 796, 800 n.2 (E.D.Pa.1980); and United States v. Heard, 439 F. Supp. 172 (E.D.Pa.1977). However, we are mindful that the standard of post trial review set out by the Supreme Court in United States v. Goldberg, 425 U.S. 94, 111 n.21, 96 S. Ct. 1338, 1348 n.21, 47 L. Ed. 2d 603 (1976), is that "the harmless error doctrine must be strictly applied in Jencks Act cases."
In Walden, the district court held that despite the destruction of the rough drafts, non-production was harmless for three reasons. First, the defendant was provided with typewritten copies of the reports. Second, credible witnesses (the agent who prepared the report and his supervisor) testified that they had examined the rough drafts and the typewritten copies were identical in content to the drafts except for spelling and grammatical changes. Third, the agents' testimony evidenced no bad faith in the destruction of the documents. Walden, supra, 465 F. Supp. at 261.
This case is disturbing because it involves the destruction of an investigative report rough draft by a DEA agent
in this Circuit after Walden.
The draft may have been prepared as much as two weeks after the critical events and Lee testified that he wrote the draft from memory. (N.T. 3.36). The officers engaged in a "group discussion of what transpired in the street" with other members of the surveillance team prior to the submission of their reports to the typist. (N.T. 3.37-38, 3.138-3.141). Cunningham had possession of Lee's handwritten report before it was submitted to the typist. (N.T. 4.12). In these circumstances, we cannot place faith in the accuracy of Lee's testimony that his rough draft and final report were identical or that the rough draft was not substantially revised at some point. It is impossible to find harmless error, for the jury's verdict could not be based on testimony tried by effective cross-examination.
The government has failed to demonstrate that no prejudice resulted to the defendants. The safeguards which other courts have cited in denying sanctions for non-production do not exist in this case. Unlike Walden, no person other than the one whose testimony would be challenged by use of the Jencks materials testified to the content of the draft. The government submitted no mechanical reproductions. See, United States v. Meisch, 370 F.2d 768, 772 (3d Cir. 1966) (carbon copy). There was no independent co-extensive information as to its content. See, Deerfield, supra at 800 n.2 (other contemporaneous notes available). Without the testimony of the government's eye witness to the transactions, the government's evidence would have been far weaker. See, Niederberger, supra at 71 (ample evidence independent of witness' testimony). Lee's draft contained the only complete descriptions of the drug deal participants, which are not matters of little import. See, Heard, supra at 174 (notations of times of little import). Even where Jencks Act statements were made available, the Court of Appeals has upheld a lower court's decision to strike the testimony of the witness because the production was untimely. United States v. Jackson, 649 F.2d 967, 971-72 (3d Cir. 1981). Here, there was not an untimely production but no production at all, and there could be none because the draft reports were knowingly destroyed.
In Walden, Judge Hannum was influenced by the fact that subsequent to the trial DEA purportedly instituted a policy requiring the retention of drafts. Walden, supra 465 F. Supp. at 260. However, at this trial Cunningham testified that it was still routine practice to destroy drafts after the final reports were produced and that it was not office policy to show rough drafts to supervisors. Three years after Walden, the government seeks to excuse its agents' conduct by arguing that the DEA retention policy "is at best confusing." (N.T. 3.23).
These conflicting representations to the court resemble those regarding FBI practice recounted in Vella, where the court stated:
To avoid future misunderstandings, we specifically adopt the precepts announced in United States v. Harrison, 173 U.S.App.D.C. 260, 524 F.2d 421 (1975), as the law in this circuit, to-wit, the rough interview notes of F.B.I. agents should be kept and produced so that the trial court can determine whether the notes should be made available to the appellant under the rule of Brady v. Maryland, 373 U.S. 83, 83 S. Ct. 1194, 10 L. Ed. 2d 215 (1963), or the Jencks Act.