D.C. Circuit Cautions On The "Serious Risk" Of Using Summary Witnesses

In drug and RICO conspiracy trial, circuit notes as "problematic" the use of a law enforcement "overview witness," reiterates previous warning of the "serious risk" of presenting such evidence and joins other circuits in "condemning the practice"; circuit outlines narrow circumstances when the prosecution may offer law enforcement testimony where the witness "could properly describe, based on his personal knowledge, how the ... investigation ... was initiated, what law enforcement entities were involved, and what investigative techniques were used," in United States v. Moore, 651 F.3d 30 (D.C. Cir. July 29, 2011) (No. 05-3050, Consolidated with Nos. 05-3051, 05-3052, 05-3053, 05-3054, 05-3064)

In the spring, the Federal Evidence Review, noted that the D.C. Circuit had weighed in with other circuits questioning the use of "overview" testimony by law enforcement witnesses. see D.C. Circuit Notes Limits On Overview Testimony. The blog noted that in United States v. Smith, 651 F.3d 30 (D.C. Cir. April 15, 2011) (No. 09-3119), the D.C. Circuit assumed, without deciding, that "[h]earsay does not become admissible merely because it is provided by a government agent in the form of an overview of the evidence." Because the law enforcement official's testimony in that case did not depend upon inadmissible hearsay and as the testimony was based on what the agent "himself heard when listening to thousands of intercepted conversations," the testimony was not erroneously admitted and in any event, if erroneous, it would be harmless. In a recent action, the D.C. Circuit has again addressed its discomfort with overview testimony and outlined the limits of its use. The circuit warned prosecutors that use of a law enforcement witness to present "an overview of the government's case-in-chief runs the serious risk" of violating the rules of evidence as well as "the high standard set" for their conduct.

In the case, six defendants were convicted in a complex drug-RICO conspiracy in their operation of an ongoing drug distribution business in the nation's capital. They challenged this result of their ten month jury trial, arguing in part that the trial court had mishandled a variety of evidence issues. While many of the issues raised were easily disposed of, there was one evidentiary issue that the circuit noted was "[m]ore problematic." This was:

FBI Agent Daniel Sparks testified as the first witness in the government's case-in-chief. His testimony provided an overview of the government's case, setting forth for the jury the script of the testimony and evidence the jury could expect the government to present in its case-in-chief. Further, he expressed his opinion, based on his training and experience, about the nature of the investigation conducted in this case.
Moore, 651 F.3d at 54-55.

The circuit noted that only recently had the "appropriateness of a government overview witness at the outset of its case" arisen as an important evidence issue in the circuit. Despite the currency of the issue, however, the circuit referred to a 1983 case as an "instructive" analysis of the use of summarization of evidence" by a non-expert witness called as part of the government's case-in-chief. In this earlier case, United States v. Lemire, 720 F.2d 1327, 1348 (D.C.Cir. 1983) "the government called toward the end of its case-in-chief an FBI agent, who was also a certified public accountant, 'to summarize the evidence about the complex cash flow through offshore companies' in a prosecution" for fraud and conspiracy. The circuit noted that at the time it approved the summary witness under FRE 602 and noting the propriety of FRE 1006 as well as under FRE 403. The three lessons indicated by the case were "First, the jury might treat the summary evidence as additional or corroborative evidence that unfairly strengthens the government's case.... Second, summary witness testimony posed the risk that otherwise inadmissible evidence might be introduced.... Third, a summary witness might permit the government to have an extra closing argument." Moore, 651 F.3d at 56 (citing United States v. Lemire, 720 F.2d 1327, 1348-50 (D.C. Cir. 1983)).

All three of these concerns were at issue in Moore. In addition, since 1983 "[o]ther circuits to address the use of overview witnesses have reached uniformly negative conclusions in view of the serious dangers of prejudice to a fair trial. The Court of Appeals for the First, Second, and Fifth Circuits have held that the use of overview testimony by the government is a “troubling development” for this very reason. United States v. Casas, 356 F.3d 104, 120 (1st Cir. 2004); see also United States v. Garcia, 413 F.3d 201 (2d Cir. 2005); United States v. Grifin, 324 F.3d 330 (5th Cir. 2003)." Moore, 651 F.3d at 56-57.

The D.C. Circuit noted the various "approach[es]" these other circuits took towards their disapproval of summary witnesses. The Circuit then articulated a more direct condemnation of the practice, examining "FBI Agent Sparks's overview testimony closely, aware that there was no voir dire before his testimony and a limiting instruction was given to the jury only after he completed his testimony, and then only with regard to opinions, not otherwise described, that he may have offered while testifying." The case was infected by "[a]ll three dangers identified by this court in Lemire."

As summarized by the circuit:

In sum, FBI Agent Sparks's testimony was improper in offering his non-expert opinions about the charged conspiracy and appellants, vouching for the reliability of the investigation and of the cooperating co-conspirator witnesses the government planned to have testify at trial, and discussing evidence that had yet to be introduced. Given the dearth of taped conversations and videotaped evidence—none as to Moore—and almost exclusive reliance on co-conspirator cooperators' testimony, the government understandably might seek at the outset to enhance its case in the jury's mind with the imprimatur of an FBI agent. But the prosecutor went too far in questioning, allowing FBI Agent Sparks to act as an expert witness with respect to gang investigations and to refer to evidence that would never be introduced at trial. The district court, in turn, failed to sustain appropriate defense objections to FBI Agent Sparks's testimony that purported to offer opinion testimony and to confirm government evidence that had yet to be introduced.
Moore, 651 F.3d at 60.

In light of these problems the circuit articulated a more vigorous warning about the use of overview testimony:

Because a witness presenting an overview of the government's case-in-chief runs the serious risk of permitting the government to impermissibly “paint a picture of guilt before the evidence has been introduced,” and may never be introduced, we join the circuits that have addressed the issue in condemning the practice. The use of overview witnesses exacerbates the “obvious dangers” this court identified in Lemire in the use of non-expert summarization evidence. Overview testimony offers an opportunity to “poison the jury's mind against the defendant or to recite items of highly questionable evidence.” Avoidance of those dangers is largely beyond the ability of the district court, much less the defense, to prevent. As the record here demonstrates, a trained law enforcement officer is likely to go as far as the questions allow, presenting a picture for the jury of a solid prosecution case based on his opinion of the strength and credibility of the witnesses the government plans to call to testify at trial for reasons made persuasive in view of the officer's training and experience. See, e.g., May 15, 2002 AM Trial Tr. at 15–16 (prosecutor asking FBI Agent Sparks why truthful information is important). After-the-fact limiting instructions can, at best, mitigate prejudice, rather than invariably eliminate its effects completely. The view of the government's case has been implanted in the mind of the jury by an agent of the Federal Bureau of Investigation who worked on the case—he should know!
Moore, 651 F.3d at 60 (citations omitted).

The circuit noted that there still might be an appropriate rule for such summary evidence. However, its presentation was subject to a variety of important restrictions in the circuit:

The government remains free to call as its first witness a law enforcement officer who is familiar with the pre-indictment investigation or was otherwise personally involved, where permissible under the Rules of Evidence and consistent with constitutional guarantees. Such a witness may, for example, be able to provide relevant background information as to the investigation's duration and scope or the methods of surveillance, based on personal knowledge. Put another way, a law enforcement officer may “describe a complicated government program in terms that do not address witness credibility,” but he may not offer “tendentious testimony.” Thus, FBI Agent Sparks could properly describe, based on his personal knowledge, how the gang investigation in this case was initiated, what law enforcement entities were involved, and what investigative techniques were used. What he could not do was present lay opinion testimony about investigative techniques in general and opine on what generally works and what does not, as illustrated by informants who pled guilty. Neither could he anticipate evidence that the government would hope to introduce at trial about the charged offenses or express an opinion, directly or indirectly, about the strength of that evidence or the credibility of any of the government's potential witnesses, including the cooperating co-conspirators.
Moore, 651 F.3d at 60 (citing Old Chief v. United States, 519 U.S. 172, 186–88 (1997); United States v. Curtis, 481 F.3d 836, 838 (D.C. Cir. 2007) (same), United States v. Flores–de–Jesús, 569 F.3d 8, 19 (1st Cir. 2009) (“Most of [the challenged testimony] was based on inadmissible hearsay.”), United States v. Grifin, 324 F.3d 330, 349 (5th Cir. 2003) (“Martin admitted that his statement ... was not based on personal knowledge but on what someone told him”)).

Despite the error in admitting the summary testimony, the circuit found that the error was "close," but ultimately harmless. This was in light of a trial record in which the witness' improper testimony was "later confirmed by admissible evidence at trial," the limited nature of the defense (e.g., limited to cross-examining testifying cooperating co-conspirators and other government witnesses), the trial judge's limiting instruction after the witness' testimony to "disregard any opinion testimony he offered," and also because the evidence of the defendants' guilt was "overwhelming." Moore, 651 F.3d at 61. The circuit advised that the problem of the law enforcement summary witness could have been easily and effectively avoided:

The inauspicious beginning of the government's case-in-chief is easily avoided in the future. No less than the court, the Department of Justice recognizes the high standard set for the prosecution by the Supreme Court.... Similarly, this court's long-held view of the purpose of the opening statement to the jury, namely to allow the prosecutor the opportunity to provide the jury with an objective overview of the evidence that the government intends to introduce at trial, has long afforded the prosecutor the opportunity to do that for which the prosecutor improperly used FBI Agent Sparks. This court now having made clear the exacerbated “obvious dangers” of the overview witness testifying about evidence yet to be admitted before the jury affords all parties clear direction to avoid unnecessary risks —for the prosecutor of an overturned conviction, for the defense of an unfair trial, and for the district court of having to retry a case.
Moore, 651 F.3d at 61 (emphasis added; citations omitted).


Federal Rules of Evidence