59 F.3d 274
42 Fed. R. Evid. Serv. 838
UNITED STATES, Appellee,
Nicholas BRIEN, Defendant, Appellant.
United States Court of Appeals,
Heard June 6, 1995.
Decided July 11, 1995.
James L. Sultan with whom Rankin & Sultan, Boston, MA, was on brief, for appellant.
Christopher F. Bator, Asst. U.S. Atty., with whom Donald K. Stern, U.S. Atty., Boston, MA, was on brief, for appellee.
Before BOUDIN, Circuit Judge, LYNCH, Circuit Judge, and SCHWARZER,* Senior District Judge.
BOUDIN, Circuit Judge.
On August 31, 1993, a grand jury indicted Nicholas Brien for armed bank robbery, 18 U.S.C. Secs. 2113(a), (d), and carrying a firearm during a crime of violence, id. Sec. 924(c)(1). In April 1994, a jury convicted Brien of the former offense, and he was later sentenced to 204 months in prison. He now appeals, raising important issues concerning (1) expert evidence on eyewitness identification and (2) courtroom identification procedure. As background, we briefly summarize the evidence and do so in the light most favorable to the government. United States v. Torres-Maldonado, 14 F.3d 95, 100 (1st Cir.), cert. denied, --- U.S. ----, 115 S.Ct. 193, 130 L.Ed.2d 125 (1994).
According to the government's evidence, Brien and an accomplice entered the Family Bank in Dracut, Massachusetts, on June 3, 1993. Brien, armed with a gun, confronted three tellers and collected money from two of them. He and his accomplice then fled with over $4,000 in a bag containing (unknown to them) an explodable red dye pack. Before and after the robbery, Brien stayed for several days with his girlfriend at the Avalon Motel in Saugus, Massachusetts. There, on June 4, an employee found some of the stolen money--identified by serial numbers and red dye--in a trash barrel outside Brien's room.
Brien was caught on July 15, 1993. On August 6, 1993, Brien was identified from a photo array by three bank tellers, including two of those whom he had directly confronted. All three testified to this effect at trial and identified Brien in the courtroom. Two of the hotel employees also identified Brien from a photo array as the man who had stayed at the hotel and also identified him at trial. The employee who had found the dye-stained money outside Brien's room testified to this effect.
Given the array of eyewitnesses, it is understandable that Brien does not now challenge the sufficiency of the evidence, but the nature of his defense at trial is pertinent to the claims he does raise on appeal. His defense was a claim of mistaken identity, based in part on inconsistent descriptions that the tellers had given of the robbers' physical characteristics after the event. Brien also offered testimony of one teller who was unable to pick Brien out from the photo array and from a customer who picked out an individual other than Brien from the photo array.
1. Brien's first claim on appeal concerns expert testimony. Prior to trial, Brien sought an in limine ruling permitting testimony from Alexander Yarmey, a professor of psychology, as an expert witness on the weaknesses of eyewitness identification. The one-paragraph description in the motion indicated that Yarmey would testify as to "the factors that affect memory, image retention and retrieval," and it provided a few sentences of detail. The trial judge asked for a proffer of testimony; Brien then submitted a three-page statement signed by counsel, somewhat enlarging upon the description.
The next day the court denied the motion in limine, stating that the proffer was too general and did not satisfy the foundational requirements for expert testimony under Daubert v. Merrell Dow Pharmaceuticals, Inc., --- U.S. ----, 113 S.Ct. 2786, 125 L.Ed.2d 469 (1993). The court made clear that, if the motion were renewed, the court wanted not the attorney's statement but "testimony of the [expert] under oath with respect to the proposed foundation evidence." In due course Yarmey filed an eight-page affidavit, which provided about 11 paragraphs of substantive opinion.
In his affidavit Yarmey disclaimed any ability to determine if a particular witness is credible or accurate in making an identification. But he set forth three elements of memory--observation, retention and retrieval; described pertinent factors that can undermine those elements (e.g., the stress of being confronted with a gun; the delay between the event and later identification); and drew attention to certain other sources that may create or compound errors in identification (e.g., reinforcement through police questioning; disparities in appearance that distinguish the suspect from others in a line-up or array).
Most of Yarmey's statements were general and but a few related directly to facts in this case. We set forth in the margin Yarmey's most detailed comment on the evidence, responding to the question, "What is [your] opinion about the suggestiveness of the photospreads used in this case?"1 Neither the affidavit nor any other submission by Brien's counsel purported to set forth in detail the scientific foundations for any of Yarmey's premises or conclusions, although Yarmey referred once to "the literature" documenting the phenomenon of "consistent error" (i.e., multiple misidentification based on a common source of error).
Two days later the district judge ruled that "in the exercise of whatever scope of discretion I have," he was excluding the testimony. The judge expressed a variety of concerns about the basis for assumptions in Yarmey's testimony and about "the fit and usefulness and misleading qualities" of the testimony. The judge stressed that he was not excluding expert testimony on identification as a matter of law. He also noted the risks of confronting the jury with battles of experts on areas within the common-sense competence of jurors.
The issue of expert testimony on eyewitness identification is an important and recurring one, and behind it lie issues even more fundamental: what factors should control the admission of such testimony, how much latitude does the trial judge have in resolving the issue, and what formulas or rules constrain the decision. In a recent case, we described the pertinent Federal Rules of Evidence but decided the case on its facts and declined to go very far in laying down general rules. United States v. Shay, 57 F.3d 126, 132-33. This case bears out the wisdom of that caution.
Broadly speaking, the expert testimony in this case involved a credibility determination within the ken of the ordinary judge and juror--unlike, say, DNA identification. But Fed.R.Evid. 702 permits expert evidence that "assist[s]" the jury; and quite possibly an expert such as a psychologist familiar with identification problems could give the jury background information about the mechanism of memory, types of errors, error rates, and other information not commonly possessed by the jury--information that may even be at odds with what a judge or juror might expect.
But helpfulness is a matter of degree, and expert evidence involves costs and risks--too obvious to need recounting--that distinguish it from lay evidence about "what happened here." Daubert itself, recalibrating the longstanding threshold requirement that the trial judge find expert evidence to be reliable, is but one facet of the difference in treatment. Indeed, trial judges have traditionally been afforded wide discretion to admit or exclude expert evidence. E.g., Hamling v. United States, 418 U.S. 87, 108, 94 S.Ct. 2887, 2903, 41 L.Ed.2d 590 (1974). But discretion is not carte blanche and, in some areas, prior law has been modified by the Federal Rules of Evidence.
In all events, for a range of reasons, trial courts have long hesitated to admit expert evidence purporting to identify flaws in eyewitness identification: for example, courts have said that the jury could decide the credibility issues itself; that experts in this area are not much help and largely offer rather obvious generalities; that trials would be prolonged by battles of experts; and that such testimony created undue opportunity for confusing and misleading the jury. Appeals courts have generally upheld rulings excluding such evidence. E.g., United States v. Fosher, 590 F.2d 381, 382 (1st Cir.1979); United States v. Purham, 725 F.2d 450 (8th Cir.1984).
Quite recently, several circuits have suggested that such evidence warrants a more hospitable reception. E.g., United States v. Rincon, 28 F.3d 921 (9th Cir.), cert. denied, --- U.S. ----, 115 S.Ct. 605, 130 L.Ed.2d 516 (1994); United States v. Stevens, 935 F.2d 1380 (3rd Cir.1991). There is more expert literature on the subject, more experts pressing to testify, and possibly more skepticism about the reliability of eyewitnesses. E.g., Loftus and Doyle, Eyewitness Testimony: Civil and Criminal (2d ed. 1992). It may be that a door once largely shut is now somewhat ajar.
We are unwilling to adopt a blanket rule that qualified expert testimony on eyewitness identification must routinely be admitted or excluded. Our Fosher decision is not a general bar to such testimony; that case upheld an exclusion as within "the broad discretion allowed a trial court." 590 F.2d at 382. But trial courts are likely to educate themselves, and us, by taking these proffers one by one. Obvious concerns are the reliability and helpfulness of the proposed expert testimony, the importance and the quality of the eyewitness evidence it addresses, and any threat of confusion, misleading of the jury, or unnecessary delay.
In this case, we sustain the district court's ruling on the ground that the district judge made clear his need for some proffer of data or literature underlying the expert's assumptions and conclusions, and the defense offered practically nothing, despite repeated opportunities to do so. In our view, this procedure was justified both in order to determine reliability under Daubert and to allow the judge to gauge whether the testimony would be helpful to the jury or would confuse or mislead instead. Nor is there any reason offered why Yarmey could not have supplied this foundation.
Brien argues that the expert literature casting doubt on eyewitness evidence is now so well established that the courts should take judicial notice of it. But Yarmey's testimony does not concern a single long-established scientific principle such as whether radar can measure speed. Rather, Yarmey offered a set of assertions whose helpfulness and reliability in this case depended on what lay underneath them. To say that eyewitnesses under stress sometimes misidentify tells the jury almost nothing beyond what common sense and argument would supply; almost everything turns on degree, data, comparable conditions, and other specifics.
If presented with a fair sample of the underlying data, the district court might have decided (as the trial judge here offered to consider) that some of the warnings were best reflected in instructions; that other portions of the proposed testimony were reliable and helpful; and that still other portions failed one or both of these criteria or met them but were outweighed by confusion or misleading character. Daubert, as well as common prudence, entitled the judge to require such underlying information, and the failure to provide it supplies an adequate basis for the trial court's decision to reject the proffer.
There is nothing to Brien's alternative argument that Fed.R.Evid. 705 entitled Yarmey to offer the expert testimony without disclosing the underlying data, leaving that to be developed by cross-examination. Rule 705 relates to the presentation of testimony at trial and, even then, is subject to the supervision of the trial judge to avoid unfairness. The rule does not impair--indeed, has nothing to do with--the trial judge's right to insist that he or she be given the underlying information by proffer as an aid to the preliminary ruling on admissibility.
2. Brien's next claim of error raises an issue which, if anything, is even more significant to the conduct of criminal trials. Every viewer of trials, or even Hollywood depictions of them, is familiar with the routine practice by which an eyewitness to the crime takes the stand, points to the defendant sitting at counsel table, and identifies the defendant as the one who committed the crime. Here, Brien complains of the alleged refusal of the trial judge "to permit [instead] non-suggestive courtroom identification procedures."
Prior to trial Brien's trial counsel, who appears to have provided Brien with an energetic and inventive defense, began by moving (with notice to the government) for permission to have the defendant seated in the spectator section of the courtroom during the trial; and by a contemporaneous motion (filed ex parte ), Brien's counsel asked leave to salt the audience with "three or four individuals of the same general description as the bank robber...." Shortly before picking the jury, the trial judge, on March 28, addressed both motions in open court, as follows:
First, the court said that the proposal to salt the audience opened the way to "the reverse of an improper identification procedure" by the government, noting that Brien's proposal would allow him to bring in anyone including an identical twin. Second, the court ruled that the defendant could sit "anywhere in the courtroom," subject to limitations required by security; but the court said it could "have a problem" with the defendant switching positions at will from counsel table to audience. Third, the court said that it would not permit ex parte submissions on the identification procedure but would allow argument by both sides.
Defense counsel said that he wanted to discuss the matter with Brien himself. Two days later, Brien's counsel renewed (not ex parte) his original request to bring in spectators similar in description to the bank robber but offered no details that might have allayed the concern expressed by the trial judge. On March 31, shortly before opening argument, defense counsel asked the court to rule on the renewed motion, amending his earlier request by indicating that Brien might choose to sit at counsel table if the motion to bring in selected spectators were denied.
The trial judge then said in part:
I do not write blank checks and sign them. Now, that's what you're asking me to do ... here. You're not disclosing to me anything about what are the arrangements that you've set up, how you expect to handle it, and I am not going to give you complete authority to do it any way you want to and give you an advance ruling that it's permissible. As I've said to you yesterday, there are problems here. You're very close to the edge, it seems to me, of some problems about whether you are undertaking to proceed in a way that will tend to mislead witnesses....
The court then denied the motion. Defense counsel made no effort to provide any further details of his proposal.
On appeal, Brien's appellate counsel has made the obvious attack on the usual practice of courtroom identification of the defendant while seated at counsel table; and he has cited us to several Second Circuit cases suggesting that that court might favor a defendant who sought a more balanced form of courtroom identification where identification was a contested issue, the defendant moved pretrial for a courtroom line-up, and the witness did not pick the defendant out in a fair out-of-court lineup prior to trial. E.g., United States v. Sebetich, 776 F.2d 412 (1985), cert. denied, 484 U.S. 1017, 108 S.Ct. 725, 98 L.Ed.2d 673 (1988). These conditions, Brien says, were met in this case.
If Brien had presented the court with a detailed plan for a fair in-court line-up, and the court had rejected the plan without a plausible justification, then on the present facts we think that a significant issue would be presented. But Brien's motion is not within a country mile of such a proposal. As the trial court sensibly explained, Brien's plan left room for a scenario fully capable of misleading the jury. To alter the standard practice, it was up to Brien's counsel to propose a plan that would guard against unfairness to either side. This, despite ample invitations, Brien's counsel declined to do.
This refusal may well have been an entirely justified trial tactic. If counsel feared that a fairly staged court room line-up would still likely result in identification of Brien as the robber, the line-up would strengthen the credibility of the witnesses and undermine counsel's misidentification argument to the jury. But whether or not counsel sought an advantage and retreated when none was offered, the trial judge was within his discretion--and we think eminently right--in refusing to endorse what the trial judge properly described as a blank check.
3. Four additional rather compact claims of error are urged by Brien (e.g., that the trial court should have granted a requested continuance). The government's answering brief provides on the surface an ample response to each of the four claims; Brien's reply brief makes no effort to meet the responses. Appraising the four claims on the merits, we do not think that any of them warrants separate discussion. In each instance the trial judge handled the matter properly, the issue is not close, and no significant legal question is presented.
By contrast, the first two issues in the case--the expert evidence and courtroom identification--do raise very difficult questions that have been ably briefed on both sides. But the difficulty of these issues, taken in the abstract, confirms the wisdom of the trial judge's approach: outlining his legitimate concerns to counsel, providing opportunities for those concerns to be addressed, and (ultimately) insisting on justifications grounded in the particulars of the case.
Of the Northern District of California, sitting by designation
"The particular male photospread in this instance (because it depicts several people of a light complexion) was of lesser reliability because witnesses could eliminate those people who do not fit the dark or medium complexion, which I understand that they observed. This, in itself, indicates suggestiveness because of the quality of the photographs. The female photospread is also suggestive because only a few of the photographs reflect women with any makeup or earrings, which I understand at least one of the witnesses observed."