Summary
excluding expert testimony on eyewitness reliability because the testimony created the risk of confusing the jury
Summary of this case from U.S. v. LesterOpinion
No. 95 CR 1075 (ERK).
June 6, 1996.
Zachary Carter, United States Attorney by Timothy Macht, Asst. U.S. Attorney, Eastern District of New York, Brooklyn, New York, for plaintiff.
The Legal Aid Society, Federal Defender Division by Cynthia A. Matthews, Brooklyn, New York, for defendant.
MEMORANDUM
The purpose of this memorandum is to set out in greater detail the reasons stated on the record on April 12, 1996, for denying defendant's in limine motion to permit Dr. Michael Leippe to testify at defendant's trial. Dr. Leippe, a professor of psychology, was offered as an expert on the subject of eyewitness identification. The motion was filed on Wednesday, April 10, along with a memorandum of law and several scholarly articles. I conducted a hearing on Friday, April 12 in order to resolve the issue in advance of the Monday, April 15 trial date.
The charges against defendant arise out of a robbery that occurred at a Burger King restaurant located on the Fort Hamilton Army Base in Brooklyn, New York. On the evening of October 12, 1995, two robbers entered the restaurant. One wore a hooded sweatshirt with the hood drawn close around his face and carried no weapons while the other wore a visor and toted a sawed-off shotgun. At the time of the robbery, the manager of the restaurant, Daveena Frazier, was seated in her office counting money in a cash drawer she had removed from the safe located in her office. Ms. Frazier was seven and a half months pregnant at the time. When she heard the commotion surrounding the robbers' entry she rolled her desk chair across her office so that she could look out of the office doorway. Tr. 48. From there, she saw Ethan Whitby, another Burger King employee, standing with his hands up facing her. Between Whitby and the manager's office, about eight feet away from Ms. Frazier, stood the gunman, facing Whitby, with his back to Ms. Frazier. Tr. 48. The gunman then turned to face Ms. Frazier and walked toward her until he was a foot away, at which time he told her to get down on the floor. Tr. 49. As he approached her, Ms. Frazier looked the gunman directly in the face. Tr. 49.
Immediately following the robbery, Ms. Frazier gave police a detailed description of her assailant. She described a young Hispanic male, about six feet tall, slim, about 160 pounds, with short dark hair, dark eyes, a slight mustache, a long nose and acne under his cheeks. Tr. 56; Tr. 114. That night, Ms. Frazier and Mr. Whitby looked at several hundred mugshots at the police station but did not pick out the robbers. Six days after the robbery, Ms. Frazier was shown a non-suggestive photo array consisting of six pictures and recognized one of them as the gunman. Tr. 75. In the months before trial, Ms. Frazier moved to Alaska and did not see the gunman or his likeness again until early April, when Ms. Frazier saw the defendant standing in the lobby of the courthouse as she was going to meet with the Assistant United States Attorney prosecuting this case. She recognized the defendant as the gunman and was struck by the "way he stared at me when I looked at him, he stared at me. It was the same way he looked that night." Tr. 76. As she walked passed him, defendant asked, "Don't you live on base?" to which she said "no" and kept walking. Tr. 77.
Based on a proffer of these facts, it appeared that Ms. Frazier's testimony would be, at the very least, a significant part of the case against defendant. Because of the fallibility of eyewitnesses and of the dangers of misidentification, particularly in cases where a single eyewitness's testimony is the only evidence from which a jury is asked to determine guilt beyond a reasonable doubt, see Kampshoff v. Smith, 698 F.2d 581, 585 (2d Cir. 1983), I agreed to take special precautions to ensure that the jury approached Ms. Frazier's testimony with appropriate caution, although I decided not to admit expert testimony.
As an initial matter, to qualify to testify under Fed.R.Evid. 702, an expert must be able to testify to "(1) scientific knowledge that (2) will assist the trier of fact to understand or determine a fact in issue." Daubert v. Merrell Dow Pharmaceuticals, Inc., 509 U.S. 579, 592, 113 S.Ct. 2786, 2796, 125 L.Ed.2d 469 (1993). I assumed, for the purpose of determining whether to admit Dr. Leippe's testimony, that the defendant's proffer satisfied the first prong of this test. Nevertheless, as is true with all relevant evidence, expert testimony should be excluded where the danger of confusing or misleading the jury substantially outweighs its probative value. See United States v. Serna, 799 F.2d 842, 850 (2d Cir. 1986), cert. denied, 481 U.S. 1013, 107 S.Ct. 1887, 95 L.Ed.2d 494 (1987). As the Supreme Court has recognized, "[e]xpert evidence can be both powerful and quite misleading because of the difficulty in evaluating it. Because of this risk, the judge in weighing possible prejudice against probative force under Rule 403 . . . exercises more control over experts than over lay witnesses." Daubert, 509 U.S. at 595, 113 S.Ct. at 2798 (quoting Jack B. Weinstein, Rule 702 of the Federal Rules of Evidence is Sound; It Should Not be Amended, 138 F.R.D. 631, 632, (1991)). Indeed, the balancing test of Rule 403 is implicit in the requirement of Rule 702 that the expert's testimony help rather than hinder the jury's deliberations.
In Serna, the Court of Appeals for the Second Circuit specifically held that a trial court has "broad discretion in admitting or excluding expert testimony under Fed.R.Evid. 702 and in excluding testimony under Fed.R.Evid. 403 because of the danger of jury confusion or unfair prejudice." Serna, 799 F.2d at 850; see also United States v. Brien, 59 F.3d 274, 277 (1st Cir.), cert. denied, ___ U.S. ___, 116 S.Ct. 401, 133 L.Ed.2d 320 (1995). The danger of confusion arises from the "aura of special reliability and trustworthiness surrounding expert testimony." United States v. Young, 745 F.2d 733, 766 (2d Cir. 1984) (Newman, J., concurring) (quoting United States v. Fosher, 590 F.2d 381, 383 (1st Cir. 1979)), cert. denied, 470 U.S. 1084, 105 S.Ct. 1842, 85 L.Ed.2d 142 (1985); see also United States v. Amaral, 488 F.2d 1148, 1152 (9th Cir. 1973). Moreover, much of what an expert would testify to can often be "fleshed out adequately by counsel through probing cross-examination and arguments pitched to the common sense of the jury." United States v. Stevens, 935 F.2d 1380, 1399-1400 (3rd Cir. 1991); see also Amaral, 488 F.2d at 1153.
In order to determine whether Dr. Leippe should be permitted to testify, defense counsel was permitted to elicit his testimony at the pretrial hearing. Dr. Leippe stated that the most important point he wanted to make — "the strongest part of what I feel I'm here for," Hearing Tr. 28 — is that "[w]e can't be more confident in the accuracy of a witness' identification just because the person himself or herself is confident" of the accuracy of the identification. Hearing Tr. 17. "[E]ven the appearance of confidence does not necessarily mean that the witness is accurate." Hearing Tr. 28. This testimony, however, did not suggest that an identification made by a witness who appeared confident in the accuracy of his or her identification is necessarily unreliable. Moreover, it entails the following corollary, which seems more appropriate to the identification here: we cannot be less confident in the accuracy of an eyewitness identification just because the eyewitness does not, in words or tone, convey confidence in the identification.
Dr. Leippe also talked about a "forgetting curve," which shows that "most forgetting" takes place "in the hours and days immediately following exposure and encoding," and that "the amount of forgetting that occurred is probably a lot greater between for example day 0 and day 6 than it would be between day 6 and day 12 or day 6 and day 20," Hearing Tr. 14. He later acknowledged, however, that "the retention is not — the six day thing isn't so important, in my opinion, as the fact that this is a stranger seen once, under the most poor conditions." Hearing Tr. 29-30. "[T]he difference between memory for once briefly seen faces and familiar faces is," Dr. Leippe emphasized, "a huge gulf." Hearing Tr. 29.
A third point in Dr. Leippe's presentation involved a phenomenon he termed "weapon focus." Dr. Leippe testified that studies had shown that facial identification was less reliable in circumstances where a weapon, such as a gun, is present. This testimony simply attached the aura of expertise to two obvious facts: 1) a gun pointed at a victim of a crime is going to focus the mind of the victim on the weapon rather than on the facial features of the perpetrator because a gun "can command attention, can rivet attention," Hearing Tr. 12, and 2) the less attention paid to the face of the perpetrator the less reliable the subsequent identification.
After listening to Dr. Leippe, I determined that a thorough and evenly balanced jury charge, one that took into account some of Dr. Leippe's concerns, would be the most effective means of guiding the jury in its evaluation of Ms. Frazier's testimony. See United States v. Rincon, 28 F.3d 921, 925-26 (9th Cir.) (holding that comprehensive jury instruction on eyewitness identification was preferable to expert testimony under the circumstances), cert. denied, ___ U.S. ___, 115 S.Ct. 605, 130 L.Ed.2d 516 (1994). Because this was to be a short trial with relatively uncomplicated evidence, a lengthy battle of experts would be distracting and confusing to the jury. Even if the government had not been planning to call its own expert, the testimony of Dr. Leippe posed the danger of confusing the jury because it suggested that the case turned on whether the jury credited his testimony or the eyewitness, Hearing Tr. 32, and because of the "aura of special reliability and trustworthiness surrounding expert testimony which ought to caution its use." Young, 745 F.2d at 766 (2d Cir. 1984) (Newman, J., concurring) (quoting Fosher, 590 F.2d at 383); see also Amaral, 488 F.2d at 1152.
More significantly, there are special considerations relating to this case that rendered extensive expert testimony unnecessary. This was not a case in which an in-court eyewitness identification was preceded by a suggestive pre-trial identification. Indeed, Dr. Leippe testified that he saw nothing unduly suggestive in the photo array used in this case. Hearing Tr. 16. Nor is this a case in which a suggestive in-court identification would be the only one in the case. On the contrary, six days after the robbery, the eyewitness picked the defendant out of a photographic array that was not suggestive and she again recognized him when she unexpectedly saw him in the lobby of the courthouse shortly before the trial. Moreover, although my ruling came in a pretrial order, by the end of the trial, it had become clear that the accuracy of Ms. Frazier's identification would not be the only, or even the primary, issue confronting the jury.
Ms. Frazier's testimony at trial came in exactly as anticipated. With respect to the identification she made from the photo array, Ms. Frazier testified as follows:
Q: When the photographs were shown to you, what did you do?
A: I looked at them.
Q: About how long did you look at this group of photographs?
A: About three to five minutes.
Q: After you looked at the photos for this period of time, what if anything did you say?
A: I pointed to the picture and said I felt this was the guy with the gun.
Q: Now, when you first looked at the photograph which you picked out as the man with the gun, what did you say?
A: That that was the man with the gun.
Q: What caused you to spend a few more minutes looking at all the photographs?
A: I felt because of the serious nature of this whole thing, that I ought to look at all the pictures.
Tr. 116. On cross examination, defense counsel sought to emphasize, among other things, that Ms. Frazier had been "startled" at the time of the robbery, Tr. 100, that having a gun pointed at her head was a new and frightening experience for her, Tr. 102, and that she might have seen the defendant on another occasion, Tr. 109.
The prosecution supplemented the eyewitness identification with an assortment of circumstantial evidence. Perhaps the most significant such evidence was Ms. Frazier's testimony, not only that she recognized the defendant when she encountered him unexpectedly in the lobby of the courthouse a week before the trial, but that he recognized her. Tr. 76-77. Another incriminating piece of evidence was that just as the arresting officers were breaking down the door to defendant's apartment, he threw a video cassette box containing $128 in U.S. currency out of his window and into a nearby tree. Tr. 134; Tr. 159. Although the amount was substantially less than had been stolen from the Burger King, the implication was that defendant was attempting to hide the remaining proceeds of his crime, even though, unknown to him, the bills taken from the Burger King were not traceable. Tr. 214.
However, what was perhaps the most incriminating evidence came from Ethan Whitby, the eighteen-year-old Burger King employee called to the witness stand by the defense. Whitby was called to testify that Louis Burrous was an acquaintance of his whose face Whitby would easily have recognized had the defendant been the gunman. Visibly anxious, Whitby then told the jury that the gunman was not Louis Burrous. Tr. 219; Tr. 226. He said that he had heard the gunman's voice during the robbery but it was not familiar to him. Tr. 230. Whitby acknowledged that he had been shown the photo array containing defendant's picture but claimed to have identified another photograph as that of the gunman and yet another as his accomplice. Tr. 240. Both of these identifications were indisputably inaccurate.
Whitby's claim that defendant was not the gunman crumbled on cross-examination, where his answers to a series of questions strongly suggested that he had, in fact, positively identified defendant as the gunman when shown the photo array. The questioning was as follows:
Q: At that point in time when you were shown the photos, you said you knew Louis Burrous, right?
A: Yes.
Q: And the FBI asked you if he was the gunman, correct?
A: Yes. I can't remember that day too good.
Q: Didn't you, when asked if Louis Burrous was the gunman, say that the gunman looked like Burrous?
A: He just — yeah, I guess I would say that.
Q: Didn't you also say that the gunman sounded like Louis Burrous?
A: I have heard Louis make a voice like that before.
Q: And didn't you also say that the gunman treated you real nice, like he knew you? Did you say that?
A: No. That was the second robber.
Q: Didn't you also say, "I'm not going to say the exact words that it was Louis Burrous, because if I do, then you'll put me on the stand to testify." Didn't you say that?
A: I can't remember.
Q: Mr. Whitby, you're afraid of Mr. Burrous, aren't you.
A: No.
Q: When the FBI showed you the photo array, didn't you say that you would be afraid to testify against Louis Burrous?
A: Why would I testify against him when he didn't do it?
Q: Didn't you say that you'd be afraid to testify against him?
A: Yes.
Q: Just a couple of weeks ago when you met with the Government at the U.S. Attorney's office, didn't you say the exact same thing, that you were afraid to testify against Louis Burrous?
A: Something like that, yeah.
Q: In fact, didn't you say you were worried that Mr. Burrous might have a gun?
A: No, I can't remember saying that.
Q: Didn't you say that you were worried that if you testified against Louis Burrous, he'd get his friends after you?
A: Yes.
Tr. 281-82.
The prosecution subsequently called the agent who showed the array to Whitby to attest to the truth of what Whitby tried, with less than complete success, to deny: that Whitby had identified Burrous as the gunman. Tr. 311. Agent Nelson testified that when Whitby was shown the photo array containing defendant's picture, "[h]is eyes were avoiding the center of the photo array," where defendant's picture was, and instead "focused on the left and then jumping to the right portion of the photo array." Tr. 309. When asked whether he knew anyone depicted in the photo array, Whitby had admitted that he knew Louis Burrous. Tr. 310. When asked whether Louis Burrous had been the gunman, Whitby had said "that it looked like Louis, it sounded like Louis and that he treated me nice like Louis, but I'm not going to tell you the exact words that it was Louis, because then you'll be able to put me on the stand and make me testify." Tr. 311.
In light of the testimony given by Whitby, a witness obviously scared out of his wits and lying to help the defendant, the case against defendant cannot be said to have turned solely on the accuracy of Frazier's identification. Instead, it turned to a significant extent on whether the jury believed that Whitby was telling the truth when he attempted to exonerate defendant or when, as both Whitby and Agent Nelson testified, he reluctantly identified defendant as the gunman when shown the photo array. If the jury believed that Whitby, whose ability to recognize defendant as the gunman was untainted by any of the factors that make identification of a complete stranger so treacherous, was being truthful when he said defendant did not do it, they would have believed that Frazier was simply mistaken. If the truth was that Whitby had recognized defendant as the gunman, then Frazier's identification, as well as the inculpatory inferences from the circumstantial evidence, would be corroborated. Under neither scenario would the jury be in a position to determine defendant's guilt or innocence based primarily on Frazier's eyewitness identification. While the eyewitness identification by Ms. Frazier was important, this was not a case where the danger of erroneous conviction justified the drawbacks associated with expert testimony.
Although I did not permit Dr. Leippe to testify, out of an abundance of caution, I gave the jury a comprehensive instruction on eyewitness identification, the text of which is appended hereto. Whether to include such a charge is a matter of discretion and may not have been necessary given the "strong indicia" that Ms. Frazier's identification of defendant was reliable. United States v. Luis, 835 F.2d 37, 41-42 (2d Cir. 1987). The identification charge began with the following instruction, which preceded the factors that the jury was told it should focus on:
The government must prove beyond a reasonable doubt that the crime charged in this case was actually committed. But more than that, the government must prove, beyond a reasonable doubt, that the defendant committed the crime. Therefore, the identification of the defendant is a critical part of the government's case. As with any other witness, you must first decide whether the identification witness is telling the truth as she or he understands it and whether she or he has any motive to lie. But you must do more than that. You must also decide how accurate the identification was — i.e., whether or not the witness was mistaken.
The charge then went on to alert the jury to the dangers of mistake in human perception and identification and cautioned it about the weak foundation upon which the identification of a stranger rested:
I want to caution you, first, that the kind of identification testimony you heard in this case must be scrutinized carefully. Scientific studies have amply demonstrated the dangers of mistake in human perception and identification.
Of course, this does not mean that the identification in this case is incorrect. I merely tell you this so that you understand the importance of carefully evaluating the evidence here.
One of the reasons for this is that when a witness testifies in court that the defendant was the one who committed the offense, what he or she is really saying is that the defendant is the person he or she remembers seeing commit the offense.
The words "remember" and "seeing" call into question factors which bear on the ability of a person to recall and the opportunity of a person to see. If you walk into a bank and see your next door neighbor fleeing with a gun and sack, what you recall later is that you saw a person whom you recognized to be your next door neighbor. When the person is a total stranger, you do not have that kind of "anchor" to rely upon and so a subsequent identification depends on your ability to accurately recall the exact features of the individual.
The charge then detailed a number of factors that the jury should consider in determining the accuracy of the eyewitness identification. One of these factors highlighted the two obvious facts underlying the proposed "weapon focus" testimony of Dr. Leippe:
You should consider whether this witness had a good opportunity to see the person — how close or far away — and the length of time the witness had to observe the person who she claims is the defendant. You should also consider whether the witness's attention was focused on the weapon that was pointed at her or on the face of the perpetrator. The better the opportunity to observe and the longer the period of observation, the more accurate the identification is likely to be. Conversely a brief encounter — a matter of seconds or minutes — tends to increase the likelihood of misidentification.
After setting out additional general factors bearing on the ability of a witness to see and accurately recall events in a case such as this, the charge focused on the critical pretrial identification made from the photographic array six days after the robbery. Specifically, the jury was given the following cautionary instruction, which to a large extent reflected Dr. Leippe's views:
You may also consider that an identification made by picking the defendant out of a group of similar individuals, or a group of photographs of similar individuals, is generally more reliable than one which results from the presentation of the defendant alone to the witness or among a group of persons with significantly different appearances. It is also more reliable than an identification made in the courtroom where a witness expects to see the accused in the courtroom.
Even if the law enforcement officers follow the most correct photographic identification procedures and show the witnesses the pictures of a number of similar individuals without indicating whom they suspect, there is some danger that the witnesses may make an incorrect identification, even if he or she is sure of his or her identification.
If there is an initial misidentification, then you must answer the question whether thereafter the witnesses retained in their memory the image of the photograph rather than the image of the person actually seen, since under those circumstances the trustworthiness of subsequent identifications is reduced.
Even if a witness is positive of his or her identification, this does not relieve you of the duty to carefully consider his or her identification testimony, especially if you find it is the only evidence that directly supports the claims that the defendant committed the offense charged.
The charge included cautionary and qualifying language regarding the certainty or positiveness of identification even though the eyewitness did not testify that she was certain or positive when she picked defendant's photograph from the array, Tr. 116, and even though her in-court identification did not convey certainty. The cautionary language was added because the FBI agent who showed her the array testified that he had asked the witness whether she was positive and that she had answered in the affirmative, Tr. 211, notwithstanding her testimony that she had only "pointed to the picture and said I feel this was the guy with the gun." Tr. 116.
Taken as a whole, the charge I read to the jury included more admonitions of the kind that were favorable to the defense than the standard charge on eyewitness identification. Indeed, when Dr. Leippe was shown an early draft of my instruction, he commented that they were "quite fair overall to the defendant," Hearing Tr. 27, and "[t]o the extent that it's not embedded in a long, long closing instruction kind of thing, I would say that this stands a better chance [of effectively instructing the jury]than a lot of the stuff that I've seen before," Hearing Tr. 36. The instruction was later modified to take into account Dr. Leippe's testimony and to make it somewhat more balanced.
Based on the requests made by the jury during deliberations, which included the testimony of both Frazier and Whitby, as well as the photographic array itself, I have no doubt that the jury understood and followed my instructions, thoughtfully considered defendant's attempt to impugn the eyewitness identification, and returned the verdict that the evidence compelled.
APPENDIX Jury Instruction Regarding Eyewitness Identification
The government must prove beyond a reasonable doubt that the crime charged in this case was actually committed. But more than that, the government must prove, beyond a reasonable doubt, that the defendant committed the crime. Therefore, the identification of the defendant is a critical part of the government's case. As with any other witness, you must first decide whether the identification witness is telling the truth as she or he understands it and whether she or he has any motive to lie. But you must do more than that. You must also decide how accurate the identification was — i.e., whether or not the witness was mistaken.
I want to caution you, first, that the kind of identification testimony you heard in this case must be scrutinized carefully. Scientific studies have amply demonstrated the dangers of mistake in human perception and identification.
Of course, this does not mean that the identification in this case is incorrect. I merely tell you this so that you understand the importance of carefully evaluating the evidence here.
One of the reasons for this is that when a witness testifies in court that the defendant was the one who committed the offense, what he or she is really saying is that the defendant is the person he or she remembers seeing commit the offense.
The words "remember" and "seeing" call into question factors which bear on the ability of a person to recall and the opportunity of a person to see. If you walk into a bank and see your next door neighbor fleeing with a gun and sack, what you recall later is that you saw a person whom you recognized to be your next door neighbor. When the person is a total stranger, you do not have that kind of "anchor" to rely upon and so a subsequent identification depends on your ability to accurately recall the exact features of the individual.
In this case, a witness testified that she did not know the defendant before the crime took place and, therefore, you must carefully weigh the following factors which bear on the ability of the witness to see and accurately recall the events:
1. You should consider whether this witness had a good opportunity to see the person — how close or far away — and the length of time the witness had to observe the person who she claims is the defendant. You should also consider whether the witness's attention was focused on the weapon that was pointed at her or on the face of the perpetrator. The better the opportunity to observe and the longer the period of observation, the more accurate the identification is likely to be. Conversely a brief encounter — a matter of seconds or minutes — tends to increase the likelihood of misidentification.
2. You should consider whether the incident in which the individual participated was significant or unusual to the witness at the time it occurred.
3. You should consider how much time passed between the crime and the first identification by the witness.
4. You should consider whether a witness has given descriptions that are inconsistent with or which differ from the actual physical appearance of the person allegedly identified and also whether the witness gave different descriptions at different times. Similarly, you should consider whether a witness has given descriptions that are consistent with the actual appearance of the person allegedly identified and also whether the witness gave the same description at different times.
5. You may also consider that an identification made by picking the defendant out of a group of similar individuals, or a group of photographs of similar individuals, is generally more reliable than one which results from the presentation of the defendant alone to the witness or among a group of persons with significantly different appearances. It is also more reliable than an identification made in the courtroom where a witness expects to see the accused in the courtroom.
Even if the law enforcement officers follow the most correct photographic identification procedures and show the witnesses the pictures of a number of similar individuals without indicating whom they suspect, there is some danger that the witnesses may make an incorrect identification, even if he or she is sure of his or her identification.
If there is an initial misidentification, then you must answer the question whether thereafter the witnesses retained in their memory the image of the photograph rather than the image of the person actually seen, since under those circumstances the trustworthiness of subsequent identifications is reduced.
Even if a witness is positive of his or her identification, this does not relieve you of the duty to carefully consider his or her identification testimony, especially if you find it is the only evidence that directly supports the claims that the defendant committed the offense charged.
On the other hand, it is not essential that the witness him or herself be free from doubt as to the correctness of his or her identification of the defendant, provided you are satisfied that the government has met its burdens of proving beyond a reasonable doubt that the defendant is guilty of the charges contained in the information. If after examining the testimony of this witness and all the circumstances under which she identified the defendant and all of the other evidence in the case, you have a reasonable doubt as to whether the defendant committed the offense charged, you should return a verdict of not guilty on both counts.