Opinion
Crim. No. 16-516 (KM)
05-01-2017
UNITED STATES OF AMERICA v. Gregory A. JONES, Defendant.
OPINION and ORDER
(Eyewitness Identification Expert)
INTRODUCTION
The defendant, Gregory A. Jones, is charged with two robberies of the same bank: one, unarmed, on May 6, 2014 ("Bank Robbery 1"); and another, armed, on September 19, 2014 ("Bank Robbery 2"). At issue are the eyewitness identifications of the defendant by bank employees. Those identifications occurred after those employees, without police involvement, went online and saw a mugshot of the defendant on a privately-run website. On February 27, 2017, I filed a Memorandum and Order ("Op.," ECF no. 30) denying the defendant's motion to suppress the bank employees' eyewitness identifications of him.
In the same Memorandum and Order, however, I granted defense counsel's oral motion for an adjournment so that the defense could retain an expert witness on the reliability of eyewitness identifications. (Op. 14-16) The government conceded that such expert testimony, in general, would be appropriate. The government reserved the right, however, to object to specific items under the standards of Fed. R. Evid. 702, once it had seen the expert report. The defense has now proffered the Report of Deryn Strange, PhD ("Strange Rpt.", ECF no. 34-1) and her c.v. (ECF no. 34-2), together with a letter brief in support of the admissibility of her testimony (ECF no. 34). The government has filed a letter brief in response (ECF no. 35), and the defendant has filed a reply (ECF no. 40).
The government concedes (Gov't Brf. 7), and I agree, that Dr. Strange is appropriately credentialed and qualified to give opinion testimony. At issue here are the government's five objections in limine to the scope of Dr. Strange's proffered testimony. Those objections are:
1. The Court should limit Dr. Strange to offering testimony about general psychological principles regarding eyewitness identifications. She should not be permitted to discuss the specific eyewitnesses or identification in this case. If Dr. Strange is permitted to testify about the facts of this case, her testimony should be limited to discussing only facts that are in evidence at trial.(Gov't Brf. 1)
2. The Court should prohibit Dr. Strange from offering testimony about wrongful convictions.
3. The Court should prohibit Dr. Strange from offering testimony about the behavior or beliefs of jurors.
4. The Court should prohibit Dr. Strange from offering testimony regarding voice identifications.
5. The Court should prohibit Dr. Strange from offering testimony about composites.
The purpose of this Opinion, then, is to determine the permissible scope of Dr. Strange's expert testimony. Like any in limine ruling, it is subject to revision in light of trial developments, but it seemed best to set some ground rules in advance, for the guidance of both sides. From this pretrial perspective, I rule that the government's objections, substantially but not entirely, are well-taken. Dr. Strange's testimony will be limited to what will assist, but not impinge upon, the jury's fact-finding role.
DISCUSSION
The relevant background facts are stated in my earlier Memorandum and Order. (Op. 2-4) Essentially, the time line is as follows:
| 5/6/2014: | Bank Robbery 1. The robber obscures his features with ascarf, glasses, and a hat, and does not display a weapon. |
| 9/19/2014: | Bank Robbery 2. The robber does not obscure his face,and he displays and discharges a gun. Shortly thereafter,two bank employees tell investigators they believe the tworobberies were committed by the same person. |
| 4/7/2015: | The State charges Mr. Jones with Bank Robbery 1 basedon a DNA match. The prosecutor's office notifies one ofthe bank witnesses that Mr. Jones has been charged |
| 4/2015: | The bank witnesses, now aware of Mr. Jones's name, goonline and find a mugshot of Mr. Jones on a privately-runwebsite. |
| 4/30/2015: | The bank witnesses tell FBI investigators that themugshot depicts the person who committed BankRobbery 2, and state that they believe, based on therobber's voice and general appearance, that he is thesame person who committed Bank Robbery 1. |
| 1/8/2016: | Mr. Jones is charged in a federal complaint with bothbank robberies. He is subsequently indicted. |
For the reasons expressed in my earlier opinion, expert testimony is appropriate here. I do not find it appropriate under the circumstances to employ the rationale that the strength of the government's case precludes or reduces the necessity of expert eyewitness identification testimony under Rule 403. The issue is one of scope.
The controlling authorities, as the parties agree, are United States v. Brownlee, 454 F.3d 131 (3d Cir. 2006), and United States v. Downing, 753 F.2d 1224 (3d Cir. 1984), which are discussed in my prior Memorandum and Order. Those cases teach that the proper scope of the testimony is defined by the well-known Rule 702 principles of "qualification, reliability and fit" imposed by Daubert v. Merrell Dow Pharmaceuticals, Inc., 509 U.S. 579 (1993), and Kumho Tire Co. v. Carmichael, 526 U.S. 137 (1999).
Brownlee, in particular, is instructive in that it embodies a principle that the Court must grant the expert scope to discuss the issues that "fit" the facts of the case. There, the district court had permitted an expert on human memory and perception to testify as to psychological factors such as cross-racial identification, the effect of hair covering, weapons focus, and exposure to multiple witnesses. 454 F.3d at 140-41. The Court of Appeals nevertheless reversed because the district court had excluded testimony as to other psychological factors implicated by the facts of the case: "(1) the comparison between the show-up and other identification procedures"; "(2) the suggestiveness of the show-up involved in this case and the effect it potentially played in the identifications"; "(3) confidence malleability"; "(4) post-event suggestiveness"; and "(5) confidence of accuracy." Id. at 141. In short, the expert must not be restricted from testifying about opinions that are pertinent to the case.
On the other hand, the expert testimony cannot be too general. It must be confined to the features of the particular eyewitness identification in the case, and address how such features bear on reliability. See Downing, 753 F.2d 1242; see also Brownlee, 454 F.3d at 141 (citing United States v. Sebetich, 776 F.2d 412, 419 (3d Cir. 1985)).
From all of this I draw the lesson that "fit" must be neither too narrow nor too broad. To take a simple example, weapons focus testimony would properly be excluded in a case where no weapon was drawn, but it might be error to exclude it where a weapon was pointed at the victim.
Those principles should be kept in mind as a backdrop to my discussion of the government's five objections to the scope of the proffered testimony.
1. Psychological principles, not specifics of this case
The first and most general of the government's objections is that "[t]he Court should limit Dr. Strange to offering testimony about general psychological principles regarding eyewitness identifications. She should not be permitted to discuss the specific eyewitnesses or identification in this case. If Dr. Strange is permitted to testify about the facts of this case, her testimony should be limited to discussing only facts that are in evidence at trial." (Gov't Brf. 1)
"Psychological" I accept as a shorthand term. The area of expertise at issue here more specifically involves human memory, perception, and cognition as they relate to the accuracy, or not, of eyewitness identifications.
Expert testimony about the reliability of human memory and perception can be very helpful to a jury that is evaluating the accuracy of an eyewitness identification. I agree with the government, however, line should be drawn between the expert's explanation of such principles, and the expert's comment on the reliability of the bank witnesses' eyewitness identifications here. The latter brand of case-specific testimony threatens to impinge upon the jury's function.
The distinction is between telling the jury how to think, and telling the jury what to think. See United States v. Mathis, 264 F.3d 321, 340 (3d Cir. 2001) (upholding admission of opinion testimony because the expert ""did not intend to tell the jury whether [the witness] was lying or telling the truth . . . [the expert] attempted to provide information that, if itself deemed credible, might cause the jury to evaluate [the witness's] testimony in a different light."). Expert testimony on the accuracy, or not, of a particular witness's eyewitness identification can "intrude[] too much on the traditional province of the jury to assess witness credibility." United States v. Lumpkin, 192 F.3d 280, 289 (2d Cir. 1999); see also United States v. Carter, 410 F.3d 942, 950 (7th Cir. 2005) (upholding discretionary exclusion of expert testimony as to credibility of eyewitness testimony, noting that "it influences a critical function of the jury—determining the credibility of witnesses.") (quoting United States v. Hall, 165 F.3d 1095, 1101-03 (7th Cir. 1999)).
Time may have cast doubt on holdings, like these, that expert testimony should ordinarily be excluded. The point remains, however, that such testimony must be treated with care, to avoid treading on the jury's prerogatives.
The government has particular concerns, which I share, about Dr. Strange's application of her opinions in a manner that amounts to an opinion about the reliability of the specific eyewitness identifications in this case. The expertise, indeed the prestige, of an expert opinion witness may have the undesirable effect of absolving the jurors. By "absolving," I mean relieving them of their responsibility, attenuating their proper sense that they, and they alone, are responsible for finding the facts. Such specific application of expertise to the witnesses' credibility oversteps the proper bounds of expert testimony under Rule 702; interferes with the jury's core fact-finding function; and for these reasons has a probative value that is substantially outweighed by the danger of prejudice under Rule 403. To admit such pointed testimony would be an unwise exercise of my discretion under the circumstances of this case.
I rule that Dr. Strange's testimony shall be confined to explanation of the psychological principles and research regarding eyewitness identifications. As summarized by defense counsel, the relevant principles would involve summarizing the psychological research as to the effect of the following: Stress; Weapon focus; Disguise and obscured views; Time estimation and exposure duration; Retention interval; Post-event information; Voice identifications; Composites; Mugshot exposure effect. The government largely concedes that these subject areas fit the facts of this case. (I deal with specific objections to testimony regarding Voice identification and Composites below.)
My ruling here tends to minimize the significance of the government's associated application that Dr. Strange be admonished to confine herself to the facts in evidence. That is a principle that is worth observing. True, in general an expert may disclose facts upon which the expert relied in forming her opinion. Fed. R. Evid. 703. Particularly in a criminal case, however, there are great dangers in referring to evidence not properly admitted. The only example given by the government, however, is the presence or not of the defendant's "salt-and-pepper" goatee in a Composite, an issue dealt with below. For this aspect, I think that at-trial policing of the evidence should be adequate.
In light of those principles, I consider certain specific objections made by the government to portions of Dr. Strange's report, lettered (a) through (e), which it highlights in bold:
a. "Clearly there was a weapon in the second robbery, which likely was the focus of the eyewitnesses['] attention, at least for some portion of the total length of the event." ([Strange Rpt.] at 5 (emphasis added).)
To summarize the research about the effect of "weapons focus," for example, is both fair and helpful to the jury; I think it goes too far, however, to testify that, at this bank on this day, the weapon "likely was the focus of the eyewitnesses['] attention." (Strange Rpt. 5) Once "weapons focus" has been explained, a discussion of whether it compromised the witnesses' perception is properly left to summation, and to the jury.
b. The encounter lasted a short period of time. The actual time available to encode any reliable facial information would have been much less and is likely to have been focused on unique features (like the goatee), rather than more fine-grained facial features. That limited opportunity to observe the perpetrators would result in a reduced ability to accurately perceive a perpetrator's appearance." (Id. at 7 (emphasis added).)
I will permit testimony about amounts of time required to encode reliable facial information and the likelihood of focus on grosser features (such as a goatee), rather than finer ones. Application of those principles to the reliability of particular witnesses' identifications is likewise a matter for summation, and for the jury.
c. "Based on the record before me, an identification did not occur until several months after Robbery #2. That is certainly sufficient time to allow forgetting and memory distortion to affect accuracy rates." (Id. at 8 (emphasis added).)
Once again, the point can be made by instructing the jury as to what is known about rates of forgetting and distortion over time. But if the testimony is simply that these witnesses could have forgotten things over the course of several months, it is probably not properly the subject of expert opinion at all.
d. Here, the letter sent to the victims acts as a form of post-event information, providing a name that allowed the victim/eyewitnesses to do their own research online. That research likely irrevocably affected their memory of the perpetrator. Indeed, as described in the following sections, providing a composite of the assailant also likely affected their memory of the perpetrator's face." (Id. at 9 (emphasis added).)
The bold passage, finding it likely that the post-event information affected these witnesses' memories, again goes too far. The point can be made in a less intrusive way.
e. "The defendant was identified in unique fashion outside of a controlled police environment and in highly suggestive conditions where none of the known procedural safeguards (blind administrators, appropriate fillers, appropriate instructions) were followed." (Id. at 12 (emphasis added).)
Once again, I will permit some explanation of the reasons for the safeguards in a police line-up, because it is useful as an illustration. I will not permit an opinion that the conditions here were "highly suggestive" or an implication that this should be equated to an unconstitutional lineup.
2. Wrongful convictions
The government next argues that Dr. Strange should not be permitted to testify about wrongful convictions in other cases. I understand, of course, that the whole object and purpose of scrutinizing eyewitness testimony is the prevention of wrongful convictions. I will of course carefully instruct the jury as to the reasonable doubt standard and other safeguards against wrongful conviction in this case.
Dr. Strange's report refers to wrongful convictions in other cases. For example, one of the ways in which researchers have assessed the reliability of eyewitness identifications has been by correlating them with DNA-based exonerations. The principles of eyewitness identification emerging from such studies can be adequately explained, however, without reference to wrongful convictions in other cases.
Evidence of wrongful convictions in extrinsic cases has minimal probative value. The guilt or innocence of Sacco and Vanzetti, see Def. Brf. 4, has no bearing on the guilt or innocence of Mr. Jones, which will depend on the quality of evidence introduced in this trial. That minimal probativeness is substantially outweighed by the prejudicial emotional impact of a wrongful conviction. See Fed. R. Evid. 403. Nor do I wish to open the door to the government's responding by instituting a collateral mini-trial on whether the allegedly wrongful extrinsic convictions were in fact wrongful. I am determined that we shall try this case, not some other one. Such extrinsic evidence would result in jury confusion, waste of time, and undue delay, within the meaning of Rule 403.
I will not permit expert testimony about wrongful convictions in other cases.
3. Behavior or beliefs of jurors
Dr. Strange's testimony is helpful and pertinent insofar as it bears on the factors that affect the reliability of eyewitness identifications. I have permitted such testimony. Testimony as to juror behavior and attitudes, however, is a different matter.
First, the defense did not seek, and I did not grant, leave to delay trial to obtain an opinion about jury behavior and decision making. Second, I am not convinced from Dr. Strange's c.v. that her qualifications lie in that area.
Third, such opinions raise my concerns, already expressed, about intruding upon the jurors' fact-finding role. For example, Dr. Strange's Report seems to question or criticize the credulity of jurors. "Jurors," she writes, "Infer Accuracy from Confidence," or rely too heavily on this factor. (Strange Rpt. 12-13) Once again, summarizing psychological research about a lack of correlation between a witness's confidence and her accuracy is fair game; telling the jurors how to weigh such evidence is not.
I rule that Dr. Strange must confine her testimony to the analysis of eyewitnesses, not jurors.
4. Voice identification
The government objects to the portions of Dr. Strange's report that opine about the accuracy of voice identifications. Here, I generally side with the defense.
I will exclude Dr. Strange's opinion that people best recognize the voices of close associates, and are less certain about the voices of people they do not know. That is not a matter on which lay jurors need guidance.
The remainder of the testimony, however, would potentially be helpful. The studies upon which Dr. Strange relies, it is true, involve voice lineups, in which a witness is comparing exemplars. I accept that the situation here—in which ear-witnesses to two crimes, occurring four months apart, stated that the voice was the same—is different. Nevertheless, the research about accuracy rates and retention intervals may be helpful to the jury. To the extent that the real-life situation here is distinguishable from those in the voice line-up studies, that should provide grounds for cross-examination, not exclusion.
5. Composites
Composites, as referred to here, are what used to be called an Identikit or a police sketch—i.e., the assemblage of a portrait from a description of someone. Dr. Strange opines that the accuracy of composites in depicting the intended person tends to be quite low. I will not rule in limine on this aspect, without an evidentiary context.
The expert testimony issue must be distinguished from some related issues. The government proffers that it has no intention of introducing the composite in evidence. From that I gather there will be no argument that the identification or apprehension of Mr. Jones resulted from a witness's coming forward after seeing the composite. Still, the implication is that the composite was circulated publicly. The defense may wish to include a jury voir dire question to ensure that the jurors were not exposed to the composite before trial. Or the defense may wish to use the composite on cross-examination of a bank witness who participated in its creation, to suggest that the witness's contemporaneous physical description was not accurate. None of these issues, without more, really implicate the admissibility of Dr. Strange's expert testimony.
To the extent that the composite does not resemble Mr. Jones, the defense may not want to undermine it. The government, at any rate, responds that the same witnesses verbally described the salt-and-pepper goatee in a questionnaire, which would seem to be legitimate rebuttal. It strikes me that this evidence is of limited usefulness to either side. Still, that is for them to decide. --------
As to this limited aspect, I think that at-trial policing of the evidence will suffice. Counsel will alert the court before attempting to introduce any composite evidence, whether directly, on cross-examination, or via expert testimony.
ORDER
Accordingly, IT IS this 1st day of May, 2017
ORDERED that the submissions of the defendant (ECF no. 34) and the government (ECF no. 35), construed as cross-motions in limine, are GRANTED IN PART AND DENIED IN PART, in accordance with the foregoing Opinion.
/s/_________
KEVIN MCNULTY
United States District Judge