Wisconsin Cases November 27, 2019: United States v. Clayborne

Case Description

[425 F.Supp.3d 1048]

Benjamin P Taibleson, United States Department of Justice (ED-WI) Office of the US Attorney, Milwaukee, WI, for Plaintiff.

Daniel W Stiller, D Stiller LLC, Milwaukee, WI, for Defendant.

ORDER OVERRULING DEFENDANT'S OBJECTIONS (DKT. NOS. 83, 84), ADOPTING JUDGE JONES'S RECOMMENDATIONS (DKT. NOS. 65, 79) AND DENYING MOTIONS TO SUPPRESS EVIDENCE (DKT. NOS. 57, 58)

HON. PAMELA PEPPER, Chief United States District Judge

On January 25, 2019, the defendant filed two motions to suppress evidence. Dkt.

[425 F.Supp.3d 1049]

Nos. 57, 58. In his first motion, he asked the court to suppress cell-site location information (CSLI) obtained under two different orders issued under 18 U.S.C. § 2703(d). Dkt. No. 57 at 1–2. In the second motion, the defendant asked the court to suppress two eyewitness identifications that he asserted resulted from "unduly suggestive" procedures in violation of the Fifth Amendment of the Constitution. Dkt. No. 58 at 3. Magistrate Judge David E. Jones recommended that this court deny both motions. See Dkt. Nos. 65, 79. The court adopts Judge Jones's recommendations and denies both motions to suppress.

I. Standard of Review

Rule 59(b) governs a district court's referral of motions to suppress to magistrate judges. Fed. R. Crim. P. 59(b). Parties have fourteen days to file "specific written objections" to a magistrate judge's report and recommendation on a motion to suppress. Fed. R. Crim P. 59(b)(2). When reviewing a magistrate judge's recommendation, the district judge must review de novo the portions of the magistrate judge's recommendations to which a party timely objects. 28 U.S.C. § 636(b)(1) ; Fed. R. Crim. P. 59(b)(2), (3). The court can "accept, reject, or modify, in whole or in part, the findings or recommendations made by the magistrate." 28 U.S.C. § 636(b)(1).

II. Motion to Suppress Evidence (Orders Obtained Under the Stored Communications Act) (Dkt. No. 57)

A. Background Facts

On August 10, 2016, the government applied for an order under the Stored Communications Act, or "SCA" ( 18 U.S.C. § 2703(d) ), seeking more than a week's worth of CSLI about a cellphone belonging to the defendant, as part of an investigation relating to a carjacking. Dkt. No. 57-1 at 1-6. Magistrate Judge Joseph granted that application shortly after the government filed it. Dkt. No. 57-1 at 8. Over a year later, on November 27, 2017, the government applied for a second order in the same carjacking investigation. Dkt. No. 57-2 at 1-6. Magistrate Judge Duffin granted that application. Dkt. No. 57-2 at 8. For a more complete recitation of the facts, see Magistrate Judge Jones's Report and Recommendation, which the court adopts. Dkt. No. 65. The court will recount relevant facts below.

B. Procedural History and Arguments

In January 2019, the defendant moved to suppress "all evidence obtained as a direct and derivative result of the § 2703(d) orders." Dkt. No. 57 at 1. He argued that under the Supreme Court's decision in Carpenter v. United States , ––– U.S. ––––, 138 S. Ct. 2206, 201 L.Ed.2d 507 (2018), the court must exclude the CSLI evidence because the government obtained it without a warrant. Dkt. No. 57 at 1–2.

[425 F.Supp.3d 1050]

The defendant acknowledged the Seventh Circuit's decision in United States v. Curtis , 901 F.3d 846 (7th Cir. 2018), in which it held that courts were not required to exclude evidence obtained under § 2703 orders obtained before Carter was decided, but asserted that Curtis didn't apply because the government wasn't acting in good faith when it applied for the § 2703(d) orders. Dkt. No. 57 at 1–2.

Specifically, the defendant argued that the government knew when it applied for the orders that § 2703(d) was being challenged, and therefore that the government should have applied for a warrant rather than using the warrantless order procedure. Id. The defendant argued that the Supreme Court cases creating the good faith exception dealt with only police officers and law enforcement officials who could not be expected to know about legal challenges to laws. Id. at 8–9. He asserted that federal prosecutors, not law enforcement officers, apply for § 2703(d) orders. Id. The defendant argued that unlike law enforcement officers, it was the job of federal prosecutors and other government lawyers to stay abreast of changes in the law. Id.

The defendant also argued that the prosecutors in his case knew that § 2703(d) was under constitutional attack. Id. at 9–10. He asserted that in August 2016 and November 2017, when Judges Joseph and Duffin issued the § 2703(d) orders, (1) neither the Supreme Court nor the Seventh Circuit had addressed the law's constitutionality; (2) the Fourth Circuit had ruled the statute unconstitutional ("at least initially"); and (3) there was at least one challenge to the statute pending in this district. Id. The defendant contended that because government lawyers should be imputed with awareness of the cases, the government was not acting in good faith when it asked for the § 2703(d) orders. Id. at 8–9.

Magistrate Judge Jones recommended that this court deny the defendant's motion. Dkt. No. 65. He explained that even though § 2703(d) was under challenge when Judges Joseph and Duffin issued the orders, "it does not necessarily follow that the government obtained those orders in bad faith." Id. at 5–6. He pointed out that at the time of Judge Joseph's August 2016 order, the Eastern District of Michigan and the Sixth Circuit had rejected challenges to § 2703(d). Id. at 6. He noted that this court had held in United States v. Wheeler , 169 F. Supp. 3d 896 (E.D. Wis. 2016), that "cell users did not have a Fourth Amendment reasonable expectation of privacy in historical CSLI." Id. He pointed out that at the time of Judge Duffin's November 2017 order, the Supreme Court hadn't decided Carpenter , and the government had taken the position in briefing and oral argument before the Supreme Court that § 2703(d) orders were constitutional. Dkt. No. 65 at 6–7. For these reasons, Judge Jones concluded that the government did not act in bad faith when it requested the orders. Id.

The defendant objected. Dkt. No. 83. Rather than objecting to any specific portion of the recommendation, the defendant renewed his original arguments with two clarifications. Id. at 4 ("The totality of circumstances is set forth in great detail in all of the briefings previously filed herein. To rehash those facts would be superfluous and cumulative."). First, the defendant cited Ninth and Tenth Circuit case law for the proposition that the government bears the burden of proving good faith, and that it must do so by showing that its choice to seek § 2703(d) orders instead of warrants was "objectively reasonable." Id. at 4. The defendant argued that because the Supreme

[425 F.Supp.3d 1051]

Court had granted certiorari in Carpenter in June 2017, the government could not show prove that it was objectively reasonable to rely on the constitutionality of § 2703(d) orders when it applied for the November 2017 order. Id. at 5. Second, the defendant reiterated that the government had chosen to seek § 2703(d) orders rather than warrants, despite knowing that the Supreme Court had granted certiorari. Id. at 5. The defendant argued that the government should be "penalized" for this violation of the Fourth Amendment, not "lauded" for it. Id.

C. Analysis

The defendant's argument boils down to an assertion that when lower federal courts disagree about a statute's constitutionality, and the Supreme Court has accepted certiorari on the question, the government cannot in good faith rely on that statute. That is not and cannot be the law. Until the Supreme Court strikes down a statute, all parties—the government included—have a good-faith basis for relying on that statute. Exclusion of evidence is not "required when law enforcement agents act in good-faith reliance upon a statute or ordinance that is subsequently held to be unconstitutional. Nor is exclusion required when law enforcement agents act in good-faith reliance upon a statute or ordinance that is subsequently held to be unconstitutional. United States v. Peltier , 422 U.S. 531, 95 S.Ct. 2313, 45 L.Ed.2d 374 ... (1975), Michigan v. DeFillippo , 443 U.S. 31, 99 S.Ct. 2627, 61 L.Ed.2d 343 ... (1979)." Illinois v. Gates , 462 U.S. 213, 256, 103 S.Ct. 2317, 76 L.Ed.2d 527 (1983). The defendant's argument that government lawyers are not entitled to exercise that same reliance has no support in the law or in the chronology of this case.

In August 2016, there was no reason for the prosecutors to believe that § 2703(d) was constitutionally infirm. The defendant points to the Fourth Circuit's 2015 decision in United States v. Graham , 796 F.3d 332 (4th Cir. 2015) ( Graham I ), which held that § 2703(d) violated the Fourth Amendment. Dkt. No. 57 at 10. Graham , however, was not binding in the Seventh Circuit, and in March 2016—five months before Judge Joseph issued the August 2016 order—this court had rejected the Graham court's logic. See United States v. Wheeler , 169 F. Supp. 3d 896, 910 (E.D. Wis. 2016). As of Judge Joseph's August 2016 order, the Fifth and Eleventh Circuits had ruled that the statute was constitutional. In re U.S. for Historical Cell Site Data , 724 F.3d 600 (5th Cir. 2013) ; United States v. Davis , 785 F.3d 498, 511 (11th Cir. 2015). And in May 2016—three months before Judge Joseph issued the August 2016 order—the Fourth Circuit abrogated the portion of Graham I that held § 2703(d) unconstitutional. United States v. Graham , 824 F.3d 421, 438 (4th Cir. 2016) ( Graham II ) ("[T]he Government legally acquired [CSLI] records through § 2703(d) orders."). These cases were decided months before the government requested the first order in August of 2016. There was no circuit split at the time the government applied for the first order, and the weigh of authority was in favor of the constitutionality of the statute.

The defendant also argued to Judge Jones (though not in his objection to this court) that by the time the government applied for the August 2016 order, at least one defendant had challenged the constitutionality of § 2703(d) before a court in the Eastern District. Dkt. No. 57 at 9 n.3. The defendant cited United States v. Vic Chambers , Case No. 16-cr-13 at dkt. no. 81.

The attorney who originally represented the defendant and who filed the instant motion to suppress—Dan Stiller—also represented defendant Vic Chambers; the prosecutor in Chambers was AUSA Margaret

[425 F.Supp.3d 1052]

Honrath. On July 20, 2016—less than a month before the government applied for the August 2016 order in this case—Attorney Stiller filed a motion to suppress the CSLI evidence in Chambers , making a similar argument to the one he made in this case. Id. at Dkt. No. 81. Given that, the defendant's assertion that a defendant had made a challenge to § 2703(d) before a judge in the Eastern District of Wisconsin at the time the government applied for the August 2016 order is true—as far as it goes. But AUSA Honrath was not the prosecutor who applied for the order in this case; it was AUSA Mel Johnson. See Dkt. No. 57-1 at 6 in the instant case. The defendant provides no evidence indicating that on August 10, 2016, AUSA Johnson knew about the motion Stiller had filed three weeks earlier in a different case assigned to a different prosecutor. Even if AUSA Johnson had known about the motion filed in the Chambers case, the fact that another defendant had challenged the statute would not constitute proof that Johnson sought the § 2703(d) order in this case in bad faith, particularly given the fact that this court—Judge Pepper—had rejected a similar challenge in Wheeler in March of 2016.

On September 14, 2016, Judge Joseph recommended that the district judge deny Chambers' motion, Chambers , 16-cr-13 at dkt. no. 96; Chambers objected, id. at dkt. no. 99, and Judge Lynn Adelman accepted Judge Joseph's recommendation and denied the motion to suppress on October 26, 2016, id. at dkt. no. 111. A month later, on November 27, 2017, AUSA Rebecca Taibleson applied for the second § 2703(d) motion in this case. Dkt. No. 57-2 at 6. By the time AUSA Taibleson filed that second application, two judges in this district—Judge Pepper in Wheeler and Judge Adelman in Chambers —had rejected constitutional challenges to § 2703(d). While it is true that the Supreme Court had accepted certiorari in Carpenter five months earlier, that fact did not convince Judge Adelman to rule in Chambers's favor. And as Judge Jones pointed out, at the time AUSA Taibleson applied for the second order, the government had argued before the Supreme Court that § 2703(d) did not violate the Fourth Amendment.

The court agrees with Judge Jones that there is no evidence that the government acted in bad faith in applying for the two § 2703(d) orders. It was objectively reasonable for the government to rely on the constitutionality of § 2703(d) orders in August 2016 and November 2017. The court will adopt Judge Jones's recommendation and will deny the motion to suppress.

III. Defendant's Motion to Suppress Evidence (Eyewitness Identifications) (Dkt. No. 58)

A. Background Facts

The defendant has not disputed the facts Judge Jones laid out in his recommendation, so the court reproduces them here.

On July 7, 2016, Michael Guster became the victim of an attempted carjacking. See Defendant's Motion to Suppress Evidence 4-5. As Mr. Guster parked his car in his garage, two masked men approached with guns drawn. Id. at 4. The men fired shots, and a bullet went through both of Mr. Guster's legs. Id. at 5. Mr. Guster, legally carrying a handgun, shot back at the two men and hit one of them who was later identified as Eric Booker. Id. The two armed and masked men fled the scene and approximately ten minutes later, a man stripped down to his underwear showed up at the door of nearby resident, P.P. Def.'s Mot. to Suppress 5. The man asked P.P. to provide him with some clothes and then he retreated. Id. at 5-6.

Shortly after the attempted carjacking, Mr. Guster was interviewed by a detective. Mr. Guster described the person

[425 F.Supp.3d 1053]

that shot him as a "black male, 23-25 years old, medium complexion, medium build, 6'0-6'1?, 180-190 pounds, 1-2? afro, wearing a mask which covered his mouth and nose." Def.'s Mot. to Suppress 6. P.P., who was also interviewed by a detective, described her visitor as a "black male, 6'0, mid 30s, medium build, caramel complexion, bald, no scars or tattoos." Id.

By August 1, 2016, Jack Clayborne was identified as a suspect through video surveillance from July 7, 2016, which showed a wounded Eric Booker being dropped off at a hospital by someone driving a Ford Taurus with Illinois license plates. Def.'s Mot. to Suppress 6-7. Law enforcement ran the plates which listed to Jack Clayborne. Id. at 7. An FBI agent contacted Jack Clayborne who explained that he'd given his Taurus to his son—also named Jack—who took it to Milwaukee and returned it around July 10. Id.

In February 2018, law enforcement officers offered Mr. Guster and P.P. an opportunity to identify Mr. Clayborne as the person they observed. Def.'s Mot. to Suppress 7. FBI Agent Lucker and Detective [Klabunde] put together a photo array comprised of six photographs, one of Mr. Clayborne and five random others purported to resemble Mr. Clayborne. Id. at 8. Detective [Klabunde] first presented P.P. with form instructions and then he showed her the six photos in sequential fashion, with Mr. Clayborne's photo appearing fourth. Id. P.P. hesitated between pictures two and four, but described herself as "pretty certain" that the fourth photo depicted the unclothed man that came to her door. Id. Detective [Klabunde] repeated the process for Mr. Guster who also picked out Mr. Clayborne from the photo array because of his nose and mouth shape. Id. at 9.

Dkt. No. 79 at 2-3.

B. Arguments

The defendant argued to Judge Jones that the identification procedures which resulted in his identification were unduly suggestive. Dkt. No. 58 at 3. The defendant asked Judge Jones to find the identifications inadmissible, and to prohibit Guster and P.P. from identifying him at trial. Id. at 22.

The defendant made a "three-pronged argument" that the identification procedures were flawed. Id. at 3. First, he asserted that his case was "ill-suited" for a photographic identification procedure because the headshots used in photo lineups do not "capture certain aspects of a suspect's broader physicality," such as height, weight, scars or tattoos that the witness might have focused on. Id. at 9–11.

Second, the defendant argued that the procedures deviated from the best practices recommended by then-deputy Attorney General Sally Yates in her 2017 memorandum captioned "Eyewitness Identification: Procedures for Conducting Photo Arrays." Id. at 11. He asserted that the Yates Memo described the Department of Justice's accepted best practices for conducting photo identifications. Id. at 11. The defendant contended that, contrary to the recommendations in the memo, the photo array was not administered by someone uninvolved with the case, and the government had created the array based on the defendant rather than on the witness descriptions. Id. at 11–13. He pointed out that none of the photos depicted people with afros, even though it was the only above-the-neck characteristic described by either witness. Id.

Third, the defendant cited

[425 F.Supp.3d 1054]

United States v. Williams , 522 F.3d 809, 811 (7th Cir. 2008) in support of his argument that data demonstrated the photo array was unfair because it was skewed toward certain individuals, one of whom was the defendant. Id. at 13-19. The defendant began this argument by discussing a 1981 article written by a professor at the State University of New York, discussing "[t]he idea of testing a series of photos used for purposes of eyewitness identification for bias based on its functional or effective size ...." Id. at 14. The defendant described the professor's theory:

The inquiry is fairly simple. In instances where an eyewitness gives a description of a perpetrator and is later shown an array or series of photos from which the eyewitness makes an identification, the legitimacy of the process can be tested by providing the eyewitness's description, as well as the photos shown to that witness, to a statistically significant group of mock witnesses. Each mock witness is given the relevant eyewitness description and then called upon to view the photos with instruction to select the individual most closely matching the eyewitness's description of the perpetrator.

A scientifically effective photo array—one in which all persons included bear an adequate resemblance to the person described by a witness—would be expected to yield an outcome in which each photo is selected a roughly equal number of times. In that case, the array's effective size ( i.e. , its number of candidates viable in relation to the actual eyewitness's description (would be the same as its actual size ( i.e. , the number of photos comprising the array).

Things become different, though, where different photos within the array are picked at rates differing to a statistically significant degree ( i.e. , deviation from what would be expected by chance) from equal. When that happens, the photographed candidates who are selected a statistically significant fewer number of times are viewed as non-viable, a circumstance causing the overall array to have an effective size less than its actual size. And when the array's effective size decreases to a statistically significant degree below its actual size, the array is viewed as flawed.

As to any photo (or photos) selected with a statistically significant grater degree of frequency, the overall array is viewed as skewed towards any such photo (or photos) and away from other photos.

Id. at 14-15.

Based on this theory, the defendant's counsel commissioned a study by then-Ph.D. candidate Kaitlin M. Ensor of Stony Brook University. Id. at 15. See also , Dkt. Nos. 58-6, 58-7. He gave Ensor the six photos law enforcement had shown to Guster and P.P. and the verbatim descriptions they had given. Id. Ensor solicited and "qualif[ied]" two groups of forty-eight people as mock witnesses. Id. at 16. Each group received the photos in the order in which they appeared in the evidence; one group was given Guster's description, the other was given P.P.'s description. Id. The group given Guster's description selected one photo twenty-two times, another fourteen times, the defendant's eight times, another photo three times and another once; one photo was not selected by anyone. Id. at 17. The group given P.P.'s description selected the defendant's photo 27 times, while the photo selected next most often was picked only seven times and the remaining photos were selected between three and five times each. Id.

Ensor concluded from this data that only three of the six photos shown to the group given Guster's description were "viable" photos (one of which was the defendant), and that only one of the photos shown to the group given P.P.'s description—the

[425 F.Supp.3d 1055]

defendant's—was viable. Id. at 18-19. The defendant argued that this data showed that the photo array from which Guster and P.P. identified the defendant was unfairly biased against the defendant. Id.

C. Judge Jones's Recommendation

Judge Jones recommended that this court deny the motion. Dkt. No. 79 at 9. He recounted the standard for "unnecessarily suggestive" procedures under the Fifth Amendment:

The Supreme Court of the United States has long recognized a due process check on the admission of eyewitness identification, but only when the police have arranged suggestive circumstances leading the witness to identify a particular person as the perpetrator of a crime. See Perry v. New Hampshire , 565 U.S. 228, 232, 132 S.Ct. 716, 181 L.Ed.2d 694 (2012). To determine whether due process prohibits the introduction of an out-of-court identification at trial, the trial court must first decide whether the police used an unnecessarily suggestive identification procedure. Id. at 235, 132 S.Ct. 716. If they did, the court must next consider whether the improper identification procedure so tainted the resulting identification as to render it unreliable and therefore inadmissible. Id.

Dkt. No. 79 at 4. Judge Jones went on to say that "Mr. Clayborne's challenge fails at step one." Id.

Judge Jones first concluded that even if the headshots shown to the witnesses didn't capture characteristics such as height or build, "absent a showing of improper law enforcement activity," the question of whether the identifications were reliable was a question for the jury. Id. at 4–5. He also expressed doubt that "a photo array that fails to capture every aspect of a witness's description constitutes improper police conduct." Id. at 5.

As for the deviations from the practices recommended in the Yates memo, Judge Jones explained that having the photo array administered by someone uninvolved with the case wasn't the only way to achieve "blind"—un-suggestive—administration of the array. Id. at 5-6. He pointed out that the officers used another "blind" procedure—sequential administration, which involved "putting each photograph in its own folder, shuffling the folders, and standing where the administrator [could not] see which photographs the witness [was] viewing." Id. at 6. Judge Jones cited United States v. Ford , 683 F.3d 761, 765 (7th Cir. 2012), in which the Seventh Circuit opined that "[w]itnesses shown a sequential lineup are more likely to compare each person in it only with their memory of the offender, rather than choose whichever person looks the most like what the witness remembers." Id.

As to the defendant's argument that the law enforcement officers had selected the filler photos to match the defendant, rather than to match Guster's description of the perpetrator (in which Guster indicated that the perpetrator had a 1-2? afro), Judge Jones noted that selecting filler photos that "generally fit the witness's description" ... "is not always practicable." Id. at 7. The government had explained in its response to the motion to suppress that the defendant "had a shaved head by the time the photo was taken that was used in the photo array," dkt. no. 60 at 6, and Judge Jones noted (as had the government) that the instructions that accompanied the array told the witnesses that characteristics such as hairstyles might be different in photos, dkt. no. 79 at 7. He concluded that the difference in hairstyle did not render the array unduly suggestive. Id.

[425 F.Supp.3d 1056]

Finally, Judge Jones found the data produced by the study the defendant commissioned from Ensor "probing." Id. at 9. But he noted that

the article that the study hinges on repeatedly warns that "a statistically significant departure does not necessarily indicate a meaningful departure from fairness in the legal context." ECF No. 58-4:6. Consequently, the study's conclusion that the photo arrays utilized in this case were flawed does not itself advance the conclusion that law enforcement used an unnecessarily suggestive identification procedure. This is especially given that no federal case known to the Court has discussed the use of effective size assessments to establish the suggestiveness of an identification procedure. See Response 17.

Id.

D. The Defendant's Objection

The defendant objected both generally and on three specific grounds. Dkt. No. 84.

The defendant first argued that Judge Jones's conclusions regarding the defendant's "ill suited" argument did not "fully take into consideration the specific facts" of his case. Id. at 3. He argued that law enforcement didn't have to use a photo array to identify him. Id. at 3-4. He asserted that while P.P. described someone who was six feet tall, with a medium build and no scars or tattoos, none of the photos in the array provided information about the subjects' height, build or lack of scars or tattoos. Id. at 4. He asserted that Guster described a masked perpetrator, which meant that his description "was limited to non-facial characteristics: ‘black male, 23-25YOA, medium complexion, medium build, 6'00-6'1?, 180-190 lbs, 1-2? afro, black shirt and tan shorts, armed with a black handgun. Wearing a mask of some sort, which covered his nose and mouth." Id. The defendant asserted that while the subjects in the photo array were black males, and that they arguably had medium complexions, they didn't show height, weight, clothing or weapons, and none of them had a 1-to-2-inch afro. Id.

Next, regarding his Yates memo argument, the defendant asserted that Judge Jones "overlooked the fact that the DOJ procedures specifically call for ‘blind administration’ of photo arrays." Id. at 6. The defendant explained the purpose of the "blind administration" recommendation—to make sure that the administrator of the array can't knowingly or unknowingly cue the witness as to the identity of the suspect. Id. He also repeated his assertion that the officers selected the fillers based on similarity with the defendant, rather than similarity to the descriptions given by Guster and P.P. Id. He concluded by asserting that while sequential administration "maybe a manner upon which to conduct a photo array, it is not the recommended manner prescribed by the DOJ." Id. at 7. He contended that law enforcement chose convenience over following the recommended DOJ procedures, and asserted that Judge Jones didn't take this into account. Id.

Third, regarding the "effective size" study the defendant had commissioned, he argued that Judge Jones "failed to take into consideration other salient factors in the study such as the data and the interpretation of that data as relates specifically to this case." Id. at 8. He asserted that he had "spelled out in excruciating detail" those data and their interpretation in his initial filing, and criticized Judge Jones for dismissing it solely because "a statistical significant departure may not lead to unfairness." Id. The defendant acknowledged that no federal case had relied on effective size assessments in finding an identification procedure unduly suggestive, but he urged this court to do so because "the law is ever evolving," asserting—without citing

[425 F.Supp.3d 1057]

authority—that "[t]he U.S. Supreme Court has recently stated that even precedent (i.e. stare decisis) may no longer be applicable as times change in this country." Id. at 8.

E. Analysis

In 2012, the Supreme Court stated in Perry v. New Hampshire , 565 U.S. 228, 232, 132 S.Ct. 716, 181 L.Ed.2d 694 (2012), that it had "recognized ... a due process check on the admission of eyewitness identification, applicable when police have arranged suggestive circumstances leading the witness to identify a particular person as the perpetrator of a crime." The Court explained that

[w]e have not extended pretrial screening for reliability to cases in which the suggestive circumstances were not arranged by law enforcement officers. Petitioner requests that we do so because of the grave risk that mistaken identity will yield a miscarriage of justice. Our decisions, however, turn on the presence of state action and aim to deter police from rigging identification procedures, for example, at a lineup, showup, or photograph array. When no improper law enforcement activity is involved, we hold, it suffices to test reliability through the rights and opportunities generally designed for that purpose, notably, the presence of counsel at postindictment lineups, vigorous cross-examination, protective rules of evidence, and jury instructions on both the fallibility of eyewitness identification and the requirement that guilt be proved beyond a reasonable doubt.

Id. at 232-233, 132 S.Ct. 716. The reason for linking exclusion to improper police conduct "is to deter law enforcement use of improper lineups, showups, and photo arrays in the first place." Id. at 241, 132 S.Ct. 716. Even when an identification procedure is "infected by improper police influence," the Court said, it was not "automatically excluded;" only if there is "a very substantial likelihood of irreparable misidentification" must the court prohibit the government from presenting it at trial. Id. at 232, 132 S.Ct. 716, quoting Simmons v. United States , 390 U.S. 377, 384, 88 S.Ct. 967, 19 L.Ed.2d 1247 (1968). Courts use a totality-of-the-circumstances approach to determine whether a very substantial likelihood of irreparable misidentification exists. Id. at 239-40, 132 S.Ct. 716.

In his motion to suppress, the defendant argued the unreliability of eyewitness identifications. Dkt. No. 58 at 1-3. He argued that using an array of headshot photos wasn't a process well suited to accurate identification in a case where witnesses had provided descriptions of height, weight and build. Id. at 9-11. He argued that the two law enforcement officers who administered the photo array were involved in the investigation and knew which photo was the defendant's. Id. at 11-12. He argued that the officers did not use photos of people with afros, even though Guster had described the person who shot him as having a 1-2? afro. Id. at 13. And he argued that the study he commissioned demonstrated that while the officers showed Guster and P.P. six photos, Guster effectively selected from only three valid photos, and P.P. effectively had only one valid photo available to her. Id. at 13-17. The only one of these arguments that came even close to an allegation that the officers "rigged," or improperly arranged, the array was his argument that the officers who administered the array were involved in the investigation.

In his reply brief in support of the motion, the defendant spent most of his ink defending the validity of his effective size assessment study. Dkt. No. 64. Toward the end of that reply, he argued that that Guster's identification was unreliable because Guster had described the person

[425 F.Supp.3d 1058]

who attacked him as having a one-to-two-inch afro, while the photo of the defendant in the array showed the defendant with a bald head. Id. at 7. The defendant repeated a claim he'd made in his motion—that he was bald "as a result of male hair loss." Id. at 7, referencing Dkt. No. 58 at 13 (in which the defendant described himself as "a man who experiences male pattern baldness "). He conceded that P.P. had described the man who appeared at her house as bald, but argued that this showed that Guster and P.P. had seen different men. Dkt. No. 64 at 8. In support of this argument, the defendant also pointed to the fact that Guster had described someone 23 to 25 years old, while P.P. described someone in his mid-thirties. Id. He asserted that Guster had only a brief opportunity to view the person who shot him (given that Guster was looking at two men and was taking cover and loading his own weapon during part of the interaction). Id. Adding the fact that Guster said the person who shot him wore a mask that covered his nose and mouth, the defendant argued that "there exists a legitimate reason in terms of the reliability of the eyewitness identifications to conclude that one of [the witnesses] got it wrong," and that the witness who likely got it wrong was Guster. Id. at 8-9.

The defendant's reply memo was based on challenges to the reliability of the identifications, without any allegation that the officers rigged or impermissibly arranged the array. The defendant's objection to Judge Jones's recommendation reiterates the arguments he made in his prior pleadings; the closest it comes to alleging improper police conduct is his assertion that the officers weren't required to use an eyewitness identification procedure, and his claim that the officers "strayed" from the Yates memo recommendations.

The first step of a suggestiveness inquiry is that the court must determine "whether the police used an unnecessarily suggestive identification procedure." Perry , 565 U.S. at 235, 132 S.Ct. 716. Due process concerns arise "only when law enforcement officers use an identification process that is both suggestive and unnecessary." Id. at 238-39, 132 S.Ct. 716. If the police did not use an unnecessarily suggestive procedure, the court does not move on to the second step—whether that "improper identification procedure so tainted the resulting identification as to render it unreliable and therefore inadmissible." Id. at 235, 132 S.Ct. 716.

Judge Jones found—and this court agreed—that the defendant's motion fails at the first step of the inquiry. The court starts with P.P.'s identification and looks to the defendant's own description of that process. The defendant says that in the winter of 2018—just shy of a year and a half after the shooting—an FBI agent told Detective Klabunde that the defendant was a suspect. Dkt. No. 58 at 7. According to the defendant, "reports authored on and after July 7, 2016, reflect Detective Klabunde's significant role in the investigation's early stages," but he fails to describe that role. Id. at 8 n.7. Klabunde put together the photo array. Id. at 8.

P.P. identified the person who came to her door as being bald; all six photos depict men with no hair. Dkt. No. 58-1. She described a black male; all six photos depict black males. She indicated that the person had no scars or tattoos; none of the photos show men with facial scars or tattoos (although the person in the fifth photograph

[425 F.Supp.3d 1059]

appears to have an injury on his right cheekbone). P.P. described the person who came to her door as having a "caramel" complexion; the men in the first and fifth photographs appear to have lighter skin tones than the other four men, and only the witness could know whether any of them matched what she considered to be a "caramel" skin tone. P.P. described someone in his mid-thirties; the men in the photos do not appear to be teenagers or middle-aged to elderly men. There is no distinguishing characteristic in the defendant's photo, nothing about him that makes him stand out from the other five subjects.

Klabunde and the FBI agent visited P.P. on February 26, 2018. Dkt. No. 58 at 8. Klabunde gave P.P. the array instructions, then showed her the six photos "in sequential fashion," with the defendant's photo being the fourth of the six. Id. Klabunde indicated that P.P. hesitated on the second photo and said she wanted to return to it, then hesitated on the fourth photo (the defendant's), saying she wanted to return to it. Id. She then indicated that she was "pretty certain" that the fourth photo (the defendant's) showed the man who had come to her door. She said that the second photo looked similar, but that the person in the second photo had "a thinner face." Id. She then "became ‘certain the person in folder #4 was the same person she saw on 07-07-16 because the shape of the head, skin complexion, shape of the nose, style of facial hair, and physical build were all the same.’ " Id. at 8-9.

Nothing in this recitation indicates that the FBI agent or Klabunde did anything manipulative or suggestive. The defendant asserts that Klabunde knew which photo was the defendant's, but there is nothing in his description of what Klabunde did that indicates Klabunde gave away his knowledge to P.P. In fact, it appears that P.P. was considering both the second and the fourth photos, but that she selected the fourth photo due to specific characteristics.

The government's response added details about the identification procedure, which support the conclusion that the agent and Klabunde did nothing suggestive. The government recounts that each of the six photos was in its own folder, "shuffled by the officer, and not visible to the law enforcement officer at the time it was visible to the witness," which meant that the officer didn't know the order of the photos until the witness had viewed all six of them. Dkt. No. 60 at 2-3. Neither the agent nor Klabunde made any suggestive comments during the array. Id. at 3. And the instructions given to P.P. advised her that she "should not feel you have to make an identification. It is as important to exclude innocent persons as it is to identify the perpetrator." Id. , citing Dkt. No. 58-2.

In the absence of evidence of an unnecessarily suggestive identification procedure, the court need not go on to consider whether P.P.'s identification of the defendant was tainted; there was no unnecessary suggestion by which her identification could be tainted.

Next, the court turns to Guster's identification. According to the defendant, the FBI agent and Klabunde visited Guster two days later, on February 28, 2018, and gave him the same instructions they'd given P.P. Dkt. No. 58 at 9. Klabunde then used the same procedure he had used with P.P.—he showed Guster "the same six photos which he again presented in sequential fashion." Id. This time, the defendant's photo was third in the array, and

[425 F.Supp.3d 1060]

according to Klabunde, when Guster reached it, he selected the defendant. Id.

The defendant has alleged that Klabunde selected filler photos that matched the defendant, rather than filler photos that matched Guster's description of a man in his early to mid-twenties with a one-to-two-inch afro. He ignores the fact that Klabunde did select photos that matched the witness's description—just as the Yates memo recommended—but the witness description was P.P.'s. Klabunde showed that same array to Guster, along with instructions telling Guster that the person who shot him "may or may not be included in the photos," and that "[y]ou should not feel you have to make an identification." Dkt. No. 58-2. There is no evidence that Klabunde said, "I know you described a person who had an afro, but keep in mind that he might have shaved his head since you saw him," or anything similar. The record indicates only that Klabunde showed Guster each of the folders after they'd been shuffled, and that Guster selected the defendant. Nothing about this process was unnecessarily suggestive.

In urging the court to find that the identification processes were unduly suggestive, the defendant focused on two of the Yates memo recommendations—the recommendation that the filler photos should generally match the witness's description of the suspect and the recommendation that the administrator avoid suggesting to the witness which photo contains the suspect. The Yates memo contained over thirty recommendations, and the procedures the officers used with both P.P. and Guster followed most of them. Dkt. No. 58-3. Neither the defendant nor photos of the defendant were nearby when the officers conduct the arrays (Yates recommendation 1.2). Only one suspect was included in the array (Yates recommendation 2.1). As the court has stated, the photos resembled P.P.'s description of the perpetrator (Yates recommendation 2.2). The array had at least five filler photographs (Yates recommendation 3.1). The array did not point to or suggest the defendant to the witness (Yates recommendation 3.2). The photos were of similar size, background, format and color, and bore no labels that reflected the subjects' identities or the sources of the photos (Yates recommendation 3.4). There is no need for the court to go through all the recommendations to make its point.

The recommendations allow administrators discretion to employ either sequential or simultaneous procedures (Yates recommendation 4.1); Klabunde used the sequential procedure. Judge Jones cited the Ford case, in which the Seventh Circuit speculated that the sequential procedure might encourage witnesses to compare photos to their memories, rather than to the other photos. The defendant points to United States v. Johnson , 745 F.3d 227, 229 (7th Cir. 2014), in which the court noted that "some recent research has called into question the view that sequential presentation of photographs is superior to photo spreads." He neglects the part of the Johnson opinion which explains that "[t]he Supreme Court has not adopted a rule that only ‘the best’ approach (as the latest social science research identifies the best current understanding) can be used," and that Perry precludes a federal court of appeals from requiring one approach or the other. Johnson , 745 F.3d at 229.

Yates memo recommendation 5.1 is one of the two recommendations the defendant alleges that the FBI agent and Klabunde didn't follow. It reads as follows:

The administrator must ensure that he or she does not suggest to the witness—even unintentionally—which photograph contains the image of the suspect. Oftentimes, the best and simplest way to achieve this is by selecting an administrator

[425 F.Supp.3d 1061]

who is not involved in the investigation and does not know what the suspect looks like.

Dkt. No. 58-3 at 4. This is not, as the defendant would have the court believe, a mandate that the administrator must not be involved in the investigation. And the memo states, on the first page, that it is "not intended to create, does not create, and may not be relied upon to create any rights, substantive or procedural, enforceable at law by any party in any matter civil or criminal. Nothing in these procedures implies that an identification not done in accordance with them is unreliable or inadmissible in court." Id. at page 3, n.1.

The defendant's arguments that the head shots did not show height, weight or build and that Guster's identification of the defendant was not reliable for various reasons go to whether the identifications were so tainted by any unnecessary suggestion that they were unreliable. Because there was no unnecessary suggestion, any errors in the identifications can't be chalked up to suggestion. The defendant will have at his disposal at trial the tools Justice Ginsberg described in Perry for attacking the reliability of the identifications—vigorous cross-examination, protective rules of evidence and jury instructions on the fallibility of eyewitness identification. Absent unnecessary and undue suggestion, there is no reason for this court to use the tool of exclusion to deter law enforcement from using unduly suggestive procedures.

A final note on the defendant's effective size study. Judge Jones characterized the results of the study as "probing," and it provides food for thought about how humans process verbal descriptions of physical characteristics when comparing those verbal descriptions to visual representations. But while Ensor concluded from the data that the array "was not fair, based on flawed selection of lineup candidates"—a conclusion reached by using tools such as "Pearson's Chi-Square ... goodness-of-fit test" and "Tredoux' E," which this court is woefully unqualified to analyze—she provided no insight into what rendered the array allegedly unfair, or how Klabunde or any other officer would have had any way to know that it was allegedly unfair. She did not state what it was about the array that was unfair, and with the court's limited understanding of what Ensor did, it can't imagine how the study would have revealed the answer to that question.

The defendant started his discussion of effective size of photo arrays by stating, "[t]he Seventh Circuit has explained that ‘it takes data rather than intuition’ to answer questions about circumstances capable of contributing to misidentification. United States v. Williams , 522 F.3d 809, 811 (7th Cir. 2008)." Dkt. No. 58 at 13. The defendant lifted this fragment out of the context of the entire sentence, and out of the context of the facts, of Williams.

Williams moved to suppress three witness identifications resulting from a lineup in which five participants wore navy blue slippers and the sixth—Williams—wore white tennis shoes. Williams , 522 F.3d at 810. He argued that the fact that he alone was wearing white tennis shoes made the lineup unduly suggestive. Id. The Seventh Circuit began by observing that even if the white tennis shoes made the lineup suggestive, "suggestiveness is only part of the legal standard," and it pointed out that courts should exclude eyewitness identifications "only if the pretrial procedure ‘was so impermissibly suggestive as to give rise to a very substantial likelihood of irreparable misidentification.’ " Id. , quoting Simmons v. United States , 390 U.S. 377, 384, 88 S.Ct. 967, 19 L.Ed.2d 1247 (1968).

Nonetheless, the Seventh Circuit considered whether the lineup was unduly suggestive. Williams's attorney argued that it was "common sense" that the white tennis

[425 F.Supp.3d 1062]

shoes rendered the lineup suggestive. Id. at 811. The Seventh Circuit disagreed, asserting that "[i]f there is one thing known about eyewitness identification, it is that ‘common sense’ misleads more often than it helps." Id. (citations omitted). The court explained that experience makes people confident that they've seen someone before, without giving them any reliable method of testing whether the confidence is justified, observing that

[p]eople confuse certitude with accuracy and so are led astray. Psychologists have established that certitude often is unwarranted. It takes data rather than intuition to answer questions such as "can non-uniform footgear in a lineup lead to misidentification?"

Id. For this reason, the court asked the parties to submit memos "addressing the social-science literature," but Williams's counsel told the court he had nothing to add. Id. The government submitted a memo, which the court said contained studies that discussed "optimal identification procedures ... but [did] not address the question at hand: do differences in the clothes worn by participants in lineups lead to false identifications?" Id.

In Williams , the defendant presented the court with a factor—the fact that he wore different shoes from the other five participants in the lineup—but no data to show that that factor was unduly suggestive. In other words, the defendant said, "there was a difference between me and the other participants in the lineup," without showing that that difference made it more likely that witnesses would have mistakenly selected the defendant. In this case, the defendant presents data that he claims shows that the array made the witnesses more likely to select him, without tethering that data to any difference between his photo and the photos of the other five subjects or anything about the identification procedures themselves. The Williams court concluded that a difference without data to show its impact didn't establish suggestiveness; this court concludes that data without an identified difference does not establish suggestiveness.

IV. Conclusion

The court OVERRULES the defendant's objections to Magistrate Judge Jones's recommendations. Dkt. Nos. 83, 84.

The court ADOPTS Judge Jones's recommendations. Dkt. Nos. 65, 79.

The court DENIES the defendant's Motion to Suppress Evidence (Orders Under the Stored Communications Act). Dkt. No. 57.

The court DENIES the defendant's Motion to Suppress Evidence (Eyewitness Identifications). Dkt. No. 58.