Michigan Supreme Court Lansing, Michigan
Megan K. Cavanagh, Chief Justice Brian K. Zahra Richard H. Bernstein Elizabeth M. Welch Kyra H. Bolden Kimberly A. Thomas Noah P. Hood,
Justices
Order
March 20, 2026
169071
PEOPLE OF THE STATE OF MICHIGAN,
Plaintiff-Appellee,
v
MAURICE DESHAWN LEE,
Defendant-Appellant.
SC: 169071
COA: 375227
Wayne CC: 22-007738-FC _________________________________________/ On order of the Court, the application for leave to appeal the July 30, 2025 order of the Court of Appeals is considered, and it is DENIED, because we are not persuaded that the question presented should be reviewed by this Court.
WELCH, J. (concurring).
In this case, defendant alleges that his jury venire lacked Black male representation. While I concur in the Court's denial order, I write to highlight the seemingly impossible burden we have placed on defendants who argue that a jury venire is not composed of a fair cross section of the community. Defendants who wait to challenge a specific venire after it is selected are told their claims cannot survive because underrepresentation of a specific group in a single venire results from chance. Yet, when a defendant like the one here challenges the venire early—allowing more time for discovery to obtain data of underrepresentation in venires over time—the claim is rejected as unripe. Additionally, we require defendants to set forth specific data from venires over time in order to succeed on a challenge—but that data is largely unavailable to the public. We can and should do better.
I. FACTS
Defendant Maurice Deshawn Lee is awaiting trial on charges of armed robbery (MCL 750.529) and felonious assault (MCL 750.82) in the Third Circuit Court (Wayne County). Before any jury pool was summoned, defendant moved for an evidentiary hearing to challenge the racial composition of his jury venire. Defendant alleged systematic underrepresentation of Black men in Third Circuit Court jury venires and sought an evidentiary hearing to produce data upon which he can base his "fair cross section"
challenge. The Third Circuit denied the motion, explaining that the challenge was not ripe
1
2
because there was not yet a venire in defendant's case.1Defendant filed a motion for reconsideration, which the court also denied. The Court of Appeals denied leave to appeal in a substantive order. Defendant now seeks leave to appeal in this Court.
II. BACKGROUND
The Sixth Amendment of the United States Constitution provides that "[i]n all criminal prosecutions, the accused shall enjoy the right to a speedy and public trial, by an impartial jury . . . ." The Sixth Amendment's right to a jury trial applies to the states through the Fourteenth Amendment. See Duncan v Louisiana, 391 US 145, 156 (1968). Likewise, the Michigan Constitution promises trial by an impartial jury. Const 1963, art 1, § 20. An "essential component" of this right to a jury trial is selection of jurors "from a representative cross section of the community . . . ." Taylor v Louisiana, 419 US 522, 528 (1975). The guarantee of a representative cross section does not mean that a specific jury must be representative of the community. Id. at 538. Rather, it means that "the jury wheels, pools of names, panels, or venires from which juries are drawn must not systematically exclude distinctive groups in the community and thereby fail to be reasonably representative thereof." Id.
Under federal law, to make out a prima facie case alleging a violation of the fair cross section requirement, a defendant2must show:
(1) that the group alleged to be excluded is a "distinctive" group in the community; (2) that the representation of this group in venires from which juries are selected is not fair and reasonable in relation to the number of such persons in the community; and (3) that this underrepresentation is due to systematic exclusion of the group in the jury-selection process. [Duren v Missouri, 439 US 357, 364 (1979).]
Once the defendant has made a prima facie case, the burden shifts to the state to show that a significant state interest justifies the systematic exclusion. Id. at 368. The Supreme Court has not adopted a specific statistical test or threshold to evaluate what constitutes "fair and reasonable" under the second prong of the Duren test. Berghuis v Smith, 559 US 314, 329-330 (2010). In Duren itself, however, the Court relied on a simple "absolute disparity" test, noting the difference between the distinct group in the
2
3
community at large and in jury venires during the time of the defendant's trial. See Duren,
439 US at 361-363. In that case, a Louisiana statute gave women an automatic exemption from jury service on request, and the court treated women who did not respond to summons as having claimed this exemption. Id. at 361-362. The Supreme Court considered the statistics that the defendant presented: census data showed that women were 54% of the adult inhabitants in the county where the defendant's trial was held, yet during the relevant periods, approximately 15% of people in venires were female. Id. at 362-363. Without further statistical analysis, Duren concluded that "[s]uch a gross discrepancy between the percentage of women in jury venires and the percentage of women in the community requires the conclusion that women were not fairly represented in the source from which petit juries were drawn . . . ." Id. at 366.
On the third prong, the Duren Court emphasized the duration of the exclusion. It concluded that the defendant's "undisputed demonstration that a large discrepancy occurred not just occasionally, but in every weekly venire for a period of nearly a year manifestly indicates that the cause of the underrepresentation was systematic—that is, inherent in the particular jury selection process utilized." Id. at 366. This Court has since elaborated on the second Duren prong, layering additional statistical requirements onto the United States Supreme Court's framework. First, this Court held that "a court must examine the composition of jury pools and venires over time using the most reliable data available to determine whether representation is fair and reasonable." People v Bryant, 491 Mich 575, 599-600 (2012).3Effectively, this means
3
4
that defendant must show both an individual and systematic harm—he must show that the venire from which his jury was called is unrepresentative and that venires in the circuit generally are unrepresentative. Second, this Court mandated that courts apply "all the relevant statistical tests for evaluating the data regarding representation of a distinct group without using any one individual method exclusive of the others." Id. at 582. In Bryant, this Court canvassed several statistical methods. First, it acknowledged that the absolute-disparity test—the difference between the percentile representation of the group in the relevant community and in jury venires—can make it impossible for a defendant to show underrepresentation if the relevant group makes up a small share of the population. Id. at 604. However, it reiterated that even when a group is small, a court must consider the results of the absolute-disparity test in conjunction with the other tests. Id. Second, the Court explained that the "comparative disparity" test often skews and overstates underrepresentation and accordingly should be given little weight. Id. at 608-
609. Third, the Court discarded the "standard deviation" test, proposing that it is a measure of the randomness of a given disparity, not the extent or fairness of the disparity. Id. at
610. Fourth and finally, it noted that the "disparity of risk" test was another statistical test for the court to consider. Id. at 615. Yet, without further reasoning or reliance on statistical expertise, it held that only disparities of risk exceeding 50% should be deemed unfair and unreasonable. Id.4
III. DISCUSSION
I am concerned that we have created an impossible burden for any defendant alleging a fair cross section claim and that meritorious claims can never meet the threshold we have established. While I question whether Bryant was correctly decided, this Court made clear that a defendant must present multiple statistical analyses of a jury venire and use "the most reliable data" in order to make a prima facie fair cross section challenge. Bryant, 491 Mich at 583.5In reality, however, the lack of publicly available data proves another." Bryant, 491 Mich at 638 (KELLY, J., dissenting). In other words, in Justice KELLY's view, the court should be able to determine that one test is more appropriate or more accurate for measuring the disparity given the facts in a specific case.
4
5
fatal for virtually every defendant challenging a jury venire before a court can ever reach the merits.6Defendants get caught in a procedural trap whereby it is hard (if not impossible) for them to access the data they need to prove a claim—no matter when they raise the claim.
Defendants making a claim after a venire is assembled are told their claims cannot survive because the state mandates a facially neutral, randomized jury selection process and underrepresentation in a single venire results from chance.7Further, courts are under pressure to keep trials moving and are understandably reluctant to stop proceedings to allow the extensive discovery that would be necessary to prove disparities over time. For example, in People v Williams, the defendant raised his fair cross section challenge right after the venire was assembled. People v Williams, unpublished opinion of the Court of Appeals, issued November 5, 2025 (Docket Nos. 361717 and 361718), pp 6-8. The trial court granted an evidentiary hearing that very day, but the defendant had little time to prepare. Id.
After hearing from the director of jury services, the Williams Court emphasized the facially neutral process for assembling juries and dismissed the defendant's challenge, to use the zip code of each person sent jury summons and census data showing the racial makeup of each zip code to estimate the number of African-Americans summoned over the course of three months. Id. at 602-603.
5
6
stating that "it was a randomized jury pool." In other words, the quick evidentiary hearing held on the day of trial did not give the defendant a chance to uncover and analyze any larger dataset that would show representation over time. The Court of Appeals held that the defendant was not entitled to relief on appeal because he "presented no data concerning 'the composition of venires over a significant time period rather than just the defendant's individual venire.' " Id. at 8, quoting Bryant, 491 Mich at 600 (emphasis added). While the Court of Appeals was correct that the defendant presented no data, how was the defendant supposed to obtain and analyze such data in mere hours?
Yet, in an effort to obtain necessary data, defendants making an earlier claim before a single venire is assembled are told their claims are unripe. For example, in this case, defendant moved for data about larger trends over time prior to trial. Defendant's motion included requests for jury qualification questionnaires, geographic information for each juror summoned, geographic information on jurors summoned who did not respond, and geographic information where summons were returned as undeliverable for all of the Third Circuit between 2022 and 2024. The court suggested that defendant was "certainly free to bring his challenge if and when his claim ripens," but stated that it would not hold an evidentiary hearing before a venire was assembled because "[n]either the law nor principles of judicial economy support" extensive discovery for a claim that is "purely speculative."
Generally, I believe that our practices safeguard defendants' Sixth Amendment rights the vast majority of the time. I also note that efforts are underway to ensure more transparency and data access in our courts. Indeed, our State Court Administrative Office recently partnered with the Criminal Justice Administrative Records System at the University of Michigan Institute for Social Research to evaluate the representativeness of juror pool data at the congressional district level. The results were clear: "[M]ost comparisons between the juror pool and population estimates were very similar or within a few percentage points."8Creating this snapshot, however, required access to confidential data and review by academics who are experts in data gathering. Further, if a disparity existed, the report alone would likely be insufficient to meet the requirements of Bryant because it represents only a snapshot in time.
Defendants, and the broader community, need ready access to jury demographic data to monitor and ensure that the jury selection process remains fair and representative— and can raise a challenge when it is not. While randomized jury selection procedures are necessary to maintaining a fair system, they are not alone sufficient. We know that there
6
7
are mistakes and glitches in such processes.9Indeed, we need look no further than Bryant itself, in which a computer programming error resulted in the jury pool being drawn from a subset of zip codes and "a disproportionately larger number of jury questionnaires [went] to zip codes with smaller African-American populations and disproportionately fewer questionnaires [went] to zip codes with larger African-American populations." Bryant,
491 Mich at 588.10The error in Bryant only came to light two years later "when a local high school teacher, Wayne Bentley, completed a study of minority representation on Kent County juries." Ambrose v Booker, 684 F3d 638, 641 (2012).
Further, facially neutral systems can systematically exclude certain groups in subtle and shifting ways. Certain communities are already less likely to have driver's licenses or state IDs, and consequently less likely to be on the list of possible jurors.11And, even when in the system, some groups are less likely to have an up-to-date address on file with the Secretary of State.12Given these headwinds, it is necessary to ensure that the jury pool list remains at representative levels over time.
7
I, Elizabeth Kingston-Miller, Clerk of the Michigan Supreme Court, certify that the foregoing is a true and complete copy of the order entered at the direction of the Court. March 20, 2026
t0304
8
Clerk In 2021, the New Jersey Supreme Court observed that a fair cross section guarantee in the absence of readily available data was nothing more than an illusory promise. State v Dangcil, 248 NJ 114, 146 (2021). Stressing that "the validity of a jury-selection process is not static," they urged that "[o]ur courts, jury managers, and stakeholders must remain vigilant in the event that the facially neutral practices we approve of here result in underrepresentation of a cognizable group despite all best efforts." Id. And, critically, they ordered their state court's administrative office to begin collecting juror demographic data. Id. As of 2023, court staff provide that aggregated data if requested by an attorney in any jury trial.13
New Jersey provides an excellent example of using and providing data in a manner that ensures public trust in our institutions. Our State Court Administrative Office is already leading on many data efforts. I believe that Michigan, like New Jersey, can also work to ensure better access to jury venire data to build public trust in our courts. Accessible and trusted data at least will enable litigants to check the validity of a venire and make challenges on the merits.
BOLDEN, J., joins the statement of WELCH, J.
8