STATE OF IOWA, Appellee, vs. KENNETH LEE LILLY, Appellant.
No. 20–0617
IN THE SUPREME COURT OF IOWA
Submitted October 20, 2021—Filed February 4, 2022
Appeal from the Iowa District Court for Lee (North) County, Mary Ann Brown, Judge.
The defendant appeals the district court’s denial on remand of his motion challenging the representativeness of the jury pool under the fair-cross-section requirements of the United States and Iowa Constitutions. AFFIRMED.
McDermott, J., delivered the opinion of the court, in which all justices joined. Mansfield, J., filed a concurrence, in which Appel, J., joined. McDonald, J., filed a concurrence, in which Christensen, C.J., and Waterman, J., joined.
Martha J. Lucey, State Appellate Defender, and Shellie L. Knipfer (argued), Assistant Appellate Defender, for appellant.
Thomas J. Miller, Attorney General, and Louis S. Sloven (argued) and Andrew Prosser, Assistant Attorneys General, for appellee.
David S. Walker, Windsor Heights, and Russell E. Lovell, II (argued), Des Moines, for amicus curiae NAACP.
A jury in North Lee County found Kenneth Lilly guilty of aiding and abetting a bank robbery. Lilly, an African-American, appealed his conviction, arguing that his right to an impartial jury under both the United States and Iowa Constitutions had been violated because neither his jury nor even the jury panel contained any African-Americans. On appeal, we remanded the case to give Lilly an opportunity to develop his impartial-jury arguments in response to refinements to how a defendant must prove a constitutional violation that we explained in several cases after his trial. The district court ultimately rejected Lilly’s further-developed claims. Lilly now appeals that ruling, arguing that the district court erred in holding that he failed to prove a violation.
I. The Issue on Remand.
We described the underlying facts from Lilly’s trial and earlier procedural history of this case in the opinion filed in Lilly’s initial appeal and will forego restating them here. See State v. Lilly (Lilly I), 930 N.W.2d 293, 296–98 (Iowa 2019). Pertinent to this appeal are the facts that the parties developed on remand related to the only remaining issue in the case: Lilly’s fair-cross-section claim.
In State v. Plain (Plain II), we defined the terms “jury pool” (the members of the community selected for jury duty and summoned and reporting to the courthouse),
Evidence offered at Lilly’s hearing on remand showed that of the people summoned who indicated their race on a summoned-juror questionnaire, one person marked “Other,” one marked “American,” one marked “Asian,” one marked “Japanese,” and one marked “White/Black.” Ultimately, none of the potential jurors in Lilly’s pool were African-American, and (thus) none of the members of Lilly’s jury were African-American. Lilly called only one witness at the hearing, the jury manager for Lee County, who testified about how the jury selection process worked in the county at the time of Lilly’s trial.
II. The Duren/Plain Elements.
The
A defendant establishes a prima facie violation of the fair-cross-section requirement by showing that (1) a group alleged to have been excluded is a “distinctive” group in the community, (2) the group’s representation in jury pools is not “fair and reasonable” when considered against the group’s percentage in the community, and (3) the group’s underrepresentation “is due to systematic exclusion of the group in the jury-selection process.” Id. at 822 (quoting Duren v. Missouri, 439 U.S. 357, 364 (1979)). The defendant bears the burden of proof
to show a prima facie violation of the fair-cross-section requirement. Plain I, 898 N.W.2d at 821–22; Lilly I, 930 N.W.2d at 299; see also Duren, 439 U.S. at 363–64.
The State concedes the first Duren/Plain prong and thus that African Americans constitute a distinctive group in the community. The dispute centers on the second and third prongs. The district court held that Lilly failed to prove either one. We review challenges alleging the denial of constitutional rights—in this case, the right to an impartial jury—de novo and thus evaluate the evidence anew without deferring to the district court’s findings. Lilly I, 930 N.W.2d at 298.
III. Lilly’s Proof of Causation Under Duren/Plain’s Third Prong.
We will begin our analysis on the third prong, since an inability to establish any one of the three Duren/Plain elements is fatal to a defendant’s fair-cross-section challenge. In Lilly I, we explained that to establish the third prong a defendant must prove that the underrepresentation resulted from a particular feature (or features) of the jury selection system. Id. at 306. The defendant, in other words, “must tie the disparity to a particular practice” and show that the practice caused the systematic exclusion of the distinctive group in the jury selection process. Id. at 307.
Lilly points to a single jury management practice to prove his claim of African-American underrepresentation in jury pools. He targets the lists—voter registration, driver’s license, and nonoperator identification—that are combined to form
that low-income people tend to register to vote and to acquire driver’s licenses and nonoperator identification cards at a lower rate than other members of the community. He next asserts that African-Americans make up a higher percentage of low-income people in Lee County. Taking these premises together, Lilly infers that African-Americans register to vote and get driver’s licenses and nonoperator identification cards at lower rates than other races. African Americans are, following this logic, underrepresented in the lists from which jury pools are sourced. From this conclusion, Lilly argues that failing to supplement the source list with other lists that might include more lower-income people amounts to “mismanagement” resulting in the systemic exclusion of African Americans.
A. Analysis Under the Sixth Amendment.
Lilly presents his arguments both under the
that might include, for instance, practices for updating juror address lists, excusing potential jurors for hardship or other reasons, and enforcing jury summonses. 930 N.W.2d at 308. These common jury practices fall within a state’s “broad discretion,” according to the Supreme Court in Berghuis, and will not sustain a Sixth Amendment cross-section challenge. 559 U.S. at 333 (quoting Taylor v. Louisiana, 419 U.S. 522, 537–38 (1975)).
The practice of using the state’s own voter registration list, motor vehicle operator list, and nonoperator identification list to construct a source list from which to draw jury pools amounts to a commonplace, run-of-the-mill practice. Courts in jurisdictions around the country have upheld the use of voter registrations lists (without the addition of another list) as a jury-pool source. See United States v. Orange, 447 F.3d 792, 800 (10th Cir. 2006) (“The circuits are ‘in complete agreement that neither the Act nor the Constitution require that a supplemental source of names be added to voter lists simply because an identifiable group votes in a proportion lower than the rest of the population.’ ” (quoting United States v. Test, 550 F.2d 577, 586 n.8 (10th Cir. 1976))); United States v. Sanchez, 156 F.3d 875, 879 (8th Cir. 1998) (“We have consistently upheld the use of voter registration lists to select jury pools.”). Lilly acknowledges that “[i]t has been the practice for district courts to solely use the lists provided”—the very lists he challenges in this case—to create jury pools. These lists are the only ones that the
than” the run-of-the-mill
B. Analysis Under the Iowa Constitution.
Lilly’s claim under
But Lilly still must prove that the challenged practice caused the systematic exclusion of the group in the jury-selection process. Id. at 307–08. And here Lilly’s claim falters in multiple ways. As an initial matter, Lilly attempts to apply deductive reasoning showing causation by constructing a syllogism that, in simple form, goes as follows:
Low-income residents are underrepresented on the source list;
African-Americans are more likely to be low-income residents;
Therefore, African-Americans are underrepresented on the source list.
But Lilly doesn’t prove the key premise. He asserts that low-income people register to vote and acquire driver’s licenses and nonoperator identification cards at lower rates than others in the community, but he offers no evidence to establish this fact. The proposition is supported by supposition, not proof. “Mere
speculation about the possible causes of underrepresentation will not substitute for a credible showing of evidence supporting those allegations.” Lilly I, 930 N.W.2d at 307 (quoting Paula Hannaford-Agor, Systematic Negligence in Jury Operations, 59 Drake L. Rev. 761, 790 (2011) [hereinafter Hannaford-Agor]).
Lilly presented no testimony about the demographic make-up of the people included on the source list. He offered no evidence that low-income people actually register to vote or get driver’s licenses or nonoperator identification cards at lower rates than others. And he offered no evidence that African Americans in particular register to vote or get driver’s licenses or identification cards at lower rates than others. The amicus curiae, the NAACP, attempts to fill this void in part by attaching an email from a former Iowa Department of Transportation director as an exhibit to its appeal brief, but that email was never offered or entered in the district court and is beyond the record for our consideration on appeal.
Even if Lilly had introduced evidence on that particular point, proof of actual causation in the record nonetheless remains absent. In Lilly I, we said that a defendant’s proof of causation will “almost always require expert testimony” to (1) identify “the precise point of the juror summoning and qualification process in which members of distinctive groups were excluded from the jury pool” and (2) offer “a plausible explanation of how the operation of the jury system resulted in their exclusion.” Lilly I, 930 N.W.2d at 307 (quoting Hannaford-Agor, 59 Drake L. Rev. at 790–91). Lilly called no expert witness whatsoever, let alone one that pinpointed the procedural step in which African-
Americans were excluded and offered a plausible explanation for how the summoning and qualification process brought it about. He has failed to show (even assuming his supporting premises had been
As a fallback, Lilly argues that requiring the defendant to establish causation under the third prong of the Duren/Plain analysis assigns the burden to the wrong party. He urges instead that once the defendant shows that a distinctive group has been underrepresented in jury pools under the second prong, the burden should shift to the State to show affirmatively that the group has not been systematically excluded. But in Lilly I we analyzed the amicus’s argument on burden shifting and explained at some length why the burden to prove causation appropriately resides with the defendant. Id. at 305–06. In Lilly’s initial appeal, we expressly assigned the burden of proof on these issues to Lilly, stating that “the defendant must prove that the practice has caused systematic underrepresentation.” Id. at 307–08 (emphasis added). He has not met his burden.
Because Lilly failed to deliver on his burden under the third prong, which on its own is sufficient to affirm the district court’s denial of his claim, we need not take up his arguments relating to the second prong’s requirement to establish actual underrepresentation of African-Americans in his jury pool.
IV. Conclusion.
In Lilly I, we conditionally affirmed Lilly’s conviction and remanded for a determination on his fair-cross-section challenge. We now affirm the district court’s holding on remand that Lilly failed to prove a violation of his right to an impartial jury, and we affirm his conviction.
AFFIRMED.
All justices concur. Mansfield, J., files a concurrence, in which Appel, J., joins. McDonald, J., files a concurrence, in which Christensen, C.J., and Waterman, J., join.
MANSFIELD, Justice (concurring).
I join the majority opinion. I write briefly to respond to the suggestion of those who partially dissented in State v. Lilly (Lilly I), 930 N.W.2d 293 (Iowa 2019), that after one remand, it is time to overrule Lilly I and abandon the effort.1
The
First, instead of examining numbers that have no statistical validity whatsoever, like the absolute disparity between the percentage of African-Americans in the jury pool and the percentage in the overall population, we focused the analysis on the only measure that has statistical validity: whether there is an underrepresentation of African-Americans in the jury pool that is unlikely to be due to chance. Lilly I, 930 N.W.2d at 302. We also explained that to make the analysis more sound, the baseline should “reflect the population that
Second, we required the defendant to prove, again using proper methods of proof, that a specific practice was causing the systematic underrepresentation of a distinctive group. Id. at 308. This was a middle position between those who argued that the State should have the burden of justifying its jury selection practices whenever the defendant proved underrepresentation and those who argued current practices should be held per se constitutional. See id. at 307. It’s a logical line to draw. If a practice leads to systematic exclusion of African-Americans from jury pools, who cares what that practice is? The important thing is that it is exclusionary. As we put it in Lilly I, “If a practice that leads to systematic underrepresentation of a distinctive group in jury pools can be identified and corrected, there is no reason to shield that practice from scrutiny just because it is relatively commonplace.” Id. at 307–08.
Our recent caselaw has led to several helpful developments. Attorneys on both sides are assembling more accurate data about the jury-eligible populations in specific counties. Those same attorneys are using statistical analyses instead of the old-fashioned, discredited, correlation-means-causation kinds of arguments. (Doing the statistics is not that difficult.) No longer do we get to rely only on our own assumptions. Instead, our caselaw has put the spotlight on specific practices and enabled us to learn whether, in fact, they lead to systematic underrepresentation of distinctive groups. None of these things would have happened without Lilly I and related cases, at least not to the same degree.
Here, the defendant has been unable to establish a constitutional violation. That does not mean we were wrong in Lilly I.
In today’s concurrence, the Lilly I partial dissenters claim the mantle of precedent. They assert that we decided categorically forty-six years ago that using voter registration lists to summon jurors did not violate the
Absent a showing of systematic discrimination, other courts have consistently turned back constitutional assaults on the use of voter registration lists as the sole source of names for jury duty.
Defendant had the burden of establishing a prima facie case that the Black Hawk County selection procedure systematically excluded a particular group. Trial court properly ruled the record made here failed to meet that test.
Id. (citations omitted) (emphases added). In other words, we held on that record that the use of voter registration lists didn’t violate the Iowa Constitution. The Lilly I majority is consistent with that holding; the Lilly I partial dissent is not.
In the present appeal, the NAACP, as amicus, asks us again to eliminate the defendant’s burden of proof on systematic exclusion, shifting the burden to the State to prove why it cannot “mak[e] the necessary reforms” to secure a jury pool that represents a fair cross-section: once the defendant proves underrepresentation. The NAACP raised a similar argument as amicus in Lilly I, and a majority of this court rejected it. Lilly I, 930 N.W.2d at 305–07 (“[A]t this time, we are not prepared to embrace the NAACP’s proposal.”). Thus, the NAACP wants us to depart from the law of the case. That is something we don’t ordinarily do, and I would not favor doing it today.
However, we also said in Lilly I, “We may be willing to impose such an obligation
For the foregoing reasons, I concur in the majority opinion and also submit this concurrence.
Appel, J., joins this concurring opinion.
McDONALD, Justice (concurring).
I agree with the majority that Lilly failed to provide any evidence in support of his fair-cross-section claims arising under the federal and state constitutions and that Lilly is not entitled to any relief, and I join the court’s opinion. The law concerning fair cross-sections applied in the majority opinion is the law of the case with respect to Lilly’s claims, but I would revisit this area of law in the appropriate case. The court’s recent fair-cross-section jurisprudence under the
The practices being challenged in this court’s most recent fair-cross-section cases are standard practices used for decades all over the country. Forty-six years ago, this court rejected the same challenge raised in this case. See State v. Williams, 243 N.W.2d 658, 662 (Iowa 1976). Rather than continuing to adjudicate these same issues for the next half-century, I would hold, in the appropriate case, that run-of-the-mill jury management practices cannot establish a claim of systematic exclusion under the Iowa Constitution. See State v. Veal, 930 N.W.2d 319, 364–65 (Iowa 2019) (McDonald, J., concurring in part and dissenting in part) (discussing federal standard); State v. Lilly (Lilly I), 930 N.W.2d 293, 318 (Iowa 2019) (McDonald, J., concurring in part and dissenting
in part) (collecting cases); State v. Williams, 929 N.W.2d 621, 645 (Iowa 2019) (McDonald, J., concurring in part and dissenting in part).
Christensen, C.J., and Waterman, J., join this concurring opinion.