801 F.3d 567
6th Cir.2015Background
- Joseph Ambrose was convicted in Kent County, Michigan (April 2001) of armed robbery, carjacking, and felony-firearm; he did not object at trial to the racial composition of the venire.
- In July 2002 press reports and county studies revealed a jury-selection computer glitch (April 2001–early 2002) that underrepresented African-Americans in Kent County venires.
- Ambrose sought state post-conviction relief asserting a Sixth Amendment fair-cross-section claim; state courts denied relief as procedurally defaulted because he failed to object at trial.
- Ambrose filed a § 2254 habeas petition; the district court granted conditional relief, finding cause (the glitch was not reasonably discoverable), actual prejudice, and a prima facie fair-cross-section violation.
- On appeal this court (Ambrose v. Booker) remanded to determine whether Ambrose showed actual prejudice sufficient to excuse procedural default, instructing courts to examine the strength of the case as central to that inquiry.
- On remand the district court relied on expert testimony (Dr. Sommers) that racially diverse juries are less likely to convict and again granted conditional relief; the Sixth Circuit reversed, holding Ambrose failed to show actual prejudice under the proper standard.
Issues
| Issue | Ambrose's Argument | Respondent's Argument | Held |
|---|---|---|---|
| Whether Ambrose may excuse his procedural default (cause and actual prejudice) based on the jury-selection glitch | Cause: glitch was not discoverable at trial; Prejudice: a properly selected (more diverse) jury would have been less likely to convict (expert social-science evidence) | Cause conceded on remand; argues Ambrose failed to show the required actual prejudice to excuse default | Cause established earlier, but Ambrose failed to show actual prejudice; procedural default not excused |
| Proper standard for "actual prejudice" to excuse procedural default | Apply a less-demanding "less likely to convict" standard (Hollis) | Apply Strickland’s "reasonable probability the result would have been different" standard | Strickland standard applies: reasonable probability the result would have been different |
| Admissibility / weight of general expert testimony that diverse juries convict less often | Expert testimony (Dr. Sommers) supports that more diverse juries are less likely to convict and thus a different outcome was reasonably probable | Such testimony is irrelevant absent an individualized assessment of this case, risks stereotyping, and cannot substitute for analysis of the trial transcript and strength of evidence | Court rejected reliance on Dr. Sommers for actual-prejudice showing; general racial-stereotype evidence insufficient |
| Whether the record shows a prima facie fair-cross-section violation and if it would have altered the verdict given the trial evidence | The venire was systematically underrepresenting African-Americans due to the glitch; a properly assembled venire would likely have produced a different outcome given weaknesses in the prosecution’s case | Even accepting underrepresentation, the evidence against Ambrose was strong (eyewitness IDs, prompt statements); no reasonable probability of a different result | Court did not reach merits of fair-cross-section claim because procedural default was not excused; on prejudice review the court found the prosecution’s evidence strong and no reasonable probability of a different outcome |
Key Cases Cited
- Ambrose v. Booker, 684 F.3d 638 (6th Cir. 2012) (remanding to assess actual prejudice to excuse procedural default)
- Hollis v. Davis, 941 F.2d 1471 (11th Cir. 1991) (discussing "less likely to convict" formulation in jury-selection context)
- Strickland v. Washington, 466 U.S. 668 (1984) (prejudice standard: reasonable probability result would have been different)
- Batson v. Kentucky, 476 U.S. 79 (1986) (racially based juror exclusions forbidden; cautions against stereotyping)
- Berghuis v. Smith, 559 U.S. 314 (2010) (Supreme Court discussion of statistical measures and jury composition concerns)
- Huffman v. Wainwright, 651 F.2d 347 (5th Cir. 1981) (example recognizing special prejudice where jury composition and facts make mixed jury likely to change outcome)
- Garcia-Dorantes v. Warren, 801 F.3d 584 (6th Cir. 2015) (discussing prejudice formulations and interpretation of Ambrose)