Antonio Garcia-Dorantes v. Millicent Warren
801 F.3d 584
6th Cir.2015Background
- Antonio Garcia-Dorantes was convicted (Sept. 10, 2001) of second-degree murder and assault; trial centered on intent/self‑defense where eyewitness memory was imprecise and intoxication was significant.
- After conviction, a July 2002 press report revealed a Kent County jury‑selection computer glitch (Apr. 2001–early 2002) that disproportionately excluded minorities by sampling only the first 118,169 state records sorted by zip code.
- Garcia‑Dorantes raised a Sixth Amendment fair‑cross‑section claim on direct appeal; state courts rejected it as procedurally forfeited for failing to object at trial; Michigan Supreme Court denied leave.
- In federal habeas proceedings, the district court found Garcia‑Dorantes established cause and actual prejudice to excuse the procedural default and that he made a prima facie Duren violation; it granted conditional habeas relief.
- The Sixth Circuit (lead opinion) affirmed: applied the Ambrose/Strickland standard for actual prejudice, rejected generalized expert testimony about diverse juries, and held the absolute and comparative disparities (3.45% abs.; 42% comp. for African‑Americans) satisfied Duren’s second prong given the small minority population and the proven systematic exclusion.
- Dissent argued (1) Supreme Court guidance in Berghuis v. Smith counsels skepticism of abstract disparity metrics in a county with a small minority pool, (2) record uncertainty about the petit jury’s actual composition, and (3) clerks may have manually supplemented panels.
Issues
| Issue | Plaintiff's Argument | Defendant's Argument | Held |
|---|---|---|---|
| Whether petitioner showed cause and actual prejudice to excuse procedural default of fair‑cross‑section claim | Computer glitch was only revealed post‑trial; cause exists; actual prejudice shown because reasonable probability a properly selected jury would have reached a different result given close issues of intent | No actual prejudice: trial evidence strong; generalized testimony about diverse juries irrelevant; statistical chance petit jury would differ is low | Affirmed: cause shown; actual prejudice met under Strickland/Ambrose standard — reasonable probability that a properly selected jury could have produced a different outcome |
| Whether the jury venire violated the Sixth Amendment fair‑cross‑section right under Duren | The glitch produced systematic exclusion and produced significant underrepresentation (absolute and comparative disparities); small minority population makes comparative disparity especially probative | Disparities are not large enough; reliance on absolute/comparative measures is misleading in counties with small minority percentages; Berghuis v. Smith cautions against finding constitutional violation here | Affirmed: petitioner established Duren’s three prongs — distinctive group, systematic exclusion (proved), and unfair/unreasonable representation (3.45% abs.; 42% comp. sufficient given context) |
| Relevance of general expert testimony that diverse juries convict less often | Testimony supports that a properly selected (more diverse) jury would be less likely to convict | Testimony is general, not case‑specific, risks stereotyping, and does not meet Strickland’s individualized prejudice requirement | Court rejected general expert as insufficient; prejudice must be shown with case‑specific analysis |
| Whether government has a significant state interest justifying the selection process | N/A (petitioner showed prima facie violation and the glitch was inadvertent) | County may argue administrative reasons or non‑benign causes | No significant state interest shown; the glitch was inadvertent/cost‑saving transfer error, so burden not met by state |
Key Cases Cited
- Ambrose v. Booker, 684 F.3d 638 (6th Cir. 2012) (explains actual‑prejudice standard for excusing procedural default in fair‑cross‑section claims and directs Strickland‑style inquiry)
- Strickland v. Washington, 466 U.S. 668 (1984) (establishes "reasonable probability" prejudice standard to undermine confidence in outcome)
- Duren v. Missouri, 439 U.S. 357 (1979) (sets three‑prong test for Sixth Amendment fair‑cross‑section claims)
- Berghuis v. Smith, 559 U.S. 314 (2010) (addresses statistical measures of underrepresentation and warns against rigid reliance on absolute/comparative disparity in jurisdictions with small minority pools)
- Batson v. Kentucky, 476 U.S. 79 (1986) (prohibits race‑based peremptory strikes; warns against stereotyping jurors by race)
- Hollis v. Davis, 941 F.2d 1471 (11th Cir. 1991) (applies Strickland‑type actual‑prejudice analysis to jury‑selection exclusion claims)
- Taylor v. Louisiana, 419 U.S. 522 (1975) (a defendant is not entitled to a jury of particular composition; fair process is required)
