Hedgpeth v. Pulido (07-544) (formerly Chrones v. Pulido)

Oral argument: Oct. 15, 2008

Appealed from: United States Court of Appeals, Ninth Circuit (May 30, 2007)


Michael Pulido was sixteen in 1996 at the time of his conviction for robbery-murder under the California Penal Code. He was convicted by a jury after it received instruction on two valid theories and one constitutionally impermissible theory of guilt, an aiding-and-abetting theory that lacked proper explanation of the “contemporaneity” requirement of robbery-murder under California law. In 2005, Pulido obtained a writ of habeas corpus from the District Court for the Northern District of California on the grounds of alternative-legal-theory error. The Ninth Circuit upheld the grant of the writ, relying primarily on Stromberg v. California, 238 U.S. 359 (1931), and its own precedent to find that alternative-legal-theory error was “structural” in nature and thus required automatic reversal of the conviction. The attorney general of California sought review before the Supreme Court, maintaining that more recent Supreme Court precedent requires review for “harmlessness” of the constitutional error, rather than automatic reversal. This case will clarify the Antiterrorism and Effective Death Penalty Act of 1996’s provision requiring federal courts to limit the granting of writs of habeas corpus to those state decisions that are “contrary to” or “unreasonable” in the light of “currently established federal law” as determined by the Supreme Court.

· [Question(s) presented]

· [Issue]

· [Facts]

· [Discussion]

· [Analysis]

Question presented

Stromberg v. California, 283 U.S. 359 (1931), required the reversal of the judgment if a general verdict could have rested on an instruction that defined a constitutionally defective alternative theory of criminal liability. However, a modern line of cases, including Neder v. United States, 527 U.S. 1 (1999), establishes that error in instructing on an element of a charged crime is not “structural error,” so as to require automatic reversal, but is instead “trial error” and, as such, may be harmless.

The question presented is:

Did the Ninth Circuit fail to conform to “clearly established” Supreme Court law, as required by 28 U.S.C. § 2254(d), when it granted habeas corpus relief by deeming an erroneous instruction on one of two alternative theories of guilt to be “structural error” requiring reversal because the jury might have relied on it?



Whether erroneous jury instructions given during a murder trial can be deemed “harmless” error and the murder conviction upheld, or whether such erroneous instructions require a grant of habeas corpus in a federal court.



Early in the morning on May 22, 1992, a Shell gas station attendant in San Mateo, California was shot dead and the cash register removed from his store. Chrones v. Pulido, 487 F.3d 669, 672 (2007). Michael Pulido testified that, from his uncle’s car outside the station, he watched his uncle go into the station convenience store, unaware that his uncle was in possession of Pulido’s .45 caliber weapon. Id. Pulido further testified that he entered the store only after hearing the gunshot, at which point he saw the attendant bleeding on the floor, and that he then ran out and got into the passenger side of his uncle’s car again without having touched anything. Id. When his uncle reemerged with the cash register, he threw it onto Pulido’s lap and forced him at gunpoint to open the register, remove the money, and dispose of the register in nearby bushes sometime later. Id. Pulido’s fingerprints were later found on a can of Coke inside the store, and on the register. Id.

At trial in California’s state court system, the jury was instructed on one of three possible theories of Pulido’s guilt for murder: (1) that Pulido shot the attendant himself, (2) that Pulido aided and abetted his uncle in the shooting, or (3) that Pulido joined the robbery only after the murder. Brief in Opp. at 1; Brief for Respondent at 3, 19. He was convicted of the murder and of an aggravating “special circumstance,” which is California’s version of the felony-murder rule. Cal. Pen. Code § 190.2(a)(17)(A); Chrones, 487 F.3d at 672. In California, the “special circumstance” requires the murder to be committed during the time the defendant is participating in the robbery (called a “contemporaneity requirement”), such that the timing of Pulido’s involvement in the robbery (i.e. before or after the shooting) was a key point. See Cal. Pen. Code § 190.2(a)(17)(A).

A typo in the jury instructions allowed the jury to find the “special circumstance” satisfied without the otherwise required finding of contemporaneity. See Chrones, 487 F.3d at 672. The jury returned a unanimous guilty verdict on Pulido’s murder conviction and the “special circumstance” conviction. See id. However, the jury was deadlocked on whether Pulido had physically manipulated the handgun during the crime, indicating that some jurors had found him guilty of murder based on the aiding and abetting theory that had been erroneously explained to them in the instructions. See id.; see also Brief for Respondent at 17. This theory should have, but did not according to the actual instructions, required the jury to find that the robbery and intent to aid and abet his uncle were contemporaneous. See id.; see also Brief in Opp. at 1. Both parties admitted on appeal that the erroneous instructions constitute constitutional error. See Brief for Respondent at 20, Brief for Petitioner at 11.

After the California Supreme Court upheld his conviction in May 1997, Pulido sought habeas corpus review in federal court on the theory that the jury instructions contained prejudicial error (an instruction that is inaccurate in a way that could cause prejudice to the defendant). See Chrones, 487 F.3d at 672. In March 2005, the United States Circuit Court for the Northern District of California granted review as authorized by 28 U.S.C. § 2254(d), on the grounds that the state court proceedings produced a decision that was “contrary to,” or involved an “unreasonable application” of “clearly established federal law, as determined by the Supreme Court of the United States.” Pub. L. No. 104-132, Title I, § 104, 110 Stat. 1218 (1996); 2008 WL 6142229 (N.D. Cal. 2005). The United States Court of Appeals, Ninth Circuit affirmed the grant of habeas corpus, and the attorney general of California filed a petition for certiorari. Chrones, 487 F.3d 669; see also Petition for Certiorari. The Supreme Court granted certiorari on February 15, 2008. 128 S.Ct. 2004.



While Hedgpeth v. Pulido (formerly Chrones v. Pulido) turns on a technical issue, the Supreme Court’s decision on the availability of “harmlessness” review for jury instruction error will have a number of practical consequences. First, the decision will contribute to the resolution of an inherent tension in the Antiterrorism and Effective Death Penalty Act of 1996 (“AEDPA”): how to limit the number of habeas corpus petitions, yet do so in a constitutionally acceptable manner. See Tale of Two Laws Revisited: Investigating the Impact of the Prisoner Litigation Reform Act and the Antiterrorism and Effective Death Penalty (“Tale of Two Laws”) at 46. Further, the decision will affect the role of juries in the court system, consequently limiting or expanding the recourse available to convicted criminals by regulating judicial oversight of jury verdicts where erroneous instructions were received. Finally, it will affect how criminal attorneys and defendants approach criminal trials, weighing the different benefits to plea bargains versus jury trials.

This decision will affect the flexibility of federal courts to grant habeas petitions under the AEDPA while significantly reducing current, documented discrepancies in application of the AEDPA across the federal circuits. See Tale of Two Laws at 61. A finding that the writ was improvidently granted supports the “finality” camp of legal theorists who want to limit grants out of respect for both federal and state systems. See Tale of Two Laws at 47. First, this camp recognizes that the validity of a state’s criminal justice process during habeas proceedings is suspended. See id. And, from a federal viewpoint, focusing efforts on the creation of an effective system to catch errors is a better use of scarce federal resources than trying to catch each error on review. See id. Conversely, a finding in agreement with the Ninth Circuit supports the principle that every petitioner should have the right of access to federal review, regardless of the administrative impact on the federal court system. See id. at 46. Supporters of this perspective argue that the state’s judicial processes may be deficient or difficult to access. See id. at 46–47. A decision either way will certainly contribute to the numerous factors that contribute to whether the AEDPA meets its goals of limiting petitions (one study has thus far found the AEDPA to actually result in increased habeas filings by state inmates), although the overall effect of the AEDPA’s supposed restrictions on numbers of filings is unclear. See id. at 46–49; see also John Scalia, “Prisoner Petitions Filed in U.S. District Courts, 2000, with Trends, 1980-2000, Bureau of Justice Statistics (2002) at 4.

A decision for California would show confidence in the American jury system, while a decision for Pulido would validate a judicial oversight function. Although almost a quarter of the adult population has served on a jury, up from 16% in 1977, by and large the public believes juries to be non-representative of their communities. See How the Public Views the State Courts: A 1999 National Survey, National Center for State Courts (“National Survey”), at 8, 30. A decision for California would emphasize that despite this perceived mismatch between communities and jury makeup, a bedrock of the jury system remains that a jury’s determination of the facts is final and not reviewed by courts. See Brief for Respondent at 17, 40. On the other hand, a decision for Pulido would validate the view that a court, when unsure what facts the jury relied on, must reverse, which could consequently lead to more reversals. See Brief for Respondent at 39; Brief for Petitioner at 14. A result for Pulido would be consistent with a study finding that Americans’ lack of faith in juries may be merited, as research shows juries do not generally understand instructions, even clear ones. See Judicial Notebook, Monitor on Psychology, Vol. 39, No. 6 (June 2008). Moreover, juries in felony-murder cases try to mete out justice by degrees in considering the degree of guilt of the accomplice. See id. Either way, courts’ protection of constitutional rights enjoy may well continue to enjoy an excellent reputation in the community: currently, nearly 85% of respondents across the categories either “somewhat” or “strongly” agreeing that courts do a good job protecting these. See National Survey at 34.

The decision may also have implications for those deciding whether to pursue a jury trial or take a plea bargain. “The rate of civil and criminal jury trials has steadily declined in recent years, eclipsed by non-trial dispositions such as settlement, plea agreements, and summary judgment.” The State of the States Survey of Jury Improvement Efforts: A Compendium Report at 4. A decision for California would tacitly acknowledge the fact that when a jury does participate, it is more likely to make character judgments and believe or disbelieve the evidence as a whole than to rely on a specific instruction. See id. at 4. If the Supreme Court agrees, as the Criminal Justice Legal Foundation maintains, that the typo in the instructions and the jury’s numerous questions do not constitute “grave doubt,” the bar will be raised for what does constitute doubt under the Brecht standard. See Brief for Amicus Curiae Criminal Justice Legal Foundation in Support of Petitioner at 21–23. A decision for Pulido, however, would mean that the contested evidence on the permissible grounds for conviction indicates jury confusion, and requires reversal, lowering the “grave doubt” bar. See Brief for Respondent at 37. If criminal defendants believe the court will allow juries to receive erroneous instructions, the possibility that this will negatively impact their trial could be additional impetus to accepting a plea bargain.



Congressional Intervention:  The Antiterrorism and Effective Death Penalty Act of 1996

The Antiterrorism and Effective Death Penalty Act of 1996, Pub. L. No. 104-132, 110 Stat. 1214 (the “AEDPA”) greatly limits federal courts’ power to grant writs of habeas corpus. See Final Technical Report: Habeas Litigation in U.S. District Courts, Nancy J. King et al., at 7–9. The provision at issue in this case, 28 U.S.C. § 1254(1), is one such restriction. See id. at 9. To grant relief, a federal court must find that the state courts issued a decision that was either “(1) contrary to, or involved an unreasonable application of clearly established federal law as determined by the Supreme Court of the United States, or (2) based on an unreasonable determination of the facts in light of the evidence presented in the state court proceeding.” 28 U.S.C. § 1254(1).

The parties disagree on whether the Supreme Court of California’s decision meets this standard, primarily because they disagree on what constitutes the Supreme Court’s “established federal law.” See Brief for Petitioner at 14–15; Brief for Respondent at 21. The Ninth Circuit, and Pulido in his response, find the origins of today’s habeas jurisprudence in 1931’s seminal Stromberg v. California, which held that alternative-legal-theory error requires reversal. Brief for Respondent at 21; 238 U.S. 359 (1931).Alternative-legal-theory error occurs when, as in this case, the judge instructs the jury on both a constitutionally correct, and a constitutionally impermissible, theory of guilt. In Stromberg, as the Court could not determine that the jury did not rely on the impermissible legal theory, the verdict was reversed. See 238 U.S. at 368.

On the other hand, California maintains that the Court repudiated this simplistic stance when it held, beginning with Rose v. Clark and extending through Neder v. United States, that not all constitutional instructional errors (errors that offends the guarantees offered by the Constitution, such as alternative-legal-theory error), require reversal. See Brief for Petitioner at 15, 17; 478 U.S. 570 (1986); 527 U.S. 1 (1999). This theory of “harmless error” or “trial error” was first established in Chapman v. California. See Brief for Petitioner at 17; 386 U.S. 18 (1967). Consequently, California argues that because Supreme Court constitutional error theory has broadened to include harmless error analysis, it now encompasses alternative-legal-theory error analysis as well and thus overrules a cut-and-dry approach reversals. See Brief for Petitioner at 16 n.3.

Is Structural Error an Outdated Concept?

Much of the debate in this case stems from the Ninth Circuit’s use of the term “structural error” to describe the failure to correctly instruct the jury on the contemporaneity requirement of felony-murder under the California Penal Code. See Cal. Pen. Code § 190.2(a)(17)(A). Despite the Ninth Circuit’s wording, none of the four briefs submitted in this case advocates the proposition that alternative-legal-theory error should be integrated into the “very limited class of cases” that constitute structural error. Johnson v. United States, 520 U.S. 461, 468 (1997). On the contrary, Brecht v. Abrahamson confirmed that trial error does not require automatic reversal, is quantifiable, and requires the determination of whether the error “had substantial and injurious effect or influence in determining the jury's verdict.” 507 U.S. 619, 623 (1993); Kotteakos v. United States, 328 U. S. 750, 776 (1946).

Pulido maintains that structural error has its place, but not in this case. See Brief for Respondent at 18, 51. He reminds the Court that structural errors, which go deep into the constitution of a trial, “vitiate[] all the jury’s findings,” such that the scope and effect of the error cannot be quantitatively assessed. Sullivan v. Louisiana, 508 U.S. 275, 281 (1993) (emphasis in original). While supporting the continued relevance of legal errors that “vitiate” all findings, Pulido repudiates the label “structural error” as used by the Ninth Circuit. Brief for Respondent at 51. Pulido argues that it is only when legal theories presented are inaccurate, like in this case, that the court steps in to deem the error “structural” and reverses the conviction. See id. at 28–29. Pulido contends that despite a finding of fact, juries are maladapted to this task. See id.

According to California, however, Stromberg has no continued relevance because “clearly established federal law” on structural error is limited to particular errors previously identified by the Supreme Court. See Brief for Petitioner at 12. California argues that a further structural error category would only be appropriate where the error is so fundamental as to deprive a reviewing court of the possibility of knowing if a jury would have arrived at the same conclusion nonetheless, and the error in this case is insufficient to meet that standard. Id. at 12, 17 (citing Neder, 527 U.S. at 8). California contends that the factors leading to the admitted constitutional error constitute nothing more than trial error, and that the Supreme Court should recognize at least that it was not unreasonable of the Supreme Court of California to treat it as such. Id. at 18–19. To constitute structural error, California argues, an error must be more serious than mere trial error that proves to be prejudicial. Id. at 24–25.

While the Court’s amici find other ways of disagreeing with the Ninth Circuit’s certainty that alternative-legal-theory error is structural, amicus curiae United States takes an intermediary position. See Brief for Amicus Curiae United States in Support of Petitioner (“Brief for U.S.”) at 22–23. The United States argues that it believes that the verdict’s possible dependence on an impermissible legal theory falls squarely within Stromberg, but that the Court should reconcile all errors under a harmlessness review as no logical distinction persists between types of errors. See id.

Brecht and the Outcome under Harmlessness Review

Both parties agree that alternative-legal-theory error is not always reversible error and that the applicable standard is Brecht, but each party also believes the consequent analysis results in a clear victory for its side. Brief for Respondent at 18, 23, 26; Brief for Petitioner at 26–27. Brecht requires a court to ask whether the instructional error “had a substantial and injurious effect or influence in determining the jury's verdict.” Brecht, 507 U.S. at 631. Pulido views Stromberg as merely a “categorical application” of the Brecht error review, while California views the Stromberg and Rose/Neder lines of cases as irreconcilable. See Brief for Petitioner at 15–16, Brief for Respondent at 34.

California maintains that the erroneous instruction, had it been analyzed under the correct harmlessness analysis established in Brecht, would have been revealed to be “simply superfluous.” Brief for Petitioner at 19. According to California, the instruction did not preclude the jury from finding all the facts for a permissible conviction. Id. at 20. Even without a finding of contemporaneity, California argues, the error is trial error because it did not prevent the jury from finding all necessary facts beyond a reasonable doubt. Id. at 21–22. In addition, California maintains, the error falls short of vitiating all the jury’s factual findings, as similar cases with such incomplete verdicts have shown. Id. at 22. The United States, in its amicus curiae brief, adds that Brecht’s “grave doubt” standard must rest on an objectively sufficient evidentiary basis for disputing the facts. Brief for the U.S. at 29–30. This, the United States argues, would in turn allow a reviewing court to determine whether the jury could have found otherwise than it did were it not for the erroneous instruction, and consequently whether reversal is required. See id.

Pulido, on the other hand, casts Stromberg as a “rigorous form” of the same type of review that goes on in Brecht, and minimizes its connection to the term, “structural error.” See Brief for Respondent at 14. This view reconciles the district court’s “harmlessness” and the Ninth Circuit’s “structural error” theories under an umbrella of Brecht-inspired analysis. Id. at 12–13. Pulido argues that the Ninth Circuit’s determination that Stromberg applies in this case merely recognized the fact that no error is harmless if it is based on evidence in genuine dispute. Id. at 18. Pulido concludes that this genuinely disputed fact then unfailingly presents a “grave doubt,” thus satisfying Brecht with Stromberg analysis. See id. Pulido argues that because he not only presented evidence that would have allowed the jury to find him not guilty, but also contested the contemporaneity of the robbery and his intent to aid and abet a murder, Neder’s “harmless error” analysis requires reversal just as Stromberg does. See id. Further, Pulido contends, since judicial weighing of evidence is inappropriate, the reversal should be automatic. See id. at 29. This reconciles Stromberg and Brecht as both an error that the court cannot coherently evaluate and as a “grave doubt.” See id. at 33. As a result, Pulido maintains, there is no tension; the two sets of rules announced in Stromberg and Brecht “work in tandem.” Id. at 41.



Hedgpeth v. Pulido requires the Supreme Court to clarify habeas corpus law by interpreting 28 U.S.C. § 2254(d), which requires federal courts to limit grants to those state decisions that are “contrary to” or “unreasonable” interpretations of established federal law. The Antiterrorism and Effective Death Penalty Act of 1996, of which this provision is a part, was intended to reduce federal habeas petitions while maintaining a path for redressing constitutional wrongs. Michael Pulido, convicted of felony-murder under California state law in a trial comporting a jury instruction on an unconstitutional basis of guilt, seeks confirmation that the Ninth Circuit properly relied on the established Supreme Court law embodied by Stromberg v. California, 238 U.S. 359, (1931). The attorney general of California, on the other hand, seeks reversal of the Ninth Circuit and confirmation that the established federal law requires harmlessness review and that the constitutional error in the state court system was harmless under the standard established by Neder v. United States, 527 U.S. 1 (1999). A decision either way has implications for criminals seeking habeas corpus, the power of federal courts to review state decisions, and the role of the jury in criminal trials.


Prepared by: Victoria Bourke

Edited by: Hana Bae

Additional Sources


Written by 

Edited by