JOSEPH PERRONE v. UNITED STATES OF AMERICA
No. 16-2437
United States Court of Appeals For the Seventh Circuit
ARGUED JANUARY 4, 2018 — DECIDED MAY 14, 2018
Before WOOD, Chief Judge, and HAMILTON and BARRETT, Circuit Judges.
Appeal from the United States District Court for the Southern District of Illinois. No. 3:14-cv-00281-DRH — David R. Herndon, Judge.
BARRETT,
I.
At approximately 4 a.m. on April 17, 2008, Terry Learn and her coworker Madonna Narog went to Narog‘s hotel room, where they did heroin and cocaine for several hours. They left the hotel around 8 a.m. to purchase more cocaine, about fifty dollars’ worth for Learn and twenty-five dollars’ worth for Narog. They returned to the hotel and did cocaine until close to noon, when Learn left for her shift at Roxy‘s Night Club. Narog saw Learn again around 2 p.m., when Narog went to the club to pick up some money, and again at 8 p.m., when Narog was beginning her shift and Learn was ending hers.
After her shift, Learn met her boyfriend, Joseph Perrone, and went back to his home. According to Perrone, the two made a suicide pact. After watching Learn inject herself with a mixture of cocaine and water, Perrone told her that she had not taken enough to kill herself. He then prepared and injected 7.5 grams of cocaine into Learn. Perrone later told the police that Learn convulsed, fell to the floor, and died immediately after he injected her for the last time. He did not specify the time at which he administered the final injection, but he said that it happened on April 18th. It was therefore at least four hours after Narog saw Learn at the shift change and at least twelve hours after Narog last saw her do any drugs not distributed by Perrone.
Perrone moved Learn‘s body to her apartment. He wiped his fingerprints off the syringe and put it into Learn‘s hand. As he stipulated in his plea agreement, he aimed “to create the false impression that Terry Learn had died alone in her own residence.” The body was not discovered until April 26th, when a concerned neighbor flagged down police to report that she had not seen Learn in several days. Police officers discovered Learn‘s body in her apartment. According to the coroner‘s report, the cause of death was “[c]ombined toxicity with cocaine, ethanol and opiates.”
Several months later, Perrone was arrested on an unrelated firearms charge. He chose that time to confess to police that he had killed Learn, describing what he had done as “premeditated murder.” During this interview, he told the police that he gave Learn one injection of an unspecified amount of cocaine; during a second interview a few weeks later, he said that he injected Learn with 7.5 grams of cocaine in three separate injections of 2.5 grams each.
The government obtained an indictment against Perrone for distributing a controlled substance in violation of
On the day before Perrone was sentenced, the Seventh Circuit decided United States v. Hatfield, 591 F.3d 945 (7th Cir. 2010), which held that the “death results” enhancement requires the government to prove that “ingestion of the defendants’ drugs was a ‘but for’ cause of the death[].” Id. at 948. Hatfield rejected jury instructions that used vaguer, less demanding language to describe the necessary causal relationship; it said that the district court could not summarize the “death results” enhancement as requiring the jury to find only that the illegal drugs “played a part” in the victim‘s death. Id. at 949.
At sentencing the next day, the district court applied the “death results” enhancement and sentenced Perrone to 240 months’ imprisonment. Before imposing the sentence, the district judge said that he had reviewed Perrone‘s Stipulation of Facts to see “what impact, if any, the Rex Hatfield case was going to have on this case.” Perrone‘s attorney did not engage this point with the judge, nor did he inform Perrone about Hatfield. Instead, he once again agreed that the sentencing enhancement applied. Perrone did not appeal his sentence. He eventually received an 80-month reduction of his sentence for assistance to the government, a possibility contemplated by the plea agreement and that Perrone and the district court had discussed at his sentencing hearing.
Four years later, the Supreme Court decided Burrage v. United States, 134 S. Ct. 881 (2014), which effectively ratified Hatfield‘s standard of causation. The Court held that the “death results” enhancement ordinarily requires the government to prove that the victim would have lived but for the unlawfully distributed drugs. Id. at 888. In Burrage, the victim died with multiple drugs in his bloodstream, including metabolites from heroin that had been distributed by the defendant. Although morphine, a heroin metabolite, was the only drug present at a level above the therapeutic range, the government‘s experts could not say whether the victim would have lived if he had not taken the heroin. They testified only that heroin was a “contributing factor” to a death caused by “mixed drug intoxication.” That testimony dovetailed with instructions requiring the jury to find “that the heroin distributed by the Defendant was a contributing cause of [the victim‘s] death.” Id. at 886. The Court said that the statute requires the government to show more than that the distributed drug contributed to the victim‘s death. The enhancement applies when “death or serious bodily injury results from the use of [the distributed] substance,” which means that the substance must be a “but for” cause of the death. Id. at 887–88.
Within a month of Burrage, Perrone filed a petition to vacate or alter his sentence pursuant to
II.
Perrone‘s strongest argument is that he is actually innocent of the “death results” sentencing enhancement. He claims that when he entered his plea agreement and pleaded guilty, he did not know that the enhancement required but-for causation. Since then, Burrage has made the standard clear, and under it, he says, there is insufficient evidence to show that Learn‘s death resulted from the cocaine he gave her. Although Perrone generally waived his right to raise collateral challenges, the waiver excludes collateral attacks based on “any subsequent change in the interpretation of the law” by the Supreme Court that is declared retroactive and renders Perrone innocent.
A.
The government contends that Perrone procedurally defaulted his Burrage claim by failing to raise it on direct appeal. McCoy v. United States, 815 F.3d 292, 295 (7th Cir. 2016) (“A claim cannot be raised for the first time in a
Perhaps because of this overlap, both parties assume (as did the district court) that the “actual innocence” exception to procedural default is available to Perrone. That assumption is doubtful. The point of the exception is to ensure that “federal constitutional errors do not result in the incarceration of innocent persons.” Herrera v. Collins, 506 U.S. 390, 404 (1993). The Supreme Court has flagged the possibility that actual innocence might be enough to justify collateral relief in a capital case on the theory that the execution of one who is actually innocent violates the Eighth Amendment. Id. at 405. Apart from that potential exception, however, the Court‘s “habeas jurisprudence makes clear that a claim of ‘actual innocence’ is not itself a constitutional claim, but instead a gateway through which a habeas petitioner must pass to have his otherwise barred constitutional claim considered on the merits.” Id. at 404. This is a problem for Perrone. He does not ask us to determine that he is actually in-nocent so that we can consider a claim of constitutional error; rather, his innocence of the “death results” enhancement is the error he asks us to correct.1
B.
Perrone‘s petition claims that the Supreme Court‘s narrowed interpretation of the “death results” enhancement renders him actually innocent of causing Learn‘s death. The Supreme Court has held that when a subsequent statutory interpretation narrows the elements of a crime, revealing that the petitioner has been convicted and sentenced for “an act that the law does not make criminal,” the petitioner has suffered “a complete miscarriage of justice” that justifies relief under
The parties agree that Perrone has a cognizable claim,2 but they disagree about the standard we should apply in as-sessing whether Perrone is actually innocent of causing Learn‘s death. Perrone argues that the standard should be the one that Schlup v. Delo provides for determining “actual innocence” in the context of procedural default. In Schlup, the Court held that “a petitioner must show that it is more likely than not that no reasonable juror would have found petitioner guilty beyond a reasonable doubt.” Id. at 327. The government, in contrast, says that Perrone must show that it is more likely than not that no reasonable judge would have found him guilty by a preponderance of the evidence. The government‘s insistence on a “preponderance” standard is grounded in its belief that we must treat Perrone‘s petition as asserting innocence of a sentencing factor rather than innocence of an element of the crime.
The government stresses that when Perrone pleaded guilty and was sentenced, the “death results” enhancement was
The government‘s position is inconsistent with our holding in Krieger v. United States, 842 F.3d 490 (7th Cir. 2016). There, in holding that Burrage announced a substantive rule that applies retroactively on collateral review, we described the “death results” enhancement as the Supreme Court did in Burrage: as an element of the crime. Krieger, 842 F.3d at 500 (“[T]he rule announced in Burrage altered the range of conduct that the law punishes.“); Burrage, 134 S. Ct. at 887 (characterizing the “death results” enhancement as “an element that must be submitted to the jury and found beyond a reasonable doubt“). Had we thought ourselves bound by Alleyne‘s non-retroactivity (which we acknowledged) to treat Burrage as changing only the scope of a sentencing factor, we presumably would have relied on Narvaez when we held that Krieger‘s Burrage error was cognizable under
To be sure, Krieger did not address the standard of review that would be applicable when a court collaterally reviews whether there is sufficient evidence to support application of the enhancement according to Burrage‘s standard of but-for causation. In that respect, Krieger technically
We thus reject the government‘s attempt to slice Burrage‘s characterization of the “death results” enhancement (as an element of crime) away from its definition of what application of that enhancement requires (but-for causation). Because Alleyne is not retroactive, Perrone could not get relief on the ground that the “death results” question went to a judge rather than jury. Once he is before us with a cognizable claim, however, there is no reason for us to describe his claim as something it is not. Burrage, unlike Alleyne, is retroactive, and it makes clear that Perrone‘s claim goes to his innocence of a crime, not a sentence. Whether the government has proven an element of the crime is always a question for the jury. That means that Perrone‘s burden is to show that it is more likely than not that no reasonable juror would have found him guilty beyond a reasonable doubt of causing Learn‘s death. Schlup, 513 U.S. at 327.
C.
This dispute is about causation, so we will begin by clearly stating what “but for” causation requires. It does not require proof that the distributed drug was present in an amount sufficient to kill on its own. The Court explained in Burrage that death can “result[] from” a particular drug when it is the proverbial “straw that broke the camel‘s back.” 134 S. Ct. at 888. As the Court put it: “if poison is administered to a man debilitated by multiple diseases, it is a but-for cause of his death even if those diseases played a part in his demise, so long as, without the incremental effect of the poison, he would have lived.” Id. Here, then, the fact that other substances in Learn‘s bloodstream played a part in her death does not defeat the government‘s claim that her death resulted from the cocaine Perrone gave her. A jury could have found him guilty of causing her death if it concluded beyond a reasonable doubt that Perrone‘s cocaine pushed her over the edge.
In Burrage, the Court left open the possibility that the government could prove causation another way: it said that strict “but-for” causation might not be required when “mul-tiple sufficient causes independently, but concurrently, produce a result.” 134 S. Ct. at 890. In other words, the “death results” enhancement might apply to a defendant who distributes a lethal dose of cocaine to a person who also consumes a lethal dose of heroin. The government suggests that a jury could have found Perrone guilty on this theory as well, because Perrone gave Learn a lethal amount of cocaine. We need not decide whether this second theory is viable, however, because there is sufficient evidence to have permitted a jury to find Perrone guilty on the first.
Perrone admitted that he distributed 7.5 grams of cocaine to Learn in a deliberate attempt to kill her, that he personally injected Learn with cocaine intending to kill her, and that she convulsed and died immediately after he injected her. That in itself strongly supports the conclusion that Learn‘s death resulted from the cocaine Perrone administered. It also distinguishes Perrone from the defendants in Burrage, Hatfield, and Krieger, none of whom stated that they had distributed the drug to the user with the intent to kill.
The best evidence on Perrone‘s side is the coroner‘s report, which listed the cause of death as “[c]ombined toxicity with cocaine, ethanol and opiates.” This, Perrone says, is the kind of evidence the Court
Before the grand jury, the government seemed focused on eliciting testimony that the cocaine Perrone distributed was independently sufficient to kill Learn. Nanduri repeated the conclusion she reached in the autopsy report: that Learn‘s cause of death was the combined toxicity of cocaine, ethanol, and opiates.4 When the prosecutor followed up with a question about which substance was “primarily responsible for her death,” Nanduri clarified that “if she just had cocaine in her system and the other two drugs were not present, then cocaine would be the toxic agent that killed her.” The prosecutor returned to this point a few minutes later, saying “I don‘t mean to beat this into the ground, but it is a very important point for us. It is your testimony under oath that to a medical certainty this quantity of cocaine found in this woman‘s blood would have killed her all by itself?” Nanduri replied “yes.” Nanduri repeated several more times before the grand jury that the cocaine in Learn‘s bloodstream was a lethal dose. She did not, however, testify that alcohol and morphine in Learn‘s system were not lethal. She said that the alcohol was not at a level she would expect to be fatal, but she expressed uncertainty about the role the morphine had played in Learn‘s death. Although Learn‘s morphine level was low, Nanduri explained that whether such a low dose could kill a person depends on numerous factors, including the person‘s past history of using opiates. She also said that a person‘s morphine level might be deceptively low if the person became comatose and continued metabolizing the morphine before dying.
Nanduri‘s testimony thus does not establish that cocaine was the but-for cause of Learn‘s death. But the government has testimony from another expert, Dr. Chris Long, who did expressly state that Learn would have lived but for the cocaine. Long prepared a toxicology report on Learn‘s body in 2008. After reviewing his report in 2014, he confirmed that the alcohol would not have killed Learn “absent the cocaine” and that “[t]he opiate is of no significance.” A reasonable juror could credit Long‘s testimony.
Perrone‘s best response is to say that even if cocaine caused Learn‘s death, the cocaine that killed her was not the cocaine he gave her. Learn had, after all, done a fair amount of cocaine with Narog the day before. And given her pattern of cocaine use, it is at least possible that Learn did some cocaine during her shift at work. Perrone may have injected cocaine into a woman with an already-lethal amount of cocaine in her body.
This evidence helps Perrone, but only a little. There is no evidence that Learn acquired or took any cocaine between starting her shift at noon and meeting up with Perrone sometime after the 8 p.m.
That Learn arrived at Perrone‘s without enough cocaine in her system to kill her is bolstered by what happened when she got there. Perrone said that the two had a suicide pact, which suggests that they both thought Learn needed to consume more drugs if she wanted to end her life. And after Learn injected herself with a dose of cocaine, Perrone himself made the judgment that what she had taken was not enough to kill her. According to his own statement, he gave Learn an additional 7.5 grams of cocaine because he concluded that she would not die unless she had more. He told the police that Learn convulsed and fell to the floor immediately after he injected her, and he later characterized what he did as “premeditated murder.” In his stipulation of facts, Perrone admitted that Learn died “immediately after receiving the injection.”
Given this evidence, Perrone cannot carry his burden of showing that it is more likely than not that no reasonable juror would have voted to find him guilty beyond a reasonable doubt. He is thus not entitled to relief on the ground that he is actually innocent of causing Learn‘s death.
III.
Perrone also argues that his sentencing counsel was constitutionally ineffective for failing to tell him about (and possibly not even knowing about) Hatfield. Had Perrone been aware of Hatfield, he says, he might have sought the court‘s permission to withdraw his plea on the ground that it was not knowing and voluntary. Because he did not know that the “death results” enhancement required the government to show but-for causation, he did not correctly understand what he was pleading to when he stipulated that he “caused” Learn‘s death.
To prevail on his ineffective assistance claim, Perrone must show both deficient performance and prejudice. Strickland v. Washington, 466 U.S. 668, 687 (1984). On the performance prong, he “must overcome the ‘strong presumption that counsel‘s conduct falls within the wide range of reasonable professional assistance.‘” Wyatt v. United States, 574 F.3d 455, 458 (7th Cir. 2009) (quoting Strickland, 466 U.S. at 689). On the prejudice prong, he must show that “but for counsel‘s errors, there is a reasonable probability that the result would have been different.” United States v. Graf, 827 F.3d 581, 584 (7th Cir. 2016). In the context of a guilty plea, a petitioner demonstrates prejudice by “show[ing] that there is a reasonable probability that, but for counsel‘s errors, he would not have pleaded guilty and would have insisted on going to trial.” Hill v. Lockhart, 474 U.S. 52, 59 (1985). As with the first prong, there is a presumption that the petitioner has not suffered prejudice. Graf, 827 F.3d at 584–85.
Even assuming that Perrone could show “deficient performance” on the part of his counsel, it is unlikely that he could satisfy the “prejudice” prong. The evidence of causation was strong and his plea agreement gave him the opportunity to obtain a sentence reduction for cooperating with
At the end of the day, however, any difficulties Perrone has on the merits do not matter because his claim is barred as untimely. Although he was not required to bring his ineffective-assistance claim in his direct appeal, Massaro v. United States, 538 U.S. 500, 509 (2003), he was still required to comply with
Perrone does not dispute that his claim is untimely. Instead, he attempts to escape the bar by contending that the government forfeited its statute-of-limitations defense. Perrone raised his ineffective-assistance claim before the district court for the first time in his reply to the government‘s response brief. He faults the government for not asserting the limitations defense in response, but the district court‘s local rules prohibited the government from filing a surreply. True, the district court chose to treat Perrone‘s reply brief as an amended petition, so it appears with the benefit of hindsight that the government could have filed a new response. But the filing was denominated as a reply brief, not as an amended petition; it did not reproduce the claims that had appeared in Perrone‘s original petition; and Perrone had repeatedly told the district court that his court-appointed counsel was not authorized to amend his petition. It was therefore reasonable for the government to conclude that it lacked the ability under the local rules to respond to the newly raised claim of ineffective assistance. And even if this were forfeiture, we would find it excused due to the understandable confusion in the district court. See Wood v. Milyard, 566 U.S. 463, 471 (2012) (allowing a court of appeals to consider even sua sponte “a nonexhaustion argument ‘inadvertent[ly]’ overlooked by the State in the District Court“). Because Perrone filed his petition after
IV.
The district court also correctly denied Perrone‘s motion for an evidentiary hearing on his claims. A petitioner under
V.
For the reasons stated above, the judgment of the district court is AFFIRMED.
