United States v. Hatfield
591 F.3d 945
7th Cir.
2010
Check Treatment
Docket
POSNER, Circuit Judge.

A jury convicted the defendants of conspiracy to burglarize pharmacies, 18 U.S.C. §§ 2118(b), (d), and to distribute controlled substances (including morphine, methadone, oxycodone, fentanyl, alprazolam, cocaine, and hydrocodone), the use of which resulted in death or serious bodily injury, 21 U.S.C. §§ 841(a)(1), (b)(1)(C), 846 — specifically, four deaths, plus a serious bodily injury to a fifth user of the defendants’ drugs. The defendants were sentenced to life in prison, as authorized by section 841(b)(1)(C). The principal issue presented by the appeals concerns the wording of the jury instruction explaining the meaning of the statutory term “results from.” The exact statutory language is “if death or serious bodily injury results from the use of such substance [the defendant] shall be sentenced to a term of imprisonment of not less than twenty years or more than life.”

The instruction began by stating that the jury had “to determine whether the United States has established, beyond a reasonable doubt, that the [victims] died, or suffered serious bodily injury, as a result of ingesting a controlled substance or controlled substances distributed by the defendants or by a defendant.” But then it added that the controlled substances distributed by the defendants had to have been “a factor that resulted in death or serious bodily injury,” and that although they “need not be the primary cause of death or serious bodily injury” they “must at least have played a part in the death or in the serious bodily injury.” The defendants’ lawyer asked that the addition, suggested by the prosecutor, be stricken as a confusing gloss on “results from.” The district judge refused.

Causation is an important issue in many cases in a variety of fields of law and has been so for centuries. Yet it continues to confuse lawyers, in part because of a proliferation of unhelpful terminology (for which we judges must accept a good deal of the blame). In the space of three-and-a-half pages in the government’s brief, one finds the following causal terms: proximate cause, actual cause, direct cause, but-for causation, contributing causation, contributory causation, significant causal connection, sole cause, factor in the victims’ injuries, concurrent cause, meaningful role, possible cause, remote cause, and cause in *948 fact. Black’s Law Dictionary (8th ed.2004) lists 26 terms in the entry for “cause.” The prosecutor was unable at oral argument satisfactorily to differentiate or explain the causal terms listed in his brief, or the three causal terms added to the instruction — “a factor that resulted in,” “primary cause,” and “played a part.”

The parties agree that the statutory term “results from” required the government to prove that ingestion of the defendants’ drugs was a “but for” cause of the deaths and the bodily injury. The death or injury need not have been foreseeable, e.g., United States v. Houston, 406 F.3d 1121, 1124-25 (9th Cir.2005); United States v. Soler, 275 F.3d 146, 152-53 (1st Cir.2002), but the government at least must prove that the death or injury would not have occurred had the drugs not been ingested: “but for” (had it not been for) the ingestion, no injury. That is the minimum concept of cause. See Price Waterhouse v. Hopkins, 490 U.S. 228, 240,109 S.Ct. 1775, 104 L.Ed.2d 268 (1989); Movitz v. First National Bank of Chicago, 148 F.3d 760, 762-63 (7th Cir.1998); United States v. Marlatt, 24 F.3d 1005, 1007 (7th Cir.1994). Is it the entire concept? Is it what “primary cause” and “played a part” would have conveyed to the jury?

At argument the government’s lawyer said that “played a part” refers to but-for causation. But his understanding of but-for causation turned out to be incorrect. For we asked: suppose the ingestion of an illegal drug weakened the victim’s health to the point where he later died of another condition that would not have killed him had he not ingested the drug. Maybe he was healthy until he ingested it, and after and because he ingested it his immune system failed and he died from an overdose of drugs, obtained from someone else, that would not have killed him but for his weakened condition. The government’s lawyer said that ingesting the first drug would not have been a but-for cause of the death. But it would have been: had the victim not ingested it, he would not have died when he did.

Probably what the government’s lawyer meant is that a but-for cause is not always (in fact not often) a cause relevant to legal liability. And that is true, and critical. Suppose a defendant sells an illegal drug to a person who, not wanting to be seen ingesting it, takes it into his bathroom, and while he is there the bathroom ceiling collapses and kills him. Had he not ingested the drug, he would not have been killed. But it would be strange to think that the seller of the drug was punishable under 21 U.S.C. § 841(b)(1)(C).

“Cause” in law, as in life generally, is an opportunistic concept: ordinarily it is the name we attach to a but-for cause (the better term is “necessary condition,” since most but-for causes aren’t considered causes at all) that we’re particularly interested in, often because we want to eliminate it. We want to eliminate arson, but we don’t want to eliminate oxygen, so we call arson the cause of a fire set for an improper purpose rather than calling the presence of oxygen in the atmosphere the cause, though it is a but-for cause just as the arsonist’s setting the fire is. We say that the cause of the death of the drug taker in the bathroom was the improper design or construction of the ceiling rather than the sale of the drug. The reason is that the sale of the drug did not increase the risk posed by the unsafe ceiling — did not increase the risk that this sort of mishap would occur. Brackett v. Peters, 11 F.3d 78, 82 (7th Cir.1993); Zuchowicz v. United States, 140 F.3d 381, 387-89 and n. 7 (2d Cir.1998); Restatement (Third) of Torts § 30 and comment a and illustration 1 (2005). Punishing a drug seller does not reduce building accidents. Punishing him *949 more severely because of the buyer’s death in the bathroom would not cause drug dealers to take care to prevent their sales of drugs from leading by so indirect a route to the death of a buyer; there is no way, in our example, that the seller could have prevented the ceiling from collapsing.

The concept of “marginal deterrence” is pertinent here. More-serious crimes are punished more severely than less-serious ones in part to ensure that criminals are not made indifferent between committing the lesser and the greater crime; if they’re going to commit crimes, at least they should commit the less serious ones. As we explained in United States v. Beier, 490 F.3d 572, 575 (7th Cir.2007), “were robbery punished as severely as murder, a robber would have an increased incentive to murder his victim in order to eliminate a key witness.” See also United States v. Newsom, 402 F.3d 780, 785-86 (7th Cir.2005); Lust v. Sealy, Inc., 383 F.3d 580, 591 (7th Cir.2004); Tracey L. Meares, Neal Katyal & Dan M. Kahan, “Updating the Study of Punishment,” 56 Stan. L.Rev. 1171, 1173-80 (2004). We want drug dealers not to kill their customers inadvertently. But in our hypothetical case of the falling ceiling, nothing the drug dealer did made death more likely. So we would not call the sale of the drugs the “cause” of the death in that case even though it was a necessary condition of it because, had the sale not occurred, the buyer probably would not have been in the bathroom when the ceiling collapsed.

We cannot see what the government’s list of causal terms contributes to an understanding of causation as we have just explained it — especially a jury’s understanding of it since the terms in the list are for the most part unfamiliar to people who haven’t studied law. We particularly don’t understand what a jury would make of “primary cause” and “played a part,” even though those do not sound like technical legal terms, albeit “primary cause” is listed in Black’s law dictionary as a synonym for “proximate cause” — which confuses things further because “proximate cause” usually implies foreseeability, see, e.g., James River Ins. Co. v. Kemper Casualty Ins. Co., 585 F.3d 382, 386-87 (7th Cir.2009); Back v. Hastings On Hudson Union Free School District, 365 F.3d 107, 127-28 n. 21 (2d Cir.2004); United States v. Hanousek, 176 F.3d 1116, 1123 (9th Cir.1999); Restatement (Third) of Torts § 29 comment j (2005), which we know is not required in our case.

In our bathroom-ceiling hypothetical, did taking the drug “play a part” in the taker’s death? In a sense, it did. Was it the “primary cause” of the death? Surely not, but might a jury think it a “secondary cause”? And that a secondary cause was enough to convict? Maybe “played a part” means “was a secondary cause” — for the jury was instructed that it did not have to find that the use of the defendants’ drugs was the primary cause of the deaths or the injury. Might it have thought that if death follows an overdose, the overdose must have “played a part” in the death, even if the death might have occurred without the overdose? Who knows?

The defendants’ objection to the instruction was well taken. All that would have been needed to satisfy it was to eliminate the addition to the statutory language, which was a good deal clearer than the addition and probably clear enough. Elaborating on a term often makes it less rather than more clear (try defining the word “time” in a noncircular way); it is on this ground that some courts, including our own, tell district judges not to try to explain to a jury the meaning of “beyond a reasonable doubt.” United States v. Bruce, 109 F.3d 323, 329 (7th Cir.1997); *950 United States v. Desimone, 119 F.3d 217, 226-27 (2d Cir.1997); United States v. Oriakhi, 57 F.3d 1290, 1300 (4th Cir.1995). Probably the same is true of “results from.”

The government’s lawyer told us that he got the additional language for the instruction from other circuits’ opinions, such as the Eighth Circuit’s opinion in United States v. Monnier, 412 F.3d 859, 862 (8th Cir.2005), and indeed it is the principal case on which the government relies. The opinion did refer to “primary cause” and “played a part,” but it was trying (perhaps not terribly successfully) to explain the difference between a test of causation that requires merely that the defendant’s act be a “contributory cause” (which the court seems to have equated to a “but for” cause) and a test of “proximate cause,” or foreseeability. In any event, the opinion was not quoting from or approving a jury instruction. No case has approved the language that was added to the instruction in this case at the prosecutor’s behest.

Earlier the Eighth Circuit had held, consistently with the Houston and Soler decisions that we cited earlier, that “results from” in section 841(b)(1)(C) does not require proof that the death or bodily injury of the user of the defendant’s drug was foreseeable. United States v. McIntosh, 236 F.3d 968, 972-73 (8th Cir.2001). The defendant in that case had manufactured methamphetamine with another person, who gave a coffee filter that had been used in that process and still had some meth on it to still another person, who gave it to the person who died from ingesting it. It may not have been foreseeable to the defendant that this person, or perhaps that anyone, would die from his meth; and he had even given instructions that the person who ended up dying should not be given any meth. The court held that none of these circumstances mattered because the statute imposes strict liability for a drug offense that results in death. That holding — the irrelevance of unforeseeability — has nothing to do with causation. Instead it illustrates the distinction between cause and legal responsibility, which the interpretation of “results from” as imposing strict liability collapses, though only to the extent of dispensing with any need to show that the defendant should have foreseen the consequence of his lethal act.

We have some misgivings about interpreting “results from” in the statute to impose strict liability. That could lead to some strange results. Suppose that, unbeknownst to the seller of an illegal drug, his buyer was intending to commit suicide by taking an overdose of drugs, bought from that seller, that were not abnormally strong, and in addition the seller had informed the buyer of the strength of the drugs, so that there was no reasonable likelihood of an accidental overdose. Yet the cases are unanimous and emphatic that section 841(b)(1)(C) imposes strict liability — see, besides the Houston, Soler and McIntosh cases, cited earlier, United States v. Robinson, 167 F.3d 824, 830-31 (3d Cir.1999), and United States v. Patterson, 38 F.3d 139, 144-45 (4th Cir.1994)though they might not push their interpretation that far, and though their reasoning might be thought by legal realists somewhat wooden. The cases emphasize the “plain meaning” of the statute, by which they mean simply the omission of any reference to foreseeability or state of mind, and point out that criminal statutes commonly do specify the required state of mind or other ground of culpability (such as negligence) rather than leaving it to be filled in by the judges (as under the Model Penal Code, which provides that proof of guilt of a statute that does not specify a state of mind or other standard of culpability requires proof of at least recklessness, American Law Institute, Model Penal *951 Code § 2.02(3) (1962)). And from this they infer that the omission of any such requirement from section 841(b)(1)(C) was deliberate, and so liability must be strict.

A realistic consideration, however, supports the conclusion: strict liability creates an incentive for a drug dealer to warn his customer about the strength of the particular batch of drugs being sold and to refuse to supply drugs to particularly vulnerable people. And strict liability does not offend against the principle of marginal deterrence in this instance because it does not give the seller an incentive to commit a more serious crime, as in the case where robbery is punished as severely as murder. In any event, the defendants in this case do not challenge the interpretation of the statute as imposing strict liability on them for death or injury to recipients of their drugs.

Still, there was error in the instruction, as we have found. But errors in instructions are not reversible if they are harmless. E.g., Neder v. United States, 527 U.S. 1, 8-10, 119 S.Ct. 1827, 144 L.Ed.2d 35 (1999); United States v. Ramsey, 406 F.3d 426, 432 (7th Cir.2005). Although the government does not argue that the error in the “results from” instruction (if it was an error, as we hold that it was) was harmless, if we were convinced it was we would not reverse, United States v. Parmelee, 42 F.3d 387, 391-94 and n. 6 (7th Cir.1994); see also United States v. Jewel, 947 F.2d 224, 228 n. 5 (7th Cir.1991); United States v. Giovannetti, 928 F.2d 225 (7th Cir.1991) (per curiam); United States v. Gonzalez-Flores, 418 F.3d 1093, 1099-1102 (9th Cir. 2005) — to do so would give the defendants a shot at acquittal were they fortunate enough to have an unreasonable jury at their retrial. But we do not think it was harmless. The evidence regarding the cause of the serious injury of the one victim and the deaths of the others, though strong enough to justify a verdict of guilt beyond a reasonable doubt, was not conclusive. In each case the victim was found to have taken multiple drugs, some probably or possibly not distributed by the defendants. In the case of the nonfatal injury (respiratory arrest), the testifying physician thought it more likely that the drug probably supplied by the defendants had caused the injury rather than the cocaine that the victim had also ingested, but he did not rule out the possibility that the cocaine was responsible. With regard to another victim, the medical evidence was that the methadone he apparently received from one of the defendants “would have been sufficient to kill him.” But he had another drug in his system and it is unclear how a juror would have fitted that evidence to the “played a part” and “primary cause” templates that he was asked to use to interpret “results from.”

So the case must be retried; for guidance on remand we’ll address the defendants’ challenges to the district court’s other rulings.

1. The special verdict form concerning the drug user who was seriously injured omitted the date of the overdose and thus, the defendants argue, “constructively amended” the indictment, United States v. Pigee, 197 F.3d 879, 887 (7th Cir.1999); United States v. Willoughby, 27 F.3d 263, 266 (7th Cir.1994), which specified the date. But the jury instructions referred to the count of the indictment that did so and the judge twice reminded the jury of it, so there is no reasonable likelihood that the jury convicted the defendants on the basis of an overdose not charged in the indictment.

2. One of the dead was an informant in another case against one of the *952 two defendants. That case was dismissed on motion by the prosecutor when the informant died. The government was permitted to present certified documents from that case, including a criminal complaint alleging that the defendant had sold oxycodone to the informant and an order dismissing the case because of the informant’s death, to back up its argument that the defendants had planned to kill her in order to stop her from testifying. The evidence consisted of public records, which usually are admissible even though they are hearsay, Fed.R.Evid. 803(8), but there is an exception for the use in criminal cases of records that set forth “matters observed by police officers and other law enforcement personnel.” These are not admissible. Rule 803(8)(B).

The police officer who had signed the criminal complaint in that case testified at the trial of the present case about the proceedings in that other case, including the allegations in the complaint that he had drafted. So he was available for cross-examination. That might seem to cure any objection to the introduction into evidence of the records of that case. “The apparent concern of the drafters [of the exception in Rule 803(8)(B)] was that use of records in criminal cases would cause ‘almost certain collision with confrontation rights.’ ” United States v. Blackburn, 992 F.2d 666, 671 (7th Cir.1993). And during floor debates on the rule, “concern was expressed that [without the exception, Rule 803(8)] would allow the introduction against the accused of a police officer’s report without producing the officer as a witness subject to cross-examination.” 2 McCormick on Evidence § 296 (6th ed.2006). But there is more to the exception than a concern with unavailability of cross-examination. There is also a concern that reports by law enforcers are less reliable than reports by other public officials because of law enforcers’ adversary relation to a defendant against whom the records are sought to be used. United States v. Rosa, 11 F.3d 315, 332-33 (2d Cir.1993); United States v. Hernandez-Rojas, 617 F.2d 533, 535 (9th Cir.1980). Moreover, the police officer’s key testimony in this case' — that the prosecutor had dismissed the case because he didn’t have an informant who could testify — was hearsay; he was testifying to the truth of what someone else, the prosecutor, had told him.

All this is of no moment, however, because the key document is the order dismissing the criminal complaint, and although it does mention the reason the prosecutor gave for asking the court to dismiss the complaint, the order is a public record of the court’s reason (the informant’s death) rather than a record of observations by law enforcement officers. See United States v. Lechuga, 975 F.2d 397, 398-99 (7th Cir.1992). The case is thus like United States v. Hernandez-Rojas, supra, which held that the law-enforcement exception did not bar the admission into evidence of a warrant of deportation. The purpose of the exception — to exclude records created in an adversarial setting and therefore likely to be tendentious — was inapplicable to the notation that the defendant was to be deported to Mexico. That was “a ministerial, objective observation, which has inherent reliability because of the Government’s need to keep accurate records of the movement of aliens. It has none of the features of the subjective report made by a law enforcement official in an on-the-scene investigation, which investigative reports lack sufficient guarantees of trustworthiness because they are made in an adversary setting and likely to be used in litigation.” 617 F.2d at 535.

Furthermore, the truthfulness of the reason for the dismissal of the other case (whether the real reason, or the reason *953 given by the prosecutor and repeated by the police officer in his testimony) was secondary to the key fact that the dismissal established — which was uncontested— that the victim of the overdose was a government informant in a case against one of the defendants. That supplied motive for the overdose that killed the informant, and while motive is not an element of 18 U.S.C. § 841(b)(1)(C) (remember that liability is strict), proof of the defendants’ lethal motive increased the likelihood that the victim had died from the defendants’ drugs rather than from a drug that she had obtained elsewhere.

3. The defendants sought to introduce in evidence out-of-court statements by a man named Willbrand, recorded in a police report, to the effect that he along with three other persons, rather than either of the defendants, had committed one of the pharmacy burglaries that the defendants were accused of. Although the statements were against Willbrand’s penal interest, the district judge refused to allow them into evidence under Fed. R.Evid. 804(b)(3). His ground was that the circumstances didn’t “clearly indicate the trustworthiness of the statement” because Willbrand had changed his story twice before admitting his involvement in the burglary. United States v. Jackson, 540 F.3d 578, 588-89 (7th Cir.2008). Yet there is no suggestion that he knew the defendants; in none of his versions of the burglary did he implicate them; and his ultimate version was corroborated by the fact that he had dialed 911 while the burglary was in progress and had stated accurately in his initial statement to the police that one person had broken into the pharmacy by shattering the glass on the front door, which was evidence that Willbrand had indeed been present at the commission of the crime. And the evidence against the defendants concerning this particular burglary was pretty weak. So Willbrand’s statements should have been admitted. The error was harmless, however, because the government presented evidence that the defendants had committed between 85 and 100 burglaries of pharmacies, and they were not charged with specific burglaries. But should the government in the new trial that we are ordering decide to present evidence that the defendants committed this particular burglary, Willbrand’s statement should be admitted.

Since the judge’s evidentiary rulings did not amount to reversible error, the new trial that we are ordering will be limited to the “results from” charge. The convictions for conspiracy to burglarize pharmacies and distribute controlled substances were supported by overwhelming evidence unrelated to the evidence of the causes of the injury and deaths. The defendants do not argue that the erroneous instruction on the death charge contaminated the jury’s consideration of the other charges; nor would such an argument be plausible given the overwhelming evidence of the defendants’ guilt of those charges.

Reversed and Remanded.