I dissent. The majority conclude that the “wanton disregard of human life” definition of implied malice required a finding of the defendant’s subjective awareness of the life threatening risk and that there was no error in instructing the jury in the language of the 1983 revision of CALJIC Nos. 8.11 or 8.31. The majority rely on several cases which use language the same as or similar to the quoted phrase.
However, our concern is not with legal definitions and precedents to define the term. We must view the instruction as “a reasonable juror could have interpreted the instruction.” (Sandstrom v. Montana (1978) 442 U.S. 510, 514 [61 L.Ed.2d 39, 45, 99 S.Ct. 2450].) When this is done, it is clear that the quoted phrase encompasses cases not only where defendant realized the risk but also where he acted recklessly violating his duty to care for the child even though he did not realize the risk of his conduct. Moreover, the balance of the instruction, stating “conscious” disregard of the risk as an alternative to “wanton” disregard of the risk, made it likely that a jury would understand that two tests were stated and only one required subjective awareness of the risk. In the circumstances of this case, where the evidence was not only sharply conflicting but was also subject to conflicting inferences and where the jury engaged in lengthy deliberations and focused on the instruction, the error must be held prejudicial.
The instruction provided: “Malice is implied when the killing results from an intentional act involving a high degree of probability that it will result in death, which act is done for a base, antisocial purpose and with a wanton disregard for human life when the killing results from an intentional act, the natural consequences of which are dangerous to life, which act was deliberately performed by a person who knows that his conduct endangers the life of another and who acts with conscious disregard for life.” (CALJIC No. 8.11 (1983 rev.) (4th ed. pocket pt.), italics added.) The definition of second degree murder used virtually identical language. (CALJIC No. 8.31 (1983 rev.) (4th ed. pocket pt.).)
It is settled that “a finding of implied malice depends upon a determination that the defendant actually appreciated the risk involved, i.e., a subjective standard. (People v. Phillips [(1966)] 64 Cal.2d [574,] 588 [51 Cal.Rptr. 225, 414 P.2d 353].)” (People v. Watson (1981) 30 Cal.3d 290, 296-297 [179 *1224Cal.Rptr. 43, 637 P.2d 279], italics in original.) Although the word “disregard” includes rejection after appreciation or consideration, it is not limited to such rejection but is also often used in a broader sense where there has been no subjective awareness to mean “to give no thought to: to pay no attention to.” (See Webster’s New Internal. Dict. (3d ed. 1961) p. 655.) Thus the term “disregard for life” encompasses not only situations where the actor subjectively appreciates the risk to life but also where the risk is not contemplated or appreciated. The word “wanton” does not eliminate or even reduce the danger that a jury might imply malice when it was of the view that' the defendant did not appreciate the risk to life.
The danger is accentuated because in the second part of the instruction the jury is told that it may imply malice on the basis of knowledge of the risk and ‘ conscious disregard” for life and this second definition is separated from the first by an “or.” While the use of the word “or” does not necessarily mean that the instruction contains two differing definitions of malice, the instruction viewed as a whole certainly implies that there are two definitions stated in the alternative. (See e.g., Houge v. Ford (1955) 44 Cal.2d 706, 712 [285 P.2d 257].) The prosecutor in his argument said there were two. Both instructions are concerned with “disregard” for life; one adds the element of knowledge and modifies disregard with “conscious,” while the other modifies with “wanton” and adds the elements of a high degree of probability of death and a “base antisocial purpose.” The jury that focuses on the definitions can only conclude that while they may and do overlap they are different and the difference lies in the difference between “wanton,”! on the one hand, and “conscious” and the knowledge requirement on the other.
Although the difference may not be strikingly apparent and perhaps is a subtle one when the instructions are read to the jury, the difference must be viewed as apparent to the jury when, as here, the jury after two days of deliberation specifically asks for a written copy of the instruction and has it in the room to study during the next two days of deliberations.
There are cases, as the majority point out, that have phrased the definition of implied malice in terms of “wanton disregard” and cases which have treated the “wanton disregard” definition as the same as the “conscious disregard” definition. (Maj. opn., ante, at pp. 1218-1219.) It is unclear whether the courts were erroneously suggesting that malice may be found on an objective basis or whether they failed to appreciate that the “wanton disregard” standard would permit conviction of murder not only where the defendant is conscious of the risk to life but also where he did not turn his mind to the obvious risk.
*1225In any event, our function is not to interpret the instruction on the basis of legal precedent. Rather, we must interpret the instruction as a reasonable juror could have interpreted it. (Sandstrom v. Montana, supra, 442 U.S. 510, 514.) The cases relied upon by the majority may not be held to authorize convictions of second degree murder for persons who did not actually appreciate the risk to human life involved. Because the plain language of the instruction does so, it is erroneous. (People v. Protopappas (1988) 201 Cal.App.3d 152, 162-164 [246 Cal.Rptr. 915]; People v. James (1987) 196 Cal.App.3d 272, 290-291 [241 Cal.Rptr. 691].) The fact that the instruction also authorizes proper convictions does not save it.
The determination whether the error was prejudicial in the instant case involves evaluation of the danger that under the evidence the jury would convict defendant of murder even though it believed that he did not appreciate the risk of his conduct.
Defendant testified that he picked up his two-year-old stepdaughter, Jaclyn, from the babysitter after work and took her home. She had been fussy, and he gave her a small amount of wine in a baby bottle. While she was on the couch watching television he went into the kitchen and began to prepare dinner. He heard a loud noise, ran out of the kitchen area, and saw Jaclyn at the middle landing of the stairs. Finding that she was not breathing, he started using mouth-to-mouth resuscitation, and called for aid. He denied giving Jaclyn any cocaine or having any cocaine in the house.
The paramedics came and took Jaclyn to the hospital where a surgeon sought to relieve the pressure on the brain by drilling holes in the skull. The medical personnel were unsuccessful, and she died a little over an hour and a half after the paramedics arrived at defendant’s home.
Death was due to injury of the brain causing swelling with portions of the brain being forced through the surgical incisions. There was also a fracture of the back of the skull which all of the doctors either said or assumed was related to the brain injury. The expert witnesses also agreed that the impact causing the fracture was due to a padded instrumentality because of the limited amount of external bruising. The expert witnesses further agreed that a great amount of force was required to cause the fracture because it was caused by impact with a padded instrumentality and occurred in the thicker part of the skull. About a year after the death, the body was exhumed and a cervical injury was found.
Chemical analysis showed a small amount of alcohol in the body. Cocaine was found in the stomach in the amount of 25 milligrams and in the liver in the amount of 1.5 micrograms. Although a radio-immuno assay test *1226indicated: cocaine in the blood, the gas chromatogram, which is accurate to 50 nanograms, detected no cocaine in the blood. The autopsy surgeon testified that according to his office’s practices the minimum fatal range for cocaine ingestion was considered between 0.1 and 20 micrograms per gram of tissue level. The doctor said that because the amount of cocaine found in the liver Was within the range, cocaine was a possible cause of death. In addition, he said that because cocaine speeds up the heartbeat, it could be a contributing cause since the ultimate cause was heart failure. The doctor conceded that the amount of cocaine found could be ingested without a fatal reaction.
He further testified that when cocaine is absorbed through the mucus membrane of the mouth or nose, absorption can be achieved within a 30- to 40-minuté period. Absorption through the stomach is a little slower with the peak level achieved at approximately the one hour level. As cocaine is absorbed jit is broken down into its basic components. When cocaine is absorbed into the blood, it is distributed and will stimulate the brain. Although his testimony as to absorption appears to show that the cocaine was ingested shortly or immediately before death because none was found in the blood and almost all remained in the stomach, he stated that his findings were consistent with cocaine ingestion shortly before the skull fracture. On cross-examination, he said that the cocaine could have been ingested after the fracture. However, on neither direct nor cross-examination on this subject was the doctor’s attention called to the fact that decedent survived for more than one and one-half hours after the skull fracture.
The autopsy surgeon testified that the brain injury was not caused by shaking the child because, if it had been, there would be bruising on both sides of the brain and no bruise was found on the side of the brain opposite the injured side. He did not testify as to the cause of the fracture.
The prosecution and the defense each presented two eminent doctors to testify with respect to the cause of the skull fracture. Dr. Noguchi was chief medical examiner with the Los Angeles County Coroner’s office and has an international reputation as a forensic pathologist, teaching and lecturing in many places. Dr. Chadwick is a pediatrician from San Diego, has academic credentials and has written extensively on child abuse. They both concluded that the fracture was not caused by striking the head on the edge of one of the carpeted stairs because the height of the stairs was not sufficiently great to develop sufficient force to cause the fracture. Dr. Noguchi added that bruises to the knees and elbows which would be expected in a fall down the stairs were not found. He said a blow delivered by hand with great force could have caused the injury. While Dr. Chadwick stated that a blow delivered by hand could possibly have caused the injury, it was more likely *1227to occur as a result of decedent being slammed hard against a carpeted surface. Although such would not explain the cervical injury, that injury might have been caused by shaking the child. The doctor also said that the injuries could have been inflicted by slamming the child against a wooden chair covered by a towel.
Doctor Foltz is professor of neurosurgery at the University of California, Irvine, was a founding member of the International Pediatric Neurosurgery Society, author of one hundred and ten papers over twenty-two years, and during the last five years had three or four pediatric head trauma cases per month. Doctor Root, who worked for the San Bernadino County Coroner’s office, had performed over 15,000 autopsies, provided one of the 37 resident training programs approved in the United States, and was an associate clinical professor of pathology at Loma Linda University Medical Center. The defense doctors rejected the claim that the fracture could have been caused by a blow from the hand. They stated that for a hand blow to generate enough force to cause the fracture the head would have to be held rigid and then there would be bruises at the place where the head was held. In the absence of rigidity the body would absorb much of the force resulting from the padded instrumentality preventing fracture. There were no bruises that would have been necessary if the head had been held rigid. Dr. Foltz said he had treated several children with similar fractures resulting from falls from high chairs. He did not indicate whether the fractures were caused by the head striking the high chair or the floor.
The defense experts were of the view that the fracture could have been caused by decedent falling and striking the edge of a stair.
In determining whether the error in instructing on implied malice was prejudicial, it is probable that if the jury found that defendant caused the fracture by hitting decedent with the great amount of force required to cause the fracture with a padded instrument, it would further infer that defendant knew of the risk of death whether defendant hit her with his hand, threw her on the floor, smashed her against the towel-covered chair, or some other active conduct developing the great force required. Any error in instructing on “wanton” disregard probably would not be prejudicial.
However, the expert evidence as to whether decedent fell and struck her head on the stairs was equally divided, and there is a substantial question whether the jury determined beyond a reasonable doubt that the fracture occurred as a result of active conduct by defendant rather than the fall. Moreover, as a matter of common sense, it must be recognized that the destructive force is greater when the injured person walks into the punch than when the head is free to roll with the punch. Even Doctor Chadwick, *1228although stating that the fracture could be caused by the hand, indicated his belief that the fracture was caused by decedent going to the padded surface rather thán bringing the padded surface to decedent. I cannot agree with the majority’s assertion (maj. opn., ante, at p. 1222) that the jury rejected the defense’s claim that the fracture was caused by the fall.
In the circumstances of the instant case, there is a substantial likelihood that the jury, under the “wanton” instruction concluded that malice was implied on the basis of reckless behavior of defendant whether or not decedent’s fracture was caused by a fall down the stairs. Although, as pointed out above, the testimony of the autopsy surgeon as to absorption of cocaine seems to rule out defendant as a source of the cocaine, the surgeon also testified that his findings were consistent with ingestion of the cocaine prior to the fracture, and defense counsel did not argue that the time element excluded defendant as a source of the cocaine. Thus, the jury probably concluded, as the majority concludes (maj. opn., ante, at p. 1222), that defendant furnished the cocaine. If so, the jury under the instructions given may have further concluded that malice was implied on the basis of the cocaine evidence, the evidence that defendant furnished some wine and the evidence that he failed to secure the gate at the top of the stairway, because they showed a “wanton disregard” of the risk to the life of decedent. The jury may have done so under the erroneous instruction even though it was of the view that defendant had not appreciated the risk or even though it was unable to agree whether defendant was aware of the risk to life caused by the combination of his several actions.
In his argument to the jury, the prosecutor treated the implied malice instruction! as providing two definitions of implied malice. The jury deliberated for four days. After two, it requested a copy of the instruction defining murder, and it was given an instruction which erroneously permitted conviction on the basis of “wanton disregard” of the risk to life without requiring a finding that defendant appreciated the risk and consciously disregarded it.
The prejudice standard of “harmless beyond a reasonable doubt” of Chapman v. California (1967) 386 U.S. 18, 21 [17 L.Ed.2d 705, 709, 87 S.Ct. 824, 24 A.L.R.3d 1065] is applicable to an error in instruction completely removing the intent element of a crime from the jury’s consideration and to contradictory and irreconcilable instructions on intent. (People v. Lee (1987) 43 Cal.3d 666, 671-676 [238 Cal.Rptr. 406, 738 P.2d 752].) The error in the instant case was clearly prejudicial under this test. Indeed, had the jury been properly instructed, it is reasonably probable that the jury would have concluded that defendant did not appreciate the risk to life *1229created by his conduct although he should have done so and that he was guilty of involuntary manslaughter rather than second degree murder.
The majority conclude that defendant was not prejudiced because the jury was also instructed pursuant to CALJIC No. 8.51 (1983 rev.) (4th ed. pocket pt.) as follows: “There are many acts which are lawful but nevertheless endanger human life. If a person causes another’s death by doing such a dangerous act in an unlawful or criminally negligent manner without realizing the risk involved, he is guilty of manslaughter. If, on the contrary, he had realized the risk and acted in total disregard of the danger of life involved, malice would be implied and he could be guilty of murder.” (Maj. opn., ante, at p. 1222, fn. 2.)
The instruction, purporting to distinguish murder and manslaughter, does not eliminate or even reduce the prejudice flowing from the “wanton disregard” instruction. If anything, the instruction exacerbates the prejudice in the circumstances of the instant case. The jury is told by the instruction that if there was criminal negligence without realization of the risk involved, the crime is manslaughter. The jury may well have concluded that defendant did not realize the risk involved as pointed out above, but it might also have concluded that defendant’s conduct was more reprehensible than negligence, amounting to reckless and wanton misconduct. Viewing the instructions as a whole, the jury would erroneously determine that defendant was guilty of murder under the “wanton disregard” instruction even though he had not recognized the risk of death.
I would affirm the judgment of the Court of Appeal.
Appellant’s petition for a reharing was denied February 15, 1990.