Opinion
WALLIN, J.Mark Edwin Taylor appealed his second degree murder conviction, based on Adrian Obregon’s death arising from Taylor selling him a few dollars’ worth of phencyclidine (PCP). We originally reversed, holding the statute upon which Taylor’s felony-murder conviction was based (Health & Saf. Code, § 11379.5) could be violated in a manner not inherently dangerous to life. The Supreme Court granted review and retransferred the case here with instructions to vacate our prior opinion and reconsider the case in light of People v. Patterson (1989) 49 Cal.3d 615 [262 Cal.Rptr. 195, 778 P.2d 549]. We did so, reversed the murder conviction again and remanded the case for consideration under Patterson’s “high probability of death” standard, and affirmed in all other respects. The trial court found the standard had been met, reimposed the judgment, and Taylor appealed again. We affirm in part and reverse in part again.
The facts surrounding Obregon’s death have not changed. However, we elaborate on the expert evidence concerning the dangers of PCP because it formed the basis for the trial court’s ruling.
Adrian Obregon went to the beach with three friends. On their way, they purchased some beer. On arrival, they swam for a few minutes and drank some of the beer. Witnesses’ descriptions of the surf that day ranged from “pretty calm” to “a little big” to “real strong.” Obregon was a fairly good swimmer.
*1088Taylor and another person approached the foursome and asked if they wanted to purchase some “water,” meaning PCP. Obregon said he did, and he and one of his friends paid Taylor about $2. Taylor dipped an unfiltered cigarette in his vial of PCP. He lit the cigarette, took several puffs, and passed it around. Obregon and his friend took turns smoking the cigarette until it was gone. Obregon and two newcomers smoked another PCP-laced cigarette in this fashion.
After smoking, Obregon sat facing the ocean and listening to the radio with headphones; he was “high.” He began scooting toward the water. At the water’s edge, he turned around so his head was pointed toward the water. He allowed several waves to hit him and carry him, appearing to enjoy it. A wave broke over him and rolled him onto his stomach; he did not get up. The surf carried Obregon out to sea and neither the lifeguard nor others on the beach could find him.
Obregon’s body turned up on the beach the next morning. An autopsy revealed a subdural brain hemorrhage, consistent with an impact against wet sand. Significant amounts of PCP were present in the blood, brain, and liver. The blood-alcohol level was .058 percent. The autopsy surgeon identified the cause of death as saltwater drowning, with PCP use and the subdural hemorrhage as contributing factors.
A complaint charged Taylor with one count of murder, seven counts of sale, furnishing, or transportation of PCP, and one count of possessing PCP for sale. At trial, Dr. Steven Lerner, the prosecution expert, testified that PCP was used experimentally as a general anesthetic in the 1950’s. At the normal anesthetic dosage of .25 milligrams per kilogram of body weight, it would render the patient unconscious without medical risk. Vital signs would remain stable. It was discontinued because of side effects, such as assaultive and belligerent behavior in the recovery room.
Ninety to ninety-five percent of PCP users have a pleasant experience. Some chronic users have ingested the drug on a daily basis for years. They build up incredibly high tolerances, up to 250 milligrams per sitting, which is 10 to 15 times the general anesthetic dose for surgery.
Although not familiar with the legal standard, Dr. Lerner testified PCP is inherently dangerous to human life. Although the drug can be toxic, most people who die from the drug do not overdose, but succumb as a result of its “behavioral effects,” such as drowning or falls from heights. Users’ spatial perceptions are affected; they lose their sense of up and down, and where their limbs are. They can drown in shallow water and are unsafe drivers. The *1089drug also interferes with their ability to respond appropriately in emergency situations. Users can become violent and have been killed by police as a result. They have also attacked and killed others.
Defense expert Dr. Orm Aniline testified PCP is not a toxic substance by nature, although any substance has some toxic properties depending on how it is used and the sensitivity of the user. PCP was used as an anesthetic because doctors did not need to worry about seizures or cardiac and respiratory problems as they did with other anesthetics. It is still widely used for “headlight surgery” in remote areas because of its high margin of safety, and on animals with sensitive respiratory systems, such as elephants.
PCP is not an inherently dangerous drug; one can be given PCP without creating a substantial risk of death. Concerns about death from an overdose do not arise until an individual has ingested 1,000 milligrams, over 50 times a high street dosage and over 40 times the average anesthetic level. The odds of dying from a dose of PCP are less than 1 in 10,000, and possibly much less. It approximates the risk of death from a shot of lidocaine in the dentist’s office.
Dr. Aniline acknowledged there are potential negative side effects to PCP use. It can cause a coma, seizures, neurological problems (including difficulty with balance and movement), respiratory problems, and exacerbation of preexisting conditions (such as heart seizures and psychosis).
Montebello Police Officers1 Robert Carcano and Louis Ojeda had significant training on PCP and street experience with users. To their knowledge, although approximately 500 to 600 PCP arrests were made in their jurisdiction each year, no “PCP deaths” had ever occurred there.
The trial judge ruled that the sale of PCP is inherently dangerous to human life and allowed the murder charge to go to the jury. Taylor was found guilty on all counts and sentenced to 15 years to life on the murder count, terms being stayed on the companion drug offenses. On appeal, we reversed the murder conviction, reasoning under controlling authority that the felony-murder theory relied upon was invalid because the sale, transportation, or furnishing of PCP was not inherently dangerous to human life when “viewed in the abstract (People v. Taylor (May 9, 1989) G005345 [nonpub. opn.]; see also People v. Burroughs (1984) 35 Cal.3d 824, 829-830 [201 Cal.Rptr. 319, 678 P.2d 894]; People v. Henderson (1977) 19 Cal.3d 86, 93 [137 Cal.Rptr. 1, 560 P.2d 1180].)
The Supreme Court granted review and subsequently transferred the case here with directions to vacate our opinion and to reconsider the matter in *1090light of People v. Patterson, supra, 49 Cal.3d 615, decided four months after our opinion was filed. Patterson held that to be “ ‘inherently dangerous’ ” the underlying felony, viewed in the abstract, must carry “ ‘a high probability’ that death will result.” (Id. at pp. 622, 627.) We reconsidered the matter with the guidance of Patterson, and reversed and remanded to the trial court with instructions to determine whether sale or furnishing of PCP posed a high probability of death. (People v. Taylor (Sept. 27, 1990) G005345 [nonpub. opn.].) The trial court found it did, relying solely on the expert testimony from the trial.
I
Before analyzing the trial court’s finding that furnishing2 PCP poses a high probability of death, we must determine the correct standard of review. Neither of the parties tackles the issue, although Taylor seems to treat the issue purely as a matter of law,3 while the Attorney General seems to view it, at least partially, as an issue of fact.4 It is the former.5
“As has often been said, ‘A judgment or order of the lower court is presumed correct. All intendments and presumptions are indulged to support it on matters as to which the record is silent, and error must be affirmatively shown. This is not only a general principle of appellate practice but an ingredient of the constitutional doctrine of reversible error. [Citations.]’ [Citation.]” (Atlantic Richfield Co. v. State of California (1989) 214 Cal.App.3d 533, 538 [262 Cal.Rptr. 683].)
However, deference need not always be given to the trial court. For example, statutory interpretation is a question of law (Dean W. Knight & *1091Sons, Inc. v. State of California ex rel. Dept. of Transportation (1984) 155 Cal.App.3d 300, 305 [202 Cal.Rptr. 44]), and appellate courts review such issues independently, at least where there is no conflicting extrinsic evidence offered to aid in interpretation. (See Atlantic Richfield Co. v. State of California, supra, 214 Cal.App.3d at pp. 538, 542 [appellate court gave deference to trial court’s interpretation of statutory language where extrinsic evidence had been offered].)6
Although second degree felony murder is not expressly set forth in the Penal Code, the courts have long held that certain felonies not listed in Penal Code section 189 (supporting a first degree conviction) will support a finding of second degree murder if a defendant kills during the perpetration of one of them. (People v. Mattison (1971) 4 Cal.3d 177, 184 [93 Cal.Rptr. 185, 481 P.2d 193].) The courts have limited the bounds of second degree felony murder by limiting the qualifying felonies to those which are “inherently dangerous to human life.” (Ibid.) Thus, even if the rule lacks a statutory basis (see People v. Dillon (1983) 34 Cal.3d 441, 472, fn. 19 [194 Cal.Rptr. 390, 668 P.2d 697]; but see People v. Taylor (1980) 112 Cal.App.3d 348, 356-357 [169 Cal.Rptr. 290] [finding a statutory basis]), the courts’ interpretive process is akin to statutory interpretation.
One court has held that when the trial court receives extrinsic evidence concerning a statute’s meaning, an appellate court will uphold any reasonable construction by it. (Atlantic Richfield Co. v. State of California, supra, 214 Cal.App.3d at pp. 538, 542 [trial court’s interpretation of oil drilling term, “at the well,” was reasonable].) However, we do not agree such deference is proper.7 The Atlantic Richfield court forged its rule by stating the general rule on interpreting writings, such as contracts and deeds, contained in Witkin’s treatise on California Procedure (9 Witkin, Cal. Procedure (3d ed. 1985) Appeal, §§ 292-295, pp. 303-306), and applying it to statutory interpretation, without discussion or citation to authority.
This position is at odds with the general proposition that appellate courts independently determine the meaning of statutes. (Sutco Construction Co. v. Modesto High School Dist. (1989) 208 Cal.App.3d 1220, 1228 [256 Cal.Rptr. 671].) Witkin recognizes as much, expressly categorizing statutory interpretation solely as a matter of law, and interpretation of an instrument or writing as a question of law or mixed question of fact and law. (9 Witkin, Cal. Procedure, supra, Appeal, § 242, at p. 247.)
*1092Further, statutory interpretation does not fit within the definition of mixed questions of law and fact. “Mixed questions are those ‘in which the historical facts are admitted or established, the rule of law is undisputed, and the issue is whether the facts satisfy the [relevant legal] standard, or to put it another way, whether the rule of law as applied to the established facts is or is not violated.’ [Citation.]” (People v. Louis (1986) 42 Cal.3d 969, 984 [232 Cal.Rptr. 110, 728 P.2d 180], italics added.) Here, it is precisely the rule of law which is disputed: Viewed in the abstract, is furnishing PCP an inherently dangerous felony carrying a high probability death will result?
This is not the type of question on which appellate courts traditionally defer to the discretion of the trial court, such as change of venue, inconvenient forum, injunction and receivership, and continuances. (See generally, 9 Witkin, Cal. Procedure, supra, Appeal, § 276, at pp. 286-287 [illustrating areas of trial court discretion]; see also Hurtado v. Statewide Home Loan Co. (1985) 167 Cal.App.3d 1019, 1023, 1025 [213 Cal.Rptr. 712], disapproved on other grounds in Shamblin v. Brattain (1988) 44 Cal.3d 474, 479, fn. 4 [243 Cal.Rptr. 902, 749 P.2d 339] [giving further illustrations].) All of those matters are extremely fact intensive and vary from case to case.
The issue of “inherently dangerous felonies” cries out for a uniform rule of law. To paraphrase Cardozo, it would not do to have PCP’s “inherent dangerousness” decided different ways among different litigants. (See Cardozo, The Nature of the Judicial Process (1921) p. 33.) The only way to assure uniformity is for the appellate courts to decide the issue as a matter of law, without deference to the trial court’s ruling.
Even if we viewed the matter as a mixed question of law and fact, we would independently review the question of inherent dangerousness. In People v. Louis, supra, 42 Cal.3d 969, the Supreme Court suggested it should independently review the question of due diligence in securing a witness’s presence at trial. (Id. at p. 988.) In doing so the court engaged in an extensive analysis of the procedure for reviewing mixed questions.
“ ‘[T]here are three distinct steps in deciding a mixed fact-law question. The first step is the establishment of the “basic, primary, or historical facts: facts ‘in the sense of a recital of external events and the credibility of their narrators . . . .’” [Citations.] The second step is the selection of the applicable rule of law. The third step—and the most troublesome for standard of review purposes—is the application of law to fact or, in other words, the determination “whether the rule of law as applied to the established facts is or is not violated.” [Citation.]’ ” (42 Cal.3d at p. 985.)
“ ‘The [trial] court’s resolution of each of these inquiries is, of course, subject to appellate review. The appropriate standard of review for the first *1093two of the . . . court’s determinations—its establishment of historical facts and its selection of the relevant legal principle—has long been settled. Questions of fact are reviewed under the deferential, clearly erroneous standard. [Citation.] Questions of law are reviewed under the non-deferential, de novo standard. [Citations.]’ ” (42 Cal.3d at p. 985.)
“ ‘Structurally, appellate courts have several advantages over trial courts in deciding questions of law. First, appellate judges are freer to concentrate on legal questions because they are not encumbered, as are trial judges, by the vital, but time-consuming, process of hearing evidence. Second, the judgment of at least three members of an appellate panel is brought to bear on every case. It stands to reason that the collaborative, deliberative process of appellate courts reduces the risk of judicial error on questions of law. Thus, de novo review of questions of law, like clearly erroneous review of questions of fact, serves to minimize judicial error by assigning to the court best positioned to decide the issue the primary responsibility for doing so. . . . [f] [Further,] [u]nder the doctrine of stare decisis, appellate rulings of law become controlling precedent and, consequently, affect the rights of future litigants.’ ” (42 Cal.3d at p. 986.)
“ ‘The appropriate standard of review for a [trial] judge’s application of law to fact may be determined ... by reference to the sound principles which underlie the settled rules of appellate review just discussed. If the concerns of judicial administration—efficiency, accuracy, and precedential weight—make it more appropriate for a [trial] judge to determine whether the established facts fall within the relevant legal definition, we should subject [the] determination to deferential, clearly erroneous review. If, on the other hand, the concerns of judicial administration favor the appellate court, we should subject the [trial] judge’s finding to de novo review. Thus in each case, the pivotal question is [whether] the concerns of judicial administration favor the [trial] court or [whether] they favor the appellate court.’ ” (42 Cal.3d at pp. 986-987.)
“ ‘If. . . the question requires us to consider legal concepts in the mix of fact and law and to exercise judgment about the values that animate legal principles, then the concerns of judicial administration will favor the appellate court, and the question should be classified as one of law and reviewed de novo. [][]... [T]he concerns of judicial administration will generally favor the appellate court, justifying de novo review. This is so because usually the application of law to fact will require the consideration of legal concepts and involve the exercise of judgment about the values underlying legal principles.’ ” (42 Cal.3d at p. 987.)
By that reasoning, we should review de novo the question of inherent dangerousness. Repeatedly, the Supreme Court has expressed the *1094notion that sound policy reasons require that the second degree felony-murder rule be construed as narrowly as possible. (See, e.g., People v. Patterson, supra, 49 Cal.3d at p. 622.) The appellate courts are best equipped to fashion the boundaries of the rule and de novo review for that purpose is required.
One might wonder why, if the issue is a question of law to be independently reviewed, the Supreme Court in People v. Patterson, supra, 49 Cal.3d 615 ordered the matter remanded to the trial court, and why we did the same here. The answer is that the determination of a “high probability of death” depends, at least somewhat, on evidence as to the dangerousness of the underlying felony in the abstract, evidence which is best garnered from testimony by experts. Appellate courts have often used evidence gathered by trial courts and referees as the basis for de novo legal rulings. (See, e.g., People v. Ledesma (1987) 43 Cal.3d 171, 219 [233 Cal.Rptr. 404, 729 P.2d 839] [incompetency of counsel]; McCoy v. Hearst Corp. (1986) 42 Cal.3d 835, 842 [231 Cal.Rptr. 518, 727 P.2d 711] [defamation]; People v. Leyba (1981) 29 Cal.3d 591, 596-597 [174 Cal.Rptr. 867, 629 P.2d 961] [search and seizure]; People v. Kelly (1976) 17 Cal.3d 24, 31, 40-41 [130 Cal.Rptr. 144, 459 P.2d 1240] [reversing trial court determination that a scientific process was sufficiently reliable to yield admissible evidence].)
Here, the trial court and prosecution opted to rely upon the expert testimony presented at trial.8 The court made no determination of credibility or express findings of fact, but merely concluded there was “substantial evidence” that furnishing PCP creates a high probability of death. The only implied finding the court might have made was that Dr. Lerner’s testimony that PCP is inherently dangerous was correct, and Dr. Aniline’s testimony to the contrary was not.9
We need not give deference to such a finding because it was not factual. It dealt with the ultimate issue of dangerousness. Although experts are properly allowed to testify on such issues (Evid. Code, § 805), here it is the very issue which this court must decide de novo as a matter of law. Thus, the implied finding, if it was such, is not the type which requires deference. (People *1095v. Louis, supra, 42 Cal.3d at pp. 986-987.)10 Any other conclusion would divest appellate courts of the power to interpret the law and place it in the hands of expert witnesses. Using the record provided, we will answer the legal question posed.
II
None of the cases since People v. Patterson, supra, 49 Cal.3d 615 have decided whether furnishing PCP (or any drug) carries a high probability of death. (See People v. Coleman (1992) 5 Cal.App.4th 646 [7 Cal.Rptr.2d 40] [kidnapping for robbery; high probability of death]; People v. Morse (1992) 2 Cal.App.4th 620, 646 [3 Cal.Rptr.2d 343] [reckless or malicious possession of a bomb; high probability]; People v. Lee (1991) 234 Cal.App.3d 1214, 1228-1229 [286 Cal.Rptr. 117] [child endangerment; not inherently dangerous]; People v. Pearch (1991) 229 Cal.App.3d 1282, 1299 [280 Cal.Rptr. 584] [kidnapping; high probability]; People v. Ordonez (1991) 226 Cal.App.3d 1207, 1228 [277 Cal.Rptr. 382] [kidnap for ransom; high probability].) However, a number of pre-Patterson cases dealt with the issue, often with little analysis.
In People v. Poindexter (1958) 51 Cal.2d 142 [330 P.2d 763], a case credited with presaging the announcement of the second degree felony-murder rule (People v. Satchell (1971) 6 Cal.3d 28, 35-36 [98 Cal.Rptr. 33, 489 P.2d 1361, 50 A.L.R.3d 383]), the Supreme Court held furnishing heroin to a minor who died as a result supported a second degree murder conviction. (Id. at p. 149.) The court reasoned, “[There] was uncontroverted testimony that [the minor] died from narcotics poisoning, and that taking a shot of heroin was an act dangerous to human life.” (Ibid.)
Ureta v. Superior Court (1962) 199 Cal.App.2d 672 [18 Cal.Rptr. 873] relied on Poindexter, and held that death from injected morphine poisoning establishes the corpus delicti for murder because one might reasonably infer the morphine was illegally furnished. (Id. at p. 676.) Without analysis or citation to authority, the court stated, “It makes no difference whether deceased or another actually injected the narcotic. The person who furnished him the narcotic is liable even though deceased did the actual administering of it.” (Ibid.)
People v. Mayfield (1964) 225 Cal.App.2d 263 [37 Cal.Rptr. 340] distinguished Poindexter and Ureta, and held the defendant, who aided and *1096abetted, by his copurchase, a self-administration of heroin resulting in death, could not be convicted of felony murder. (Id. at pp. 266-268.) However, the court apparently based its holding on its conclusion that the copurchasers would be guilty, at most, of misdemeanor use of drugs. (Ibid.; see also People v. Cline (1969) 270 Cal.App.2d 328, 332 [75 Cal.Rptr. 459, 32 A.L.R.3d 582] [interpreting the holding in Mayfield].)
In People v. Williams (1965) 63 Cal.2d 452 [47 Cal.Rptr. 7, 406 P.2d 647] the Supreme Court held that conspiracy to possess methedrine was not inherently dangerous. (Id. at p. 458.) In doing so, the high court held for the first time that the felony in question must be viewed in the abstract. (Id. at p. 458, fn. 5.) Therefore, it ignored the facts of the stabbing in making its determination on inherent dangerousness. (Id. at pp. 454-455, 458.)
The court in People v. Cline, supra, 270 Cal.App.2d 328 relied on the bald statement in Ureta v. Superior Court, supra, 199 Cal.App.2d 672 that no distinction should be made between administering and furnishing a drug. (People v. Cline, supra, 270 Cal.App.2d at p. 332.) The court held furnishing phenobarbital was an inherently dangerous felony, ignoring the then-recent “felony in the abstract” rule of People v. Williams, supra, 63 Cal.2d 452. (People v. Cline, supra, 270 Cal.App.2d at p. 333.)
The Cline court reasoned as follows: “The trial judge found that defendant’s act in furnishing a restricted dangerous drug to the deceased in violation of law was inherently dangerous to human life. His finding in this respect is amply supported by the evidence. It was the uncontroverted testimony of the pathologist that the consumption of phenobarbital in unknown strength was dangerous to human life. There was clear evidence that within a period of one-half hour this drug was consumed in considerable quantity by [the victim] in defendant’s presence and with his knowledge. Even defendant admitted that the deceased consumed 15 of these pills within one-half hour.” (270 Cal.App.2d at p. 333.) The court also looked to legislative intent to declare the drug “dangerous to human life.” (Ibid.) It relied upon “all of [these] reasons” to find second degree felony murder was supported by furnishing a restricted dangerous drug without a prescription. (Id. at pp. 333-334.)
Despite the Cline court’s reliance upon the facts of the case and legislative intent to reach its result, the court in People v. Taylor (1970) 11 Cal.App.3d 57 [89 Cal.Rptr. 697] cited Cline as authority for the proposition that furnishing drugs is inherently dangerous in the abstract. (Id. at p. 59.) The Taylor court noted that Cline “cited [People v. Phillips (1966) 64 Cal.2d 574, 582 [51 Cal.Rptr. 225, 414 P.2d 353] [reiterating the ‘abstract’ rule] and the *1097Supreme Court denied a hearing.” (People v. Taylor, supra, 11 Cal.App.3d at p. 59.)
Without protracted discussion or dispute, People v. Mattison, supra, 4 Cal.3d 177, 185-186 and People v. Calzada (1970) 13 Cal.App.3d 603, 605-606 [91 Cal.Rptr. 612] added administering methyl alcohol and driving under the influence of narcotics, respectively, to the list of inherently dangerous drug-related felonies. Then came the second People v. Taylor, supra, 112 Cal.App.3d 348. Relying without analysis on Mattison,11 Poindexter, Taylor, Cline, and Ureta, the court rejected the argument that administering heroin is necessary for application of the felony-murder rule, and held furnishing was sufficient. (Id. at p. 358.) Finally, in People v. Edwards (1985) 39 Cal.3d 107 [216 Cal.Rptr. 397, 702 P.2d 555], the Supreme Court alluded to the second Taylor and Cline, but based its reversal on the failure to instruct that the defendant might escape culpability if he merely aided or abetted a purchase of heroin, relying on People v. Mayfield, supra, 225 Cal.App.2d 263. (People v. Edwards, supra, 39 Cal.3d at pp. 113-116.)
Two notable observations can be made about these “drug related” second degree felony-murder cases. The first is obvious: By predating People v. Patterson, supra, 49 Cal.3d 615, none of the cases used the Patterson's “high probability of death” standard. Second, although People v. Taylor, supra, 11 Cal.App.3d 57 paid lip service to the need to look at the felony in the abstract (id. at p. 59), none of the cases actually performed an analysis of drug furnishing from that vantage point.12 As we shall discuss, these omissions yielded an improper conclusion.
While the “drug related” second degree felony-murder cases were evolving, other cases adhered to the rule that inherent dangerousness must be viewed in the abstract. (People v. Patterson, supra, 49 Cal.3d at p. 625; People v. Burroughs, supra, 35 Cal.3d at pp. 830-832 [practicing medicine without a license; not inherently dangerous]; People v. Henderson, supra, 19 Cal.3d at pp. 93-96 [false imprisonment; not inherently dangerous]; People v. Satchell, supra, 6 Cal.3d at pp. 36-41 [firearm possession by felon; not inherently dangerous]; People v. Lopez (1971) 6 Cal.3d 45, 51-52 [98 Cal.Rptr. 44, 489 P.2d 1372] [escape from penal facility; not inherently dangerous]; People v. Nichols (1970) 3 Cal.3d 150, 163 [89 Cal.Rptr. 721, 474 P.2d 673] [auto arson; inherently dangerous]; People v. Phillips (1966) *109864 Cal.2d 574, 582-584 [51 Cal. Rptr 225, 414 P.2d 353] [grand theft; not inherently dangerous].) Of these cases, four give particular insight into application of the “abstract” rule.
Patterson itself expressly distinguished furnishing from administering for analyzing “inherent dangerousness.” (People v. Patterson, supra, 49 Cal.3d at pp. 624-625.) In People v. Burroughs, supra, 35 Cal.3d 824, the Supreme Court pointed out that the statute making the unlicensed practice of medicine a felony included cirbumstances where there was a risk of great bodily harm, serious mental or physical illness, or death. (Id. at p. 830.) Although the other conditions could involve a great risk of death under certain circumstances, the court held the felony would not support second degree murder because in the abstract, the statute could be violated without such risk. (Id. at pp. 830-831, 833.)
Similarly, in People v. Henderson, supra, 19 Cal.3d 86, the court reasoned that false imprisonment by fraud and deceit carried no great risk of death, and declined to focus on violent means of violating the statute. (Id. at pp. 93-95.) And, in People v. Lopez, supra, 6 Cal.3d 45, the court refused to consider the possibility of violent resistance in determining whether escape is inherently dangerous. (Id. at pp. 51-52.)
We apply this wisdom to Taylor’s crime. Pursuant to the mandate of People v. Patterson, supra, 49 Cal.3d 615, we look only to the offense he committed within the purview of Health and Safety Code section 11379.5, sale or furnishing of PCP (at p. 625), and ask whether it inherently carries a “ ‘high probability’ ” of death (id. at p. 627). However, to answer the question we must ask what the term means.13
Like the trial court, we do not understand the term “high probability” to mean greater than 50 percent. Very few activities in life, even noncriminal conduct, carry such a risk. And, for two reasons, we eschew relying solely on statistical comparisons of death rates from furnishing PCP and those involving furnishing other drugs or noncriminal activities such as driving, boating, and hang gliding.
*1099First, we have been presented with no such statistical analysis. Neither Dr. Lemer nor Dr. Aniline used such an approach.14 Second, courts have hesitated to turn the law into a pure statistical analysis. (See, e.g., People v. Collins (1968) 68 Cal.2d 319, 320, 327, 329-332 [66 Cal.Rptr. 497, 438 P.2d 33, 36 A.L.R.3d 1176].)15
However, a “high probability” is apparently greater than a “ ‘substantial risk.’ ” (People v. Patterson, supra, 49 Cal.3d at pp. 628-629 (conc, and dis. opn. of Lucas, C. J.) Because these terms lack analytical precision, we look to implied malice cases for guidance. Implied malice, which People v. Watson (1981) 30 Cal.3d 290, 300 [179 Cal.Rptr. 43, 637 P.2d 279] instructs must involve a high probability of death, has been found where the defendant: shot at a cup the victim was holding in front of her face (People v. Thomas (1953) 41 Cal.2d 470, 480 [261 P.2d 1]); initiated a gun battle (People v. Washington (1965) 62 Cal.2d 777, 782 [44 Cal.Rptr. 442, 402 P.2d 130]); used a robbery victim as a shield (Pizano v. Superior Court (1978) 21 Cal.3d 128, 136 [145 Cal.Rptr. 524, 577 P.2d 659]); and drove drunk at high speeds through city streets (People v. Watson, supra, 30 Cal.3d at pp. 300-301.) Thus, imminent deadly consequences must be inherent in the act.
Viewed in this context, furnishing or selling PCP does not carry a high probability of death in the abstract. Merely conveying the drug to another does not require that it be consumed; completion of the crime does not depend upon consumption.16 And, unlike a live explosive (see People v. Morse, supra, 2 Cal.App.4th 620), mere possession of PCP is not inherently dangerous.
*1100Even when consumed, PCP does not carry a high probability of death. It is almost impossible to overdose on it. Although Dr. Lerner related a great number of particular instances where ingestion of PCP led to tragic consequences, he did not suggest that such outcomes were inevitably imminent whenever any amount of PCP is ingested. Indeed, his testimony suggests the vast majority of users, 90 to 95 percent, do not suffer any untoward effects on any given occasion. The chronic users taking heavy doses of the drug are the ones who most frequently meet untimely demises.17 Furnishing or selling PCP, when viewed in the abstract, does not carry with it an inherently high probability of death.
By saying this, we do not condone the sale or use of illegal drugs in any amount. Some risk of death is always present. As usage continues, the probability of adverse consequences rises. And these consequences are not always death. Drug and alcohol abuse ruins untold lives of users and their loved ones. Relationships and job performance suffer from an activity that has no social utility. These are but a few of the reasons for drug laws, education concerning the danger of drug use, and other social measures aimed at ameliorating this serious problem.18
However, we decline to act as a superlegislature and enact the crime of murder for those rare and ofttimes strange instances when death results from *1101furnishing PCP.19 Although we recognize the viability of the felony-murder rule, we adhere to the steady commitment of our courts to avoid “ ‘ “extension [of the rule] beyond its required application.” ’ ” (People v. Burroughs, supra, 35 Cal.3d at p. 829.)
The judgment is reversed as to the conviction for second degree murder. The superior court is directed to dismiss the murder count. Within 30 days after the opinion becomes final, the district attorney may amend the information to include a count of involuntary manslaughter. If he does not do so, the total sentence is modified to five years. In all other respects, the judgment is affirmed.20
Sonenshine, J., concurred.
Obregon was from Montebello.
Although the statute encompasses importation, transportation, furnishing, administering, sale and giving away of PCP (Health & Saf. Code, § 11379.5, subd. (a)) and the evidence here supported a sale as well as furnishing, we will use the latter term to describe Taylor’s conduct and in discussing similar drug statutes unless the context requires otherwise.
Taylor argues, “The trial court’s determination . . . was erroneous on its face,” and, “[T]he trial court erred in charging the jury on second degree felony murder . . . .”
He concludes, “[The] evidence supported the trial court’s ruling.”
The dissent asserts that Government Code section 68081 mandates an opportunity for the parties to brief the question of the appropriate review standard. Hardly. The statute only requires such an opportunity when the decision is “based upon an issue which was not proposed or briefed by any party . . . .” (Gov. Code, § 68081, italics added.) The decision here is not based upon the standard of review; it is based upon the law concerning inherently dangerous felonies. The standard of review merely sets the procedural guidelines for making the decision. Further, the standard of review question is present in every case, although the parties often ignore it. The purpose behind section 68081 is to prevent decisions based on issues on which the parties have had no opportunity for input. (Schneider v. Kaiser Foundation Hospitals (1989) 215 Cal.App.3d 1311, 1315, fn. 2 [264 Cal.Rptr. 227].) They certainly had the opportunity here.
However, as we discuss below, we disagree that the existence of extrinsic evidence alters the standard of review.
We distinguished Atlantic Richfield in Orange County Employees Assn. v. County of Orange (1991) 234 Cal.App.3d 833, 840-841, footnote 5 [285 Cal.Rptr. 795]. Thus, we had no cause to consider the correctness of its rule.
Taylor asked the trial court to take judicial notice of statistics on PCP deaths and other types of deaths for comparative purposes. The trial court declined, opting to rely on the trial record, and the district attorney did not object to that limitation or seek to produce additional evidence.
Thus, contrary to the dissent’s assertion, “the matter we are faced with here” is not fact intensive. (Dis. opn., post, at p. 1103) The facts of Taylor’s case are irrelevant to deciding whether furnishing PCP is inherently dangerous because that question is decided in the abstract. (People v. Patterson, supra, 49 Cal.3d at p. 622.) We cannot imagine an issue less “fact intensive,” as that term has been traditionally and appropriately used.
Another reason for refusing to give deference to any implied finding regarding the experts’ conclusions is they did not deal with the precise question presented here. Both testified on whether PCP is inherently dangerous without reference to the Patterson “high risk of death” standard, and Dr. Lerner admitted he was unfamiliar with the legal standard for inherent dangerousness.
The court’s citation to People v. Mattison, supra, 4 Cal.3d 177 is surprising. The Mattison court discussed the first People v. Taylor, supra, 11 Cal.App.3d 57, but did so in an unrelated context. (People v. Mattison, supra, 4 Cal.3d at pp. 184-186.)
Perhaps the Attorney General recognized these facts. He does not cite any pre-Patterson cases in support of his argument on the issue of inherent dangerousness.
We do not take Chief Justice Lucas’s statement that the standard could virtually never be met in a drug-furnishing case as an invitation to throw up our hands and abandon any try at analyzing the issue. (People v. Patterson, supra, 49 Cal.3d at p. 628 (conc, and dis. opn. of Lucas, C. J.).)
Dr. Aniline did say the chances of dying from an overdose were less than 1 in 10,000. He did not opine about the odds of dying from other causes while high on PCP. Taylor asked us to take judicial notice of the statistics provided in his opening brief concerning death rates for PCP and various travel and recreational activities, and the Attorney General voiced no opposition. But Taylor provided no scientific analysis, of which we also might take judicial notice, concerning the statistical import of those numbers.
We also opt not to consider the testimony of the Montebello police officers. It is historically interesting, but we do not give it weight in our analysis because it does not purport to involve a statistically reliable or significant sample. Although the outcome should not be based solely on statistics, neither should it be based upon isolated incidents which lack any statistical context.
The dissent urges “plain, ordinary common sense” tells us the drug will always be consumed. (Dis. opn., post, at p. 1105.) One need only read the newspapers about the tons of contraband seized by law enforcement every year to know that is not true. Nonetheless, we do not dispute that furnished drugs are usually consumed. However, the dissent seems to miss the thrust of viewing the underlying felony in the abstract. We reiterate: The question is not what happened in this case or what usually happens. The question is whether the statute can be violated in a nondangerous (or less than high-probability-of-death) fashion. (See, e.g., People v. Henderson, supra, 19 Cal.3d at pp. 93-95 [false imprisonment can be committed by fraud and deceit].) The dissent turns this rule on its head and hypothesizes a worst case scenario of toxic dermal osmosis. (Dis. opn., post, at pp. 1105-1106.)
Dr. Lemer’s testimony was as follows: “A. Probably 90 to 95 percent of people that use PCP have pleasurable experiences. However, in the chronic users we’ve studied, if you follow them longitudinally over a long enough period of time, all of them uniformly talk about what they would consider problems with the drug, where either they have—they’ve been bizarre, unusual or violent in their behavior toward others. ... [1] Q. Are you saying then it’s okay to take one or two puffs off a PCP cigarette as long as you don’t use it again? [f] A. No, not at all. Because depending on whether you take those one or two puffs, you could either put you own life in jeopardy, if you’re at the ocean or the beach, near a body of water. If you take one or two puffs and you happen to be in an automobile, you’re clearly putting everyone around you, their life in jeopardy by smoking because your driving abilities are compromised. [1] Q- That is whether you’re a chronic user or not? [1] A. Makes no difference.”
This testimony shows that PCP can be dangerous in particular circumstances, and no prudent person should ingest it. But Dr. Lerner’s opinion is far from saying all PCP use in the abstract inevitably carries a high probability of death. Although he pointed out that potential danger exists when any amount is consumed, his scenarios ending in death involved chronic users.
The dissent, seizing on the prosecution expert’s use of the term “pleasurable” to describe user’s experiences, implies we are drug use apologists, seeking to dilute the risk of drug use by touting the good times it purportedly engenders. We do not. Recreational drug use has no social utility. Our society has paid a high price for the experimentation of prior years. We have expressed that concept in the preceding paragraph and stress it here. The dissent’s innuendo to the contrary is unfortunate and diverts us from our task to dispassionately apply the law. Our colleague’s position on this case is puzzling. When the case was before the court in 1989 and the trial court had found furnishing PCP was inherently dangerous using the same evidence it later used to rule after the latest remand, our colleague “concur[red] in the judgment under the compulsion of People v. Burroughs . . . .” (People v. Taylor, supra, (May 9, 1989) G005345 (conc. opn. of Moore, J.).) Now that the case is before us with the same *1101factual basis but a more stringent requirement for inherent dangerousness (People v. Patterson, supra, 49 Cal.3d at p. 628 (conc, and dis. opn. of Lucas, C. J.)), he dissents.
The present case is a good example of the rather fortuitous circumstances under which a death may occur. Although Dr. Lemer testified that PCP users are attracted to water, he related no other instances in which a user had been swept to a death at sea. Another example would be if we were to find that furnishing codeine without a prescription was inherently dangerous. (See Health & Saf. Code, §§ 11054, subd. (c)(4) and (5), and 11352, subd. (a).) A person giving a friend with a severe headache Tylenol-3 resulting in a deadly allergic reaction could be sent to prison for 15 years to life. (Pen. Code, § 190, subd. (a).)
Taylor raises no issues relating to his other convictions. Our reversal on the murder count renders his other contentions moot.