People v. Bonilla

Titone, J. P.

(dissenting in part and concurring in part). In the absence of a “brain death” standard, which I, too, am unwilling to adopt as a matter of judicial fiat, the evidence is legally insufficient to establish that the defendant caused the decedent’s death. Evidentiary insufficiency aside, the trial court erred in refusing to define for the jury the point at which death occurs. Inasmuch as these questions have been preserved for appellate review, the judgment should be modified by reducing the conviction for manslaughter in the first degree to assault in the second degree. Alternatively, I would order a new trial.

I

We all agreé that it is inappropriate to judicially adopt a brain death standard. My reasons rest upon both moral considerations and the need for judicial restraint in an area peculiarly statutory.

First, judicial enunciation of such a standard is contrary to a long-standing tradition in this State, only recently reaffirmed (see Murphy v American Home Prods. Corp., 58 NY2d 293, 301-302 ), that an appellate court should not, within the context of adversary litigation, usurp the legislative prerogative to change a settled rule of law when the standard sought to be substituted is itself a changing flexible one such as brain death (see Dolphin Lane Assoc. v Town of Southampton, 37 NY2d 292, 296 [“If a change is to be made in the procedures for locating shore-side boundary lines to conform more precisely to hydrographie data, in our view, such innovation should be left to the Legislature.”]). 1 Since medicine is an evolving science, what con *423 stitutes an appropriate definition of death might require some elasticity. 2

The simple observation that life is the primary value of all living organisms is so basic that it is often unstated. Yet, it is the foundation of our constitutional guarantee of life and serves as a recognition that society retains the ultimate right to decide what criteria should be used to declare that life is extinguished. Before a new definition of death is adopted, a thorough exploration of the standard and its effects is necessary (see Murphy v American Home Prods. Corp., 58 NY2d 293, 301-302 , supra). The task of such an articulation should be deferred to the Legislature. “Unlike the Legislature, the courts are neither equipped nor empowered to prescribe substantive or procedural rules for all, most, or even the more common contingencies. Our role, especially in matters as sensitive as these, is limited to resolving the issues raised by facts presented in particular cases” (Matter of Storar, 52 NY2d 363, 370 ).

We cannot defer to the medical profession on this question. Its expertise lies in providing health care and not in defining death on a moral or philosophical basis (cf. Matter of Storar, supra, pp 376-378). Recognition of this limitation is implicit in the protocol on brain death issued by the Harvard Medical School in 1968 (205 J Amer Med Assn 337), which essentially urged that benefits would come about by society’s acceptance of the brain death standard (see Ufford, Brain Death/Termination of Heroic Efforts to Save Life — Who Decides?, 19 Washburn LJ 225, 234).

It cannot be denied that death has a philosophical component. The process of death consists of two levels. One relates to the death of the body which is a physical phenomenon occurring in a series of measurable events. The other is associated with the passing of the person, which is *424 a nonphysical process, poorly defined and largely unmeasurable. This second element is connected with moral and philosophical concepts not within the concern of the medical profession. (See Showalter, Determining Death: The Legal and Theological Aspects of Brain-Related Criteria, 27 Cath Lawyer 112, 114-116 and authorities cited therein.)

Brain death as it is defined in the record before us refers only to the death of the body. While I believe that the common law is adaptable enough to expand and include a new physical definition of death, it is a poor vehicle for sorting the different views of a noncohesive society or to assess the value judgments implicit in the acceptance of such a standard (cf. Murphy v American Home Prods. Corp., supra; Matter of Storar, supra).

I must stress that I do not deny the biological fact of brain death. Actually, a brain death standard is implicit in the common-law notion that death occurs when the heartbeat and circulation ceases (O’Hara, Medical-Legal Agreement on Brain Death: An Assessment of the Uniform Determination of Death Act, 8 J of Contemp L 97, 99). Death has always been defined in terms of its causes and for all practical purposes death is never declared until brain death has occurred. When circulation and respiration cease, the time of death is not fixed until after the brain has been starved of oxygen long enough to die completely, about five minutes (Biorck, When is Death?, 1968 Wis L Rev 484, 493; Walker, Cerebral Death, p 41).

In contrast, under the brain death standard the process of brain death occurs over a much longer time. The usual testing interval is 24 hours and it is possible, using the criteria advanced by the medical profession, to declare brain death while portions of the brain are still alive but in the process of dying (see O’Hara, op. cit., pp 99-100).

As a society we are acclimated to thinking that the transition from person to corpse occurs in a moment (see Evans v People, 49 NY 86 ). The brain death standard requires an adjustment in our viewpoint. No one is prepared to bury a “corpse” whose heart is still beating. Yet, acceptance of a brain death standard may not make such a course unthinkable.

*425 Having watched a significant number of people die slowly from various causes, I personally recoil from a standard the purpose of which is to value a dying person as the sum of his spare parts. Nevertheless, I cannot deny that there are social and moral costs in mistaking and confusing a person with his corpse. If the capacity to think, respond and regulate bodily functions is irretrievably lost, it is not appropriate to continue treating the corpse as a person and thereby deprive other living persons of needed organs which can be salvaged from the corpse. (Compare Brain Death, 148 America 234, with Paris & Cranford, Brain Death, Profile and Catholic Confusion, 147 America 345.)

Again, the words “irretrievably lost” must be stressed, because in none of my reading has the margin for error under the brain death standard been discussed. Nonetheless, there have been macabre mistakes. Do the mistakes relate to the level of competence of the declaring physician or to the state of the art? Can protocols which were developed by whole medical schools be modified, as occurred in this case, and still be effective? I have no answers to these inquiries and the record provides none. They are dilemmas to be resolved by the Legislature.

Another aspect for the Legislature’s attention is the medical profession itself. Utilizing the brain death standard, many members believe, will result in criminal and civil liability for physicians (Paris & McCormick, Living — Will Legislation, Reconsidered, 145 America 86, 88, quoting Prof Robert A. Burt of Yale Univ). Will the public’s fear of falling victim to what it perceives as the profession’s arrogance, avarice and ambition, increase when it recognizes that the criteria for declaring brain death, unlike common-law death, is peculiarly within the profession’s expertise and control? Will it be alleviated by requiring that the declaration be made by more than one physician?

Finally, the Legislature must consider the impact of brain death on traditional concepts of criminal law. For example, in this case the medical examiner could not testify to the time of death. He was dependent on the hospital’s timing of the declaration because the physiological breakdown of the brain is very slow. Nevertheless the *426 timing of death is a critical factor in presenting and refuting an intervention. An additional aspect of the causation problem is that the subsequent organ harvesting operation has the potential to kill from shock and pain, but is unrelated to the victim’s treatment for the wounds inflicted by a defendant. Perhaps, a penal statute fixing the same penalty for criminal acts causing brain death or a vegetative state as those that cause common-law death, may prevent causation problems without the necessity of adopting the brain death standard (see discussion, infra, p 435, n 7, and Capron, The Purpose of Death: A Reply to Professor Dworkin, 48 Ind LJ 640 [recommending such an approach]).

Another area for concern derives from the recognition that brain death is a declaration based on interpretations of delicate variables and that death could be seen as an opinion rather than as fact (cf. Kovaler & Vermel, The Legal Definition of Death, World Health, Nov., 1982, p 5). It is for the Legislature to decide whether such uncertainty should be interjected in the criminal sphere or whether it should bring “some order and design [to] an area which is so very important to society and which is already replete with complexity and confusion”. (Robinson, Determination of Death Legislation, 27 Cath Lawyer 246, 253.) Consequently, I agree that the court properly declines the invitation to judicially adopt a new standard to replace the common-law definition of death.

II

Defense counsel’s application for a trial order of dismissal, made at the close of the People’s case, was posited on the grounds that there was an “intervening factor [,] [t]he intervening factor being the removal of the deceased’s kidneys and spleen for transplantation” and that the People failed to prove “that the specific act of the defendant was the intentional and the direct and immediate cause of death of the deceased”. By that motion, the issues now before us were “brought to the attention of the Trial Judge in ways that pinpointed the legal question”, thereby preserving them for appellate review (People v Cobos, 57 NY2d 798, 800 ; see, also, People v Cona, 49 NY2d 26, 37, n 3 ).

*427 At the onset, I would note that the People have changed tunes on appeal. At trial, the People’s initial theory hinged on a finding of brain death. They found the definition of common-law death unacceptable. Now the People claim that common-law death is the applicable standard and that brain death is a nonissue which they mention only in passing because the defendant has raised it. As the Court of Appeals has made clear, preservation requirements apply to prosecutor and defendant alike and the People’s failure to preserve will be binding upon them on a defendant’s appeal (People v Bell, 48 NY2d 913, 915 ).

Prior to trial, the court informed the attorneys that it would charge brain death. The defendant had no objection. There is no indication in the record, however, that either the court or the attorneys were of the view that because brain death was to be charged common-law death could not be. Common-law death was in fact the culmination of the defendant’s argument. He asserted that brain death had not been proven and that the organ harvesting operation, which was unrelated to the victim’s treatment, had caused the common-law death of the victim. The defendant maintained this position consistently throughout the trial.

Although the defendant used a faulty brain death standard as the premise of his argument, he was not dependent on the concept. If the court had initially taken the position that brain death was not a permissible standard, the defendant’s major argument would still have been that the transplant operation caused the common-law death of the victim. Thus, brain death and common-law death were not mutually exclusive on the facts of this case. The defendant merely employed both definitions to separate the two stages of his intervening cause defense.

At the end of the trial, the court and the attorneys discussed the instructions to be given to the jury. During the discussion on causation, the defendant asked the court to define the organ harvesting operation as a secondary agency unrelated to the victim’s treatment, because if the jury agreed with the prosecutor “they will never get to the kidney transplant — that he is dead, and that’s it, and we caused it”.

*428 Immediately after this request, the defendant asked the court to charge a definition of death, specifically requesting a definition of brain death. The court refused. The defendant then attempted to explain how brain death related to a cessation of respiration or common-law death. But it should not be assumed that the defendant abandoned his two-step analysis of causation by first requesting a charge on brain death. The court’s reply indicated a view which would preclude a subsequent request for a charge on common-law death:

“the court: We have no law in New York that defines death. What you are asking me to do is decide this as a matter of law when there is no law to guide me.

“mr. feldman [Assistant District Attorney]: What are you going to —

“the court: I am not going to.

“mr. feldman: You are just going to use the term death?

“the court: Yes.

“mr. moser [defendant’s attorney]: What I am actually asking the Court is that I am submitting to you that it is a question of law. Absent the law, you have got to make the law. We have to have a starting point. And if a jurist is not going to start from someplace, we are never going to get it. I don’t think it is a point where you are actually usurping the Legislature in this state. I think what you have to do is to say and define death so that the jury can determine as a question of fact when death did occur. In that way, if they find that everything was properly done and accept the testimony of my expert or, in reverse, if they accept the testimony of the People’s witnesses, then they can determine that death occurred at the time it was pronounced. If they accept my position, then they have to determine that death followed the

“the court: I think it’s a question for them to determine.

“mr. moser: Yes, when death occurred is to be determined. How can they possibly determine when something occurred if they don’t know what that something is?” (Emphasis supplied.)

*429 It is interesting to note that the court did not limit its ruling to brain death and that the defendant did not limit his argument and the use of the term death to brain death.

The court then asked the prosecutor if he too were requesting a charge on death since the court felt that if the prosecutor joined in the request there would be no problem with instructing the jury on a definition of death. Various definitions were explored. The discussion, however, did more to highlight the differences in the understanding of the attorneys as to brain death than to resolve the problem of the charge. The court then read one final statutory definition to the attorneys: “A person shall be medically and legally dead if (a) in the opinion of a physician duly authorized to practice medicine in this state, based on ordinary standards of medical practice, there is the absence of spontaneous respiratory and spontaneous cardiac functions and because of the disease or condition which directly or indirectly causes these functions to cease, or because of the passage of time since these functions ceased, attempts at resuscitation would not, in the opinion of such physician, be successful in restoring spontaneous life sustaining functions, and in such event death shall have deemed to have occurred at the time these functions ceased. Or, in the opinion of a consulting physician who shall be duly licensed and a specialist in the field of Neurology, Neuro-surgery or Neuro-encephalography, when based on ordinary standards of medical practice, there is an absence of spontaneous brain function and spontaneous respiratory functions, and in the opinion of the attending physician, based on the ordinary standards of medical practice and considering the absence of the aforesaid spontaneous brain functions and spontaneous respiratory functions and the patient’s medical records further attempts at resuscitation or continued support of maintenance would not be successful in restoring such spontaneous function, in such event death shall have been deemed to have occurred at the time when these conditions first coincided.”

This incorporated both common-law and brain death definitions. Yet, with respect to brain death, it must be noted that lack of spontaneous respiration was added to *430 this particular formulation. This additional element created total confusion in the discussion that ensued.

As the following colloquy demonstrates, the trial court was under a continuing misimpression that it was being asked by the defendant to choose between brain death and common-law definitions rather than to.separate the definitions as per the factual stages of the causation defense. The prosecutor objected twice to the inclusion of the definition of common-law death. Defense counsel agreed to the common-law definition of death as contained in the statute read by the court. He mistakenly understood the prosecutor’s position, however, as an objection to the lack of respiration being included as an element of the definition of brain death rather than as an objection to common-law death.

“the court: * * * Now, you see, there [are] two different things. You’re asking me to pick one.

“mr. feldman: The first one creates a problem.

“mr. moser: You know what that definition does?

“the court: You don’t , want me to define death. You want me to define death the way you want it.

“mr. moser: No, I will accept it for the respiration end of it. [Common law death]. 3 This last thing incorporates every possible definition. [Brain death with lack of respiration.] “mr. feldman: The first one is unacceptable.

“mr. moser: Including the additional definition, and that’s what we just don’t want. It would be great for me because his heart was working. His heart was pumping * * *

“the court: My decision at this point is not to define death. I have read you five different definitions from God knows how many states. We don’t have a definition here, and I am not going to select one or arbitrarily decide which should be used. I don’t even think it’s a responsibility of mine to do as a Judge.” (Emphasis supplied.)

The next day defense counsel went to a fall-back position made necessary by the court’s refusal to define death *431 pursuant to successive stages of the causation defense. He asked that two certified questions, designed to insure that the jury thought in accordance with the structure of his defense, be presented to the jury. The first question required the jury to define death. The second asked that they determine when it occurred. The court declined. At the conclusion of the charge the defendant renewed his requests.

In the context of the court’s attempt to settle the question of whether brain death should be charged, an instruction on common-law death was suggested by defense counsel. The distinction offered by counsel went unnoticed by the court which held to its earlier decision of refusing to charge any definition of death.

Given the confusion engendered by the court’s misunderstanding of the defense strategy and the court’s announced determination that no charge on the definition of death would be given, it would have been futile for defense counsel to further press his position (Kalisch-Jarcho, Inc. v City of New York, 58 NY2d 377, 383 ). Defense counsel maintained a consistent defense strategy throughout the trial and explained the need for a charge on the definition of death. He requested both a charge on brain death and common-law death. A “mere emphasis of one prong of attack over another or a shift in theory on appeal, will not constitute a failure to preserve” (People v De Bour, 40 NY2d 210, 215 ) and will not constitute a waiver (People v Le Mieux, 51 NY2d 981, 982-983 ). In face of the court’s refusal to charge any definition, defendant “without success * * * expressly or impliedly sought or requested a particular ruling or instruction” and thus “is deemed to have thereby protested the court’s ultimate disposition of the matter * * * sufficiently to raise a question of law with respect to such disposition * * * regardless of whether any actual protest thereto was registered” (CPL 470.05, subd 2; see People v Cobos, 57 NY2d 798, 800 , supra; People v Le Mieux, supra).

Ill

The majority opinions resolve the troublesome issues presented by holding that the People have proven beyond a reasonable doubt that the defendant’s conduct in inflicting *432 the head wound on Miranda was the cause of the latter’s death even under its common-law definition. They reason that the infliction of such wound prevents a finding that the victim’s death was solely attributable to the purported secondary agency of misdiagnosed brain death and the subsequent operation for organ harvesting. Proof of brain death is irrelevant since “but for” that wound the victim would not have been in a position to die from the operation.

The difficulty with this approach is that it blurs the distinction between an accused’s culpable act and intervening causes. “The basic theory of causation is that the culpable act or omission must be a causa sine qua non (inevitable or necessary cause)” (Clark & Marshall, Crimes [7th ed], § 4.01, p 210). Since the concept of “ ‘causing death’ normally involves the notion of shortening life and not merely determining the manner of dying” (Hart & Honor, Causation in the Law, p 220), “one who hastens the victim’s death is a cause of his death” (LaFave & Scott, Criminal Law, § 35, p 250). So, if the defendant has inflicted a wound which would prove fatal and a third party comes along while the victim has but hours to live and kills him instantly, the third-party’s act substantially hastening death constitutes the cause of death and the defendant cannot be convicted of homicide (State v Scates, 50 NC [5 Jones] 420; State v Wood, 53 Vt 560; 1 Hale, The History of the Pleas of the Crown, pp 427-428; Hall, General Principles of Criminal Law [2d ed], p 267; LaFave & Scott, op. cit., p 250; cf. Clark & Marshall, Crimes [7th ed], § 4.01, pp 216-219). 4

State v Wood (53 Vt 560, supra) illustrates the flaws in the analyses of Justices Rubin and Niehoff. In that case, the trial court instructed the jury that although the victim died of the wound inflicted by another, acting independently, and not that inflicted previously by the defendant, “still if the wound inflicted by Wood was mortal, and would in course of time have killed [the victim], if he had not previously died from the wound inflicted by [the third person], and although he did not die of the wound by Wood, yet the latter could be convicted of murder” (53 Vt, at p *433 566). As the Vermont Supreme Court explained in reversing Wood’s conviction (p 566), “The court was in error in the assumption that a man can be convicted of murder although his act does not cause the death. The question does not turn upon the moral aspect of the case. The intent to murder may be never so plain, yet if something intervenes to prevent the consummation of the intent, if death does not follow from the act of the accused, he is not in law a murderer”.

True, when the culpable act produces the intervening cause the defendant remains liable for all foreseeable consequences (Clark & Marshall, Crimes [7th ed], § 4.01, p 213; Focht, Proximate Cause in the Law of Homicide — With Special Reference to California Cases, 12 So Cal L Rev 19, 33-34). Medical treatment for a wound is an intervening cause of this sort because acts of ordinary medical malpractice are foreseeable: “mere negligence in medical treatment is not so abnormal that the defendant should be freed of liability” (LaFave & Scott, op. cit., p 259; see People v Kane, 213 NY 260 ). After all, defendant’s conduct exposed the victim to the medical procedures and it is not unjust to hold him responsible for the attendant risks.

Nonetheless, responsibility does not extend to all intervening acts to which the victim would not have been exposed “but for” the defendant. The defendant is not liable for acts of gross negligence or intentional malpractice (LaFave & Scott, op. cit., p 259; cf. People v Stewart, 40 NY2d 692 [operation on unrelated hernia]). He is likewise not liable for coincidences, as, for example, if the victim dies in a traffic accident attributable to the negligence of the ambulance driver who is taking the victim to the hospital (LaFave & Scott, op. cit., p 259). The victim would not have been in the ambulance “but for” the defendant’s conduct. But the traffic accident was too remote; it was not foreseeable (cf. People v Kibbe, 35 NY2d 407). As LaFave and Scott point out (op. cit., p 259), for causation purposes “the medical negligence cases have usually emphasized that the negligence aggravated the wound inflicted by the defendant”.

*434 Reliance upon this theory of intervening cause is therefore misplaced. Defendant may have exposed the victim to treatment for the wound but not the organ harvesting operation because the physician’s acts were intentional, not negligent. It should be self-evident that such operation hardly constitutes an accepted method of treatment for an injury.

Moreover, the declaration of brain death and subsequent operation were not foreseeable. The whole purpose of such an operation is to benefit other patients. It does not absolutely follow brain death unless the permission of the next of kin is obtained. The medical procedure is still relatively uncommon and the state of the art is far from the point where brain death determinations are certain (LaFave & Scott, op. cit., § 67, p 533). Flat EEGs may also be caused by trauma, cardiac arrest and barbiturate overdoses (Hamlin, Life or Death by EEG, 190 J Amer Med Assn 112). 5

None of the cases cited by Justice Rubin involves comparable issues of causation. In Commonwealth v Golston (373 Mass 249, cert den 434 US 1039 ), State v Shaffer (223 Kan 244) 6 and Swafford v State (_ Ind _, 421 NE2d 596 ), there was evidence that death had occurred prior to the disconnection of the life support systems and the jury was instructed that it would have to so find in order to return a conviction. While no instruction was given in Cranmore v State ( 85 Wis 2d 722 ), there was no conflict in' medical testimony, all the experts testified that death had taken place prior to the operation. State v Fierro (124 Ariz 182) did not involve removal of organs and the court viewed the termination of life support systems as just passively stepping aside (see, also, People v Mitchell, 132 Cal App 3d 389, 396-397, where the court noted that different considerations might pertain in organ transplant cases).

As commentators have noted, causation cannot supply an escape from determination of the appropriate standard of ascertaining when death has occurred. If the patient *435 was, in fact, dead before the life support system was disconnected, then, what the physicians did after that point is totally irrelevant (see, e.g., O’Hara, Medical-Legal Agreement on Brain Death: An Assessment of the Uniform Determination of Death Act, 8 J of Contemp L 97, 120-121).

Adoption of the formulation set forth by Justices Rubin and Niehoff also leads to a maze of contradictions. Under their theory, defendant’s criminal liability hinges upon whether the individual “pulling the plug” too soon is a physician or a member of the decedent’s family. If the latter, the defendant would be relieved of criminal responsibility, no matter how laudable the motive and no matter how terminal the patient; if a physician, defendant would remain responsible irrespective of how erroneous his diagnosis of the prospects of death may be. Responsible jurisprudence cannot accept such logic.

In sum, I am not willing to strain settled principles of causation solely to sustain a conviction of homicide, as opposed to assault, simply because the defendant intended to commit a brutal crime. “The doctrine that a person whose conduct excites moral disapproval may be punished for doing what he has not done is * * * a dangerous one” (Hart & Honor, op. cit., p 294). 7

*436 IV

From a review of the record, I am convinced that the People failed to establish defendant’s guilt of manslaughter in the first degree beyond a reasonable doubt. There is simply no evidence that, under any standard, the victim “died” before the organ transplant operation was performed. To further highlight the prosecution’s failure to carry its burden on the question of death and its causes, it should be emphasized that the record contains no affirmative evidence that the victim’s respiration and circulation terminated at any point after the transplant operation was complete and the respirator was stopped, except that which can be deduced from the fact that the victim was dead in the common-law sense at the time that the autopsy was conducted by the medical examiner. On the proof adduced at trial it is entirely possible that the victim suffocated in the body bag on the way to the medical examiner’s office.

The reason for these deficiencies in proof is that the People relied exclusively on brain death, a standard we have determined is not appropriate for judicial adoption. The People were forced into this strategic position because the hospital record and the testimony of the medical experts would not support a finding of common-law death prior to the transplant operation and the People tried to avoid the causation problem generated by the transplant operation by fixing the death of the brain prior to the operation. The People’s failure to adduce proof of common-law death after the operation is consistent with their reliance on brain death and is tantamount to an admission that the transplant operation was an independent intervening cause which relieved the defendant of criminal responsibility for the death of the victim. Accordingly, on appeal, the People cannot overcome this failure of proof by belatedly urging this court to focus on the ultimate fact of death.

The testimony of all of the medical experts, indicating that the victim would have probably died from the bullet wound to the head, merely demonstrates that the defendant inflicted a mortal wound. It does not establish that the defendant’s conduct caused the death. Indeed, according to the medical examiner, if the victim was not brain *437 dead at the time of the operation, the operation would have killed him. There is a scanty medical record. The doctor who made the brain death determination did not refute the experts’ testimony concerning the necessity of certain procedural tests which were absent here. Moreover, he did not testify as to the results of certain tests or that certain tests were performed twice. The electroencephalograms, which are meant to be confirmatory evidence, did not confirm the brain death determination. Finally, there was no testimony that this physician was qualified to modify the elements of established protocols.

With respect to the medical examiner’s testimony, there are also difficulties because in his answers he never distinguished common-law death from brain death. Further, while he did not rely on the hospital’s determination of brain death to reach his opinion that the bullet wound caused death, his own finding of brain death was a necessary factor in his conclusions as to the cause of death. As noted above, he testified that if the victim was not brain dead at the time of the operation, the operation would have killed him and that he could not fix the time of brain death.

While it is the sole province of the jury to resolve issues of credibility (People v Gruttola, 43 NY2d 116, 122 ; People v Joyiens, 39 NY2d 197, 203 ), we are required to review the entire record to determine whether the evidence is sufficient in quality and quantity to justify a finding of guilt beyond a reasonable doubt (People v Reed, 40 NY2d 204, 208 ; People v Santos, 38 NY2d 173, 175-176 ). When, as here, the proof of a key element of the crime itself consists of no more than mere surmise and speculation, we are obliged to set the conviction aside (People v Montanez, 41 NY2d 53 ; People v Reed, supra; People v Williams, 35 NY2d 783 ).

This, however, does not require dismissal of the indictment. While due to the lack of proof of the causation element, the record does not establish the defendant’s guilt of manslaughter in the first degree, i.e., that with intent to cause serious physical injury to another person, he caused the death of such person (Penal Law, § 125.20, subd 1), all of the other elements have been proven beyond a reasonable doubt. This would support a conviction of assault in *438 the second degree (Penal Law, § 120.05, subd 1). Accordingly, the appropriate corrective action would be to reduce the conviction to that charge (CPL 470.15, subd 2, par [a]) and remit for resentencing (CPL 470.20, subd 4; see People v Graham, 36 NY2d 633 ). 8

V

Putting to one side the questions of whether a common-law or brain death standard is appropriate and whether there was, in the abstract, sufficient evidence adduced in support of either, it is evident that the trial court’s refusal to define death constituted serious error.

A properly instructed jury could have found causation lacking because the physicians, in their zeal to obtain the victim’s organs, shortened their patient’s life but a matter of hours (see Hart & Honor, op. cit., pp 308-309 [discussing an English case to that effect]; People v Mitchell, 132 Cal App 3d 389, 396-397, supra [finding causation where physicians merely disconnected the life support systems and indicating that a different result might have obtained had the deceased’s organs been removed for transplant purposes]; Runk, Brain Death: The Emerging Common Law Definition in Criminal Homicide, 6 Western State U L Rev 295, 305). In other words, the jury could have found that the infliction of the wound was the opportunity for, rather than the cause of, the organ harvesting operation.

Without a definition of death, the time that death occurred cannot be determined and the time of death affects causation (see State v Williams, 247 Ga 200 [indictment that did not set forth time and date of death was jurisdictionally defective]). This question was taken from the jury here by the trial court’s refusal to supply it with a standard for ascertaining the point at which death had occurred (cf. State v Meints, 212 Neb 410 [jury instruction patterned after generally accepted medical criteria of death]). The choice of a proper definition is a legal question and cannot be left solely to the jury’s discretion.

*439 It is axiomatic that a guilty verdict can never be directed and thus a court has no power to instruct the jury that any fact essential to conviction has been established as a matter of law "[n]o matter how conclusive the evidence was” (People v Walker, 198 NY 329, 334 ; see, also, People v Weiss, 290 NY 160, 171 ). Yet, no matter how disguised, that is the prop that supports the thesis adopted by a majority of this court. I cannot cast my vote to support it for “even the criminal most deserving of punishment is entitled, under our system, to a fair and impartial trial” (People v Herman, 255 App Div 314, 315 ; see, also, People v Donovan, 13 NY2d 148, 154 ; People v Bai, 7 NY2d 152, 155 ).

VI

For the reasons I have stated, I would modify the conviction of manslaughter in the first degree to one of assault in the second degree and remit for resentencing. Alternatively, I would remit for a new trial. 9

. I understand that the Los Angeles County District Attorney’s office, which employs brain death criteria by statute, picks among several standards, depending upon the expert called upon in the case. I would be loath to hinge guilt or innocence upon an expert’s opinion of whether or not a homicide has occurred (cf. People v Ciaccio, 47 NY2d 431, 439 ; People v Williams, 6 NY2d 18, 23 ; People v Graydon, 43 AD2d 842, 843 ).

. The Kansas statute, for example, has been both praised (Mills, The Kansas Death Act: Bold and Innovative, 285 New Eng J Med 968) and condemned (Kennedy, The Kansas Statute on Death, 285 New Eng J Med 946). (See, also, Capron & Kass, A Statutory Definition of the Standards for Determining Human Death: An Appraisal and a Proposal, 121 U of Pa L Rev 87 [critical of Kansas statute, proposing different criteria]; O’Hara, Medical-Legal Agreement on Brain Death: An Assessment of the Uniform Determination of Death Act, 8 J of Contemp L 97,118 [calling for uniformity!; Robinson, Determination of Death Legislation, 27 Gath Lawyer 246, 247 [same]; Note, The Uniform Determination of Death Act: An Effective Solution to the Problem of Defining Death, 39 Wash and Lee L Rev 1511, 1520 [same].) In this case, there was testimony to the effect that the criteria shift almost weekly.

. Common-law death is defined as meaning a “total stoppage of the circulation of the blood, and a cessation of the animal and vital functions consequent thereon, such as respiration, pulsation, etc.” (25A CJS, Death, § 1, p 546).

. Of course, the defendant could still be charged with attempted murder (see People v Dlugash, 41 NY2d 725 ).

. I note with interest the recent indictment of two Los Angeles physicians for murder (New York Times, May 6, 1983, p 14, col 6). Newspaper accounts indicate that they may have acted too quickly in removing a victim’s organs (New York Times, May 6, 1983, p 14, col 6; March 10,1983, p 18, col 1; Feb. 7,1983, p 10; Sept. 15,1982, p 16; Aug. 20,1982, p 8; Newsweek, Feb. 14,1983, p 76; Newsweek, March 21,1983, p 52; Debate on Boundaries of Life, Time, April 11, 1983, pp 68-69).

. Parenthetically, Kansas has adopted a statutory definition of brain death.

. As Justices Rubin and Niehoff point out, causation is not legislatively defined in this State. Perhaps this is because the question of causation is really one of fact and degree for which mechanical rules cannot be laid down; the general guides are “not tests, but clues” (Perkins, Criminal Law [2d ed], p 694, quoting Cardozo, The Paradoxes of Legal Science). No statute can be drafted which would encompass the myriad factual permutations (see, e.g., Hart & Honor, Causation in the Law, p 357, criticizing the formulation contained in section 2.03 of the Model Penal Code, as not providing specifically for those “cases where causal problems arise because, although the accused did not intend it, another human action besides [the] accused’s is involved in the production of the proscribed harm. These are treated merely as one kind of case where harm may or may not be ‘too accidental’ in its manner of occurrence”). The California Joint Legislative Committee for the Revision of the Penal Code proposed a two-step formulation for intentional crimes in “but for” circumstances; causation would be established if the intended result in fact occurred and “(i) is within the purpose or contemplation of the defendant, whether the purpose or contemplation extends to natural events or to the conduct of another, or, if not, (ii) involves the same kind of injury or harm as that designed or contemplated and is not too remote, accidental in its occurrence or dependent on another’s volitional act to have a just bearing on the defendant’s liability or on the gravity of the offense” (Penal Code Revision [Tent Draft No. 32], § 408, quoted in Kadish & Paulsen, Criminal Law [3d ed], p 325). The proposal was never enacted.

. Assault in the second degree is the only applicable lesser-included offense. Assault in the first degree contains elements additional to those specified for manslaughter in the first degree (cf. People v Glover, 57 NY2d 61 ) and defendant’s acquittal of murder, under the facts of this case, must be deemed to constitute an implicit acquittal of attempted murder as well (People v Ressler, 17 NY2d 174, 179 ).

. After this opinion was released, the California Court of Appeal, Second Appellate District, granted a writ of prohibition against the prosecution of the doctors in the case referred to in footnote 5 above, thereby reversing a Superior Court Judge who had, essentially, overruled a magistrate’s determination and directed the reinstatement of a murder information (Barber v Superior Ct., 147 Cal App 3d 1006). That court, in the course of its opinion, observed that a murder prosecution was a “poor way” to resolve the complex legal, medical and moral issues involved and urged prompt legislative action. I am, of course, in total agreement.