definition of that term, including the ancient common law definition." Neither suicide nor accessoryship to suicide are punishable any longer. If the victim has thus consciously terminated her own life-that she used a tool supplied by her husband, the slug, makes no difference-the husband cannot be guilty of murder, but only of attempted murder. With respect to the death itself, the husband is no more liable than the razor blade manufacturer who produced the blade with which the decedent in Lewis cut his throat. Nor does it make any difference whether the weapon or tool is supplied lawfully or unlawfully. (The slug as a suicide tool might be said to have been supplied unlawfully!)

(b) The outcome would be different if the wife died in ignorance that rescue was easily possible. In that case the husband's conduct has resulted in fruition. That the slug did not produce death by loss of blood or the like, but, rather, by an acute purulent meningitis does not alter his liability. No perpetrator can ever predict the causal train in its most minute concrete details. While I must concede that if, under (a), the husband is found guilty of murder, he cannot complain, for the liability is just as he expected, it is nevertheless true that, logically speaking, it was not he who succeeded in terminating his wife's life, but, rather, it was the wife herself. The husband is nevertheless guilty of attempted murder.

With this caveat, on the meaning of cause as effective, i.e., not neutralized, cause, in mind, it is also easy to solve the (Benge) case of the railroad foreman who was found guilty of manslaughter." The foreman's negligently reading the timetable and removing the tracks when it was improper to do so, clearly is a conditio sine qua non, a condition which cannot be imagined absent without failure of the result. But the question is: did the condition which the foreman interposed become effective, i.e., was it a cause to begin with? The answer is that it did not become effective or operative, by reason of the fact that the defendant foreman himself neutralized it-set it at naught-immediately upon creating it, by interposing safety measures fully capable of averting any harm. It took an independent causal constellation to produce the harm." (The liability of engineer and flagman will be discussed below.)

46 Blackstone, 4 Commentaries, 189.

47 Text at note 30, p. 180, supra.

48 Causation would fail, furthermore, for lack of the teleological element, quae vide, text under (2), infra.

2. The finalists have solved the problem of the evaluation once and for all in a general way which obviates the necessity of making separate individual inquiries in each particular case. This can be demonstrated by Poor Richard's Case, as presented with a variety of contributing causes." We had assumed that all the actors (rider and superiors, scouts, quartermaster, officers) considered the possibility of the harm and acted nevertheless as they did. They would, therefore, incur liability for recklessness (there would be even less difficulty if negligence sufficed), provided, however, that the conduct of each is a sine qua non. Whether it is or not is a question of fact which can be simply answered by reference to the meaning of sine qua non. Much of our own confusion at common law stems from the fact that we have never fully realized that there are indeed two possible ways of defining the sine qua non formula:

(a) Conduct is causal if, without it, and considering all remaining actually present circumstances, the result would not have occurred.50 Under this formula none of our defendants could be deemed to have caused the harm since, imagining the conduct of each one absent, successively, there still would remain a sufficient amount of activity amply sufficient in the aggregate to have produced the result. This sine qua non formula might recommend itself to a democratic society in which human dignity is a goal in itself.

51

(b) of several causes which can be imagined absent alternatively but not cumulatively without failure of the result, each one is a conditio sine qua non. Under this formula clearly all of our defendants would be deemed to have caused the result. This formula might recommend itself to societies with less stress on human dignity as a goal in itself.

It can be seen that after having made the original choice as to either of these two possible sine qua non formulas—a choice which our law has not made the result achieved with the factualteleological causation formula stands a good chance of winning the day through its independence from vague correctives.

One might wish to subject this new causation concept-the modern theory of causation to a further test, for which the Benge

4 Text post note 28, p. 179, supra.

5 This is the formulation of Spendel, Die Kausalitätsformel der Bedingungstheorie für die Handlungsdelikte, 81 (1948).

51 Welzel, Das Deutsche Strafrecht, 42 (6th ed., 1958).

52

case may serve us. We are here no longer concerned with the foreman, whom previously I found not liable for lack of proper causation. The question now arises whether the engineer, or the flagman, or both, have caused the harm. The teleological element (probably a relatively objective foreseeability sufficient for manslaughter) gives us no trouble. The problem centers solely around the sine qua non requirement.

As to the flagman, under sine qua non formula (a), we must imagine the flagman's 460 yards shortage in moving ahead, absent, i.e., absent the negligent conduct in question, the flagman would stand at 1,000 yards. All other factors remain constant, i.e., the negligent engineer spots the flagman only when the engine is alongside the flagman. This gives the engineer exactly 1,000 yards to stop the train, and this the train can do in fact. The harm would not have resulted, but for the flagman's shortage in moving. The flagman's conduct was causal of the harm.

Similarly, under sine qua non formula (a), we must imagine the engineer's 500 yards shortage in spotting the flagman absent, which would mean that the engineer would have spotted the flagman at 1,040 yards, which would have given him ample time to stop the train. Without the engineer's negligent conduct in question, the harm would not have occurred. The engineer's conduct was causal.

In this particular case the same result of liability for both actors would follow under the more stringent sine qua non formula (b).

There is one limitation upon the causation formula of the finalists, as indeed there is on any other causation theory: Causation can never be viewed as an isolated principle. It functions, and can function only, in the entire body of criminal law, comprised of all the principles. If it is this which the relevancy theory of causation " meant when it required conformity with the penal norm as part of the causation concept, it certainly scored a valid point. This must be briefly demonstrated:

53

The King v. De Marny led to the conviction of a newspaper editor for "aiding and abetting the publication in England of obscene literature, and the sending through the post in England a packet the sending of which is prohibited by the Post Office (Protection)

52 Text at note 30, p. 130, supra.

53 Text at note 13, p. 198, supra, for comparison.

M.E.

14

Act, 1884."" The defendant had published in his newspaper advertisements offering books, catalogues and photographs for sale by the advertiser. His conduct was the sine qua non, under either formula, for the subsequent publication, mailing and dissemination in England of the materials in question. Moreover, it was proved that the defendant had knowledge that such unlawful dissemination, etc., would be the consequence of his conduct. Thus, the teleological element was present as well, i.e., the defendant acted finally," with the result in mind. But is causation the only element on which criminal liability rests? No: The finding of causation is but one prerequisite for conviction, for unless mens rea -here scienter of the obscene nature of the disseminated publication-be established in addition, the crime is not complete. And so the court in effect held.

A difficult case, previously discussed in this paper, likewise led to an incorrect result because of the failure of the court to consider the case within the totality of criminal law principles and doctrines:

Root's conduct in agreeing to and engaging in a race with the decedent was causal in the sine qua non sense. If he consciously considered the decedent's death as possible or likely, he might be credited with the necessary recklessness for involuntary manslaughter. Causal imputation would seem to follow." On closer analysis, however, criminal liability (though not causal imputation) would have to be excluded for reasons which have nothing to do with causation:

Neither suicide nor self-mutilation nor self-endangering acts are unlawful as such. Both actors committed a self-endangering act in concert. But co-principalship in a lawful self-endangering act cannot raise to unlawfulness what is lawful if done singly (barring application of the conspiracy concept). Nobody, legally speaking, is his brother's keeper, unless the law imposes a legal duty of care and protection. There was no such duty in the instant case. Truly, both have violated the traffic laws, and the survivor will be made liable for his violation. But homicide liability can follow therefrom only if we arbitrarily decree that every self-endangering act or every

14 The King v. De Marny [1907] 1 K.B. 388; 21 Cox C.C. 371; repr. Hall, Cases and Readings on Criminal Law and Procedure, 73 (1949).

* See also Smith v. California, 80 S.Ct. 215 (1959).

54 Accord, a decision of the German Federal Supreme Court, 7 B.G.H.St. 112.

traffic offense resulting in death should constitute involuntary manslaughter. Such I do not understand to be the law. It follows that the decision in the Root case was wrong for reasons which have nothing to do with causation.

Lastly, we should test the modern causation theory on the challenging Pennsylvania felony-murder cases:

If causation were the only requirement for criminal liability, murder liability would have to be affirmed in these cases if, indeed, the defendants meant to inflict death or foresaw it as certain; manslaughter liability would seem to follow if the defendants merely took a gambler's risk that death might follow," provided however, in either case, that the defendants' conduct would qualify as causa sine qua non. We have here a course of conduct constituting a psychological stimulation directed at the victim, or the police, to use deadly force. The defendants' conduct cannot be imagined absent without failure of the homicide. In effect, the defendants stimulated the killing of a co-felon or an innocent bystander, which act of killing constitutes justifiable homicide on the part of the robbery victim or police officer who fired the shot. The law provides that to incite, etc., another to commit a crime, amounts to accessoryship before the fact to that crime, if the other actually perpetrates the crime and the instigator is absent. Here, however, neither was the instigator absent, nor did the other commit a crime, for justifiable homicide is no crime. Since presence of the inciter makes for co-principalship, it might be argued that the police officer or the robbery victim are co-principals with the robbers. Not only is this an absurd suggestion, but it also fails technically for the reason that what police officer or robbery victim has committed simply is not criminal. The law, however, also provides that killing through intentional employment of an innocent agent makes for liability as a principal in the person who employed the innocent agent. But the defendants did not intentionally employ an innocent agent to procure human death. It follows that, under proper considerations of all the principles and doctrines of our criminal law, while the defendants have caused death, they are not guilty of murder

57 But under the archaic wide-open felony-murder doctrine, all teleological considerations would be immaterial and murder liability would follow as a matter of course if robbery coincides with death. Pennsylvania has such a statute or, at least, the majority of the court interpreted it that way. See Musmanno J., dissenting, in Commonwealth v. Bolish, 391 Pa. 550, 563; 138 A. 2d 447, 454 (1958).