the word 'proximate' is unfortunately used, and seems often to mislead the inquirer, and to produce misapprehension of the real rule of law. That which is the actual cause of the loss, whether operating directly, or by putting intervening agencies - the operation of which could not be reasonably avoided in motion, by which the loss is produced, is the cause to which such loss should be attributed." 17

"It [Bacon's Maxim] is not intended to apply to a cause remote as regards distance, or remote in point of time, for so long as the loss is in the sequence of the antecedent, without other independent coöperating cause, neither distance nor time can affect the question.'

" 18

"The expression, indeed, remoteness is calculated to mislead, since a man may be held the cause of, and liable for, damage which may be a very remote consequence of his conduct, provided there be no intermediate cause to which it can be more properly referred." 19

Another alleged test of the existence of causal relation is one which is occasionally found in instructions to juries, but is not likely to be sustained in a well-considered opinion of an appellate court. It is sometimes called "The But for Rule" or "But for which Rule," and sometimes "The Causa sine qua non Rule."

17 Martin, C. J., in Brady v. Northwestern Ins. Co., 11 Mich. 425, 445 (1863). 18 Guthrie Smith, Damages, 380.

19 Dicey, Parties, 411.

As to the general subject, see also 4 Am. L. Rev. 210, 214; 9 Col. L. Rev. 144, note 20; 15 HARV. L. Rev. 542-543; Terry, Leading Principles of Anglo-American Law, § 542; 33 Can. L. J. 713, 714; Willard, Law of Personal Rights, 235.

Various words and phrases used in reference to causation are criticised by Mr. Green in the following passage in 8 Am. L. Rev. 518-519:

"It is said that a recovery can be had for 'certain but not for uncertain damage'; for 'the proximate and natural consequences of the act complained of, but not for remote or consequential loss,' that 'the damage must arise naturally'; that 'it must be the fair, legal, and natural result of the act'; or, as Mr. Greenleaf expresses it (2 Greenl. Ev. 210), 'the damage to be recovered must always be the natural and proximate consequence of the act complained of.""

66

Now all these expressions are vague: they mean little; and in the majority of instances in which they are employed they probably mean nothing. No person who uses one of them, if asked what he means by it, can give a well-defined explanation. Such sentences are not a solution of a difficulty; they are stereotyped forms for gliding over a difficulty without explaining it. When a court say this damage is remote, it does not flow naturally, it is not proximate; all they mean, and all they can mean, is that under all the circumstances they think the plaintiff should not recover. They did not arrive at the conclusion of themselves by reasoning with those phrases, and by making use of them in their decision they do not render that decision clearer to others. The employment of such phrases has never solved one single difficulty. .

This test or rule affirms that the defendant's tort is the legal cause of the plaintiff's damage if, but for the commission of defendant's tort, the damage would not have happened.

If this statement is turned round and put in a negative form, it can generally be applied as a test of what is not the cause of an event.

Generally speaking, a defendant is not liable, unless it be true that, but for his tortious act, the damage would not have happened. A defendant's tort cannot generally be considered the legal cause of plaintiff's damage, if that damage would have occurred just the same even though defendant's tort had never been committed.20

But it is a mistake to apply the "but for" rule affirmatively, as constituting a sole sufficient test of causation, as was done in the charge to the jury in Gilman v. Noyes.21

The "but for" requirement is generally one of the indispensable elements to make out legal cause. But it is not the only requisite element; it is not per se an all-sufficient element. The fact that the damage would not have happened, "but for" the commission. of the defendant's tort, does not invariably justify the conclusion that the tort was the cause, in the legal sense, of the damage. At the present day, a causa sine qua non is not regarded as being necessarily the causa causans.22 There is no absolute rule that defendant is the legal cause if there is a tortious act of his anywhere in the chain of antecedents, no matter how far back. The defendant's tort must be distinctly traceable as one of the substantial efficient antecedents; as having had a substantial share in subjecting plaintiff to the damage. A tort very remote in time or space may have practically spent its force and may not have been potentially oper

20 See Sowles v. Moore, 65 Vt. 322, 26 Atl. 629 (1893). The only exception we can think of is where two or more tortious causes are simultaneously operating, each being independent of the other, and each being alone sufficient to produce the damaging result; e. g., X. and Y., each acting independently of the other, simultaneously shoot at B., and each hit him in a vital spot. B. is instantly killed; and each shot alone was sufficient to produce death. X. could not escape on the plea that the death would have resulted just the same if his tort had never been committed. See Watson, Damages for Personal Injuries, §§ 60-64, and § 43. See also Corey v. Havener, 182 Mass. 250, 65 N. E. 69 (1902).

21 57 N. H. 627 (1876).

22 See Ladd, J., Gilman v. Noyes, 57 N. H. 627, 631-632 (1876). "The primitive conception of a sufficient legal cause was a causa sine qua non." Professor Bohlen in 21 HARV. L. REV. 234-235. At an early day the "but for" rule prevailed.

& Maitland, History of the Common Law, 2 ed., 470.

2 Pollock

ative at the time of the harm: or its effect may have been infinitesimal.23 To constitute the tortious conduct of a defendant the legal cause of the damage suffered by the plaintiff, "its connection must be something more than one of a series of antecedent events without which the injury would not have happened." 24

Probably there are cases where no injustice would result, if the "but for" rule was given to the jury as the sole test of causative relation. Judge Ladd, whose opinion is always entitled to consideration, seems to have thought that Gilman v. Noyes was a case of this class.25 But Judge Ladd, in the same opinion, explicitly denied that the "but for" rule could be universally applied as the sole test in determining the existence of causal relation.26

Another alleged test of the existence of causal relation is based upon the distinction between a cause and a condition.

But although we may reject Mr. Mill's view that we have no right to give the name of cause to any one out of a number of antecedent conditions27- still the alleged distinction does not solve the question of the existence of causal relation. It is simply a restatement of the original problem in a different form of words.

"The distinction between cause and condition would be valuable, if there were any definite standard for determining what is a cause and what is a condition. The only standard by which this can be deter23 It "has so far expended itself, that its influence in producing the injury is too minute for the law's notice." Bishop, Non-Contract Law, §§ 41, 44. And see 9 HARV. L. REV. 84, 85.

24 Hart, J., Pittsburgh Reduction Co. v. Horton, 87 Ark. 576, 577, 113 S. W. 647, 648 (1908); Schoultz v. Eckhardt Mfg. Co., 112 La. 568, 36 So. 593 (1904); Gilman v. Noyes, 57 N. H. 627 (1876); Lord Dunedin, in Dunnigan v. Cavan & Lind Mfg. Co., 48 Scot. L. R. 459, 461 (1911).

25 In Palmer v. Concord, 48 N. H. 211 (1868) the court practically held that the legislature intended to give this definition to the word "caused," as used in a certain

statute.

26 It is a mistake to regard the "but for" theory as exactly equivalent to the "theory of the logicians," as stated by Mill. It would seem that the theory of the logicians would render nobody liable. Mill says that all antecedents are equally causes, or rather parts of the cause (the cause being the sum of all the antecedents), and that we have no right to single out any one of them and call it the cause. The "but for" theory adopts Mill's premise; that all antecedents are, in one sense, causes; but, unlike Mill, says that plaintiff may single out any one tortious human antecedent and hold him liable.

27 Mill, Logic, 9 Eng. ed., 378-383.

mined is the same as that which determines a proximate from a remote cause; . . . Accordingly, 'condition' or 'occasion,' while affording a convenient verbal distinction, is, in use, likely to mislead thinkers into a conviction that they have something which they have not." 28 Pollock says: ". . . the contrast of cause' and 'condition' is dangerous to refine upon;

" 29

There is an alleged rule, known as the "Last (or Nearest) Wrongdoer Rule," of which Dr. Wharton is an especial advocate.30

This rule is, in substance, as follows:

The legal cause is the last (or nearest) culpable human actor to be found in the chain of antecedents; i. e., the one acting last before, or nearest to, the happening of the damage to plaintiff.

This rule is sometimes propounded, not as a supplement to the rule of liability for probable consequences; but as furnishing, per se, a complete and all-sufficient doctrine of legal cause.

Taking this rule literally, the test under it would be to trace back the links in the chain of antecedents,31 until we come to a wrongful act of a responsible human being; to the last or nearest wrongful act of a free human agent. The person doing that act is the legal cause, (and semble is responsible even though the result could not have reasonably been anticipated).

The above rule is a good working rule to apply in a hurry. In a great majority of cases it gives the correct result. But it will not always do so.

1. The last wrongdoer is not always liable as the legal cause of the damage.

2. Moreover, the last wrongdoer, if himself liable, is not necessarily the only party liable. An earlier wrongdoer may sometimes be suable at the election of plaintiff.32

29 Pollock, Torts, 8 ed., 464, note l. See also 9 Col. L. Rev. 144, note 23.

30 See Wharton, Negligence, 1 ed., Appendix, bottom paging 823, et seq., also in text, §§ 85-99, and 134-145.

31 See Mr. Green's criticism of the phrase "chain of causation,” 4 Am. L. Rev. 211,

212.

32 See Watson, Damages for Personal Injuries, § 74. When the damage results from the simultaneous concurrent acts of two independent wrongdoers, neither could escape on the ground that he was not the last wrongdoer.

As to the first point:

If the fault of the nearest wrongdoer is a very remote link in the chain of antecedents, he may not always be considered the legal cause of the damage. The force which he set in motion may be regarded as having become exhausted before the happening of the damage. It may no longer have been potentially operative at the time when the damage occurred; or its effect then may have been infinitesimal. It may have "so far expended itself, that its influence in producing the injury is too minute for the law's notice." " The defendant's tort, in order to constitute it a legal cause, must be distinctly traceable as one of the substantial efficient antecedents; as having had a substantial share in subjecting plaintiff to the damage.

This idea is expressed by Professor Beale in the following words:

"One at least of the factors of the act of injury must in a fair sense be due to the defendant. If the force he set in motion has become, so to speak, merged in the general forces which surround us, or in the language of Bishop has 'exhausted itself' like a spent cartridge, it can be followed no further. Any later combination of circumstances to which it may contribute in some degree is too remote from the defendant to be chargeable to him."

"Conceivably, of course, we might resolve every act of injury into its ultimate human forces, and charge each person who had set one of these forces in motion with his share of the act of injury. This would take us back to Adam in every case. Human knowledge is too small to perform such a task with justice, and time too short for the determination by this method of a single case. For their own protection, and for the security of the public at large, the courts refuse to go so far; beyond a certain point the operation of a force is called remote, and is disregarded." 34

As to the second point (the possible liability of an earlier wrongdoer).

Suppose that there are two tortious human actors, A. and B., in the chain of antecedents, not acting in concert; that A.'s tort began earlier; and that B.'s tort, which began later, immediately preceded the happening of the damage to the innocent plaintiff and

33 See Bishop, Non-Contract Law, §§ 41 and 44, and 2 Bishop, New Criminal Law, § 637.

34 9 HARV. L. REV. 84-85, and 85, note 2.