CHAPTER EIGHT: SCOPE OF DUTY (THE CORE OF THE DUTY-

RISK ANALYSIS)
Dean Leon Green, along with Prof. Wex S. Malone, fomented a Copernican revolution in the
analysis of the “causal” components in the common-law negligence cause-of-action. The law
review Article below provides a good overview of Dean Green’s diagnosis of the indeterminacy
and analytic weakness of traditional proximate cause, along with an outline of his solutions. For
further reading, see Leon Green, Rationale of Proximate Cause (1927) and Leon Green, The
Litigation Process in Tort Law (2d ed. 1976.

“The Causal Relation Issue in Negligence Law,” 60 Michigan Law

Review 543 (1962)
Leon Green*

[1] Two significant legal studies of "Causation” one English, one American -- have been
recently published.5 The English book brings to the subject more scholarly learning and a more
comprehensive examination of its literature than any other book that has been written. The authors
are devoted disciples of causation principles and make a stout defense of the causation concept
as the structural core of negligence law. They examine the philosophical, common sense and
semantic backgrounds of causal concepts as the basis of legal liability, find that they have merit,
and launch extended, and sometimes devastating, attack upon theories that question their adequacy,
though in some instances their understanding of the subject of their attack seems only peripheral.
They give slight attention to the administrative, economic, moral, and other environmental factors
that have conditioned the decisions of courts indeed they greatly discount policy considerations as
limitations on liability-nor do they adequately consider the procedural apparatus of the litigation
process in allocating the functions of judge and jury. Whatever the disagreement with respect to
the details of the analysis, arguments, and the use made of the numerous cases reviewed by the
authors, their ambitious and enlightened discourse excites admiration. Their exposition is chaste,
crisp, and clear, and their arguments have appeal. In the main their efforts are given to reading
sense into traditional and historical usages of the cause concept though they reject much of the
metaphorical and metaphysical terminology found in the opinions of the courts and the writing of
commentators. Nothing short of the reading of the book will suffice to indicate their serious
probing of all phases of causation in the law and the high quality of advocacy devoted to bolstering
the waning influence of cause doctrines in the de termination of the liabilities and reparation of
litigants.
[2] The American book is limited to "factual causation." The thesis of the authors is
painstakingly presented and cannot be adequately summarized in any brief space. The thorough
detail of their arguments make for difficult but profitable reading. They are interested primarily

* Professor of Law, University of Texas.

5
Becht & Miller, THE TEST OF FACTUAL CAUSATION IN NEGLIGENCE AND STRICT LIABILITY CASES (1961); H.L.A Hart
& A. HONORE, CAUSATION IN THE LAW 1959). The latter English book explores the causation concept in tort, contract,
and criminal law together with certain Continental theories. The references to the book in this article are limited to
negligence law.
402
in the development of their own theory and devote slight attention to the consideration or criticism
of theories advanced by others, except those found in the barrier breaking article recently published
by Professor Malone.6 A brief chapter is given over to a discussion of the English book, which
appeared at the time their book was ready for publication. While much of the authors' exposition
is readily acceptable their basic premises seem highly doubtful.7
[3] Both books have one fatal weakness in common. The authors of each book overload the
causal relation issue with difficulties more readily and more adequately dealt with in the
consideration of other issues. This is a vice found in most of the decisional and textual writings on
the problems of legal liability. The American book, even though expressly restricted to a
consideration of factual causation, founders on this reef. The cases and problems posed by the
authors are difficult because of issues foreign to factual causation. Some of the hypothetical cases
employed to test their theory more nearly resemble riddles than situations found in real life, as
fantastic as some of the occurrences of everyday life may be. To their credit, however, they admit
the breakdown of their theory when it becomes apparent that it gets the wrong result.8
[4] The causal relation problem is doubtless found in its most in tractable form in
negligence cases. If it can be brought under control there, similar treatment is available in
other cases. It is not believed that causal relation can be dealt with in isolation as is attempted
by the American authors, or that it can be dealt with generally as is done by the English
authors. Causation runs through every negligence case but its peculiar significance in a
particular case can be grasped only by setting it off against the other issues in the case. Because
of the close kinship in meaning between cause on the one hand and blame, fault, culpability,

6
Malone, Ruminations on Cause-in-Fact, 9 STAN. L. REV. 60 (1956).
7
The operative theory of Factual Causation seems to be based on the assumption that "acts" and "omissions" must be
distinguished, and also that in determining causal relation, the negligent segment of a defendant's conduct must be
separated from his total conduct. See BECHT & MILLER, op. cit. supra note 1, at 21, 26, 27. The attempt to link injury to
negligence instead of to conduct makes much trouble for them in their analysis of specific cases and leads to numerous
highly questionable conclusions.
8
The following cases are taken from BECHT & MILLER, op. cit. supra note I, at 90.
Case 1. "The defendant turns left without giving the statutory signal and the plaintiff, driving in (coming
from) the opposite direction, collides with him. The plaintiff, how- ever, was not keeping a lookout far
enough ahead to have seen a signal if one had been given. The failure to signal is not a cause of harm,
because, tracing hypothetically from the omission, one can see that a signal would not have avoided the
accident. This problem involves the odd circumstance that the contributory negligence is also not a cause of
the harm, as, tracing hypothetically from the omission, there would have been no signal to see if the plaintiff had
looked. The plaintiff loses because he cannot prove a causal relation between the defendant's negligence and the harm,
not because his own negligence was a hypothetical cause of the harm."
A few pages over at 95, the authors REVert to the same situation with this change: Case 2. "But suppose that
the driver who kept no lookout had a passenger in his car who was hurt in the accident and sued both defendants.
While the conduct of each defendant was a simple cause of the harm it is clear as before that the negligence of neither
of them, tracing hypothetically, was a cause of it. To allow recovery against either or both would contradict the axiom
that the negligence, and not merely the conduct, must be a cause of the harm. On the other hand, each defendant would
be saved only by the other's negligence; to deny relief on such facts seems morally indefensible to us, and we would
accordingly permit recovery though admitting that the causal relation usually required is lacking."
It is difficult to visualize such occurrences without raising other questions such as: (1) did D see or should he
have seen P's car, and (2) could the driver P have seen D's car though no signal was given? But accepted as stated, the
solution given seems weird. As a matter of fact, in both cases the defense of no causal relation runs afoul of the
proposition that conduct or other fact relied on as a defense must itself be causally related to the victim's injury. Here it
is conceded that the driver's failure to see a signal that was not given is not causally related to his own or his
passenger's injury. D's driving his car across the path of the driver's car was a cause of the collision. He owed a duty
both to the driver and his passenger not to drive across the path of the oncoming car without giving a signal. He
violated his duty to both, and he has no defense in either case.
403
and responsibility on the other hand, the causal relation problem almost inevitably draws to its
consideration the more weighty considerations of these competing and, as frequently used,
synonymous concepts. This is not surprising for it was not so long ago that the common law of
our forebears was based on the moral principle that one who hurt another, however
innocently, should compensate him.9 Throughout my years of teaching I have never ceased to
be amazed by the tyranny that our elemental concepts wield over the human mind and its
thought processes. Even now after liability has been greatly modified by negligence law we
run ahead in our mind to grapple in terms of cause with the ultimate problems of
responsibility though causal relation is obvious. For this reason I place great emphasis on the
necessity of dealing with the causal relation issue in the light of all the other issues of a case.
If the other issues are not clearly formulated there is great danger that they will become
confused with the cause issue and convert it in a twinkling into some other issue underlying
responsibility. Because of the over loading of the cause issue, and in order to indicate some of
the differences of importance I have with my American and English friends, I think it well to
set out briefly the process which gives the causal relation issue its full significance, but no
more.
[5] The orthodox analysis of a negligence case seems adequate for this purpose. (1) Did
defendant's conduct contribute to the victim's injury (the causal relation issue)? (2) Was the victim
protected under the law against the defendant's conduct with respect to the injury inflicted on him
(the duty issue)? (3) Did defendant violate his duty under the law with respect to the victim's injury
(the negligence issue)? (4) What is the evaluation in money of the losses suffered by the victim
as a result of his injury (the damage issue)? There may arise in the particular case any number of
issues subordinate to each of these basic issues, as well as affirmative defensive issues to each of them.
1. Affirmative Conduct
[6] The beginning point of all tort liability is affirmative conduct, and the first step in
establishing a defendant's liability is to identify him and connect his conduct with the victim's
injury. Somewhere along the line of defendant's conduct the doing of something which contributed
to the victim's injury must be found. At common law there is no liability for doing nothing unless
there has been some undertaking which imposes an obligation upon the defendant to do something.
The undertaking may involve the defendant's own conduct or that of someone who stands in some
relation to him, for example, as a servant. Omissions to act are merely incidents in a longer line of
affirmative conduct and are easily resolved into it. Thus it is that tort law is designed to give
protection to the victim against danger incident to some activity of the defendant.10 In
establishing causal relation between the victim's hurt and the defendant's undertaking, the detailed
acts, or omissions to act, after the undertaking is put in operation are important only to indicate the

9
See Green, The Thrust of Tort Law (The Influence of Environment), 64 W. VA. L. REV. 1 (1961).
10
The point of no-action no-liability is developed in the "Good Samaritan" doctrine. Depue v. Flatau, 100 Minn.
299, 111 N.W. 1 (1907); Tullgren v. Amoskeag Mfg. Co., 82 N.H. 268, 133 Atl. 4 (1926). Some courts have held
that one whose conduct hurts another innocently is under no duty to render aid to him. Union Pac. Ry. v.
Cappier, 66 Kan. 649, 72 Pac. 281 (1903); Osterlind v. Hill, 263 Mass. 73, 160 N.E. 301 (1928) ; Buchanan v.
Rose, 138 Tex. 390, 159 S.W.2d 109 (1942); Boyer v. Gulf, Colo. & S.F. Ry. Co., 306 S."W.2d 215 (Tex. Civ. App.
1957) . These and other cases of the same import are believed to be insupportable. Hardy v. Brooks, 118 S.E.2d
492 (Ga. App. 1961); Johnson v. Rea, Ltd., [1961] 1 Weekly L.R. 1400, 78 L.Q. REV. 6 (1962). The very fact of
injury as a result of a person's conduct, however innocent, should place him under a duty to render aid, and failure to
do so should make him liable for further injuries suffered as a result of violating his duty. The "hit and run" statutes,
now quite universal, reflect this attitude. Gregory, Gratuitous Undertakings and the Duty of Care, 1 DE PAUL L. REV.
30 (1951); McNiece and Thornton, Affirmative Duties in Tort, 58 YALE L.J. 1272 (1949); Comment, The Failure To
Rescue; A Comparative Study, 52 COLUM. L. REV. 631 (1952). See Note, 75 HARV. L. REV. 641 (1962) .

404
course of conduct. For example, a person while driving a car may put his foot on the accelerator
and collide with the victim, or he may fail to put his foot on the brake, or simply fail to stop, with
the same result. The affirmative undertaking to drive the car is the important factor in causing the
victim's injury. Pressing the accelerator, or failing to use the brake, or simply failing to stop in time,
are significant only as details which give color to the driver's conduct and which may render it
negligent. The driving of the car and the collision are sufficient basis for finding the causal relation
issue favorably to the victim.
[7] An obligation imposed by law to do something by virtue of some relation assumed would
not require different treatment were the defendant to fail to meet the obligation unless it was
imposed to meet a specific situation.11 Consider the licensing cases,12 the un licensed driver who runs
down a pedestrian and is charged with negligence.13 In such cases there is usually no question that the
driver's conduct in driving the motor vehicle contributed to the victim's injury. Nothing more needs
to be known on the issue of causal relation. To attempt to link the victim's injury to an absence
of a driver's license would be impossible as well as uncalled for. If the absence of a license has any
relevance at all it must be to some other issue. Did the driver owe a duty to the pedestrian to
have a license? Was he negligent in not having a license with respect to the injury suffered by the
pedestrian? If the absence of a license were relevant to show the driver's incompetence that would
go only to the negligence issue. Even for that purpose the factual data incident to the collision would
overshadow any inference that could be drawn from the absence of a license. Moreover, if in some
case the absence of a license might be relevant as a circumstance to bolster some other circumstance,
it might well be excluded on the basis that it would tend to prove too much, i.e., give too great weight
in the minds of jurors. The only certain generalization that can be made is that it has no relevance at all
to the causal relation issue, and there is no general rule that would make it relevant to other issues.14
2. The Causal Relation Issue
[8] The causal relation issue in a negligence litigation is greatly simplified by the fact that it is
restricted to the conduct of the parties; in the first instance to that of the defendant. It does not
initiate an exploratory search for all the causes that contributed to the victim's injury, or a search
for the cause, or the proximate or the legal cause. Nor does it initiate a search of the why of the
defendant's conduct, i.e., a search for the motive, reason, impulse or other accounting in explanation
of his conduct. A philosophic or scientific exploration of defendant's conduct may be relevant to
11
Many statutes impose specific duties. Injury contributed to by the conduct of the defendant in violating
such a statute seldom, if ever, raises any issue of causal relation between the injury and the conduct. The
issue of negligence may be foreclosed by the statute. But frequently the duty imposed does not include
the risk of injury suffered by the victim.
12
BECHT & MILLER, op. cit. supra note 1, at 141; HART & HONORE, op. cit. supra note 1, at 111ꞏ14, 193 et seq. The
authors are troubled by the attempts made by courts to link the victim's injury to the absence of a license,
and rightly so. No issue of causal relation is involved unless the statute is mandatory, and even then it is
between injury and conduct, and not between absence of a license and injury.
13
Wysock v. Borchers Bros., 104 Cal. App. 2d 571, 232 P.2d 531 (1951); Mundy v. Pirie-Slaughter Motor
Co., 146 Tex. 314, 206 S.W.2d 587 (1947).
14
Brown v. Shyne, 242 N.Y. 176, 151 N.E. 197 (1926) is regarded as a tough case soluble on the basis of
causation. But there was little doubt about the fact that the chiropractor's treatments contributed to the patient's
paralysis. The failure to have a license to practice was not offered to show causal relation but to show
incompetence and thus negligence. The relevance of a license to the negligence issue was doubtful, but
Judge Lehman so confused the problem with his discussion of proximate cause that no two agree on what
he did, except (1) to hold the absence of a license should not have been admitted to be considered as
some evidence of negligence, (2) to break the back of the opposition of the medical association to the
licensing of chiropractors by the legislature. An act was passed the following year providing for licensing
chiropractors and further providing that an absence of a license was prima facie evidence of negligence. N.Y.
EDUC. LAW § 6513.
405
other issues but not to the causal relation issue. The inquiry is limited to the fact of defendant's
contribution to the injury. The search for proximate, legal or other causes is designed to determine
whether the defendant's conduct should be condemned and he be made to compensate for his
victim's injury. Numerous lawyers and judges use proximate cause when they mean causal relation
but that is only one of several confusing usages of the term. The only relevance the consideration
of other cause factors may have in the determination of the causal relation issue is the light they
may shed on whether defendant's conduct contributed to the injury.
[9] Also it is well to understand that it is not important to the causal relation issue that
defendant's conduct in whole or in part was lawful, unlawful, intentional, unintentional, negligent,
or non-negligent. The moment some moral consideration is introduced into the inquiry the issue is
no longer one of causal relation. Causal relation is a neutral issue, blind to right and wrong. It
is so easy to think and speak of defendant's negligence15 as the cause of a victim's hurt that it is
frequently overlooked that causal relation is the beginning point of liability and must be established
or tentatively assumed before issues involving duty, negligence, dam ages, and the defensive issues
can be determined. There may be causal relation but no basis for the later inquiries, but in absence
of causal relation plaintiff has no case, and all other inquiries become moot.
[10] There is great advantage in considering the causal relation issue first. This was done in the
recent cigarette-cancer case with excellent results.16 Conduct is a factual concept; the victim's hurt
is a factual concept; causal relation is a factual concept. Duty, negligence, and damages are legal
concepts and depend upon different considerations from those involved in the determination of
causal relation. Only after liability has been determined can negligence be merged with conduct,
and even then for clarity's sake the merger should be called negligent conduct.
[11] In the overwhelming number of cases it is very clear that a defendant's conduct contributed
to the victim's hurt, but in what respect the conduct was negligent, if at all, may be a serious
problem. For example, in a highway collision case defendant may be charged with driving his
automobile at excessive speed, with bad brakes, inadequate lights, failure to keep a lookout or to
give a signal, driving while intoxicated, or the violation of any combination of traffic regulations.
If it is shown that defendant in driving the automobile collided with the victim, the causal relation
issue between defendant's conduct and the hurt is settled once and for all, irrespective of the
violation of any police regulation or common-law rule. Whether defendant's violation of any one
or more of the regulations was negligent is another and distinct problem. Great difficulty and
great confusion frequently arise at this point. It is here that the authors of Factual Causation17
go astray by insisting that causal relation must be found between the segment of defendant's
conduct that is negligent and the victim's injury. A plaintiff in his pleading frequently pinpoints
the specific conduct of the defendant which he alleges contributed to his injury and which was
negligent. The specific conduct may be characterized as some act or omission. This narrows and
restricts the issue of negligence to the specific conduct. But it does not narrow or restrict defendant's
conduct to the specifications as a cause of the injury. For example, the charge is excessive speed.
The driving of the car which resulted in the collision is a factual cause; the excessive speed
characterizes the driving as negligent. The negligence issue cannot be determined until causal relation
between defend ant's driving and the victim's hurt is established or provisionally assumed, although

15
It is a professional habit to speak of defendant's negligence as a cause-a habit the writer has not been able
to correct in his own usage. Where used as factual causation in most instances the context is clear that it
is the defendant's conduct the user has in mind. Musko v. Walton, 358 P.2d 49 (Colo. 1960); Stevenson v.
City of Kansas City, 187 Kan. 705, 360 P.2d I (1961) ; Martin v. Commercial Standard Ins. Co., 350 S.W.2d
664 (Tex. Civ. App. 1961) ; Byrnes v. Stephens, 349 S.W.2d 611 (Tex. Civ. App. 1961) .
16
Pritchard v. Liggett & Myers Tobacco Co., 295 F.2d 292 (3d Cir. 1961) (cigarette cancer case).
17
BECHT & MILLER, op. cit. supra note 1.
406
the same evidence may support both issues.18 Causal relation is the factual basis for finding that
the speed was negligent or not. It may be undisputed that the victim was hurt as a result of the
defendant's driving, but his speed may be found not excessive and hence not negligent.19 Put
another way, it is defendant's conduct that inflicts the hurt, but it is the law that makes his conduct
negligent. Negligence must be based on causal relation, but causal relation can never be based on
negligence in the air.20
[12] Assume that defendant was speeding along a street crowded with pedestrians and vehicles.
Two blocks away the victim was run down. Defendant's speeding two blocks back would not be
relevant to show that he ran down the victim, but if it were shown that the victim was run down by
defendant, his speeding two blocks back would be relevant to show that he was still speeding when
he ran down the victim, and hence was negligent. If it were found that defendant was not speeding
at the time he ran the victim down, some other phase of his conduct would have to be found
negligent in order to impose liability upon defendant even though causal relation between his
driving and the injury were conceded. Causal relation between defendant's conduct and the
victim's hurt must of necessity precede a finding that defendant was negligent with respect to the
victim and the injury he suffered; but as simple as this requirement may seem hundreds of courts
and many writers put "the cart before the horse" by insisting that negligence must be found as a

18
Kentucky Power Co. v. Kilbourn, 307 S.W.2d 9 (Ky. 1957); Sims v. Dixon, 224 Ore. 45, 355 P.2d 478
(1960); Law v. Uinta Oil Refining Co., 12 Utah 2d 229, 364 P.2d 1024 (1961) ; Byrnes v. Stephens, 349
S.W.2d 611 (Tex. Civ. App. 1961) . This does not mean that a single act may not be the cause of a
victim's hurt and also be negligent, but usually a line of conduct must be known before either causal relation or
the negligent quality of the conduct can be determined.
19 The confusion at this point is illustrated in the case of Kabzenell v. Stevens, 168 Cal. App. 2d 370, 336
P.2d 250 (1959) . Plaintiff was a passenger on defendant's bus. The driver drove the bus so close to the
inside of a mountainous road that a protruding root struck the glass in a window and plaintiff's finger was
cut off by broken glass. Plaintiff alleged that D was negligent in driving at an excessive speed, failing to
blow his horn, driving too far off the pavement, etc. Defendant's defense was the fear of collision with
an oncoming truck around the curve. The jury found a verdict for defendant. Among other points, plaintiff
insisted that the failure to blow the horn was negligent and that if it had been blown the truck driver
would have heard it and defendant would not have had to pull over and would have avoided the root.
The court in support of the jury's verdict argued that the jury could have found no causal connection
between a failure to blow the horn and plaintiff's injury. This was a false issue. The plaintiff had been hurt
as a result of defendant's conduct. Whether he was negligent in the several respects alleged was correctly
submitted and the jury found for defendant. That was the only question the court had before it and the
attempt to justify the verdict by speculating that the jury could have found no causal relation between
one aspect of defendant's conduct and the plaintiff's hurt was not required.
20 Benedick v. Potts, 88 Md. 52, 56 (1898): "Until you know what did occasion an injury, you cannot say that
defendant was guilty of some negligence that produced the injury." Thus it is that the res ipsa loquitur cases
illustrate this point so vividly. Manley v. New York Tel. Co., 303 N.Y. 18, 100 N.E.2d 113 (1951). Plaintiff
claimed he was knocked out while talking over the phone but did not prove what knocked him out. Lack
of causal relation made res ipsa loquitur unavailable. But the fellow at the other end of the line offered
proof that he was knocked out by electricity and on the basis of res ipsa loquitur recovered $25,000.
Seeley v. New York Tel. Co., 281 App. Div. 285, 120 N.Y.S.2d 262 (1953). National Lead Co. v. Schuft, 176
F.2d 610 (8th Cir. 1949); Burr v. Sherwin-Williams Co., 42 Cal. 2d 682, 268 P.2d 1041 (1954); Emigh v.
Andrews, 164 Kan. 732, 191 P.2d 901 (1948); Sullivan v. Crabtree, 36 Tenn. App. 469, 258 S.W.2d 782 (1953).
Also see Hansen v. Phagan, 361 P.2d 977 (Colo. 1961) (neither causal relation nor negligence could be
shown); Kentucky Power Co. v. Kilbourn, 307 S.W.2d 9 (Ky. 1957) (causal relation and negligence based on
meager evidence); Frye v. City of Detroit, 256 Mich. 466, 239 N.W. 886 (1932) (negligent conduct clear but
no causal relation could be shown); Wank v. Ambrosino, 307 N.Y. 321, 121 N.E.2d 246 (1954) (causal relation
but not sufficient evidence of negligence) . See Malone, Contrasting Images of Torts: The Judicial Personality
of Justice Traynor, 13 STAN. L. REV. 779, 798 (1961) .

407
basis of finding causal relation between it and the victim's injury.21 What they do is to draw in some
other issue which they seek to determine on the basis of cause.
[13] It may be further observed in this connection that even though both causal relation and
defendant's negligence are clear, there still may be no liability. Courts frequently make great efforts
to show there is no causal relation between the victim's injury and defendant's negligence when the
problem is whether the risk is one within the protection of the duty owed by the defendant to the
victim.22 There are also many cases in which a defendant has violated some statute, is found
negligent, and the plaintiff has been injured as a result of the defendant's conduct, but defenses of
assumed risk and contributory negligence such as stepping out in front of a speeding car, crossing
against a red light, or other violation by the victim of traffic regulations imposed for his protection,
preclude the risk to which he has been subjected as beyond the scope of the defendant's duty. It is
in these cases, especially, that the causal relation issue may be so overburdened with considerations
pertinent to other issues that causation in negligence law becomes a nightmare for courts,
practitioners, students, and teachers.23
[14] The simplest and most conclusive defense a defendant can make to the issue of causal
relation is to deny that he did anything to contribute to the victim's hurt. This brings into question
his conduct with respect to the injury. The plaintiff must offer enough evidence to support a
reasonable inference of causal relation, and this may be impossible.24 Suppose in an action for
wrongful death the victim was found dead, his body in the highway near his car parked beside the
highway. Marks on his clothing indicate that he was crushed by the wheels of a truck. Several
trucks, including that of the defendant, are known to have passed along the highway shortly before
the body was found, still warm. Did defendant's conduct in driving his truck kill or contribute to
the death of the victim?25 All the truck drivers testify that they saw the victim's car but all deny that
they saw him or his body. Plaintiff may be able to prove circumstantial details that point to the
defendant and the defendant may offer evidence to counter the details. The problem for the trial
judge is the sufficiency of the evidence to support a reasonable inference that defendant's conduct
in driving the truck killed or contributed to the death of the victim. It is for the jury to draw the
inference if the issue is submitted. There is no test that will be of any decisive value to the judge

21
Probably the Texas Special Issue Submission practice is the worst offender in this respect. On each
specification of negligent conduct, it is usual to submit three or more issues. Example in substance: I. Did
the party keep a proper lookout? 2. Was the failure to do so negligent? 3. Was such negligence a proximate
cause of the injury? See the thorough study and exposition of the practice in HODGES, SPECIAL ISSUE SUBMISSION IN
TEXAS 106 et seq. (1959) . The "proximate cause issue" is resolved on the basis of "foreseeability" and thus is
essentially a repetition of the negligence issue. See Dellwo v. Pearson, 107 N.W.2d 859 (Minn. 1961) and
authorities cited for the rejection of this usage of "foreseeability."
22
Smith v. Sharp, 82 Idaho 420, 354 P.2d 172 (1960); Berg v. New York Cent. R.R., 391 Ill. 52, 62 N.E.2d
676 (1945) and cases relied on therein; State v. Hatfield, 197 Md. 249, 78 A.2d 754 (1951); Emery v.
Rochester Tel. Corp., 271 N.Y. 306, 3 N.E.2d 434 (1936); Davis v. Atlas Assur. Co., 112 Ohio St. 543, 147 N.E.
913 (1925) . Sometimes the efforts are to show there is a causal relation when the basic issue is something
else. Socony-Vacuum Oil Co. v. Marshall, 222 F.2d 604 (1st Cir. 1955); Werkman v. Howard Zink Corp., 97
Cal. App. 2d 418, 218 P.2d 43 (1950); Henningsen v. Markowitz, 132 Misc. 547, 230 N.Y. Supp. 313 (Sup.
Ct. 1928); Hall v. Coble Dairies, 234 N.C. 206, 67 S.E.2d 63 (1951); Hines v. Morrow, 236 S.W. 183 (Tex.
Civ. App. 1921).
23
The cases are legion in which the contributory negligence, last clear chance, discovered peril and kindred
doctrines have been resolved into proximate cause, sole proximate or some other cause doctrine. See Green,
Proximate Cause in Texas Negligence Cases, 28 TEXAS L. REV. 471, 621, 755 (1950).
24
Sheptur v. Procter & Gamble Distrib. Co., 261 F.2d 221 (6th Cir. 1958) ; Hansen v. Phagan, 361 P.2d 977
(Colo. 1961); Mosko v. Walton, 358 P.2d 49 (Colo. 1960); Terry v. Timberlake, 348 S.W.2d 919 (Ky. 1961);
Sanders v. McMichael, 200 Okla. 501, 197 P.2d 280 (1948).
25 Eley v. Cunon, 121 Cal. App. 2d 280, 263 P.2d 86 (1953); Comet Motor Freight Lines v. Holmes, 203
S.W.2d 233 (Tex. Civ. App. 1947).
408
in the performance of his function or to the jury in drawing a reasonable inference, other than their
respective capacities to evaluate the evidence.26 If their evaluations are favorable to the plaintiff a
succession of other difficult issues will still remain and success on the causal relation issue may be
an empty victory.27
[15] The function of the judge in determining the sufficiency of the evidence to raise any issue,
and particularly the causal relation issue, is seldom given the importance to which it is entitled.28
It is sometimes confused with the substantive issue when the problem is whether the court will
permit the jury to have the issue, and in absence of a jury it is not always clear whether the judge is
passing on the sufficiency of the evidence or is deciding the substantive issue. The formula whether
reasonable minds may draw different inferences from the evidentiary data is disarmingly simple,
but it does not simplify the determination the judge must make in close cases. It is here that the
"substantial," "appreciable," "material" factor formula is at times invoked. Its value for the judge
in performing his function, or its value for a jury in passing on the issue is slight, to say the most.
There is no substitute in the performance of their respective functions for their capacity to draw
inferences. But if the judge must have some expression to articulate his own judgment or to caution
a jury to weigh the evidence carefully, no better term has been suggested, and it is certainly no
more vague or incomprehensible than the term "preponderance of the evidence" so generally used.
The fact that "substantial" cannot be defined, further analyzed or broken down into lesser terms,
frets the authors of Causation in the Law no end,29 as it does all those who do not keep in mind the
necessities of the procedural apparatus of the litigation process.
[16] In some extreme cases, such as injuries resulting from the simultaneous discharge of
firearms by two or more persons,30 or injuries suffered during a surgical operation in which several
persons participated,31 the plaintiff has no means of identifying the person or persons whose
conduct resulted in his hurt, but the court may place the burden on the defendants to offer evidence
to show the factual details of the incident, and further to show that their con duct did not contribute
to the victim's injury.32 In determining whether from all the evidentiary details of the incident
in its factual environment an inference can be drawn that the defend ants' conduct contributed to
the victim's hurt, it is very clear that the court may be greatly influenced by policies which it
considers essential to the ends of justice.33 And if the issue of causal relation is submitted to the

26
Causal relation cannot be reduced to simpler terms. As BECHT & MILLER, op. cit. supra note 1, so aptly
noted: "Who can describe or define 'yellow' "?
27
Greene v. Atcheson, T. & S.F. Ry., 120 Cal. App. 2d 135, 260 P.2d 834 (1953); Texas & Pac. Ry. v.
Shoemaker, 98 Tex. 451, 84 S.W. 1049 (1905); Annot., 40 A.L.R.2d 881 (1955).
28 Eitel v. Times, Inc., 221 Ore. 585, 591, 352 P.2d 485 (1960) : "The evaluation of the evidence to
determine whether a verdict can stand is one of the most difficult tasks presented to the courts." Smith, The
Power of the Judge To Direct a Verdict, 24 COLUM. L. REV. 1ll (1924).
29 See HART & HONORE, op. cit. supra note 1, at 92, 98, 117, 155, 216-18, 263-66, 276. The authors' frustration
seems to be that the terms cannot be defined. The terms reasonable, ordinary, average, prudent, foreseeable, and
many more such terms used by the profession are equally undefinable. Moreover, the terms approved by the
authors, such as abnormal, coincidental, occasioning, necessary, relevant (Chs. 5 and 6) have a similar
weakness. But the point is not a matter of definition but the capacity of all such terms to call forth the
judgment of someone, judge or jury, on the issues such terms put in focus for judgment.
30
BECHT & MILLER, op. cit. supra note 1, at 104; Summers v. Tice, 33 Cal. 2d 80, 199 P.2d 1 (1948); Norling
v. Carr, 211 F.2d 897 (7th Cir. 1954); Annot., 5 A.L.R.2d 98 (1949).
31 Ybarra v. Spangard, 93 Cal. App. 2d 43, 208 P.2d 445 (1949); Ybarra v. Spangard, 25 Cal. 2d 486, 154
P.2d 687 (1944); Cassidy v. Ministry of Health, (1951) 2 K.B. 343.
32
Authorities cited in notes 26, 27 supra.
33
Professor Malone in his article, Ruminations on Cause-in-Fact, 9 STAN. L. REV. 60, 81-88 (1956), has clearly
demonstrated how in dealing with the causal relation problem the courts do not close their eyes to the

409
jury there is little doubt that they in their con sideration of the issue will be influenced by the whole
factual environment and whether they think the victim should be compensated by the defendants.
Even though the issue of causal relation is determined favorably to the plaintiff, whether
defendant's conduct was negligent must also be subjected to a similar weighing of the sufficiency
of the evidence and the jury's determination of the issue. At both steps in the trial of a case rules
of law can only set the stage for the ultimate judgment of those who administer the law.34
[17] Sometimes, though not often, it is thought to be a problem how much causal relation must
be found between a defendant's conduct and the victim's hurt in order to resolve the issue in the
plaintiff's favor. If the evidence is sufficient to support a reason able inference that the defendant's
conduct contributed to the hurt, or that reasonable men may draw different inferences, nothing
more is required. Let it be noted that this is a different problem from how much the victim has been
hurt by defendant's conduct,35 or how much he should be compensated. That judges in the
exercise of their function to determine the sufficiency of the evidence for submitting the causal
relation issue to the jury will differ in their conclusions, and that juries in the exercise of their
functions will also differ, must be accepted as a necessary risk of the litigation process. That is the
way of the courthouse. Any qualification that causal relation must be substantial or material is
only a caution that the finding must rest on intelligent and rea sonable considerations.36 The judge
should not submit the issue in absence of substantial, appreciable, or material evidence to support
a finding either way on the issue.
[18] Although causal relation cannot be reduced to any lower terms and cannot be measured by
any yardstick other than the good judgment of judge and jury, various tests have been suggested
and utilized for its determination. The tests usually befuddle far more than they clarify. The "but-
for" test is quite widely supported as a reliable test for causal relation. "But for A would B have
happened?”37 The question focuses inquiry on another issue more difficult than the original and
may not be answerable. "Event A is not a cause if event B would have happened without A.”38 This
may or may not be true in many cases and its truth can never be demonstrated for sure. Tests of
this character have the same vice as any "if," or any analogy. They take the eye off the ball. They
are essentially argumentative, and may fall short even as arguments, for when they seem to work
they are not needed, and where light is needed they throw another shadow.39 A defendant's con duct
may contribute to the victim's injury even though there are other causal factors that would have
caused the same or similar injury to the victim. This is demonstrated in the fire cases multiple fires

practical effects of their decisions at this point but yield to policy considerations as they do on other issues
in the case.
34
It has often been observed how impossible it is to hold juries within the bounds of the instructions they
receive and how in fact the very heart of jury trial is their power to moderate the law. McCormick, Jury
Verdicts Upon Special Questions in Civil Cases, 27 J. AM. JUD. SOC'Y, 84, 85 (1943); Pound, Law in Books and
Law in Action, 44 AM. L. REV. 12 (1910); Wigmore, A Program for the Trial of Jury Trial, 12 J. AM. JUD. SOC'Y
166 (1929).
35
In cases involving disease, former injuries, or injuries by other persons, how much the defendant has hurt
the victim, if at all, may be an exceedingly difficult problem. Belt v. St. Louis-S.F. Ry., 195 F.2d 241 (10th
Cir. 1952) ; Taylor v. Pole, 16 Cal. 2d 668, 107 P.2d 614 (1940); Farmer v. McColm, 364 P.2d 1059 (Colo.
1961); Tullgren v. Amoskeag Mfg. Co., 82 N.H. 268, 133 Atl. 4 (1926); Cohenour v. Smart, 205 Okla. 668,
240 P.2d 91 (1951); Landers v. East Tex. Saltwater Disposal Co., 151 Tex. 251, 248 S.W.2d 731 (1952), 31
Texas L. REV. 226; Carr v. Martin, 35 Wash. 2d 753, 215 P.2d 411 (1950); Farley v. Crystal Coal &: Coke Co.,
85 W. Va. 595, 102 S.E. 265 (1920).
36
Anderson v. Minneapolis, St. P. & S. Ste. M. Ry., 146 Minn. 430, 179 N.W. 45 (1920).
37
BECHT & MILLER, op. cit. supra note 1, at 13-21; HART & HONORE, op. cit. supra note 1, ch. V.
38
Ibid.
39
Kabzenell v. Stevens, 168 Cal. App. 2d 370, 336 P.2d 250 (1959); Byrnes v. Stephens, 349 S.W.2d 611 (Tex.
Civ. App. 1961).
410
(some of innocent, some of unknown, and some of negligent origin), merging fires, and fires
contributed to by the victim's own conduct.40 Suffice it to say that the issue of causal relation is
seldom difficult in these cases though the extent of duty, the defendant's negligence, and the
damages suffered may be exceedingly difficult.
[19] A so-called rule of "probabilities" is not infrequently put into play as a test of causal
relation. It is a highly seductive "decoy." It has the same vice as the "but-for" test, namely, it takes
the focus off the defendant's conduct and goes abroad for other causes. At most it is an argument
and, though a legitimate one for the advocate, can never rise to the level of a test of causal relation.
This is readily demonstrated. The issue is whether defendant's conduct contributed to the victim's
injury. To turn the issue into whether the defendant's conduct caused the injury is a different issue,
and then to require that defendant's conduct must be found to be more probable than not the cause
of the injury completely perverts the original issue. The judge's function is merely to say whether
there is sufficient evidence to raise the issue that defendant's conduct contributed to the injury. There
are no limits put on what he considers in reaching his judgment. When he says that he will not
allow a jury to guess or speculate about the matter he is merely saying the evidence is insufficient
to raise the issue. But if he finds the evidence sufficient to require the submission of the issue it is
not his function to tell the jury that they must find it "more probable than not" that defendant's
conduct caused the victim's injury. The plaintiff's burden of proof requires only that the jury find
from a preponderance of the evidence that defendant's conduct contributed to the victim's injury.
[20] A victim's hurt as the result, at least in part, of a defendant's conduct may be highly
improbable and yet admittedly true,41 while on the other hand it may be highly probable and yet
the result of other cause factors.42 Since there are other causes in every case that also contributed to
the victim's hurt, to seek to find whether defendant's conduct was the cause is literal nonsense, and
equally so is the attempt to find whether the "probabilities" are that it was the cause. The
"probability" test as employed on the issue of causal relation is an "argumentative decoy" not
infrequently employed by the courts when it is not needed, or to reach an insupportable judgment.
Perhaps it has been brought over to the causal relation issue because of confusing the use made of
the concept as a rule of admissibility of evidence, and also as a test of the negligence issue in which

40
BECHT & MILLER, op. cit. supra note 1, at 22. See Peaslee, Multiple Causation and Damages, 47 HARV. L. REV.
1127 (1934). Also see such cases as Douglas, Burt & Buchanan Co. v. Texas & Pac. Ry., 150 La. 1038, 91 So.
503 (1922); Illinois Cent. R.R. v. Wright, 135 Miss. 435, 100 So. 1 (1924).
41
See Tullgren v. Amoskeag Mfg. Co., 82 N.H. 268, 133 Atl. 4 (1926), for the example given of a motorist
taking the inside of a sharp curve on a mountain road where only three or four cars per day travel and it is
highly improbable that the motorist will meet another car coming around the curve. But a collision does occur
and P is hurt. Causal relation is clear and negligence may be found, in spite of the slight probability that a
collision would occur. The only test of the former is, did it happen? The test of the latter is, would a
reasonable person have taken the risk?
42
Assume the body of P's decedent is found on the tracks of X railroad with his feet severed and missing.
He was known to have walked that way about the time X's train passed. All the circumstances would
indicate that X's train probably severed decedent's feet, but there is no basis for finding that X was negligent
in the operation of its train. P sues D for decedent's death, and after extended investigation finds sufficient
evidence to support a reasonable inference that D's train ran over P's decedent on an intersecting track and D's
trainmen placed his body where it was found on X's track. The evidence is highly conflicting and it would
seem highly improbable that D's train• men would do such a thing. But if the proof will support a
reasonable inference that they did, and the fact is found by a jury, causal relation will be established, and
also it can be inferred from the same evidence that they were negligent in operating D's train so as to run
over decedent. Probability is not a test of either causal relation or negligent conduct, and even as an
argument may be highly treacherous if pressed to the extreme. The facts here assumed are suggested by Texas
&: N.O. R.R. v. Ozuna, 266 S.W.2d 896 (Tex. Civ. App. 1954) .
411
probability of harm is frequently employed interchangeably with foreseeability of harm.43
[21] In Kuhn v. Banker,44 the defendant doctor had failed to take an X-ray of a bony union he
had sought to make, although the patient told him the bones were scraping and had not joined. She
dragged her leg around for some five months with his assurance that the union had been made.
When she consulted another surgeon it was too late to attempt another union, and the expert
testimony was that if a union had been attempted at the time of the patient's complaint to the
defendant it was less probable that a union would have resulted than that it would not have resulted.
On this basis a judgment was directed for defendant and affirmed by the appellate courts. This
highly meritorious case was made to turn on a wholly false issue. The issue was not whether a
complete and successful union of the bones had been probable, but rather what injury the patient
had suffered as a result of defendant's failure to take another X-ray and make another attempt at
union, or employ other remedial measures, which failure was conceded to have been negligent. It
was without doubt that the patient had lost at least one chance in three or four for a union and
full recovery if timely treatment had been given, and that she suffered further impairment of her
condition over a period of five months without remedial treatment, together with great physical
pain and emotional agony and other more calculable losses. The Ohio courts were somehow held
in bondage by a rule of "probabilities" that was resolved into a rule of law which left no basis for
damages unless she could show the probability of a complete union of her broken limb. Other
courts have been led astray by the same or similar arguments.45
[22] Perhaps the cases in which the causal relation issue is most difficult are those in which the
defendant has failed to provide some safeguard for the victim's protection. Usually in these cases
the other issues will be found favorable to the victim if a causal relation between his hurt and the
defendant's conduct can be found. Such are cases in which the failure to provide fire escapes,46
lifeboats and lifesavers, lighting for stairways, and the like.47 The causal relation issue in these
cases seems at first blush to call for a judgment on what would have happened had defendant
performed his duty-and here it is that "but for" and "probability" arguments have their greatest
play. As arguments they are valid, but it is error to state the issue in these terms. The causal
43
Probability is usually associated with the negligence issue. See HARPER & JAMES, THE I.AW OF TORTS § 19.4, at
1068 (1956); PROSSER, TORTS 200 (2d ed. 1955). See Eitel v. Times, Inc., 221 Ore. 585, 352 P.2d 485 (1960) (citing
PROSSER, supra, at 222). The court does not reject the test of probabilities but clearly indicates that the
matter involved is one of drawing inferences and if a reasonable inference can be drawn that defendant
contributed to the victim's injury it is for the jury to draw it. Also, it rejects the cliche that an inference
cannot be based on an inference. See also Sims v. Dixon, 224 Ore. 45, 355 P.2d 478 (1960). The fact that other
reasonable inferences in conflict can also be drawn is unimportant. See also Jack Cooper Transp. Co. v. Griffin,
356 P.2d 748 (Okla. 1960); Martin v. Commercial Standard Ins. Co., 350 S.W.2d 664 (Tex. Civ. App.
1961).
44
133 Ohio St. 304, 13 N.E.2d 242 (1938).
45
Ramberg v. Morgan, 209 Iowa 474, 218 N.W. 492 (1928); Cole v. Shell Petroleum Corp., 149 Kan. 25, 86
P.2d 740 (1939); Wright v. Clement, 287 Mass. 175, 190 N.E. 11 (1934); Bowles v. Bourdon, 148 Tex. 1, 219
S.W.2d 779 (1949); Barker v. Heaney, 82 S.W.2d 417 (Tex. Ct. Civ. App. 1935); cf. Dunham v. Village of
Canisteo, 303 N.Y. 498, 104 N.E.2d 872 (1952), 31 TEXAS L. REV. 446 (1953); Sundquist v. Madison
Rys., 197 Wis. 83, 221 N.W. 392 (1928).
46
Burt v. Nichols, 264 Mo. I, 173 S.W. 681 (1915); Weeks v. McNulty, 101 Tenn. 495, 48 S.W. 809 (1898);
Radley v. Knepfly, 104 Tex. 130, 135 S.W. Ill (19II); Berry v. Farmers Exchange, 156 Wash. 65, 286 Pac.
46 (1930) ; See Malone's excellent discussion of these cases, supra note 2, at 77-81.
47
In Ford v. Trident Fisheries Co., 232 Mass. 400, 122 N.E. 389 (1919), the law re quiring adequate life-saving
apparatus ready for use afforded no protection against the risk of the seaman's being lost when no one of
the ship's crew knew when or how he had fallen overboard, even though D's putting to sea in heavy
weather contributed to his falling overboard; DiNicola v. Pennsylvania R.R., 158 F.2d 856, 857 (2d Cir. 1946);
Reynolds v. Texas &: Pac. Ry., 37 La. Ann. 694 (1885); Arrington v. Young, 366 P.2d 400 (Okla. 1961)
(questionable); Malone, supra note 2, at 75-79.
412
relation issue does not change its character; it is still, did defendant's conduct in failing to provide
safety facilities contribute to the victim's in jury? To ask whether he would have escaped unscathed
had the facilities been provided may present a false issue heavily weighted against the victim and
one that can seldom, if ever, be answered. The arguments may range far and wide but the court's
function is to determine whether the evidence is sufficient to warrant the submission of the issue,
and if so, it is the jury's function to say whether defendant's conduct contributed to the victim's
injury. If the judge is denied the power to comment on the weight of the evidence, he should not
be permitted to formulate a rule of law that gives him the power to bind the jury's judgment even
more effectually than would his comments.
[23] It will be noticed in these cases that the factual details of the immediate environment out
of which the case arose must be shown with considerable particularity. Causal relation will not be
presumed from the fact of injury or from a showing defendant has violated his duty.48 It may be a
case in which the duty and its violation did not include the risk to which the victim was subjected.49
But if the details indicate that the victim's safety was put in peril by defendant's conduct it is for a
jury to find whether that conduct did in fact contribute to the injury he suffered. The difficult
question is the sufficiency of the evidence to raise the issue. So long as the evidence will support
an inference that defendant's conduct contributed to the victim's injury, even though other
inferences can be drawn that it did not, or that his injury was due to other causes, it is for the
jury and not the judge to draw the inference.50 There are many close cases at this point and there
is no doubt, as so ably demonstrated by Professor Malone,51 that judges are influenced by policy
considerations in their determination of the sufficiency of the evidence as a basis for submission of
the causal relation issue. In cases in which it is clear that the defendant has been callous to the
victim's safety they will strain to the very limits of good judgment to permit the jury to pass on the
issue. In other cases in which the inferences are even more clearly based the courts may strain the
other way, and they have numerous formulas which rationalize their straining in either direction.52
[24] In other types of cases it may also be extremely difficult to establish a connection between
the plaintiff's injury and the defendant's conduct. In many cases the parties may have to rely almost
wholly on scientific proof, i.e., the opinion of experts, and they may differ widely in their
opinions.53 Whatever the complexity, if enough evidence is adduced to support a reasonable
inference that defendant's conduct played a part in the result, the issue is for the jury. The same
multiplicity of causes and the difficulties of determining if defendant's conduct played any part in
the victim's injury will also have great, doubtless even greater, significance in determining the
issues of duty, negligence, and damages, as well as the defensive issues raised by the defendant.

48
See discussion by Judge Brown in Schlichter v. Port Arthur Towing Co., 288 F.2d 801 (5th Cir. 1961) (no
evidence to support issue of causal relation or negligence).
49
Stacy v. Knickerbocker Ice Co., 84 Wis. 614, 54 N.W. 1091 (1893); RESTATEMENT, TORTS § 432 (1934) ; BECHT &
MILLER, op. cit. supra note 1, at 89. Cases of this type are frequently dealt with as involving a causal relation
issue. It is usually very clear what happened and that what defendant had done contributed to the victim's
injury, but the risk was not within the scope of defendant's duty. Smith v. Sharp, 82 Idaho 420, 354 P.2d 172
(1960), and Texas &: Pac. Ry. v. Bigham, 90 Tex. 223, 38 S.W. 162 (1896), submit to ready solution on this basis.
50
Sanders v. Atcheson, T. &: S.F. Ry., 65 N.M. 286, 336 P.2d 324 (1959); Pitchfork Land & Cattle Co. v. King,
346 S.W.2d 598 (Tex. 1961).
51
Malone, supra note 2, at 72.
52
Id. at 68.
53
Pritchard v. Liggett &: Myers Tobacco Co., 295 F.2d 292 (3d Cir. 1961) (cigarette cancer case); Pryor v. Lee C.
Moore Corp., 262 F.2d 673 (10th Cir. 1959); Hagy v. Allied Chemical & Dye Corp., 122 Cal. App. 2d 361, 265 P.2d
86 (1953) (hearing denied 1954); Dorsey v. Muilenburg, 345 S.W.2d 134 (Mo. 1961) ; Jack Cooper Transp.
Co. v. Griffin, 356 P.2d 748 (Okla. 1960); Seattle-First Nat'l Bank v. Rankin, 367 P.2d 835 (Wash. 1962);
Gregory v. Shannon, 367 P.2d 152 (1961); Richison v. Nunn, 57 Wash. 2d I, 340 P.2d 793 (1959);
Konieczka v. Mt. Clemens Products Co., 360 Mich. 500, 104 N.W.2d 202 (1960).
413
"When causal relation becomes confused with these issues it is rarely submitted to a jury by intelligible
instructions. In most American state courts it is combined with other issues, usually false issues
such as proximate cause, legal cause, and the like, or combined with the negligence foreseeability
formula,54 thereby submerging the causal relation issue in a complex of doctrinal terminology that
serves primarily as a source of errors for appellate review and frequent reversals. Perhaps the only
virtue this confusion of issues and terminology has is that the instructions are frequently so in
comprehensible that juries, where permitted, give a general verdict based on their sense of justice
under the facts, and where special issue verdicts are required on a multitude of fractional issues,
answer the issues to support a judgment for the party they think should prevail.
3. Other Issues
[25] Whether a victim suffered hurt as a result of defendant's con duct is a question in
negligence cases wholly apart from whether the defendant should be made to compensate the
victim. Whether defendant should be made to compensate the victim for his hurt, and how much,
involves a congeries of inquiries which can be re solved only by other considerations. Tort law for
some six hundred years was grounded in deep morality; one who hurt another should make
recompense; unlawfulness, intent, malice, and other such moral-laden terms were the touchstones
of tort law.55 In cases of negligent hurts these terms have given way to terms almost bereft of moral
content such as duty, breach of duty, assumption of risk, accident, contributory negligence, last
clear chance, and others. They may still have a moral coloration but their chief substance is made
up of other factors subsumed under the concept of "fault."56"Cause," "fault," and "blame" have a
large area of common meaning and are frequently used synonymously. This accounts in large part
for the numerous attempts to find solutions of duty, violation of duty, and even damages on the
basis of "causes." Whenever an attempt is made, as is too frequently done, to trace causal relation
from a defendant's negligence to a victim's hurt, the inquiry is turned into one of responsibility-
ought the defendant be made to compensate the victim for his hurt.57
A. Duty
[26] The second inquiry in the determination of a defendant's responsibility for his conduct is
directed at his duty to the victim and whether it included the risk of injury which the victim has
suffered. This is a policy problem-a matter of law. The usual inquiry made of an advocate for the
plaintiff in a doubtful case is: "Have you any law to support your claim?" This is the whole import
of the duty concept. There is no mystery involved and it requires no further definition. In most
cases the law has already been determined by former decisions or by statute under which a
doctrine or rule governing defendant's conduct is firmly established. But if defendant's conduct is
different in some significant respect from that involved in former cases, it must be determined
whether the rule or doctrine relied on should be expanded, modified, or reformulated to include or

54
Stevenson v. City of Kansas City, 187 Kan. 705, 360 P.2d I (1961). A typical instruction is found in Baumler
v. Hazlewood, 347 S.W.2d 560 (Tex. 1961) . There was no question that the conduct of both plaintiff and
deceased contributed to plaintiff's injury. The only issues were negligence and contributory negligence. The
instruction combined a causal relation formula with the "foreseeability" negligence formula as a formula for
"proximate cause." This combination of the two issues into a hybrid formula serves no good purpose and
may result in great injustice. For the use of the foreseeability formula, see Green, Foreseeability in Negligence
Law, 61 COLUM. L. REV. 1401 (1961) .
55
Green, The Thrust of Tort Law, 64 W. VA. L. REV. I (1961); Stone, Touchstones of Tort Liability, 2 STAN.
L. REV. 259 (1950).
56
GREEN, TRAFFIC VICTIMS, TORT LAW AND INSURANCE 9 (1958).
57
This is what happens when an issue of proximate cause based on "foreseeability" is interposed as an issue
in a negligence case. It may be either a partial substitute for or crude duplication of the negligence issue.
See note 50 supra.
414
exclude the conduct of defendant which contributed to the victim's hurt. If there is no law at hand,
the court may be called upon to make new law, and this it does however it may rule. It is at this
stage of the judicial process that a court exercises its "law-making" function, which may be and
usually is, influenced by some administrative, economic, moral, and/or other factors, summed up
as justice.58Whatever the impelling factors may be, they involve considerations that go beyond the
interests of the immediate parties to the litigation and include the interests of "the people" in the
administration of the law in the case before the court, and also in future cases that may come before
the courts. Unless the court can bring defendant's conduct and the victim's protection within the
scope of some doctrine or rule of law as determined by the court, there is no duty under which
defendant is obligated to compensate the victim for his hurt. It may be emphasized that after the
courts have given sanction to some legal concept or rule of law, it will sooner or later fail to respond
to the "facts" of somewhat similar later cases. Since courts must of necessity rely heavily upon legal
concepts and rules for the administration of law, their concepts and rules must be under continuous
reconsideration in order that they may be adjusted to accommodate the facts of new cases.59 It is in
this way that the law grows and escapes sterility.60
[27] In the determination of a defendant's duty and whether it covers the injury his victim
suffered, matters of policy as indicated above are of prime importance. The problem may be very
complex. Many of the complexities arise from the fact that the conduct of other persons, including
that of the victim himself, and other factors such as natural phenomena and animals, and the
combination and conjunction of any one or more of these factors may play a part in causing the
victim's hurt. These factors are frequently identified as concurring, intervening, supervening,
superseding, independent, dependent, proximate, remote, efficient, and other causes, and may be
dealt with as though they were essential to the determination of causal relation between defendant's
conduct and his victim's injury.61 As already indicated, they are without significance on that issue
so long as the defendant's conduct is found to have contributed to the victim's injury. But they are
of great significance in the determination of the extent of defendant's duty; in many cases to extend
his liability to the full extent of the hurt or hurts contributed to by his conduct would throw too
heavy burdens on the courts in other cases, burden enterprise beyond its financial capacity to
respond and still do business, serve no purpose in the distribution of losses or in imposing
precautions against dangers to others, and be morally unjust in imposing penalties out of all

58
GREEN, JUDGE AND JURY 74 (1930) ; Green, supra note 50.
59
See opinion of Justice Smith in Elbert v. City of Saginaw, 363 Mich. 463, 474, 109 N.W.2d 879, 884 (1961), for
significance of duty issue and how, when stated in terms of proximate cause, "we obscure the whole process of
decision, we scramble the functions of judge and jury, and we conceal the critical issues of policy involved
shaping judgment." See also Gronn v. Rogers Constr., Inc., 350 P.2d 1086 (Ore. 1960), for excellent discussion of
the necessity to weigh interests and risks. In this case a contractor was held not liable for loss of minks' young
even though engaged in extra-hazardous blasting near plaintiff's farm.
60
This is graphically indicated in the recent rejection of many important immunities developed during the
1800's. Unborn infant: Keyes v. Construction Serv. Inc., 340 Mass. 633, 165 N.E.2d 912 (1960); Woods v.
Lancet, 303 N.Y. 349, 102 N.E.2d 691 (1951). Charitable hospital: Noel v. Menninger Foundation, 175 Kan.
751, 267 P.2d 934 (1954); Parker v. Port Huron Hospital, 361 Mich. 1, 105 N.W.2d 1 (1960).
Governmental immunity: Muskopf v. Coming Hosp. Dist., 55 Cal. 2d 211, 359 P.2d 457 (1961); Molitor v.
Kaneland Community Unit Dist., 18 Ill. 2d 1, 163 N.E.2d 89 (1959). Michigan has recently followed suit,
Williams v. City of Detroit, 364 Mich. 231, 111 N.W.2d 1 (1961), 60 M1CH. L. REV. 379 (1962). Trespass:
Martin v. Reynolds Metals Co., 221 Ore. 86, 342 P.2d 790 (1959). For numerous other advances, see Green,
supra note 51.
61
RESTATEMENT, TORTS §§ 430-461 (1934) .
415
 proportion to the failure to exercise care in the particular case.62 The courts will go so far and no
farther in ex tending protection to a victim by shifting the loss to a defendant. And few rules can
be formulated to stake out the limits of liability for any particular case. Thus in setting the bounds
of liability, a great English judge has said we must depend upon the "good sense" of the judge,63
or as Holmes and Cardozo put it, "The law does not spread its protection so far."64 Nevertheless
many courts and many writers have attempted to devise specific causation rules and formulas for
this purpose, but they have merely cluttered the books with metaphysical rubbish to confuse
students, advocates, and judges for years to come.
[28] The authors of Causation in the Law do an excellent job of discarding much of this
rubbish, but they also seek to preserve some of it for further use. Their arguments are not without
merit but they are far more relevant to other issues of a case than they are to causal relation. Their
analysis and classification of cases on the basis of necessary conditions,65 conditions causally
irrelevant,66 causally relevant factors and conditions not sine qua non,67 causing harms,68 factors
negativing causal connection,69 voluntary human conduct,70 non-voluntary conduct,71

62
Those were the underlying policies that gave impetus to the development of negligence law during the
nineteenth century; Waube v. Warrington, 216 Wis. 603, 258 N.W. 497 (1935) . See GREEN, op. cit. supra
note 52. The reaction that set in late in that century and continued during the 1900's is indicated in
Green, supra note 51.
63
Denning, L.J., in King v. Phillips, [1953] I Q.B. 429, 442, quoting Lord Wright in Bourhill v. Young,
[1943] A.C. 92, 110.
64
Robins Dry Dock & Repair Co. v. Flint, 275 U.S. 303, 309 (1927); H. R. Moch Co. v. Rensselaer Water Co.,
247 N.Y. 160, 165, 159 N.E. 896, 897 (1928) .
65
HART & HONORE, CAUSATION IN THE LAW 103 (1959). The discussion here develops the difference between
causal relation and policy. Causal relation is considered on the basis of sine qua non in the light of a three-
way classification of conditions. It is believed that the refinements are too subtle and complex for general
usage. In negligence cases on the causal relation issue all the court needs to know about causation is that
defendant's conduct contributed to the injury. As arguments in a particular case the matters considered by
the authors could be quite effective if wrung dry of their sophistication.
66
Id. at 108. The discussion under this section deals with the difficulties that arise when a victim's injury
is attempted to be tied to defendant's negligence rather than to his conduct. The problem is usually whether
the conduct was negligent, or whether the risk was within the scope of defendant's duty. It is most
unfortunate to deal with a problem as one of causal relation when it is not involved.
67
Id. at ll6. The discussion here considers the problem when two or more causal factors are involved, either or
any one of which may h a ve brought about the same injury. In practically all cases numerous contributing
causes can be identified, but the important question in a litigation is whether the defendant's conduct
contributed to the injury. There may be serious questions of duty, negligence, and damages involved, but
the causal relation problem offers no unusual difficulties.
68
Id. at 126-51. The authors here come to grips with the problem of limiting a defendant's liability, even though
he contributed to a victim's injury, because other cause factors also contributed to the injury. They would
limit liability by classifying the other cause factors as voluntary human conduct, non-voluntary conduct,
abnormality, animal behavior, non-voluntary but abnormal conduct, with numerous refinements. They give
numerous examples but seldom do their examples involve a causal relation issue other than the sufficiency
of the evidence to raise such issue as between defendant's con duct and the injury. Practically all the cases
given to support their analysis and classification involve the scope or extent of defendant's duty. The authors
simply refurbish the traditional and largely bankrupt causation doctrines as a basis of limiting liability. They
either ignore or reject the analytical formula requiring a clear formulation of the issues and their allocation
to judge and jury, respectively.
69
Id. at 128.
70
Id. at 129.
71
Id. at 134.
416
 abnormality,72 occasioning harms,73 each with refinements, are entirely too complex to be reliable
where use must be made of them by thousands of practitioners and judges. Their scheme of things is
but another attempt to solve the important issues of a case by tracing and evaluating causes, and would
continue the endless discussions found in the opinions of appellate judges who seek to rationalize their
decisions on the basis of causation metaphysics.
[29] In order to project their re-development program, the authors clear away what may be
called the "slum theories" which in some areas have overgrown the traditional causation concepts.
The "foreseeability" and "risk" theories bear the brunt of their "bull dozing" operations.74 By
identifying the "risk" theory as a variation of the "foreseeability"75 theory they give it a twist that
makes it an easy victim of their attack.76 But there is another and more significant rationale of
the risk theory which divorces it from foreseeability and gives it an independent status. The authors
identify and make way with it as "atomistic" and dependent upon "intuition.''77 It would appear that

72
Id. at 151.
73
Id. at 171. The discussion here is directed to a distinction between causing harm and inducing or
occasioning harm, and becomes quite refined. The authors acknowledge the general tendency to treat the
problems discussed as involving the extent of duties. It may be observed that the traditional causation
doctrines serve no useful function in this area, whatever their values may be in other areas.
74
The terms "slums" and "bulldozing" are not used offensively but only to fill out the redevelopment figure
of speech.
75
HART & HONORE, op. cit. supra note 61, at 230. Much of the authors' discussion of the "foreseeability"
concept as lacking precision or a body of reality is acceptable. They do not seem to give the term its
"functional" value, that is, its useful employment to call forth someone's judgment on the facts of a case,
but in and of itself no test of anything. For this view, see my article, Green, Foreseeability in Negligence
Law, 61 COLUM. L. REV. 1401 (1961).
76
HART & HONORE, op. cit. supra note 61, at 256. For this interpretation of the "risk theory" the authors refer to
Professor Seavey's article, Mr. Justice Cardozo and the Law of Torts, 52 HARV. L. REV. 372 (1939) , and his
and their interpretation of Palsgraf v. Long Island R.R., 248 N.Y. 339, 162 N.E. 99 (1928). My own
interpretation is found in GREEN, op. cit. supra note 54; GREEN, The Palsgraf Case, 30 COLUM. L. REV. 789
(1930). The trouble with Palsgraf is that the issue to which Judge Cardozo is talking is not clearly
formulated. Causal relation in the case is clear. Either the insufficiency of evidence to raise an issue of
negligence with respect to the plaintiff, or the absence of negligence itself with respect to her and the injury
she suffered, seems clear enough to support his judgment. The interpretation given by the authors, and
perhaps Professor Seavey, is that of no duty to her. That there was a duty to Mrs. Palsgraf as a waiting passenger
is clear, but whether defendant's duty included the risk that befell her is seemingly what the authors think is
involved, and that on this basis Judge Cardozo decided that the risk was determined on the basis of
"foreseeability." Despite some of the language in his opinion, that interpretation is not consistent with his
opinions in H. R. Moch Co. v. Rensselaer Water Co., 247 N.Y. 160, 159 N.E.2d 896 (1928) decided the
same year, Wagner v. International Ry., 232 N.Y. 176, 133 N.E. 437 (1921), and Ultramares Corp. v.
Touche, Niven &: Co., 255 N.Y. 170, 174 N.E. 441 (1931), nor in fact, with MacPherson v. Buick Motor Co.,
217 N.Y. 382, 11 N.E. 1050 (1916), and Hynes v. New York Cent. R.R., 231 N.Y. 229, 131 N.E. 898 (1921) .
In the last mentioned cases, Judge Cardozo clearly finds the source of duty in the law. In Palsgraf the
important fact is that the highest appellate court, after all the evidentiary data and arguments were in, and
after consideration of all the factors involved, simply decided that the injury suffered by Mrs. Palsgraf was
not a risk within the scope of any duty owed her as a passenger. If this is an acceptable interpretation, many
more factors were involved than mere "foreseeability." On the other hand, if the court decided that there
was no sufficient evidence to raise an issue of negligence, then "foreseeability" was a highly pertinent
consideration.
77
HART & HONORE, op. cit. supra note 61, at 256. The authors' characterization of the theory of risks advocated by
me for some thirty-five years as "atomistic" and dependent upon "intuition" is merely due to their failure to
comprehend it. Personally, I am greatly flattered by the fact that they accept some of the phases of the
theory explicitly and even more implicitly. Their selection of the two cases as analyzed by me to show the theory's
weakness has merit. Mahoney v. Beatman, in GREEN, op. cit. supra note 54, at 226, could probably be more

417
 they do not fully comprehend this theory and are in too big a hurry to dismiss it. In this theory it
is not the consequences that defendant as a reasonably prudent per son should have foreseen as
"within the risk"78 of his conduct that controls the limitation of his liability, but the risks that the
court, after the conduct has taken place and the injury has been suffered, considers should fall
within the "scope," "ambit," "zone," "radius," or "orbit" of the defendant's duty under the law,
leaving foreseeability as a jury formula for the determination of the violation of duty, the
negligence issue. Under this theory many risks may be actually known to the defendant and still
no liability will be imposed; for under this theory of risks the court may take into consideration any
one or more of the factors catalogued by the authors, plus the economic and moral environments,
the difficulties of administration, the former decisions of the courts, statutes, and any and all other
pertinent factors brought to the court's attention, as they affect the parties litigant as well as the
public good and on this basis the court can determine for what risks, as reflected by defendant's
conduct and the victim's injuries, the defendant should be held responsible-subject, of course, to
the jury's findings on the issues of negligence and damages.79
[30] As I understand the authors it is at this point they fear what they call "intuitive" judgment,
and the consideration of policies on such a wide scale, and would have the court controlled by
rules of causation. I know of no advocate of any theory who would agree that a court should rely
exclusively or even largely on its intuitive judgment in any case, as important as intuition based on
experience and training is in the making of all judgments. The influences that control a judge in
making a decision could never be fully catalogued even by the judge himself. He is bound in
numberless ways more securely than he could be bound by hoops of steel. His environment,
language, learning, experience, sense of justice, associates, constituency, profession, personal
decency, rules of law, procedural restraints, and other influences hold him on his judicial course
and are so binding that he seldom dares to correct bearings he knows to be erroneous until he has
found great support elsewhere. And even when he does dare he often seeks to bring the new
boundary he sets within the confines of antiquity. Moreover, no rules of causation not in accord
with these influences can control his judgment. The only virtue of causation rules is that any
meaning required by these influences can be read into them and the result thus will appear to have
been reached by learned and profound considerations. But I think it safe to say that no rule or
formula can be written that can ever be depended upon to supersede the power of enlightened
judges to meet the responsibility that the environment in its totality imposes upon a court.
[31] It has been observed many times that most cases fall into patterns and are foreclosed on

conventionally treated as a last clear chance case and get the same result. Hines v. Morrow, in GREEN,
RATIONALE OF PROXIMATE CAUSE 21 (1927), could have been easily decided the other way on the ground that
defendant's duty was only owed the state, and was not a basis of civil liability to the plaintiff, as is done in
the snow removal cases, Hanley v. Fireproof Building Co., 107 Neb. 544, 186 N.W. 534 (1922) . Since the
state is under no liability to travelers, the statute was doubtless enacted to relieve the state of the financial
burden of keeping the highway in repair. At any rate, the defendant did not explore the defenses available but
relied on proximate cause to his undoing, inasmuch as the plaintiff's advocate interpreted proximate cause to
mean causal relation and the court accepted this interpretation. I use the case in my teaching to indicate how
an advocate can himself become entrapped in some cause doctrine set to trap his opponent.
78
HART & HONORE, op. cit. supra note 61, at 257. The authors argue that the phrase "within the risk" has both a
restrictive and extensive aspect. This would seem to make it a synonym for "duty" when in fact it is more
nearly a synonym for "harm." Whether a risk or harm falls within the scope of a defendant's duty would seem
to make much more sense than to say whether "the harm is within the risk.'' On their interpretation of the
"risk theory" the authors can claim an easy victory.
79 See FLEMING, TORTS (2d ed. 1961) . Much is found in this excellent book to support the views expressed in
this article but the author would doubtless stop considerably short of the views here expressed. See
particularly id., ch. 9, "Remoteness of Damage," at 178.

418
the duty issue before they reach the courthouse; others fall victim to some procedural or factual
detail; a few require extended consideration on the law. The danger is not that these few will set
new boundaries for the law but rather that they will be tailored to some pattern that makes their
decision grotesque. It is here it would seem that litigants, profession, and everyone would hope
that the court is not so restricted that it cannot give a case fresh thought. And of necessity it must
if the law is to serve its functions. The law "front," much as is true of the weather, is always on
the move; a thousand cases each year mark its fluctuations and movement-an inch here, a foot
there, a step backwards perhaps, and now and then some breakthrough; a new front forms without
the slightest aid of causation rules-perhaps in spite of them.80 The most that rules and formulas
can do in this eternal turbulence of the law is to aid the judge's guidance, and no one would
discount them for that purpose, and equally so no one should desire them to create an impenetrable
intellectual barrier for him.81
B. The Negligence Issue
[32] It must not be overlooked that even though the defendant's conduct may have contributed
to the victim's injury, and even though the court has ruled that the risk of injury was within the
scope of defendant's duty to the victim, another and frequently a more serious hurdle must be
successfully surmounted in maintaining the plaintiff's case. Did the defendant violate his duty to
the victim, i.e., was his conduct negligent? Here both judge and jury have important functions, and
both functions may rest heavily on the "foreseeability" formula. The sufficiency of evidence to

80
Vivid examples are found in Battalla v. State, 10 N.Y.2d 239, 176 N.E.2d 729, 219 N.Y.S.2d 34 (1961),
renouncing the long-standing New York doctrine of no liability for injury due to fright without physical impact,
and in Keyes v. Construction Serv., Inc., 340 Mass. 633, 165 N.E. 912 (1960), renouncing the long-standing
Massachusetts doctrine denying recovery for physical injury to an unborn infant. See also the examples in
note 56 supra.
81
In order to demonstrate how the duty concept relieves the causation concept of most of its burdens and
confusing terminology, let us assume a difficult case suggested by my associate Professor Wayne Thode. D
sold a car to X knowing it had bad brakes but did not disclose the fact to X. X had a rear-end collision
with P's car due to X's failure to stop in time at a red light. X did not attempt to use the brakes, but his
speed was such that had the brakes been good and he had used them the collision would not have been
avoided. Both P and X suffered injuries. P sued D on the ground that D was negligent in selling the car to
X without disclosing to him that the brakes were bad. D defended on the ground that if the brakes had been
good the collision would not have been avoided because of X's speed and failure to make use of the brakes.
D's selling the car to X contributed to P's injuries. D was under a duty to P and to X and other users of
the highway to refrain from selling the car to X knowing it had defective brakes without disclosing that fact
to X. The policy of the law is to reduce the risks of highway collisions by applying pressure on those
who create the risks.
X's buying and driving the car contributed to P's injury. X owed no duty to anyone to use the defective
brakes, and did not violate any duty to anyone in failing to use them.
Was the risk of driving at a speed which would not permit X to avoid collision with P's car within the
scope of D's duty to X and to P, in selling the car to X without dis closing the fact that the car had defective
brakes? Would X have bought and driven the car if D had disclosed to him that the brakes were defective?
Under the evidence this could be a jury issue. In the absence of a finding, whether by court or jury, that
X would have purchased and driven the car knowing the brakes were defective, it cannot be presumed that
he would have risked his own safety and that of other users of the highway by driving a car with
defective brakes. The risk of driving the car on the highway, irrespective of the speed with which it was
driven, falls within the scope of D's duty to X and to P with respect to the injuries suffered by them.
It may be observed that by slight variations in the facts other problems would be raised; that an action
by X against D could be defeated without defeating P's action against D, or that D and X might be held as
joint tortfeasors. The more difficult the problem the more bewildering the attempt to find a solution by use of
causation doctrines. In fact no acceptable solution in any difficult tort case can be found other than on the
basis of duties and risks, and then the translation of the results into "causes" is unnecessary.
419
raise the issue of negligence may be as puzzling as in the determination of the sufficiency of the
evidence to raise the causal relation issue. It is not infrequently confused with the scope of duty
issue. But more frequently, and especially by appellate courts, it is seemingly confused with the
substantive negligence issue. In numerous cases the courts have said the defendant, as a reasonably
prudent person under all the circumstances of his conduct, could not have foreseen any such risk
or harm that befell the victim, when all they meant was that the evidence was insufficient to raise
the issue of negligence. In many cases this is expressed in terms of no proximate cause as a matter
of law,82 the proximate cause formula itself being only a rephrasing of the foreseeability negligence
formula. There may be numerous other factors involved but the foresee ability of harm bulks large
at this step of the litigation process. And even though the issue of negligence is submitted to the
jury, there is no assurance that their verdict will be favorable to the plaintiff.
[33] The formula for submitting the issue of negligence is comprehensive. It is designed to ascertain
from a layman's standpoint whether defendant's conduct should be condemned and the victim
compensated. In absence of a jury the judge must perform this function presumably governed by the
same formula. The formula presents two inquiries: (I) should the defendant as an ordinarily prudent
person, under all the circumstances of his conduct, have reasonably foreseen some hurt to the
victim of the same general nature as he suffered, and (2) did the defendant fail to exercise
reasonable care to avoid the hurt? It calls for the evaluation of defendant's conduct in all of its
environmental details as reflected by the evidence. The abstract standard is the foresight of the
ordinarily prudent person, both transparent fictions, but fictions well designed to focus attention
upon the reasonableness of the defendant's conduct in the particular environment, without
obscuring the factual data on which the jury must give their verdict.83 Further, it is important to
notice that the ultimate judgment is, did defendant exercise reasonable care to avoid the hurt?84
[34] It is at this point that the "acts" and "omissions," by supplying the details of the defendant's
conduct, may, and usually do, become important. The negligence concept by its very terms is "failure
to exercise care at some stage of an undertaking assumed by the defendant." Whether the conduct
is characterized by positive action or by doing nothing to avoid the consequences, it is still but a
phase of the affirmative undertaking. "Acts" and "omissions" are of no significance other than to indicate
the details of conduct which resulted in the failure to exercise care.
[35] In the determination of the negligence issue under the com prehensive negligence formula,
the jury consider the import of the same evidence that the judge considered in imposing a duty.
But in contrast to those of the judge, their considerations are restricted to the parties to the litigation,
the environment of their conduct, and the justice of compensating the plaintiff at the expense of the
defendant. They are not concerned with the administration of future cases between other litigants,
economic burdens on enterprise in general, and other such policy considerations. Nevertheless, the
jury are faced with the multitude of other cause factors that played a part in the injury of the victim.
It is their function to find whether under all the circumstances (including this multitude of cause
factors) the defendant's conduct was negligent and whether he should be penalized for his conduct
82
A group of Illinois cases indicate how "proximate cause" is utilized for defeating liability though causal
relation and duty are clear and negligence has been determined by the jury. Merlo v. Public Serv. Co., 381
Ill. 300, 45 N.E.2d 665 (1942); Berg v. New York Cent. R.R., 391 Ill. 52, 62 N.E.2d 676 (1945); Seith v.
Commonwealth Elec. Co., 241 Ill. 252, 89 N.E. 425 (1909). But see Bennett v. New York &: Queens Elec.
Light &: Power Co., 294 N.Y. 334, 62 N.E.2d 219 (1945) (similar facts). Incredible as the results in the Illinois
cases may seem, their influence in other cases has been great.
83
See Green, Foreseeability in Negligence Law, 61 COLUM. L. REV. 1401 (1961), for extended discussion.
84
"Natural and probable consequences" is the phrase frequently used to mark the foreseeability of the danger
which a defendant must use care to avoid. The negligence formula does not require "probability." Even
though there is only a "possibility" yet if the danger is such that the "ordinarily prudent person" would not
take the risk, a defendant is negligent if he does not use care to avoid it. See note 37 supra.
420
or whether plaintiff should have to bear his loss without compensation. Here again the courts have
evolved numerous rules and formulas to guide, if not to govern, the jury's judgment.85 It may be that
courts have so greatly overdone their function of instructing the jury on this issue as practically to
destroy the basic theory of jury trial. Perhaps under the pressures of advocates they have so proliferated
and refined their instructions that juries can only vaguely, if at all, comprehend their meanings with
the result that in many cases juries are left without effective guidance and are freed to do what they
think should be done, if they know how to do it without committing reversible error. This issue is the
heart of a negligence case and if fairly and clearly submitted, the jury's response will in most
cases be intelligent and acceptable.
[36] A good example of how the negligence formula can be fouled by the useless introduction
of causation doctrines is found in the recent case of Leposki v. Railway Express Agency, Inc.86
Defend ant's driver parked his truck uphill but pitched in toward the curb so that the intake pipe
permitted gasoline to drip into the gutter. During the driver's absence two young boys saw the
gasoline and one dropped a lighted match in the gutter beneath the pipe and the gasoline caught
fire. Being unable to extinguish the fire, the driver called the fire department and moved the truck
to a position in front of plaintiff's home. Flames and gasoline spread to the house, causing property
damage and personal injuries to the plaintiffs for which they instituted suit.
[37] One of the defenses was that the boy's act in igniting the gasoline constituted an intervening
act that superseded defendant's negligence. On the basis of an inadequate instruction on
foreseeability the jury found for defendant. The court of appeals on the basis of the Restatement
of Torts reversed, saying in part: "The act is superseding only if it was so extraordinary as not to
have been reasonably foreseeable. . . . [T]he events are viewed retrospectively and not
prospectively.... Whether an intervening act constitutes a superseding cause is a question that is
more readily resolved in hindsight, and that which appears to be extraordinary in the abstract may
prove to be otherwise when considered in light of surrounding circumstances that existed at the
time of the accident." Then follows a detailed statement of what the jury should have been told to
consider retrospectively, concluding with this statement: "Considering the intervening act in light
of the circumstances that prevailed at the time defendant parked the truck, can it be said that the
boy's act was extraordinary?"
[38] Several observations are pertinent. (1) Nearly all state courts would be precluded by the
comment on the weight of the evidence rule from giving such instructions as the court here said
should have been given.87 (2) The rules of causation developed by the Restatement are legitimate
only by way of educating the judge with respect to his functions in the determination of the
risks that may fall within the scope of defendant's duty, and the sufficiency of the evidence.88 Even
for these purposes they fall far short and may be quite misleading and confusing. They introduce
several types of "causes" where other factors are more significant and far more reliable as a basis

85
For an excellent discussion of how this function can be overdone, see the opinion of Justice Shinn in
Werkman v. Howard Zink Corp., 97 Cal. App. 2d 418, 218 P.2d 43 (1950) (hearing denied). See alsoꞏ
Socony Vacuum Oil Co. v. Marshall, 222 F.2d 604 (1st Cir. 1955); Mosley v. Arden Farms Co., 26 Cal.
2d 213, 157 P.2d 372 (1945) (Traynor, J.); Green v. Keenan, 10 Ill. App. 2d 53, 134 N.E.2d 115 (1956). A
jury's reaction is indicated in Kindle v. Armstrong Packing Co., 103 S.W.2d 471 (Tex. Civ. App. 1937).
86
297 F.2d 849 (3d Cir. 1962) .
87
James, Sufficiency of the Evidence and Jury-Control Devices Available Before Verdict, 47 VA. L. REV.
218, 237 (1961); Wright, Adequacy of Instructions to the Jury, 53 MICH. L. REV. 505, 509 (1955) . The
Restatement's proliferation of causation rules as used in many cases provides the judge with further controls
of the jury without seeming to violate the rule against commenting on the weight of the evidence.
88
See RESTATEMENT, TORTS §§ 430-462 (1934). No one can exaggerate the burden placed on the
administration of negligence law by the attempt to work out detailed rules of causation by which liability
can be determined in complex cases.
421
of judgment. In the case before the court causal relation was so clear as to raise no issue, and so
was defendant's duty. Running the case through the complicated coils of causation threw no light
on defendant's duty. Defendant was using in its business a highly dangerous explosive on the
highway near plaintiff's house against which plaintiffs could provide no protection. Indeed, after
the driver knew the truck was on fire he moved it in front of the plaintiff's house. The risk imposed
was clearly within the scope of defendant's duty, and it only remained for the jury to say whether
defendant's conduct was negligent. (3) The Restatement rules of causation are far too
complicated to be understood by jurors in their determination of the negligence issue. Perhaps not
more than ten percent of the profession, including the judges, who administer negligence law
can employ them so as to avoid entanglements and conflicts. Even the seemingly simple line
between "ordinary" and "extraordinary" depends so largely on subjective reactions that it cannot
be said to have stability. (4) The easily stated but comprehensive negligence formula discussed
above indicates to jurors as clearly as can be done what they are called upon to determine without
obscuring the factual data they must consider. The refinements of causation can only blur and
bewilder. "Under the surrounding circumstances" gives the jury a wide leeway for judgment and
equal leeway to the advocate's argument. Moreover, the formula centers the inquiry where it
should be centered; in the case here on the reasonableness of defendant's conduct in failing to
avoid injury to the victims in view of the danger imposed on nearby residents and their property.
The formula is given in part in terms of "foreseeability," but no one could ever think that the jury
could do more than use their hindsight in considering whether the defendant was negligent in view
of all the circumstances which were shown to have surrounded its conduct.89 The well-known
habit of young boys to play with matches and start fires was simply one of the circumstances that
gave color to defendant's conduct.
C. The Damage Issue
[39] Let us assume that in the particular case the jury have found causal relation between
defendant's conduct and the victim's hurt; that the judge has found a basis in law for imposing a
duty on the defendant as to the particular risks or injuries suffered by the victim; and that the jury
have found that defendant was negligent as to these injuries; there is still a very important issue to
be resolved. What losses has the victim suffered and how much in dollars and cents should he be
compensated? The great latitude this issue gives the jury for reconciling what may seem to be
doubtful if not unjustified findings on other issues is sometimes overlooked. Theoretically, if the
other issues are found favorably to the plaintiff, under the law, he should be compensated for the
full amount of any losses within the coverage of the duty which the defendant has violated.90 In
courthouse history full compensation is rarely given. Here again the rules and formulas are
numerous by which a judge may guide or control the jury's award. But with all the controls a court
may exert, the jury have a wide latitude for judgment, and so long as the amount is not outrageously
large or small it will usually be permitted to stand in absence of some overt misconduct of the jury
or in absence of some vital error committed in the trial process.91
[40] As a matter of fact as well as theory, the whole case, with all the details of the conduct
89
The writer has dealt more fully with the "foreseeability" formula in Green, Foreseeability in
Negligence Law, 61 COLUM. L. REV. 1401, 1417 (1961).
90
The duty a defendant has violated frequently does not cover all the risks and losses a plaintiff has
incurred. Many losses will be eliminated at some preliminary step as on pre-trial, by motion, or in the
exclusion of evidence. See GREEN, RATIONALE OF PROXIMATE CAUSE 132-95 (1927) .
91
The extremes are indicated in Vaughn v. Menlove, 7 Car. & P. 525, 173 Eng. Rep. 232 (C. P. 1837) where
the plaintiff lost several cottages by fire, value £500, but had a jury verdict for only £,5, and in Seifert v.
Los Angeles Transit Lines, 364 P.2d 337, 15 Cal. Rptr. 161 (1961), where $134,000 was awarded for pain
and suffering based on some $53,000 economic loss. Traynor, J. dissented. Missouri-Kan.-Texas R.R. v.
Edwards, 361 P.2d 459 (Okla. 1961) ($650,000 award reduced to $278,793).
422
of all the parties, all the cause factors, all the needs and hardships, are considerations that may find
lodgment in the minds of twelve men and influence them to reach a common conclusion despite
the wide differences in. notions and vagaries that may possess them. The process of reconciling
their differences, and reconciling causes, duties, and the conduct of the parties through
compromise, to the end that the defendant shall not be made to bear the full loss the victim may
have suffered is doubtless the strongest bastion of jury trial. There can be no division of liability or
damages on a basis of causal relation alone. If any division can be made successfully, which is at
least doubtful in complex cases, it must be made on a comparison of so-called "faults"-duties and
violation of duties-and whether the function is performed by judge or jury its exercise may take
into account a multitude of factors with at best only an approximation of justice. It is believed
that what juries do here, and perhaps do better out of their own sense of justice if not too severely
policed by the judge, is what the advocates of comparative negligence would have them do by
formal instructions.92 For it is here that the jury may greatly modify and assuage any seeming
severity reflected in their findings on the issues of causation and negligence in the victim's favor,
and in the rejection of the defenses and excuses offered by the defendant. It may be that many
judges, and especially appellate judges in their examination of a paper record do not trust the
reaction of a jury, but if they sat where the jurors sat, with all the limitations and restrictions
placed on juries, their own reactions would probably be very similar.

92
See Maloney, From Contributory to Comparative Negligence: A Needed Law Reform, 11 FLA. L. REV.
135 (1958) .
423