INTENTIONAL HARMS

RICHARD A. EPSTEIN*

INTRODUCTION

T HROUGHOUT the history of our law, intentional harms have formed a cate-
gory of wrongs separate and apart from the rest of the law of tort. The reason
for treating intentional harms differently from accidental ones is not difficult
to see. With the intentional infliction of harms, it is not necessary to decide
which of two innocent persons should be requiredto bear the loss in question.
The element of intention makes the case an easy one, by allowing the loss to
be placed upon the person who willed it, upon the person who is "bad" in the
strongest sense of the word. The importance of mens rea in the criminal law
reinforces that dominant impression: the man who harms another and who
does so intentionally is in general a person worthy of punishment, and if he
is worthy of punishment, why should he not be required to pay the injured
party for the harm which he has wilfully inflicted? The theme of special
treatment for deliberate harms also echoes in other corners of the law. Thus
proposals for schemes of no-fault insurance to compensate victims of traffic
accidents often exclude deliberate harms from their coverage.' And the work-
men's compensation law, although designed to remove the "fault principle"
from the field of industrial accidents, likewise gives special attention to the
deliberate infliction of harm.2
There is also popular and intuitive support for a special treatment of de-
liberate harms. We do not need utilitarian analysis to decide that it is generally
* Professor of Law, University of Chicago. I wish to thank Walter
J. Blum, Robert C.
Ellickson, John H. Langbein, and J. Jerry Wiley for their helpful comments on earlier
drafts of this paper. I should also like to thank the American Bar Foundation for the
support it has given for my work. The views here are of course my own, not those of
the Foundation.
1 The isolation of intentional harms has been clear since the first definitive proposal
for compulsory insurance, the Columbia Plan, was published in 1932. Columbia Univ.,
Council for Research in the Social Sciences, Comm. to Study Compensation for Auto-
mobile Accidents, Report 237 (1932). See also Robert E. Keeton & Jeffrey O'Connell,
Basic Protection for the Traffic Victim: A Blueprint for Reforming Automobile In-
surance 304-05 (1965).
2 Thus the "wilful misconduct" rule limits
recovery in the event of the deliberate
violation of a safety rule by an employee. See generally Arthur Larson, The Law of
Workman's Compensation ? 32 (1973).

391

This content downloaded from 129.127.145.240 on May 08, 2018 20:13:09 PM

All use subject to University of Chicago Press Terms and Conditions (http://www.journals.uchicago.edu/t-and-c).
392 THE JOURNAL OF LEGAL STUDIES

a bad thing for one person deliberately to maim or kill another, or to take or
destroy his property.3 What offends the moral sense is that one person
deliberately uses his own power to deprive another of the very rights which
he claims and defends for himself. The objection is not to the costs and
benefits of the thing, but to the willingness of one person to exercise his
dominance over another without his consent. The belief that every person is
entitled to be free of purposive invasions of his person or property need not
(and cannot) be reduced to lower terms.
The importance that is attached to deliberate harms makes it necessary to
determine what harms are deliberate and what harms are not. In the easy
case, the harm in question is the specific end desired by the person who in-
flicts it. But the category of deliberate harms extends beyond this case to
two other situations. In the first of these the actor has a substantial or certain
knowledge that one consequenceof his conduct will be the harm in question,
even though he might not wish for it to occur. In the second, the actor knows
that his acts might well cause harm, but is recklessly or consciously indifferent
to their consequences.There are some difficult cases of degree at the edge of
both of these categories, for example, where foreseeability of harm shades
into knowledgethat harm will occur as a consequenceof certain acts. But even
within these gray areas, the basic distinction between deliberate and accidental
harms retains much of its original force. A manufacturermay know from ex-
perience that a certain fraction of his output will be dangerouslydefective, but
it does not follow that he intended to cause the harm resulting from a particu-
lar defective unit where he did not know that it was defective before he
allowed it out of his control.4
It is possible, then, to identify a class of deliberate harms. The question I
address in this paper is how the law should treat that class of harms. The
law of tort has from its earliest times sought to harmonize three distinct
legal theories of liability for the infliction of harm: strict liability, negligence,
and deliberate harms. While the theory of deliberate harm is the easiest to
accept, its very acceptance raises serious questions about the total structure

3 An attempt to explain in utilitarian terms the law's condemnation of deliberate harms

may be found in Guido Calabresi & A. Douglas Melamed, Property Rules, Liability
Rules, and Inalienability: One View of the Cathedral, 85 Harv. L. Rev. 1089, 1125-27
(1972); Richard A. Posner, Economic Analysis of Law 67-68, 358 (1973).
4 But see Richard A. Posner, supra note 3, at 66: "Most accidental injuries are in-
tentional in the sense that the injurer knew that he could have reduced the probability
of the accident by taking additional precautions. The element of intention is unmistakable
when the tortfeasor is an enterprise which can predict from past experience that it will
inflict a certain number of accidental injuries every year." Nonetheless, defective products
cases always have been regarded as concerned with accidents, and not intentional harms,
no matter whether liability has been bottomed on negligence or strict causal principles.
The distinction between accidental and deliberate harm may for some purposes be
immaterial, but it is not unintelligible.

This content downloaded from 129.127.145.240 on May 08, 2018 20:13:09 PM

All use subject to University of Chicago Press Terms and Conditions (http://www.journals.uchicago.edu/t-and-c).
 INTENTIONAL HARMS 393

of the law of tort once we shift our focus from the particular case to the
general theory that decides it. If the element of intention is crucial to plaintiff's
case, why should he ever recover for accidental harm? Why is a suit for de-
liberate harms not only the easiest case in tort but the only case as well? Or
to reverse the point, if we allow recovery under any circumstances for acci-
dental harms, why should the plaintiff ever be requiredboth to plead and to
prove that the defendant intended to harm him? What intelligible grounds
are there for having three theories of tort instead of a single theory of tort?
In attempting to answer these questions, I shall employ the same analytical
frameworkthat I have developed for accident cases.5 Briefly, that framework
assumes that at the first round of pleading in a law suit the plaintiff must give
some good reason for upsetting the initial balance in favor of the defendant.
The reasons that he gives, even if sufficient to state a cause of action (and
here we are concernedwith formal structure, not with what reasons will count
as sufficientin the given case), do not conclude the lawsuit in the plaintiff's
favor. They only create a substantive presumption (the word having in this
context no evidentiary importance) in favor of recovery. One responseopen to
the defendant is to concede the strength of the prima facie case but to allege
in avoidance good reasons of his own to restore, in whole or in part, the
original balance. These reasons, like the plaintiff's, again do not conclude the
case, but only raise a presumptionthat the defendant has defeated or dimin-
ished the plaintiff's claim. The plaintiff can revive his case by alleging at a
third stage of argument new matter which, though immaterial to the prima
facie case, now works to reestablish his claim in whole or in part. The process
of new pleas continues until one of the parties chooses to join issue on a
question of law or fact. The use of this system with its multiple stages both
allows and requires us to establish the appropriaterelationships among those
concepts regardedas relevant to the tort law, and to do so in a manner which
makes clear the role of the intention to harm.
With this formal frameworkfirmly in mind, the overall plan is as follows.
In the first three parts of this paper, I examine the way in which deliberate
harms are treated in physical injury cases. In the first part I deal with the
explication of the concept of intention to harm offered in the tort law, and,
briefly, with its relationship to the criminal doctrine of mens rea. In it I
conclude that the existing tort law extends the concept of intention to harm
far beyond its proper limits, and in so doing robs it of the explanatory power

5 See Richard A. Epstein, Pleading and Presumptions, 40 U. Chi. L. Rev. 556 (1973)
[hereinafter cited as Epstein, Pleadings and Presumptions], for a more detailed account
of the method. For its application, see Richard A. Epstein, A Theory of Strict Liability, 2
J. Leg. Studies 151 (1973) [hereinafter cited as Epstein, Strict Liability], and Richard A.
Epstein, Defenses and Subsequent Pleas in a System of Strict Liability, 3 J. Leg. Studies
165 (1974) [hereinafter cited as Epstein, Defenses].

This content downloaded from 129.127.145.240 on May 08, 2018 20:13:09 PM

All use subject to University of Chicago Press Terms and Conditions (http://www.journals.uchicago.edu/t-and-c).
394 THE JOURNAL OF LEGAL STUDIES

that it might otherwise have. In the second part I review the causal theories
of strict liability and then proposea set of rules that allows us to accommodate
both accidental and intentional harms (now properly defined) within the
frameworkof a single theory. The essence of the accommodationis that the
defendant's intention to harm, while insufficient to state a prima facie case,
is nonetheless sufficientat the third stage of argument to overrideany affirma-
tive defense that the defendant might properly raise to allegations of
merely accidental harm. After noting the consequences of treating the in-
tention to harm in this manner, I examine in the third part of the paper
some of the most common excuses and justifications offered for the infliction
of intentional harm, with special attention to necessity, self-defense, and
consent. Each of these issues can be raised in the fourth round of the pleadings,
and each is in principle subject to important qualifications which can be set
out in subsequent rounds.
Once the role of intention to harm is established in physical injury cases, I
then considerits place in connection with economic losses apart from physical
injury. In this context, I try to show in the fourth part of the paper that
it is a mistake to adopt the common approach that creates a prima facie tort
for the intentional infliction of economic harm, subject to a justification based
upon the defendant's self-interest, applicable except when he behaves mali-
ciously. Instead I conclude that the general pattern of arguments raised in
physical injury cases requireswith only limited exceptions that there shall be
no action for the infliction of economic losses no matter what the defendant's
mental state. With both physical and economic harms thus examined, it is
then possible to conclude that the intention to harm has a secure place within
a properly formulatedlaw of torts, but one subordinateto the places occupied
by liberty, property, and causation.

I. Mens Rea AND THE INTENTION TO HuARM

The current treatment of intentional harms derives its moral force from
the fact that the tort law demands compensation only for the very acts for
which the criminal law demands punishment. With that basis for judgment
we might expect the treatment of the intentional element in the intentional
tort to draw its inspiration from the account of mens rea developed with
such patience in the criminal law. As a general proposition, a person will be
held criminally responsibleonly for consequenceshe intended to bring about.
The doctrine is softened, to be sure, by taking into account the closely related
concepts of recklessness and substantial knowledge of harm."But even with

6 For a much more detailed account of mens rea, see generally H. L. A. Hart, Punish-
ment and Responsibility chs. 5 & 6 (1968). The doctrine is qualified, and rightly so, to
the extent that if a person intends some grievous bodily harm and commits some greater

This content downloaded from 129.127.145.240 on May 08, 2018 20:13:09 PM

All use subject to University of Chicago Press Terms and Conditions (http://www.journals.uchicago.edu/t-and-c).
 INTENTIONAL HARMS 395

these modifications,mens rea operates as a limitation on criminal responsibility

that is independentof causation. Thus if a man accidentally sets a ship on fire
when he lights a match in the course of a theft of rum, he has committed theft,
but not arson.7Likewise, if he throws a stone at another person with whom
he has a running fight, and breaks a window of a nearby building, he cannot
be held criminally for the destruction of the window that he did not intend
to break.8 In both cases the accused caused the harm in question; only the
doctrine of mens rea stands between him and criminal responsibility.
Within the tort law, however, the mental element does not work to restrict
liability otherwise called for on causal grounds, but operates in quite a differ-
ent manner. Consider Vosburg v. Putney,9 a case of assault and battery
brought by one schoolboy against another, for damages inflicted by a kick
just below the knee. At the time of the kick, the plaintiff had nearly, but not
completely, recovered from a prior injury. Given the weakened condition of
his leg, the defendant's kick, which otherwise would have been harmless,
caused the plaintiff permanent injury, for which the jury assessed damages
of $2500. One point of law before the Wisconsin court concerned the precise
nature of the defendant's mental state necessary to make out the plaintiff's
prima facie case of assault and battery. If the criminal standard of mens rea
controls in the tort context, then the plaintiff fails to make out a
cause of action. The jury found by special verdict that the defendant, in
kicking the plaintiff with his foot, intended to do him no harm. The court,
however, held that the action could be maintained even if it were shown only
that the intention to commit the act was "unlawful." Moreover, it ruled that
"if the intended act is unlawful, the intention to commit it is necessarily un-
lawful."10The sentence is enigmatic, but the most sensible interpretationthat
can be impressed on it is that, as it is prima facie criminal to touch someone
else deliberately, it is prima facie tortious to do that same thing.""Yet once
harm than he intends, the doctrine of mens rea will not shield him from the consequences.
"If a man once begins attacking the human body in such a way, he must take the
consequences if he goes farther than he intended when he began." Reg. v. Serne, 16
Cox Crim. Cas. 311, 313 (1887). One exception to the general rule that has enjoyed
support holds that the accused who kills another person in the course of the commission
of a felony will be held for murder even if he did not intend to kill his victim. That
doctrine represents an uncomfortable departure from the traditional requirements of
mens rea, and has been effectively criticized on that ground. See, e.g., Thomas B. Ma-
cauley, A Penal Code Prepared by the Indian Law Commissioners, Note M, 64-65
(1837), as cited in Monrad G. Paulsen & Sanford H. Kadish, Criminal Law and Its
Processes 226 (1962); Report of the Royal Commission on Capital Punishment, 1949-
1953, 36 (1953).
7 Reg. v. Faulkner, 13 Cox Crim. Cas. 550 (1877).
8 Reg. v. Pembliton, 12 Cox Crim. Cas. 607 (1874).
9 80 Wis. 523, 50 N.W. 403 (1891).
to Id. at 527, 50 N.W. at 403.
11 There is an alternative account of the opinion whose logic is even more difficult to

This content downloaded from 129.127.145.240 on May 08, 2018 20:13:09 PM

All use subject to University of Chicago Press Terms and Conditions (http://www.journals.uchicago.edu/t-and-c).
396 THE JOURNAL OF LEGAL STUDIES

the proposition is put in that fashion, the cleavage between the tort and
criminal law on the question of intention becomes apparent, as the doctrine
of mens rea insulates the defendant from punishment for aggravated assault.
The same cleavage is found in Mohr v. Williams.'2 There the defendant
performedan operation on the plaintiff's left ear after he received the plain-
tiff's permission to perform a similar operation on the right ear. Under
anesthesia it turned out that the left ear was more diseased than the right.
The defendant made the medical judgment to tend first to the left ear and
performed an operation on it that was both skilfully and successfully done.
Again the legal issue concerned the precise mental state of the defendant
essential to the plaintiff's prima facie case. The court held that the above
facts disclosed a good cause of action for assault and battery, even with the
total lack of evidence of an "evil" intent. The way in which the court finessed
the requirement of deliberate harm is, moreover, of great importance.
[T]he act of defendantamountedat least to a technicalassault and battery. If
the operationwas performedwithoutplaintiff'sconsent,and the circumstances were
not such as to justify its performancewithout, it was wrongful; and if it was
wrongful,it was unlawful.. . [A]ny unlawfulor unauthorizedtouchingof the
person of another,except it be in the spirit of pleasantry,constitutesan assault
and battery. In the case at bar, as we have alreadyseen, the questionwhether
defendant'sact in performingthe operationupon plaintiffwas authorizedwas a
questionfor the jury to determine.If it was unauthorized,
then it was, withinwhat
we have said, unlawful.It was a violent assault,not a mere pleasantry;and, even
thoughno negligenceis shown,it was wrongfuland unlawful.The case is unlikea
criminalprosecutionfor assaultand battery,for there an unlawfulintent must be
shown. But that rule does not apply to a civil action, to maintainwhich it is
sufficientto showthat the assaultcomplainedof was wrongfuland unlawfulor the
result of negligence .... Vosburg v. Putney, 80 Wis. 523, 50 N.W. 403.13

The court appears to have read the intentional element out of the intentional
tort. In its stead we find the pale requirementof "unlawfulness,"here taken
to show that the defendant has no good affirmativedefense to a prima facie
case that can easily be made strict in form-to wit, you cut into my left ear.14

follow. The defendantkicked the plaintiff after the class had been called to order, and
the defendant's acts could be regarded as unlawful because they were in defiance of
a school rule. Yet here the violation of the school rule is quite complete whether or not
the defendant hurt the plaintiff. In either case (harm or no harm), school discipline is the
appropriate sanction for the violation of the school rule. The reliance on that rule, more-
over, does not make the harm intentional. It does, however, make it difficult to decide
the case where the injury is inflicted in the classroom seconds before the teacher calls
the class to order.
1295 Minn. 261, 104 N.W. 12 (1905).
13Id. at 270-71, 104 N.W. at 16.
14 On the distinction between ultimate issues of fact and conclusions of law see Epstein,

This content downloaded from 129.127.145.240 on May 08, 2018 20:13:09 PM

All use subject to University of Chicago Press Terms and Conditions (http://www.journals.uchicago.edu/t-and-c).
 INTENTIONAL HARMS 397

Yet even that view of the case is not uniformly followed, for the court
makes explicit reference to the defendant's mental state by noting that there
can be no assault and battery where the touching is done in "a spirit of pleas-
antry." It then treats all touchings that are not so motivated as though they
were "violent," even though done with the intention to help the plaintiff.
That logic could not be defended if the mens rea requirement were taken
seriously. The defendant only meant to perform an operation knowing that
he did not have the requisite consent; by no stretch of the imagination did
he intend to harm the patient in a manner that would make him criminally
responsible for any harm that might ensue.
In Garratt v. Dailey15 the defendant was a five-year-old child who pulled
out the chair of the plaintiff, an elderly and arthritic woman, just as she was
about to sit down. The woman fell to the ground and suffered serious injury.
If the case were tried on the theory that the defendant deliberately inflicted
an injury upon the plaintiff, then recovery should be barred; the defendant
did not intend to hurt the plaintiff and did not have specific knowledge that
such harm was likely to result. Nonetheless, the court was prepared to let
the plaintiff recover if she could show that defendant's act was done "with
the intent of causing the plaintiff's bodily contact with the ground."'6Again,
the defendant is held liable in tort for consequences for which he is not
criminally responsible.
The last case of this sort to be considered involves the doctrine of "trans-
ferred intent," by which the defendant in a tort action will be held prima
facie responsiblefor the harm he has inflicted upon the plaintiff by virtue of
the fact that he intended to harm someone else. In Talmage v. Smith17 the
defendant threw a stick at some boys, intending to frighten them off his land.
He was held liable when his stick put out the eye of the plaintiff who was on
the defendant's roof with the other boys, though unseen by the defendant. If
we treat the criminal law as the appropriatemodel for decision, there should
be no recovery at all because the particular harm caused was wholly unin-
tended. At worst, the defendant would be subject to prosecution for the
attempted assault on the boys at whom he threw the stick. Instead the court
allowed the plaintiff's action for the full damages because the defendant would
have committed an intentional tort if he had struck one of the boys whom
he intended to frighten.

Pleadings and Presumptions 561-66. Each of the substantive pleas considered in this
paper satisfy all of the formal tests for ultimate issues of fact.
1546 Wash. 2d 197, 279 P.2d 1091 (1955).
16 That point of view is hinted at in Charles
O. Gregory & Harry Kalven, Jr., Cases
and Materials on Torts 23 (1969), where the authors note that the defendant "hit her
with the ground," a heroic attempt to force the case into the trespass mold.
17 101 Mich. 370, 59 N.W. 656 (1894).

This content downloaded from 129.127.145.240 on May 08, 2018 20:13:09 PM

All use subject to University of Chicago Press Terms and Conditions (http://www.journals.uchicago.edu/t-and-c).
398 THE JOURNAL OF LEGAL STUDIES

II. ACCIDENTS AND INTENTIONAL HARMS

Before we can solve the puzzles raised by these four cases, we must answer
the questions already posed with respect to the relationship of intentional
harms to accidents. That task in turn requires the development of a unified
theory of tort. I believe that it is possible to develop such a theory, but in
order to do so I must retrace ground that I have canvassed in two recent
articles devoted to the proper treatment of accidents.18
These articles deal with the perennialtension between theories of negligence
and those of strict liability. The basic conflict between these two theories is
illustrated by the divergent treatment they require when the defendant has
taken reasonable (but unavailing) steps to avoid harming the plaintiff. The
general principle of the law of negligence is that the defendant should never
be held liable in a case of this sort. Being reasonable,his conduct is not blame-
worthy in any moral sense. Given that its expected benefits exceeded its ex-
pected costs, it is conduct that is beyond economic criticism as well. As there
is no theory that makes the defendant's conduct wrongful, the plaintiff must
go without compensation.
The negligence approach ignores the fact that a lawsuit is always a com-
parative affair. The defendant's victory ensures the plaintiff's defeat. Why
then should it be decisive that the defendant acted reasonably? Perhaps the
plaintiff acted reasonably as well, if indeed he acted at all. Why should we
prefer the injurer to his victim in a case where one must win and the other
lose? Once a defendant is allowed to excuse himself on the grounds that he
acted with due regard for the plaintiff, it follows that he will be able to keep
for himself the benefits of his own actions even as he imposes their costs
upon a stranger. The crucial question is whether or not the defendant should
be allowed to force (and here the words should be taken literally) others to
bear his costs because prior to the accident he made a decision that was
rational in the case. The major premiseof the theory of strict liability is that,
prima facie, he should not be allowed to help himself by taking or destroying
the plaintiff's person or property. If in the course of activity conducted for
his own gain, the defendant had harmedhimself or damaged his own property,
he would be requiredto bear that loss himself even if the expected gains were
worth the risk involved, and there is no reason why that result should not be
sought by the legal system as well when the initial harm is to the person or
property of another. As a matter of fairness between the parties, the defen-
dant should be requiredto treat the harms which he has inflicted upon another
as though they were inflicted upon himself. The question of his negligence or
intention is at the outset of the suit quite immaterial.
In order to make good on the promise of a system of strict liability, it is
18 Epstein, Strict Liability 151; Epstein, Defenses 165.

This content downloaded from 129.127.145.240 on May 08, 2018 20:13:09 PM

All use subject to University of Chicago Press Terms and Conditions (http://www.journals.uchicago.edu/t-and-c).
 INTENTIONAL HARMS 399

still necessary to accomplish three distinct tasks. The first is to give an ac-
count of the notion of causation implicit in the general argument of fair-
ness made on behalf of a theory of strict liability.19The second is to work out
in detail the exceptions to the initial causal premise, for the proposition that
the defendant must always be held liable when he has hurt the defendant is
not capable of principled defense. The third-and main concern of this paper
-is to show how the notion of intention to harm can be made an integral
part of a system of tort law that begins with an initial concern about causation.

A. Causation Revisited
It is necessary to consider four basic causal paradigms in order to account
for the use of causal language in most common cases of physical harm. The
approachused here does not begin with the loose test of "but for" causation,
which is then cut down to size by the "policy" considerations of proximate
cause. Instead we begin with the simplest model of causation and then ex-
tend it by degrees to more complex situations.
The first paradigm of causation--one accepted in all legal systems-is the
direct and immediate application of force by one person against the person or
property of another: it is the case of A hits (wounds, kills, etc.) B. The
second paradigm involves the common case where A frightens (shocks, terri-
fies) B. Here the element of force is also prominentbut only in its threatened
application as viewed from the standpoint of B, whose reactions to A's conduct
must be taken into account to complete the causal chain. Although the causal
link is thus attenuated, it is by no means broken, as the grammaticalparallel
between the propositions "A hit B" and "A frightened B" reveals. The third
paradigm of causation takes the form "A compelled B to hit C," where the
acts of a third person, B, must be taken into account to complete the causal
chain between A and C. Nonetheless, despite this complication, the use of
force or the threat thereof again shows the closeness between this paradigm
and the two that precede it.20
The fourth major paradigm of causation applicable to most tort cases con-
cerns the creation of dangerous conditions which then result in harm, harm
immediately caused in any of the three above senses. Since the harm to the
plaintiff can be completed only upon the occurrence of some subsequent act
or event that works upon the condition that the defendant created, all of
the situations to be mentioned are best regardedas instances of indirect (but
not remote) harm. The first of the cases to be noted concerns things which
are made dangerous on account of their position. The simplest case of this

19 Epstein, Strict Liability 160-89.

2o For an analysis of the actions among A, B and C, see Epstein, Strict Liability
174-77.

This content downloaded from 129.127.145.240 on May 08, 2018 20:13:09 PM

All use subject to University of Chicago Press Terms and Conditions (http://www.journals.uchicago.edu/t-and-c).
400 THE JOURNAL OF LEGAL STUDIES

sort is one in which A drops a brick upon B's head: the release of the brick
alone is sufficientto allow the force of gravity to work upon it. And it is only
a short extension to the case where A places that brick on the edge of a table
where a gentle push will allow the force of gravity to pull it down.
A second common instance of dangerous conditions involves the manufac-
ture of defective products which cause (by force) harm when used in their
ordinary manner. A weak bolt on an industrial machine may give way and
release a piece of wood that strikes its operator. The intermediate act of the
plaintiff does not break, but rather completes, the causal connection between
the defendant'sact and the plaintiff'sinjury.
A last instance of dangerousconditions concerns cases where the defendant
has blocked the plaintiff's right of way. Again the element of force becomes
prominent when harm results after the plaintiff crashes into the blockade.
And again the plaintiff's application of force only provides the last causal
link between the defendant's act and his own harm.

B. Defenses in Accident Cases

The second task of the law of tort concerns the evaluation of the excuses
or justifications that the defendant might urge on his own behalf. Of those
that should be rejected, the one most relevant to our concerns here is that of
private necessity, whereby the defendant inflicts harm upon the plaintiff in
order to protect himself against the harm threatened by some natural event.21
As with prima facie cases, effective defenses should show not only that the
defendant is not to blame (in any moral sense) but also that he should be
preferredto the plaintiff, who may be as blameless as himself, in any contest
between them. On the question of liability we must choose between the plain-
tiff who had no option at all and the defendant who had at least the option-
hard as it may be-between suffering the harm himself and inflicting it upon
another. If, under conditions of necessity, a defendant had damaged his own
property, he could not call upon a stranger to compensate him for his loss.
If under those same conditions he harms the third party instead of himself,
he should be required to make compensation for the loss suffered. The argu-
ments just made apply with equal force to infancy and insanity, and to com-
pulsion by the acts of third parties.
The three principal types of valid defense are causal defenses, assumptionof
risk, and plaintiff's trespass. In connection with each of these three defenses,
we again encounter the traditional conflict between strict and negligence
21 For discussion, see Francis H. Bohlen, Incomplete Privilege to Inflict Intentional
Invasions of Interests of Property and Personality, 39 Harv. L. Rev. 307 (1926) [here-
inafter cited as Bohlen, Incomplete Privilege], reprinted in his Studies on the Law of Torts
614 (1926) [hereinafter cited as Bohlen, Studies]; Epstein, Strict Liability 157-60;
Epstein, Defenses 169.

This content downloaded from 129.127.145.240 on May 08, 2018 20:13:09 PM

All use subject to University of Chicago Press Terms and Conditions (http://www.journals.uchicago.edu/t-and-c).
 INTENTIONAL HARMS 401

theories.Assumptionof risk, for example,might be a sufficientdefensein

and of itself; or it might be a defense only if the plaintiff's conduct were
unreasonable.The appropriatenessof the reasonablenessrequirementis thus
called into play in connection with defenses based upon plaintiff's conduct.
Given the preference for the strict formulation of the prima facie case, the
same treatment should be accorded affirmative defenses. There is no reason
why the defendant's conduct should be judged by one standard and the plain-
tiff's conduct by another. A strict standard for defenses will indeed bar
plaintiff's recoveryin cases where it might otherwise be allowed, but the result
in question need not be improper for that reason alone. The purpose of the
tort law is not to provide the plaintiff, at defendant's expense, with compensa-
tion in every case of injury. It is to determine whether the compensation is
warranted on the facts of each case, where the same standard of judgment
is applied to the conduct of both parties.
A theory of strict liability, then, has as one of its elements the development
of a strict theory of defenses. We turn first to the causal arguments open to
the defendant.22Suppose the defendant had struck the plaintiff, causing him
harm. Under the paradigms already developed, he would have a good cause
of action on the strength of the causal argument.If now the defendant alleges
that the plaintiff struck him as well, that affirmative defense (also strict in
form) should be sufficient to require an apportionment of the loss, since
neither party demonstrates a causal priority over the other. Similarly, if A
struck B, and could show that C compelled him to do so, A would have an
action against C even though he remains liable to B. In the special case where
B and C are the same person, the action over against B can be dispensed
with and an affirmativedefense substituted in its place.
The same line of argument applies to the other causal paradigms. For ex-
ample, suppose A drives into B's car when B has blocked A's right of way.
B has a good prima facie case based upon the paradigm of force, but A has
a good affirmative defense because B blocked his right of way. The same
principle applies as well to objects dangerous because of the potential force
that they contain. Take a case where A strikes an object which explodes,
harming B. B's prima facie case-that A set off the explosion-could be met
by the affirmativedefense that B created the dangerouscondition in question.
Parallel examples can of course be constructed for cases of objects dangerous
because of internal defects or dangerousbecause of position. Indeed one great
strength of the causal analysis developed here is that it permits a principled
(and predictable) resolution of the troublesomeproblems of proximate cause
even in those frequent cases of accidents that involve multiple causation.
There is no need to dwell at great length upon the other affirmativedefenses.

22 See, for a more complete discussion,

Epstein, Defenses 174-85.

This content downloaded from 129.127.145.240 on May 08, 2018 20:13:09 PM

All use subject to University of Chicago Press Terms and Conditions (http://www.journals.uchicago.edu/t-and-c).
402 THE JOURNALOF LEGALSTUDIES

We have already noted that assumption of risk is a defense that is strict in

form. The issue is much the same with respect to the plaintiff's trespass: the
strict form of that defense is made out by showing the plaintiff entered the
defendant's land. The stage is now set to take into account the defendant's
intention to harm the plaintiff.

C. The Third Stage: The Place of the Intention to Harm

The defendant'sintention to harm, I contend, becomes relevant in the third
stage of argumentonly after the defendant has interposed a sufficientaffirma-
tive defense to the prima facie case. It is true that this approach will not
eliminate all the vagueness created by the use of the language of intention.
But this approach does reduce substantially the number of cases in which
the difficulties with the language of intention will have to be confronted.
In most tort cases, it is the causal language that will bear the tensions of
the system, for the mental element, which bears the stress at the outset in
criminal cases, becomes relevant only at a later stage in the tort analysis. In
order to show the importanceof reserving the intentional element to the third
stage of a case, it is necessary first to show how the defendant's intention
to harm ties in with each of the affirmativedefenses already discussed, and
second to show the consequencesthat follow once the element of intention is
no longer regardedas part of the prima facie case.
Let us first examine cases in which the plaintiff's conduct raises one of the
causal defenses previously mentioned. Take first the case of a head-on colli-
sion between A and B in which the paradigm of force alone is operative.
Here B should be allowed to recover in full over A's partial causal defense
where he can show that A intended to cause him harm. The sequence of pleas
is: (1) A hit B, (2) B hit A, (3) A intended to harm B.23 Given the causal
involvement of both parties, the intention to harm decides the case. Straight-
forwardextensionsof the position are also possible where more than one causal
paradigm is involved. Suppose B could show that A struck him, and is then
met with A's defense that B blocked his right of way. B again will have the
benefit of a good reply over this complete defense if he can show that A
intended to cause him harm. The sequence of pleas is as before, except that
"B blocked A's right of way" is substituted for "B hit A." The same argu-
ments apply, moreover, regardless of which of the causal paradigms-be it
defective products, unstable position, or whatever-is applicable to the plain-
tiff's (B's) conduct.
The familiar doctrine of last clear chance contains a strong suggestion of
23 In those cases in which the harm is deliberate on both
sides, apportionment is again
the preferable solution. "B intended to harm A" is a good rejoinder in the fourth stage
of the action, subject to the same pleas in defeasance discussed in the next part of this
article.

This content downloaded from 129.127.145.240 on May 08, 2018 20:13:09 PM

All use subject to University of Chicago Press Terms and Conditions (http://www.journals.uchicago.edu/t-and-c).
 INTENTIONAL HARMS 403

this position, especially if the elements of knowledgeand conscious indifference

prominent in some formulations of the rule24are given adequate attention.
One case, close to the line, that illustrates the approach is Kumkumian v.
City of New York.25The plaintiff's decedent had made his way onto the city's
subway tracks. The defendant's train ran over his body and set off the trip-
ping mechanism that stopped the train. The motorman twice reset the device
without looking under the train before the deceased's mangled corpse was
found "steaming"between the third rail and the running rail. The court per-
mitted the jury to find for the plaintiff if it believed that the death "did not
come about through mere lack of vigilance in observing the tracks, but rather
as a result of . . [defendant's agents'] own willful indifference to the
emergency called to their attention by the automotive equipment, to which
clear warning they paid no heed."26Even on these facts, the plaintiff's case
is a difficult one because the defendants knew that there were many ways of
setting off the tripping mechanism that did not signal danger to human life.
There would have been no doubt about a judgment for the plaintiff, however,
if the motorman had decided to restart the train knowing that the decedent
was beneath it, for then his decision would have been tantamount to a deci-
sion to inflict an intentional harm. For the harm caused after the discovery
of the decedent's peril, that intention should override any defense resting
upon the decedent's blocking the train's right of way.
In applying the doctrine of last clear chance, however, the courts do not
always pay sufficient attention to the defendant's mental state. The conse-
quences of this view are unfortunate; where the harm in question is accidental,
the plaintiff should at best be allowed to recover for some portion of the loss
on causal principles alone, but not for the full extent of his losses, as he does
when the last clear chance rule is applied.27The point is illustrated by Peter-
24In the languageof Cardozo:
The doctrineof the last clear chance,however, is never wakened into action unless and
until there is brought home to the defendantto be chargedwith liability a knowledge
that anotheris in a state of presentperil, in which event there must be reasonableeffort
to counteractthe peril and avert its consequences.. . . Knowledgemay be established
by circumstantialevidence,in the face even of professionsof ignorance. . . but knowledge
there must be, or negligenceso recklessas to betoken indifferenceto knowledge.
Woloszynowskiv. New York Central R.R., 254 N.Y. 206, 208-09, 172 N.E. 471, 472
(1930).
25305 N.Y. 167, 111 N.E.2d 865 (1953).
26Id. at 176, 111 N.E.2d at 869.
27Within the negligencesystem, the doctrineof last clear chanceis used to get around
the completebar that contributorynegligenceraises to the plaintiff'saction. The avoid-
ance of the one unjust result is offset, however, by the creation of another, for the
plaintiff is by no means entitled to full recovery. Even within the negligencesystem,
there is no need for a last clear chance rule if a thoroughgoingsystem of comparative
negligenceis adopted. The same is true in the causal system I have put forward, even
if apportionmentis in accordancewith somewhat differentprinciples.Epstein, Defenses
179-84.

This content downloaded from 129.127.145.240 on May 08, 2018 20:13:09 PM

All use subject to University of Chicago Press Terms and Conditions (http://www.journals.uchicago.edu/t-and-c).
404 THE JOURNALOF LEGALSTUDIES

son v. Burkhalter.28The defendant was driving his car in a northerly direc-

tion toward an intersection. Fifty feet from the crossing, he noticed the
plaintiff looking backwards over his shoulder while riding his motorcycle
easterly toward the intersection. The defendant continued on his way through
the intersection (where he had the right of way), taking no further notice
of the plaintiff until the plaintiff was within six feet of his car, still looking
backwards. On the causal question, the defendant could claim both that he
had the right of way and that the plaintiff hit him. The only arguments for
the plaintiff would have to stem from some statutory duty to look out for the
plaintiff after he was noticed. If the only choice were between full recovery
and no recovery, the strength of the defendant's causal case suggests that the
latter is the better result. The dissent appears to have much the better of
the argument when it protests:

The majorityrequirethat the defendantin enteringand crossinga street inter-

sectionat a lawful rate of speed,and havingthe right of way over plaintiff,should
have looked right, looked left, looked ahead (all of which he did do) and have
anticipated that plaintiff riding a motor scooter approaching and entering the
same intersection from the defendant's left at a speed not over 30 miles per hour
would for more than 75 feet of his travel look neither right or left nor ahead but
only behind him, would violate the law by failing to yield the right of way to
defendant, and crash into defendant's car. For failing to anticipate all these things
and avoid the accidentdefendantis liable to plaintiffin damageson the theory
of last clear chance.29

Thus far we have examined the way in which the plaintiff's plea of the
defendant's intention to harm ties in with causal pleas that the defendant
can raise on his own behalf. Analogous arguments apply when we consider
either of the noncausal defenses-plaintiff's trespass or his assumption of risk
-that are open to defendants. The position with respect to assumptionof risk
was hinted at by the court in Vosburg v. Putney when it stated:
Had the parties been upon the play-groundsof the school, engagedin the usual
boyish sports, the defendantbeing free from malice, wantonnessor negligence,
and intendingno harmto plaintiffin what he did, we shouldhesitate to hold the
act of the defendantunlawful,or that he could be held liable in this action.
Some considerationis due to the impliedlicense of the play-grounds.30

28 38 Cal. 2d 107, 237 P.2d 977 (1951).

29 Id. at 114, 237 P.2d, at 981.
30 80 Wis. 523, 527, 50 N.W. 403, 403-04 (1891). In this context, the use of negligence
to defeat the defense of assumption of risk could not refer to the poor execution of
maneuvers in the course of the game, as for example, by a badly timed block in football.
But it could well cover the use of dangerous equipment, for example an unpadded steel
knee guard, prohibited by the applicable regulations.

This content downloaded from 129.127.145.240 on May 08, 2018 20:13:09 PM

All use subject to University of Chicago Press Terms and Conditions (http://www.journals.uchicago.edu/t-and-c).
 INTENTIONAL HARMS 405

In effect the argument could be formulated as asserting that once the defen-
dant is able to prove (even if only by inference from custom or conduct)
that the plaintiff assumed the risk of accidental injury, the plaintiff makes
out a good reply by showing that the defendant intended to harm him. The
appropriatesequence of pleas is (1) A kicked B, (2) B assumed the risk of
harm from A, (3) A intended to harm B. The defense of assumption of risk
indicates that the defendant has agreed to bear the risk of the accident he
might suffer even if the accident was caused by the plaintiff. Where there is
such an agreement,the appropriatereply is that, prima facie, deliberate harm
was not in the class of risks the plaintiff agreed to assume. Where the risk
assumed by the plaintiff rests solely on his decision to run the known hazard
in order to enjoy some gain, the implication is perhaps somewhat more difficult
to draw; but it seems only fair to conclude that the plaintiff did not issue
an invitation for the defendant to harm him at his own pleasure.
Parallel arguments apply to the case of plaintiff's trespass, where the
general rule has been that the defendant is liable for any harm that he in-
flicts wilfullly or wantonly.31Suppose the defendant could deliberately inflict
some harm upon the plaintiff solely because he was a trespasser. How then
could we limit the application of the principle? There is no ready limitation
that can be placed upon the defendant's conduct once we have decided that
his intention to harm is not relevant to the case. If his intentions are indeed
irrelevant, it makes no sense even to ask for any justification, no matter
what its form, for the harm intended. How does one justify an accident?
If the intention to harm is immaterial, why is not an owner at liberty to kill
a trespasser?And if he could kill him, then he could spare him as an act of
mercy, only to condemnhim to a life of slavery instead. It is difficult to know
what counts as an argumentin support of the owner'sprivate right of conquest

31 This view of the problem has long been adopted at common law. William L. Prosser,
Handbook of the Law of Torts 361-64 (4th ed. 1971) [hereinafter cited as Prosser,
Torts]. Once the trespasser is discovered, many jurisdictions require the owner to conduct
himself with reasonable care. That approach is somewhat broader than the position taken
in the text, but, since it does not abandon the subjective requirement of knowledge, it
occupies a halfway place between intention to harm and negligence, even if it is not
tantamount to recklessness. In any event, it does show that a test of strict liability is
inapplicable once the defendant is able to interpose the plaintiff's trespass as a defense.
There is, of course, a clear parallel between the tests here and those used under the
doctrine of last clear chance.
There are, however, a number of recent decisions that have tended to eliminate the
mental element entirely by using a straight negligence approach in the case of the tres-
passing plaintiff. See, e.g., Smith v. Arbaugh's Restaurant, 469 F.2d 97 (D.C. Cir. 1973);
Rowland v. Christian, 69 Cal. 2d 108; 70 Cal. Rptr. 97, 443 P.2d 561 (1968). It is
difficult to know how far the trend will go, as the cases in which the issue has been
raised have not in general involved trespassers. Rowland v. Christian was a case of a
residential visitor, and Smith v. Arbaugh's Restaurant was a case of a health inspector
who entered private premises under public authority. For criticism of the straight negli-
gence approach see Epstein, Defenses 202-04.

This content downloaded from 129.127.145.240 on May 08, 2018 20:13:09 PM

All use subject to University of Chicago Press Terms and Conditions (http://www.journals.uchicago.edu/t-and-c).
406 THE JOURNAL OF LEGAL STUDIES

over trespassers and difficult in consequence to know what beyond a simple

conception of liberty could be urged against it. The issue is that of individual
autonomy, of the degree of sacrifice that one individual can demand as of
right from another. On that question, the defendant's intention to harm plays
a crucial role in the formation of the law of torts. The plaintiff's trespass
should excuse the defendant in a case of simple accident, but it does not
justify his actions where the harm is deliberately inflicted.
The plaintiff's reply based on the intention to harm has important substan-
tive ramifications.In the first place, this sequence of pleas explains why at
commonlaw contributorynegligencehas never been regardedas a good defense
to an intentional harm.32Contributorynegligencehas the same role in the neg-
ligence system that causal pleas and assumption of risk occupy in a system
of strict liability. It serves in accident cases as the comprehensiveaffirmative
defense based on plaintiff's conduct. Pleading the defendant's intention to
harm in the prima facie case anticipates and forecloses any affirmative de-
fense that is appropriateonly to accident cases. However, while the inclusion
of the intentional element in the prima facie case may yield the correct result
on the question of liability, it is not without its drawbacks,as it raises a false
conflict between the theory of strict liability and that of intentional harms.
Suppose that the defendant hurt the plaintiff, that no affirmativedefense was
available, and that the defendant did not intend to hurt the plaintiff. If the
prima facie case required the allegation of an intentional harm, the plaintiff
would fail, and there would be no need to consider the problems that might
otherwise be raised by the absence of an affirmativedefense. Yet if we treat
the.case as initially one of strict liability, the plaintiff has a good prima facie
case to which the defendant has no response. Where issues are raised in their
proper sequence, the plaintiff will be able to recover; where he is forced to
raise the issue of intention in the prima facie case, he cannot.
Another advantage of reserving the issue of intention until the third stage
of the argument is that it helps clarify the relationship between rules of lia-
bility and those of the remoteness of damage. Seavey's famous aphorism,
"prima facie at least, the reasons for creating liability ought to limit it,'"33
appeals to a principle of symmetry to explain why "reasonableforesight" is
the appropriate test of proximate cause under a general law of negligence.
Within a system of strict liability, however, the symmetry principle cuts in
the opposite directionand demandsthe rejection of the foresight test; if causa-
tion is prima facie the test of liability, so too should it be the test of damage.34
32 Prosser, Torts 426.
83 Warren A. Seavey, Mr. Justice Cardozo and the Law of Torts, 52 Harv. L. Rev. 372,
48 Yale L.J. 390, 404, 39 Colum. L. Rev. 20, 34 (1939).
34 Epstein, Strict Liability 162 n.34.

This content downloaded from 129.127.145.240 on May 08, 2018 20:13:09 PM

All use subject to University of Chicago Press Terms and Conditions (http://www.journals.uchicago.edu/t-and-c).
 INTENTIONAL HARMS 407

If, however, the case is pleaded out to the plaintiff's reply, then that principle
of symmetry now requiresthat the plaintiff be allowed to recover for the harm
only if it is both caused and intended by the defendant. It will no longer be
sufficient to rely on the causal connection alone. The intention to harm may
make the case easy in an intuitive sense, but the need to plead and prove it
serves to restrict, not expand, the scope of the defendant's liability. Thus if
the defendant strikes a plaintiff who is in an extrasensitive condition, he will
normally be prima facie liable for all the harm he caused whether or not
he intended it. The maxim is that the tortfeasor takes his victim as he finds
him. But after some affirmative defense, say assumption of risk, is estab-
lished, the plaintiff should be able to recover only for the harm that the de-
fendant intended to inflict upon him. He should therefore be restricted to
nominal damages when the defendant intended to touch him in ignorance of
his delicate condition. Where the intentional element is incorporatedinto the
prima facie case, it is difficult to reach this result. If liability depends on both
intention and causation, why should remoteness of damages depend on causa-
tion alone when the conditions that create liability also limit it?35 Yet the
departure from theory is countenanced in order to allow the courts to reach
the proper results on remoteness of damages obtained as a matter of course
when strict causal principles are allowed to govern the prima facie case.36
We are now in a position to resolve the puzzles created by the four cases of
"intentional" harm set out earlier in the paper.
In Vosburgv. Putney, the prima facie case is strict; the plaintiff says that
the defendant kicked him below the knee. Assumption of risk cannot, as the
court notes, be made out on the facts, and the plaintiff's weakened condition is
not an issue, owing to the maxim that the defendant must take his victim as he
finds him. Having forestalled the affirmativedefense, there is no occasion to
consider a reply based upon the intention to harm.
Similar observations dispose of Mohr v. Williams, the case of an operation
without proper consent. The prima facie case is strict. There is no need to
worry about whether the defendant touched the plaintiff in a spirit of pleas-

35 Thus: "There is a definite tendency to impose greater responsibility upon a defendant

whose conduct has been intended to do harm, or morally wrong. More liberal rules are
applied as to the consequences for which he will be held liable." Prosser, Torts 30. See
also Derosier v. New England Telephone & Telegraph Co., 81 N.H. 451, 463-64, 130 A.
145, 152 (1925); Ralph S. Bauer, The Degree of Moral Fault as Affecting Defendant's
Liability, 81 U. Pa. L. Rev. 586 (1933).
36 Note, too, that the exclusive use of a causal test on the issue of remoteness of
damages will lead to errors in those cases in which the defendant has a good affirmative
defense, for then the allegation of intention is necessary to revive the plaintiff's case and
cannot, accordingly, be fictionalized. See the discussion of Talmage v. Smith, infra,
at 408.

This content downloaded from 129.127.145.240 on May 08, 2018 20:13:09 PM

All use subject to University of Chicago Press Terms and Conditions (http://www.journals.uchicago.edu/t-and-c).
408 THE JOURNAL OF LEGAL STUDIES

antry, and no need to characterize his conduct as violent in order to make

out the prima facie case. Once the touching and consequent harm is shown,
the plaintiff wins because the defendant has no affirmativedefense. The ques-
tion of intention to harm never arises.
There is no need either to dwell on the question of intent in Garratt v.
Dailey either. That case differsfrom Vosburgv. Putney only in that the causal
paradigm of dangerous conditions (defendant took plaintiff's chair out from
under her) is involved in the statement of the prima facie case. Since the case
does not give rise to any affirmativedefense the question of deliberate harm
is not reached and need not be considered.
The most difficult case to evaluate is Talmage v. Smith. The appropriate
prima facie case is that the defendant put out plaintiff's eye when he hit him
with a stick. The defendant could respond by pointing to the plaintiff's tres-
pass, and the plaintiff could prevail only if he could show that the defendant
had the requisite intention to harm him. Since the element of intention is
now in the reply, the extended notions of intention used in the three cases
just consideredare no longer applicable. On the facts of the case, the plaintiff
must build his own case on what the defendant intended to do to a third
party stranger, as no harm was intended him.37 Under the criminal law,
the defendant could be held for his attempt to harm the third person (and
pleas as well any defenses applicable to attempts), but could not, given the
rules of mens rea, be held responsiblefor the harm sufferedby the plaintiff.38
As a tort matter, if the third party had been injured by the defendant, he
could no doubt have raised the defendant'sintention to harm him in the reply
if that were necessary for his recovery. And if the case had not been com-
plicated by the plaintiff's trespass, he could recover on a causal theory and
need not rely on the fiction of transferred intent. But where the plaintiff's
trespass is part of the case, that fiction is inappropriate, not because it is
unnecessary but because it is insufficient. The plaintiff must fail because he
cannot make out the sufficientreply that the defendant intended to harm him.

37 The analysis in the text hearkens back to some of the language used in Palsgraf v.
Long Island R. Co., 248 N.Y. 339, 162 N.E. 99 (1928), where Cardozo in his articulation
of the duty concept speaks of "the futility of the effort to build the plaintiff's right upon
the basis of a wrong to someone else." Id. at 343, 162 N.E. at 100. Again: "What the
plaintiff must show is a 'wrong' to herself, i.e., a violation of her own right, and not
merely a wrong to someone else." Id. at 343-44; 162 N.E. at 100. Indeed, the rejection
of strict liability led Cardozo to view the doctrine of transferred intent with some caution.
"Under this head [of acting at one's own peril], may fall certain cases of what is known
as transferred intent, an act wilfully dangerous to A resulting by misadventure in injury
to B (Talmage v. Smith, 101 Mich. 370, 374). These cases aside, wrong is defined in
terms of natural or probable, at least when unintentional." Id. at 344, 162 N.E. at 101.
38 A criminal version of the doctrine of transferred intent had been urged by Coke.
E. Coke, 3d Institute 56. It has not found much acceptance in the modern law. See, e.g.,
Reg. v. Sernm,16 Cox Crim. Cas. 311 (1887).

This content downloaded from 129.127.145.240 on May 08, 2018 20:13:09 PM

All use subject to University of Chicago Press Terms and Conditions (http://www.journals.uchicago.edu/t-and-c).
 INTENTIONAL HARMS 409

III. EXCUSES AND JUSTIFICATIONS

Once the defendant's intention to harm the plaintiff is recognized as an

appropriatereply to an affirmativedefense, a host of other substantive issues
may be raised, as the defendant may introducein the fourth stage of argument
new matter in the nature of excuse or justification either to defeat or to
diminish the plaintiff's claim. The common tendency today is to treat excuses
and justifications together under the single rubric of "privilege."39But that
usage is unwise. The two words, excuse and justification, are as a matter of
common usage by no means verbal equivalents. A defendant will point to the
weaknessesof his personal condition in order to excuse himself for harm inten-
tionally caused. He will, however, point to the plaintiff's conduct when he
wishes to justify the intentional infliction of harm. As excuses and justifica-
tions raise quite different problems, and have quite different consequences,
they are best considered separately.

A. Excuses
The most commonexcuses to be consideredare those of infancy and insanity
on the one hand, and that of private necessity on the other. In response to a
prima facie case based on causal allegations alone, each of these excuses is
insufficient,as indeed they are treated (with some difficulty) even within the
frameworkof negligence law. Even if they are insufficient at the second stage
of a case, they need not be rejected as well at the fourth. To the contrary,
they could be raised at the fourth stage of the argument only if immaterial
at the second. Thus in order to reach the problem of personal excuses, it
must necessarily be shown first that the defendant has a good defense at the
second stage of the argument (based upon defenses already discussed) and
then that the defendant intended to cause the harm in question. Deliberate
harms are given special treatment because they mark off, as a first approxima-
tion, the class of bad acts. The introduction of an excusing condition in the
fourth stage weakens the nexus between deliberate harms and bad conduct.
The identification of these excuses is best made, now that the intention to
harm is in issue, by following the lead of the criminal law, such as it is. Thus
infancy looks as though it should be given no effect at all, except perhaps
for children of very tender years. Those of insanity or compulsion on the
other hand should be given at least partial and, perhaps, full effect. Suppose
there is a head-on collision between two cars, one driven by A, and the other
by B, where A deliberately harms B in order to escape from personal peril.
With B as plaintiff, the sequence of pleas is as follows: (1) A hit B, (2)
B hit A, (3) A intended to harm B, (4) A intended to harm B only to escape

39 Bohlen, Incomplete Privilege 307-08, reprinted in his Studies 614-15 (1926).

This content downloaded from 129.127.145.240 on May 08, 2018 20:13:09 PM

All use subject to University of Chicago Press Terms and Conditions (http://www.journals.uchicago.edu/t-and-c).
410 THE JOURNAL OF LEGAL STUDIES

from a private necessity. At best the necessity neutralizes intention; as such

it follows not that there is a complete defense to the plaintiff's case, but that
the case should be decided on the strength of the first two pleas, with appor-
tionment of damages the indicated solution. Alternatively,if necessity has only
partial effect, then A should be charged with some additional portion of the
damages, not his on causal principles alone, in order to reflect the intentional
element of the case.
The modern rules of tort liability, being predicated upon negligence or in-
tent, require different results. In the first place, pleas (1) and (2), being
strict in form, are insufficient. B, however, could cure this defect by alleging
in his prima facie case that A meant to harm him.40 At this point, if private
necessity is allowed to operate as a complete defense, the plaintiff is barred
from recovering any fraction of his damages on causal principles alone, con-
trary to the result requiredabove. If, alternatively, private necessity is treated
as a partial defense, damages are apportioned,but not in accordancewith the
rules developed under the staged system. To illustrate the point, suppose that
the necessity plea when given partial effect requires that 80 per cent of the
harm be treated as accidental and 20 per cent of it as deliberate. Under the
current rules of law, the plaintiff will be able to recover for only 20 per cent
of his loss, as he has no case with respect to the accidental portion of the harm.
With the alternative system, the plaintiff will be able to recover that same 20
per cent for deliberate harm, plus that portion of the accidental harm for
which the defendant should be held accountable on causal principles alone.
The staged mode of argumentationtherefore allows us to give full effect to
each of the four features of the case, whereas the modern system denies all
independentsignificanceto the causal contentions.

B. Justifications
When we move to the class of justifications, two pleas are of central im-
portance: self-defense and consent. Neither of these pleas could arise in the
absence of the deliberate infliction of harm, and for that reason alone they too
must be reserved to the fourth stage of argument. In each of these cases,
the defendant argues that the plaintiff's conduct itself provides a reason that
makes it right and proper for the defendant to act as he did, even though
his intention appears to make his conduct wrongful. Both these justifications
function as complete, if defeasible, bars to the plaintiff's entire case. Their
consequences therefore are quite unlike those which are attached to personal
40 The consequences of rejecting private necessity as a defense are different, moreover,
within the framework of the two systems. Under the current law the plaintiff recovers
for all of the harm inflicted; in the staged alternative developed here, he recovers only
for that harm which is both caused and intended. On the relationship between the measure
of damages and the system of staged pleading see Epstein, Pleadings and Presumptions 569.

This content downloaded from 129.127.145.240 on May 08, 2018 20:13:09 PM

All use subject to University of Chicago Press Terms and Conditions (http://www.journals.uchicago.edu/t-and-c).
 INTENTIONAL HARMS 411

excuses, which when given full effect only requires the apportionmentof loss
in accordancewith the principles, largely causal, applicable to accident cases.
There is little that need be said to support the recognitionof either justifica-
tion by the law of torts. The argument for self-defense, that it is proper to
meet force with force, has a strong intuitive appeal that antedates any utilitar-
ian justification that might be given it. Its acceptance vindicates the impor-
tance attached to both liberty and property, for it allows the use of force
against those who threaten to take them away. The case for the recognition
of consent as a defense in case of the deliberate infliction of harm can also
be made in simple and direct terms. The self-infliction of harm generates no
cause of action, no matter why inflicted. There is no reason, then, why a person
who may inflict harm upon himself should not, prima facie, be allowed to have
someone else do it for him.41Individual autonomy, the organizingprinciple of
the tort law, is not violated when one person inflicts harm upon another at
the latter's insistence.
The problemswith both self-defense and consent do not concern their suffi-
ciency; instead, they concern first their scope and second the qualifications
that must be attached to them.
1. Self-Defense. The sequence of pleas with B as plaintiff and A as de-
fendant that raises the question of self-defense is as follows:
(1) A hurts B (in any of the causal senses already developed);
(2) B attacks or frightens A;42
(3) A intends to harm B;
(4) A does so in order to defend himself from B's attacks.
The requirement of B's attack or threat will sometimes involve issues of
characterization.Has A "read into" B's conduct a threat that is not present,
or has B acted in a suggestive, even if harmless, manner? How, for example,
do we treat the common case where a man reaches inside his pocket when
approaching another person on a dark and empty street? On this type of
question, there tends to be a marked convergence between the theories ad-
vanced here and the traditionallaw of negligence, but there are still differences

41 There will of course be questions of the effectiveness of the consent under the crim-
inal law. There one could view, though in a somewhat formal sense, the function of the
system as being to monopolize the use of force in the hands of the state save in those
rare instances, such as self-defense, where it is not possible for the state to protect the
individual. On that view of the subject, it is clear, for example, why attempts are a
category of actionable wrongs under the criminal law even though they create no liability
under the tort law. It also explains why the criminal law in many jurisdictions views
suicide as a crime. If it is so viewed, moreover, then it follows that the consent of the
victim should be no defense to a criminal prosecution, even where it is in tort an effective
justification for the intentional infliction of harm.
42 The allegation in the text is applicable only to cases of defense of the
person. Where
the defense of real property is at stake, the analogous allegation is that the plaintiff
entered the defendant's land.

This content downloaded from 129.127.145.240 on May 08, 2018 20:13:09 PM

All use subject to University of Chicago Press Terms and Conditions (http://www.journals.uchicago.edu/t-and-c).
412 THE JOURNAL OF LEGAL STUDIES

between the two points of view. For example, the negligence law will take
into account the defendant's fears generated by the conduct of third parties,
whereas a system of strict liability will not.43 But even with differences of
this sort to one side, it is clear that the evaluation of the defendant's response
presents irreducible questions of fact no matter what theory of liability is
adopted.
The requirement of the actual or apparent threat to the defendant does
raise, however, an important question of principle, especially in connection
with the use of various mechanical devices for the protection of person or
property. In some cases, the defendant may surroundhis premises with a high
wall, which he may then top with broken glass or barbed wire. In still others
he may take the extreme measure of setting up spring guns or similar devices
that are triggered by some act of the intruder himself. In all of these cases,
the harm is deliberate; in some, in the strong sense that the harm to the
intruder was the defendant's sole object when he set the device; in others
in the sense that he knew of the substantial probability that harm would result
from their use, even though the prime object was to frighten intruders away.44
The crucial question therefore is whether the defendant can justify his con-
duct. In cases of actual invasion with the threat of harm, the issue of self-
defense, coupled with its possible exceptions, is squarely raised. It is not
raised, however, when the defendant's mechanicalapparatus harms the "inno-
cent" trespasser,one whose entry threatens harm to neither person nor prop-
erty.45At this point the argument shifts from a justification of the deliberate
infliction of harm to a (cost) justification of the decision to install the
mechanical device in the first instance. The acceptability of this economic
justification in turn rests on the initial choice between the principle of negli-
gence and that of strict liability.
Within the frameworkof a negligence system, the defendant should be able
to justify his decision to use a mechanical device by showing that it is a
reasonable decision. That inquiry can be undertaken in the usual common
sense terms characteristicof the common law approachto negligence, or with
the explicit comparison of costs and benefits required by an economic ap-

43 See Epstein, Strict Liability 173.

44 It makes no difference whether the specific purpose for setting the device was to
injure or to deter. There have, of course, been cases where that distinction has been made
(see, e.g., Bird v. Holbrook, 4 Bing. 628, 130 Eng. Rep. 911 (C.P. 1828)) but it is of no
importance once we accept the broader accounts of intention to harm, discussed supra
at note 4.
45 Bird v. Holbrook, supra note 44, allowed the plaintiff to recover in the classic case
of this sort. The plaintiff climbed into the defendant's garden in order to recover a
peahen for a woman who had lost it. While in the garden, he triggered the wires of a
spring gun and was injured in consequence. The defendant set the gun to protect
his tulip crop from loss by theft, and posted no notice of the gun's presence because
he wished to apprehend the thief.

This content downloaded from 129.127.145.240 on May 08, 2018 20:13:09 PM

All use subject to University of Chicago Press Terms and Conditions (http://www.journals.uchicago.edu/t-and-c).
 INTENTIONAL HARMS 413

proach. The first approach was taken by Bohlen and Burns in their classic
article on the subject,46and the second in Posner's recent article.47But the
results are much the same,48 in that both are prepared to allow the cost
justification where it is made out on its facts. Thus high walls and exposed
barbed wire may be used as a matter of course because they are efficient
deterrents that hold out little chance of harm to innocent persons, be they
trespassers or not. Spring guns and similar mechanical devices of course
present more complex questions, raising as they do the prospect of deadly
harm. With respect to them, no per se rule can capture the range of relevant
considerations.The authors look to the place in which the devices are used.
Thus a dwelling house merits greater protection than a warehouse, which in
turn merits more protection than an unused shack. They look as well to
the time of their use-night is more acceptable than day because of the
greater difficulty in the prevention of dangerous or felonious entry at night.
The alternative means of protection are also material: cheap and easy alter-
natives reduce the likelihood that the use of a spring gun should be privileged.
The question of notice also plays a role:49 the defendant who posts signs is
in a better position than one who does not, because he has taken reasonable
steps to enable potential victims to avoid harm. This analysis will, of course,
lead to the conclusion that there will in principle be times in which the
defendant could escape liability for the harm he has deliberately inflicted
upon a bare trespasser.50
The pattern of argumentthat produces this result is, however, unacceptable
under a system of strict liability. In this context, as in all others, the reason-
ableness of the defendant's antecedent conduct is the defendant's concern
alone. He should, of course, make the cost-benefit analysis, as he will, if he
is rational. But the calculations are of no concern to the plaintiff. If the
original decision to set the gun was unwise, but the plaintiff is a person against
whom the use of force is justified in self-defense, there is no reason to give
the plaintiff a windfall by disallowing the plea. The defendant's act might be

46 Francis H. Bohlen & John J. Burns, The Privilege to Protect Property by Dangerous
Barriers and Mechanical Devices, 35 Yale L.J. 525 (1926) [hereinafter cited as Bohlen &
Burns].
47 Richard A. Posner, Killing or Wounding to Protect a Property Interest, 14 J. Law &
Econ. 201 (1971).
48 The central points of the position are, in Posner's view: (1) the avoidance of a
"blanket permission or blanket prohibition" on the use of force, id. at 221; (2) the re-
jection of such formulas as no man should be allowed to do by indirection (mechanical
devices) what he could not do directly, id. at 225; (3) whether the plaintiff's entry was
felonious under state law should not determine whether the defendant is entitled to the
privilege, id. at 221. These results were for the most part reached by the common law,
if not by the Restatement. Id. at 205-08.
49 See, e.g., Bird v. Holbrook, supra note 44.
50 See Bohlen & Burns
546, for a hint of this position.

This content downloaded from 129.127.145.240 on May 08, 2018 20:13:09 PM

All use subject to University of Chicago Press Terms and Conditions (http://www.journals.uchicago.edu/t-and-c).
414 THE JOURNAL OF LEGAL STUDIES

imprudent at the outset, but it is justified in the end. If, on the other hand,
that original decision was prudent, but the act unjustified because the plain-
tiff threatened no harm to the defendant, then the defendant should be held
liable. He should not be allowed to force on others the costs of his own mis-
takes, however reasonable. Bohlen and Burns rightly note the parallel to the
case where a defendant injures the plaintiff in an effort to protect himself from
the attacks of a third party.51 But once the defendant is held liable in that
situation, as he is in a system of strict liability, then it follows that he should
be held liable here as well.52The defendant who uses mechanical devices will
no longer be in a position to discriminate between dangerous and innocent
intrusions, as he was when he defended his land in person. It may be wise to
take that risk in order to minimize the possibility of personal injury. It does
not follow, however, that as between himself and the simple trespasser, the
latter should take the risk of his miscalculations.53
One caveat must still be entered. Questions of reasonablenessof the sort
suggestedseem quite appropriatewhen the question at hand is whether general
legislation should prohibit the use of spring guns and the like where they are
apt to do more harm than good.64As with injunctions sought by private par-
ties, there is no hard and fast rule of decision and utilitarian considerations
must play some part in a decision that covers a vast multitude of possible
cases. But these considerationsare not relevant after the event to determine
whether the facts in the very case justified the use of force. Indeed, even
where the defendant's conduct does not comply with the statutory require-
ments, he still should not be deprived of his justification of self-defense in
the particular case. The better approach is to treat the decision in the case
as independent of the statutory scheme, allowing the defense in the private
action while at the same time subjecting the landowner to the statutory
penalties.
Many cases involving the defense of property will be decided in the same
way regardless of the view taken towards cost-justifications and the defense
of property. Thus Bohlen and Burns are, I believe, correct in suggesting that
there should be a per se rule that denies recovery to a plaintiff who falls while

51 Id. at 535.
52 Epstein, Strict Liability 158-60.
53 See, Bohlen & Burns 526. The same argument was made by counsel for the defendant
in Bird v. Holbrook, 4 Bing. 628, 637-38, 130 Eng. Rep. 911, 914-15 (C.P. 1828). The
concern just raised supports the position taken both by Bohlen and Burns and by Posner
that it is proper to reject the maxim that one cannot proceed by indirect means where
direct means are prohibited. Given a theory of strict liability, however, that maxim makes
sense, because it says in effect that the degree of risk created by the defendant's conduct
is his own concern and does not have any effect upon the outcome of the case.
54 See An Act to Prohibit the Selling of Spring Guns . . ., 7 & 8, George IV, c. 18
(1827). That statute drew the kinds of distinctions appropriate to the legislative function.
It allowed the use of spring guns only between sundown and sunrise, and then only for
the protection of dwelling houses upon the posting of adequate notice.

This content downloaded from 129.127.145.240 on May 08, 2018 20:13:09 PM

All use subject to University of Chicago Press Terms and Conditions (http://www.journals.uchicago.edu/t-and-c).
 INTENTIONAL HARMS 415

attempting to scale a high wall that the defendant has erected to protect his
property.55That result, however, need not rest on the asserted reasonableness
of the defendant's conduct; indeed the logic of any reasonablenesstest is not
easily invoked to support the per se results required here. Instead, the case
for the defendant should be made out on causal grounds. The only causal
paradigmof aid to the plaintiff is that of conditions made dangerousby virtue
of position.56Yet here it was the plaintiff who placed himself in the position
from which he could fall when he climbed the wall. Hence, the plaintiff's
causal argumentmust fail, and the defendant wins not because he can justify
the use of the wall to protect his property in economic terms, but because
the plaintiff fails to state a prima facie case in causal terms.
It is also possible in cases involving the defense of property for the defen-
dant to prevail in a system of strict liability even after the plaintiff has made
out a prima facie case on causal grounds. For example, if the plaintiff cuts
himself on the defendant's barbed wire as he tries to scale the wall, there is,
perhaps, the prima facie case of dangerous conditions, but there is also the
affirmativedefense of assumption of risk, in those frequent cases where the
plaintiff knew of the wire when he chose to scale the wall.
Let us turn now to the limitations applicable once self-defense is made out.
It is often said that the force used in self-defense must be "reasonable,"a
term that conceals far more than it reveals. To systematically approach the
problem, it is better to consider each individual factor that in an intuitive
sense bears on the question of reasonableness.To facilitate that analysis we
shall consider first the insufficient and then the sufficient exceptions to the
plea of self-defense.
The first possible but insufficient limitation concerns the case of the "in-
nocent attacker," one whose attack is neither negligent nor wilful. The
Restatement of Torts in its treatment of self-defense takes no position on
whether the defendant can assert the privilege of self-defense in cases of this
sort.57 The reason for this hesitation is clear within the framework of the
negligence law. If the plaintiff had harmed the defendant, he could not have

55 Bohlen & Burns 528.

56 The causal arguments could never be thrown into such sharp relief with the "but for"
test of causation: but for the erection of that wall, there would have been no harm. Since
that test does not yield a clear result, it leaves an open invitation to ask the "policy"
question of proximate cause, from which it is a short step to the issue of cost justification
raised by the negligence formula.
57 "The Institute expresses no opinion as to whether there is a similar privilege of
self-defense against conduct which the actor recognizes, or should recognize, to be entirely
innocent." Restatement (Second) of Torts ? 64. For an excellent discussion of the prob-
lems of self-defense and the innocent aggressor, see George P. Fletcher, Proportionality
and the Psychotic Aggressor: A Vignette in Comparative Criminal Theory, 8 Israel L. Rev.
367, 375 (1973) [hereinafter cited as Fletcher, Proportionality]. Although Fletcher dis-
cussed the question in the context of the criminal law, his conclusions are easily trans-
ferred to the tort law governing intentional harms.

This content downloaded from 129.127.145.240 on May 08, 2018 20:13:09 PM

All use subject to University of Chicago Press Terms and Conditions (http://www.journals.uchicago.edu/t-and-c).
416 THE JOURNALOF LEGALSTUDIES

been held responsiblefor the harm he caused. One way to reach this result is
to say that a person (here the plaintiff) can escape liability in the ordinary
case by showing that he did not act with negligence or intent. If that premise
is accepted, as it is in the modern law, it is not difficult to argue that the same
showing should rebut the defendant's justification of self-defense, thereby
limiting the scope of the plea solely to cases of "wrongful"attack. The view
is unfair, however, because it requires the defendant to surrender his own
personal integrity to the force of the plaintiff, the sort of sacrifice that it is
hard to ask, and impossible to demand, of any person. Though fairly required
by the law of negligence, it is not difficult to see why the Restatement shies
away from it.
This problem of the "innocent aggressor" cannot arise at all under a
system of strict liability. Here the plea of self-defense no longer turns on the
moral wrong of the plaintiff's attack; at this stage we need not trouble our-
selves with the distinctions amongst wilful, negligent and innocent aggressors
that pervade the Restatement.58Instead it rests on the judgment that as
between a victim and his assailant, the victim is prima facie to be preferred
in any action between them. That judgment allows the injured party his prima
facie case as plaintiff for the harm inflicted. Likewise, it prima facie allows
him to meet force with force, and thus to invoke as defendant the plea of
self-defense. The issue is not one of punishment of the assailant. His moral
innocenceis immaterialboth to the prima facie case based on causal principles
and to the justification of self-defense.
A second qualificationto the plea of self-defense-one based on defendant's
improper motive-likewise deserves rejection.59The plaintiff should not be
allowed to argue that the defendant is not entitled to the privilege because
the plaintiff is a person whom the defendant disliked and wished to harm in
any event. If there is no obligation to make a gift of services to a friend,
why should it be required on behalf of an enemy?60Malice itself cannot be
the source of a duty. The plaintiff fails in the Good Samaritan case because
of his inability to satisfy the causal requirementof the tort law, even where
the defendant refuses to give aid out of malice alone.
The case of self-defense raises a direct parallel. It is of course possible that
a person entitled to use force on his own behalf will decide not to exercise
that right. But there is no way in which a plaintiff whom the defendant hates
should be able to compel the waiver of privilege which could not be demanded
by the defendant's closest friend. The traditional maxim often quoted on the
question of motive is "maliciousmotives may make a bad case worse, but they
58 Restatement (Second) of Torts ? 63-66.
59 It is indeed rejected by the Restatement (Second) of Torts ? 63, comment e.
60oOn the question of the Good Samaritan, see generally, The Good Samaritan and the
Law (James M. Ratcliffe ed. 1966). For my own views, see Epstein, Strict Liability
189-204.

This content downloaded from 129.127.145.240 on May 08, 2018 20:13:09 PM

All use subject to University of Chicago Press Terms and Conditions (http://www.journals.uchicago.edu/t-and-c).
 INTENTIONAL HARMS 417

cannot make that wrong which is in essence lawful."61As stated, the proposi-
tion amounts only to a conclusion of law because it does not enable us to
determinewhat conduct is "in essence lawful."62But if it is agreed that a man
is entitled, prima facie, to use force in self-defense, what counts as conduct
which is in essence lawful has already been established by independent means.
The only issue then left is whether that plea of self-defense can in turn be
overriddenby pointing to the defendant's motive. On the arguments already
advanced, they cannot.
There are of course good exceptions (at the fifth stage of argument) to
the plea of self-defense. One of the most important allows the plaintiff to
show that the defendant used excessive force in the defense of his person or
property. The plea recognizes that self-defense, which legitimates the use of
force, also limits its use; retaliation is not permitted under the guise of self-
defense. The defendant is, however, allowed to put himself first, and need
concern himself with the plaintiff's welfare only after he has vindicated his
own interests, as here the law maintains, as indeed it should, the sharp dis-
tinction between aggressor and victim. No close cost-benefit analysis is in-
voked in order to insure that the defendant's use of force will advance the
social good; no requirementis imposed upon him to minimize the total harm
to the two parties.63 Indeed the defendant is allowed to use force even though
the resulting harm to the plaintiff is greater than the benefit to himself, as
when he must kill the plaintiff in order to save himself from a serious but not
deadly attack that is by no means certain of success. And where the de-
fendant exceeds his justification, the appropriate measure of damages is no
longer the total harm caused, or for that matter the harm both caused and
intended. It is only the harm caused in excess of that necessary for self-
defense.64
The argument has not been drawn to a close. It is still possible in a sixth
stage of argument for the defendant to override the plaintiff's last plea. As
stated, the plea of excessive force is strict: it demands of the defendant that
he use the best conceivable means to protect himself from attack, and holds
him responsible for all harm to the plaintiff that could have been avoided if
that standard had been met. As such it demands that the defendant be re-
quired to choose the best alternative on behalf of a person whose conduct
has made it most difficult for him to do so.65 As a matter of fairness between
the parties, the defendant should be able to show that he did all within

61 Prosser, Torts 24. The maxim also has great importance in connection with cases of
harm to trade or advantageous relations. See infra, at 437-38.
62 Prosser, Torts 24.
63 The same anti-utilitarian bias is also found in Fletcher, Proportionality 374.
64 See Restatement (Second) of Torts ? 63, comment 1, ? 82, for the same rule.
65 For a parallel argument in connection with the exceptions to the plea of assumption
of risk, see Epstein, Defenses 195-97.

This content downloaded from 129.127.145.240 on May 08, 2018 20:13:09 PM

All use subject to University of Chicago Press Terms and Conditions (http://www.journals.uchicago.edu/t-and-c).
418 THE JOURNAL OF LEGAL STUDIES

his power to minimize the harm to the plaintiff once he protected his own
interests. The plea as stated is subjective, and allows the defendant to take
into account all of his personal weakness and problems that would be of no
consequencein a simple case in which he inflicted harm, not upon his attacker,
but upon a stranger. Rejected at the second stage of argument, these personal
excuses are material for now they show why the defendant should not be
held for using means not generally adopted in self-defense.66
The position just taken is not uniformly accepted. Many accounts of the
limitations on self-defense insist that the defendant act as a "reasonable"
man, as a man of "ordinary firmness," in response to an attack."6 The
reasonableman of the law of negligence becomes the model victim in cases of
self-defense. But the preferred result surely must be that the victim of an
attack must do, as against his attacker, only that which is within his power
to do. The reasonableconduct of the reasonableman might offer one guide to
the question whether the defendant tried to avoid unnecessary harm to the
plaintiff, but the ultimate judgment should be on the subjective standard,
leaving the defendant free to show why he could not behave like the reasonable
man. With good-faith effort as the appropriate standard, the defendant's
motive, though immaterial in itself, now counts as relevant evidence on the
question of whether the defendant sought to avoid the use of excessive force.

866This approach allows the latter-day reconciliation of two early English cases that
are of great importance on the question whether the early common law adopted a theory
of negligence or strict liability. In the Thorns case, Y.B. Mich. 6 Ed. 4, f. 7, pl. 18 (1466),
as reprinted in C. H. S. Fifoot, History and Sources of the Common Law 195 (1949), the
defendant cut some thorns that fell on the plaintiff's land, and was held for the damage
he caused both by their cutting and by his entrance on defendant's land to recover them.
During the argument of the case, Choke, J. suggested that the defendants plead that
they did all within their power to hold the thorns in, the kind of excuse rejected by his
fellow judge, Littleton. Littleton is right, I believe, on the facts, because the excuse sug-
gested by Choke can be raised only in the second stage of the argument, there being
no independent ground of defense. On the insufficiency of that plea see Epstein, Defenses
172. That excuse, however, does have much more appeal in Mitten v. Faudrye, Popham
161, 79 Eng. Rep. 1259 (K.B. 1626), where the defendant was not held liable when his
dogs chased away sheep that had trespassed upon his land, because he did all that was
within his power to keep them in. This case can be read to support the more "lenient"
interpretation of Thorns urged by Choke. See C. H. S. Fifoot, supra, at 190. That line
is indeed hinted at by Crew, C.J., who calls the Thorns case a "hard" case. Popham 162,
79 Eng. Rep. 1260. Yet the two cases can be clearly distinguished, because only Mitten v.
Faudrye involves the use of excessive force in the defense of property to which the
response of "best efforts" can be raised in the sixth stage of argument. The point is
important because the use of staged pleadings allows a defender of the basic principle
of strict liability (in the prima facie case) to place principled limitations upon its scope.
67 Restatement (Second) of Torts ? 63, comments i & j. The Restatement does hedge
the position somewhat by taking into account "the fact that the other's conduct has put
the actor in a position in which he must make a rapid decision." Id. ? 63, comment j.
But there is still no complete shift to the subjective position because the next sentence
demands that we evaluate the defendants against "what a reasonable man in such an
emergency would believe permissible. . . ." Id. ? 63, comment j.

This content downloaded from 129.127.145.240 on May 08, 2018 20:13:09 PM

All use subject to University of Chicago Press Terms and Conditions (http://www.journals.uchicago.edu/t-and-c).
 INTENTIONAL HARMS 419

But even on this view the defendant's motive is not an ultimate issue on the
question of liability. It becomes material only if the plaintiff can first establish
that the defendant's force was excessive. Even then it only helps raise an
uneasy inference that the defendant yielded to his worst instincts when he
used excessive force in self-defense.68
The questions about the excessive use of force are, I think, susceptible to a
straightforwardsolution because they do not require us to "balance" the in-
terests of the plaintiff against those of the defendant. The problems of balance
create, however, grave difficulties when the question shifts to whether the
defendant may use "deadly force" required in self-defense. Take the simple
case where the defendant is able to defend some odds and ends of little value
only by killing or maiming the plaintiff, who wants to take them.69Excessive
force is not the issue, but deadly force is. Here it is quite possible to take
the position that defendant should be entitled to use deadly force. The princi-
ple that controls is that right need never yield to wrong, where the justifica-
tion of self-defense has shown the defendant to be in the right and the
plaintiff in the wrong.70To deny the use of deadly force necessarily allows
the plaintiff to profit from his own wrong even when the defendant is able
to stop him.
To be sure, in this extreme case it is most awkward and disagreeable to
argue for the use of deadly force.7 Yet the utilitarian argumentat the root of
the discontent lacks a cutting edge. It cannot be generalized to cover the cases
that arise as the value of the defendant'sinterests continue to increasein value.
We could try to "balance" the interests of the parties in question, but that
balance is not a straight economic one, as the defendant is not required to
yield to force whenever the benefits to his aggressor outweigh the costs to
him, even if he retains an action for damages of uncertain worth. Thus while
the rule that allows the defendant to use deadly force in defense of his
dwelling house (any dwelling house) might be taken to represent only the

68 In certain situations, moreover, it is possible to carry the argument a step further.

Suppose the defendant was a police officer required by statute to take particular steps in
response to a violent attack. In that case, it may well not be dispositive that he acted
in the best of faith, given his duty to conform to particular standards. Here, therefore,
it is proper to allow the plaintiff to show that all or some of the harm could have been
avoided if the defendant had complied with the appropriate standards of care. The case
thus turns on a kind of negligence standard after the argument winds its way through
seven stages. Here the appropriate duties in question are not judicially created under the
general foresight doctrines, but are derived instead from specific statutes or regulations.
As the theory of the case has thus changed, the plaintiff's appropriate measure of damage
is that harm which could have been avoided had the requisite care been taken.
09 Cf. Katko v. Briney, 183 N.W. 2d 657 (Ia. 1971).
70 See Fletcher, 378-79, for a comparative law discussion of the position.
71 In Katko v. Briney, 183 N.W. 2d 657 (Ia. 1971), the jury not only found for plain-
tiff, but also awarded punitive damages.

This content downloaded from 129.127.145.240 on May 08, 2018 20:13:09 PM

All use subject to University of Chicago Press Terms and Conditions (http://www.journals.uchicago.edu/t-and-c).
420 THE JOURNAL OF LEGAL STUDIES

utilitarian judgment as to the total costs and benefits of the rule, more likely
it rests at least in part on the general principle that no assailant should be
able to force his victim to sacrifice his own substantial interest for the sole
benefit of the assailant. The rule that a man may stand and fight to protect
his honor, although often rejected in recent years, stems from that same basic
principle.
The problem is insoluble in general terms because there is no common
measureby which to reconcile two concerns.The calculus of costs and benefits
does not take into account the rights and the wrongs of the case, and the
strictures against sacrifice do not take into account its costs and benefits.
We must proceed therefore with caution to identify (in the sixth stage of
argument) the kinds of interests that may be protected by deadly force. The
distinction between protection of a dwelling house and of an empty shack, or
that between serious personal injury and personal honor, raise questions of
degree now transmutedinto distinctions in kind. There is little we can expect
by way of general theory once the easy paradigms are set out. To "balance"
may be the distinctive feature of negligence law; but the few cases in which
it is truly unavoidable show its weakness as a general technique for solving
legal problems.
One other possible limitation upon the plea of deadly force deserves special
mention. The justification of self-defense does not require a showing that the
plaintiff's attack was meant to cause harm. Therefore the issue of plaintiff's
mental state still may be introduced into the argument (at the sixth stage)
to justify the use of deadly force in defense of property even in cases where
the property is not of substantial worth. Here the defendant in effect seeks
to imitate the plaintiff's argument, which in the third stage rested on the
defendant's intention to harm him. Yet the plaintiff's use of the defendant's
intention only justified a money judgment for the harm deliberately caused.
In this context the defendant's use of the plaintiff's intention is seized upon,
not to sanction a money judgment, but to justify the killing of another per-
son, and then for acts which in themselves constitute but minor offenses under
the criminal law. In the end therefore the plaintiff's intention does not have
much weight in this context, but should at best serve as a minor offset to
the plaintiff's recovery once the plea of deadly force is incorporatedinto the
system.
2. Consent. The second principledjustification for the intentional infliction
of harm is consent. In the usual case it is made out in accordance with the
general principles applicable to all consensual arrangements, such that the
overt expressionof consent will be sufficient, unless it is known by the other
party that the consent was not intended.72Those cases which raise difficulties
72 See, e.g., Cunard Lines v. O'Brien, 154 Mass. 272, 28 N.E. 266 (1891), where the
court found that the defendant had consented to a vaccination when she held out her
arm to an immigration official who administered the shot, even though she did not want it.

This content downloaded from 129.127.145.240 on May 08, 2018 20:13:09 PM

All use subject to University of Chicago Press Terms and Conditions (http://www.journals.uchicago.edu/t-and-c).
 INTENTIONAL HARMS 421

arise when the defendant tries to establish the plaintiff's consent by referring
not to an agreementbetween the parties but to the combination of the plain-
tiff's action and knowledge.
Let us return to the case where the defendant sets a spring gun that is set off
by an intruder who knows of its use. The issue of consent arises in the
following manner. The plaintiff first alleges that the defendant created a
dangerous condition that caused him harm, a case to which the plaintiff's
setting off the device is no defense. The defendant could now plead in his
defense either that the plaintiff assumed the risk or trespassed against his
land. In either event he could be met with the successful reply that the harm
was deliberately inflicted, given his knowledge of the substantial prospect that
harm would be caused. Nonetheless the plaintiff could be met with the justifica-
tion of consent if he had actual notice of the gun's use.73 Knowledge alone is
not the equivalent of consent but the knowledge coupled with the plaintiff's
own decision to enter that land after he knew of its dangerouscondition could
well be treated as consent. In their treatment of the question Bohlen and
Burns argue that there is no consent to the injury because the trespasser
merely braves a danger. As such, they continue, he does not "deliberately
encounter" the certainty of danger necessary to make out consent by impli-
cation.74 But the point about certainty cannot be decisive in this context.
When the defendant hurts the plaintiff with his mechanical device, he does so
deliberately, even though the outcome is not certain; assume otherwise, and
the plaintiff's trespass or assumption of risk will be sufficient defense to the
prima facie case, and one to which the plaintiff has no reply. If the substantial
knowledge of the defendant establishes his intention to harm, it is certainly
arguable that the same degree of knowledge should make out the plaintiff's
consent, if he continues with his trespass.
The plaintiff's consent, once established, still may of course be overridden
by misrepresentationor duress.7"As with assumption of risk, duress or mis-
representation does not deny the fact of consent, but only gives a reason
why the consent should not be operative in the particular case. The distinction
between the denial and the exception is not merely verbal. Consider the case
where a third party is the source of the duress or misrepresentation.Suppose
X told the plaintiff that he had to submit to the defendant's medical experi-

73 If the notice were posted, but not seen, the issue of consent (or for that matter
assumption of risk) would have to be decided for the plaintiff. Note, too, that notice
itself is immaterial in terms of our general theory of tort because it looks only to the
reasonableness of the defendant's conduct, and not to the way in which that conduct
affects, as a matter of fact, the relationship between the parties. Where the notice is
prominent, however, or the dangerous condition apparent, it will be most difficult for
the plaintiff to show that he did not have the requisite knowledge.
74 Bohlen & Burns, at 538.
75 See, e.g., Fowler V. Harper & Fleming James Jr., The Law of Torts 40 (1956) [here-
inafter cited as Harper & James].

This content downloaded from 129.127.145.240 on May 08, 2018 20:13:09 PM

All use subject to University of Chicago Press Terms and Conditions (http://www.journals.uchicago.edu/t-and-c).
422 THE JOURNAL OF LEGAL STUDIES

ments certain to result in physical harm or suffer in the alternative some

greater harm at X's hands. After the plaintiff submits to the defendant's ex-
periments,he brings an action for assault and battery, to which the defendant
respondswith the defense of assumptionof risk. Once the plaintiff alleges that
the defendant intended to harm him, the case will then turn on the plea of
consent. If the issue of duress could be raised on a general denial, then the
plaintiff would prevail. (Indeed the case would never reach the fourth stage
because the plaintiff could deny as well the plea of assumption of risk.) But
once it is recognizedthat the source of the compulsion is crucial, the plaintiff
has no exception to the plea of consent, even though he has an action
against X for compelling the consent to the experimentin question.76
Duress and misrepresentationare not the only exceptions to the defense of
consent. Indeed, in many consensual situations the defendant is under a duty
not only to tell the truth, but to disclose to the plaintiff all material facts
that may influence his decision to give consent.77The most important context
of this duty to disclose is that of medical malpractice. There the basic legal
position is that the patient is generally entitled to know about all material
risks inherent in the proposed course of treatment.78The duty is said to rest
upon the autonomy of the patient, upon his right to control what is done
with his body, no matter how unwise.79So defended, the principle of "informed
consent" has a strong appeal, particularly to those who believe that the
protection of an individual autonomy is the first end of tort law. But there
are dangers as well. Viewed as a tort question, the physician's duty to disclose
is but one outgrowth of the general duty of care imposed by the law of negli-
gence. There is, however,anotherview of the question that restricts application
of the doctrine of informed consent. The relationship between physician and
patient is at bottom consensual, much like that between the buyer and seller
of goods. The appropriate function of the law in consensual arrangementsis
to approximate those terms which the parties themselves would have chosen
if they had specified all of the incidents of their relationship.On this view, the

76 For parallel arguments with assumption of risk, see Epstein, Defenses 192.
77 Logically, the discussion that follows should be viewed as concerning the possible
exceptions to the defense of assumption of risk, since there is, it is clear, no intention to
harm the patient. See the discussion of Mohr v. Williams, supra at 396, 407. Yet it is
universally discussed under the heading of informed consent, a term justified only because
attention is placed on the consent to the operation and not to the injury.
78 For the early recognition of the doctrine, see Schloendorff v. Society of the New York
Hospital, 211 N.Y. 125, 105 N.E. 92 (1914). For recent cases that strongly support this
general position, see, e.g., Canterbury v. Spence, 150 U.S. App. D.C. 263, 464 F.2d 772
(1972); Natanson v. Kline, 186 Kan. 393, 350 P.2d 1093, rehearing denied, 187 Kan. 186,
354 P.2d 670 (1960); Wilkinson v. Vesey, 110 R.I. 606, 295 A.2d 676 (1972). See gen-
erally Informed Consent and the Dying Patient, 83 Yale L.J. 1632 (1974), [hereinafter
cited as Informed Consent], and material cited at id. 1633 n.8.
79 See, e.g., Natanson v. Kline, 186 Kan. 393, 406-07, 350 P.2d 1093, 1104 (1960).

This content downloaded from 129.127.145.240 on May 08, 2018 20:13:09 PM

All use subject to University of Chicago Press Terms and Conditions (http://www.journals.uchicago.edu/t-and-c).
 INTENTIONAL HARMS 423

physician-patient relationshipdiffers from other consensual arrangementspre-

cisely because the parties to it generally have a different view of their legal
relations. The rules requiringdisclosurein medical cases representa good first
approximationof the respective rights and duties of physician and patient, and
do so in a way in which the regime of caveat emptor does not. Most patients
want to know the relevant facts, and most physicians doubtless want them to
know them.
The crucial point, however, concerns the relationship between the rules of
tort and contract. The approach here allows the parties to individuate their
relationshipand thereby to supplant the "tort" rules fashionedby the courts;801
and it does not matter that the private agreement between physician and
patient restricts the scope of the duty of disclosure. The autonomy of the
physician is at stake every bit as much as that of the patient; and it is violated
whenever a duty he has disavowed is imposed upon him for his patient's
benefit. It has been said that waivers of the right to the appropriatedisclosure
should be viewed with suspicion, given that the patient stands in awe of his
physician.8' Yet as stated the argument surely proves too much. If the physi-
cian has such power, then why will he be unable to obtain the patient's con-
sent, even if he makes all of the necessary disclosures? If anything, it seems
more likely that a patient can ask for the information even if unable to
evaluate it once received. The relationship between physician and patient is
difficult enough as matters now stand; it is only made more difficult if treated
as a problem of public policy, and not of private arrangement.

IV. EcoNOMIC HARMS

A. The Prima Facie Tort
Thus far our inquiry has focused upon the intention to harm only in cases
of physical harms. Its place in the law of tort, in particular its relationship
to the problems of causation and justification, can be better understood if
studied in yet another context, that of deliberate infliction of economic losses
wholly apart from physical injury. In cases of this sort it is often said that the
intentional infliction of economic loss gives rise to a "prima facie" tort, one
that bears close resemblanceto the intentional infliction of physical harm or of
emotional distress. In this part I hope to show that this resemblanceis false

80 In this context, it is possible to use the standard of who can "best avoid the costs"
associated with the illness or injury to formulate the standard set of implied terms, as it
is not difficult to assume that the patient and physician enter into their relationship for
their mutual benefit. See Informed Consent 1645-47. See generally Guido Calabresi, The
Cost of Accidents (1970).
81 Canterbury v. Spence, 150 U.S. App. D.C., 263, 274 n.36, 464 F.2d
772, 783, n.36
(1972).

This content downloaded from 129.127.145.240 on May 08, 2018 20:13:09 PM

All use subject to University of Chicago Press Terms and Conditions (http://www.journals.uchicago.edu/t-and-c).
424 THE JOURNAL OF LEGAL STUDIES

at every crucial point, at least if the approachtaken toward physical injuries

in this series of articles is sound.
The classic case of Mogul v. McGregor82provides a convenient vehicle for
examining the common law principles governing economic harms. There the
defendant shipownersentered into an agreementamong themselves, the object
of which was to obtain exclusive control of the "tea harvest" shipped on the
China-to-Englandrun. Their agreement called for the firms to carry tea at
a very low rate in order to make it unprofitable for outsiders to compete
successfully with them. As an additional part of their scheme, the defendants
offered a five per cent rebate on each shipment.83Under the terms of their
offer the entire rebate was, however, to be forfeited in the event that the
customer at any time shipped his tea by a carrier not a member of the defen-
dant's association. The plaintiff was a shipping firm that was denied entrance
to the defendants' association. It commencedits action for loss of trade after
the defendant's strategy had made it impossible for it to conduct its own
trade at a profit. As the suit was for harm between strangers, the plaintiff
could recover, if at all, only in tort. The case went to the House of Lords,
where it was unanimouslydecided that the plaintiff'scomplaint stated no cause
of action at common law. The most famous exposition of the governing
principles, however,is contained in the judgment of Bowen, L.J., in the Court
of Appeal:

All personalwrongmeansthe infringementof some personalright. "It is essential

to an actionof tort,"say the Privy Councilin Rogersv. RajendroDutt, "that the
act complainedof shouldunder the circumstancesbe legally wrongfulas regards
the party complaining;that is, it must prejudiciallyaffect him in some legal
right; merelythat it will, howeverdirectly,do a man harm in his interest,is not
enough."... Now, intentionallyto do that whichis calculatedin the ordinarycourse
of eventsto damage,and whichdoes, in fact, damageanotherin that otherperson's
propertyor trade, is actionableif done without just cause or excuse. Such in-
tentionalaction when done without just cause or excuse is what the law calls a
maliciouswrong.The acts of the defendantswhich are complainedof here were
intentional,and were also calculated,no doubt, to do the plaintiffsdamagein
their trade. But in order to see whether they were wrongfulwe have still to
discuss the questionwhetherthey were done without any just cause or excuse.
Such just cause or excusethe defendantson their side assert to be found in their
own positive right (subject to certain limitation) to carry on their own trade

82 23 Q.B.D. 598 (1889), aff'd, [1892] A.C. 25.

88 The plaintiff tried to call these provisions "penalties" in order to cast the pall of
illegality over them. But Bowen rightly noted that they could not be accurately described
as such, as they represented inducements to customers that the defendants were not
obligated in the first instance to make. 23 Q.B.D. 598, 611 (1889).

This content downloaded from 129.127.145.240 on May 08, 2018 20:13:09 PM

All use subject to University of Chicago Press Terms and Conditions (http://www.journals.uchicago.edu/t-and-c).
 INTENTIONALHARMS 425

freely in the mode and mannerthat best suits them, and which they think best
calculatedto securetheir own advantage.84

Bowen then accepted the defendants' account of just cause, particularly

since "there was here no personal intention to do any other or greater harm
than such as was necessarily involved in the desire to attract to defendants'
ships the entire tea freight of the ports, a portion of which would otherwise
have fallen to the plaintiff's share."85
The general actionability for the deliberate infliction of harm makes appar-
ent sense if the intentional tort is treated as the archetypical tort in physical
injury cases. Yet that parallel fails once we adopt the position, urged here,
that the defendant's intention to harm the plaintiff forms no part of the prima
facie case, but becomes material only in reply to a good affirmativedefense.
It might be argued that the general rule should be displaced in Mogul v. Mc-
Gregor because it involves the protection of trade, a species of intangible
property. Yet that is not the case. Patents and copyrights, for example, create
property rights in their owners that cannot be reduced to physical possession.
Yet once their subject matter is identified, they share the essential features of
tangible property in that they too create rights of exclusive use and control
good against the entire world. These interests, moreover, are protected by
actions against infringement that are strict in form, as the defendant's in-
tention to appropriateplaintiff's copyright or patent for his own use forms no
part of the prima facie case.86Again, the recent proposals to create a system
of private property rights in the electromagneticspectrumcontemplate actions
that are strict in form, although here too only intangibles are involved.87
Bowen's version of the prima facie tort also requires us to give an accurate
account of the interest in trade to be equated with an interest in property. In
the sense already mentioned, the right of property gives to its owners the
exclusive possession or use of the subject matter in question, a right negative
in form that imposes upon strangers only the correlative duty to forbear,

84 23 Q.B.D. 598, 613 (1889). For a sympathetic restatement of Bowen's position see
Harper & James ? 6.12. The same line is taken by Holmes, in his Privilege, Malice
and Intent, 8 Harv. L. Rev. 1 (1894). He argues that the trade cases are all cases of
"temporal damages" that turn on the question of justification. Id. at 2-3. He finds that
justification in the "policy" of "privilege," based upon the belief that the defendant's
activity did in general more harm than good. Id. at 3-7. Holmes' position is subject to
the general criticisms made of the Bowen opinion.
85 23 Q.B.D. 598, 614 (1889).
86 Anthony W. Deller, Deller's Walker on Patents ? 507-514 (1972); Melville Nimmer,
Nimmer on Copyright ? 148 (1964).
87 See Arthur S. De Vany, Ross D. Eckert, Charles J. Meyers, Donald J. O'Hara &
Richard C. Scott, A Property System for Market Allocation of the Electromagnetic
Spectrum: A Legal-Economic-Engineering Study, 21 Stan. L. Rev. 1499, 1540-42 (1969).

This content downloaded from 129.127.145.240 on May 08, 2018 20:13:09 PM

All use subject to University of Chicago Press Terms and Conditions (http://www.journals.uchicago.edu/t-and-c).
426 THE JOURNALOF LEGALSTUDIES

and not an affirmativeduty to act. The right to trade,if it meansonly the

right to offer one's property or labor to another on whatever terms he sees
fit, conforms as well to this pattern of negative rights, for it too does not
impose upon a stranger the correlative duty to accept the offer as stated. Yet
if the right to trade is so construed,then the defendants in Mogul v. McGregor
collectively have not infringed upon or interfered with the plaintiff's interest
in trade. It can still make whatever offers for carriage it chooses even after
the defendants have formed their combination to control the China tea trade.
It is true that they will not be accepted, but of that the plaintiff cannot com-
plain as no one is bound to accept them on his behalf.
In Mogul v. McGregor,however, Bowen proceeded on the assumption that
there was a good prima facie case. That assumption requires us to redefine
both the nature of the plaintiff's right to trade and the acts that constitute
actionable interferencewith it. To turn first to the definition of the right, it is
best understood as the right to trade successfully, the right to trade at some
sort of profit. Where the right is defined in this manner, however, it is no
longer possible to preserve the close parallel between rights of property and
rights of trade which Bowen took for granted in his articulation of the prima
facie tort. In order to vindicate the right to trade in this second and broad
sense, it is necessary to identify some class of individuals under a correlative
duty, not to forbear, but to contract with the plaintiff, for without that
guaranteed custom it is difficult to see how the plaintiff can make the profit
assuredhim under the broaderdefinition of the right. It is therefore no longer
possible to say that this form of property right is negative in form and good
against the entire world. In truth it imposes upon part of the world (though
there is no easy account of which part of the world) an affirmativeduty to
contract.88
The uneasiness with the right in its new and broad form is furthered,more-
over, when we ask the kinds of act that must be taken to constitute actionable
interferencewith it. Mogul v. McGregor, as Bowen was careful to point out,
does not involve the threat or the use of force; defamation or misrepresenta-
tion; or the inducement of breach of contract.89The only source of inter-
88 23 Q.B. 598, 614 (1889). It is possible to develop an alternative account that gives
the plaintiff the exclusive right to make offers of carriage on that particular run, one
that gives it, as it were, an exclusive franchise at common law. That formulation allows
us to escape the problem of identifying those who are under an affirmative duty to deal
with the plaintiff, but it is in turn subject to insuperable objections of its own. First,
why should the plaintiff be entitled to that franchise? Why not the defendants, or some
of them, or some third party? Second, assume that we can decide which of those who
desire it should receive it, how then do we justify restrictions placed upon those who
might prefer to deal with other carriers? Both of these objections point out the havoc
that is done to any semblance of the equal distribution of rights amongst private persons.
The franchise argument was not pressed in Mogul v. MacGregor.
89 23 Q.B.D. 598, 614.

This content downloaded from 129.127.145.240 on May 08, 2018 20:13:09 PM

All use subject to University of Chicago Press Terms and Conditions (http://www.journals.uchicago.edu/t-and-c).
 INTENTIONAL HARMS 427

ference with the plaintiff's rights is the offers they have made to attract
people who might otherwise have done business with the plaintiff, rightly
characterizedby Bowen as "competition to the bitter end." Yet that formula-
tion of interference demands that the plaintiff show why those persons to
whom the two offers are directed should be bound to accept that offer which
they wish for whatever reason to reject. It also raises in stark form the
fundamental question of formal equality of rights between traders, indeed
between all private parties. To see the point one must ask how the plaintiff
would wish to treat the case if the roles of the parties were reversed, say in
a counterclaimby the defendant for the loss of his trade to the plaintiff. Can
the defendant turn the plaintiff's case against him and argue that his right to
trade at a profit is protected by the law, while the plaintiff's offers to third
parties constitute acts of actionable interference?
Given the broad definition of the right to trade and the kinds of action
which must be held to constitute interference with it, some limitations must
be placed upon the plaintiff'scase if trade itself is to flourishwithout a constant
stream of lawsuits. The first limitation Bowen suggests forms part of the
prima facie case: the defendant must deliberately interfere with the plaintiff's
trade. While that requirementcould serve, if desired, as an effective limitation
on recovery in cases of physical injury, it is doubtful that it can perform
that same role in the trade cases. The very object of trade is to take customers
away from a rival. It is difficult, therefore,even to conceive of a case in which
the defendant acted without specific intention to win the plaintiff's customers
away or, at least, without the knowledge that his success is certain to insure
that result.
The second limitation-the one that really matters-upon the prima
facie tort is contained in the justification open to the defendants. According
to Bowen, they can establish "just cause" by showing that they sought to
advance their own self-interest by carrying on their own business at a profit.
Admit, however, that the plaintiff has stated a good prima facie case, and
the justification offeredis insufficient.Self-interest is a good (the best) reason
to search for legal arguments to support one's position, but it is not a sub-
stitute for those legal arguments. The defendant cannot justify inflicting
physical harm upon the plaintiff in order to advance his own self-interest.
The existence of a private necessity does not excuse the defendant from paying
for the harm he has caused, even if he has profited thereby. A man cannot take
another'sgoods to better his own lot. Every thief could say "nothingpersonal"
to his victim. Likewise, it is immaterial if the taking was for the benefit of
society at large and not the defendant alone. It will not help a thief to give
the stolen property as alms to the poor. In such a case its owner has two
causes of action, one in trespass for the taking, and the other in conversion or
restitution against the subsequent takers.

This content downloaded from 129.127.145.240 on May 08, 2018 20:13:09 PM

All use subject to University of Chicago Press Terms and Conditions (http://www.journals.uchicago.edu/t-and-c).
428 THE JOURNALOF LEGALSTUDIES

The weakness of Bowen's frameworkis further revealed when we examine

his treatmentof the question of malice. In the physical injury cases, the plain-
tiff could not override the justification of self-defense, for example, by show-
ing that the defendant acted out of spite or ill will. In that context malice was
material only as evidence on the question of whether the defendant did all
that was in his power to avoid harm to the plaintiff, once it was established
that the force used was in excess of that necessary for the task of self-defense.
In this context, however,malice is said to play a differentrole, as an exception
to a justification premised upon the plaintiff's self-interest. Yet if the argu-
ments about malice in the physical injury cases are correct, the malice should
be an immaterial issue in connection with economic harms as well. Once
it is settled that there is justification for whatever reasons to harm someone
the defendant likes, it is impossible to see why the defendant should be
stripped of that justification when the plaintiff is someone whom he despises.

B. The Causal Analysis

There is to my mind no doubt that the result reached in Mogul v. Mc-
Gregoris correct. The task is to show why it is correct given the principles of
liability developed for both accidents and deliberate harms. In order to do
that we must make a radical shift in orientation. We must not, as did Bowen,
consider the case as turning on the justification for harm deliberately inflicted.
Instead we must underminethe prima facie case,90by showing in causal terms
that the plaintiff cannot make out the invasion of a protected interest, here
the right to trade only in its narrow sense, as the right to make offers on
whatever terms the plaintiff sees fit.
1. Causation and the Narrow Right to Trade. It requires only an easy
extension of the basic causal paradigms to move from the protection of
property to the protection of trade, when that term refers only to a person's
right to make offers to the world on whatever terms and conditions he chooses.
In the famous case of Tarleton v. McGawley,91the defendants fired shots at
some canoes occupied by African natives who were heading toward the plain-
tiff's ships intent upon trade. Several of the natives were killed, and the
others retreated to land, causing the plaintiffs to lose the benefits of the
expected exchanges, not only for that day but for the entire trading season.

9o0" '[E]xcuse or justification' is only needed where an act is prima facie wrongful."
Allen v. Flood [1898] A.C. 1, 128 (per Lord Herschell).
91Peake 270, 170 Eng. Rep. 153 (K.B. 1793). Indeed, the identification of the plain-
tiff's protected interest was taken for granted in the case. The defendant rested his case
in large measure on a plea in avoidance, that the plaintiffs' action was barred because
they had not first obtained the permission from the local king to trade with his natives.
That plea was rejected on the ground that the law in question was a jus positivum that
should be enforced by the king, not the defendant.

This content downloaded from 129.127.145.240 on May 08, 2018 20:13:09 PM

All use subject to University of Chicago Press Terms and Conditions (http://www.journals.uchicago.edu/t-and-c).
 INTENTIONAL HARMS 429

The plaintiff's interest in trade here only involves his right to make offers on
terms he thinks fit but, given the arguments made in connection with Mogul
v. McGregor, it is still an interest entitled to protection. With his interest
established, the plaintiff has a cause of action based on the theory that the
defendant used force to drive away those who wanted to trade with him.
Lord Kenyon in Tarleton v. McGawley, asserted that the action lay only
because the harm inflicted was deliberate.92There is, however, no reason why
the cause of action should not be strict, even if few cases of the forceful in-
terference with trade take place by accident. The intention to harm forms,
as with physical injuries, no part of the prima facie case; it is material only
by way of reply to override a valid affirmativedefense.
The position just taken clashes with the current law, which denies the
plaintiff recovery for economic losses even where the defendant acted negli-
gently. The result is sometimesdefended on the causal ground that the damage
is too remote.93The negligence test of "reasonable foresight," however, sug-
gests the opposite conclusion, as it is most common for parties with expensive
plant or equipment to enter into contracts for its use. If the "foresight" test
is conveniently put aside, the remoteness question can be approachedon ex-
plicit causal grounds. Yet the causal argumentsapplicable in Tarleton v. Mc-
Gawley still decide the case for the plaintiff. The conclusion of remotenessis,
therefore, perhaps best understood as a covert policy judgment in favor of
barring recovery, based upon the fear of exposing the defendant to multiple
lawsuits each time he damages the person or property of another.
The differentcontentions are well illustrated in the Georgiacase of Byrd v.
English.94 There, while the defendant's servants were excavating on the
defendant's land, they (negligently) built up an unstable wall of earth
(causation by creation of dangerous conditions), which fell upon and broke
the wires of the electric company that carried electric current to the plaintiff's
plant. The plaintiff was forced to suspend the operation of his plant for
several hours until the wires could be repaired,no alternative source of power
being available. He then sued the defendant for the economic loss suffered on
account of the delay.
Of crucial importance in the case is that the plaintiff's contract with the
electric company exempted it from liability for the loss of service caused by
the acts of a third party. The case therefore is not one of the inducement by
force of a breach of contract. Instead, the situation, as in Tarleton v. Mc-
Gawley, is one where the defendant's acts make it impossible for the plaintiff
to deal with a third party on terms that work to their mutual benefit. Given
92 Peake 270, 273, 170 Eng. Rep. 153, 154 (1793).
9 See, e.g., Anthony v. Slaid, 52 Mass. (11 Metc.) 290 (1846); Byrd v. English, 117
Ga. 191, 43 S.E. 419 (1903); Harper & James 509.
94 117 Ga. 191, 43 S.E. 419 (1903).

This content downloaded from 129.127.145.240 on May 08, 2018 20:13:09 PM

All use subject to University of Chicago Press Terms and Conditions (http://www.journals.uchicago.edu/t-and-c).
430 THE JOURNAL OF LEGAL STUDIES

this identificationof the plaintiff'sprotected interest in Tarleton v. McGawley,

the case cannot be decided for the defendant on narrow causal grounds. The
force applied by the defendantcaused the plaintiff's economiclosses, no matter
what he intended.
The case must therefore be argued on the assumption that the liability in
question should be denied in order to spare the defendant from the crushing
burdensthat would follow in cases of this sort. To examine that claim, assume
that the electric company had obligated itself unconditionally to provide the
plaintiff with power.In that case, as the court conceded,95the electric company
could recover consequential damages from the defendant for the amount it
had to pay the plaintiff pursuant to their contract. If that burden will not
crush the defendant, then why cannot the plaintiff now sue him for those
same losses in his own right? The court distinguished the two cases on the
ground that the plaintiff here only has rights under his contract with the
electric company, to which the defendant is in no way connected.96But the
plaintiff's claim does not rest on the contract. His protected interest is the
right to offer to purchasepower from the electric company on whatever terms
he sees fit. The history of past agreements, discharged to their mutual satis-
faction, is needed only to establish that the loss flowed from the defendant's
interferencewith the plaintiff's right. Indeed, if the electric company had been
liable for the power failure occasioned by the defendant's acts, that would
mean only that either the plaintiff or the company could have sued for the
loss,97with the recovery of the one barring the action of the other.98
Even if this is all correct, we must still contend with the specter of multiple
suits if actions for economic loss were allowed. One protection against that
possibility is to requirethe plaintiff, as in all cases, to mitigate his damagesby
seeking alternative means of supply, a possibility which was not open in this
particular case. That single rule will in most cases eliminate or reduce to
nominal amounts recovery for economicloss, given the few situations in which
the plaintiff has but one source of supply.99Again, if mitigation is not suffi-
cient, a convenient dollar minimum could bar those trivial cases which in all

95 Id. at 194, 43 S.E. at 420.

96 Id. at 194, 43 S.E. at 420.
97 The situation is that raised by Robin's Dry Dock and Repair Company v. Flint, 275
U.S. 303 (1927). The plaintiff was the time charterer of a ship owned by a third party.
The defendant had (negligently) damaged the ship in a manner that made him liable to
its owner. The court held that the plaintiff could not recover because its damages were
too remote. See, for excellent criticism, Harper & James 500-06.
98 Id. at 504.
99 Note, too, that this result will follow even in cases of inducement of breach of
contract, for there too mitigation should be possible where the plaintiff has numerous
alternatives. This explains, perhaps, why so much of this kind of litigation involves
opera singers or movie stars, each of whom can be regarded as a unique sort of good.

This content downloaded from 129.127.145.240 on May 08, 2018 20:13:09 PM

All use subject to University of Chicago Press Terms and Conditions (http://www.journals.uchicago.edu/t-and-c).
 INTENTIONAL HARMS 431

likelihood will not be brought in any event because of the high costs of their
prosecution. Finally, as was mentioned above, the defendant should be able
to implead all possible plaintiffs in a single lawsuit to protect himself from the
risk of double payment for a single economic loss. Once these precautions
are observed, the "policy" argumentsused in Byrd v. English lose their force.
Strict liability extends with perfect symmetry to it and Tarleton v. McGawley,
and there is no reason why the legal results should not be altered to conform
to the theory.
2. Causation and the Broad Right to Trade. The causal principles ap-
plicable to physical injuries are of no further assistance to the plaintiff once
we shift to the second account of the right to trade-the right to trade at a
profit. True, the plaintiff in Mogul v. McGregor could say that "the defendant
deprived him of his business," as in one sense happened in Tarleton v. Mc-
Gawley. Yet that account of the two cases only conceals the essential differ-
ence between them, the differencebetween force and persuasion.No victim can
"walk away" from the threat or use of force and still retain his full complement
of rights. Where there is only competition in trade (whether between indi-
vidual firmsor by firms acting in concert) the situation is much more complex.
The defendant intent upon "destroying his competition" cannot do so by his
own physical acts. He must persuade those persons, who are entitled to do
business with whomever they choose, that it is better to do business with him
than with the plaintiff. On strict causal terms, the necessary presence of these
possible customers means that the case does not involve the simple two-party
situation implied by the phrase "defendant drove plaintiff out of business."
The case, therefore, presents no parallel whatsoever to the straightforward
trespass case of "A hit B" even though the plaintiff may speak, however
earnestly, of what the defendant has done to him.
The plaintiff's causal case must therefore rest, if it is to succeed at all,
on the three-party causal models applicable to cases of physical injury. Yet
here there is surely no case of compulsion; the defendant gave the third
party a choice which he preferred to the plaintiff's, but did not force him to
accept it. Likewise, the paradigm of dangerous conditions is of no assistance
to the plaintiff. Under that paradigm,the plaintiff has a cause of action against
either the third party or the defendant, with the former having an action over
in the event that he is required to compensate the plaintiff. Here, of course,
the plaintiff has no action against the third party, who is under no duty to
contract with him; thus the need for an action over never arises. Indeed, the
logic of that action, while clear in the case of physical injury, makes little
sense in this context. The defendant gave the third party a choice which he
preferred to those otherwise available to him; he did not subject him to a
set of external constraints not of his own making. We can decide the causal
question for the plaintiff only if the rational response of a third party in the

This content downloaded from 129.127.145.240 on May 08, 2018 20:13:09 PM

All use subject to University of Chicago Press Terms and Conditions (http://www.journals.uchicago.edu/t-and-c).
432 THE JOURNAL OF LEGAL STUDIES

usual trade case is part of the causal chain. But that response, being both
deliberate and independent, is a novus actus interveniens, and the plaintiff's
causal case is no longer tenable.100The damage is too remote.
This discussion of causation and economic harm does not imply that causal
arguments (and hence tort recovery) can be successfulonly where the plaintiff
has used force. But it does give us the necessary clue as to the circumstances
in which the tort law can be extended beyond cases involving force. The
essential point is that the causal paradigms in all situations in which action-
ability is warranted involve the nonreciprocal causal relation between the
defendant's conduct and the plaintiff's protected interest, be it of person,
property, or trade. So long as that intimate relationship between causation
and property is preserved, actions should be allowed, even with intangibles,
and even without the threat or use of force. In the case of patents or copy-
rights the causal question is nonreciprocalas the defendant infringes upon
the plaintiff's interest, and not the other way around. Likewise, it is the
defendant who passes off his own goods as though they were those of the
plaintiff. It is the defendant who misleads the plaintiff by his false repre-
sentations.101The defendant induces breach of contract between the plaintiff
and a third party. In all these cases the causal relations are nonreciprocal.
The use of nonreciprocalcausal arguments cannot, however, be extended
to the "right" of trade in its broad sense at stake in Mogul v. McGregor.
The plaintiff has no right to demand that third parties do business with him.
He cannot show therefore that the defendant's offers invaded his protected
interests, be they of property or trade. All that is established is that he is
left worse off by virtue of the agreements that the defendant reached with
third parties.
The common law captured the difference between being worse off and

1oo See Epstein, Strict Liability 177-89. H. L. A. Hart & A. M. Honorb, Causation in
the Law 129-51 (1959). It is, of course, these causal arguments which explain why it is
that the protection against intentional physical invasions was recognized at an early point
in the development of the law, whereas the judicial examination of the intentional inflic-
tion of economic harms only developed in the late part of the nineteenth century. Hart
and Honore do not appear to grasp the importance in causal terms of the shift from
physical to economic harms. Id. at 173-74. Indeed these separate patterns of historical
development should warn us against treating the assault and battery cases like the eco-
nomic harms cases.
101 Indeed, the misrepresentation cases can be made quite a bit more complex, as for
example, where the defendant makes a misrepresentation to a third party that induces
him not to deal with the plaintiff. Nonetheless, a quite straightforward extension of
the rules of misrepresentation is sufficient to cover the three-party situation. See, Allen v.
Flood [1898] A.C. 1, 141-43, where the House of Lords held the plaintiff stated a good
cause of action where misrepresentations were made to a third party in order to induce
him not to contract with the plaintiff. Note, too, that where there is a material mis-
representation there is no need to place reliance upon some undifferentiated notion of
"unfair" competition to permit the plaintiff his recovery.

This content downloaded from 129.127.145.240 on May 08, 2018 20:13:09 PM

All use subject to University of Chicago Press Terms and Conditions (http://www.journals.uchicago.edu/t-and-c).
 INTENTIONAL HARMS 433

the invasion of a protected interest in the Latin expression damnum absque

iniuria, harm without legal injury. Properly understood, the phrase is not
double talk; it refers to those cases in which a party is worse off economically
even though he has suffered no invasion of his protected interests. The
cleavage between the ordinary and legal use of harm and causation is, of
course, the source of endless problems. The legal system cannot be in total
harmony with the popular sentiment that uses these broad causal expressions
invoked by losers in the competitive struggle. But unless the distinction be-
tween legal and economic harm is observed, there will be no principled way
to maintain the parallel between actions for the invasion of physical and
economic interests, and no principled way to avoid the endless political ques-
tions of deciding whose economic self-interest the law must serve. We must
therefore shun the causal arguments tailored to the extended right to trade.
In all their particulars they bear no relationship to the general arguments
applicable to the physical injury cases and, for that matter, to those which
are applicable to the right to trade in its narrow sense.

C. The Two Views of Economic Harms Compared

In the examination of the trade cases we have thus far distinguished be-
tween two approaches. The first of these, that propounded by Bowen in
Mogul v. McGregor, concedes the causal question to the plaintiff, and then
decides the case on the defendant's justification of self-interest and the ex-
ception to it based upon his malice or ill-will. The alternative approach treats
the causal question-the question of what legal right of the plaintiff was
involved102--as decisive and, resolving that in defendant's favor, never
reaches the twin issues of privilege and malice. The two approachesyield the
same result in Mogul v. McGregor but require quite different results in other
situations. To show further the consequences that flow when one course is
adopted instead of the other, it is useful to consider several cases, mostly
English, that could be decided on either approach.'03
In The Mayor of Bradford v. Pickles,o04the defendant gathered rainwater
that was percolating through his own land. If he had let the water take its
natural course, it would have been drawn by gravitational forces to the city's
land, where the city could have collected it for use in its water supply. The
House of Lords held that the city's complaint did not state a cause of action
even though it alleged that the defendant gathered the water "maliciously."

102 That was the

point of the language of the Privy Council in Rogers v. Rajendro
Dutt, 13 Moore, P.C. 209 (1860), quoted by Bowen in Mogul v. McGregor, supra, at 424.
103 The English cases are discussed because they, more than the American cases, raise
the basic issues in the area. The recent American cases, moreover, turn in large measure
on statutory provisions that have reduced the importance of common law principles.
104 [18951 A.C. 587.

This content downloaded from 129.127.145.240 on May 08, 2018 20:13:09 PM

All use subject to University of Chicago Press Terms and Conditions (http://www.journals.uchicago.edu/t-and-c).
434 THE JOURNAL OF LEGAL STUDIES

The opinion noted that it made no difference whether the defendant had
acted out of sheer spite or out of a desire to collect the water in order to sell
it to the city. That assertion is not consistent with Bowen's approach to the
prima facie tort. On his view, we must first concede both the damage to the
plaintiff from the defendant's act and its justification founded upon the
defendant's self-interest. At that point the case could turn on the question of
malice, and as such it is crucial whether the defendant acted out of sheer
spite or a desire for economic gain.
Since the defendant's motive was regarded as immaterial by the House of
Lords, its decision is consistent with the alternative view, argued here, that
allows the defendant to prevail on causal groundsalone. On that view, plaintiff
can succeed only where the defendant took the plaintiff's property. That could
be shown if the water in question had become part of a stream in which
definite rights of appropriationhad been established. But percolating rain-
water when it falls belongs to neither plaintiff nor defendant until taken under
control; yet the plaintiff's cause of action for diversion requires it to show
the diversion of its water. Where it makes that showing, malice is immaterial,
since the prima facie case is, here as elsewhere, in principle strict. True, the
defendant is worse off without the water than he is with it; but the same
can be said of the plaintiff as well. To avoid the problem of reciprocalharm,
ownership first must be established to decide which economic harm is com-
pensable and which is not.1'5 Once that interest is identified, the defendant's
intention to harm (bound up in the issue of malice) becomes relevant only as
a reply to a valid affirmativedefense. It cannot make up the defect in a prima
facie case that fails because it cannot identify the plaintiff's proprietary in-
terest.
The importance of the legal interest is also revealed in the analysis in the
earlier case of Keeble v. Hickeringill.106There the plaintiff used duck decoys
to attract wild fowl to his lands, fowl he wished to capture and sell. The
defendant constantly dischargedfirearmsin order to drive the fowl away from
the plaintiff's decoy, and was successful in that end. The plaintiff brought
his cause of action for the intentional infliction of harm, and the court,
speaking through Holt, C.J., held that he had a right to recover. Under the
theory of prima facie tort, the decision is correct. The plaintiff is able to
show in a broad sense the intentional infliction of harm, given that the defen-
dant's action has made him worse off. The justification of self-interest is
weak on the facts, and in any event arguably is rebutted by malice.

105 To return to the electromagnetic spectrum for a moment, ask who interferes with
whom when two parties broadcast on the same frequency. That question can be answered
only after some system of property rights is established in the spectrum. At that point,
one party can say of the other that he broadcast on my frequency. See generally R. H.
Coase, The Federal Communications Commission, 2 J. Law & Econ. 1 (1959).
106As reported 11 East 573, n.(a), 103 Eng. Rep. 1127 (Q.B. 1706).

This content downloaded from 129.127.145.240 on May 08, 2018 20:13:09 PM

All use subject to University of Chicago Press Terms and Conditions (http://www.journals.uchicago.edu/t-and-c).
 INTENTIONAL HARMS 435

The case, however, assumes a different complexion if the plaintiff is

required to show the defendant's invasion of his protected interest. True, an
action lies if the plaintiff owned the fowl driven off by the defendant. Indeed
the theory of recovery should be strict: "defendant frightened away plaintiff's
fowl" states a good prima facie case, and references to defendant's mental
state are quite beside the point. The difficulty in Keeble is that the plaintiff
has only an economic interest, an economic desire to capture the fowl, not a
proprietaryinterest because he has captured them.107Holt was aware of the
difference, for he noted the plaintiff would have no action if the defendant
had used his own lure to attract the fowl that otherwise would have gone to
plaintiff's land. Yet doubtless even that tactic would have been improper if
the plaintiff had already taken possession of the birds. Holt also tried to
defend his position by reference to the schoolmaster cases.
One schoolmastersets up a new school to the damageof an antient school, and
therebythe scholarsare alluredfrom the old schoolto cometo his new. (The action
was held there not to lie.) But supposeMr. Hickeringillshould lie in the way
with his guns, and fright the boys from going to school, and their parentswould
not let them go thither; sure that schoolmastermight have an action for the loss
of his scholars.?08

But here too the analogy is imprecise. The case of the young scholars is
reminiscent of Tarleton v. McGawley: it involves forceful interference with
the right to do business with those who will do business with you. But in the
case at hand, the fowl did not deal with the plaintiff on a contractual basis,
much less want to do so.
The same general problem is illustrated by the important and difficult

107 On this point there would be a question whether defendant's shooting amounted to
a nuisance, such that the loss of the fowl could be regarded as consequential damages.
But while that approach might be accepted on the facts to support a judgment for the
plaintiff, it makes it unnecessary to treat the case as a cause of action based on malice
alone. See Allen v. Flood, [18981 A.C. 1, 133. See also Hollywood Silver Fox Farm, Ltd.
v. Emmett, [1936] 2 K.B. 468. Here the plaintiff had an ongoing dispute with the
defendant over the placement of a large sign on the plaintiff's land. When the plaintiff
would not remove his sign as the defendant requested, the defendant had his son shoot
his gun near the property line in a manner which was calculated to, and which did in
fact, frighten the plaintiff's vixen such that they were unable to breed. The court relied
on Keeble v. Hickeringill and allowed the plaintiff's action. Here, however, as the animals
in question were the property of the plaintiff, the prima facie case could well be strict
in form: defendant's shooting frightened the plaintiff's fowl. The question of the intention
to harm, moreover, could be crucial to the case if we accepted, at least with respect to
animals, an excuse based on the extrasensitive nature of the plaintiff's vixen. At that
point, the specific intention to harm, on the model developed earlier, could well override
this affirmative defense; and, as the defendant would be hard pressed to show any
justification for his conduct (self-interest not being an adequate justification), the plain-
tiff could recover. On this view of the case, the defendant's intention to harm moots the
importance of the extrasensitive condition of the plaintiff's vixen.
108 11 East at 576, 103 Eng. Rep. at 1128.

This content downloaded from 129.127.145.240 on May 08, 2018 20:13:09 PM

All use subject to University of Chicago Press Terms and Conditions (http://www.journals.uchicago.edu/t-and-c).
436 THE JOURNAL OF LEGAL STUDIES

case of Allen v. Flood.109The defendant Allen was the representativeof the

ironworkersunion, forty of whose memberswere in the employ of the Glengall
Iron Company. The ironworkershad long opposed having members of the
rival shipwrightsunion doing any ironworkon the ships on which they worked.
The plaintiffs, Flood and Taylor, were members of the shipwrights union
who were hired by the Glengall Company to do woodworkon its ships. When
the ironworkerslearned of that fact, they became quite upset because the
plaintiffs shortly before had done ironwork for another firm on other ships
that the ironworkersclaimed to be theirs. Allen, as their representative, told
Glengall that all of his men would walk off the job unless the plaintiffs were
fired.110The continued service of the ironworkerswas essential to Glengall's
economic survival; that of the plaintiffs was not. The firm complied with the
demands and dismissed the plaintiffs, and the action against Allen followed.
It was agreed that all of the men, both shipwrights and ironworkers,worked
on contracts terminable at will. The ironworkersdid not threaten breach of
contract when through Allen they said that they would walk off the job, and
the Glengall Company committed no breach of contract when it dismissed the
plaintiffs. The theory of the plaintiff's cause of action was that the defendant
"maliciously" induced the Glengall Company not to enter into contractual
relationships beneficial to the plaintiffs. The trial court allowed the case to
go to the jury on the strength of that theory, without taking it upon itself to
give an account of "malice." The jury returned a verdict of ?20 for each
plaintiff, and a judgment in that amount was affirmedby the Court of Ap-
peal."' The House of Lords then took the unusual step of asking eight judges
of the Queen's Bench for advisory opinions on the question whether the
plaintiff's allegation is disclosed a good cause of action. These judges, by a
margin of six to two agreedwith the courts below.112The House of Lords then
took its own counsel, reversed the decisions below, and gave judgment to the
defendant by a six to three vote.
The opinion of Lord Halsbury best reveals the logic of those who voted
to sustain the plaintiff's cause of action. The first question that he addressed
is whether any legal right of the plaintiff was infringed. His answer is clear
enough: "I think the right to employ their labour as they will is a right both
recognizedby the law and sufficientlyguarded by its provisions to make any
undue interference with it an actionable wrong."113 He then defended the

109[18981 A.C. 1. For the statement of facts in the reports, see [18981 A.C. 1, 2-4.
110 The threat was even broader in that ironworkers were prepared to walk off the job
whenever the shipwrights were allowed to do ironwork [18981 A.C. 1, 3.
111Flood v. Jackson, [1895] 2 K.B. 21.
112 [18981 A.C. 1, 11-67.
113 [18981 A.C. 1, 71. In support of this position, he quotes Sir William Erle: "Every
person has a right under the law, as between him and his fellow subjects, to full freedom

This content downloaded from 129.127.145.240 on May 08, 2018 20:13:09 PM

All use subject to University of Chicago Press Terms and Conditions (http://www.journals.uchicago.edu/t-and-c).
 INTENTIONAL HARMS 437

"right of the plaintiffs to pursue their calling unmolested" as part of that

"liberty of action" basic to the principles of the common law.114The right
thus established, he then concluded that the defendants infringed that right
by "intimidationand coercion"by threateningto walk off their jobs at a time
when their employer could ill afford to do without their services."5 On this
view of the case, the prima facie tort is made out, and given Halsbury's
characterizationof the defendant's acts, the presence of malice overrides any
justification based upon self-interest. The problem with this argument rests
in its characterization of the plaintiff's right as the right to dispose of his
labor as he will. As stated, it corresponds to the right of trade in the broad
sense discussed in connection with Mogul v. McGregor.Accept, however, that
the right to trade, or the right to work, only comprehendsthe plaintiff's right
to offer services on whatever terms he sees fit, and the case collapses."16The
plaintiffs are unable to show the invasion of a legal right of any sort or
description, as the employer was under no affirmativeduty to provide them
with jobs.
Even though it deals with unions and not firms, Allen v. Flood is but an
instant replay of Mogul v. McGregor. Stripped to its essentials, the plaintiff's
case alleges only that Glengall Iron Company, which was entitled in its own
self-interest to deal with either union, chose not to deal with the plaintiffs as
the price of dealing with the defendant's union. Glengall could not complain
because the terms were no longer as attractive as before, because it could not
dictate the terms on which the defendant's men must work when they owed
it no obligation to work at all. The plaintiff's men could not complain be-
cause their right was only to make whatever offers they saw fit, not to have
them accepted by those who find it in their self-interest to reject them. On
this view of the case, we do not have to consider the question whether the
defendant could justify his conduct in terms of his self-interest or whether
the plaintiff could prove it malicious in any of the protean senses of that term.
The model of the prima facie tort put forwardby Bowen plays no part in the
case,117and the decision of the House of Lords is correct on all its points. One
might argue that the proposition "an act lawful in itself is not made unlawful
because it is actuated by a bad motive" is not the most elegant way to defend

in disposing of his own labour or his own capital according to his own will." [18981 A.C.
1, 75, quoting Erle on Trade Unions 12 (1869).
114 [18981 A.C. 1, 72.
115 [18981 A.C. 1, 80, 83.
116 The point was anticipated quite well in counsel's argument for Allen: "What is there
unlawful in telling an employer that if he continues to employ A, B will leave his service ?
It is said that A has a right to employment. What he has is a right to make a contract
of employment if the employer desires it. . ... He has a right to sell his labour-but only
to someone who wishes to buy it." [18981 A.C. 1, 4.
117 Noted by Lord Herschell, [18981 A.C. 1, 139-40.

This content downloaded from 129.127.145.240 on May 08, 2018 20:13:09 PM

All use subject to University of Chicago Press Terms and Conditions (http://www.journals.uchicago.edu/t-and-c).
438 THE JOURNAL OF LEGAL STUDIES

the result. But if that proposition is understood as saying that the causal
defects in the prima facie case cannot be cured by alleging malice, or at least
the intention to harm, then its application is entirely proper. With both
physical and economic harms, the defendant's mental state should be material
only if the legal argumentprogresses to its third stage.
Tuttle v. Buck118presentsin a somewhatdifferentsetting the same questions
about causation, privilege, and motive raised in Allen v. Flood. The plaintiff
was a barber by trade, who for many years did a profitable business in the
village of Howard Lake, Minnesota. The defendant was a banker and a man
of large means, who out of his dislike of the plaintiff installed a succession of
rival barbersin town, his purpose being to take away the plaintiff's trade. To
achieve that end he subsidized his barbers so that they could operate at a
profit even though they charged lower rates than the plaintiff. In addition,
"by threats of his personal displeasure," he persuaded the plaintiff's former
patrons not to have the plaintiff cut their hair.
The court allowed the plaintiff's action, holding that "when a man starts
an opposition place of business, not for the sake of profit to himself, but
regardlessof loss to himself, and for the sole purpose of driving his competitor
out of business, and with the intention of retiring upon the accomplishment
of his malevolent purpose, he is guilty of a wanton wrong and an actionable
tort."119That result is correct given Bowen's account of the prima facie
tort, with malice the crucial issue. On the theory advanced here, however,
the case is wrongly decided because the plaintiff could not identify his pro-
tected interest. The case is one in which both parties have made offers to third
persons, where those of the defendant's barbersgenerally have been accepted.
Judge Elliott, who wrote the opinion for the court, regarded the complaint
as insufficientbecause it did not allege that "the defendant intentionally ran
the business at a loss to himself, or that after driving the plaintiff out of
business the defendant closed up or intended to close up his shop."120But
even if the defendant wanted to lose money, there is still no prima facie case,
because there is no interferencewith the plaintiff's right to offer his services
on whatever terms he sees fit. Again the parallels to the physical injury cases
are instructive. With both physical and economicharms the question of motive
and the question of costs and benefits are immaterial to the prima facie case,
on which the causal issue remains decisive. Only if we are prepared to accept
two theories of tort, one for physical injury and one for economic harm, is it
possible to allow the plaintiff to recover in Tuttle v. Buck.
One further variant of the problem of economic harm requires mention. It
is that of combination or, more ominously, conspiracy. Where there is no
118 107 Minn. 145, 119 N.W. 946 (1909).
119Id. at 151, 119 N.W. at 948.
120Id. at 151, 119 N.W. at 948.

This content downloaded from 129.127.145.240 on May 08, 2018 20:13:09 PM

All use subject to University of Chicago Press Terms and Conditions (http://www.journals.uchicago.edu/t-and-c).
 INTENTIONAL HARMS 439

invasion of a protected interest, plaintiffs have sought to close the gap in

their claim by arguing that, although the act of a single defendant is not
actionable in itself, that same act becomes actionable when done by many in
concert, particularly if done to hurt the plaintiff. There are hints of that
position in Bowen's opinion in Mogul v. McGregor,121 and it is at the root
of the decision of the House of Lords in Quinn v. Leathem.122
In that case, the plaintiff (Leathem) was a butcher, and the defendants
(Quinn and others) were officers in a trade union. In order to promote their
cause in a trade dispute with the plaintiff, the defendants approached one
Munce, a buyer of the plaintiff's goods, and told him that he must not do
business with the plaintiff during the dispute,123if he did not want the defen-
dants to call Munce's employees, all union members, off the job. Here, as in
Allen v. Flood, these contracts of employmentwere at will, thus ruling out any
action for inducement of breach of contract.124The House of Lords decided
for the plaintiff, holding that the defendants acted in a conspiracy that in-
terfered with the plaintiff's right of trade. Allen v. Flood was treated as a
wholly different case because "all that Allen did was to inform the employers

12123 Q.B.D. 598, 616 [1889]. It is quite clear,however,that Bowen did not think that
conspiracyshould play the role that it enjoys today in the antitrust law. Thus, for ex-
ample,he notes that the contractamong the defendantswas illegal on groundsof public
policy, but holds that the only consequenceof illegalityis that the contract could not be
enforcedby the parties to it in the event of breach.Id. at 619-20. It did not represent
the kind of "unlawfulend" that the tort law rendersactionableas conspiracyin restraint
of trade. Indeed, the late nineteenthcentury trade cases show in the main little concern
with modernantitrust-typeissues. Of particularinterest is Lord Herschell'sobservation:
"I am aware of no ground for saying that competitionis regardedwith special favor by
the law; at all events, I see no reason why it should be so regarded."Allen v. Flood
[1898] A.C. 1, 140-41.
122 [19011 A.C. 495.
123The disputeconcerned,among other issues, the questionwhether the plaintiffcould
satisfy the union's demands by having his nonunion workers become members of the
union, with the plaintiff paying all of their membershipfees. The defendantsdid not
accept these terms, both because they were displeasedthat the plaintiff's workers did
not join the union at some earliertime, and because their own membersstood in need
of jobs. Their refusalto accept these terms was describedas "reprehensible" and "mali-
cious,"but it is difficultto see, however much they inconveniencedboth the plaintiffand
his employees,why they were not in the rationalself-interestof the defendantsand their
union members.One should well rememberBowen's warning that "malice"is one of
the most slipperyof legal concepts.Mogul v. McGregor,23 Q.B. 598, 612-13 (1889), and
Lord Herschell'swarning that the use of the term without definition representsone of
the greatestthreatsto individualliberty. [18981 A.C. 1, 118.
124Note, however, that if the defendantshad threatenedthe inducementof breach of
contract by Munce'semployees,we would have a differentcase, becausenow more than
persuasionis used to secure a favorable settlement with the plaintiff. The case is like
Tarletonv. McGawley,with only one difference:the threat of inducementof breach of
contracttakes the place of the use of force. It is as though the defendant said, "I will
take your propertyif you deal with the plaintiff,"conductthat gives both a good cause
of action.

This content downloaded from 129.127.145.240 on May 08, 2018 20:13:09 PM

All use subject to University of Chicago Press Terms and Conditions (http://www.journals.uchicago.edu/t-and-c).
440 THEJOURNAL
OF LEGAL
STUDIES

of the plaintiffsthat most of theirworkerswouldleave themif they did not

dischargethe plaintiffs."125The characterization of Allen as an "informant,"
however, makes that case a trivialone that could have been decidedat trial
on a narrowquestionof fact, namely,that he had not procuredthe Glengall
Iron Companyto dismissthe plaintiffs.Yet the jury foundthat suchprocure-
mentwas the preciseeffectof Allen'swarningsto Glengall,therebygivingthe
case its greatimportanceon a point of law.126
Quinn v. Leathem, therefore, is distinguishable from Allen v. Flood only if
the defendants'combinationmadetortiousthoseacts otherwiselawfulif done
by each individually.1'2The allegationof malicein Allen v. Flood, even if
takenin the senseof the specificintentionto harmbornonly of ill will, was
not sufficientto makegood the causaldefectsin the plaintiff'scase. So here
the allegationof combinationis insufficientto make good that same defect
in Quinn v. Leathem,even if coupledwith an allegationof malice. Lord
Halsbury,who was on the Courtin Quinnv. Leathem,had a much better
sense of the problemwhen in Mogul v. McGregorhe wrote: "I think this
questionis the first to be determined:What injury,if any, has been done?
Whatlegal righthas beeninterferedwith? Becauseif no legal righthas been
interferedwith, and no legal injuryinflicted,it is vain to say that the thing
may have been doneby an individual,but cannotbe doneby a combination
of persons."128Thus, supposeone persontold Muncethat he wouldnot deal
with himif he, Munce,dealtwith the plaintiff.Is the plaintiff'sharmany less
if Muncecomplieswith that demand?If not, why no actionin the one case,
but an actionin the other?
The point is againmadeclearif we makethe obviouscomparisonsto the
physicalinjurycases.Therethe fact of combinationonly worksto maketwo
or more defendantsjoint tortfeasors.Proof that two defendantsacted in
concertdoes not enablethe plaintiffto dispensewith his causal allegations,
even if it allowshim to chargeeach defendantfor the harmjointly caused.
Combinationdoes not createliabilityin a physicalinjurycase preciselybe-
causethe causalparadigmsdo. Thoseparadigms,moreover,dependsolely on
the natureof the acts in question,not the numberof peoplewho perform

125 [1901] A.C. 495, 532. See also, for Lord Halsbury'ssimilarviews, [1901] A.C. 495,
506-07.
126Id. at 534. With this reinterpretationof its facts, there is no wonder that Lord
Halsbury (who dissentedin Allen v. Flood) took a very narrow view of the scope of
precedent.[19011 A.C. 495, 506. The very point raised by Lord Halsburyhad been dis-
cussedand dismissedin Allen v. Flood. [18981 A.C. 1, 117. Indeed, the distinct nature of
Allen'sthreatswas the basis for liabilityin the Courtof Appeal.Flood v. Jackson, [18951
K.B. 21, 37 (per Lord Esher, M.R.).
127 Even this point of distinction is odd. Allen v. Flood looks like a clear case of
combination,as Allen had the full supportof the ironworkers.
128[1892] A.C. 25, 38.

This content downloaded from 129.127.145.240 on May 08, 2018 20:13:09 PM

All use subject to University of Chicago Press Terms and Conditions (http://www.journals.uchicago.edu/t-and-c).
 INTENTIONAL
HARMS 441

them.Persuasion(whichincludesthe threatnot to do thosethingswhichyou

are not obligatedto do) is not transmutedinto coercionbecausemanyuse it
to achievetheir commonend, even if it becomesmore effectivethereby.Is
speechto be bannedbecauseit persuades?Similarly,inducementof persons
not to deal with plaintiffsdoes not becomeinducementof breachof contract
when done by many insteadof one. Statementsof fact do not becomemis-
representationswhenassertedby manyinsteadof one. In all casesthe causal
allegationsform an indispensablepart of the primafacie case.

CONCLUSION
This articleis the last of three in whichI have soughtto give a coherent
accountof the tort law that allowsus to approachall cases, fromthe simple
trespassto traderegulation,froma singlepointof view.The majorassumption
of thesearticlesis that, as a substantivematter,the tortlaw shouldbe seenas
a systemof correctivejustice that looks to the conduct,broadlydefined,of
the partiesto the case with a view towardthe protectionof individualliberty
andprivateproperty.The firstdemandof that inquiryis to identifythe class
of interests,be they of liberty or property,that are entitledto public pro-
tection.The secondproblemis to give an accountof the permissible"moves"
that allowprivatepartiesto switchfromone set of entitlementsto another.129
On that point, my implicitpresupposition is that the permissiblemoves are
the consensualarrangements betweenthe partieswhose entitlementswill be
changedby the switch,as for example,by contract,by gift, by inheritance
and the like.130The role of the law of tort in this schemeis to police, as it
were, this system of moves by rectifyingchangesin entitlementsbrought
aboutby impermissiblemeans.In orderto developthe appropriaterules of
rectification,it is, I believe,best to articulatethe rules of torts in termsof
the pleasthat eachpartyinvolvedmighturgeon his own behalf.The firstof
the particulartasks of the tort law is developinga workableaccountof the
conceptof "causingharm"which identifiesthe means of invasionof the
individualinterestsalready defined.The second portionis to set out the
limitationson causationthat operateboth by way of defenseand subsequent
plea, includingthose which relate to the intentionto harm.The theory of
justicewhichis necessarilyinvokedin the case is necessarilyhistoricalin that
it looks to what given individualshave done to upset the initial equilibrium
betweenthemselvesand othersin orderto determineif redressis in order.As
129See, for use of this term, Robert Nozick, DistributiveJustice, 3 Phil. & Pub. Affairs
45 (1973), reprintedin his Anarchy, State, and Utopia 149 (1974), whose entire his-
torical theory of justice is consistentwith the set of rules developedhere.
130The law of contracts,gifts, and wills has as one of its major functionsthe identifica-
tion of those consensualarrangementsthat should not be treated as moves of the game,
for reasonsof fraud,duress,incapacityand the like.

This content downloaded from 129.127.145.240 on May 08, 2018 20:13:09 PM

All use subject to University of Chicago Press Terms and Conditions (http://www.journals.uchicago.edu/t-and-c).
442 THEJOURNAL
OF LEGAL
STUDIES

such, it does not turn upon any preconceived notion of what should be the
appropriate end-state distribution of goods and services, and indeed it will
accept any such end-state so long as there is no defect in the process in which
it was reached.
I have given my view of the legal results that are required once these
premises are accepted, and it is quite clear that they conflict, often in result
and always in manner, with those results that are required by utilitarian
principles. A defense of the initial premise of a system of corrective justice
against its utilitarian alternative is, of course, difficult to make because there
is no commonmeasureof ethical discoursethat allows the two to be compared
with each other. I do not know how to examine the basic conflicts between
the two systems within the compass of a single page or paragraph and will
close with but a single observation. The view that makes torts the study of
resource allocation has not as yet developed any straightforwardset of rules
consistent with its own initial premises to govern even routine cases of harm.
Instead it gives us a list of considerationsto be taken into account before the
definitive answer can be reached in a given case or class of cases.131Every
legal system must take them into account at some point in its development,
but it is, I believe, a grave mistake to suppose that tort questions can be
argued only in utilitarian terms. There is an irreducible ethical base to the
tort law, one that compels us to recognize that hard choices and unpleasant
results cannot be avoided by arguing that further empirical researchis needed
to resolve the traditional conflicts within the law of tort in a principled
matter.132 Most of the positions which I have defended rest upon belief in the
importanceof individual autonomy within the social order. Those conclusions
are, of course, subject to attack, but only within the frameworkof an alterna-
tive theory of corrective justice, and not by an economic theory of social
control.

131See Peter A. Diamond, Single Activity Accidents,3 J. Leg. Studies 107 (1974), to
get some sense of the incrediblycomplexmathematicalapparatusneeded to formalizea
model of the accident law designed to minimize accident costs, even under the most
restrictiveset of assumptions.
132 "The questionwhethera generalsubstitutionof strict for negligenceliability would
improve efficiencyseems at this stage hopelessly conjectural;the question is at bottom
empiricaland the empiricalwork has not been done."RichardA. Posner,Strict Liability:
A Comment,2 J. Leg. Studies205, 211-12 (1973).

This content downloaded from 129.127.145.240 on May 08, 2018 20:13:09 PM

All use subject to University of Chicago Press Terms and Conditions (http://www.journals.uchicago.edu/t-and-c).