Conditional Fault of The Law of Tort

Conditional Fault in the Law of Torts
Author(s): Robert E. Keeton

Source: Harvard Law Review, Vol. 72, No. 3 (Jan., 1959), pp. 401-444
Published by: The Harvard Law Review Association
Stable URL: http://www.jstor.org/stable/1337892 .
Accessed: 05/02/2011 07:33
Your use of the JSTOR archive indicates your acceptance of JSTOR's Terms and Conditions of Use, available at .
http://www.jstor.org/page/info/about/policies/terms.jsp. JSTOR's Terms and Conditions of Use provides, in part, that unless
you have obtained prior permission, you may not download an entire issue of a journal or multiple copies of articles, and you
may use content in the JSTOR archive only for your personal, non-commercial use.
Please contact the publisher regarding any further use of this work. Publisher contact information may be obtained at .
http://www.jstor.org/action/showPublisher?publisherCode=harvardlaw. .
Each copy of any part of a JSTOR transmission must contain the same copyright notice that appears on the screen or printed
page of such transmission.
JSTOR is a not-for-profit service that helps scholars, researchers, and students discover, use, and build upon a wide range of
content in a trusted digital archive. We use information technology and tools to increase productivity and facilitate new forms
of scholarship. For more information about JSTOR, please contact support@jstor.org.
The Harvard Law Review Association is collaborating with JSTOR to digitize, preserve and extend access to
Harvard Law Review.
http://www.jstor.org
VOLUME 72 JANUARY 1959 NUMBER 3
HARVARD LAW REVIEW I
CONDITIONALFAULT IN THE LAW OF TORTS

RobertE. Keeton*
Decisionsimposingstrict liabilityfor harmscausedby nonnegligent,
risky conducthave been viewedalmostuniformlyas instancesof lia-
bility eithout moralfault; risk-spreadingcapacityhas been said to
be an emerginginfluence,threateningto supplantfault. This Ar-
ticle urges, instead, that these decisions are founded on a moral
concept- for convenience called "conditionalfault"- and that
recognitionof its influencecan aid in understandingthe law of torts.
I. FAULT AS A BASIS FOR Loss SHIFTING
T HE usual startingpoint for explainingbases of liability in

tort is the assertion that the machinery of adjudication will
not be set in motion without good reason. Its operation is costly.
From the community point of view, its use to shift a loss is not
worth the price unless the loss shifting serves some good purpose.
This axiom of adjudicatory inertia is rarely challenged.
It is occasionally said that the primary function of tort law
is the compensation of losses. This statement is at best a half-
truth. Even apart from other reasons opposing liability, the
axiom of inertia reminds us that sometimes compensation should
not be allowed; courts should leave a loss where they find it unless
good reason for shifting it appears. Moreover, the mere fact that
one member of society is compensated when the court shifts the
loss is not such a reason, since the gain is offset by the loss shifted
to another member of society. In modern Anglo-American tort
*Professor of Law, Harvard Law School. B.B.A., University of Texas, 1940,

LL.B., I94I; S.J.D., Harvard, I956.
I gratefully acknowledge my indebtedness to Professors Warren A. Seavey and
Lon L. Fuller for their helpful criticisms of a draft of this article. Of course, they are
not responsible for any heresies expressed.
40I
402 HARVARDLAW REVIEW [Vol. 72
law, fault has been considered the one generally acceptable

reason for such loss shifting. For more than a century, at least,
fault has been the principal theme of tort law.
This theme may have even more ancient lineage.' The evidence
is inconclusive,2but there is reason for believing that early com-
mon-law conceptions of responsibility for harms one had caused
were rustic definitions of fault. The notions of causation and
fault are close kin. Picking one or more responsible causes from
the multitude of antecedents of a given incident is very close to
finding fault.' It may be that, to early common-law lawyers and
even until the nineteenth-century flowering of the concept of
1An interesting range of opinions is presented in (i) Wigmore, Responsibility

for Tortious Acts: Its History (pts. I-3), 7 HARV. L. REV. 3I5, 383, 44I (I894)
(suggesting that law began with an amoral concept that a man acts at his peril and
gradually developed toward the theme of liability based on fault); (2) HOLMES,
THE COMMON LAW lectures I, 3, 4 (I88I) (suggesting that in early law liability
was associated with revenge, and was based on the thought that someone or some-
thing was to blame, and that though in a sense law always measures liability by
moral standards, it is continually transmuting these moral standards into external
or objective ones, from which the actual guilt of the party concerned is wholly
eliminated); and (3) Isaacs, Fault and Liability, 3I HARV. L. REV. 954 (I9I8)
(suggesting that culpability has always been an important factor in determining
rules of liability, and that the degree of emphasis on culpability has alternately
waxed and waned).
2 From the decisions, these conclusions seem warranted: (i) the gist of liability
in trespass was that defendant's act directly produced the injury; (2) probably
the complete absence of any negligence would defeat the action; (3) but if it
would, this was matter of defense to be affirmatively pleaded and proved by
defendant; it came in, so to speak, by the back door; (4) the concept of
negligence was not dominant enough - at least in this form of action - and
not well enough worked out in the mind of the profession so that its opposite
was ever successfully formulated as a defense to an action of trespass before
the nineteenth century.
2 HARPER& JAMES, TORTS 748-49 (I956). (Footnotes omitted.)
Though liability was in some areas stricter than it is today it has never been
true that "a man acts at his peril," if by that phrase it is meant that a man is
liable for whatever harm he does to another. Even the so-called absolute liability
in trespass was subject to grounds of justification and excuse. See Winfield, The
Myth of Absolute Liability, 42 L.Q. REV. 37 (I926).
3 Cf. 3 HOLDSWORTH,A HISTORY OF ENGLISH LAW 378-80 (5th ed. I942); 8 id.
at 446-47 (2d ed. I937). Holdsworth takes the position that during the medieval
period (io66-I485) civil liability was not concerned with fault but rather rested
on a primitive basis that one is liable for harms caused by his acts; this principle,
he contends, continued to occupy a major role in civil liability up into the nine-
teenth century. He observes, however, that even in the medieval period, one's
liability was limited to the proximate consequences of his acts. "[Tlhis limitation
of liability was perhaps accepted as a measure of obvious justice without a precise
analysis of its consequences and bearing upon the prevalent theory of civil liability.
The conception of negligence is latent in such a limitation; but in this period this
latent consequence has not been discovered." 3 id. at 379-80.
1959] CONDITIONALFAULT 403
negligence was at hand, there was no such concept as an unavoid-

able accident in the modern sense of one for which neither the
deity nor any person is to be blamed.4
In early common law, conduct was judged more by results than
by intentions.5 But that is not to say that it was judged by a
standard which disregardedfault in favor of causation. Probably
it is nearer to the mark to say that conduct causing harm was
considered morally blameworthy. Close adherence to moral
blameworthiness may have been sacrificed to gain supposed ad-
vantages of objective standards of judgment. It is particularly
likely that such a sacrifice will seem sensible whenever we are
either uneasy about our concepts of moral responsibility6 or dis-
'One might so interpret the report of Weaver v. Ward, Hob. I34, 8o Eng.
Rep. 284 (K.B. c.i6i6). Judgment was given the plaintiff upon his demurrer to
the defendant's plea that the plaintiff and the defendant were soldiers engaged in
training when, by accident, his musket was discharged and plaintiff was wounded.
The report states, "and, therefore, no man shall be excused of a trespass . . .
except it may be judged utterly without his fault. As if . . . the defendant had
. . . set forth the case with the circumstances so as it had appeared to the Court
that it had been inevitable, and that the defendant had committed no negligence
to give occasion to the hurt." Ibid. It seems a permissible inference from this
ambiguous passage that its author thought that an unintended hurt was either
negligently caused by defendant or not caused by defendant; in the latter event,
it was "inevitable" and occurred utterly without his fault.
It has been said that this was the first of a line of cases in which "inevitable
accident" came to be recognized as a ground for escaping liability. Bohlen, Liability
in Tort of Infants and Insane Persons, 23 MICH. L. REv. 9, I4 (I924). But the lack
of any intimation in the report that this idea was an innovation in i6i6 suggests
that it was commonly assumed before being stated in Weaver v. Ward. The state-
ment, "but he ought to say that he could not do it any other way, or that he did
everything that was in him to keep them out, etc., or otherwise he shall pay
damages," appears as early as The Case of the Thorns, Y.B. 6 Ed. IV, f. 7, pl.
i8 (I466).
Arguing that there has never been a time since the early I500's when a de-
fendant in trespass was not allowed to appeal to some standard of blame or fault,
Wigmore observes: "[E]ven up to the i8oos we find court and counsel constantly
interchanging 'inevitable accident' and 'absence of negligence or blame.'" Wigmore,
Responsibility for Tortious Acts: Its History (pt. 3), 7 HARv. L. REV. 44I, 444
(I894).
' See, e.g., HOLDSWORTH, A HISTORY oF ENGLISH LAW 375 (5th ed. 1942); Ames,
Law and Morals, 22 HARv.L. REV.97 (I908).
6 Cf. Jaffe, Damages for Personal Injury: The Impact of Insurance, i8 LAW &
CONTEMP. PROB. 2I9, 220 (1953):
Whether fault is an effect of free will is a question that each decides for himself,
but the law, following the common instinct, treats fault as a moral dictum.
. . . There is, however, no denying that for most of us the question of the
individual's moral responsibility for fault arouses deep and discomforting
metaphysical distress. The modern man is thus driven to avoid where possible
solutions that invoke the concept of fault.
trustful of the capacity of fact-finders to make the finer dis-

criminations required in the application of a subjective standard
-one dependent upon the personal characteristics and the state
of mind of the defendant.
Tort law provides many illustrations of the use of relatively
mechanical rules even though a more personalized standard of
judgment would commend itself were it considered administrable.
The objectively oriented standard of negligence is an important
example. Also, the difficultiesof making a personalized judgment
of the plaintiff's harm and its causes in part explain the past re-
luctance of courts to extend liability in cases of emotional injury.
One factor in the current trend of expanding liability for emo-
tional injury is increasing confidence in fact-finding as relatively
more reliable expert evidence becomes available.
In any event, if fault was not from the first the main theme of
the common law of what we now know as torts, it came to be such
before the present day. Moreover, many current trends toward
broader liability represent no deviation from the theme of fault,
but rather a movement toward closer adherence to it. Trends to-
ward broader liability in cases of emotional injury,7 failure to
benefit one in peril,8 and occupiers' liability,9 are examples, as is
the trend toward the narrowing or elimination of immunities
enjoyed by charities, governmental units, and persons in family
relations.'0 Another example is the trend toward comparative
negligence," which, even after allowance for jury aberrations,
is a closer approach to distribution of loss in proportion to fault
than the older decisional rule that contributory negligence is a
complete bar. The older rules in all these instances are illustra-
7I HARPER & JAMES, TORTS 665-9I (1956); 2 id. at I03I-39; PROSSER, TORTS
38-47, I76-82 (2d ed. I955).
8 2 HARPER & JAMES, TORTS I044-53 (I956); PROSSER, TORTS i82-90 (2d ed.
I955).
92 HARPER & JAMES, TORTS I430-533 (1956); PROSSER, TORTS 427-77 (2d ed.
I955). In England, the Occupiers' Liability Act, I957, 5 & 6 Eliz. 2, ch. 3I, does
away with the rules under which the liability of the occupier differs according to
whether the visitor is an invitee or a licensee; ? 2 provides instead that the occupier
owes "the common duty of care" to all his lawful visitors, "except in so far as
he is free to and does extend, restrict, modify or exclude his duty to any visitor
or visitors by agreement or otherwise."
10 I HARPER & JAMES, TORTS 643-52 (1956); 2 id. at I6I3-32, 167I-75; PROSSER,
TORTS 770-93 (2d ed. I955).
11 2 HARPER & JAMES, TORTS I238-4I (I956); PROSSER, TORTS 298-99 (2d ed.
I955).
1959] CONDITIONAL FAULT 405
tions of the failure of tort law to carry through its theme of fault;
they are deviations toward a narrower scope of liability than the
theme of fault would indicate. The trend toward broader liability
in the cases thus far noted may be viewed as a movement toward
a finer tailoring of tort law to fault, rather than a challenge to
fault as the basis of liability.
II. MODERN ADVOCACY OF RISK-SPREADING CAPACITY

AS A BASIS FOR TORT LIABILITY
In recent years, it has been urged with increasing vigor that a

loss should be shifted from plaintiff to defendant if defendant is
a more efficient loss distributor. The gist of the argument is that
in this way a loss will be spread more generally in the community
among those who benefit from the activity out of which the loss
arises.
A. Various Forms of the Argument

The argument may be presented in a variety of ways. First,
if one thinks of this process as occurringafter the loss, he envisions
redistributionof loss; the loss should be placed on the one with the
greater capacity for spreading it. Second, if one thinks of the
process as occurringbefore the loss, he envisions the accumulation
of a fund; the better distributor of loss is the one with the greater
capacity for accumulating from widely scattered sources a fund
for bearing such losses whenever they occur.
From the anticipatory point of view of accumulating a fund in
advance of a loss, the matter appears not simply as one of loss
distribution but rather as one of risk management. The fund is
being accumulated by setting aside from the proceeds of each par-
ticular activity a sum estimated to be sufficient for its share of the
losses resulting from that type of activity. The contribution due
from each activity is dependent on the amount of risk. For this
reason, there is a third form in which the argument is sometimes
expressed-it is said that losses should be shifted to the better
risk-bearer. The self-evident similarity to insurance explains the
fact that, at other times, the argument is expressed in a fourth
form, as one for the application of the insurance principle in tort
law. These different forms of expression are likely to produce
different overtones, but they are in fact variations on a single

theme. For convenience, we may refer to it as the notion of risk-
spreading capacity in contrast with the notion of fault.
In whichever one of these forms the argument is expressed
loss distribution, fund accumulation, risk management, or ap-
plication of the insurance principle-it may be based on the
capacity of the class of persons such as defendant in comparison
with the class of persons such as plaintiff. On the other hand it
may be based on the capacity of the individual plaintiff and
defendant in the light of whatever insurance or other provision
each has arranged in advance of the accident. To the extent that
the latter form of the argument is relied upon to support liability
deriving not from agreementbut regardless of manifested intent,12
there are special objections to it. Imposing liability because the
defendant has a liability-insurance policy, for example, operates
partly as a deterrent to effective risk spreading; it adds liabilities
not sufferedby those less concernedwith forehandedmanagement
of risk. These added liabilities mean added costs in higher insur-
ance premiums. They make it impossible to choose relatively
low-cost insurance against the same limited risks for which unin-
sured persons are responsible; they present the hard choice be-
tween higher-cost insurance and no insurance. They thus tend
to discouragesome risk-spreadingarrangementsas well as to offend
12 Some policies issued to legal entities with immunity from certain tort liability
specify by endorsement that the insurance company shall not assert the immunity;
in those circumstances liability up to the amount of the insurance coverage is
derived from agreement. But even in the absence of such an endorsement, some
decisions hold that immunity is not available to defeat recovery if the defendant
has a liability-insurance policy in an amount at least equal to the claim. E.g.,
O'Connor v. Boulder Colo. Sanitarium Ass'n, I05 CoIo. 259, 96 P.2d 835 (I939);
McLeod v. St. Thomas Hosp., I70 Tenn. 423, 95 S.W.2d 9I7 (I936). Perhaps
these decisions can be rationalized as interpretations of allegedly ambiguous language
in the insurance policies to the effect that the company agrees to pay on behalf of
the insured all sums which the insured shall be legally obligated to pay as damages
because of injury from defined hazards. But the failure of the opinions to take
notice of the terms of the policies suggests that the opinions may support the
proposition that immunity is waived by taking out the insurance policy, even
if its terms state explicitly that it is to cover only the liabilities that would have
existed in the absence of insurance. See also Michael v. St. Paul Mercury Indem.
Co., 92 F. Supp. I40 (W.D. Ark. I950), applying a statute which provides that an
injured person shall have a cause of action against the liability insurer of "any
cooperative non-profit corporation, association or organization or . . . any other
organization or association of any kind or character not subject to suit for tort,"
regardless of the terms of the policy itself.
g959] CONDITIONAL FAULT 407
one's sense of fairness by imposing heavier burdens on those

whose conduct is more praiseworthy in providing assurance of
financial responsibility."3For purposes of the present inquiry, set
aside this more doubtful form of the argument which is based
on the risk-spreading capacity of the individual defendant and
plaintiff. Consider, instead, that form of the argument which
deals with the persons in classes, without regard to whether a
policy of insurance has been obtained by either person.
B. Does This Theory Explain Recent Developments?

It is plain that risk-spreading capacity has had significant
influence on tort law, and will have more in the future. Instances
of strict liability are generally cited as proof of this influence.
With diffidencein the face of the array of contrary opinions,14 I
13 Cf. Note, 33 MINN. L. REv. 634, 64o-46, 652-54 ('949). But cf. id. at 657-58,
arguing that, in the absence of an express exclusion of liability for injuries to
members of the insured's family or household, liability should be imposed and
coverage should be afforded, though the circumstances are within the scope of a
previously-recognized immunity. See also Ehrenzweig, Assurance Oblige -A Com-
parative Study, i5 LAW & CONTEMP. PROB. 445, 450-5I (I950), expressing pref-
erence for a rule based on the availability of insurance rather than the purchase
of insurance (that "insurability obligates," not that "insurance obligates"). Never-
theless Professor Ehrenzweig, a vigorous advocate of risk distribution through
enterprise liability, appears to approve decisions in which the courts relied upon the
fact that the particular defendant had liability insurance. Id. at 452; EHRENZWEIG,
NEGLIGENCE WITHOUT FAULT 26-27, 78 (95I).
14 Witness the comments of eminent authorities on torts. For example, Dean
Prosser, in discussing what he prefers to call strict liability, states:
The courts have tended to lay stress upon the fact that the defendant is acting
for his own purposes, and is seeking a benefit or a profit of his own from such
activities, and that he is in a better position to administer the unusual risk
by passing it on to the public than is the innocent victim. The problem is dealt
with as one of allocating a more or less inevitable loss to be charged against
a complex and dangerous civilization, and liability is placed upon the party
best able to shoulder it. The defendant is held liable merely because, as a matter
of social engineering, the conclusion is that the responsibility should be his.
This modern attitude, which is largely a thing of the last four decades, is of
course a far cry from the individualistic viewpoint of the common law courts.
PROSSER,TORTS 3I8 (2d ed. I955).
Professors Harper and James, in the introduction to their treatise, describe
this development as follows:
In the late nineteenth and early twentieth centuries powerful pressures were
forming to ameliorate the burdens of the victims of industrialism and to
distribute the advantages of those who had exploited its potentialities. In the
law of torts, this movement slowly took the form of a search for the best risk
bearer and the most efficient loss administrator.
i HARPER & JAMES, TORTS xlii (1956).
Though not attributing such importance to the influence of risk-spreading
submit that the areas of strict liability which have developed do

not bear out the conclusion that risk-spreading capacity has had
or will have the kind and degree of influence on tort law which
is implied in characterizingthese recent developmentsas "a search
. . . for the most efficient loss administrator."15 Nor do they
bear out the conclusion that these trends are merely social engi-
neering, with a basis in social morality but not personal blame."6
There are other circumstanceswhich help to explain these develop-
ments. They are suggestive of an element of personal blame and
of limits within which the idea of risk-spreading capacity will be
influential in the development of tort law. The most important
decisional development cited in support of conclusions concerning
the influence of risk-spreading capacity is that which has been
variously called strict liability, absolute liability, and liability
without fault for harm resulting from nonnegligent risky conduct.
In the English and Dominion cases this development is identified
with Rylands va.Fletcher 17 and its progeny. In the United States,
some courts have accepted the doctrine of that case, some have
accepted the doctrine of liability for uncommon ultrahazardous
activities, as expressed in the Restatement of Torts,18 and some
have rejected both.'9 We do not pause now to consider whether
courts nominally rejecting such doctrines have allowed the new-
comer into the parlor by a backdoor such as nuisance.20 It is
capacity in these developments, Professor Seavey joins in viewing them as con-

clusions of social justice on grounds other than individual blameworthiness. Re-
ferring to various examples, including liability for harm caused by nonnegligent
blasting and liability for speaking laudatory words which turn out to be defamatory,
he says:
To say that these acts are wrongful is in substance to misdescribe them. They
are tortious, but only because in the conflict of interests the courts have come
to the conclusion that justice is best served by placing the loss upon the actor
or speaker.
SEAVEY,COGITATIONS ON TORTS 6 (1954). See also Seavey, Speculations as to
""Respondeat Superior," in HARVARD LEGALESSAYS 433, 439-45 (I934).
15 The phrase is quoted from i HARPER & JAMES,TORTSXlii (1956).
16 For expressions of this conclusion, see the quotations from Dean Prosser and
Professor Seavey in note I4 supra.
17 [I868] 3 H.L. 330.
18?? 5I9-24 (I938).
19 2 HARPER & JAMES, TORTS 794-802 (I956); PROSSER, TORTS 33I-38 (2d ed.
I955).
20 See 2 HARPER & JAMES, TORTS 8I9 (1956); PROSSER, TORTS 336, 393 (2d ed.
I955).
I9591 CONDITIONAL FAULT 409
enough for our purposes that a substantial body of precedent for

such liability has arisen, and that some basis other than the tradi-
tional rationale of fault must be found to explain it. The terms
strict liability, absolute liability, and liability without fault do
not suggest a new rationale.
Is the risk-spreadingcapacity of the classes of defendants held
liable in these cases a satisfactory explanation of the results? If
so, one would expect that such liability would be extended to
other activities in which participants have similar risk-spreading
capacity. But this expectation is not borne out by a comparison,
for example, of blasting, aviation, and railroading. The risk-
spreading capacity of persons engaged in railroading appears to
be as great and as evident as that of persons engaged in aviation.
In both instances, there are available the mechanism of fares to
pass along costs to large groups of users and the mechanism of
liability insurance to protect against catastrophic losses from
individual accidents. Both those engaged in railroading and
those engaged in aviation appear to have an edge over blasters
as to risk-spreadingcapacity. The blaster's costs must be passed
along somewhat more deviously than by fares, and arguably less
effectively in the sense of spreading the cost in proportion to the
the benefits ultimately realized by a large group of persons. Also,
it is somewhat more troublesome to obtain satisfactory insurance
to protect against catastrophic losses from individual blasting
accidents. Yet, among these three activities the area of greatest
development of this broader liability is blasting, and the broader
liability is much more evident in aviation than in railroading.
A possible explanation of the contrast among these three areas
is that it arises from a lag in the acceptance of a new notion as a
basis for tort liability. Before yielding to the temptation to take
this easy way out, perhaps we could usefully consider still another
concept of liability without negligence or intendedharm - namely,
the concept of unjust enrichment. Though usually assigned to
the rubric of quasi-contract, liability on the theory of unjust en-
richment is not derived from agreement. Why this apparent
deviation from the principal theme of tort law that liability
beyond that attached to manifested agreement is not imposed
in the absence of fault?
4IO HARVARDLAW REVIEW LVol.72
III. UNJUSTENRICHMENTANDVINCENTV.
LAKEERIETRANSP.CO.
It is often suggested that Vincent v. Lake Erie Transp. Co.2"is
explainable as a case of unjust enrichment- that is, the ship-
ownerwho kept his ship at the dock through the storm rather than
risk harm to it in the open water was enriching himself at the
expense of the dockowner whose dock was damaged as the ship
beat against it.22 But explaining Vincent by a generalization that
enrichment through another's loss is a ground for liability may
open a door to considerable extensions of "unjust enrichment."
Professor Dawson's comment, though made without particular
reference to disputes about ships and docks, is peculiarly apt in
this context:
"[A] generalprincipleprohibitingenrichmentthroughanother's
loss appearsfirst as a convenientexplanationof specificresults; it
is an instrumentfor quite practicaland intelligiblepurposes.Yet
once the idea has been formulatedas a generalization,it has the
peculiarfaculty of inducingquite sober citizensto jump right off
the dock."23
21 IO9 Minn. 456, I24 NW. 22I (I9IO). Defendant's ship was moored to
plaintiff's dock for unloading; a storm arose and had grown violent by the time
unloading was completed. The lines between the ship and the dock were kept fast,
and as soon as one parted or chafed it was replaced. The wind and waves constant-
ly pitched the ship against the dock, causing damage to the dock fixed at $500.
Defendant was held liable.
22 Among the authorities supporting this explanation is RESTATEMENT, RESTI-
TUTION ? I22 (I937):
A person who is privileged to harm the land or chattels of another while acting
to preserve himself or a third person or to preserve his own things or those of a
third person is under a duty of restitution for the amount of harm done,
except where
(a) the harm which he seeks to avert is threatened by the things which he
destroys or by the tortious conduct or contributory fault of the owner or
possessor, or
(b) his act reasonably appears to be necessary to avert a public catastrophe,
or
(c) he is exercising his privilege as a member of the public to enter land
adjacent to a highway which has become impassable.
See also Bohlen, Incomplete Privilege To Inflict Intentional Invasions of Interests
of Property and Personality, 39 HARV.L. REV.307 (I926).
23 DAWSON,UNJUST ENRICHMENT 8 (I95I). The passage continues:
This temporary intoxication is seldom produced by other general ideas, such as
"equity," "good faith," or "justice," for these ideals themselves suggest their
own relativity and the complexity of the factors that must enter into judgment.
The ideal of preventing enrichment through another's loss has a strong appeal
to the sense of equal justice but it also has the delusive appearance of mathe-
matical simplicity. It suggests not merely the need for a remedy but a measure
of recovery. It constantly tends to become a "rule," to dictate solutions, to
impose itself on the mind.
1959] CONDITIONALFAULT 4II
A. Gain and Loss

The use of "unjust enrichment" to explain liability in cases
like Vincent presses us to recognize that the term "enrichment"
includes not only gain and avoidance of loss, but also avoidance
of risk of loss.24 If it is proved at trial that the ship would have
been lost had it left the dock, the case presented is avoidance of
loss. On the other hand, if it is proved that there was only a
chance of its being lost (let us say, a one-in-ten chance) and that
probably the ship would have sufferedneither damage nor destruc-
tion, but that a prudent shipowner would not have taken that
chance, the case presented is avoidance of risk of loss. If we do not
consider the latter a kind of gain, the case for restitution disap-
pears. It is not a satisfactory answer to this argument that the
shipowner received the use of the dock. If no special weight is
given to the purpose of that use, the measure of the gain should
be rental value.25 If, on the other hand, the purpose of the use is
considered it appears that the value to the shipowner is in the
avoidance of risk of loss. Thus we are pressed to recognize
avoidance of risk of loss as a kind of gain, or else to deny resti-
tution in such cases while granting it in cases involving avoidance
of loss. But realization of advantage or avoidance of disad-
vantage to the shipowner at the expense of the dockowner is
as clear in one case as in the other. Arguably even if the ship
is lost despite the reduction of risk by remaining at the dock, the
shipownerhas avoided a disadvantage at the expense of the dock-
owner. But if the ship is saved, the avoidance of disadvantage at
the expense of the dockownerseems beyond dispute.
The similarity between avoiding loss and avoiding risk of loss
is made more apparent by considering application of these con-
cepts at trial. If avoidance of loss were the critical fact in deter-
Is it possible that the concept of risk-spreading capacity has similar impact on

the mind of man?
24 Compare RESTATEMENT, RESTITUTION ? i, comment b (I937). Though not
explicitly recognizing avoidance of risk of loss as enrichment, it plainly points to
that conclusion by stating that one confers a benefit "where he saves the other from
expense or loss," including that conferred when the physician attends an insensible
person "who is saved subsequent pain or who receives thereby a greater chance
of living." A greater chance of living is surely equivalent to a reduced risk of early
death.
25 This would not be equal to the damage to the dock, in the absence of a
question-begging construction of a fictional rental price in the form of an agree-

ment to pay for damage to the dock.
4I2 HARVARD LAW REVIEW [Vol. 72
mination of legal liability, it would be proved at trial with less

than certainty. Under the customary standard of proof, the fact-
finder would be asked whether, from a preponderanceof the evi-
dence, it is more probable than not that the loss would have oc-
curred had the ship not remained at the dock. Even under a
more stringent standard of proof, it is clear that something less
than certainty would be required; the standard of proof of guilt
in criminal cases, the most stringent known to the law, falls short
of certainty. The difference,therefore, between avoidance of loss
and avoidance of risk of loss is at most one of degree of proba-
bility. Moreover, this difference may be less significant than a
difference as to the probable severity of loss if it should occur.
For example, compare the risk of loss of an entire ship with the
certainty of small damage to the ship.
An additional reason for our being pressed toward recognizing
avoidance of risk of loss as enrichment is that it is probable that
Vincent itself was a case of avoidance of a risk of loss that prob-
ably would not have occurred, but might have been so severe that
a prudent shipownerwould not have taken the chance. One would
expect that, if the members of the court had considered it relevant
to distinguish between avoiding loss and avoiding risk of loss,
their opinions would have contained some discussion as to which
type of case was presented by the facts before them.
Unjust enrichment by gain from conduct which also produces
loss to another is only one of the possible forms of enrichment to
the detriment of another in the broad sense of change in the al-
location of advantages and disadvantages. Other forms of such
enrichmentinvolve modificationof risk - such things as avoiding
or reducing existing risks to oneself while creating or increasing
risks to others. Also, some form of risk modification may appear
in combination with loss to another, gain to oneself, or avoidance
of loss to oneself. Making a less safe product in order to econo-
mize on cost is such a combination involving enrichment to the
detriment of another in this broad sense.
B. What CircumstancesAre To Be Comparedin Determining

Whether There Has Been Unjust Enrichment?
In addition to implying a broad concept of "unjust enrichment,"
the determination of the liability of the shipowner in Vincent
at least implicitly involved other debatable steps of reasoning.
I959] CONDITIONAL FAULT 4I3
"Enrichment" implies change. It implies a comparison between

different sets of circumstancesexisting at differentpoints of time.
Suppose we assume as our starting point a time when the load-
ing of the vessel had been completed, the ship was ready to depart
but for the impending storm, and the weather was already bad
but was not yet causing damage to the dock or the ship. If we
assume that at that time the shipowner had a legal duty to cast
off from the dock,26and could be expected in fact to do so, his storm
risk was great and the dockowner's risk small. Comparisonwith
the later situation of the ship remaining and battering the dock,
or the still later situation of the accomplished damage after the
risk to the dock had materialized in harm, indicates that the ship-
owner's failure to cast off resulted in his enrichment; incidentally,
the fact that the shipowner tied the ship more securely is su-
perfluousunder this assumption. If, on the other hand, we assume
that at the earlier time selected for comparison the shipowner
was entitled to keep his ship at the dock through the period of
an approaching storm, the shipowner's risk was smaller and the
dockowner's risk larger at this starting point. If this latter as-
sumption is made, it is arguable that there was enrichment in a
factual sense by the shipowner's assertion of his legal right to
remain, thus causing the valuation appraisal of the risks to the
ship to rise less rapidly, as the storm approached,than would have
been the case had he cast off. But clearly there would be no un-
just enrichment from merely failing to cast off; rather, the change
would be only a realization by shipowner and dockowner of the
advantages and disadvantages inherent in their respective legal-
factual situations. Moreover, there would be no unjust enrich-
ment from tying the ship more securely if we assume that the
privilege of doing so was included in the proposition that the
shipowner had no duty to cast off.27 On the other hand, if we
26 It might be argued that such a duty to cast off should be recognized as one
of the terms of the agreement for the use of the dock. Some differences of opinion
about Vincent are due to this problem of contract interpretation rather than to
disagreement over a rule of law to be applied regardless of contract terms.
27 The majority in Vincent may have been influenced by the defendant's conduct
in tying the ship more securely.
[T]he lines were kept fast, and as soon as one parted or chafed it was replaced,
sometimes with a larger one. . . But here those in charge of the vessel
deliberately and by their direct efforts held her in such a position that the
damage to the dock resulted, and, having thus preserved the ship at the expense
of the dock, it seems to us that her owners are responsible to the dock owners
to the extent of the injury inflicted.
IO9 Minn. at 457, I24 N.W. at 222.
assume that this privilege was not included, the determinationof

enrichment from tying it more securely would depend on factual
estimates as to the probability of harm in the absence of such
action. Thus, one's assumption as to the allocation of rights and
duties, advantages and disadvantages, at the starting point is a
vital step in his determination that there has or has not been an
unjust enrichment of one to the detriment of the other by subse-
quent events. It is easy to fall into the error of thinking we have
demonstrated by logic a proposition which we assumed without
explicit consideration.
C. Justification
Still another trap awaits us. As already noted, it is not merely
enrichment, but rather unjust enrichment, which must be found
as a basis for liability. Our daily lives are filled with activities
which change the allocation of advantages and disadvantages
among us, but do not amount to unjust enrichment. A common
example is the nonnegligent driving of one's automobile. It in-
creases risks to others while gaining the benefits of rapid and
convenient travel for oneself. It might be said that there is no
change in the legal aspect of the allocation of risks in this instance,
because we assume as a starting point that each citizen is subject
to the legal risk of physical harm from nonnegligent driving. But
if we decline to assume this legal conclusion in describing the
circumstances under comparison, it is apparent that there is a
change in the factual aspect of the risk to each one of us in the
vicinity of a nonnegligently operated vehicle. Thus there is a
kind of enrichmentof oneself to the detriment of others when one
operates an automobile nonnegligently; it is not, however, unjust
enrichment. It does not give rise to liability.
Another common form of enrichment which is not unjust en-
richment is the use of one's own land, for example by building
a service station upon it, which interferes with the use and value
of adjoining lands but, given all the circumstances, falls short of
being an unreasonable interference and thus a nuisance. Sub-
stantial changes in land use usually affect the value of other land
in the vicinity. But the law of nuisance does relatively little in
the way of compensatingfor this shifting of values, and little more
has been done outside the law of nuisance.28
28 British efforts to deal with this problem in land-use planning are instructive.
Their experience is suggestive, first, of recognition that enrichment of some land-

Illustrations could be multiplied, but these should serve to

make the point that in reaching the conclusion that the shipowner
was enriched unjustly at the expense of the dockowner one passes
through a debatable step of assertion, explicitly or by assumption,
that the new allocation of advantages and disadvantages is legally
improper (either because of its nature, or because of the method
by which it was brought about, or because of a combination of
the two). It is apparent, on reflection, that most and perhaps
all activities commonly assumed to be permissible lead to conse-
quences which amount to enrichment to the detriment of another
in the broad sense of changes in the allocation of advantages and
disadvantages, especially when we regard reduction of risk as
an advantage and increase of risk as a disadvantage.
D. Is Use or Intended Impact Essential

to Liability?
Arguably one may distinguish between enrichment gained by
using another's property or causing an intended impact upon it
and, on the other hand, enrichment gained by causing a risk of
harm to another'sproperty without using it. The same distinction
owners at the expense of others occurs when changes in land use are effected, and
second, of a growing realization of the complexities encountered when, by statute,
it is sought to declare this enrichment unjust and to provide a legal remedy. Sec-
tions I8-24 of The Town & Country Planning Act, I932, 22 & 23 Geo. 5, ch. 48,
provided that any person injuriously affected by a planning scheme could recover
compensation from the authority responsible for carrying out the scheme, and the
authority could recover 75%0 of a resulting increase in value, called a "better-
ment." In fact, little compensation was paid under these provisions, though plan-
ning decisions were often affected by potential liability. HEAP, AN OUTLiNE OF
PLANNING LAW 8 (2d ed. I955). The Town & Country Planning Act, 1947,
IO & ii Geo. 6, ch. 5I, abandoned the compensation-betterment provisions of the
I932 act and nationalized the development value of land (including the right to
make any material change in its use) by taking from the owner the right of de-
velopment and compensating in part at least for such taking. The Town &
Country Planning Act, I954, 2 & 3 Eliz. 2, ch. 72, retreated from the national
expropriation of development rights from the owner, but did not revive the com-
pensation-betterment scheme of I932. See HEAP, op. Cit. Supra, at 13, I27-28,
I36-3 7.
The I954 Act has been termed a compromise attempt (because there is no
perfect solution) at dealing with the compensation-betterment problem of land
planning. It is a compromise because, on the one hand, it makes no further
attempt at the ad hoc collection of betterment (such attempts having failed
under the I932 Act and also, being in the form of development charges, under
the I947 Act) whilst, on the other hand, it provides for the payment of limited
compensation on the imposition of planning restrictions on development.
Id. at 137. See also HAAR, LAND-USE PLANNING 545-56 (I959); HAAR, LAND PLAN-
NING LAW IN A FREE SOCIETY94-1O6 (I951).
4I6 HARVARD LAW REVIEW [Vol. 72
might be made between using another's person or causing an in-

tended impact upon it and, on the other hand, causing a risk of
harm to his person without using it. Thus the person who, to
protect himself against assailant A's gunshots, takes cover behind
bystander B, is using B's person as a shield and might be held
liable for resulting harm,29whereas the one who shoots back
at A in self-defense is merely causing a risk of harm to bystander
B and, if not negligent, is not liable even though harm within that
risk results.30
Such a distinction is implicit in the terms of incomplete privi-
lege as it has developed in tort law,3"and in a comparisonbetween
the treatment in tort law of emergency cases in the area of negli-
gence and those in the area of intentional torts. Defendants A
and B, driving down the highway one behind the other, are con-
fronted with a sudden emergency which arises without negligence
on the part of either. Defendant A makes a prudent decision to
swerve, realizing that there is a risk of his striking the parked
and unoccupied car of P, but hoping that he can stop before doing
so. The impact of his car upon P's car is neither intended nor
negligent, and he is not held liable. Defendant B makes a prudent
decision to swerve, realizing that he will certainly strike Q's
parked and unoccupied car, but realizing also that the danger to
himself and his car will be much greater if he takes any other
available choice. He has exercised the incomplete privilege to
cause an intended impact on the chattel of another to benefit
himself and is held liable for the harm done to Q's car.32
29 This conclusion is based on speculation; see Laidlaw v. Sage, 158 N.Y. 73,
52 N.E. 679 (i899), which is inconclusive on this point. The argument for liability
is somewhat stronger, and has firmer roots in common-law trespass, when defendant
pulls B in front of him than when he takes cover behind B without touching B.
And, when defendant pulls B in front of him, the argument for liability is stronger
as to harm resulting, for example, from B's fall caused by defendant's contact with
B, than as to harm resulting from assailant A's gunshot.
30E.g., Morris v. Platt, 32 Conn. 75 (I864).
31
RESTATEMENT, RESTITUTION ? I22 (I937) is broadly phrased in terms of a
person "who is privileged to harm the land or chattels of another" while exercising
an incomplete privilege. (Emphasis added.) But in RESTATEMENT, TORTS ?? I97,
2I7 (I934) the incomplete privilege concerns conduct which, but for the privilege,
would amount to the kind of entry on land involved in trespass to land, or the kind
of use or "intentional intermeddling" involved in trespass to chattels. Section 2I7,
concerning ways of committing nondispossessory trespass, does not include cases in
which the actor does not intend to cause a contact with the chattel, though he knows
there is a risk of such contact.
32 Perhaps liability would not be imposed if B could show that he was
acting
1959] CONDITIONAL FAULT 4I7
Of course, it is a commonplace observation that two fact situa-

tions involving small differences fall on opposite sides of a line
between liability and nonliability. The criticism that two cases
so nearly alike should not be treated so differently could be made
as to two cases barely on one side and the other of almost any
other line between liability and nonliability.33 Nevertheless, the
similarity between these cases suggests at least that the legal rule
which treats them differently, though well entrenched, may not be
impregnable.
Vincent can be sustained without attack upon this distinction,
since the shipowner was using the dock. Indeed, the distinction
may serve to reconcile cases which have reached the result of
nonliability in circumstances rather similar to those of Vincent.34
Nevertheless, the use of the rationale of unjust enrichment to
explain Vincent suggests that gaining advantage at another's
disadvantage is the basis for liability imposed on the user. The
fact that advantage has been gained at another's disadvantage
is equally clear where there has been no use but rather a reduc-
tion in risks of loss to the actor and an increase in risks of loss to
the other. The distinction between using another'sperson or prop-
to save the lives and property of others, and not merely to benefit himself. See
pp. 429-30 infra.
3 See Holmes, The Theory of Torts, 7 AM. L. REV. 652, 654 (I873), reprinted
in 44 HARV.L. REV. 773, 775 (I93I):
The growth of law is very apt to take place in this way: Two widely different
cases suggest a general distinction, which is a clear one when stated broadly.
But as new cases cluster around the opposite poles, and begin to approach each
other, the distinction becomes more difficult to trace; the determinations are
made one way or the other on a very slight preponderance of feeling, rather
than articulate reason; and at last a mathematical line is arrived at by the
contact of contrary decisions, which is so far arbitrary that it might equally
well have been drawn a little further to the one side or to the other. The
distinction between the groups, however, is philosophical, and it is better to
have a line drawn somewhere in the penumbra between darkness and light, than
to remain in uncertainty.
34 E.g., Esso Petroleum Co. v. Southport Corp., [I956] A.C. 2I8 (oil company
not liable for expenses of cleaning oil from shore to which it was carried by wind
and tide after it was discharged to lighten a tanker which had run aground);
Cordas v. Peerless Transp. Co., 27 N.Y.S.2d I98 (N.Y. City Ct. I94I) (cab driver
and his employer were held not liable when armed bandit jumped into a taxicab,
driver accelerated rapidly, slammed on his brakes and leaped from the cab, which
then veered to the sidewalk and struck a pedestrian). Not only was the shipowner
using the dock in Vincent, but also it is arguable that damage to the dock was
substantially certain to occur, whereas in these cases there may have been only
a risk. Professor Morris also suggests, as a distinction between Vincent and Cordas,
that liability in Vincent serves a public policy of reducing the dockowner's
incentive to cast the ship loose, whereas no promise of compensation is needed to
affect the pedestrian's behavior. MORRIS, TORTS 45-46 (I953).
4I8 HARVARDLAW REVIEW [Vol. 72
erty and increasing risks to his person or property may be useful

in expressing a limit in scope of the concept of unjust enrichment
as customarily understood today; but it is far from clear that this
limit is sound. Perhaps better methods can be found for fixing
essential limits upon the generalization that enrichment through
another's loss is a ground for liability.
IV. CONDITIONALFAULT
A. Definition and Illustration
The broad concept of unjust enrichment implicit in Vincent
involves us in considerations relevant to the whole range of tort
law. Even in the absence of contributory fault, tort law some-
times holds the shifting of advantages and disadvantages per-
missible. This may be done by saying that there is no prima facie
tort, as in the illustration of a service station which interferes
with use of adjoining lands short of a nuisance, and also in cases
of prudent motoring, prudent construction and use of a railroad
grade crossing rather than a more expensive overpass, and per-
missible puffing of one's wares. At other times this is done by
saying that there is a privilege, as in cases of self-defense and
cases of defamatory communication in the course of judicial
proceedings. What determines whether such enrichment is unjust
or permissible? What determines whether conduct which gains
advantages for the actor while causing disadvantages to another
gives rise to liability?
The answer in most of tort law has been based on fault. Cases
such as Vincent, on the one hand, and on the other hand cases of
strict liability, as for ultrahazardous activities, have been re-
garded generally as independent diversions from the main theme
of fault.35 But it is useful to view them as closely related aspects
of a single concept which, though different from fault as we use
it in tort law generally, bears some important similarities. Though
there is no magic in the term, "conditional fault" is a convenient
way to designate this concept.
Perhaps the concept of conditional fault can best be introduced
by some hypothetical cases. As to each case, ask yourself whether
the conduct of the actor is blameworthy. "Blameworthy"is being
35 But see EHRENZWEIG,NEGLIGENCEWITHOUT FAULT 56-57 (I95I) (observing

that there is a significant analogy between these areas of liability).
used here in the extra-legal sense of something the actor ought not
to do, by some standard of judgment-other than a legal standard
- i.e., by one of that group of standards ordinarily referred to
as moral standards. In view of the ambiguities in the word
"blameworthy," it may be helpful to consider these subsidiary
questions: Apart from the question whether the actor's conduct
falls below standards enforced by legal sanction, does it fall
below standards which you would consider appropriate for guid-
ing your own conduct? For evaluating the conduct of another?
Does it fall below standards generally accepted in the community
as appropriate for guiding and evaluating the conduct of an
individual?
Case I-A: The actor is a person engaged in nonnegligent blast-
ing. Before commencinghis operations, he posted a bond for pay-
ment of damages resulting without negligence, as well as those
negligently caused, and in an amount which a prudent observer
would consider adequate to cover all injuries of which there is any
substantial risk. I assume that substantially all who consider this
case will answer that there is nothing morally blameworthy in this
conduct.
Case I-B: The situation is the same as in Case I-A, except that
the actor has posted only a bond to cover injuries resulting from
negligence, and no bond to cover the risk of injuries from non-
negligent blasting, and he is a shoestring operatorwith no financial
assets. He cannot pay for certain injuries which experience with
blasting has taught us may well occur. Is his conduct blame-
worthy? I suggest that most persons will answer "yes." He ought
not to engage in blasting operations, even though conducted with
great care, if he is unpreparedto compensate for resulting harm.
It has been man's experience that blasting, even when conducted
with great care, involves a severe risk. That risk should not be
created, even to gain the benefits of improved, less expensive,
socially useful construction, without making provision for com-
pensation of injuries.
Case II-A: The actor is a shipowner, and has no contractual
arrangementwith the owner of a dock. To avoid the risk of loss
of the ship or severe damage in an impending storm, the ship-
owner brings his ship to the dock and remains there through the
storm, knowing that it is likely that damage will be done to the
dock.36 He does so with the intention of paying for such damage,
38 This variation on the
Cf. Ploof v. Putnam, 8I Vt. 47I, 7I Atl. i88 (igo8).
and in fact does pay for all the damage to the dock which occurs.
I assume, again, an answer that there is nothing morally blame-
worthy in this conduct.
Case II-B: The situation is the same as in Case II-A, except
that the actor has no intention of paying for the damage to the
dock, and refuses to do so voluntarily. I suggest that many will
answer "yes" - his conduct is morally blameworthy.
Case III-A: The actor is a person engaged in a business - any
business involving an ordinary degree of risk to its employees.
Let us say he is operating a grocery store. He is a person of very
limited assets, but he carries workmen's-compensationinsurance,
and thus provides scheduled benefits even for the injuries suffered
by the employee who negligently slips on a banana peel which
the employee himself has dropped. There is nothing morally
blameworthyin the employer'sconduct.
Case III-B: The situation is the same as in Case III-A, except
that the employer does not carry any insurance covering injuries
to his workers. I suggest that many persons will answer "yes"
- his conduct is morally blameworthy. Moreover, this same an-
swer will be given by many if the facts are changed to the case of
an employer with very limited assets who carries only liability
insurance and no workmen's-compensationinsurance, and whose
business is in a jurisdiction where the statutes do not require the
carrying of workmen's-compensationinsurance, but merely seek
strongly to encourage the practice by the statutory withdrawal of
common-law defenses. That is, the sense of moral obligation to
carry insurance for the protection of workers, though interrelated
with the legal obligation, is more fully developed than the latter.37
In Cases I-B, II-B, and III-B, it appears likely that risk-
spreading capacity is a factor affecting the suggested moral judg-
ments. If you ask a layman to explain his reactions to these cases,
typically one of the reasons given is that the actor can easily in-
sure or otherwise pass along the cost. As to Case II-B, however,
fact situation in Vincent is used here to avoid distractions incident to the problem
of construing the contract between the parties in Vincent. See note 26 supra.
3 Cases of injuries to seamen and those covered by the Federal Employers'
Liability Act present quite different problems from other employment cases, as to
existence or nonexistence of a moral obligation to provide for nonnegligently
caused injuries. It is apparent that most injuries from these types of work are
compensated under a negligence rationale, and that the prevailing attitude among
the employees themselves, as expressed through their unions, is to avoid any en-
croachment of the workmen's-compensation principle as if it were the plague.
I959] CONDITIONAL FAULT 42I
this is true of both shipowners and dockowners; risk-spreading

capacity affords no affirmativebasis for choosing between shifting
the loss and letting it lie.38 Moreover, even when the actor's group
is a superior risk bearer, this factor is only one among several
affecting the moral judgments. If it were otherwise, we would
feel the same sense of moral obligation in all other cases involv-
ing such superior risk-spreading capacity.
B. The Attitude TowardMotoring

Case IV-A: The actor is a person engaged in prudent motor-
ing, is normal in that he is relatively judgment-proof, still owing
several installments on the car, but is in the still unusual situation
of having substantial accident insurance for the protection of
persons injured by his motoring,39as well as adequate liability
insurance.40 His conduct is not blameworthy, though he non-
negligently causes an injury - e.g., when his vehicle gets out of
control without negligence and strikes a parked car in which the
victim is sitting.
Case IV-B: The situation is the same as in Case IV-A, except
that the actor has made no advance provision for compensation
of injuries other than those caused by his negligence and refuses
afterward to pay for harm nonnegligently caused to the victim in
the parked car. Is his conduct blameworthy? Though it is now
possible and feasible to obtain at least some accident insurance
for the benefit of such a victim of nonnegligentmotoring, I assume
that most persons will answer the question "no." Why? It is not
solely because risk-spreading capacity is less apparent or less
familiar in the present stage of the development of accident in-
surance. Probably another influential factor is that few motoring
accidents arise without negligence, and the relatively slight risk
of injury from prudent motoring does not tip the moral scales
toward an obligation to provide against that risk. Still another
factor, perhaps, is the nearly universal use of cars.
38 Cf. Momus, TORTS 44 (1953).

39 I.e., insurance providing that the company shall pay specified benefits to
persons injured by the policyholder's motoring, without regard to whether the
policyholder is negligent. "Medical payments" coverage is an example of this
type of insurance.
40e., insurance by which the company agrees to pay on behalf of the policy-
holder sums which the latter is legally obligated to pay to third persons who are
injured by his motoring.
In any event, whatever may be the complex of factors that

produces these moral judgments, the legal distinctions which im-
pose liability without negligence in cases of employee injuries
and injuries from blasting, but do not impose liability without
negligence in motoring cases, are consistent with community moral
attitudes toward the conduct of the actors involved. This is not
to say, however, that no change in the present method of handling
automobile injuries is to be expected. The contrary is true.
While, in general, tort law should not take the initiative of move-
ment in a direction that is offensive to the moral standards of the
community, neither should it lag far behind the developing moral
standards. A change in the attitude of the community concerning
the moral responsibility of motorists to compensate the victims
of motoring has been in progress for years. There has been an
increasing prevalence of the view that it is blameworthy for a
person of no great means to enjoy the benefits of motoring with-
out having liability insurance for the protection of potential vic-
tims of his negligence. This view has supported the legal develop-
ments of compulsory liability insurance in two states now,4' and
other forms of financial-responsibilitylegislation in all others.42
These statutes have been aimed not at developing liability without
negligence, but rather at assuring financial ability to pay, within
moderate limits, for the injuries caused by negligence. It seems
unlikely, however, that the trend will stop at this point. On the
other hand, it may stop short of the adoption of an accident-
insurance plan which is as completely divorced from negligence
as is true of workmen's compensation. A likely possibility is the
development of a system combining relatively modest but certain
compensationwith additional recovery dependentupon negligence.
One important reason for doubting that the workmen's-com-
pensation principle will be accepted completely in this area is a
difference as to the focusing of moral responsibility. The em-
ployer is in a position superior to that of the worker with respect
to capacity for making the work safe. It is primarily the em-
ployer, rather than the individual worker, who has the capacity
for developing a safe place to work, safe tools, and safe methods,
and the capacity for controlling the hazardous activities of other
41 MASS. GEN. LAWS ch. 90 (1932); NEW YORK VEHICLE & TRAIc LAW
?? 93-93k.
42 Legislation of all the states is collected in INS. L. REP. (Auto.) 500I-7205
(I958).
workers. Thus, there is one person, the employer, upon whom

devolves the primary responsibility for making the worker safe
and compensating for his injury if safety precautions are in-
effective. In motoring accidents, on the other hand, two or more
drivers are usually involved; their opportunities and capacities
for avoiding accidents are ordinarily about equal. We are en-
couraged to distinguish between them according to the character
of their conduct because of our desire to fix legal responsibility
consistently with notions of moral responsibility. Negligence still
suggests itself as a basis for making such distinctions. Arguably
this desire could be served by giving to such distinctions only the
role of affecting insurancepremiums,but at the present moment it
seems likely that they will and should continue to have more sig-
nificance. Even though the force of the moral attitude toward the
motorist causing injury to another is diluted by the availability to
him of liability insurance, the moral attitude as to what the in-
jured person deserves remains unaffected. Moderate, scheduled
benefits meeting at least a subsistence level may seem satisfactory
if the injured person alone was at fault, but full redress for his
economic loss seems more appropriate if he was the victim of a
motoringaccident caused solely by the fault of another. The sac-
rifice in workmen's-compensationlegislation of a similar desire to
give full redress to the worker injured through negligence was
not as great as it would be in motoring cases. Prior to legislation,
relatively few employees could overcome the hurdles of proving
fault and defeating the defenses of contributory negligence, as-
sumption of risk, and the fellow-servant rule. For injured work-
ers as a group, the sacrifice of higher recoveries in these few cases
was a small price to pay for a legislative change allowing recovery
of at least moderate benefits in substantially all cases of injury.
On the other hand, such a sacrifice would affect a high percent-
age of persons injured from motoring, and it is unlikely that it
would be acceptable. The alternative of full compensation for
all economic loss from motoring accidents also seems an unlikely
one, for the present at least, because of the common prediction
that the cost in insurance premiums would rise sharply. If this
guess is proved erroneous, as it may well be in view of the high
percentage of victims now compensated by settlements at least
equal to the economic loss, the prospects for a closer approach to
the workmen's-compensationprinciple in motoring cases would
improve.
C. How Much Compensation?

It may be suggested that one who provided only the statutory
workmen's-compensationbenefits would not be considered blame-
worthy with respect to the financial misfortune of the victim of a
serious accident. Yet, those benefits might be far short of full
compensation for the harm suffered,and normally would be much
less than the customary measure of damages in tort cases. Prob-
ably the community's moral judgment is that conditional fault
is avoided if reasonableprovisions have been made for compensa-
tion, even though the compensation is not equivalent to the tort
measure of damages. Perhaps the moral judgment of what is
reasonable for the individual employer is so strongly influenced
by what the statutes provide that they substantially control that
judgmentat present.
In workmen's-compensation cases, the legal discrimination
between the employer who has provided specific, moderatebenefits
and the employer who has provided none is assured by statutory
fiat. Might the same discriminationbe made by a court as between
a defendant who had no insurance and one who had obtained
accident insurance for his blasting operations? Or for his air-
craft operations? In certain aircraft cases, the Warsaw Conven-
tion requires a modest sum of compensation regardless of fault.43
If aircraft operators made a similar provision by means of acci-
dent insurance,with contract provisions making the benefits avail-
able for all harms resulting from aircraft operations, would it not
be less likely that courts and legislatures would impose strict
liability? We are not necessarily tied to old notions of the meas-
ure of damages in cases of fault in the traditional sense; rather,
when conditional fault is the basis of liability, it is expectable
both that the advance provisions sufficient to negate conditional
fault, and the amount of recovery allowed when conditional fault
is shown, will vary from the traditional measure of damages in
tort cases. For example, in cases of conditional fault a limitation
to economic loss, omitting any compensation for pain and suffer-
ing, seems a likely development.44
43Though the United States is not a signatory of the convention, it adheres

to the convention rules as the result of a presidential proclamation. 49 Stat. 3000
('935).
44 Cf. Jaffe, Damages for Personal Injury: The Impact of Insurance, I8 LAW
& CONTEMP. PROB.2I9 (I953).
'959] CONDITIONALFAULT 425
D. Some CriticismsAnticipated
One anticipated criticism of the concept of conditional fault is
related to the interdependence of moral and legal obligations.
Among the eight illustrations examined, each of the three cases
in which a sense of individual blameworthiness can be perceived
is also a case in which a legal obligation has been recognized. It
might be argued that the moral obligation exists only because of
the imposition of the legal obligation. It is true that any standard
of legal liability, whether based on blameworthiness of conduct
or on some wholly independent ground, invokes that moral sup-
port which is incident to respect for law as an essential instrument
of social order. For this reason, it is impossible to account for
the moral sense of the community, concerning conduct which
subjects one to legal liability, as something existing wholly apart
from legal developments. The development of the legal rule in-
fluences the development of moral attitudes. This is apparent,
for example, in Case III-A and Case III-B of the illustrations (the
workmen's-compensationproblems), in which the legislative en-
actment has a strong bearing on our present moral attitudes con-
cerning the blameworthinessof an individual employer's conduct.
Decisional rules also have their impact on moral attitudes. But
this demonstrates only an influence, and not that moral attitudes
are produced solely by legal rules. One indication that moral atti-
tudes have vitality beyond that born of respect for the legal rules
of liability is seen in the fact that the scope of the moral obligation
is in some instances broader. For example, there is now a widely
held sense of moral obligation that an employer should provide
workmen's-compensationbenefits for his employees, even in the
majority of states in which the law gives him the option of doing
without such insurance and paying for only those injuries negli-
gently caused, though with the disadvantage that defenses of
contributory negligence, assumption of risk, and the fellow-serv-
ant rule are withdrawn. Another example may be seen in a com-
parison between Case II-B (the shipowner'suse of the dock with-
out compensation) and the circumstances of a recent English
case, holding an oil company not liable for the expenses of clean-
ing oil from the shore to which it was carried by wind and tide
after it was discharged to lighten a tanker which had run
aground.45I suggest that most persons would regard the conduct
45 Esso Petroleum Co. v. Southport Corp., [I956] A.C. 2I8.
of discharging the oil without provision for compensation, either

before or after harm occurred, as morally blameworthy.46 Yet
this recent English decision stands against liability, and there is
other English authority indicating that the principle of Vincent
may not be accepted there.47 Moreover, it is not at all clear that
Vincent will be accepted generally in the United States, and since
the case of the dischargedoil can be distinguished from Vincent,48
it is still more doubtful that our courts would have imposed liabil-
ity on the oil company. Therefore, the scope of the recognized
moral obligation is broader than that of the legal obligation thus
far developed.49
A second anticipated criticism of the concept of conditional
fault is based on the observationthat the characteristicdistinguish-
46 Compare the suggestion that "one who is under no legal liability for damage
caused to another may yet think it right and proper to offer some measure of
compensation." SALMOND,TORTS 25 (I2th ed. Heuston I957). The supporting foot-
note states: "Contrast the attitude of the successful appellants in Bolton v. Stone,
[1951] A.C. 850 ('The cricket clubs of this country who supported the appeal to
the House of Lords have done everything that they can to see that Miss Stone
does not suffer financially'; 68 L.Q.R. 3), and Esso Petroleum, Ltd. v. Southport
Corporation [I956] A.C. 2i8." In Bolton v. Stone the plaintiff was denied recovery
for injuries suffered when she was struck by a ball which went over the fence and
into a public way due to an exceptional though not unprecedented hit.
47 See Cope v. Sharpe, [I9I2] i K.B. 496 (a fire started on plaintiff's land;
defendant was gamekeeper for one Chase to whom plaintiff had leased the shooting
rights on the land; defendant started a back fire to protect nesting pheasants;
the main fire was put out before it reached the back fire; defendant, having acted
reasonably, was held not liable for harm done by the back fire). But cf. Gilbert
v. Stone, Style 72, 82 Eng. Rep. 539, Aleyn 35, 82 Eng. Rep. 902 (I647) (trespass
quare clausum fregit, and taking of a gelding; the court sustained plaintiff's de-
murrer to defendant's plea that he acted for fear of his life, being threatened by
twelve armed men). Should a distinction be made between duress by fire and
duress by men? Between damaging land and entering land to take personalty? Was
the privilege to protect the pheasants by back fire reasonably implied in the terms
of the lease in Cope v. Sharpe? In this respect Cope v. Sharpe involves a contract-
interpretation problem not unlike that in Vincent, see note 26 supra. For a dis-
cussion of other English cases bearing on the problem, see WINFIELD, TORT 6i-68
(6th ed. Lewis I954).
48 See note 34 supra.
4 Cf. Williams, The Aims of the Law of Tort, in 4 CURRENT LEGAL PROBLEMS
137, I42 (I95I), which suggests that there is a sense of moral obligation to make
compensation ("ethical compensation"), even in the absence of legal obligation,
when one borrows a friend's book and loses it nonnegligently. But elsewhere in
the same paper Professor Williams argues that the principle of ethical compensa-
tion, requiring culpability, does not justify "strict liability" (which phrase he
appears to use in the sense of liability without a requirement of intent or negli-
gence). Id. at I5I-52. Thus it appears doubtful that he would approve the argu-
ment in the text of the present article.
1959] CONDITIONALFAULT 427
ing the three cases (involving a blaster, a shipowner, and an em-

ployer) in which it was suggested that blameworthiness was
found from those involving no blameworthiness was the failure
of the actor to provide compensation. It might thus appear that
the moral judgments were based solely on failure to pay. If they
were so based, we would have been engaged in the question-beg-
ging process of elaborating these two propositions: (i) The actor
should pay because his conduct is blameworthy. (2) His conduct
is blameworthy because he did not pay or make provision for
payment. But it is a misstatement of these moral judgments to
characterizethem in the form of the second of these two proposi-
tions. They are based upon a combination of factors, only one
of which is failure to make provision for payment. For com-
parison, consider the fact that an intent to strike without consum-
mation of the blow does not amount to battery, nor does the blow
without the intent, nor do both, if the victim consented. Intent
is only one of several requisites of battery. So also, failure to
provide compensation is only one of several requisites of the
blameworthiness which we have found in these cases. Enrichment
of the actor at the expense of another is a second requisite, and
there appear to be others as well. It is to be expected that we shall
develop different sets of requisites by developing categories of
liability for conditional fault, just as we have categories of liability
for fault.
E. Blameworthiness- Three Classes of Conduct

Though it appears that the community sense of morality finds
blameworthiness in the nonnegligent, risky conduct as to which
courts have imposed liability, yet this is distinctly different from
the blameworthiness found in most cases of negligence or inten-
tional tort. The sanctions which the courts impose are sometimes
different. In view of these differences, it may be helpful to identify
three types of conduct with which tort law must be concerned.
This is a classification from the point of view of blameworthiness
and does not coincide with the three-fold classification of tort
liability as intentional torts, negligence, and strict liability.
Type one: It is the moral sense of the community that one
should not engage in this type of conduct, even though he makes
provision for compensationof resultinglosses.
Type two: It is the moral sense of the community that one
should not engage in this type of conduct, because of risk or cer-
tainty of losses to others, without making reasonable provision

for compensation of losses.50 But if he makes such provision, his
conduct is permissible.
Type three: It is the moral sense of the community that this
type of conduct is permissible, and there is no moral obligation to
provide for resulting losses, unless the conduct violates some legal
standard not concernedwith blameworthiness. When such a legal
standard exists, it is a true deviation from the principle that lia-
bility in tort is based on blameworthiness. Of course the legal
standard when promulgated invokes moral support for law,
but the standard itself is based on some policy other than the
principleof liability for blameworthyconduct.
Type one is fault. The legal result usually reached in cases in-
volving this type of conduct is liability or, from another point of
view, absence of privilege. Type three is absence of fault. The
legal result usually reached in cases involving this type of conduct
is nonliability or, from another point of view, privilege. Type two
is conditional fault. From another point of view, this type of
conduct supports the legal conclusion of conditional privilege."1
In this type fall the cases of privilege to use the property of
another in circumstances of private necessity, the privilege being
subject to a duty to compensate for any harm done. Vincent is
50 Cf. Stallybrass, Dangerous Things and the Non-Natural User of Land, 3

CAMB. L.J. 376, 387-88 (I929):
The principle of law behind all these cases is, it is submitted, that if a man
takes a risk, which he ought not to take without also taking upon his own shoul-
ders the consequences of that risk, he shall pay for any damage that ensues.
In every case the question really is: Was the risk one which the defendant was
entitled to take only on condition of paying compensation to those injured
thereby irrespective of any negligence on his part?
Though this passage is a statement of legal principle, it is at least suggestive of
the view that the legal principle is supported by a moral judgment of the conduct,
and by standards concerned with individual blameworthiness.
51 A comment about terminology may be needed to avoid confusion. The
privilege in a case such as Vincent, referred to in the text above as a conditional
privilege, is often called an incomplete privilege, see Bohlen, Incomplete Privilege
to Inflict Intentional Invasions of Interests of Property and Personalty, 39 HARV.
L. REV. 307, 3I3 (I926), and is sometimes called an incomplete conditional priv-
ilege, see I HARPER & JAMES, TORTS 6i (I956). In the latter of these forms of
terminology, however, the idea intended to be suggested by the word "conditional"
is different from that in the text above; as used by Harper and James, "conditional"
refers to the requirements that the purpose of the actor in destroying or invading
the other's interest in property be a proper one and that his action be reasonable
under the circumstances. In the face of this different usage of "conditional," I
choose it here only for lack of a better word to serve, as this one does, to describe
both the kind of fault and the kind of privilege.
I959] CONDITIONAL FAULT 429
such a case of conditional privilege. Cases of so-called strict

liability for nonnegligent risky conduct, such as blasting, are also
of this type. The blaster is privileged to engage in the reasonable
use of explosives, despite the fact that his conduct imposes risks
upon others, but his privilege is subject to a duty to compensate
for harm done.
The analogy between conditional privilege to use the property
of another because of private necessity and conditional privilege
to make reasonable use of explosives is very close. It might be
argued that they should be treated exactly alike. Probably a
difference is justified, however, in favor of a broader privilege,
i.e., a less restrictive condition upon the privilege, in the case of
private necessity, at least when preservation of life is involved.
This point may be seen if we assume, first, the case of a blaster
who is judgment-proof and has no bond or liability insurance. I
submit that his blasting could be enjoined. Though blasting under
given circumstances may be reasonable conduct for one who is
able to compensate the victims of foreseeable injuries, it is un-
reasonable conduct under circumstances which are identical
except for the fact that the blaster is judgment-proofand without
bond or insurance. The conduct in the latter situation should be
enjoined as an unreasonable interference with the use and enjoy-
ment of lands within the area of risk - as a nuisance. The con-
duct of blasting is of such a nature that it is the moral sense of the
community that it should not occur without an advance provision
for compensation.
On the other hand, it may well be that in the cases of private
necessity involving the preservation of one's own life, the condi-
tion would not be an advance provision for compensation, but
rather only an obligation to compensate after loss has occurred.
A reason for the distinction lies in the absence of opportunity to
make advance provisions in the emergency situation. Another
arguable ground for the distinction is that the interest in pres-
ervation of life deserves greater weight than interests concerned
with property. But this ground has the weakness that, though
one does not ordinarily think of blasting as being closely asso-
ciated with the preservation of life, such an association can be
clearly seen in cases of blasting to construct a hospital, and it is
often present in other cases as well.
If, assuming still another variation of the facts, the case of
private necessity is one involving prudent conduct of the actor for
the preservation of the lives of others and not his own life, it is
less likely that his conduct, including failure to pay for resulting
harm, would be considered blameworthy. Arguably, at least, this
type of case involves no moral blameworthiness and should be
treated in law as a case of absolute privilege or no fault, rather
than a case of conditionalprivilege or conditional fault.
The classifications of liability according to negligence and in-
tentional tort do not fit perfectly within this threefold classifica-
tion concerning the degree of fault. Probably most negligence
cases should be placed within the first classification- conduct
of the type which is blameworthy, irrespective of provisions for
compensation.52But some cases of liability for negligence involve
only conditional fault. For example, in some instances, failure of
a railroad company and its servants to measure up to the hyper-
critical standards of "negligence" in FELA cases 53 cannot be
regardedas blameworthy in the sense of the first classification-
i.e., irrespective of provision for compensation. But the same
moral obligation underlying workmen's compensation supports
an allowance of compensation to the injured worker, though it
may not dictate that the FELA measure of compensation be
allowed in preference to that of a workmen's-compensation
act, or vice versa. Other negligence cases which do not involve
conduct of the first type of blameworthiness are those in which,
because of the use of an objective standard for judging whether
conduct is negligent, liability is imposed on one whose substand-
ard capabilities have been the primary reason for the actor's
failure to measure up to legal requirements. Probably most such
cases can be included properly within the second classification
conditional fault - as to which it is the moral sense of the com-
munity not merely that the legal result of liability is fair and just,
but also that part of the justificationis in a kind of personal blame-
worthiness. Perhaps the taint of blame in these cases arises from
the belief that, even though the actor was trying as hard as would
the law's ordinary prudent person, he could have made up for his
deficient capacities by trying harder; the occurrence of the acci-
52 It would seem that most motoring accidents, for example, result in part
at least from factors other than a substandard capacity. Speed is an illustration.

For a view which places greater emphasis on limitations of capacity in contrast
with personal blameworthiness, see 2 HARPER & JAMES, TORTS 734-4I (I956);
James & Dickinson, Accident Proneness and Accident Law, 63 HARV.L. REV. 769
(I950).
"See Note, 69 HARV.L. REv. 1441 (i956).
1959] CONDITIONALFAULT 43 I
dent was not beyond his control. The view that one should try
to make up for deficient capacities would support the inclusion
of some of these cases within the first classification of uncondi-
tional fault.54
There remain at least a few cases, such as those imposing lia-
bility on the mentally deranged adult for failure to measure up to
ordinary prudence,55which clearly fall in the third classification
of conduct that which is not even conditionally faulty. These
cases representtrue deviations from the principle that tort liability
is based on blameworthiness.
Among the intentional torts, without doubt, may be found
illustrations of conditional fault, as well as some illustrations of
true deviation from the principle that tort liability is based on
blameworthiness. The forced purchase of the hat which, it is
generally supposed, is converted by being taken from the hat rack
by a nonnegligentmistake 56 is such a deviation, as are some cases
of liability for trespass to land. But in many instances of so-
called strict liability for interference with the possessor's interest
in land, and perhaps in most instances of substantial harm, it will
be found that the interference has been part of a reallocation of
advantages and disadvantagescomparableto that in Vincent. For
example, the person who nonnegligently sets fire to a structure
while using it in the mistaken but reasonable belief that he has the
owner's consent to do so 57 has gained the advantage of use while
subjecting the other to the disadvantage of increased fire risk to
his property. At least when the owner has not contributed to the
mistaken belief that he has consented, a compelling argument of
conditional fault is available. A legal rule of liability in such cases
is supported by a moral judgment that the individual who so uses
another's land without compensating for harm done is blame-
worthy; indeed, it is probable that this moral judgment would
extend to the case of such use with the consent of the owner,
"4 This notion of an obligation to make up for deficiencies by trying harder to
avoid accidents is expressed in the legal rules regarding "physical" deficiencies

such as blindness, see, e.g., Weinstein v. Wheeler, I4I Ore. 246, 15 P.2d 383 (1932);
Smith v. Sneller, 345 Pa. 68, 26 A.2d 452 (1942), and small stature which impairs
one's ability to see road hazards while driving, see Mahan v. State ex rel. Carr, 172
Md. 373, I9I Atl. 575 (i937).
55 PROSSER, TORTS 793 (2d ed. I955); RESTATEMENT, TORTS ? 283 (Supp. 1948).
56 See, e.g., SEAVEY, COGITATIONS ON TORTS 52 (I954).
57 Cf. Southern Counties Ice Co. v. RKO Radio Pictures, Inc., 39 F. Supp. 157
(S.D. Cal. 194I); Newsom v. Meyer, 102 Conn. 93, I28 Atl. 699 (1925); Wyant
v. Crouse, 127 Mich. I58, 86 N.W. 527 (1901).
absent some agreement as to responsibility for the risk of such

harm from fires nonnegligentlystarted by the user.
In the area of defamation, there are additional instances of
liability without blameworthiness.58 But here, as in trespass to
land, some of the cases of so-called strict liability appear to be
based on conditional fault. Is there not an analogy, in the extra-
hazardous character of the activity, between blasting and mass
communicationof derogatory information?
When a remedy in the form of a money judgment is allowed
against one who by "innocent misrepresentation" intentionally
induced another to enter into a contract,59conditional fault is
involved. It is of particular interest that the rationale of unjust
enrichmenthas been used to justify the result in such cases.60
In most cases of nuisance, the defendant's conduct falls in the
first classification- fault; his conduct produces an interference
with the interests of others which is unreasonable irrespective of
whether he makes provision for compensation. But some nuisance
cases involve only conditional fault. This appears to be true, for
example, of mining activity in the hills of Tennessee which the
court would not enjoin unconditionally because it was extremely
valuable both to the defendants and to the community; instead,
the court decreed that an injunction would issue unless the defen-
dants paid damages or gave security for such payment to the com-
plaining landowners.6"A decree denying an injunction on such a
condition may be viewed as a determinationthat the interference
is not unreasonable if compensation is provided, but is unreason-
able otherwise.62 Compare the case of blasting. It has been sug-
58 i HARPER & JAMES, TORTS 362, 367-70 (1956); PROSSER,TORTS 602-04 (2d
ed. 1955); RESTATEMENT,TORTS ?H 579, 580 (I938).
59E.g., Aldrich v. Scribner, 154 Mich. 23, II7 N.W. 581 (I908).
60 [I]t is not necessary that the person making the statement should either
know that it is untrue or be recklessly and consciously ignorant whether it be
true or not. It is sufficient if it be false in fact. It must be said, however, that
in the cases in which this principle has been applied, the defendant obtained
what the false representations caused the plaintiff to lose.
Id. at 28, II7 N.W. at 583. (Emphasis added.)
61 Madison v. Ducktown Sulphur, Copper & Iron Co., II3 Tenn. 331, 83 S.W.
658 (1904).
62 If the analysis in this article is sound, "unreasonable" has been used both
in negligence cases and in nuisance cases in a sense broad enough to include some
cases falling in the second classification of blameworthiness. Note also that "un-
reasonable" in the context of nuisance is not identical with "unreasonable" in the
context of negligence. Nonnegligent blasting by the judgment-proof defendant is
not "unreasonable" as that term is used in reference to conduct claimed to be
gested that an injunction would be granted against a judgment-

proof blaster, who made no provision for compensation,though not
against one who was financially responsible.63
F. The Moral Element in ConditionalFault

In the controversial writings about the basis of liability for
harm caused by nonnegligent, risky activity, the question whether
such liability can be brought within a rubric of "fault" has been a
point of contention. It has been noted that in part the controversy
is one of definition.64"Fault" has been used to characterize, first,
individual conduct falling short of some set of generally accepted
moral standards for individual behavior, second, individual con-
duct falling short of a legal standard for determining whether
some relief should be granted, and third, individual conduct falling
short of standards for judging whether conduct is socially desir-
able or instead is antisocial. These different senses are suggested
negligent, but is "unreasonable," absent satisfactory provision for compensation,

as that term is used in reference to interference claimed to be a nuisance. Shifting
usage of "unreasonable" has produced interesting clashes of opinion.
Professor Seavey asserts that nuisance is distinct from the principle of Rylands v.
Fletcher and from liability for ultra-hazardous activities, and that there can be no
liability without fault in nuisance since, by definition, nuisance is an unreasonable
interference. Seavey, Nuisance: Contributory Negligence and Other Mysteries, 65
HARv. L. REV. 984, 985-86, 988 (1952).
On the other hand, Dean Prosser, employing a somewhat different meaning of
"unreasonable," avoids contradiction within his own terminology by referring to
the principle of Rylands v. Fletcher as one of strict liability rather than liability
without fault. PROSSER, TORTS3I8 (2d ed. 1955). He asserts first, that Rylands
v. Fletcher is based on the defendant's "maintaining a condition or activity which
involves an unreasonable risk of harm to persons or property in its vicinity,"
PROSSER, The Principle of Rylands v. Fletcher, in SELECTED ToPics ON THE LAW
OF TORTS 135, i85 (I953), and second, that many courts rejecting this principle
in name apply it in other guises, of which nuisance is one. Id. at 159-77; PROSSER,
TORTS 336, 393 (2d ed. I955).
Professors Harper and James, though agreeing that many courts reluctant to
approve Rylands v. Fletcher by name are achieving similar results by "manipula-
tion of other techniques," 2 HARPER & JAMES, TORTS 8I9 (I956), decline to
accede to Dean Prosser's different usage of the term "unreasonable" and criticize
his position as an effort "to assimilate the principle of Rylands v. Fletcher to
the 'unreasonable risk' analysis of negligence." Id. at 8oI-02.
I suggest that when allowance is made for different preferences in form of ex-
pression, there is less substantive difference among these friendly scholars than
their verbal assaults appear to indicate.
63 See p. 429 supra.
64 PROSSER, TORTS 317 (2d ed. 1955); PROSSER, The Principle of Rylands v.
Fletcher, in SELECTEDToPics ON THE LAW OF TORTS I35, I8I (I953).
by the terms "individual fault," "legal fault," and "social fault."

Plainly the concept of "legal fault" is not helpful in explaining
the basis of liability. It is a metaphoricalappeal to include within
"fault" those cases of legal liability which do not involve moral
blameworthiness, either individual or social.65 The term "fault"
used in any sense which strips the moral element away and substi-
tutes a purely legal meaning becomes merely another way of say-
ing the actor is subject to legal liability.66
Some writers have found a concept of social fault underlying
liability for a harm.caused by nonnegligent, risky conduct,67but
most have balked at the notion that there is individual fault in
these cases.68 If the speculations about the illustrative cases given
above are sound, however, those actors who made no provision
for compensating the victims of their activities were individually
blameworthy under moral standards generally accepted today,
though the quality of blame is different from that found in most
cases of negligent or intentional torts. That difference justifies
the use of different sanctions by the courts, but it does not justify
the conclusion that the use of any sanction in the cases of condi-
tional fault is a departure from the general principle of tort law
that the shifting of loss is based on individual fault.
The moral element in the legal rules based on conditional fault
involves more than the mere fact that they do not offend the com-
munity's sense of fairness. These types of cases do not present
situations of moral indifference to the choice of one or another
method of social engineering. Moreover, the moral element is
BCf. KEIGWN, CASES ON TORTS I9 (3d ed. I929):

[T]he body of ethical principles to which our phraseology refers [if we say
that the ground of liability in tort is moral delinquency] is not the moral code
of popular speech, but an artificial and somewhat sublimited morality, which
is formulated by the law and is called morality only by a use of that term
which is almost metaphorical.
See also PROSSER, TORTS I6 (2d ed. I955).
88fSee Bohlen, The Rule in Rylands v. Fletcher, 59 U. PA. L. REv. 298, 3I3
87
See, e.g., PROSSER, TORTS I6, 3I8 (2d ed. I955); PROSSER, The Principle of
Rylands v. Fletcher, in SELECTED ToPics ON THE LAW Op TORTS I35, I80-8I (I953).
But see 2 HARPER & JAMES, TORTS 786-87 (i956): "Many activities there are which
create situations fraught with danger to others, but because of the general social
utility of the activities, the risk is not regarded as unreasonable. There is, accord-
ingly, neither ethical nor social fault in the carrying on of such activities."
88 See, e.g., 2 HARPER & JAMES, TORTS 786-87 (I956); PROSSER, TORTS I6, 318
(2d ed. I955); Seavey, Speculations as to "Respondeat Superior," in HvARvD LEGAL
ESSAYS 433, 439-45 (1934).
19593 CONDITIONAL FAULT 435
not adequately explained by the suggestion that the conduct of

the individual employer or blaster has been antisocial but not
blameworthy when no provision has been made for compensation
of the expectable injuries from nonnegligent operations. Social
morality is an important influence on law, but there is in these
cases a sense of blame against the individual as well. Of course
the assertion that individual blameworthiness is sensed in these
cases is not demonstrableby logic. Rather its validity is depend-
ent upon an observation of the moral views of the community.
The appeal for the reader's agreement with this assertion is an
appeal for confirmationfrom his own observations.
To say that a moral discrimination as to individual blame-
worthiness underlies the legal discrimination between prudent
blasting and prudent motoring is not, however, to say that we can
usefully seek or express the requisites of legal liability in terms of
the existence or nonexistence of blameworthiness. We do not do
so with respect to traditional theories of liability for fault, such
as negligence, nor should we hope to do so with respect to this
developing area of liability for nonnegligent, risky conduct. For
legal standards more rigid content than that of moral standards
is needed in order to serve the interests of stability, predictability,
and impartiality in the administration of justice. Inevitably the
question of blameworthiness,either as it was asked in the illustra-
tive cases in this article or in some other form, will evoke more
differences in its responses than would be tolerable for legal
standards.
The references in this article to moral standards generally ac-
cepted in the community as appropriate for judging the moral
quality of an individual's conduct are not intended to suggest that
there is a community moral sense which is an existing datum
awaiting our discovery and application. Rather, this community
moral sense is something in the process of articulation. It is more
in the nature of a supposed coincidence, for the time being, of an
indeterminatenumber of potential opinions about the moral qual-
ity of certain conduct. It is a supposed rather than certain coin-
cidence of potential rather than actual opinions, because the ques-
tions are often of a type as to which few persons have opinions,
and fewer still have considered opinions. The number is inde-
terminate because legal developments are not guided by such a
precise rule as, for example, the majority view; rather, as new
moral standards develop, either more or less conducive to loss

shifting than the old, legal developments often lag until there is
overwhelming acceptance of the moral standards. Occasionally,
on the other hand, legal developments give sanction to evolving
moral standards which would otherwise be unacceptable to the
majority.
To a very considerable degree, the community moral sense
which influences judicial and legislative lawmakers is a product
of reason and understanding. For example, the moral judgment
of the conduct involved in the case illustrations discussed above
representsa judgment supported in significant measure by reason,
which has been brought to bear in both appellate arguments and
legislative deliberations. Sometimes the reasons for such judg-
ments are patent, but often they are only vaguely perceived, and
they are somewhat better understood after we are put to the test
of deciding a succession of cases calling for refinements and dis-
criminations. Our legal rules are better expressed in terms of the
results in such cases, and grounds of distinction from cases de-
cided differently,than in abstract terms of the quality of an actor's
conduct. It is a part of the common-law method to build up a
body of case results and a body of opinions somewhat more gen-
eralized than the declarations of results themselves, and contin-
ually to test these one against the other. One type of generaliza-
tion produced is the statement of requisites for each of numerous
categories of liability. It is possible to develop statements of re-
quisites for the recognition of liability based on conditional fault
in various types of fact situations, just as this has been possible
in the areas of liability based on fault in the traditional sense.
This development of categories of liability for conditional fault
is already well underway,as in the articulation of requisites of lia-
bility for uncommon,ultrahazardousactivities.
G. Risk-Spreading Capacity as an Element in Conditional Fault

We speak of liability based upon fault. In fact, of course, fault
is a conclusion. If we press the analysis further, seeking reasons
for finding conduct faulty, we discover a great variety of reasons,
some but not all of which will exist in any given case of fault. Also,
we discover a variety of opposing considerations. And a given
reason pointing toward the conclusion of fault never controls the
I9591 CONDITIONALFAULT 437
decision except in a sense comparable to that in which a pair of

like votes controls when the margin of victory is one vote - except
in the sense of swinging the balance away from the side where it
would have been absent this factor. In this sense, of course, risk-
spreading capacity affects decisions, and probably somewhat more
today than in time past. But it is misleading to say that it is thus
supplanting fault as a basis of liability. Instead, it is an influence,
along with others, in the constant process of change in the mean-
ing of the concept of fault. Probably it has been one of the factors
influencing the gradual change in the meaning of negligence- a
change toward applying a more stringent standard than in earlier
times.69 This factor is also influential in the development of lia-
bility for risky conduct which does not infringe the test of negli-
gence even in its more stringent modernform.
We have seen 70 that among both intentional torts and negli-
gence cases are instances of liability not based on blameworthy
conduct, even in the less severe form of conditional fault. It does
not appear, however, that risk-spreading capacity has been a
significant factor in the development of these deviations from lia-
bility based on blameworthiness. It is a matter of interest, more-
over, that even as to areas of liability for nonnegligent, risky con-
duct, it is rare to find a judicial opinion recognizing that superior
risk-spreading capacity is either a requisite of liability or an
influence upon the decision.7' One reason for this lack of reliance
upon risk-spreading capacity may be found in the fact that ra-
tionally using it as a factor affecting either the moral conclusion
of blameworthinessor the legal conclusion of liability requires a
comparison between plaintiff and defendant, rather than merely
an appraisal of the defendant's capacity.72 Earlier in this
69
See Feezer, Capacity to Bear Loss as a Factor in the Decision of Certain Types
of Tort Cases (pts. 1-2), 78 U. PA. L. REV. 805 (I930), 79 U. PA. L. REV. 742
(I93I) -
70 See pp. 430-32 supra.
71 As one of the rare examples, see Bridgeman-Russell Co. v. City of Duluth,
I58 Minn. 509, I97 N.W. 97I (I924), in which liability was imposed on the defendant
city, without proof of negligence, for harm resulting from a broken water main.
The opinion noted that under its holding, the whole community bore the loss,
and that if the opposite result had been reached, the burden would have been too
heavy upon one person. Id. at 5II, I97 N.W. at 972.
72 Cf. MORRIS, TORTS 246-53 (I953); Morris, Hazardous Enterprises and Risk
Bearing Capacity, 6i YALE L.J. I,72, II76-77 (1952).
article " it was suggested that the better form of the argument
for risk-spreadingcapacity focuses attention upon the class rather
than upon the individual defendant. The risk-spreadingcapacity
of the defendant's class is a good reason for shifting the loss only
if it is substantially better than that of the plaintiff's class. Thus,
as increasingly broad accident-insurancecoverage becomes avail-
able, the argument of superior risk-spreading capacity becomes
less often applicable since both groups (that to which plaintiff
belongs and that to which defendant belongs) will have a capacity
for efficient risk-spreading. Moreover, development of new types
of insurance is influenced by the legal rules placing responsibility
on one or another class of persons; it seems likely that insurance
will become available to any group on which responsibility for
bearing a risk is placed.
The theme of placing legal responsibility upon the superior
risk bearer, if carried to its ultimate implications, would stand
in opposition to the shifting of losses through tort law. Were
people not inspired to obstruction or co-operation depending on
their moral judgments, a plan of accident insurance carried by
each person to cover injuries to himself would be more efficient
than either a plan of liability insurance or a plan of accident in-
surance carried by each person to cover injuries his activities
cause to others. The plan of insurancecovering one's own injuries
would avoid many difficult problems of legal cause inherent in
any plan fixing responsibility for injuries to others. If each per-
son were carrying insurance for injuries to others rather than to
himself, hearings would sometimes be required to determine
whether an injury arose out of motoring, and if so, whose motor-
ing, just as hearings have been required in workmen's-compensa-
tion cases to determine whether injury arose out of employment,
and if so, for whom. For example, cases of successive impacts
separated by seconds, days, or years would require hearings for
the determinationof responsibility for the end result of disability;
there might be doubts as to what impacts contributed to the dis-
ability, and in what degrees, and doubts as to what motorists con-
tributed to each impact. Of course, a plan of accident insurance
for one's own protection as a possible victim would have to be
compulsory to be effective. The average man probably would
7 See pp. 406-07 supra.

gamble that the accidents would happen only to somebody else,

and would save the insurance premium to buy a fancier car for
himself. But the necessity of compulsion is also true of any plan
requiring provision for injuries to others. If the system of insur-
ance for one's own protection seems less feasible than the system
of insurance for the protection of others harmed by one's activi-
ties, perhaps the reason lies more in what persons affected con-
sider to be fair than in the efficiency of one system over the other
apart from such moral attitudes.
Even more efficient, but for the possible effect of moral atti-
tudes, would be a plan under which all injuries, of whatever na-
ture and from whatever cause, were compensated out of a single
fund, to which all contributed- either a government fund sup-
ported by taxes, or an insurance fund supported by compulsory
contributions from all. Yet it is plain that we are nowhere near to
acceptance of that form of loss distribution in our law. The sac-
rifice of other interests which it would entail is one we are not
preparedto make. Two such interests are noted here.
Responsibility for injuries caused to others serves as a selector
of socially useful activities and methods. Without such responsi-
bility, some governmental authority would be needed to decide,
for example, whether one who is under contract to remove a ledge
may use one form of explosive or another, or must not blast at
all, and whether manufacturersof apricot-puffcookies must incur
the expense of electronic devices to catch and remove each cookie
which contains wire.74 Though economic incentive is only one
factor among many in the complex motivation for accident pre-
vention, it can be an important one. In the absence of responsi-
bility for injuries caused, it tends to work against safety. If all
injuries from whatever cause were compensated out of a general
fund, the cost to blasters and cookie manufacturers in the form
of their respective shares of the increased assessment on all con-
tributors to the fund would be less than the cost to them of mak-
ing ledge removals and cookies safe.
On the other hand, if liability is imposed upon one whose activi-
ties cause injury to another, a control factor is present in the form
of an increase in price of the product to recoup the cost of com-
pensating for the injuries. This factor is present when liability
is based either on negligence or on conditional fault. For the
7 See Bowman Biscuit Co. v. Hines, I5I Tex. 370, 251 S.W.2d I53 (I952) (wire
found in cookie).
440 HARVARD LAW REVIEW [Vol. 72
individual, this control factor serves as an incentive to avoid

hazards to others - an incentive which, however, is somewhat
diluted by liability insurance. For society, this control factor
serves as a selector separating from socially useful activities,
which can pay their way even with this added cost, the socially
undesirableactivities which cause more harm than they are worth.
It serves to improve the opportunity for free and rational choice,
both for the individual and for the group, in matters concerning
the worth of products and activities.
The antisafety factor inherent in a social order based on the
compensationof all losses from a single fund, with no effort to fix
responsibility according to causes, would produce much more
serious consequences than the relatively mild impact on highway
safety which one may envision as a consequence of shifting from
a rule of responsibility for negligently inflicted injuries to a rule of
responsibility for all injuries that one's motoring causes. The fact
that insurance against liability is permissible tends to reduce the
incentive to be careful, especially if those with bad safety records
are not charged higher insurance rates. But from the community
point of view, there is still inherent in the rule of responsibility
(either in the form based on negligence or in the form based on
causation regardless of negligence) a tendency toward discour-
agement of unduly risky activities, though probably it is less
forceful in the latter form than when the negligent driver pays
more, either in tort judgments or insurance premiums, than the
careful driver. While it is not likely under any system that this
inherent control factor would lead to such high-cost compulsory
insurance that few could afford to drive, the rise in the cost of
motoring with each rise in harm caused would at least keep
society informed of a real cost of motoring in the toll of accidental
injuries. To some extent, we are so informed today by liability
insurance rates; probably we would be even better informed
under a broader rule imposing responsibility for all injuries one's
motoring causes, regardless of negligence. Moreover, such infor-
mation on the real cost of motoring helps to provide a rational
basis for decisions as to what resources should be committed to
developing safer means of transportation than the popular com-
bination of power in the vehicle and impatience and bravado in
the driver.
A second important interest which would be sacrificed by the
adoption of a plan of compensating all losses from a single fund
1959] CONDITIONAL FAULT 44I
is that of fairness. Under currently prevailing views, equal pay-

ments per capita would be regarded as grossly unfair to careful
persons who cause relatively few of the losses and risks. Few
would argue that one is morally obligated to carry accident in-
surance covering harms to himself. Such sense of obligation as
has developed with respect to life insurance is no exception; the
risk of harm against which it protects is not harm to oneself, but
rather to one's dependents. The moral obligation is one between
a person and his dependents, not between him and others whose
activities might cause his death.
There are sensible reasons for placing upon the blaster, rather
than upon everyone as a potential blasting victim, the moral and
legal responsibility for making advance provision against the
hazard of nonnegligent blasting accidents. One reason for prefer-
ring this method is that it serves aims of deterrence of undesirably
risky conduct and encourages safety precautions. A second and
perhaps more important reason is that it better spreads the loss
in proportion to benefits realized. That is, it effects distribution,
according to the principle of unjust enrichment, among those who
realize benefits from the blasting. It would be difficult and per-
haps impossible to find a basis for a class definition which would
include the victim and would also conimend itself to one's sense
of fairness and reason as a definition of a group which ought to
bear the loss. On the other hand, one aspect of such a class defini-
tion which includes the blaster can be found readily in the unjust
enrichmentnotion - the notion that those who benefit by receiv-
ing the products of blasting activities ought to bear the losses if
they can be distributed at a reasonablecost. They should pay, as
part of the price of construction, the equivalent of an insurance
premium or contribution to a fund to cover risks from blasting
accidents. The accent is not upon efficiency but upon fairness of
distribution.
The influence of risk-spreading capacity in tort decisions is
primarily in the role of what might be called an enabling or rebut-
tal factor. It removes a barrier which might have prevented the
imposition of liability based primarily on other grounds. This role
is clearly seen, for example, in the impact of liability insurance on
charitable immunity.75 Such an immunity is a barrier to normal
75 See, e.g., Pierce v. Yakima Valley Memorial Hosp. Ass'n, 43 Wash. 2d i62,
260 P.2d 765 (I953).
liability for fault, a barrier based to a significant degree on the

financial weakness of a defendant charity. It was argued by
plaintiffs that a charitable hospital, like one operated for profit,
should be liable for harm resulting from negligence in its opera-
tions. Defendants answered that such tort liability would be dis-
astrous to many socially desirable charities. This answer is now
met with the plaintiffs' rebuttal that the threat of tort liability
need not be a disaster to any charity, since adequate liability in-
surance is available; the governingboard of the charity can substi-
tute the ascertainable costs of insurance premiums for the risk of
ruinous judgments. Liability insurance does not, of course, elimi-
nate or even reduce the cost of compensatingvictims; in fact it in-
volves an increase in such costs due to administrativeexpenses and
profits to the insurers, which more than offset savings from effi-
ciency in handling claims. But liability insurance can eliminate or
reduce the threat of tort judgments disastrous to the individual
charity. With the cost thus made relatively certain for the indivi-
dual charity, would it not be a narrowconception of its charitable
purpose to exclude from its beneficence the victims of the care-
lessness of its own servants? In this instance, it is quite clear that
risk-spreadingcapacity, in the form of availability of liability in-
surance, is used to bring th'eliability of charities into line with the
theme of liability based on fault. Similarly, though less obviously,
risk-spreadingcapacity operates in such an enabling role in cases
of conditional fault. In many jurisdictions, the small retail grocer
is liable to the consumer of canned foods for harm resulting from
unfitness of the food.76 The prima facie ground of liability is
difficult to grasp because of the difficulty of finding material dis-
tinctions between selling food and selling some other product
which, if defective, is likely to cause harm to persons, or between
injuries from the food and injuries from the container. Perhaps
products taken internally were thought to involve special hazards,
though this ground of distinction is being eroded by cases of lia-
bility for products used externally.77 To the extent that special
hazards are recognized in these cases, there is an analogy to the
liability of blasters. In any event, it seems erroneous to conclude
76
2 HARPER & JAMES, TORTS I599-60I (1956); PROSSER, TORTS 508-IO (2d ed.
I955).
" See, e.g., Graham v. Bottenfield's, Inc., I76 Kan. 68, 269 P.2d 413 (I954)
(hair dye).
that the grocer's capacity for risk spreading is the basis of his
liability; such liability is not imposed on retailers of other prod-
ucts though a similar and often superior capacity for risk spread-
ing exists. On the other hand, risk-spreading capacity (in the
form of the availability to the grocer of liability insurance) has
been used, and appropriatelyso it would seem, as a rebuttal to the
argument that such liability will be disastrous to many small
grocers.78The suggestion that the appropriateuse of risk-spread-
ing capacity is in the role of an enabling factor is not to disparage
its importance, however. The argument of threatened financial
ruin has been influential,79and the effective answer is likewise
influential.
If, as is here suggested, this enabling or rebuttal role is and
should be the role of risk-spreading capacity as a factor in tort
law, then sound prediction and sound development of the scope
of liability for conditional fault rests less on comparison of the
relative capacities of plaintiff and defendant, or classes including
them, to insure or otherwise spread the risk, than upon identifying
other grounds for liability in such cases as blasting and the sale
of food,
V. CONCLUSION
The realization that community notions of individual blame-

worthinessunderliethe legal discriminationsbetween, for example,
prudent blasting and prudent motoring, should help us as ob-
servers to understand the resulting legal rules, as lawyers to pre-
dict them, and as judges and legislators to develop them wisely.
78 If one is deeply concerned over the nowadays rather hypothetical small grocer,
who seems to be seldom sued, but who yet might be seriously inconvenienced
by initial liability . . . it may be observed that . . . he doubtless can and
often does have liability insurance, which he may charge as a normal expense
of the business and recoup through adding to the price of his goods, just as he
does with other business hazards like the negligence of his servant ....
.Garwood, J., in Bowman Biscuit Co. v. Hines, 2I Tex. Sup. Ct. Rep. 273, 280
(1952). On rehearing, a changed vote caused Garwood's opinion to become a
dissent, and the passage quoted above was deleted from the opinion as it appears
in the official reporter and in the Southwestern Reporter. 15I Tex. 370, 373, 25I
S.W.2d 153 (1952). Has the availabilityof liability insurancehad more influenceon
doctrine than its rare mention in opinions would indicate?
7n See. e.g., Beatty v. Central Iowa Ry., 58 Iowa 242, 247-48, I2 N.W. 332,
334 (I882); H. R. Moch Co. v. Rensselaer Water Co., 247 N.Y. i6o, i65-66, I59
N.E. 896, 897-98 (I928); Ryan v. New York Cent, RR., 35 NXY, 2I0, 2I6 (i866),
In general, this realization is likely to lead to less fear of hearty

acceptance of decisions like those in the blasting cases and in
Vincent, and to carefully considered extensions of liability based
on conditional fault; the basis in conditional fault offers some
reassuranceagainst the specter of runaway social engineeringwith
ill-consideredemphasis on risk-spreadingcapacity. Thus, modern
decisions imposing liability for nonnegligent, risky conduct
appear not as developments threatening to supplant blameworthi-
ness as the theme of liability in tort, but rather as a trend con-
sistent with that theme and with its application in relation to the
somewhat nicer and perhaps more debatable moral discrimina-
tions of conditional fault. In this perspective, these decisions are
not founded on a principle of searching for the most efficient
risk spreaderswithout regard to standards of individual responsi-
bility; they are not founded on a principle which must find its
maturity in the substitution of risk-spreadingcapacity for blame-
worthiness, or the substitution of social responsibility for indivi-
dual responsibility as the theme of tort liability. Rather, concepts
of individual responsibility, with moral content, remain vital to an
understandingof tort decisions and trends. Such concepts under-
lie liability for nonnegligent risky conduct as well as liability in
the more familiar areas of negligence and intentional torts.

Conditional Fault of The Law of Tort

Uploaded by

Document Information

Original Description:

Original Title

Copyright

Available Formats

Share this document

Share or Embed Document

Sharing Options

Did you find this document useful?

Is this content inappropriate?

Copyright:

Available Formats

Conditional Fault of The Law of Tort

Uploaded by

Copyright:

Available Formats

Conditional Fault in the Law of Torts

Author(s): Robert E. Keeton

HARVARD LAW REVIEW I

CONDITIONALFAULT IN THE LAW OF TORTS

I. FAULT AS A BASIS FOR Loss SHIFTING

T HE usual startingpoint for explainingbases of liability in

*Professor of Law, Harvard Law School. B.B.A., University of Texas, 1940,

law, fault has been considered the one generally acceptable

1An interesting range of opinions is presented in (i) Wigmore, Responsibility

negligence was at hand, there was no such concept as an unavoid-

trustful of the capacity of fact-finders to make the finer dis-

II. MODERN ADVOCACY OF RISK-SPREADING CAPACITY

In recent years, it has been urged with increasing vigor that a

A. Various Forms of the Argument

different overtones, but they are in fact variations on a single

one's sense of fairness by imposing heavier burdens on those

B. Does This Theory Explain Recent Developments?

submit that the areas of strict liability which have developed do

capacity in these developments, Professor Seavey joins in viewing them as con-

enough for our purposes that a substantial body of precedent for

A. Gain and Loss

Is it possible that the concept of risk-spreading capacity has similar impact on

question-begging construction of a fictional rental price in the form of an agree-

mination of legal liability, it would be proved at trial with less

B. What CircumstancesAre To Be Comparedin Determining

"Enrichment" implies change. It implies a comparison between

assume that this privilege was not included, the determinationof

Their experience is suggestive, first, of recognition that enrichment of some land-

Illustrations could be multiplied, but these should serve to

D. Is Use or Intended Impact Essential

might be made between using another's person or causing an in-

Of course, it is a commonplace observation that two fact situa-

erty and increasing risks to his person or property may be useful

35 But see EHRENZWEIG,NEGLIGENCEWITHOUT FAULT 56-57 (I95I) (observing

this is true of both shipowners and dockowners; risk-spreading

B. The Attitude TowardMotoring

38 Cf. Momus, TORTS 44 (1953).

In any event, whatever may be the complex of factors that

workers. Thus, there is one person, the employer, upon whom

C. How Much Compensation?

43Though the United States is not a signatory of the convention, it adheres

of discharging the oil without provision for compensation, either

ing the three cases (involving a blaster, a shipowner, and an em-

E. Blameworthiness- Three Classes of Conduct

tainty of losses to others, without making reasonable provision

50 Cf. Stallybrass, Dangerous Things and the Non-Natural User of Land, 3

such a case of conditional privilege. Cases of so-called strict

at least from factors other than a substandard capacity. Speed is an illustration.

avoid accidents is expressed in the legal rules regarding "physical" deficiencies

absent some agreement as to responsibility for the risk of such

gested that an injunction would be granted against a judgment-

F. The Moral Element in ConditionalFault

negligent, but is "unreasonable," absent satisfactory provision for compensation,

by the terms "individual fault," "legal fault," and "social fault."

BCf. KEIGWN, CASES ON TORTS I9 (3d ed. I929):

not adequately explained by the suggestion that the conduct of

moral standards develop, either more or less conducive to loss

G. Risk-Spreading Capacity as an Element in Conditional Fault

decision except in a sense comparable to that in which a pair of

7 See pp. 406-07 supra.

gamble that the accidents would happen only to somebody else,

individual, this control factor serves as an incentive to avoid