Philosophical Review

A Theory of Strict Liability by Richard A. Epstein

Review by: Jules L. Coleman
The Philosophical Review, Vol. 92, No. 4 (Oct., 1983), pp. 613-617
Published by: Duke University Press on behalf of Philosophical Review
Stable URL: http://www.jstor.org/stable/2184887 .
Accessed: 04/12/2014 10:58
Your use of the JSTOR archive indicates your acceptance of the Terms & Conditions of Use, available at .
http://www.jstor.org/page/info/about/policies/terms.jsp

.
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.

Duke University Press and Philosophical Review are collaborating with JSTOR to digitize, preserve and extend
access to The Philosophical Review.

http://www.jstor.org

This content downloaded from 158.251.6.77 on Thu, 4 Dec 2014 10:58:57 AM

All use subject to JSTOR Terms and Conditions

BOOK REVIEWS
belief that those complexities are so much excess baggage which have
nothingto do withthe relationbetween fairnessand justice and the search
for "basic and generallyacceptable" premises.
JOSHUA T. RABINOWITZ

University
ofIllinois-Urbana-Champaign

The PhilosophicalReview,XCII, No. 4 (October 1983)

A THEORY OF STRICT LIABILITY. By RICHARD A.
cisco, The Cato Institute,1980. Pp. xv, 141.

EPSTEIN.

San Fran-

The centralquestion before each court tryinga case in tortsis whether

the costsof the harm sufferedbythe plaintiff(or victim)should remainhis
responsibility-lie where theyhave fallen-or be shiftedto the defendant
(who is usually but not always the injurer). The rules that specifythe
conditionswhichmustbe satisfiedin order fora loss to be shiftedfromthe
individualwho initiallysuffersit to someone else are called "liabilityrules."
Tort law consistsof a set of liabilityrules; torttheoryconsistsof effortsto
explain and (if possible)justifyexistingliabilityrules.
Liabilityrules are basicallyof two sorts: faultand strict.In strictliability
the loss may be shiftedto the defendant even if the loss is not his fault.In
faultliability,a loss may be shiftedto the defendant only if it is his fault.
Effortsto explain existingtortlaw fallinto two general categories:those
which purport to deduce the structureand substance of tort law from
moral principles; and those which explain the law of tortsas rooted in
economic theory.Richard Epstein'sA TheoryofStrictLiabilityis the leading
contemporarystatementof a moral account of the law of torts.
Until recently,the dominant strategyfor defending tortlaw on moral
grounds relied on the role of faultin determiningliability.Because strict
liabilityimposes sanctionsin the absence of culpability,the prevailingview
was that it was not morallydefensible. In other words,a defendantcould
be liable under strictliabilityeven if he acted under what we would normallythinkof as exculpatingconditions.Fault liability,however,requiresa
measure of wrongfulnesson the defendant's behalf,and, in the received
view, liabilityis justly imposed upon the defendant to penalize his fault.
The received view could not be sustained, however,since the standard
of faultinvolvedin tortlaw does not coincide withthe conditionsof moral
fault.An individualis at faultin tortsprovided his conduct fallsbelow that
of a reasonable man of ordinaryprudence. Generally,it does not matter
whether the defendant could have done what a reasonable man would
613

This content downloaded from 158.251.6.77 on Thu, 4 Dec 2014 10:58:57 AM

All use subject to JSTOR Terms and Conditions

BOOK REVIEWS
have. While fallingshortof a standardbeyond one's grasp need not marka
moral defect,it sufficesin tortsto ascribe faultand to impose liabilityon
the basis of fault. If the charge against strictliabilityis thatit may impose
liabilityin the absence of culpability,and that,on those grounds,it penalizes unjustly,the same could be said of fault liability.The received view
had to be abandoned.
Once the traditionalview had been rejected,effortsto defend existing
tortlaw on moral grounds gave way to systematicattemptsto explain the
law in economic terms.In the economic theory,faultis simplythe failure
to take the precautions necessary to avoid an economically inefficient
harm. A harm is inefficientwheneverthe costsof preventionare less than
the costs of the harm discounted by the probabilityof its occurrence.
Because faultrequires inefficiency,
not culpability,an individualmaybe at
faultin this sense regardless of his capacityto do otherwise.
Epstein's book must be understood against this background of a failed
moral account of fault liabilityand an increasinglyaccepted economic
account of it. The central claim of the book is that the only hope for a
moral theoryof tortsinvolvesdeemphasizing fault,which the economists
have seized upon, in favor of a new interpretationof strictliability.Epstein'sstrategyis the following:both faultand strictliabilityrequire thatthe
injurercause the victim'sharm. Fault liabilityimposes the additional constrainton recoverythat the injurer be at fault in causing the harm. Previously the notion of fault was thought to carry the moral weight.That
view is mistaken.For Epstein the moral weightis to be borne by the causal
requirementinstead. The fact that one partycaused another harm suffices,in Epstein's view, tojustifyimposingprimafacie liabilityon the injurer. Because the causal requirement is common to both strictand fault
liability,both can be defended on moral grounds. The role of faultin the
defense of tortliabilityis greatlyreduced, if not entirelyeliminated;thus,
the theoryof strictliability.
Epstein's task then is to show: first,given the extraordinaryskepticism
concerningthe role of causal language in legal discourse,thatascriptions
of causal responsibilityare neither trivialon the one hand nor simply
policyjudgments on the other; next, that causing another harm is necessary to impose liabilityjustly, and sufficientto establishat least the prima
facie case; and finally,thatcertaindefenses themselvesbased on causation
as well as on other more traditionalgrounds, for example, assumptionof
risk,can defeat the primafacie case.
The current fashion in legal theoryis firstto distinguishbetween two
senses of causation-"but-for" (or cause-in-fact)and "proximatecause"then to deny that either could justifyliability(in tortsor in crimes). One
event is a but-forcause of another if the antecedent event is a necessary
614

This content downloaded from 158.251.6.77 on Thu, 4 Dec 2014 10:58:57 AM

All use subject to JSTOR Terms and Conditions

BOOK REVIEWS
condition of the subsequent event's occurrence. One event is a proximate
cause of anotherif,giventhe purposes of the inquiry,the antecedentevent
is responsible for the subsequent one. But-forcausation is too weak to
justifyimposing liability.In the law, a judgment of proximatecausation is
simplythe assertionthatresponsibilityor liabilityis warranted.Proximate
causation is thereforetoo strongto justifyimposing liability.
Epstein's task is to develop an account of causation that is both strong
and weak enough to justifyimposing liabilityon the basis of it. Epstein
does not actuallyproduce a theoryof causation. Instead, he outlines and
develops in some detail what he calls "causal paradigms." In effect,these
are instances in which ordinary individuals experience no difficultyin
employingcausal language both meaningfullyand as a basis of responsibility.For example, sentenceslike "A frightenedB" and "A forcedB" asserta
causal connection between A and B, and theirtruthmayjustifyimposing
liabilityupon A for B's harms.
The problem withEpstein's causal paradigms of frightand compulsion,
forexample, is thattheyare not normativelyneutral.It is not the factthat
A caused B harm thatjustifiesimposingliabilityupon A; it is the factthat
A frightenedor compelled B thatjustifiesliability.The case forliabilityis
built into the claim that one individual frightenedanother or that one
party compelled another. Both involve wrongfulways of causing harm
either by violatingthe other's autonomy or by causing unwarrantedpain
and anguish.
The real taskthen is to show thatcausation byitselfis both necessaryfor
liabilityand sufficientto justifyat least a primafacie case. This requires a
moral principle that imparts moral significanceupon every instance in
which A harms B-innocently as well as wrongfully.For if only wrongful
instancesof harmjustifyliability,then Epstein would have failed to provide a defense of strict liability,that is, liability without regard to
culpability.
Epstein's basic argument is this: in order that liabilityfor a loss be imposed justly,the individual mustbe responsibleforthe loss. An individual
is responsibleforan event if and onlyif he caused it. So ifA harmsB, A is
responsibleforB's loss, and ifA is responsibleforB's loss, thenliabilityfor
it isjustly imposed on him. Epstein writes:"Indeed, formost persons,the
difficultquestion is often not whether these causal assertionscreate the
presumption [of liability],but whether there are in fact any means to
distinguishbetweencausation and responsibility,
so close is the connection
between what a man does and what he is answerable for" (p. 25).
To harm an individual is to interferewith a legitimateinterestof his.
Not everylegitimateinterestis protectedby a right.There are numerous
waysin whichan individual mightharm another,forexample, in business
615

This content downloaded from 158.251.6.77 on Thu, 4 Dec 2014 10:58:57 AM

All use subject to JSTOR Terms and Conditions

BOOK REVIEWS
and in love, withouthis so doing raising the question of liability.Because
there is no general right not to be harmed, there is no general rightto
compensation whenever one suffersharm, and no reason thereforeto
suppose thatcausing harm is sufficientto establisheven a primafacie case
for liability.
The argumentfromresponsibilityalso failsbecause it relieson the claim
that because causation is necessary for responsibilityit is necessary for
liability.That Epstein holds such a view is evidenced by his solutionto the
good Samaritan problem. The problem is whetheran individualwho fails
to come to the aid of another can be liable in damages to thatperson for
harms the "good Samaritan" could have prevented. In general, the common law fails to recognize the duty to be a good Samaritan. Epstein believes thatwhereas the economic theorycannot explain the court'sposture
towardthe good Samaritan,the theoryof strictliabilitycan. The reason: to
failto come to another'said is to failto performan act. Failure to act is not
itselfan action, nor thereforeis it an event. Only events can cause other
events.By failingto come to the aid of othersthe good Samaritandoes not
cause harm: he simplydoes not preventit. Since he does not cause harm,
he fails to satisfythe causal requirement,and cannot be held liable.
There are two things wrong with this argument. First, if Epstein is
correctthat causation is necessary for liabilitybecause it is necessaryfor
then the theoryof strictliabilitywillbe unable to account for
responsibility,
a wide range of cases in which liabilityis normallyimposed. Suppose it
snows and I failto clear the snow fromthe steps leading to myhouse. The
mailman deliversthe mail to my house and upon leaving slips and injures
himself.Normally,I would be held liable for the mailman's injury.I am
liable, however,not because I harmed the mailman, but because I could
have and should have preventedhis injury.My failureto shovelis not itself
an action. Because it is not an action itcannot cause anything.If the theory
of strictliabilityrelieson causation as a necessaryconditionof bothresponsibilityand just liability,itwillfailto account formuch of negligencelaw in
which liabilityis imposed for omissions.
The lesson of thisexample, however,is thatthe theoryof strictliabilityis
wrong to insist upon causation as a condition of responsibility.Even if
liabilityrequired responsibility,it would not require causation. An individual may be responsible and thus liable forwhat he failsto do as well as
is not restrictedto
forwhat he does. The range of individualresponsibility
the events that an individual causes, and though the events an individual
causes are always in the sense of individual authorship his responsibility,
theymay or may not be thingshe is morallyresponsible for.
One can constructanother argument for the theoryof strictliability
based on Epstein'sconception of the principleof correctivejustice.Epstein
616

This content downloaded from 158.251.6.77 on Thu, 4 Dec 2014 10:58:57 AM

All use subject to JSTOR Terms and Conditions

BOOK REVIEWS
does not explicitlymake this argument in the book. Still there is some
textualsupport forit and in his mostrecentessays he has taken to emphasizing the role of correctivejustice in the theoryof strictliability.The
argumentappears to be this:ifA harmsB, thereis an imbalance,a disequilibrium,between them. Following Aristotle,this is an occasion to rectify
the imbalance, to reestablishthe equilibrium.The basis of liability,in this
forcausing harm as much as it is the
view,is not the injurer'sresponsibility
correlativeof the victim'srightto have his losses rectified.
Not everyloss, however,is a departure froma distributionthatought to
be preserved; and not everywayof departingfroma distributionis wrongful. Therefore, not everyloss is an unjustifiableone. Contraryto Epstein,
onlywrongfulor unjustifiablegains and losses in Aristotle'stheorymustas
a matter of justice be annulled. Ironically, the appeal to principles of
correctivejustice appears to provide a moral account of the faultprinciple,
notof the principleof strictliability.A loss caused bythe faultof anotheris,
in a suitablynarrow sense, a wrongfulone. Those losses are distortions
which, according to the principle of correctivejustice, ought to be rectified.The obligation correlativeto the right to have one's unjustifiable
losses rectifiedfalls on the injurer who is at fault.Thus, the principleof
correctivejustice provides the beginningsof a moral defense of fault,not
of strictliability.
A Theoryof StrictLiabilityis the only systematicattemptto deduce the
basic structureof tortlaw and the contentof tortrules fromprinciplesof
morality.It is a well-writtenand carefullyargued essay that focuses on
importantquestions in the theories of responsibilityand justice. It is especially recommended for legal philosophers who have devoted far too
much attentionto the criminallaw almost to the exclusion of tortlaw. A
TheoryofStrictLiabilitywill give the reader an idea ofjust how philosophically fertilethe private law can be.
JULES

L.

COLEMAN

of Wisconsin-Milwaukee
University

The PhilosophicalReview,XCII, No. 4 (October 1983)

UTILITARIANISM AND CO-OPERATION. By DONALD H. REGAN. New
York, The Clarendon Press, Oxford UniversityPress, 1980. Pp. xvi,
279.
willwant to study
Anyone interestedin the battleamong utilitarianisms
thisbook. Regan presentsa tightlywoven argument for the positionthat
617

This content downloaded from 158.251.6.77 on Thu, 4 Dec 2014 10:58:57 AM

All use subject to JSTOR Terms and Conditions