University of Chicago Law School

Chicago Unbound

Journal Articles Faculty Scholarship

1996

The Tort/Crime Distinction: A Generation Later

Richard A. Epstein

Follow this and additional works at: https://chicagounbound.uchicago.edu/journal_articles

Part of the Law Commons

Recommended Citation
Richard A. Epstein, "The Tort/Crime Distinction: A Generation Later," 76 Boston University Law Review 1
(1996).

This Article is brought to you for free and open access by the Faculty Scholarship at Chicago Unbound. It has been
accepted for inclusion in Journal Articles by an authorized administrator of Chicago Unbound. For more
information, please contact unbound@law.uchicago.edu.
 SYMPOSIUM
THE TORTICRIME DISTINCTION:
A GENERATION LATER

RICHARD A. EPSTEIN*

I. A CHANGE IN WORLD VIEW

This Paper marks the second occasion-both in Boston, but a genera-

tion apart-that I have grappled with the tort/crime distinction as it has
developed first at common law and today in the mass of statutes that
regulate virtually every area of human endeavor. When I first wrote
about this subject in my paper Crime and Tort: Old Wine in Old Bottles,'
my philosophical outlook was both deontological and libertarian. I un-
derstand the first of these terms less well today than I thought I did then,
but its basic import is that certain inherent qualities of actions and mental
states brand them as wrongful, regardless of the consequences they bring
about. As this one sentence summary indicates, the deontological ap-
proach typically defines its field negatively. In general it is an-
ticonsequentialist, that is, opposed to any theory that looks to the aggre-
gate consequences of certain actions to determine their moral worth.2
More specifically, this approach is also antiutilitarian, that is, opposed to
any theory where the consequences that count go to the gain or loss (or
perhaps only the pleasure and pain) of the various individuals governed
by the legal rule.
The libertarian side of the theory claims that the proper role of govern-
ment is limited to restraining the use of force and fraud in the conduct of
human affairs, thus preserving the maximum scope for freedom of action
for all persons governed by the legal order. From these premises two
conclusions seemed to follow. The criminal law should concentrate its
resources on intentional violations of the person and intentional depriva-

* James Parker Hall Distinguished Service Professor of Law, University of

Chicago. I should like to thank Dan Kahan for his very helpful comments on an
earlier draft of this Paper.
1 Richard A. Epstein, Crime and Tort. Old Wine in New Bottles, in ASSESSING THE
CRIMINAL: RESTITUTION, RETRIBUTION, AND THE LEGAL PROCESS 231 (Randy E.
Barnett & John Hagel III eds., 1977).
2 See, e.g., JOHN RAWLS, A THEORY OF JUSTICE 30 (1971) (characterizing a deon-
tological theory, i.e., "justice as fairness," as "one that either does not specify the
good independently from the right, or does not interpret the right as maximizing the
good"); see also Heidi M. Hurd, What in the World is Wrong?, 1994 J. CONTEMP.
LEGAL ISSUES 157, 160-62 (summarizing deontological theory and distinguishing it
from consequentialism).

HeinOnline -- 76 B. U. L. Rev. 1 1996

 BOSTON UNIVERSITY LAW REVIEW [Vol. 76:1

tions of property. The tort law should seek to protect these same inter-
ests against all other external deprivations, including many that are
strictly accidental, by awarding money damages to the victim from the
wrongdoer. In some cases of imminent danger injunctive relief could be
appropriate as well, although the issue of prospective3 relief injects a mass
of complications to which I shall briefly refer later.
Over the years, I have remained quite content with the basic substan-
tive orientation of the tort/crime distinction, but have become more un-
easy about its intellectual foundations. My ideal government is not quite
as small as the above theory suggests, but it is still much smaller than the
massive government in place today. The shifts in basic intellectual out-
look are perhaps more substantial. On the political theory side, a robust
libertarian theory is vulnerable to the charge that it fails to account for
the coordination, public goods, and holdout problems4 of such great con-
cern in the organization of social affairs. Thus it is not sufficient to as-
sume that the only forms of conduct accompanied by undesirable social
consequences are those involving the use of force or fraud. The libertari-
an posture seeks to minimize the use of force and fraud and their conse-
quences. However, a more comprehensive social statement seeks to max-
imize social welfare, embracing the libertarian prohibitions, but going
beyond them to allow certain forms of regulation and taxation to over-
come these otherwise intractable coordination problems. The set of
forced exactions that are totally barred under a libertarian approach have
a limited, but secure place in the larger system of social controls.
Similarly, on questions of method, I believe that the deontological ap-
proach is wrong insofar as it claims that its normative conclusions can be
denied only on pain of self-contradiction. Today many writers believe
that the protection of individual autonomy is not a primary goal of legal
rules, but that, to the contrary, any "natural" distribution of talents is
determined largely by luck and hence morally arbitrary.' Given this per-
spective, it follows that legal rules should introduce certain measures of
sharing across individuals, if not by forced labor, then by systems of taxa-
tion and regulation that redistribute the fruits of individual labor.' One
can argue against these views, but hardly on the ground that they are self-
contradictory, or even that they are morally suspect in their effort to raise

3 See infra Part IV, at 16.

4 See, e.g., RICHARD A. EPSTEIN, FORBIDDEN GROUNDS 12 (1992) (identifying
such problems as possible justifications for "government coercion or control"); RICH-
ARD S. POSNER, ECONOMIC ANALYSIS OF LAW 61-65 (4th ed. 1992) (addressing
holdout problems in connection with pollution abatement).
5 See RAWLS, supra note 2, at 100-08 (suggesting that "[t]he natural distribution [of
talents] is neither just nor unjust" and that such distributions are "simply natural
facts").
6 Id. at 274-84 (identifying the function of such measures in an ideal institutional
framework based on the principle of distributive justice).

HeinOnline -- 76 B. U. L. Rev. 2 1996

1996] A GENERATION LATER

the level of the least fortunate closer to the level enjoyed by those who
have a greater share of natural abilities and endowments. A defense of
the older regime of individual liberties and properties cannot rest on a
simple assertion that people have rights and that other individuals are not
allowed to do actions that violate those rights.7 One has to show why
any given configuration of rights is superior to its rival conceptions, an
undertaking that typically requires an appeal to consequences, less for
particular cases, and more for some overall assessment of how alternative
legal regimes play out in the long run. In a word, one has to become a
utilitarian of some stripe to justify rules in terms of the consequences they
bring about.'
A utilitarian approach is frequently attacked on the ground that the
critical outcomes needed to fuel the system are invariably indeterminate.
In many legal contexts, the choice of legal rules is hard to make, but the
inferences we can draw from a select set of difficult legal cases are far
from obvious. More concretely, it is only when individuals look to some
of the hardest questions within the legal system-whether we choose neg-
ligence or strict liability as the basic principle of tort law-that we find
both our moral instincts and our utilitarian calculations coming up short.
But those hard cases are not the ones we should chose initially to test the
moral limits of a legal rule. Rather the operative inquiry steps back a
long way and asks: Why, of all the possible tort rules that could organize
the system of tort law, do those two survive in constant tension with each
other? At this level it is not difficult to imagine the adverse consequences
of a rule requiring the victim of harm to pay damages equal to the harm
that he sustained to the party that inflicted the harm. This rule might be
ingeniously justified on the ground that it induces a high degree of victim
precaution. But don't bet on its social appeal. It takes little temerity to
say that no legal system will ever adopt so odd a position, given the mas-
sive insecurity generated by subsidizing individuals to maim or kill their
fellow human beings. Likewise a rule declaring that all promises are pre-
sumptively criminal acts would not sit well either: it is simply too costly
for the legal system to declare war on the mutual gains arising from ordi-
nary exchanges. Sometimes these presumptive gains are overridden by

7 See ROBERT NOZICK, ANARCHY, STATE, UTOPIA at ix (1974).

8 In principle, this statement could invite a long disquisition of what form of conse-
quentialism is appropriate. And even though philosophers are quite conscious of the
different versions of utilitarianism, I shall largely overlook them here. For a more
complete view of my approach, see Richard A. Epstein, The UtilitarianFoundations
of Natural Law, 12 HARV. J.L. & PUB. POL'Y 713, 718 (1989) (suggesting that "a
proper application of utilitarian principles generates many of the categorical conclu-
sions that [anti-consequentialist] natural rights thinkers first articulated and defend-
ed") and Richard A. Epstein, Postscript Subjective Utilitarianism,12 HARV. J.L. &
PUB. POL'Y 769 (1989).

HeinOnline -- 76 B. U. L. Rev. 3 1996

 BOSTON UNIVERSITY LAW REVIEW [Vol. 76:1

third-party losses, but those losses must be demonstrated by clear evi-

dence and not simply asserted.
Given the foregoing, my current view is that we should think about
legal rules in terms of the incentives they create for individual conduct, be
they for good or ill. Once we take this approach, it hardly matters where
we start with the creation of legal rights and responsibilities. The key
inquiry is not our current baseline, but whether we can imagine some
legal changes capable of generating social gains greater than their overall
costs. Thus, in an odd sense, it hardly matters whether the inquiry begins
with a Hobbesian view that natural liberty is simply the actions that indi-
viduals can get away with before they meet a greater rival force,9 or the
more congenial Lockean theory, premised on a view of natural rights
guaranteeing to all individuals a right to be free of aggression against
their persons and property.'" All individuals gain vastly from the security
of possession and exchange promised by the Lockean view of individual
rights, so that no matter what the starting point, the logic of the social
contract allows for movement in only one direction. Faced with an all-or-
nothing choice, individuals in the Hobbesian state would be prepared to
exchange their freedom to attack for blanket protection from attacks by
others." But those in a Lockean world will not be prepared to make the
same exchange in reverse. In the one case we move, and in the other we
stay put. And the end points are the same in both cases.
To be sure, both sets of exchanges are hypothetical in the sense that we
can never hope to observe actual contracts between the various individu-
als subject to a common rule. But in both cases we attribute the term
"social contract" to these mental inquiries because we seek the same re-
sult that voluntary contracts normally yield to their participants: mutual
gain judged from the ex ante perspective. 2 The object of the social con-
tract is to obtain the same result where voluntary exchanges are not pos-
sible. At that point the inquiry shifts to the form of government that will
best secure the (well-nigh) universally preferred Lockean system of
rights. Hobbes's unitary sovereign can be faulted for his evident totalitar-
ian dangers that, in turn, would drive us to more sophisticated govern-

9 THOMAS HOBBES, LEVIATHAN 86-96 (A.R. Walker ed., Cambridge Univ. Press
1935) (1651).
10 JOHN LOCKE, SECOND TREATISE OF GOVERNMENT 3-10 (Richard H. Cox ed.,
Harlan Davidson, Inc. 1982) (1689).
11 HOBBES, supra note 9, at 115-19.
12 The ex ante constraint is surely decisive because any system that imposes crimi-
nal sanctions on aggressors will not leave violators well off ex post. Indeed, one of the
greatest perils to any system of contract-voluntary or hypothetical-is the institu-
tionalization of a general principle of ex post regret, whereby those who do not like
the observed outcomes can undo the entire arrangement. When that regret contin-
gency is Imown in advance, the initial deal will not get off the ground because no one
will regard himself as a potential winner.

HeinOnline -- 76 B. U. L. Rev. 4 1996

1996] A GENERATION LATER

ment structures-complex voting systems, divided power, federalism, ju-

dicial review-in order to ensure that the state designed to preserve
individual rights and liberties does not rise up to consume them.

II. NEITHER CRIME NOR TORT

This set of introductory remarks takes us somewhat far afield. It has,

however, powerful implications for the proper relationship between tort
and criminal law, albeit in a way that is somewhat orthogonal to the stat-
ed purpose of this Symposium. The usual discussion of tort and crime
focuses on the differences between the two systems of enforcement and
attempts to articulate a principled justification for distinguishing the two.
I believe that this is an important inquiry, and one to which I will turn in a
moment. But in any comprehensive view of the subject, it is not the first
question on the table. Far more important is the question of whether
certain forms of conduct should be subject to any form of sanction-pri-
vate or public-at all. The more fundamental issue therefore concerns
the combined scope of the two sets, rather than questions of overlap and
coordination.
On this note it is critical to recognize that many of today's law enforce-
ment problems arise from the overall expansion of liability-both civil
and criminal-to criminalize types of conduct that had been unquestiona-
bly legal before the passage of new law. Inquiries that concentrate on the
role of mens rea in determining criminal liability or the reach of proxi-
mate causation in determining civil liability miss the central point: What
broad classes of conduct should be subject to either form of legal sanction
in the first place?
As mentioned earlier, the libertarian world view places powerful limi-
tations on the use of state power. A conceptual predicate of this view is
that many forms of conduct generate no adverse legal consequences re-
gardless of their public approval ratings. Decisions such as whom to mar-
ry, with whom to associate, or what to do with one's property and life lie
outside the scope of state interests.' 3 The legal system thus creates many
small islands of independence in which people are free to go their sepa-
rate ways so long as they do not trespass against the liberty and property
of their neighbors. The system is internally symmetrical with respect to
the rights it creates and the obligations it imposes. In consequence, the
libertarian view leaves a relatively narrow scope in which the state may
impose either tort or criminal responsibility.
The greatest resistance to government power comes from the contrac-
tion of the private realm and the corresponding expansion of the public
realm. Consider questions of land use. There has never been any doubt
that individuals who commit trespasses on the property of their neighbors

13 For discussion, see Richard A. Epstein, The Harm Principle-AndHow it Grew,

45 U. ToRoNro L.J. 369, 375-97 (1995).

HeinOnline -- 76 B. U. L. Rev. 5 1996

 BOSTON UNIVERSITY LAW REVIEW [Vol. 76:1

are subject to either civil or criminal liability. The same has surely been
true with respect to nuisances, such as the discharge of pollutants on land,
water, or air. Yet the potential liabilities accompanying the newer envi-
ronmental restrictions go far beyond those associated with traditional
wrongs such as nuisance or pollution. Today it is wrong to place a railway
tie on the side of a critical dune unless one first receives a permit from the
state,' 4 and it is a criminal offense to modify the habitat of any endan-
gered species located on private property.'" It hardly matters what dan-
gerous conditions are corrected or what collateral benefits are created.
Because there is rarely a particular harm to any one person, the set of
sanctions tends to be criminal in nature. But the core objection to this
form of criminal liability does not rest on the observation that these of-
fenses are strict liability offenses, although that is sometimes the case.
Nor does it rest on the possibility that individuals may not receive suffi-
cient notice that the law deems the designated conduct criminal. They
often receive an abundance of notice; public officials are always willing to
proclaim their commands from the mountain tops in order to satisfy the
most exacting due process requirements. And surely the statutes could
incorporate, explicitly or by implication, a narrow exception for cases of
abject necessity, which would seem pointless if the crime in question is
removing a rubber tire from a wetland under the jurisdiction of the Army
Corps of Engineers.
No, the objection to the criminal sanction quite simply comes from the
deep and passionate belief that government should not undertake these
exercises at control at all, or at the very least, that if it intrudes on the
lives of its individual citizens, then it should compensate them for the
losses it imposes.16 This position follows the argument on state control
outlined above.' 7 Start with a world in which disputes between neighbors
are regulated by the law of nuisance and ask whether the further regula-
tion works on balance to the joint improvement of the parties. Certain
rule adjustments surely make sense, which helps explain why privately
planned unit developments usually adopt covenants more extensive than
those required under the law of nuisance. Yet the legal restrictions men-
tioned above are more intrusive still, but their gains to affected parties
are small or nonexistent. As between the parties, we may describe the
transactions as Pareto Pessimal, meaning that both neighbors end up
worse off than before, without any enormous positive external gains to

14 Coastal Zone Management Act of 1972, 16 U.S.C. §§ 1451-1464 (1994).

15 Endangered Species Act of 1973 § 11, 16 U.S.C. § 1540(b)(1) (1994) (providing
for criminal sanctions of up to $50,000 and one year imprisonment for knowing viola-
tions).
16 See generally RICHARD A. EPSTEIN, TAKINGS: PRIVATE PROPERTY AND THE
POWER OF EMINENT DOMAIN 263-73 (1985) (addressing the applicability of the Tak-
ings Clause of the Fifth Amendment to, inter alia, land use regulations).
17 Supra notes 9-12 and accompanying text.

HeinOnline -- 76 B. U. L. Rev. 6 1996

1996] A GENERATION LATER

justify these restrictions. The protest therefore is directed in good utilita-

rian fashion at the scope of the regulation. The internal logic of criminal
law has nothing to do with the issue.
The same argument is equally applicable to other contexts as well. To-
day it is a crime to pay workers wages below some minimum established
by statute,' 8 or to discriminate in the hiring or promotion of employees, 19
or to use equipment in ways that do not comply with the mandates of the
Occupational Health and Safety Administration ("OSHA").2" In all of
these cases, the use of legal sanctions, criminal or tortious, are in my view
wholly inappropriate, even if the government were to give ample notice
of the restrictions imposed, and prosecuted only those persons who acted
in conscious disregard of the legal command. As with various forms of
environmental legislation, the argument against criminalization is quite
simply that government ought not, for substantive reasons, undertake to
regulate these activities at all. The case for government intervention is to
stabilize the social environment so that individuals can use their property
in productive ways and engage in various forms of productive labor.
Although restrictions on the use of aggression and force advance overall
social welfare, similar restrictions on building a home or running a busi-
ness have exactly the opposite effect. Such restrictions cut off the gains
from voluntary exchange but produce little, if anything, by way of exter-
nal benefits.
It is all too easy to claim that ordinary people do not have sufficient
information to be accurate judges of their own interests. Yet usually the
truth lies elsewhere. Few government officials can ever gather sufficient
reliable information to decide for other people which particular course of
action is best. Those who claim that markets do not create perfect incen-
tives for cooperative behavior likewise ignore the far greater distortions
created in the political arena when transfer payments are divorced from
productivity gains. To be sure, there are some possible exceptions to this
proposition, but the basic logic of the hypothetical contract dictates that
the first task is not to understand the overlap between crime and tort.
Rather, we must first commit ourselves to restricting the domain over
which both operate.

III. THE EXPANSION OF THE CRIMINAL LAW

Once we have made the decision to sanction these forms of conduct it
is easy to explain the inexorable expansion of the criminal law, even if we
18 Fair Labor Standards Act of 1938 § 6, 29 U.S.C. § 206 (1994) (establishing mini-
mum wage requirement).
1' 42 U.S.C. § 2000e-2(a) (1988 & Supp. V) (prohibiting employment decision
based on consideration of race, color, religion, sex, or national origin).
20 Occupational Safety and Health Act of 1970 § 17, 29 U.S.C. § 666 (1994) (pro-
viding for criminal sanctions for violating standards or orders promulgated under
OSHA).

HeinOnline -- 76 B. U. L. Rev. 7 1996

 BOSTON UNIVERSITY LAW REVIEW [Vol. 76:1

put aside the entire issue of drug offenses. 21 The first step in the argu-
ment is to show why traditional tort law will no longer do the job. Quite
simply, the answer is that there are no victims to whom a tort remedy
could be sensibly assigned. Consider once again the issues of land use
control and worker safety regulation. Turning first to land use regulation,
precisely who is hurt if an individual owner decides to turn over some soil
or build a stairway up the side of a hill? Clearly it is not the neighbors,
most of whom have engaged in similar conduct (before the restriction was
imposed) or who currently chafe under like restrictions. To be sure,
neighbors do not always get along in this or any other world, so that at
least some neighbors might well be inclined to report violators to public
officials in order to exact a fair measure of vengeance against them. But
those sorts of indirect unhappiness do not count as compensable or pro-
tectable interests under either the tort or the criminal law. So, too, with
harms to the "environment:" no private plaintiffs stand ready and avail-
able to police them. Consequently, it falls to the government by default
to fill the void said to exist. Further, in this case the government sues
only as the government, rather than in its capacity (analogous to the pri-
vate landowner) as steward or owner of public lands and waters. The
utter want of a definable, focused wrong thus propels these cases into a
large and unstructured public arena.
The second stage of the argument is more difficult. The government
can proceed in its own name to remedy the perceived wrong with either
civil fines or criminal actions. In principle it would seem that when tort
law normally remedies the relevant harm, civil fines should be the pre-
ferred public response. We utilize public officials to decrease enforce-
ment costs rather than to escalate conflicts with individual defendants.
To be sure, most regulatory statutes provide for civil penalties.
Yet enforcement rarely stops there. Instead, such statutes allow the
state-at its option-to pursue the criminal route, either by way of fine
or imprisonment. The expansion of power is often justified in part by the
fact that most of the prohibited acts are done intentionally, and hence fit
more easily in the criminal camp. We expand the scope of the actus reus
without any offsetting contraction of the relevant mens rea.
This "double whammy" is hardly defensible when the underlying con-
duct should not be subject to any sanctions at all. But once we take the
first step, the second comes easily. After all, ordinary criminal prosecu-
tions do not allow a defense for ignorance of law. It is quite enough to
intend to do the act that the law has declared harmful. The approach has
great merit as a means to dispose of false pleas of ignorance for conven-

21 On which I am tempted by a standard that eliminates all punishment for use, all
public aid for illegal drug users, and all defenses of diminished responsibility based on
drug use. See Richard A. Epstein, The Moral and Practical Dilemmas of an Under-
ground Economy, 103 YALu L.J. 2157, 2175-76 (1994).

HeinOnline -- 76 B. U. L. Rev. 8 1996

19961 A GENERATION LATER

tional, widely acknowledged crimes. Yet it holds little attraction in con-

nection with regulatory offenses made criminal by stipulation only.
Unfortunately, it is all too easy to pinpoint the motivation behind this
expansion of state power. The political discourse announces that filling in
a wetland constitutes the most pressing of social evils. We avoid explicit
comparisons to murder, rape, and theft. The political logic dictates a flex-
ible government response to meet the growing peril. Criminal sanctions
may entitle a defendant to some protections not available in civil cases: a
higher standard of proof and certain privileges against self incrimination
as the two most obvious. But, by the same token, the government comes
to wield two strategic weapons against a potential defendant: civil sanc-
tions when questions of proof are hard to resolve, and criminal sanctions
when evidence of a violation is stronger. It is very difficult for any public
official to resist a free grant of more police power. Likewise, it is equally
difficult for a legislator to deny the criminal sanction when it has been
supplied so many times before.
By degrees, we create a simple syllogism. Once we bring public offi-
cials into a case, criminal sanctions are possible. The more modest ration-
alization of efficient enforcement of diffuse public wrongs slips into the
background.
A similar analysis applies to various statutes for workplace safety. In
most instances, restrictions are no more desired by workers than they are
by management. The regulations in question usually derogate from free-
dom of contract, not provide remediation of frauds and deceit that em-
ployers sometimes practice on their workers. For example, the usual ap-
proach of OSHA is to insist that safety comes through precise compliance
with detailed and often absurd regulations about the conditions of the
premises.22 Many of these precautions are either counterproductive or
too expensive for what they achieve relative to simpler precautions that
workers could take in their day-to-day work.2" There is accordingly little
incentive on the part of employees to report violations to the govern-
ment. After all, what do they have to gain by the enforcement of a stat-
ute that makes their own work more difficult, and in some instances less
safe? Most reporting therefore is likely to come from disgruntled or fired
employees who (like the neighbors in land use cases) seize the opportuni-
ty to report violations of the law to wreak a fair measure of vengeance on
their (former) employers. The net effect is that OSHA enforcement typi-
cally requires unannounced inspections-trespasses now cloaked in the
privilege of the underlying administrative mandate. The newer genera-
tion of regulatory offenses also gravitates toward public enforcement.

22 For a recent anecdotal indictment of the follies of OSHA, see PHILIP K. How-
ARD, THE DEATH OF COMMON SENSE 12-15 (1994) (commenting on the scope and
specificity of OSHA regulations and the massive expenditures made by employers to
reach mandated compliance levels).
23 E.g., id.

HeinOnline -- 76 B. U. L. Rev. 9 1996

 BOSTON UNIVERSITY LAW REVIEW [Vol. 76:1
Once again state officials will be hard pressed to turn down the strategic
advantages that come from being able to choose between civil and crimi-
nal sanctions. Once again the political dynamic ensures a normatively
questionable expansion of public power.
All this is not to say that property damage and workplace injuries do
not happen. But in those cases, why deviate from traditional patterns of
private enforcement? The real workplace injuries are usually covered by
workers' compensation (which was consensual in its origins, and could
easily be made consensual today) and occasionally by the tort law, so for
tangible harms a private plaintiff will quickly seek redress for his wounds.
But for diffuse workplace safety issues, the absence of an identifiable
plaintiff means that the regulatory work, once unmoored from a require-
ment of actual injury, falls by necessity into the lap of the state. Once
again, the tort law will play second fiddle to the criminal law. The situa-
tion clearly moves in reverse as well: trim down the ambitious scope of
government regulation and these diffuse social harms would be the first
to disappear from the social plate. The state would concentrate on griev-
ances real victims pursue in response to real injuries. Cutting down on
both tort and crime therefore should be the first order of social business.

IV. CRIME AND TORT CABINED

Some actions still remain illegal even under standard libertarian theo-
ry, aided perhaps by a limited assist from the antitrust laws. How should
control of these wrongs be divided between tort and criminal law? In
principle, the utilitarian overlay24 looks to maximize some social welfare
function. The upshot is to use criminal and tort law to optimize the net
gain from productive activities, less the sum of the costs of harmful ones
and the costs of their prevention. The dialogue then switches to the fa-
miliar litany of optimal deterrence, which envisions harm from both
under- and over-enforcement of the law (criminal and tort) and hopes to
steer a path between the two.25 Questions of individual rights, so conge-
nial to the traditional deontological approach, are swept away by a set of
instrumental and technical concerns. The connection between the ordi-
nary expectations of citizens is washed away in a sea of jargon. The com-
prehensibility and legitimacy of the law, both criminal and civil, are sorely
tested. Or so the standard story goes.
Lest we despair, I believe that we can make the jargon intelligible. Just
as the techniques of social contract theory are equal to the task of defin-
ing individual rights, they can also explain why the traditional view of the

24 E.g., supra notes 5-8 and accompanying text.

25 See, e.g., Richard A. Posner, An Economic Theory of the Criminal Law, 85
COLUM. L. REv. 1193, 1205-14 (1985) (discussing optimal criminal penalties); Steven
Shavell, Deterrence and the Punishment of Attempts, 19 J. LEGAL STuD. 435, 448
(1990) (utilizing optimal deterrence theory to address the punishment of attempts).

HeinOnline -- 76 B. U. L. Rev. 10 1996

1996] A GENERATION LATER

tort/crime distinction should make sense. The key to rapprochement

rests on the relationship between moral discourse and practical affairs.
Ordinarily, we form views of right and wrong with a view to making judg-
ments about the conduct of others. Those judgments then guide both
individual and group behavior. Questions of enforcement are not central
to either of these inquiries, and tend therefore to recede quickly into the
background.
Legal discourse builds heavily on this moral discourse but does not du-
plicate it.26 Law and friction are inescapable companions. Enforcement
questions of all sorts and sizes necessarily intervene between the articula-
tion of a sound moral judgment and its translation into a legal command.
The undeniable proposition that enforcement is expensive implies that
some moral norms will not be enforced legally because resources are not
there to enforce them. The undeniable proposition that legal enforce-
ment is often unreliable also implies that simple proxies will often prove
more workable than complex substantive commands: the length of a knife
blade becomes a proxy for the intention to inflict bodily harm. The prac-
tical side of legal discourse treats these constraints as an inescapable com-
ponent of the overall enterprise. The peculiar jumps and starts of the
legal system frequently are little more than a workable compromise be-
tween the morally appropriate and the legally feasible. As a result, many
of our legal doctrines, even on substantive issues, take an odd form to
compensate for the remedial inadequacies of the legal system. The pro-
ject in this section is to trace this twisting process as it relates to the tort/
crime distinction.
The standard economic models of tort and crime are optimal deter-
rence models that ask what system of rules will minimize the costs of
violations and enforcement. 27 However, the equations driving this analy-
sis are not capable of formal solution. So in the mathematical tradition,
we abandon the elegance of a formal proof and adopt in its stead a
messier but sturdier system of successive approximations that bring us
ever closer to the ideal. What is required is a judicious mix of hunch and
wisdom.
This analysis suggests no unique approach, but I propose a first cut: as
between tort and crime-for the moment as between private and public
enforcement-which should come first? Historically, this is a somewhat
naive question because the development of civil law delict and common
law trespass and appeal did not distinguish sharply between criminal and
civil prosecution.2 8 Few standing state officials were around to handle

26 See H.L.A. HART, THE CONCEPT OF LAW 181 (1961) (observing that the devel-
opment of law has been "profoundly influenced" by moral discourse, but that legal
validity need not be explicitly tied to morality).
27 See, e.g., POSNER, supra note 4, at 119-77.
28 See J.H. BAKER, AN INTRODUCTION TO ENGLISH LEGAL HISTORY 571 (3d ed.
1990) ("In early societies there is no concept of the 'state.' Both compensation and

HeinOnline -- 76 B. U. L. Rev. 11 1996

 BOSTON UNIVERSITY LAW REVIEW [Vol. 76:1
any kind of case, so that private enforcement applied to all forms of anti-
social conduct, of which deliberate harms were the most pressing. It was
only with the emergence of the instruments of the modern state that
criminal enforcement became a specialized, public function. Thus the
tort law is a good place to begin because it functions with weaker social
institutions.
That conclusion is reinforced by the observation that traditional crimes
against property and persons have real victims who make ideal plaintiffs.
Just how would the legal system look if its only enforcement mechanism
was the tort law? To answer this question, we need to make some as-
sumptions about the effectiveness of this system in practice. Here I shall
hypothesize some very benevolent and wholly implausible assumptions
that tend to collapse the legal inquiry into the moral one. I shall assume
that questions of detection and proof can all be disposed of in a reliable
fashion; that the costs of bringing suit are sufficiently low that it pays
plaintiffs to sue for all but the smallest offenses; that the solvency of the
defendants is sufficiently great that they can answer for the substantial
harms they inflict; and that the substantive law can fashion decent dam-
age rules (no easy task).
To make this last point more sharply, consider two broad families of
damages rules. The first awards a sum equal to the hurt that the plaintiff
suffers; the second posits a sum equal to the gain that the defendant ob-
tains by inflicting harm. In the tort world the first of these measures is
usually preferred on the ground that an individual should be left no worse
off after the wrong committed than before. In practice the measure is
quite unattainable in connection with certain horrible injuries for which
no sum of money offers appropriate compensation. Likewise, the mea-
sure makes little sense in wrongful death cases where individual victims
cannot receive the damages payment.
However, in light of the austere assumptions set out above, these obvi-
ous difficulties may not be fatal to a privately operated system of sanc-
tions. As long as the relevant measure of damages is greater than a de-
fendant's gains, potential defendants will not commit harmful acts, even if
they do not pay full tort compensation. Thus, if an injury causes $10 mil-
lion in damages and produces a gain of $500,000, a (calculating) defend-
ant asked to pay anything over $500,000 will not undertake the action.
He is still worse off by doing the act and disgorging its benefits than by
doing nothing at all. The conclusion here is a happy one: With reliable
enforcement, perfect deterrence is possible with less than full tort dam-
ages.
However, it is also clear that even this tort-restitution cushion does not
counteract such simple realities as offenders who escape detection or who
are insolvent. Nor is the legal system a paragon of accuracy, even in sim-

retribution for wrongdoing are exacted at the instance of the wronged individual and
his kin.").

HeinOnline -- 76 B. U. L. Rev. 12 1996

19961 A GENERATION LATER

pie accident cases. But through it all, the basic intuition remains true: For
accidental harms at least, a system that puts the full burden of enforce-
ment on the injured plaintiff works better than a system of state prosecu-
tion for harms to discrete individuals. The private plaintiff has good
knowledge of the facts-or at least better knowledge than anyone else-
and strong incentives to prosecute the matter to judgment. So as a first
approximation the traditional tort system looks optimal despite its imper-
fections. Thus we can take some comfort in the older cases that stressed
the dominance of tort in cases of accidental harms.29
The same view seems to flow from liberal political theory. It is easy to
forget that one of the major issues in the classical treatments of political
philosophy concerned the scope of tort and crime, and setting the appro-
priate limits to the doctrine of self-defense, whose contours are of great
importance in the state of nature, that is, prior to the effective introduc-
tion of police forces. In this regard, it is useful to note that the Lockean
view draws a similar distinction between tortious and criminal behavior.
It is worth quoting the familiar and illuminating passages:
10. Besides the crime which consists in violating the law, and vary-
ing from the right rule of reason, whereby a man so far becomes
degenerate, and declares himself to quite the principles of human
nature and to be a noxious creature, there is commonly injury done
to some person or other, and some other man receives damage by his
transgression, in which case he who hath received any damage, has
besides the right of punishment common to him with other men, a
particular right to seek reparation from him that has done it. And
any other person who finds it just, may also join with him that is
injured, and assist him in recovering from the offender, so much as
may make satisfaction for the harm he has suffered.
11. From these two distinct rights, the one of punishing the crime
for restraint, and preventing the like offence, which right of punish-
ing is in everybody; the other taking reparation,which belongs only
to the injured party, comes it to pass that the magistrate, who by
being magistrate hath the common right of punishing put into his
hands, can often, where the public good demands not the execution
of the law, remit the punishment of criminal offenses by his own au-

29E.g., Hulle v. Orynge, Mich. 6 Edw. 4, fol. 7, pl. 18 (1466), reprinted in J.H.
BAKER & S.F.C. MILsOM, SOURCES OF ENGLISH LEGAL HISTORY 327, 330 (1986). A
participant in that case observed:
[I]f a man is shooting at the butts and his bow swerves in his hand and [the arrow]
kills a man, against his will, this (as has been said) is not a felony. If, however, he
injures a man by his archery, the man shall have a good action of trespass against
him: and yet the archery was lawful, and the injury which the other suffered was
against his will.
Id.; see also Weaver v. Ward, 80 Eng. Rep. 284 (K.B. 1616) (applying a similar distinc-
tion to the insanity defense that, although irrelevant in tort, is critical in criminal law).

HeinOnline -- 76 B. U. L. Rev. 13 1996

 BOSTON UNIVERSITY LAW REVIEW [Vol. 76:1
thority, but yet cannot remit the satisfaction due to any private man,
for the damage he has received. 0
The major reason why some power of punishment vested in the magis-
trate was "to prevent [an act] being committed again, by the right he has
of preserving all mankind, and doing all reasonable things he can in order
to that end."'" Because cases of pure accidental harm do not have the
follow-on effect, they seem not to call for criminal sanction. Locke un-
derstood that deliberate harm undermines the sense of security in other
persons on which social organization so clearly rests. In a world of scarce
resources, it is unlikely that the limited resources of the magistrate will be
directed against accidental harms.
The more difficult situation is indeed the opposite. Many deliberate
harms are detected. But often the questions of apprehension and satis-
faction of judgment are not satisfied. This minor inconvenience of the
tort system has enormous consequences both for its internal operation
and the emergence of the criminal law.
To begin with the former, ask: Why do we recognize a privilege of in-
flicting harm in self-defense? The privilege itself is messy and requires
figuring out the right rules to govern the use of excessive or deadly force,
and deciding whether the mental state of the aggressor is relevant in set-
ting proper limits on the defense. 2 But for these purposes, the central
point is that there would be no need for any privilege of self-defense if the
tort damage system worked flawlessly. In each case the aggressor would
be captured after inflicting the harm and subjected to a damage award
that left him worse off for inflicting the harm than from never acting at
all. Aggression would never pay, so no one would ever attempt it. Who
worries about the scope of a self-defense privilege if it need never be
exercised? But once we recognize that legal remedies are always imper-
fect, self-help remedies become critical to the success of the system,
which now has to sort through these questions as best it can. Better that
we run the risk of undue retaliation and concealed aggression then elimi-
nate the privilege altogether.
The breakdown of the tort remedial system also leads to the creation of
the criminal law. To be sure, with accidental harms, it might be better to

30 LOCKE, supra note 10, at 6-7. The references to the transition from collective
punishment by individuals to collective punishment by magistrates marks the transi-
tion from a state of nature. Of course noxious beasts could be killed with impunity,
and the parallel Locke draws between such conduct and private enforcement of one's
interests in the state of nature is an intentional one. Id. at 7. Further, the deadly
punishments are allowed for lesser offenses, if such is needed for public safety. Id.
("[E]ach transgression may be punished to that degree, and with so much severity as
will suffice to make it an ill bargain to the offender ...
31 Id. (emphasis omitted).
32 E.g., RICHARD A. EPSTEIN, CASES AND MATERIALS ON TORTS 35-39 (6th ed.

1995).

HeinOnline -- 76 B. U. L. Rev. 14 1996

1996] A GENERATION LATER

chalk up the plaintiff's inability to collect a judgment to experience. Yet

we do not adopt that same restrained attitude toward deliberate actions
fully capable of repetition. Self-defense will sometimes fail, even if legal-
ly allowed. But public fines will hardly fill the gap, for if the perpetrator
of the wrong has no wealth, he is no more able to pay a large fine than a
damage award. So the question is, what asset of the individual can the
law seize? Although a person might have "plenty of nothing," he always
fears losing at least two assets: his liberty and his life. Take these away
and deterrence against aggression can proceed apace even if compensa-
tion is utterly unavailable to the injured party. Such deterrence in turn
benefits others, at least indirectly, through a reduction in the crime rate
and the concomitant increase in personal security. Hence the question
becomes whether the increment of deterrence achieved through punish-
ment is worth the dislocation it creates in the lives of guilty individuals.
Although there is always some question as to how much punishment is
optimal, the criminal law's focus on deliberate harm once again speaks
eloquently to the overall social judgment: deterrence is better than inac-
tion when reparation is not possible. One imposes criminal sanctions
even if the loss of any life and liberty (including the wrongdoer's) is re-
garded as a social loss, and even though (in the modem context) the costs
of incarceration are high. Once we account for the imperfections of our
legal institutions, the need for both crime and tort, and the distinction
between them, starts to assume its customary shape.
Some additional complications are needed to fill in the outlines of any
comprehensive theory. It is a major over-simplification to assume that all
harms are caused by one of two types of human conduct: accidental or
deliberate. Tort and criminal law scholars delight in noting the many gra-
dations of mental states falling uneasily in between. Thus, next to strict
liability are rules imposing liability absent the highest levels of care; be-
side these are rules for ordinary negligence; next come rules for gross
negligence; for recklessness; for willful and wanton conduct; and for dif-
ferent shades of intention that involve conscious indifference or substan-
tial certainty that harm will ensue from the performance of certain ac-
tions. It takes no feat of deductive logic to realize that although tort and
crime divide the world into two spheres, the basic inventory of mental
states contains at least half a dozen notable members, and perhaps more
if we account for the various shades of criminal intention (purpose,
knowledge, etc.) as well.
The basic conceptual problem is just how any legal system can juggle
more distinctions than it can use. One possibility is to ignore many of the
refinements, especially in the important class of stranger cases. Thus, the
strict liability vision of the tort world does not preclude a plaintiff's re-
covery when the defendant has committed negligence. Quite the oppo-
site, one of the standard justifications for the strict liability rule is that it
dispenses with proof of negligence, which is surely present in the large
majority of civil cases. On the other side, we can simplify our treatment

HeinOnline -- 76 B. U. L. Rev. 15 1996

 BOSTON UNIVERSITY LAW REVIEW [Vol. 76:1

of criminal intention by using grades of intention solely on the more deli-

cate question of which offense (manslaughter, murder) with which to
charge a suspect. The basic divide therefore seems to be between gross
negligence on one side and reckless indifference on the other-about the
best that can be done.
Even here the inevitable complications continue to creep in at the
edges. The injection of punitive damages into civil actions once again
undermines the strict separation contemplated by classical views of the
subject."3 Many punitive damage cases involve affront and dignitary in-
terests, to which the criminal law pays relatively little regard. In other
instances punitive damages work well-perhaps too well-against insti-
tutional defendants where the solvency risk is small, so that public re-
sources can be directed to those cases where private prosecution is not
possible. The insolvent defendant is beyond the scope of the tort law in a
way that the institutional or professional defendant is not.
Several other challenges to the crime/tort line deserve brief mention
here. One key question concerns whether injunctions should join dam-
ages as a tort remedy. In principle, why not enjoin actions that result in
death or serious injury, especially because no amount of damages can
offer full compensation? In some cases, most notably in disputes between
neighbors, an injunction might also be a feasible remedy. If some harm
has already occurred, the injunction prevents its repetition. In other
cases, the fixed physical relations between parties establish who is in
harm's way, as with upper and lower riparians. Still, the risk with injunc-
tions is that they could enjoin lawful conduct, which, if allowed, would
have resulted in no harm. This overdeterrence problem is usually met by
requiring a plaintiff to show irreparable harm or imminent danger before
issuing the injunction.
Yet sometimes the likelihood of future harm may be great, but identifi-
cation of potential victims may be uncertain. Now, the familiar coordina-
tion problems make it difficult for potential victims to come together to
prevent actions likely to result in accidental harm. Resort to systems of
licensing (which have often been subject to massive forms of abuse) is
one possible response to this difficulty. This alternative works tolerably
well as long as the shift from private to public enforcement does not esca-
late civil sanctions into criminal ones. More importantly perhaps, the
idea of a proactive public police force makes better sense when there is
reason to prevent harm before it happens instead of relying on the deter-
rent effect of private litigation to achieve that result. The rise of the pub-
lic police force made this centralization of function a reality, although one
not without risk of abuse. Once again the collectivization of the injunc-
tive function should not be allowed to blur the lines between criminal

aa See Kenneth Mann, Punitive Civil Sanctions: The Middle Ground Between Crim-
inal and Civil Law, 101 YALE L.J. 1795, 1861-71 (1992) (assessing the impact of the
expansion of civil punitive damages on the tort/criminal law distinction).

HeinOnline -- 76 B. U. L. Rev. 16 1996

1996] A GENERATION LATER

punishments, appropriate for conspiracy and attempts, and a lesser set of

sanctions to deal with accidental injuries.
Finally, the view taken here helps explain something of the popular
unease concerning attempts. Clearly these events cannot be regarded as
torts, because of the want of a present victim. To keep matters in propor-
tion, the control of dangerous attempts must come through the criminal
law, now in circumstances in which the punishment of victimless offenses
makes sense, given the insecurity that failed efforts to kill or maim instill
within the community at large. But what form of punishment? The prac-
tical compromise in this area is that we punish attempts less severely than
completed offenses. It is hard to gainsay that view of the world in the
absence of any clearer compass.
Professor Steven Shavell has put his finger on the major difficulty.3 4 It
is quite impossible to think of the ideal punishment for attempted of-
fenses unless and until one first understands the status of punishment for
completed offenses. If these crimes are fully punished at the optimal lev-
el, then why bother with attempts? Punishment of the fewer completed
crimes makes clear the power of the state, and the number of attempts-
successful and otherwise-will decrease. In the end, the optimal level of
deterrence for completed offenses could remove all potential gain from
such conduct, obviating the need to punish attempts at all: indeed, there
would be no attempts to punish, for no one would dare take the risk of
capture. The analysis here is nothing more than a reprise of the basic
argument on self-defense.
But once again imperfections in the remedial structure shape the at-
tempted/completed offense distinction, as they do with the law of self-
defense. Here of course insolvency is not an issue, but as Shavell notes,
social norms place serious upper bounds on the amount of allowable pun-
ishment for any offense.3 5 This moral sentiment has a good practical
base. Allow torture and the effort to control criminals could induce wide-
spread misconduct by those officials entrusted with enforcing the criminal
law. Impose a death sentence for a first murder and the principle of mar-
ginal deterrence is violated because the punishment for the second can be
no worse than for the first, which is why the "three strikes and you're
out" rule may lead criminals to commit serious wrongs to avoid arrests
for trivial offenses.3 6 Yet placing some upper bound on punishment

34 Shavel, supra note 25, at 436-37.

35 Id at 436.
36 See, e.g., PeterJ. Benekos & Alida V. Merlo, Three Strikes and You're Out: The
PoliticalSentencing Game, FED. PROBATION, Mar. 1995, at 3, 6 (commenting on ob-
servations by police officers that offenders facing a third conviction under a "three
strikes" system are more prone to violent acts); Timothy Egan, A 3-Strike Law Shows
It's Not As Simple As It Seems, N.Y. TIMES, Feb. 15, 1994, at Al (noting that police
officers have encountered a "nothing to lose" attitude in confrontations with suspects
facing a third conviction).

HeinOnline -- 76 B. U. L. Rev. 17 1996

 BOSTON UNIVERSITY LAW REVIEW [Vol. 76:1
levels compresses the scale of punishments up and down the line. This
being the case, the only way to avoid having the second murder go for
free is to tolerate at the margin insufficient deterrence for the first and all
subsequent offenses. In brief, the law cannot preserve the right spacing
between offenses whenever there are upper bounds-practical or mor-
al-on the infliction of punishment.
It is out of this sense of frustration that the law of attempted acts, and,
by similar logic, of conspiracy and accessory liability, are born. If one
cannot increase the level of punishment on the small class of completed
offenses, then at some administrative cost it is possible to expand the net
of punishable offenses to capture near misses (which in attempted murder
cases could easily mean a large chest wound) in order to make up some of
the difference. As that is the case, there is good reason to understand
why people are so conflicted about the treatment of attempts and similar
offenses. The analytically correct solution depends on coordinating the
punishment of attempts with that of completed offenses. Yet the avail-
able information typically provides no clue how to fine tune the day-to-
day operations of the criminal law. So the system moves about in a mid-
dle range, without clear guidance. 7 Here, the economics can explain the
source of the ordinary ambivalence, even if it cannot define a precise
niche for the law of attempts.

V. LAW AND POPULAR MORALITY

Thus far I have tried to explain why a shift to an overtly consequential-
ist calculus does not require abandoning the traditional distinction be-
tween tort and criminal law. In closing, I think it useful to ask how the
moral and economic theory influences the legitimacy of the legal system
in the eyes of a wary public whose rights and duties it determines. In
order to approach that topic it is probably appropriate to say something
tentative about how the general public forms its views of legal rights and
duties.
In the absence of hard evidence, a couple of conclusions seem clear.
The enormous resistance to consequentialist thinking in ordinary life
should remind the legal profession that the vast bulk of the population
does not resonate overtly with any of the economic theorizing of which I

37 A similar issue arises in the tort law. Compensation in cases with a less than

50% chance that the defendant causes the harm in question depends on our treatment
of cases where the likelihood of causation by a defendant is greater than 50%. If
there is full compensation in the 50%-plus cases, then there should be no enforcement
in the 50%-minus cases, at least if deterrence is the goal of the civil system. But if we
utilize partial compensation for the 50%-plus cases, then we should use it across the
board. Whether that response is appropriate depends on the distribution of cases
around the 50% point. If all individual cases are below the 50% level, the result will
be underdeterrence unless we effect some adjustment. However, complications of
this sort are hard to work into popular morals.

HeinOnline -- 76 B. U. L. Rev. 18 1996

1996] A GENERATION LATER

have been guilty in this short Paper. Notwithstanding the rise of law and
economics in the academy, most people are more or less where I was on
legal thinking when I first addressed the tort/crime distinction in 1977.
They accept a strong categorical distinction between accidental and delib-
erate harms, and recognize that the latter are dangerous enough to re-
quire public control.
A major challenge to this distinction lies in the endless number of tort
suits against institutional defendants, all claiming that something more
than accidental harm is at stake. The most expensive words in the tort
lexicon are "send them a message," with its attendant invitation to puni-
tive damages as well as hefty compensatory damages. Although overused
and misused on particular occasions, the phrase has at least one saving
grace: it reminds us that most people have some sense of the deterrence
value associated with imposing both tort and criminal liability. It reminds
us that, for firms as well as individuals, reputational loss-that ubiquitous
stigma-delivers a large part of the total sanction. Once we have in-
voked "the message," it becomes increasingly difficult to defend the posi-
tion that social consequences and social utility-real or perceived-have
no bearing on popular views about civil and criminal liability.
Can these impulses be folded into a model that assigns to the criminal
law an expressive or condemnatory role missing from tort law? At one
level, the looming threat of punitive damages in tort settings constantly
blurs the distinction between the two systems. Many tort cases give rise
to expressive or condemnatory actions as well. But for these purposes, a
more vexing inquiry is whether the criminal law-even at its core-truly
reflects a social effort to regenerate fundamental norms by publicly con-
demning those who violate them. At some level, every society has to face
the task of reaffirming its values and beliefs during times of crisis. Funer-
al services that remind us of the sacrifices by those who have fallen victim
to senseless violence surely fill this role, as does the ultimate conviction of
an assailant.
But my own hunch is that the centrality of the criminal law to these
functions is limited. The enormous amount of self-examination and re-
crimination in Israel following the assassination of Prime Minister Yitzak
Rabin began at the instant of his untimely death and continued unabated
through the funeral services. It reached a fever pitch long before the be-
ginning of the trial of his assailant, Yigal Amir. Of course, this trial will
fall into the public spotlight. Yet there is little reason to think that its
inevitable outcome will shape Israeli or world opinion on the central
question of whether, and if so how, the harsh debate over peace with
Palestine induced-or drove?-Amir to commit cold-blooded murder.
Most criminal trials are not show trials. Often they involve acts of vio-
lence that ordinary people down on their luck commit on their fellow
citizens. In most cases the criminal system strains to keep up with the
inflow of business. Under these circumstances, the instrumental role of
criminal trials-locking people up-necessarily comes to dominate any

HeinOnline -- 76 B. U. L. Rev. 19 1996

 BOSTON UNIVERSITY LAW REVIEW [Vol. 76:1

possible expressive role. Routine cases often raise troublesome questions

of circumstantial evidence, and provoke protracted disputes over the
proper set of jury instructions. They offer a thousand temptations to di-
vert judge and jury from the fundamental principles of criminal responsi-
bility. At bottom, I remain convinced (if open to persuasion) that
whatever ordinary people think about the criminal law, their views are
not shaped in any material way by its day-to-day operation. The social
benefits often attributed to the "expressive" or "educative" function of
criminal trials most likely flow from some other source.
In The Hollow Hope"8 Gerald Rosenberg offers painstaking evidence
that the formation of public attitudes on such critical issues as racial de-
segregation and abortion depends little on the reasoning in judicial deci-
sions. Few people are persuaded to change their opinions, one way or the
other, by Roe v. Wade,3 9 or even Bowers v. Hardwick.4 On all these
grand issues people reach their own decisions based on information from
other sources that operate more powerfully on their intuitions and sensi-
bilities.41 Ironically, perhaps only lawyers see legal opinions as formative
influences shaping individual preferences and moral judgments.
What seems true about the premiere issues of the day carries over to
the more mundane issues of civil and criminal responsibility. I can testify
that any thoughts I had on the tort/crime distinction prior to my time in
law school were developed in complete ignorance of the Model Penal
Code or the Restatement of Torts. Rather, like other people, I derived
my views from the people with whom I spent most of my life: family,
friends, religious leaders, even newspapers and television, but rarely the
law. Even a direct confrontation with the legal system in action is unlike-
ly to change these core beliefs. Such experiences are far more likely to
become test cases that illustrate the wisdom of prior beliefs or the folly of
current practices, depending on the circumstances. At most, the opera-
tion of the system can confirm to those who care that the system is, or is
not, doing the proper job. The vast criminalization of regulatory offenses
does little to shape the views of ordinary people as to what conduct
should be made legal or not. Although academics may love to speak
about the most sophisticated form of corporate responsibility, ordinary
citizens still associate crime with aggression and molestation, as well they
should.

38 GERALD N. ROSENBERG, THE HOLLOW HOPE (1991) (questioning the capacity

of the Supreme Court to produce significant social reform).
39 411 U.S. 113 (1973).
40 478 U.S. 186 (1986).
41 ROSENBERG, supra note 38, at 157-69 (contending that the success of the civil
rights movement resulted from economic and electoral developments, shifting inter-
national demands, and other fundamental societal changes); id. at 247-65 (attributing
the end of discrimination against women to changing economics, decreasing fertility
rates, and improved education).

HeinOnline -- 76 B. U. L. Rev. 20 1996

1996] A GENERATION LATER

The great danger therefore is that the proliferation of criminal sanc-

tions, and to a lesser extent, tort sanctions, will undermine any widely
shared belief that the criminal or tort law does what ordinary people ex-
pect it will do. If someone actually tells us that building a retaining wall
on private property is as dangerous as poisoning a public well, we might
conclude that state officials do not attach a very high value to the sanctity
of life, or worse still, that committing crimes against the person is no
more serious than violating one of the countless administrative rules that
govern our lives. Regardless of their limitations, utilitarian theories do
not fall prey to that charge. They may use funny language to explain why
the older modes of doing business made sense, but they hardly represent
a frontal assault on either the categories of conduct that count as criminal
or tortious, or the relevant preconditions for civil and criminal liability.
Instead, they should be understood as a way to shore up a set of intuitions
that have proved themselves insufficient to withstand the relentless attack
by the modem regulatory state. A retrenchment in the size of govern-
ment will do enormous good in restoring public respect for state enforce-
ment efforts and redirecting scarce public resources towards more useful
ends. We need to worry about the line between tort and crime. But we
also need to shrink both domains simultaneously. On that issue at least,
both utilitarian thought and ordinary sensibilities should lead to the same
conclusion.

HeinOnline -- 76 B. U. L. Rev. 21 1996

HeinOnline -- 76 B. U. L. Rev. 22 1996