Journal of Philosophy, Inc.

Moral and Legal Obligation

Author(s): Conrad D. Johnson
Source: The Journal of Philosophy, Vol. 72, No. 12 (Jun. 19, 1975), pp. 315-333
Published by: Journal of Philosophy, Inc.
Stable URL: http://www.jstor.org/stable/2025132
Accessed: 21-04-2016 17:02 UTC

Your use of the JSTOR archive indicates your acceptance of the Terms & Conditions of Use, available at
http://about.jstor.org/terms

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.

Journal of Philosophy, Inc. is collaborating with JSTOR to digitize, preserve and extend access to The
Journal of Philosophy

This content downloaded from 14.139.209.67 on Thu, 21 Apr 2016 17:02:08 UTC
All use subject to http://about.jstor.org/terms
 THE JOURNAL OF PHILOSOPHY
VOLUME LXXII, NO. 12, JUNE 19, 1975

MORAL AND LEGAL OBLIGATION

N this paper I wish to consider anew' the question how legal

obligations differ from moral obligations. I shall be concerned
with obligations that arise from promises, though extensions
will sometimes be made beyond that sphere. Familiar criteria for
distinguishing moral from legal obligations are fundamentally
wrong partly because of a failure to find correct analogues in law to
various kinds of normative generalization employed in morality.
Here an attempt is made to identify those elements of the lawyer's
and judge's materials which are genuinely analogous to the more
often discussed elements of moral thinking.
I

A promise can give rise to a moral and a legal obligation. Thus

suppose I say to Jones (and, to forestall irrelevant objections, let it
be in the presence of ten other people): "In return for doing the
gardening while I'm gone, I'll pay you $100 a week. You have my
word on that." Jones agrees and does a conscientious job of garden-
ing during the six weeks that I am gone. Surely I have a moral obli-
gation to pay Jones $600. Pretty clearly I would also have a legal
obligation to pay Jones the money: He would have a good case in
a suit for damages.
It is important to analyze the elements which together make
promising and the obligations that arise from it possible. Consider
first the side of moral obligation. Here it is well to distinguish, as
others2 have, between the constitutive rules which determine what
actions constitute promises, and the moral principle which tells us
11 am interested in many of the same questions that interested H. L. A. Hart
in two places: "Legal and Moral Obligation" in A. I. Melden, ed., Essays in
Moral Philosophy (Seattle: Univ. of Washington Press, 1958), and The Concept
of Law (New York: Oxford, 1961), pp. 163-176.
2 See, for example, John Rawls, A Theory of Justice (Cambridge, Mass.: Har-
vard, 1971), p. 346.

315

This content downloaded from 14.139.209.67 on Thu, 21 Apr 2016 17:02:08 UTC
All use subject to http://about.jstor.org/terms
316 THE JOURNAL OF PHILOSOPHY

that promises are to be kept. The constitutive rules have an entirely

conventional-and not a moral-status, at least in the following
sense: It makes no great difference-certainly no moral difference-
that certain actions and not others constitute, in the appropriate
circumstances, the making of a promise. Presumably no objections
from moral principle could be made against the proposal that
jumping up and down twice replace uttering the words 'I promise
that . . .' as an action which, in the appropriate circumstances, con-
stitutes the making of a promise. Of course, there may be a galaxy
of other reasons why this would be a bad or unworkable proposal,
but not reasons of moral principle. The principle that promises
ought to be kept is fundamental and not open to alteration. In this
it stands in contrast to the constitutive rules. Further, what the
constitutive rules are must be construed quite liberally. In the
example above, a promise is made to Jones even though the prom-
isor does not use the word 'promise'. To formulate the constitutive
rule would be a nearly hopeless task because of the variety of other
words ('agree', 'will', 'undertake', etc.) which could be used to make
a promise; it would be impossible if we had to specify what circum-
stances are appropriate for making a promise by using these words.
Restricting ourselves again to the moral institution of promising,
we can ask what benefits are to be derived from having such an
institution. A promise is a thing of value because the promisee can
avail himself of the fact that either the promisor or those around
him (or both) have a sense of justice and adhere to the principle
that promises ought to be kept. The promisee thus has a kind of
moral property which he can use in getting actual performance. He
can appeal to the sense of justice of the promisor, or, failing that,
he can demonstrate to others that there was such a promise and
rely on their sense of justice, depending upon them to exert pressure
of various kinds on the promisor. At this point it becomes evident
why promising is governed by fairly explicit constitutive conven-
tions. The clearer and more explicit the constitutive conventions,
the clearer and less ambiguous it is to third parties that a promise
was made, or that an obligation was otherwise self-imposed.
Notice that, in making moral judgments about a person's obliga-
tions, we are willing to conclude that Ingemar, say, is obligated
to Ingemund because Ingemar "led Ingemund to believe" that
Ingemar would take care of the plants if Ingemund would only take
care of Ingemar's dog next month. But how do we know that
Ingemund really was led to believe this? And, assuming Ingemund
was led to believe this, how do we know that the belief was reason-

This content downloaded from 14.139.209.67 on Thu, 21 Apr 2016 17:02:08 UTC
All use subject to http://about.jstor.org/terms
 MORAL AND LEGAL OBLIGATION 317

able in the circumstances? We as third parties may not be able to

know these things unless we have a particularly intimate knowledge
of the parties' beliefs about each other, their (perhaps private)
customs of behavior, and so on; yet these are all relevant to mak-
ing a judgment on the moral question. Though it is less obvious,
there is also a need for the (potential) promisor to know what others
will justifiably take as his having put himself under an obligation.
In view of all this, a reasonably explicit and widely understood
convention of using specific words to put oneself under an obliga-
tion can cure the uncertainty. A new and more valuable moral
currency comes into existence, one whose value is evident on its
face. People can thus make agreements the mutual performance of
which each understands at the time of the agreement to be in his
interest. Similarly, at the time of the agreement each has the
assurance of possessing a moral claim against the other even though
he has not yet received actual performance.
Consider now the legal parallels. There are also, in a suitably
loose sense, constitutive rules to determine what constitutes a prom-
ise. It is very easy, however, to be misled about the status of these
rules. First, they are in general not the product of legislation.
Courts rely first and foremost on common sense, business practice,
community customs, and, in general, the same conventions any-
one who wished to determine whether there was an ordinary
promise would use for purposes of interpretation. This is not to say
that legislative enactments and precedent-setting judicial decisions
cannot or do not have any bearing on what the constitutive conven-
tions are; 3 it is only to say that courts take everyday conventions
much more seriously than those who believe that all contracts are
in boilerplate are apt to think. Secondly, it follows that, if the
everyday constitutive rules undergo a change, that fact will ipso
facto be relevant to the determination of people's legal obligations.
Legislation is not needed in order to recognize new ways of making
legally binding promises. A third and related point is that, with
some exceptions, the constitutive conventions are not to be re-

3 A striking example of legislative intervention resulting in constitutive con-

ventions different from those used in everyday life for the making of promises
is the Statute of Frauds (or more accurately, the various statutes of frauds
employed in different jurisdictions). A typical statute of frauds provides that
actions for damages will not be brought upon certain agreements (e.g., contracts
for the sale of land, contracts that are not to be performed within a year's time)
unless the agreement has been reduced to some writing, memorandum, or note
which has been signed by the person to be charged. The numerous exceptions
that courts have carved in the Statute shows that judicial legislation as well can
alter constitutive conventions.

This content downloaded from 14.139.209.67 on Thu, 21 Apr 2016 17:02:08 UTC
All use subject to http://about.jstor.org/terms
3I8 THE JOURNAL OF PHILOSOPHY

garded as legal rules. For the most part, they existed (or would
exist) before (or without) the legal system; and they are for the
most part not to be found written anywhere, either in pieces of
legislation or in the opinions of courts. Finally, there would be
as little prospect for successfully formulating the constitutive con-
ventions here as for promises generally. It is important to emphasize
these points because there is a danger of thinking that the law of
contracts must consist of (i) a set of rules (ii) which are or can be
formulated and (iii) which are a part of the legal system because
enacted or pronounced to be valid by some legal authority, such as
a legislature or court.
H. L. A. Hart is one who, though surely aware of the facts about
the formation and enforcement of contracts as few philosophers
are, has in his own writings tended to encourage the thought that
the power to create contractual obligations is primarily conferred
by rules that have been formulated by a legislature. These rules
provide that "certain words" being used in a certain way will give
rise to certain obligations.4 Thus on one page Hart writes at
length of the "legal rules" that confer powers on individuals; he
speaks of these as "laws"; and he gives the Wills Act of 1837 as
an example.5 What can be misleading in all this is that it obscures
the way in which the common-law development of contract law has
not been limited to the enforcement of contracts made in accord-
ance with the requirements of rigid rules of formation. It also
misleads in encouraging the idea that contracts are enforced because
they are made in accordance with specific rules of formation rather
than because of general principles, such as the principle that a
promisor ought to pay for losses that the promisee suffers in reli-
ance on that promise, or the principle that one who gets something
of value by promising ought to do what has been promised.
Just as in the moral sphere constitutive conventions are to be
distinguished from the principle requiring that promises be kept,
so in the law a distinction can be made between constitutive rules
and the principles that courts follow in enforcing contracts. These
principles have much in common with the principles that govern
what we do about ordinary promises in everyday life. In enforcing
contracts, courts are, in an important sense, enforcing moral prin-
ciples. There is certainly no one principle, moral or otherwise, that
courts employ in dealing with the contractual disputes that come
to them. Thus, generally, if A and B have entered into an agree-

4 See "Legal and Moral Obligation," p. 88.

5 The Concept of Law, p. 28.

This content downloaded from 14.139.209.67 on Thu, 21 Apr 2016 17:02:08 UTC
All use subject to http://about.jstor.org/terms
 MORAL AND LEGAL OBLIGATION 319

ment courts regard as legally binding and if A performs but B

breaches, A can recover compensatory damages enough to put him
in as good a position as if the contract had been fully performed.
This result is called for by a principle akin to one already cited:
One who gets something of value by promising ought to do what
has been promised. That this is not the only kind of principle em-
ployed can be seen in the following: Bart Baumeister and Horatio
enter into the following written but unsigned agreement: Bart is to
build a house for Horatio in return for an agreed sum. Bart goes
to considerable expense in preparing to start the project, but
then Horatio changes his mind and refuses to permit Bart to begin
building on his (Horatio's) property. The contract turns out to be
one that courts would not otherwise enforce (e.g., by the principle
just cited) because, being for the sale of real estate, it falls under
the Statute of Frauds, which requires a written and signed agree-
ment. Even though courts would not require each to perform his
part of the agreement, Horatio would probably be compelled to
compensate Bart for losses incurred in relying on the unsigned
agreement. (Horatio would have even more to pay if he had been
unjustly enriched through Bart's losses.)
There is, therefore, a straightforward sense in which a court ap-
plies, and if you will, enforces moral principles in the settlement of
disputes. This would be most evident in a first-instance case, where
it cannot be said that the court's reason for the action is merely to
follow the lead of another legal organ; but even if a precedent has
been set, one of the principles just cited will still constitute a strong
(probably primary) reason for the kind of remedy in question,
even though this reason may be supplemented by another, i.e., that
the governed justifiably rely on courts to follow precedent.
In addition to the rather large collection of principles-plausibly
deemed moral or well as legal-which courts employ, there is a
large body of principles adherence to which furthers social and in-
stitutional goals, purposes, and policies. It is the use of these princi-
ples which accounts for the fact that legal solutions to contractual
disputes sometimes differ from what common-sense morality would
require. For example, take old Mr. Sorglos. He thinks that his
neighbor's son Teddy is such a model young citizen that he writes
Teddy a letter congratulating him on his high school Good Citizen
Award and ends by saying, "In fact, I am so impressed that I'll give
you $5000 before your graduation. You have my word on that."
Though Sorglos would pretty clearly have a moral obligation to
young Teddy, he would just as certainly not be under a legal ob-

This content downloaded from 14.139.209.67 on Thu, 21 Apr 2016 17:02:08 UTC
All use subject to http://about.jstor.org/terms
320 THE JOURNAL OF PHILOSOPHY

ligation. Various reasons have been adduced to support the

common law's failure to make such a promise legally binding. A
typical reason is that, in general, it is safer to assume that a prom-
isor is absolutely serious about his promise when he offers it in
order to get something in return. Another is that, where the
promisee has suffered no material and demonstrable loss in reliance
on the promise it would be unwise for courts to intervene. They
have enough on their hands enforcing contracts in which promises
are in earnest and a clear-cut injustice would result from a failure
of the promisor to perform.6 (It would be relevant to Teddy's claim
if, relying on Sorglos's promise, he had taken out a loan and paid
nonrefundable college tuition in the expectation that Sorglos's
money would cover the rest.) Now take another example. Dita Diva,
an irascible coloratura, has signed a contract in which she agrees
to perform at the grand opening of a new center for the performing
arts. The owners of the center rightly believe that to have her per-
form on opening night would give the center a prestige and im-
portance to wlhich a money value could not be meaningfully at-
tached. A week before the opening, Miss Diva angrily announces
that she refuses to perform. In a case such as this, courts would
be unwilling to issue an injunction requiring her to perform, and
the most that the owners could obtain would be money damages.
The principle cited earlier-that one who gets something of value
by promising ought to do what has been promised-will thus re-
main, strictly speaking, unenforced. What Miss Diva has promised
is to sing at that star-studded opening and this she will not be
forced to do. Enforcement of the principle gives way to practical
considerations which become relevant when the question is whether
the apparatus of state coercion-as opposed to urgings, reminders
of one's duty, and so on-is to be used. It would obviously be
undesirable to have a diva like Dita sing under duress.
Initially, there are several things to notice about these examples.
Policy (in general, considerations peculiar to the legal system)
played a role in determining that Sorglos had no legal obligation
at all. Diva, on the other hand, does have a legal obligation, but she
will not be held to the letter of that obligation. In Diva's case
r "Such promises [as Sorglos's] . . . are lightly made, dictated by generosity,
courtesy, or impulse-often by ruinous prodigality. To enforce them by a judg-
ment in favor of those who gave nothing therefor would often bring such im-
perfect obligations into competition with the absolute duties to wife and
children, or into competition with debts for property actually received, and make
the law an instrument by which a man could be forced to be generous before
he was just." Lamar, J., in Davis & Co. v. Morgan, 1903, 117 Ga. 504, 43 S.E.
732, 61 L.R.A. 148, 97 Am.St.Rep. 171.

This content downloaded from 14.139.209.67 on Thu, 21 Apr 2016 17:02:08 UTC
All use subject to http://about.jstor.org/terms
 MORAL AND LEGAL OBLIGATION 321

too the leading consideration is policy, and though some may

prefer to regard as a moral principle the policy that a sensitive diva
should not be forced to sing, some legal considerations, such as the
importance of not getting courts involved in unnecessary litiga-
tion, seem to be more characteristically nonmoral.
In what does the difference between legal and moral obligation
consist? I shall move in the direction of an answer to that question
by considering separately some criteria that have had great currency
among philosophers. We ought to be much more puzzled by these
criteria than we are.

1. The criterion of immunity from deliberate change. No suggestion

is more typical than that legal requirements can, and moral require-
ments cannot, be created or extinguished by the act, will, or
decision of some person or group. Thus Baier says:

It is the very function of the legislator to bring into being or abolish

the illegality of something, as he thinks fit. However, the rightness or
wrongness of a type of action cannot be the logical consequence of
anyone's decision.... It is nonsense to say, "Yesterday God decreed
that killing shall no longer be morally wrong" or "The moral law
against lying was promulgated on May 1st." Morality, therefore,
cannot be any sort of law.7

Putting it somewhat differently but in very much the same spirit,

Hart says:

It is characteristic of a legal system that new legal rules can be intro-

duced and old ones changed or repealed by deliberate enactment....
By contrast moral rules or principles cannot be brought into being or
changed or eliminated in this way (The Concept of Law, 171).

And again:

. . .the idea of a moral legislature with competence to make and

change morals, as legal enactments make and change law, is repug-
nant to the whole notion of morality (173).

Not one of these passages contains a single unambiguous claim, and

some sorting out is required. First consider Baier's claim that the
rightness or wrongness of a type of action cannot be the "logical
consequence" of anyone's decision. Whatever this may mean (es-
pecially the part in scare quotes), one thing is clear: The reason it
is (prima facie) wrong for Sorglos to fail to give Teddy the money
is simply Sorglos's act of promising. Likewise, Diva made a type

7 Kurt Baier, The Moral Point of View (Ithaca, N. Y.: Cornell, 1958), p. 178.

This content downloaded from 14.139.209.67 on Thu, 21 Apr 2016 17:02:08 UTC
All use subject to http://about.jstor.org/terms
322 THE JOURNAL OF PHILOSOPHY

of action wrong by her act of promising. There are other examples.

The acceptance of public office (and many other roles and posi-
tions) makes certain kinds of act wrong that would otherwise be
permissible, and vice versa. Nor is it always necessary that the ac-
ceptance be accompanied by an explicit promise or oath. Few
would deny either that obligations are created in this way or that
these are moral obligations.
Now consider Hart's suggestion that moral "rules or principles"
cannot be brought into existence or abolished by deliberate en-
actment. This is surely correct as a claim about moral principles.
For example, the principles that one ought to tell the truth, that
one ought to do what one has promised, could hardly be abolished
by any act of mine or of any legislature, no matter how powerful.
For to abolish such a principle would seem to mean one of two
things: (a) that the principle would, as of a certain instant, cease
to be accepted by the community as a valid principle; (b) that the
principle would, as of some specified time, cease to be valid, true, or
correct simpliciter. Abolition in sense (a) is, one is inclined to
think, a psychological or sociological near-impossibility. Abolition
in sense (b) would seem to be a conceptual impossibility: In our
conception of morality, at any rate, the validity of moral principles
may depend on many things but not on someone's sayso. We now
need to ask what the legal analogues to such principles are. Surely
the natural legal analogue to the general principle that one ought
to do what one has promised is simply the principle, used as a basis
for contractual liability, that one ought to do what one has prom-
ised. It does no good to say that there are other legal principles
which sometimes conflict with this one, for there are many (some-
times conflicting) principles in what we ordinarily think of as the
province of morality. Now what can it mean to claim that the
legal principle that one ought to do what one has promised can be
abolished? The first thing to notice is that a legislature would
hardly succeed in abolishing such a principle merely by abolishing
this or that statute. A legislature could add further provisions to
the Statute of Frauds, revise or eliminate entirely the Uniform
Commercial Code (returning to the common law of contracts), but
none of this would abolish the principle in question. Suppose an
even more drastic legislative step were taken: It is enacted that no
promise or agreement in any form will be enforced by the legal
system. This would certainly be an heroic attempt to abolish a
principle from the legal system, and it would be hard to imagine
how far-reaching the repercussions would be, but the important

This content downloaded from 14.139.209.67 on Thu, 21 Apr 2016 17:02:08 UTC
All use subject to http://about.jstor.org/terms
 MORAL AND LEGAL OBLIGATION 323

point is that even this drastic action would not succeed. Recall
that, even in its heyday, the principle did not always hold sway,
for there were a variety of competing principles and considerations.
To say that a principle is a recognized legal (or moral) principle is
not to say that deviations from it are never called for; it means
rather that the principle has some (perhaps considerable) weight in
deliberations of that sort, whether legal or moral.8 Now the prin-
ciples forming the basis of contractual liability are not relevant
only in the law of contracts as we know it. They also hold sway
elsewhere. For example, suppose that the Telegraph Company, a
public utility, agrees to send A's message to B. The message is not
sent. Quite apart from the question whether A has any contractual
rights against Telegraph Company, such situations as this give
rise to Telegraph Company's liability in tort to B for failing to
transmit the message. The principle that one ought to do what
one has promised thus plays a role in a quite different area of the
law, and the result is that Telegraph Company is subjected to a
quite different measure of money damages than it would if liability
were based on contractual obligation.9 To take another example,
suppose that the state legislature's tax law provides that the rate
of taxation "shall not apply to any railway company hereafter
building and operating a line . .. until the same has been operated
for the full period of ten years. . . ." In Wisconsin & M.R. Co. v.
Powers (24 S.Ct. 107, 191 U.S. 379, 48 L.Ed. 229 (1903)), Justice
Holmes held that these words did not constitute a promise. That it
is legally relevant to ask whether it is, however, is evident both
from the fact that Holmes spent some effort in disposing of the
claim that there was a promise and from the fact that Professor
Corbin, in his treatise on contracts, says this:

It may be suggested that while the interpretation given by the court

seems not unreasonable to a disinterested party, it would not seem
so reasonable to one who had built a railroad and rendered the
desired service in reliance on the statutory assurance. It may be sound
policy to deprive legislatures of the power to bind the state by prom-
ises of future exemption from taxation.1O [my emphasis]

Indeed, it should be clear by now that to abolish the principle that

8 On principles and their dimension of weight, Ronald Dworkin has many in-
teresting things to say. See "Is Law a System of Rules?" in Robert S. Summers,
ed., Essays in Legal Philosophy (Oxford: Basil Blackwell, 1968), pp. 25-60.
9 On tort liability of this sort, see William L. Prosser, Handbook of the Law
of Torts (St. Paul, Minn.: West, 1964), p. 689, and the cases cited there.
10 Arthur L. Corbin, Corbin on Contracts, one-volume ed. (St. Paul, Minn.:
West, 1952), p. 23.

This content downloaded from 14.139.209.67 on Thu, 21 Apr 2016 17:02:08 UTC
All use subject to http://about.jstor.org/terms
324 THE JOURNAL OF PHILOSOPHY

one ought to do what one has promised would not merely mean
that the principle would sometimes be outweighed by another, or
would no longer form the basis for an entire area of the law; it
means that it would be systematically excluded from having any
weight in legal argument in any significant area of the law. Indeed,
it would not even be clear that the constitution itself, founded as
it is in an agreement, should any longer be taken seriously by
courts. I leave the reader to think of further consequences. Beyond
that, I suggest only that we abandon the idea that this legal prin-
ciple, like a city ordinance, is vulnerable to deliberate abolition.
Another claim of Hart's is that the idea of a "moral legislature"
is repugnant to the concept of morality. This is a multiply ambigu-
ous claim since it is not specified what this supposed legislature
would have the power to change (e.g., whether standards, rules,
principles, obligations, or "morals"). That a moral legislature (in
at least one interesting sense) is hardly impossible can be seen from
the simple fact that, in suitable circumstances, a legislature that has
the power to enact laws also has the power to create moral obliga-
tions. Suppose that the members of a group unanimously agree to
a constitution giving legislative powers to a subgroup. The sub-
group enacts a tax law providing that, among other things, tax
returns are to be filed by 11 PM on a specified day. Not only is there
a legal obligation thus created, but a quite specific (because of the
specificity of the act required) moral obligation is created. In fact,
these obligations are not to be distinguished by their content. Nor
is it necessary that we restrict ourselves to such ideal cases in which
there is an actual social contract giving authority to this subgroup.
In Rawlsian contract theory, there is a duty to comply with the
laws emanating from a just constitutional arrangement even if some
of these laws are themselves unjust.1' The Rawlsian legislature thus
can impose quite specific-as well as general-moral obligations on
the governed. There are, to be sure, certain obligations that a
legislature cannot create or abolish, but then there are certain
obligations that private parties cannot create or abolish when they
promise or make agreements.
It might be objected that there is a general moral obligation to
obey the law, and this no legislature creates or abolishes; that it is
rather the law that is created or abolished. This objection, how-
ever, is of little use. If it were a good argument, it could be used
to establish that we do not really impose moral obligations on
ourselves and others by promising, since there is a general prin-
11 See A Theory of Justice, pp. 350-355.

This content downloaded from 14.139.209.67 on Thu, 21 Apr 2016 17:02:08 UTC
All use subject to http://about.jstor.org/terms
 MORAL AND LEGAL OBLIGATION 325

ciple that promises ought to be kept, and this is not created or

extinguished by fiat. In both cases, a general requirement that is
not subject to deliberate change is employed to produce specific
requirements that surely are subject to deliberate change.
There is perhaps a somewhat deeper reason why many think that
there is something conceptually unfitting in the idea of a legislature
with power to impose moral obligations. Thus one might make the
claim that all that is necessary to the imposition of a legal obliga-
tion is the existence somewhere of some legal system that enacts
the appropriate law. Hence, if sufficiently perverse, the lawmakers
of a foreign country with which I have never had any association
whatever could impose any legal obligation on me they wished, so
long as they could succeed in passing the appropriate law. It
would require heroic effort to establish that they could also impose
a moral obligation on me in this way.'2 Since this question is not
essential to my discussion, my remarks will be brief and incomplete.
When a government attempts to impose a legal obligation on a
foreign citizen, this attempt is typically backed by the claim that
there is some pre-existing moral relationship between that citizen
and that government. For example, foreign visitors to a country
can be put under a legal obligation to do certain things, and
the natural argument is that they enjoy the benefits of protection
while visitors; they know or ought to know that certain things will
be expected of them; hence they owe deference to the laws of the
country, and this moral relationship exists before, and independent
of, the creation of the law imposing a legal obligation. Similarly,
a country might be able to support the claim that the citizen of a
second country owes something to the first on account of having
been raised there, having received an education there, and so
on. This claim might in turn be used to defend the legislative
action of requiring military service of this citizen. It would be ex-
tremely strange to find jurists and authorities (who have any sensi-
tivity to legal principles) arguing that, although someone may be a
total stranger who has had no connection with a given country, and
stands in no moral relationship with respect to it, that country can
impose a legal obligation on him merely by legislative fiat. It is
natural-though the foregoing is no knock-down argument-to
think that in such a case the stranger could be threatened or even

12 I have discussed this in connection with a criticism of Rawls's argument

that we have a natural duty to obey the laws emanating from a just constitu-
tional system in my paper, "Actual- v. (Rawlsian) Hypothetical-Consent,"
Philosophical Studies, forthcoming.

This content downloaded from 14.139.209.67 on Thu, 21 Apr 2016 17:02:08 UTC
All use subject to http://about.jstor.org/terms
 326 THE JOURNAL OF PHILOSOPHY

locked up by authorities of the country in question, but that they

could not succeed in imposing a legal obligation on him.

2. The criterion of relativity to goals, purposes, policies. A less

familiar but somewhat more promising criterion for distinguishing
legal from moral requirements is that the former, though not the
latter, can be (and are) made to depend upon a variety of goals,
purposes, and policies. Joel Feinberg, in a related attempt to dis-
tinguish legal from moral responsibility, suggests that this is one
important difference between the legal and the moral:

Determining legal responsibility in problematic cases often comes

down to the questions of who ought to pay or who ought to be pun-
ished and how much. These questions are rendered problematic by
conflicting interests and principles of justice, and the answers to
them usually depend on what the judge takes to be the "ends" or
"purposes" of compensation and punishment.13

And further:

In those cases where legal responsibility is problematic, moral re-

sponsibility would be absolutely undecidable in principle and there-
fore inapplicable since, in respect to moral responsibility as here un-
derstood, we are not allowed to appeal to purposes and policies (32).

Remembering the various ways in which particular policies and

goals are taken into consideration in deciding whether and how
contracts are to be enforced, it seems natural to extend this idea
to make the needed distinction between legal and moral obligation.
Thus in addition to principles-identical with everyday moral
principles-which form the basis of contractual liability, goals and
policies play a role in the question of legal obligation. The distinc-
tion between legal and moral obligation will be sharp if it is clear
that (a) there is a sharp distinction between moral principles on the
one hand and goals, purposes, and policies on the other; (b) the
question of moral obligation and liability is not made to depend
upon such goals, purposes, and policies. I shall consider each of
these in turn.
We have seen that the principle that one ought to do what one
has promised plays a fundamental role in legal thinking. For this
very reason among others it has the status of a legal principle. Con-
sider now some examples of agreements which may not be enforced
for reasons of what lawyers call "policy." An agreement may be
la "Problematic Responsibility in Law and Morals," in Doing and Deserving
(Princeton: University Press, 1970), p. 27.

This content downloaded from 14.139.209.67 on Thu, 21 Apr 2016 17:02:08 UTC
All use subject to http://about.jstor.org/terms
 MORAL AND LEGAL OBLIGATION 327

held unenforceable because it would operate in restraint of trade.

Similarly, there has sometimes been the policy, reflected in legis-
lative enactment, of not enforcing agreements not to compete after
termination of employment. Thus an employee might agree with
his employer that, should his employment be terminated for any
reason, he will not engage in a similar line of business in compe-
tition with his erstwhile employer. Where strictures on the enforce-
ment of such contracts have existed, an important reason for these
strictures seems to have been a desire not to lend legal support to
the limitation of competition. There is indeed an important dif-
ference between a policy such as this one and the principle that
promises ought to be kept. The policy may have only a temporary
existence in the legal system. Legislative enactments may well have
brought the policy into existence, and legislative enactments can
likewise be responsible for its demise. But whether or not the policy
owes its life to legislative decision, the important point is this.
Compared to the principle that promises are to be kept, the policy
that business competition shall not be limited plays a relatively ad-
ventitious role in a legal system. I have already argued that it would
be hard to imagine what the legal system would be like if the whole-
sale abolition of the principle were to succeed. In fact, we would
have a very hard time thinking of the result as anything like law or
a legal system at all. On the other hand, to abolish the policy
against aiding the restriction of business competition might be
undesirable, it might be very unwise from an economic point of
view, but it quite certainly would not have such far-reaching reper-
cussions that the whole life of the legal system would be disrupted.
Another example is provided by the quote (cited 323 above) from
Corbin, in which he speculates that it may be sound policy to de-
prive legislatures of the power of binding the state through prom-
ises. One can easily imagine a legal system in which legislatures are
recognized to have this power. No far-reaching consequences
going to the very core of a legal system and the idea of ruling
by law would be involved either in recognizing or not recognizing
this power.
There is therefore a plausible distinction to be made between
policies and principles. However, there is not always a crisp answer
to the question in what category a particular generalization falls.
An example is the general refusal of courts to enforce unconscion-
able contracts. Thus in Campbell Soup Co. v. Wentz,14 an agree-
ment under which grower Wentz was to provide Campbell with
14 172 F. 2d 80, 3d Cir. 1948.

This content downloaded from 14.139.209.67 on Thu, 21 Apr 2016 17:02:08 UTC
All use subject to http://about.jstor.org/terms
328 THE JOURNAL OF PHILOSOPHY

carrots, the contract provided, among other things, that if Campbell

was unable to process carrots grown by Wentz, Campbell would not
be obligated to receive or pay for them and Wentz would not be
permitted to sell those carrots elsewhere unless Campbell agreed.
The court found this to be much too tough a bargain to justify
compelling Wentz to specific performance of the contract (i.e., to
do what was specifically agreed rather than to pay money damages).
On the one hand, we might think of this result as required by a
policy of not lending judicial support to extraordinarily hard bar-
gains-a policy that could in theory be removed wholesale from the
law. But on the other hand, the notion that gross inequality in
bargaining position tends to weaken the consequent obligation of
the party in the inferior position is an idea too deeply entrenched
both in legal and everyday thinking to count as mere policy.'5
As to (a), then, there is a distinction to be made between prin-
ciples and policies, but this distinction is not in every case a clear
one. No doubt excessive attention to those parts of the lawyer's and
judge's materials-in particular, statutes and policies-which can
be created, changed, or eliminated has been one important source
of the idea that there is an accidental connection at best between
legal and moral principles.
I turn to (b), the question whether moral obligation is likewise
made to depend, in an interesting sense, on policy. Hart says flatly:
"Morals like law may have principles of 'public policy' and render
'void' a promise that from the start involved doing something
patently immoral." 16 Judging from the scare quotes, Hart is evi-
dently a little wary about this claim. There is initially something
strange in the idea that judgments about moral obligations arising
from promises should depend on public policy. But if we locate in
morality the real analogue to public policy in contract law, at least
some of the strangeness disappears. What courts do in recognizing
legal obligations arising from agreements is, generally, to award
15 Related issues arose in Henningsen v. Bloomfield Motors (32 N. J. 358, 161
A. 2d 69 (1960)). There the question was the implied warrantly of merchant-
ability of an automobile purchased from Bloomfield Motors. Mrs. Henningsen
had been injured when the car veered into a brick wall. The court refused to
recognize as a limitation on the manufacturer's liability the fact that Henningsen
had executed a purchase order on the back of which there was the following
language: "this warranty being expressly in lieu of all other warranties ex-
pressed or implied." The court stressed the fact that the bargaining position of
buyer and seller in such a case is very unequal and asked, "Where can the
buyer go to negotiate for better protection?" The court in Henningsen, how-
ever, does speak of the reasons for not enforcing such clauses as "public policy"
and discusses the notion of public policy at some length.
16"Legal and Moral Obligation," p. 102.

This content downloaded from 14.139.209.67 on Thu, 21 Apr 2016 17:02:08 UTC
All use subject to http://about.jstor.org/terms
 MORAL AND LEGAL OBLIGATION 329

damages or specific performance in case of breach. The analogue

in morality is criticism, reminders, blame, and various other kinds
of more diffuse social pressure in case a promise is broken. Now I
maintain that the principle that one ought to do what one has
promised always constitutes a reason for blaming or criticizing one
who has broken his promise, and this parallels the fact that the
same principle always constitutes a reason for the legal enforcement
of a promise. But it is obvious that we sometimes have a raft of
reasons for not blaming or criticizing one who has broken a
promise. And these reasons may in some instances spring from
policy. For example, a person may be too sensitive and easily in-
jured to bear criticism, or a group may decide that too much (or
too little) criticism has injured group solidarity, and they may
adopt a new policy to change that. There is a difference, to be sure,
in that there is presumably no central decision-maker who sets up
policy governing criticism, although the existence of such a central
decision-maker even in this case would not seem to be a con-
ceptual absurdity. Both legal and moral obligation, then, are in a
sense governed by policy-even public policy. A central feature of
legal obligation, however, is that to have a legal obligation is to be
under a requirement which, in a suitably broad sense, outside
parties will be prepared to enforce in a regular way, using pub-
licly available procedures to determine the fact of violation and the
way violations will be handled. Indeed, typically, a nonlegal
obligation hardens into a legal obligation when others stand ready
to assure such enforcement. But it is obvious that the centralized
and public enforcement of requirements brings with it all sorts
of policy considerations that simply would not arise when enforce-
ment operates via scattered individual criticism. For example, tak-
ing into account economic and foreign policy considerations may be
both desirable and unavoidable in the public sphere and neither
of these in the private sphere of negotiations and agreements be-
tween individuals. Thus many policy considerations that by the
nature of the case play no role in governing moral criticism come
to the fore in determining which requirements will constitute legal
obligations. This is perhaps clearest in the case of promises between
private parties.

3. The criterion of importance. Another typical suggestion is that

moral rules, but not all legal rules, have, and are perceived to have,
a fundamental importance. For Hart, who attempts to distinguish
moral and legal rules of obligation, this is manifested in essen-

This content downloaded from 14.139.209.67 on Thu, 21 Apr 2016 17:02:08 UTC
All use subject to http://about.jstor.org/terms
330 THE JOURNAL OF PHILOSOPHY

tially three ways:

. . . moral standards are maintained against the drive of strong

passions which they restrict, and at the cost of sacrificing considerable
personal interest; secondly, in the serious forms of social pressure
exerted not only to obtain conformity in individual cases, but to
secure that moral standards are taught . . . ; thirdly, in the general
recognition that, if moral standards were not generally accepted, far-
reaching and distasteful changes in the life of individuals would occur
(The Concept of Law, 169).

Legal rules, while sometimes having all these features, sometimes

have none of these features:

. . . importance is not essential to the status of all legal rules as it is

to that of morals. A legal rule may be generally thought quite unim-
portant to maintain; indeed it may generally be agreed that it should
be repealed: yet it remains a legal rule until it is repealed. It would,
on the other hand, be absurd to think of a rule as part of the morality
of a society even though no one thought it any longer important or
worth maintaining (170/1).

Notice first that some confusion may be engendered by the term

'rules of obligation': We do not know whether Mickey's obligation
to water the plants as promised counts as a rule of obligation, nor
do we know whether the obligation of anyone who earns above a
specified amount to pay a specified tax counts as a rule of obliga-
tion. Let us speak simply of obligations (or sometimes more gen-
erally of requirements), and ask how we distinguish between the
moral and the legal. Now, viewed in this way, it becomes quite
clear that some moral requirements do not manifest importance in
the sense explained. Thus suppose that there is a general under-
standing on the part of members of a family that whoever uses the
car will refill the tank. It would be wrong to ignore this require-
ment, and morally so, even though it would be ludicrous to suppose
that this requirement is maintained against the drive of "strong
passions" or that there is a general recognition that, if not generally
accepted (or adhered to) "far-reaching and distasteful changes"
would take place in the life of this small community.
Much the same point can be made even where the requirement
does not grow out of any agreement or understanding. Suppose that
in one society a custom has slowly evolved whereby one indicates
appreciation to a dinner hostess by accompanying one's eating with
what we would regard as excessively loud noises. You are invited to
dinner and you know the custom. It would be wrong for you to eat

This content downloaded from 14.139.209.67 on Thu, 21 Apr 2016 17:02:08 UTC
All use subject to http://about.jstor.org/terms
 MORAL AND LEGAL OBLIGATION 331

silently. (This does not mean that you could not have an excuse for
doing so.) If you do eat silently, it would hurt the hostess's feel-
ings, perhaps seriously. Because it is morally wrong needlessly to
hurt people's feelings, it is morally required that you not eat
silently. This custom is unimportant in the following sense: It could
be altered radically without far-reaching undesirable social changes
taking place; many-perhaps all-members of the society recognize
this and hence do not perceive the custom as important; and finally,
not only is it possible that the custom is not perceived as im-
portant to maintain, members of the society may be unanimous in
wishing that the custom were replaced by another-they may be
trapped by it.17 Still, the custom does define the way in which one
indicates appreciation at a dinner. What is important is the general
principle from which this particular requirement springs, i.e., that
one ought not to hurt people's feelings needlessly.
There is a related thought about the difference between moral
and legal requirements that must be rejected. This is the idea that,
whereas legal requirements are sometimes unjust, genuine moral
obligations are by the nature of the case consistent with what justice
requires. Consider again the obligation that each member of the
family has to refill the tank after use. This requirement may be
unjust in that it imposes unequal burdens on different members of
the family. The injustice may be so great as to justify refusing
to do what is required. But this is consistent with there being an
obligation to refill the tank; for the obligation arises from the un-
derstanding on which everyone relies, and this holds even though,
ultimately, other considerations, such as the justice of the require-
ment, may justify failure to do what is required.
One reason why it is so easy to think that moral requirements
are fundamental and important but legal requirements sometimes
are not may be the following: We tend to compare law and morality
casually and without being sure that we compare a particular level
in law with the analogous level in morality. Thus we find the com-
parison of such global moral principles as "promises ought to be
kept" with statutes making specific and dated requirements. (See
the first quotation from Baier above.) In fact, both law and morality
contain requirements that depend for their existence on the appro-
priate person's sayso, on convention, or on combinations of acts
that need not have taken place. Likewise, both law and morality

17 On the preferences people may have with respect to a social convention,

and the sense in which they may be trapped by it, see David Lewis, Convention
(Cambridge, Mass.: Harvard, 1969), pp. 88-96.

This content downloaded from 14.139.209.67 on Thu, 21 Apr 2016 17:02:08 UTC
All use subject to http://about.jstor.org/terms
332 THE JOURNAL OF PHILOSOPHY

contain principles that are not adventitious in any of these ways.

To contemporary philosophers, this latter fact is as inobvious
about the law as it is obvious about morality.
II

I shall conclude by making more explicit the conception of legal

obligation which I think emerges from the discussion so far. The
alternative view of legal obligation that I think can be defended is
this. To be under a legal obligation is to be under a requirement
that is-in a somewhat broad but hardly unnatural sense-a moral
obligation. (Of course, the converse-that moral obligations are
also legal obligations-clearly does not hold.) I would maintain
that a legal obligation is-as foreshadowed earlier in this paper-an
obligation that outside parties are prepared to enforce in a regular
way, using publicly available procedures to determine the fact of
violation and the way violations will be handled. The paradigm is
the obligation arising from a promise. This is surely a moral obli-
gation, but it becomes a legal obligation as well when courts
indicate that they stand ready to enforce obligations of this kind.
In the remainder of this paper I shall give a partial defense of this
conclusion. I wish to point out some obstacles standing in the way
of this notion of legal obligation and suggest how those obstacles
might be surmounted.
A central problem is that the great body of legal obligations under
which a citizen labors is only partially composed of obligations that
he has created or otherwise brought on himself. Contracts are the
clearest case of self-imposed obligation, and here we have the least
difficulty in saying that the obligation is moral as well as legal. Like-
wise, tort law-at least where liability is based on fault-is rela-
tively easy to handle: One who has, through his own negligence,
injured another has some moral obligation to pay or otherwise help
the injured; this obligation hardens into a legal obligation when
courts become prepared to enforce such payment. But what about
the enormous body of obligations that are simply imposed on the
citizen by the legislature or by the courts acting legislatively?
Here the difficulty is not quite so great as it might appear. Take
the easiest case first. A small community with a participatory
democracy freely and unanimously elects a legislature. Fair and
reasonable laws are enacted. These laws succeed in imposing both
moral and legal obligations. But of course some political com-
munities are large, are not pure participatory democracies, do not
have unanimously elected legislatures, and do not afford each
citizen a genuine opportunity to consent to the constitutional ar-

This content downloaded from 14.139.209.67 on Thu, 21 Apr 2016 17:02:08 UTC
All use subject to http://about.jstor.org/terms
 MORAL AND LEGAL OBLIGATION 333

rangement; yet it seems absurd to say that in such cases the legisla-
ture cannot succeed in putting all the citizens under a legal obliga-
tion. Here a more sophisticated social-contract theory can come
to the rescue. I pointed out earlier that Rawlsian contract theory in
an important sense confers on the legislature of a just society the
power to impose moral duties on citizens, even though those citizens
may have in no way actually consented to the particular law or to
the constitution. Indeed, Hart argues in a similar vein that "when
a number of persons conduct any joint enterprise according to rules
and thus restrict their liberty, those who have submitted to these re-
strictions when required have a right to a similar submission from
those who have benefited by their submission" 18 Here I do not
wish to defend any of these theories. I maintain only that, for the
typical case of a citizen living under a reasonably just regime, it is
not hard to find the flavor of moral obligation necessary-on the
account I would defend-to the claim that he is under a legal
obligation. Notice also that to say that one is under a moral obli-
gation to do X is not to say that, all things considered, he (morally)
ought to do X. Thus to say that the legislature can by fiat impose
a moral obligation on a citizen is not to say that disobedience is
never morally justified.
What about the hapless person who lives under a dictatorial
regime and (as an extreme example) was earlier kidnapped and
carried off to live as a virtual slave under that regime? The legis-
lature now makes certain requirements of him, backs them up with
the threat of penalties and all the machinery of a legal system. It
may seem strange to say that he is not under a legal obligation;
but he is not, in spite of the fact that the legal system may indeed
succeed in imposing legal obligations on other persons. What
strangeness there may be in denying that a legal obligation exists
in such pathological cases is outweighed, I think, by the fact that we
can now give a coherent and natural account of the intimate
connection between the legal and moral obligations that arise from
promises as well as of the role of principles in the law and in
morality.
CONRAD D. JOHNSON

The University of Maryland

18 "Are There Any Natural Rights?" in A. Quinton, ed., Political Philosophy

(New York: Oxford, 1967), p. 61.

This content downloaded from 14.139.209.67 on Thu, 21 Apr 2016 17:02:08 UTC
All use subject to http://about.jstor.org/terms