Oxford Journal of Legal Studies, Vol. 22, No. 3 (2002), pp.

489–495

The Necessary Connection between

Law and Morality
TONY HONORÉ

Abstract—If positivism is interpreted as requiring that nothing is law that does not
conform to socially accepted criteria, it is inconsistent with positive law. This is
because law purports to be morally in order. Hence it is always possible to argue
against a certain interpretation of the law that it is morally indefensible and there is
always a certain pressure within a legal system to render it morally defensible. In
that way critical morality necessarily becomes a persuasive source of law.

On a plausible interpretation of legal positivism, lawyers should reject it as

inconsistent with positive law.1
Positivism has stood historically for a number of different, though related,
ideas. They revolve round two problems. How is law to be identified? How does
it relate to morality? On identifying law the positivist thesis is that, as Raz puts
it, what is or is not law is a matter of social fact.2 To this thesis there are two
aspects. First, there are in every society criteria (for example accepted rules of
recognition) according to which certain items, such as constitutionally enacted
legislation, count as law. These criteria mark off what is legally authoritative
from what is not. Second, nothing that does not satisfy these socially accepted
criteria is law. Nothing is law unless it is recognized as authoritative in the society
in which it is claimed to be law.
The positivist thesis about the relation of law to morality follows from the
thesis about the identification of law. Nothing that is law according to the
accepted criteria for identifying law is disqualified from being law merely because
it is morally indefensible. In an extreme form, which Raz dubs the ‘semantic
thesis’, terms such as ‘right’, ‘duty’ and ‘obligation’ do not mean the same thing
in law and morals. A law-giver who prescribes duties or confers rights does not
even purport to prescribe morally binding duties or to confer morally defensible
rights.
In its less extreme form, however, this thesis about the relation of law to
morality is easy enough to defend. There can be evil laws and corrupt legal

1
The argument in this paper against a certain version of positivism is an argument about the nature of law but
does not imply a commitment to any particular natural law theory.
2
‘Legal Positivism and the Sources of Law’ in The Authority of Law. Essays on Law and Morality (1979) at 37.
 2002 Oxford University Press
490 Oxford Journal of Legal Studies VOL. 22

systems.3 The Third Reich had a legal system. Apartheid South Africa had a
legal system. It is true that not only these societies but every society has some
unjust laws. These laws and legal systems are imperfect because, unless rights
and duties mean something different in law and morality, laws and legal systems
claim to be morally in order. But, though imperfect, the unjust laws and corrupt
systems are real laws and legal systems. It is not plausible to say, with some old-
fashioned natural lawyers, that the morally objectionable laws and legal systems
are not laws and legal systems. Their supporters and opponents alike treat the
unjust laws as laws and the corrupt systems as legal systems. Law is a human
construct and what is treated by humans as law is law. To deny that objectionable
laws and systems are law is to express abhorrence of those laws or legal systems.
It is to stress that the laws or systems fail to impose the morally binding duties
that they claim to impose. But it is to tell a lie.
On the other hand the more extreme thesis about the relation between law
and morality—the thesis that right, duty and obligation mean something different
in law and morality—cannot be defended. Hart, it is true, thought that rights,
duties and obligations meant something different in law and morals, the moral
meaning being in some sense stronger than the legal one. No doubt from the
citizen’s point of view what the law requires is merely one consideration among
others in deciding what to do. But that is not because the law means the duties
it imposes to be treated as merely providing one consideration among others.
Theorists, including Raz, have rightly criticized Hart’s opinion,4 which stems
from a mistake about constitutional authority. The authority to make law is
futile unless it includes the authority to say what people should do all things
considered, not merely what they should do if their conduct is to be guided by
law. It would be self-defeating for a law-maker (legislator, judge or official),
when laying down duties or conferring rights, to say, ‘This is what you should
do if you want to abide by the law’ rather than ‘This is what you should do, full
stop’. The citizen may view compliance with the requirement as conditional on
his or her wishing to abide by the law, but the law-maker must at least purport
to pre-empt any reasons the citizen may have for not obeying the law. This is
not because the law-maker usually attaches a sanction to disregard of the law,
though the threat of sanction, if it exists, will be an additional reason for
conforming to law. It is rather that the law-maker must claim the authority to
lay down what is morally required of the citizen despite any reasons that appear
to conflict with what the law requires. The law-maker may be mistaken—may
even know that what is required is not morally defensible. But the claim that it
is defensible is an ineluctable feature of law making.
The law-maker’s claim that what is prescribed is morally binding may take
one of two forms. One is that what the law requires is something morally required
3
e.g. H. L. A. Hart, The Concept of Law (2nd edn, 1994) at 268: ‘morally iniquitous provisions may be valid
as legal rules or principles’. The reason Hart gives for this conclusion, viz. ‘that there are no necessary conceptual
connections between the content of law and morality’ is open to objection unless ‘conceptual’ is given a very
narrow meaning.
4
Raz, above n 2 at 39.
AUTUMN 2002 Necessary Connection Between Law and Morality 491
even apart from law, like supporting one’s family. The other is that what is
required is morally defensible apart from law, like paying VAT, and morally
binding because prescribed by law. For law will sometimes make morally binding
what was not binding apart from its being so required.5 To understand the
meaning of a law requires us, then, to attend to its normative aspect—what
Kelsen called its normative meaning, its oughtness.6
Apart from this constitutional point, it would be strange in the light of history
if terms like as ‘right’, ‘duty’ and ‘obligation’ had different meanings in law and
morality. These terms are derived from Roman law (ius, debitum, obligatio), the
system that first conceived law as a balance of rights and obligations subsisting
between individuals and between them and the state. From this legal context
the terms were extended by analogy to moral discourse, so that moral philosophers
now speak of ‘moral duties’, ‘moral obligations’ and ‘moral rights’. It would be
strange if this change of context carried with it a change of meaning, and what
is more a change from a weaker legal to a stronger moral meaning.
The difference between the legal and moral duties and rights is not a difference
of meaning. It is a difference between formal, institutionally recognized duties
and rights and their informal, non-institutional equivalents. The use of these
notions outside the law implies that we ought to treat certain actions and interests
rather as if they formed part of the institutional apparatus that makes up the
law, though they do not. If it is said that a father has a moral obligation to
support the mother of his child this means that, though perhaps not legally
bound to do so, he should regard himself somewhat as if he were legally bound
to support her. In need not be implied that the moral right or obligation should
be converted into a legal one. It remains an open question whether it would be
good policy to convert the moral obligation to support the child’s mother into
a legal one. Moral rights and obligations can stand on their own feet; but the
meaning of the terms ‘moral right’ and ‘moral obligation’ is parasitic on the
legal model from which they are historically derived.
That law-makers claim to impose morally binding requirements has an im-
portant consequence. It opens the law, when it comes to be interpreted or
applied, to challenge on moral grounds. It makes it part of the notion of law,
not something external to law, that it is open to moral criticism. One needs to
keep an apparent paradox in mind. A law may be law though morally indefensible.
But when the interpretation or application or continued existence of a law is in
issue it is always open to a litigant or citizen or legislator to argue that it should
be interpreted and applied in a way that is morally defensible or amended in
such a way that it becomes morally defensible. For brevity I shall use the term
‘interpretation’ but this needs to be understood in a very wide sense. The moral
challenge to law may take, and often has taken, the form of inviting a magistrate
such as the Roman praetor or the English Chancellor to intervene in order to
5
T. Honoré ‘The Dependence of Morality on Law’, OJLS 13 (1993) 1–17.
6
Introduction to the Problems of Legal Theory (Reine Rechtslehre, 1934) translated by B.L and S.L. Paulson
(1992) 32–5.
492 Oxford Journal of Legal Studies VOL. 22

render the application of the law morally defensible. If the law claims to be
morally in good order the moral argument, whether addressed to the judicial or
executive or legislative branch of the state, must always be available.
But what does the law’s claim to being morally in good order amount to?
Does it mean only that the interpretation of the law should be consistent with
the morality current in the society in question (‘positive morality’)? If so, the
positivist thesis that law is a matter of social fact is not affected, since it is a
matter of social fact what the morality currently practised or endorsed by a
society consists in. Raz, who supports the social fact thesis, accepts that this
theory may be consistent with there being a necessary connection between law
and current morality.7 Perhaps. But what if the law-maker’s meaning, properly
understood, is that what the law prescribes really is a moral duty, not merely
that one that fits the morality current in the society in question?
There is a distinction between two arguments concerned with morality that
can be raised when the interpretation of a law is in issue. One is that a proposed
interpretation would offend current morality and so cause unrest or be difficult
to enforce. That argument rests on the likely consequences of adopting a
particular interpretation. The other is that a proposed interpretation would be
morally indefensible, irrespective of whether it would cause unrest or be difficult
to enforce. The first argument appeals to social facts, or apprehended facts. The
second does not.
But what does it appeal to? Presumably to ‘critical’ as opposed to current
morality. At any rate it is not enough, with Dworkin, to say that the appeal is
to the best moral justification of the source-based rules and decisions, because
there may be no moral justification for some, perhaps even for many of these.
But what is the content of critical morality in this context? It cannot refer to the
philosophically correct theory of human behaviour, utilitarian, Kantian, Catholic
or whatever it may be, since there neither is nor is likely to be agreement on
what this is. But the question admits of a rational answer in the context in which
it arises. The critical or rational morality to which law-makers implicitly appeal
will comprise certain values inherent in law irrespective of the particular legal
system concerned. Law seeks to supplant unregulated violence, both within
societies and between them, by fostering the values of co-operation and peaceful
co-existence.8 The critical morality to which the law-maker implicitly appeals
relates to these values, and to any more specific values consistent with them that
the society in question may embrace. That the society embraces them does not
make the appeal to them a social fact about that society. They are relevant to
legal argument not because the society embraces them, or some of them (it may
reject others) but because they tend to promote co-operation and peaceful co-
existence.

7
Above n 1 at 39. He speaks of ‘popular’ morality but that may suggest too down-market a version of morality.
‘Positive morality’ is meant to include that practised or endorsed by the high-minded.
8
T. Honoré, Making Law Bind (1985) 106–14
AUTUMN 2002 Necessary Connection Between Law and Morality 493
If this is correct, the claim of law to be morally in good order puts in question
the social thesis, according to which what law is or is not is (wholly) a matter
of social fact. Whether something is law is rather partly, indeed largely, a matter
of social fact but partly a matter of what is morally defensible.
The argument for the existence of a necessary connection between law and
morality does not depend on the society in question possessing a constitution
that makes reference to moral criteria, for example in a bill of rights. In some
societies the list of authoritative sources includes some that incorporate moral
considerations. In that case the moral considerations are positive in the sense of
being formally recognized in the society as relevant to whether something is or
is not law.9 The necessary connection here asserted is however independent of
whether constitutional provisions of this sort are incorporated in the sources of
law recognized in a particular society. It exists by virtue of the fact that whatever
is identified as law according to criteria such as rules of recognition requires to
be interpreted and applied, not that it has any particular moral content. If moral
provisions are written into, say, a bill of rights these provisions themselves require
interpretation, which includes interpretation from the point of view of critical
morality. A person who under a bill of rights has a right to a fair trial has a right
to a trial that is objectively fair, not merely to one that fits the current morality
of the society in question. But even if there is no constitutional right to a fair
trial, the trial procedure prescribed by law falls to be interpreted in the light of
critical morality, of which fairness forms a part.
The connection here proposed is a connection between morality and present,
not future law. A person’s conduct, if subject to legal challenge, is judged by
law as it comes later to be interpreted by a court or executive official. The future
moral input into interpretation is therefore, from the citizens’ point of view, part
of the law at the time he or she acted, though from the point of view of a legal
historian it was not yet part of the law. But for the positivist thesis that what is
law is a matter of social fact it would be question-begging to confine law to
official determinations of the law, since the law by which citizens are judged
includes determinations that were not made when they did what they did. Given
that citizens are exposed to retrospective interpretations of the law governing
their conduct, the moral input into interpretation is some protection to those
whose seek to be guided by moral considerations.10
It is true that the necessary connection with morality here proposed is a
connection with morality as a persuasive, not an authoritative source of law. For
the argument that a certain interpretation of the law will be morally untenable,
even if correct, cannot be guaranteed to convince a court or official. Or it may
convince the court or official without being decisive since other considerations
of, say, the wording, systematic fit or purpose of a law stand in the way of giving
9
e.g. W. J. Waluchow, Inclusive Legal Positivism (1994).
10
On this point I disagree with the point of view expressed by John Eekelaar in his article in this issue. From
the citizens’ point of law the moral input is part of the law both because this is the way in which citizens normally
think of law, and because they are right to do so in view of the fact that the moral input may turn out to affect
their rights and obligations.
494 Oxford Journal of Legal Studies VOL. 22

effect to the morally best interpretation. The interpreter has to take account not
only of moral but of linguistic, systematic and teleological arguments. Which
interpretations are possible given the wording of the law? Which best fits the
rest of the system and the purposes of the provision in question? In the thinking
of some theorists these considerations would be arranged in lexical order, so
that one of them (let us say the moral considerations) would systematically
prevail over the others.11 In the real world this is not the case, and there are
good reasons, outside the scope of the present essay, why it should not be. The
necessary connection of law with morality does not therefore guarantee that law
will be morally reputable, merely that there will be pressure to make it morally
reputable. In what way, then, is the connection a necessary one? Does it enhance
the claims of law to our respect?
The connection between law and critical morality is necessary in that it is
not contingent. It applies to every law and every legal system. The proposed
interpretation of every law in every legal system can legally be challenged on the
ground that it is not morally defensible, whether the challenge succeeds or fails
in a particular instance.
It is also a connection that does, in my view, enhance the claims of law to our
respect. The positivist doctrine is something of a prison. How is the law to
escape from received ideas, including received ideas about morality? It is because
of the necessary connection with morality that judges, officials, writers and
teachers whose function it is to interpret the law have been able over the centuries
to mould it so as to take account of moral considerations. The same connection,
arising from the claim of law to be morally in order, has served to inspire
reformers and promote law reform. Hence the values of fairness, equity, justice,
honesty, humanity, dignity, prudence, abstention from violence and a host of
other values that conduce to co-operation and co-existence play a prominent
role in the law even when they are not incorporated in any formal source of law.
The interpreter’s mandate is to apply the law as an item that purports to be
morally in order. The requirement that a judge should do justice according to
law expresses this mandate, though obliquely.
I have argued that the view here advocated derives from positive law, since
the positive law of societies with legal systems, unlike the theory of positivism,
makes arguments addressed to critical morality admissible in the interpretation
and application of law. But is this true only of the law of certain societies? It is,
in the sense that it applies fully only to those societies in which the interpretation
of law is a serious intellectual discipline, so that a legal culture exists which in
which there are recognized experts, often professionally organized. It would not
apply to a society, if there has even been one, in which law was regarded merely
as a system of threats. Nor would it apply to one what was not institutionally
advanced enough to possess a legal culture. But it applies to any society in which
there exists, in the words of the Roman writer Pomponius, ‘unwritten law that

11
One could read R. Dworkin, Law’s Empire (1986) in this sense.
AUTUMN 2002 Necessary Connection Between Law and Morality 495
consists in interpretation.’12 In any such society there is an inbuilt pressure
towards improving the law morally and in other ways. It is what in Soviet jargon
was called a ‘permanently operative factor’. Lawyers can take some comfort
from the perception that they are committed to improving the law not by virtue
of their individual merits but by the nature of the institution to which they are
committed.

12
Digest 1.2.2.12. ‘Unwritten’ in the sense of not being embodied in a set text.