4 Law As The Union of Primary and Secondary Rules

CHAPTER FOUR:
LAW AS THE UNION OF PRIMARY AND SECONDARY RULES
H.L.A. Hart’s theory represents the present state of legal positivism today. This does
not mean that no improvements have been made since his theory was first
formulated and presented to the public in a complete form in 1961. It is simply that
these improvements have been minor. They do not pose a challenge to his system
and in fact can be incorporated within it. The content of the theory remains in
accordance with Hart’s basic original intuition that law comprises a union of primary
and secondary rules.
I. METHODOLOGY
Hart’s methodology can be divided into a theory of definition and a theory of law.
He greatly developed both topics and his methodology, just like his version of legal
positivism, remains the dominant method of analytic philosophy today.
A. Theory of Definition
Hart first introduced his theory of definitions in his inaugural lecture as Holder of the
Chair of Jurisprudence at Oxford in his article entitled: “Definition and Theory in
Jurisprudence.” He formulated the problem of his article thus:
“Questions such as those I have mentioned, ‘What is a State?’, ‘What is

the law?’, ‘What is a right?’, have great ambiguity. The same form of
words may be used to demand a definition or the cause or the purpose
or the justification or the origin of a legal or political institution. But if,
in the effort to free them from this risk of confusion with other
questions, we rephrase these requests for definitions as ‘What is the
meaning of the word “State”?’, ‘What is the meaning of the word
“right”?’, those who ask are apt to feel uneasy, as if this had trivialized
their question. For what they want cannot be got out of a dictionary,
and this transformation of their question suggests it can. This
uneasiness is the expression of an instinct which deserves respect: it
emphasizes the fact that those who ask these questions are not asking
to be taught how to use words in the correct way. This they know and
yet are still puzzled. Hence it is no answer to this type of question
merely to tender examples of what are correctly called rights, laws, or
corporate bodies, and to tell the questioner if he is still puzzled that he
is free to abandon the public convention and use words as he pleases.
For the puzzle arises from the fact that though the common use of
these words is known, it is not understood; and it is not understood
because compared with most ordinary words these legal words are in
different ways anomalous. Sometimes, as with the word ‘law’ itself,
one anomaly is that the range of case to which it is applied has a
diversity which baffles the initial attempt to extract any principle
behind the application, yet we have the conviction that even here
there is some principle and not an arbitrary convention underlying the
surface differences; so that whereas it would be patently absurd to ask
for elucidation of the principle in accordance, with which different men
69
are called Tom, it is not felt absurd to ask why, within municipal law,
the immense variety of different types of rules are called law, nor why
municipal law and international law, in spite of striking differences, are
so called.”1
Just as Kelsen, Hart was wary of political bias infecting his analysis of law. He called
this mistaken methodology, the anomaly of “theory growing on the back of
definitions.” Consequently he strove for neutrality.
“Hence, though theory is to be welcomed, the growth of theory on the

back of definition is not. Theories so grown, indeed, represent valuable
efforts to account for many puzzling things in law; and among these is
the great anomaly of legal language—our inability to define its crucial
words in terms of ordinary factual counterparts. But here I think they
largely fail because their method of attack commits them all, in spite of
their mutual hostility, to a form of answer that can only distort the
distinctive characteristics of legal language.” 2
His definition of a ‘legal right’ is therefore as follows:
“I would, therefore, tender the following as an elucidation of the

expression ‘a legal right’:
(1) A statement of the form ‘X has a right’ is true if the following

conditions are satisfied:
(a) There is in existence a legal system.

(b) Under a rule or rules of the system some other person Y is, in
the events which have happened, obliged to do or abstain from some
action.
(c) This obligation is made by law dependent on the choice
either of X or some other person authorized to act on his behalf so tahe
either Y is bound to do or abstain from some action only if X (or some
authorized person) chooses otherwise.
(2) A statement of the form ‘X has a right’ is used to draw a conclusion

of law in a particular case which falls under such rules.” 3
B. Legal Theory
Hart’s methodology may be considered as a combination of conceptual analysis and

descriptive sociology. Indeed his book, The Concept of Law provides an elucidation
of these methods in his attempt to analyze the concept “law”.
“My aim in this book has been to further the understanding of law,
coercion, and morality as different but related social phenomena.
1
H.L.A. Hart, “Definition and Theory in Jurisprudence,” Essays in Jurisprudence and Philosophy, Oxford:
Clarendon Press, 1983, pp. 21-22.
2
Id., pp. 25-26.
3
Id., p. 35.
70
Though it is primarily designed for the student of jurisprudence, I hope
it may also be of use to those whose chief interests are in moral and
political philosophy, or in sociology, rather than in law. The lawyer will
regard the book as an essay in analytical jurisprudence, for it is
concerned with the clarification of the general framework of legal
thought, rather than with the criticism of law or legal policy. Moreover,
at many points, I have raised questions which may well be said to be
about the meaning of words. Thus I have considered: how ‘being
obliged’ differs from ‘having an obligation’: how the statement that a
rule is a valid rule of law differs from a prediction of the behaviour of
officials; what is meant by the assertion that a social group observes a
rule and how this differs from and resembles the assertion that its
members habitually do certain things. Indeed, one of the central
themes of the book is that neither law nor any other form of social
structure can be understood without an appreciation of certain crucial
distinctions between two different kinds of statements, which I have
called ‘internal’ and ‘external’ and which can both be made whenever
social rules are observed.
Notwithstanding its concern with analysis the book may also be

regarded as an essay in descriptive sociology; for the suggestion that
inquiries into the meaning of words merely throw light on words is
false. Many important distinctions, which are not immediately obvious,
between types of social situation or relationships may best be brought
to light by an examination of the standard uses of the relevant
expressions and of the way in which these depend on a social context,
itself often left unstated. In this field of study it is particularly true that
we may use, as Professor J. L. Austin said, ‘a sharpened awareness of
words to sharpen our perception of the phenomena.” 4
He aims to answer the question “What is law?,” noting that the question gives rise
to a number of persistent perplexities.
“Few questions concerning human society have been asked with such
persistence and answered by serious thinkers in so many diverse,
strange, and even paradoxical ways as the question ‘What is law?’
Even if we confine our attention to the legal theory of the past 150
years and neglect classical and medieval speculation about the
‘nature’ of law, we s hall find a situation not paralleled in any other
subject systematically studies as a separate academic discipline. No
vast literature is dedicated to answering the questions ‘What is
chemistry?” or “What is medicine?’, as it is to the question ‘What is
law?’ A few lines on the opening page of an elementary textbook is all
that the student of these sciences is asked to consider; and the
answers he is given are of a very different kind from those tendered to
the student of law. No one has thought it illuminating or important to
insist that medicine is ‘what doctors do about illnesses’, or ‘a
prediction of what doctors will do’, or to declare that what is ordinarily
recognized as a characteristic, central part of chemistry at all. Yet, in
4
H.L.A. Hart, The Concept of Law, Oxford: Clarendon Press, 1961, “Preface,” p. v.
71
the case of law, things which at first sight look as strange as these
have often been said, and not only said but urged with eloquence and
passion, as if they were revelations of truths about law, long obscured
by gross misrepresentations of its essential nature.” 5
These perplexities persist despite the fact that most people know what the law
means and can identify a law very readily and unproblematically. It concerns a
deeper, more disturbing unease and disconcert.
“When we reflect on the quite general ability of people to recognize

and cite examples of laws and on how much is generally known about
a standard case of a legal system, it might seem that we could easily
put an end to this persistent question, ‘What is law?’, simply by issuing
a series of reminders of what is already familiar. Why should we not
just repeat the skeleton account of salient features of a municipal legal
system which, perhaps optimistically, we put . . . into the mouth of an
educated man? We can then simply say, ‘Such and such is the
standard case of what is meant by “law” and “legal system”;
remember that besides these standard cases you will also find
arrangements in social life which, while sharing some of these salient
features, also lack others of them. These are disputed cases where
there can be no conclusive arguments for or against their
classification.’
Such a way with the question would be agreeably short. But it would
have nothing else to recommend it. For . . . it is clear that those who
are most perplexed by the question ‘What is law?’ have not forgotten
and need no reminder of the familiar facts which this skeleton answer
offers them. The deep perplexity which has kept alive the question, is
not ignorance or forgetfulness or inability to recognize the phenomena
to which the word ‘law’ commonly refers.” 6
There are recurrent themes regarding the question of the nature of law, and it is the
resolution to these themes which provides the key to the answer to the question.
“For there are certain recurrent main themes which have formed a
constant focus of argument and counter-argument about the nature of
law, and provoked exaggerated and paradoxical assertions about law
such as those we have already cited. Speculation about the nature of
law has a long and complicated history; yet in retrospect it is apparent
that it has centered almost continuously upon a few principal issues.
These were not gratuitously chosen or invented for the pleasure of
academic discussion; they concern aspects of law which seem
naturally, at all times, to give rise to misunderstanding, so that
confusion and a consequent need of greater clarity about them may
coexist even in the minds of thoughtful men with a firm mastery of a
knowledge of law.”7
5
Id., p. 1.
6
Id., pp. 4-5.
7
Id., pp. 5-6.
72
The first issue is connected with the concept of obligation.
“The most prominent general feature of law at all times and places is
that its existence means that certain kinds of human conduct are no
longer optional but in some sense obligatory. Yet this apparently
simple characteristic of law is not in fact a simple one; for within the
sphere of non-optional obligatory conduct we can distinguish different
forms. The first, simplest sense in which conduct is no longer optional,
is when one man is forced to do what another tells him, not because he
is physically compelled in the sense that his body is pushed or pulled
about, but because the other threatens him with unpleasant
consequences if he refuses. The gunman orders his victim to hand
over his purse and threatens to shoot if he refuses; if the victim
complies we refer to the way in which he was forced to do so by saying
that he was obliged to do so. To some it has seemed clear that in this
situation where one person gives another an order backed by threats,
and in this sense of ‘oblige’, obliges him to comply wit, we have the
essence of law, or at least ‘the key to the science of jurisprudence’.
This is the starting-point of Austin’s analysis by which so much English
jurisprudence has been influenced.
There is of course no doubt that a legal system often presents this

aspect among others. A penal statute declaring certain conduct to be
an offence and specifying the punishment to which the offender is
liable, may appear to be the gunman situation writ large; and the only
difference to be the relatively minor one, that in the case of statutes,
the orders are addressed generally to a group which customarily obeys
such orders. But attractive as this reduction of the complex
phenomena of law to this simple element may seem, it has been
found, when examined closely, to be a distortion and a source of
confusion even in the case of a penal statute where an analysis in
these simple terms seems most plausible. How then do law and legal
obligation differ from, and how they are related to, orders backed by
threats? This at all times has been one cardinal issue latent in the
question ‘What is law?’.”8
The next recurring issue is also connected with the issue of obligation.
“Moral rules impose obligations and withdraw certain areas of conduct

from the free option of the individual to do as he likes. Just as a legal
system obviously contains elements closely connected with the simple
case of orders backed by threats, so equally obviously it contains
elements closely connected with certain aspects of morality. In both
cases alike there is a difficulty in identifying precisely the relationship
and a temptation to see in the obviously close connection an identity.
Not only do law and morals share a vocabulary so that there are both
legal and moral obligations, duties, and rights; but all municipal legal
systems reproduce the substance of certain fundamental moral
8
Id., pp. 6-7; italics Hart’s.
73
requirements. Killing and the wanton use of violence are not only the
most obvious examples of the coincidence between the prohibitions of
law and morals. Further, there is one idea, that of justice which seems
to unite both fields: it is both a virtue specially appropriate to law and
the most legal of the virtues. We think and talk of ‘justice according to
law’ and yet also of the justice and injustice of the laws.
These facts suggest the view that law is best understood as a ‘branch’
of morality or justice and that its congruence with the principles of
morality or justice rather than its incorporation of orders and threats is
of its ‘essence’. This is the doctrine characteristic not only of
scholastic theories of natural law but of some contemporary legal
theory which is critical of the legal ‘positivism’ inherited from Austin.
Yet here again theories that make this close assimilation of law to
morality seem, in the end, often to confuse one kind of obligatory
conduct with one another, and to leave insufficient room for
differences in kind between legal and moral rules and for divergences
in their requirements. These are at least as important as the similarity
and convergence which we may also find. So the assertion that ‘an
unjust law is not a law’ has the same ring of exaggeration and paradox,
if not falsity, as ‘statutes are not laws’ or ‘constitutional law is not law.’
It is characteristic of the oscillation between extremes, which make up
the history of legal theory, that those who have seen in the close
assimilation of law and morals nothing more than a mistaken inference
from the fact that law and morals share a common vocabulary of rights
and duties, should have protested against it in terms equally
exaggerated and paradoxical. ‘The prophecies of what the courts will
do in fact, and nothing more pretentious, are what I mean by the
law.’”9
The third recurrent issue concerns the law’s connection with rules.
“The third main issue perennially prompting the question ‘What is law?’
is a more general one. At first sight it might seem that the statement
that a legal system consists, in general at any rate, of rules could
hardly be doubted or found difficult to understand. But those who
have found the key to the understanding of law in the notion of orders
backed by threats, and those who have found it in its relation to
morality or justice, alike speak of law as containing, if not consisting
largely of, rules. Yet dissatisfaction, confusion, and uncertainty
concerning this seemingly unproblematic notion underlies much of the
perplexity about the nature of law. What are rules? What does it mean
to say that a rule exists? Do courts really apply rules or merely
pretend to do so? Once the notion is queried, as it has been especially
in the jurisprudence of this century, major divergencies in opinion
appear.”10
There are then, to Hart, three recurrent issues of law.

9
10
Id. p. 8; italics Hart’s.
74
“Here then are the three recurrent issues: How does law differ from
and how is it related to orders backed by threats? How does legal
obligation differ from, and how is it related to, moral obligation? What
are rules and to what extent is law an affair of rules? To dispel doubt
and perplexity on these three issues has been the chief aim of most
speculation about the ‘nature’ of law. It is possible now to see why this
speculation has usually been conceived as a search for the definition of
law, and also why the familiar forms of definition have done so little to
resolve the persistent difficulties and doubts. Definition, as the word
suggests, is primarily a matter of drawing lines or distinguishing
between one kind of thing and another, which language marks off by a
separate word. The need for such a drawing of lines is often felt by
those who are perfectly at home with the day-to-day use of the word in
question, but cannot state or explain the distinctions which, they
sense, divide one kind of thing from another. All of us are sometimes
in this predicament: it is fundamentally that of the man who says, ‘I
can recognize an elephant when I see one but I cannot define it.’ The
same predicament was expressed by some famous words of St.
Augustine about the notion of time. ‘What then is time? If no one asks
me I know: if I wish to explain it to one that asks I know not.’ It is in
this way that even skilled lawyers have felt that, though they know the
law, there is much about law and its relation to other things they
cannot explain and do not fully understand. Like a man who can get
from one point to another in a familiar town but cannot explain or show
others how to do it, those who press for a definition need a map
exhibiting clearly the relationships dimly felt to exist between the law
they know and other things.”11
But it is not a definition a legal theorist is looking for.
“There are of course many kinds of definition besides the very simple
traditional form which we have discussed, but it seems clear, when we
recall the character of the three main issues which we have identified
as underlying the recurrent question ‘What is law?’, that nothing
concise enough to be recognized as a definition could provide a
satisfactory answer to it. The underlying issues are too different from
each other and too fundamental to be capable of this sort of resolution.
This the history of attempts to provide concise definitions has shown.
Yet the instinct which has often brought these three questions together
under a single question or request for definition has not been
misguided; for, as we shall show in the course of this book, it is
possible to isolate and characterize a central set of elements which
form a common part of the answer to all three.”12
If not a definition, what then is the theorist after?
11
Id., pp. 13-14.
12
Id., p. 16.
75
“For its purpose is not to provide a definition of law, in the sense of a
rule by reference to which the correctness of the use of a word can be
tested; it is to advance legal theory by providing an improved analysis
of the distinctive structure of a municipal legal system and a better
understanding of the resemblances and differences between law,
coercion, and morality, as types of social phenomena.” 13
Consequently, I now turn to the substantive aspects of Hart’s theory, wherein he

attempts to answer the question ‘What is law?’ by means of resolving the three
recurrent issues of law satisfactorily. To the substance of his work therefore, I now
proceed.
II. SOCIAL RULES
Rules provide the focus of Hart’s theory. In explaining a social rule, he first
distinguishes a rule from a habit. In other words, he is concerned to explain when a
habit becomes a rule. For example, members of a particular community engage in
the habit of shaking hands as a form of greeting. Nobody told them to do so.
Perhaps one person did it to another and the practice caught on. At that point, the
practice was only a habit, and not yet a rule. It becomes a rule when it has
acquired a certain degree of importance that the members feel it to be an obligation
to shake hands upon meeting and consider people rude who refuse to do it. Prior to
that point, it was merely a practice or custom that most people complied with; in
short, merely a habit. There was no feeling or sense of obligation that they had to
shake hands upon meeting people.
From this insight, Hart distinguishes a habit from a rule in three important ways.
“First, for the group to have a habit it is enough that their behavior in
fact converges. Deviation from the regular course need not be a
matter for any form of criticism. But such general convergence or
even identity of behaviour is not enough to constitute the existence of
a rule requiring that behaviour: where there is such a rule deviations
are generally regarded as lapses or faults open to criticism, and
threatened deviations meet with the pressure for conformity, though
the forms of criticism and pressure differ with different types of rule.
Secondly, where there are such rules, not only is such criticism in fact
made but deviation from the standard is generally accepted as a good
reason for making it. Criticism deviation is regarded as legitimate or
justified in this sense, as are demands for compliance with the
standard when a deviation is threatened. Moreover, except by a
minority of hardened offenders, such criticism and demands are
generally regarded as legitimate, or made with good reason, both by
those who make them and those to whom they are made. How many
of the group must in these various ways treat the regular mode of
behavior as a standard of criticism, and how often and for how long
they must do so to warrant the statement that the group has a rule,
are not definite matters; they need not worry us more than the
13
Id., pp. 16-17.
76
question as to the number of hairs a man may have and still be bald.
We need only remember that the statement that a group has a certain
rule is compatible with the existence of a minority who not only break
the rule but refuse to look upon it as a standard either for themselves
or others.
The third feature distinguishing social rules from habits is implicit in

what has already been said, but it is one so important and so
frequently disregarded or misrepresented in jurisprudence that we
shall elaborate it here. It is a feature which throughout this book we
shall call the internal aspect of rules. When a habit is general in a
social group, this generality is merely a fact about the observable
behaviour of most of the group. In order that there should be such a
habit no members of the group need in any way think of the general
behaviour, or even know that the behaviour in question is general; still
less need they strive to teach or intend to maintain it. It is enough that
each for his part behaves in the way that others also in fact do. By
contrast, if a social rule is to exist some at least must look upon the
behaviour in question as a general standard to be followed by the
group as a whole. A social rule has an ‘internal’ aspect, in addition to
the external aspect which it shares with a social habit and which
consists in the regular uniform behaviour which an observer could
record.”14
Therefore, there exist three characteristics in order for a social rule to exist. First,
there must exist such a convergence of behavior so that deviations from the
standard of behavior call not only for criticism among those who engage in the
practice against those who do not behave in accordance with the way the majority
behaves, but also calls for conformity against the deviators. Secondly, appeal to
the standard is a good reason for behaving in accordance with the standard. Finally,
there must exist what Hart calls the internal aspect.
Hart elaborates on this internal aspect. It is not simply a matter of feelings, of

feeling bound, for example. Rather it is a critical reflective attitude:
“The internal aspect of rules is often misrepresented as a mere matter

of ‘feelings’ in contrast to externally observable physical behaviour. No
doubt, where rules are generally accepted by a social group and
generally supported by social criticism and pressure for conformity,
individuals may often have psychological experiences analogous to
those of restriction or compulsion. When they say they ‘feel bound’ to
behave in a certain ways they may indeed refer to these experiences.
But such feelings are neither necessary nor sufficient for the existence
of ‘binding’ rules. There is no contradiction in saying that people
accept certain rules but experience no such feelings of compulsion.
What is necessary is that there should be a critical reflective attitude to
certain patterns of behaviour as a common standard and that this
should display itself in criticism (including self-criticism), demands for
conformity, and in acknowledgment that such criticism and demands
14
Id., pp. 54-55.
77
are justified, all of which find their characteristic expression in the
normative terminology of ‘ought’, ‘must’, and ‘should’, ‘right’ and
‘wrong’.”15
By means of the internal aspect of rules, Hart introduces a method of scientific

inquiry, now referred to as hermeneutic, which adopts an internal attitude towards
the phenomenon or practice the scientist or theorist is investigating.
“This is the context for analysing Hart’s concept of the internal aspect
of rules. The idea is that one cannot understand a social system
unless one understands how the people who created the system or
who participate in the system perceive it. This ‘hermeneutic’ approach
—that is, giving priority to trying to understand how other people
perceive their situation is always in tension with those who want social
theory to be more scientific.
The ‘scientific’ approach in social theory would rely only on data that
was ‘objective’, data on which different observers would always agree.
The ‘scientific’ approach to legal theory might be exemplified in
various theorists’ writings: for example, Christopher Columbus
Langdell’s view of legal theory as the search for the system of basic
principles within the law, and the American legal realists (to some
extent reacting to Langdell’s views) emphasizing what judges ‘actually
do’ as contrasted with what they are saying that they are doing. Hart
also specifically mentioned the work of the Scandinavian Legal Realist
Alf Ross, who (according to Hart) ‘claimed that the only method of
representation of the law fit to figure in a modern rational science of
law was one which shared the structure and logic of statements of
empirical science.”
Hart’s argument is that whatever advantage a ‘scientific’ approach

might have, it simply is not adequate for a full understanding of law.
Law is a social institution to set up to achieve certain human purposes,
and also to give guidance to citizens. One can only understand
purposive behaviour and normative (rule-following) behaviour if one
leaves one’s spectator’s perspective and tries to understand the
perceptions of the participants in the system, that is, the perceptions
of the people who are following the rules, and who perceive
themselves as doing so. In Hart’s terms, to understand ‘any form of
normative social structure’, ‘the methodology of the empirical sciences
is useless; what is needed is a ‘hermeneutic’ method which involves
portraying rule-governed behaviour as it appears to the participants.” 16
III. THE KEY TO THE SCIENCE OF JURISPRUDENCE
Hart argued that without the idea of a rule, it was hopeless to elucidate even the
most elementary forms of law. He then explained two kinds of rules:
15
Id., p. 56.
16
Brian Bix, Jurisprudence: Theory and Context, London: Sweet and Maxwell, 2003, p. 40.
78
“It is true that the idea of a rule is by no means a simple one: we have
already seen . . . the need, if we are to do justice to the complexity of a
legal system, to discriminate between two different though related
types. Under rules of the one type, which may well be considered the
basic or primary type, human beings are required to do or abstain from
certain actions, whether they wish to or not. Rules of the other type
are in a sense parasitic upon or secondary to the first; for they provide
that human beings may by doing or saying certain things introduce
new rules of the primary type, extinguish or modify old ones, or in
various ways determine their incidence or control their operations.
Rules of the first type impose duties; rules of the second type confer
powers, public or private. Rules of the first type concern actions
involving physical movements or changes; rules of the second type
provide for operations which lead not merely to physical movement or
change, but to the creation or variation of duties or obligations.” 17
It is Hart’s contention that “in the combination of these two types of rule there lies
what Austin wrongly claimed to have found in the notion of coercive orders, namely,
‘the key to the science of jurisprudence.’” 18
He then conducted a thought experiment in order to find out the key elements to
law or a legal system. He imagined a society without a legislature, courts or
officials of any kind. He referred to it as a social structure with only primary rules of
obligation.
“If a society is to live by primary rules alone, there are certain

conditions which, granted a few of the most obvious truisms about
human nature and the world we live in, must clearly be satisfied. The
first of these conditions is that the rules must contain some form of
restrictions on the free use of violence, theft, and deception to which
human beings are tempted but which they must, in general, repress, if
they are to coexist in close proximity to each other. Such rules are in
fact always found in primitive societies of which we have knowledge,
together with a variety of others imposing on individuals various
positive duties to perform services or make contributions to the
common life. Secondly, though such a society may exhibit the tension,
already described, beyond those who accept the rules and those who
reject the rules except where fear of social pressure induces them to
conform, it is plain that the latter cannot be more than a minority, if so
loosely organized a society of persons, approximately equal in physical
strength, is to endure: for otherwise those who reject the rules would
too little social pressure to fear. This too is confirmed by what we know
of primitive communities where, though there are dissidents and
malefactors, the majority live by the rules seen from the internal point
of view.”19
17
Id., pp. 78-79.
18
Id., p. 79.
19
Id., p. 89.
79
Such a simple society will suffer from certain defects and will require
supplementation in various ways. The first of these defects is the lack of a rule of
recognition.
“In the first place, the rules by which the group lives will not form a
system, but will simply be a set of separate standards, without any
identifying or common mark, except of course that they are the rules
which a particular group of human beings accepts. They will in this
respect resemble our own rules of etiquette. Hence if doubts arise as
to what the rules are or as to the precise scope of some given rule,
there will be no procedure for settling this doubt, either by reference to
some authoritative text or to an official whose declarations on this
point are authoritative. For, plainly, such a procedure and the
acknowledgement of either authoritative text or persons involve the
existence of rules of a type different from the rules of obligation or
duty which ex hypothesi are all that the group has. This defect in the
simple social structure of primary rules we may call its uncertainty.”20
There is a second defect, that of staticity.
“A second defect is the static character of the rules. The only mode of
change in the rules known to such a society will be the slow process of
growth, whereby courses of conduct once thought optional become
first habitual or usual, and then obligatory, and the converse process of
decay, where deviation once severely dealt with, are first tolerated and
then pass unnoticed. There will be no means, in such a society, of
deliberately adopting the rules to changing circumstances, either by
eliminating old rules or introducing new ones: for, again the possibility
of doing this presupposes the existence of rules of a different type from
the primary rules of obligation by which alone society lives. In an
extreme case the rules may static in a more drastic sense. This,
though never perhaps fully realized in any actual community, is worth
considering because the remedy for it is something very characteristic
of law. Each individual would simply have fixed obligations or duties to
do or abstain from doing certain things. It might indeed be very often
the case that others would benefit from the performance of these
obligations yet if there are only primary rules of obligation they would
have no power to release those bound from performance or to transfer
to others the benefits which would accrue from performance. For such
operations of release or transfer create changes in the initial positions
of individuals under the primary rules of obligation, and for these
operations to be possible there must rules of a sort different from the
primary rules.”21
The third defect is that of inefficiency.
“The third defect of this simple form of social life is the inefficiency.
Disputes as to whether an admitted rule has or has not been violated
20
Id., p. 90; italics Hart’s.
21
Id., pp. 90-91; italic’s Hart’s.
80
will always occur and will, in any but the smallest societies, continue
interminably, if there is no agency specially empowered to ascertain
finally, and authoritatively, the fact of violation. Lack of such final and
authoritative determinations is to be distinguished from another
weakness associated with it. This is the fact that punishments for
violations of the rules, and other forms of social pressure involving
physical effort or the use of force, are not administered by a special
agency but are left to the individuals affected or to the group at large.
It is obvious that the waste of time involved in the group’s unorganized
efforts to catch and punish offenders, and the smoldering vendettas
which may result from self help in the absence of an official monopoly
of ‘sanctions’, may be serious. The history of law does, however,
strongly suggest that the lack of official agencies to determine
authoritatively the fact of violation of the rules is a much more serious
defect; for many societies have remedies for this defect long before
the other.”22
The remedy to these defects lies in supplementing the primary rules of obligation
with secondary rules.
“(Secondary) rules have important features in common and are

connected in various ways. Thus they may all be said to be on a
different level from the primary rules, for they are all about such rules;
in the sense that while primary rules are concerned with the actions
that individuals must or must not do, these secondary rules are all
concerned with the primary rules themselves. They specify the ways
in which the primary rules may be conclusively ascertained,
introduced, eliminated, varied, nd the fact of their violation
conclusively determined.”23
It constitutes a step from the pre-legal into the legal world. “The introduction of the
remedy for each defect might, in itself, be considered a step from the pre-legal into
the legal world; since each remedy brings with it many elements that permeate law:
certainly all three remedies together are enough to convert the regime of primary
rules into what is indisputably a legal system.” 24
To the defect of uncertainty, Hart supplements the simple social structure of primary
rules with the rule of recognition.
“The simplest form of remedy for the uncertainty of the regime of

primary rules is the introduction of what we shall call a ‘rule of
recognition’. This will specify some feature or features possession of
which by a suggested rule is taken as a conclusive affirmative
indication that it is a rule of the group to be supported by the social
pressure it exerts. The existence of such a rule may take any of a huge
variety of forms, simple or complex. It may, as in the early law of
many societies, by no more than an authoritative list or text of the
22
Id., p. 91, italics Hart’s.
23
24
81
rules is to be found in a written document or carved on some public
monument. No doubt as a matter of history this step from the pre-
legal to the legal may be accomplished in distinguishable stages, of
which the first is the mere reduction to writing of hitherto unwritten
rules. This is not itself the crucial step, though it is a very important
one: what is crucial is the acknowledgement of reference to the writing
or inscription as authoritative, i.e., as the proper way of disposing of
doubts as to the existence of the rule. Where there is such an
acknowledgement there is a very simple form of a secondary rule: a
rule for conclusive identification of the primary rules of obligation.
In a developed legal system the rules of recognition are of course more

complex; instead of identifying rules exclusively by reference to a text
or list they do so by reference to some general characteristic
possessed by the primary rules. This may be the fact of their having
been enacted by a specific body, or their long customary practice, or
their relation to judicial decisions. Moreover, where more than one of
such general characteristics are treated as identifying criteria,
provision may be made for their possible conflict by their arrangement
in an order of superiority, as by the common subordination of custom
or precedent to statute, the latter being a ‘superior source’ of law.
Such complexity may make the rules of recognition in a modern legal
system seem very different from the simple acceptance of an
authoritative text: yet even in this simplest form, such a rule brings
with it many elements distinctive of law. By providing an authoritative
mark it introduces, although in embryonic form, the idea of a legal
system: for the rules are now not just a discrete unconnected set but
are, in a simple way, unified. Further, in the simple operation of
identifying a given rule as possessing the required feature of being an
item on an authoritative list of rules we have the germ of the idea of
legal validity.”25
The second remedy is that of rules of change.
The remedy for the static quality of the regime of primary rules
consists in the introduction of what we shall call ‘rules of change’. The
simplest form of such a rule is that which empowers an individual or
body of persons to introduce new primary rules for the conduct of the
life of the group, or of some class within it, and to eliminate old rules.
As we have already argued . . . it is in terms of such a rule, and not in
terms of orders backed by threats, that the ideas of legislative
enactment and repeal are to be understood. Such rules of change may
be very simple or very complex: the powers conferred may be
unrestricted or limited in various ways: and the rules may, besides
specifying the persons who are to legislate, define in more or less rigid
terms the procedure to be followed in legislation. Plainly, there will be
a very close connexion between the rules of change and rules of
recognition: for where the former exists the latter will necessarily
incorporate a reference to legislation as an identifying feature of the
25
82
rules, though it need not refer to all the details of procedure involved in
legislation. Usually some official certificate or official copy will, under
the rules of recognition, be taken as a sufficient proof of due
enactment. Of course if there is a social structure so simple that the
only ‘source of law’ is legislation, the rule of recognition will simply
specify enactment as the unique identifying mark or criterion of
validity of the rules. This will be the case for example in the imaginary
kingdom of Rex I depicted in Chapter IV: there the rule of recognition
would simply be that whatever Rex I enacts is law.” 26
To the third defect of inefficiency, there correspond the rules of adjudication as a

remedy.
“The third supplement to the simple regime of primary rules, intended

to remedy the inefficiency of the social pressure, consists of secondary
rules empowering individuals to make authoritative determinations of
the question whether on a particular occasion, a primary rule has been
broken. The minimal form of adjudication consists in such
determinations, and we shall call the secondary rules which confer the
power to make them ‘rules of adjudication’. Besides identifying the
individuals who are to adjudicate, such rules will also define the
procedure to be followed. Like the other secondary rules these are on
a different level from the primary rules: though they may be reinforced
by further rules imposing duties on judges to adjudicate, they do not
impose duties but confer judicial powers and a special status on
judicial declarations about the breach of obligations. Besides these
resemblances to the other secondary rules, rules of adjudication have
intimate connexion with them. Indeed, a system which has rules of
adjudication is necessarily also committee to a rule of recognition of an
elementary and imperfect sort. This is so because, if courts are
empowered to make authoritative determinations of the fact that a rule
has been broken, these cannot avoid being taken as authoritative
determinations of what the rules are. So the rule which confers
jurisdiction will also be a rule of recognition, identifying the primary
rules through the judgments of the courts and these judgments will
become a ‘source’ of law. It is true that this form of rule of recognition,
inseparable from the minimum form of jurisdiction, will be very
imperfect. Unlike an authoritative text or a statute book, judgments
may not be couched in general terms and their use as authoritative
guides to the rules depends on a somewhat shaky inference from
particular decisions, and the reliability of this must fluctuate both with
the skill of the interpreter and the consistency of the judges.” 27
This union of primary and secondary rules comprises both the heart of a legal
system such that Hart called it the “key to the science of jurisprudence,” but also a
powerful conceptual tool by which Hart will proceed to analyze the concepts of law
and of legal systems.
26
Id., p. 93; italics Hart’s.
27
Id., pp. 94-95, italics Hart’s.
83
“If we stand back and consider the structure which has resulted from
the combination of primary rules of obligation with the secondary rules
of recognition, change, and adjudication, it is plain that we have here
not only the heart of a legal system, but a most powerful tool for the
analysis of much that has puzzled both the jurist and the political
theorist.
Not only are the specifically legal concepts with which the lawyer is
professionally concerned, such as those of obligation and rights,
validity and source of law, legislation and jurisdiction, and sanction,
best elucidated in terms of this combination of elements. The
concepts (which bestride both law and political theory) of the state, of
authority, and of an official require a similar analysis if the obscurity
which still lingers about them to be dissipated. The reason why an
analysis in these terms of primary and secondary rules has this
explanatory power is not far to seek. Most of the obscurities and
distortions surrounding legal and political concepts arise from the fact
that these essentially involve reference to what we have called the
internal point of view: the view of those who do not merely record and
predict behaviour conforming to rules, but use the rules as standards
for the appraisal of their own and others’ behaviour. This requires
more detailed attention in the analysis of legal and political concepts
than it has usually received. Under the simple regime of primary rules
the internal point of view is manifested in its simplest form, in the use
of those rules as the basis of criticism, and as the justification of
demands for conformity, social pressure, and punishment. Reference
to this most elementary manifestation of the internal point of view is
required for the analysis of the basic concepts of obligation and duty.
With the addition to the system of secondary rules, the range of what
is said and done from the external point of view is much extended and
diversified. With this extension comes a whole set of new concepts
and they demand a reference to the internal point of view for their
analysis. These include the notions of legislation, jurisdiction, validity
and, generally, of legal powers, private and public. There is a constant
pull towards an analysis of these in the terms of ordinary or ‘scientific’,
fact-stating or predictive discourse. But this can only reproduce their
external aspect: to do justice to their distinctive, internal aspect we
need to see the different ways in which the law-making operations of
the legislator, the adjudication of a court, the exercise of private or
official powers, and other ‘acts-in-the-law’ are related to secondary
rules.”28
IV. THE RULE OF RECOGNITION
Hart’s rule of recognition plays the same role as the basic norm in Kelsen’s coercive
order does. It provides the criteria by which the validity of other rules of the system
is assessed. At this stage, the distinction between the ultimacy of the rule of
recognition and the supremacy of one of its criteria deserves attention. Hart has
this to say about the latter: “We may say that a criterion of legal validity or source
28
84
of law is supreme if rules identified by reference to it are still recognized as rules of
the system, even if they conflict with rules identified by reference to the other
criteria, whereas rules identified by reference to the latter are not so recognized if
they conflict with the rules identified by reference to the same criterion.” 29
This is how Hart defines the ultimacy of a rule of recognition.
“The sense in which the rule of recognition is the ultimate rule of a

system is best understood if we pursue a very familiar chain of legal
reasoning. If the question is raised whether some suggested rule is
legally valid, we must, in order to answer the question, use a criterion
of validity provided by some other rule. Is this purported by-law of the
Oxfordshire County Council valid? Yes: because it was made in the
exercise of the powers conferred, and in accordance with the
procedure specified, by a statutory order made by the Minister of
Health. At this first stage the statutory order provides the criteria in
terms of which the validity of the by-law is assessed. There may be no
practical need to go farther; but there is a standing possibility of doing
so. We may query the validity of the statutory order and assess its
validity in terms of the statute empowering the minister to make such
orders. Finally when the validity of the statute has been queried and
assessed by reference to the rule that what the Queen in Parliament
enacts is law, we are brought to a stop in inquiries concerning validity:
for we have reached a rule which, like the intermediate statutory order
and statute, provides criteria for the assessment of the validity of other
rules; but it is also unlike them in that there is no rule providing criteria
for the assessment of its own legal validity.”30
Moreover, the rule of recognition is not merely presupposed, as Kelsen’s basic norm
is, but its validity is assumed because its existence can be shown as a matter of
fact. This was explained in the previous chapter. In such a manner is its validity
and existence distinguished from the other rules or laws in the legal system. “In
this respect, however, as in others a rule of recognition is unlike other rules of the
system. The assertion that it exists can only be an external statement of fact. For
whereas a subordinate rule of a system may be valid and in that sense ‘exist’ even
if it is generally disregarded, the rule of recognition exists only as a complex, but
normally concordant, practice of the courts, officials, and private persons in
identifying the law by referenced to certain criteria. Its existence is a matter of
fact.”31
Hart also distinguishes efficacy from validity in a manner bearing a great similarity
with that of Kelsen.
“Some of the puzzles connected with the idea of legal validity are said
to concern the relation between the validity and the ‘efficacy’ of law. If
by ‘efficacy’ is meant that the fact that a rule of law which requires
certain behaviour is obeyed more often than not, it is plain that there is
29
Id., p. 103.
30
Id., pp. 103-104, italics Hart’s.
31
Id., pp. 107.
85
no necessary connexion between the validity of any particular rule and
its efficacy, unless the rule of recognition of the system includes
among its criteria, as some do, the provision (sometimes referred to as
a rule of obsolescence) that no rule is to count as a rule of the system
if it has long ceased to be efficacious.
From the inefficacy of a given rule, which may or may not count
against its validity, we must distinguish a general disregard of the rules
of the system. This may be so complete in character and so protracted
that we should say, in the case of a new system, that it had never
established itself as the legal system of a given group, or, in the case
of a once-established system, that it had ceased to be the legal system
of the group. In either case, the normal context or background for
making any internal statement in terms of the rules of the system is
absent. In such cases it would be generally pointless either to assess
the rights and duties of particular persons by reference to the primary
rules of a system or to assess the validity of any of its rules by
reference to its rules of recognition. To insist on applying a system of
rules which had either never actually been effective or had been
discarded would, except in special circumstances mentioned below, be
as futile as to assess the progress of a game by reference to a scoring
rule which had never been accepted or had been discarded.
One who makes an internal statement concerning the validity of a

particular rule of the system may be said to presuppose the truth of
the external statement of fact that the system is generally efficacious.
For the normal use of internal statements is in such a context of
general efficacy. It would however be wrong to say that statements of
validity ‘mean’ that the system is generally efficacious. For though it is
normally pointless or idle to talk of the validity of a rule of a system
which has never established itself or has been discarded, none the less
it is not meaningless nor is it always pointless. One vivid way of
teaching Roman Law is to speak as if the system were efficacious still
and to discuss the validity of particular rules and solve problems in
their terms; and one way of nursing hopes for the restoration of an old
social order destroyed by revolution, and rejecting the new, is to cling
to the criteria of the old regime. This is implicitly done by the White
Russian who still claims property under some rule of descent which
was a valid rule of Tsarist Russia.” 32
Moreover, given the centrality and importance of the rule of recognition, the officials
of the legal system must exhibit the appropriate attitude towards it.
“(The rule of recognition), if it is to exist at all, must be regarded from

the internal point of view as a public, common standard of correct
judicial decision, and not as something which each judge merely obeys
for his part only. Individual courts of the system though they may, on
occasion, deviate from these rules must, in general, be critically
concerned with such deviations as lapses from standards, which are
32
86
essentially common or public. This is not merely a matter of the
efficiency or health of the legal system, but is logically a necessary
condition of our ability to speak of the existence of a single legal
system. If only some judges acted ‘for their part only’ on the footing
that what the Queen in Parliament enacts is law, and made no criticism
of those who did not respect this rule of recognition, the characteristic
unity and continuity of a legal system would have disappeared. For
this depends on the acceptance, at this crucial point, of common
standards of legal validity. In the interval between these vagaries of
judicial behaviour and the chaos which would ultimately ensue when
the ordinary man was faced with contrary judicial orders, we would be
at a lost to describe the situation. We would be in the presence of a
lusus naturae worth thinking about only because our awareness of
what is often too obvious to be noticed.”33
On the other hand, the citizens need not exhibit such an attitude. Most of them
must merely display what Austin called a ‘habit of obedience’ towards the laws of
the legal system.
“Here (in a complex modern state) surely the reality of the situation is
that a great proportion of the ordinary citizens—perhaps a majority—
have no general conception of the legal structure of its criteria of
validity. The law which he obeys is something which he knows of only
as ‘the law’. He may obey it for a variety of different reasons and
among them may often, though not always, be the knowledge that it
will be best for him to do so. He will be aware of the general likely
consequences of disobedience: that there are officials who may arrest
him and others who will try him and send him to prison for breaking
the law. So long as the laws which are valid by the system’s tests of
validity are obeyed by the bulk of the population this surely is all the
evidence we need in order to establish that a given legal system
exists.”34
This brings me to discuss the two necessary and sufficient conditions for the
existence of a legal system.
“There are therefore two minimum conditions necessary and sufficient

for the existence of a legal system. On the one hand those rules of
behaviour which are valid according to the system’s criteria of validity
must be generally obeyed, and, on the other hand, its rules of
recognition specifying the criteria of validity and its rules of change
and adjudication must be effectively accepted as common public
standards of official behaviour by its officials. The first condition is the
only one which private citizens need satisfy: they may obey each ‘for
his part only’ and from any motive whatsoever; though in a healthy
society they will in fact often accept these rules as common standards
of behaviour and acknowledge an obligation to obey them, or even
trace this obligation to a more general obligation to respect the
33
Id., pp. 112-113.
34
Id., pp. 111.
87
constitution. The second condition must also be satisfied by the
officials of the system. They must regard these as common standards
of official behaviour and appraise critically their own and each other’s
deviations as lapses. Of course it is also true that besides these there
will be many primary rules which apply to officials in their merely
personal capacity which they need only to obey.
The assertion that a legal system exists is therefore a Janus-faced

statement looking both towards obedience by ordinary citizens and to
the acceptance by the officials of the secondary rules as critical
common standards of official behaviour. We need not be surprised at
this duality. It is merely the reflection of the composite character of a
legal system as compared with a simpler decentralized pre-legal form
of social structure which consist only of primary rules. In the simpler
structure, since there are no officials, the rules must be widely
accepted as setting critical standards for the behaviour of the group.
If, here, the internap point of view is not widely disseminated there
could not logically be any rules. But where there is a union of primary
and secondary rules, which is, as we have argued, the most fruitful
way of regarding a legal system, the acceptance of the rules as
commons standards for the group may be split off from the relatively
passive matter of the ordinary individual acquiescing in the rules by
obeying them for his part alone. In an extreme case the internal point
of view with its characteristic normative use of legal language (‘This is
a valid rule’) might be confused to the official world. In this more
complex system, only officials might accept and use the system’s
criteria of legal validity. The society in which this was so might be
deplorably sheeplike; the sheep might end in the slaughter-house. But
there is little reason for thinking that it could not exist or for denying it
the title of the legal system.”35
V. COMMENTS AND CRITICISM
Hart’s theory of law can be criticized on four points. The first concerns his
distinction between primary and secondary rules. The next two are external
criticisms due to Dworkin, who challenged Hart’s claim regarding the existence of a
rule of recognition and maintained that his theory was beset by what he called the
semantic sting. Finally, Raz’ internal criticism faulted his theory for lacking
sufficient purity in positivist values by advocating what was later called an
‘inclusive’ version of positivism, or soft positivism, in contrast to his ‘exclusive’
positivism or hard positivism.
A. The Conflation of the Distinction between Primary and Secondary Rules
Hart’s distinction between primary and secondary rules can be made more precise
as he appears to conflate three different distinctions.
“Hart lays great emphasis on another distinction among rules. In

Chapter V he distinguished primary from secondary rules, and believes
35
88
that this distinction furnishes him with ‘the key to the science of
jurisprudence’. But the distinction is uncharacteristically unclear. It
seems to be a conflation of at least three different distinctions. There
is a distinction between rules imposing duties and rules conferring
powers; there is a distinction between simple legal rules and somewhat
more sophisticated meta-rules; and there is the ghost of a positivist
distinction between rules concerning actions involving physical
movements or changes, and those which lead to the creation of duties
or obligations.”36
1. The distinction between rules concerning actions involving physical movements

and those which create duties
It is the social aspect of the rule rather than its physical component which render
rules relevant and significant to law and legal theory.
Laws seldom characterize actions by reference to mere physical

movement, but construe them in a social context and often with
reference to the agent’s intention. Stumbling into you differs from
banging into you only in that I didn’t mean to: if I swat a horse-fly that
is about to feast on your sunbathing thigh, I have not assaulted you,
although I may have hurt you as much as if I had been administering
corporal punishment: and physical actions which would be quite illegal
if performed by me on you, would be entirely lawful if done by a
policeman on a suspected criminal resisting arrest. Or, to take a very
different case, my obligation to pay taxes derives from a rule Hart
would reckon as primary, since it requires men to do certain actions
whether they wish to or not, and yet is characteristically discharged by
my signing a cheque, which varies the duties and obligations of a
banker. It is a mistake to try to peel off the social or legal
characterization of actions from a basic description in behaviouristic
terms. Most of our actions are social actions, undertaken for social
reasons in a social context, with social consequences and often endued
with a social significance.”37
2. The distinction between rules which impose duties and rules which confer
powers
Rules which impose duties and rules which confer powers are not so much distinct
as correlative. Hart also mistakenly implies that one is more fundamental than
another, which is not necessarily the case.
“The distinction between rules which impose duties and rules which
confer powers likewise softens under scrutiny. Not that it is a useless
distinction—it remains a valuable tool for analytic jurisprudence—but it
is neither as fundamental nor as directed as Hart supposes. In spite of
the inadequacies he perceives in Austin’s analysis, and the ways in
36
J.R. Lucas, “The Phenomenon of Law,” Law, Morality, and Society, ed. by P.M.S. Hacker and Joseph Raz, Oxford,
Clarendon Press: 1997, pp. 85-98.
37
Id., p. 87.
89
which having a legal obligation cannot be reduced to being obliged by
the threat of sanctions to act in a particular fashion. Hart tends to
assume that having an obligation is more fundamental than having a
power, so that rules imposing the former are primary, and those
imposing the latter are only the secondary. But really the rules are
correlative. The rules proscribing my driving a self-propelled vehicle on
the public highway, unless I and it are licensed, and the rules
prescribing the procedure for obtaining a driving and a vehicle license
interlock. I cannot explain the one except with reference to the other.
So, too, although less obviously, with Hart’s favourite examples of
solemnizing a marriage or making a will. The laws prohibiting rape,
adultery, seduction, and fornication, are clearly primary rules in Hart’s
view, but equally clearly presuppose an already intelligible concept of
marriage—no man, however attractive, can ever hope to seduce his
own wife. Equally, the laws against theft presuppose laws of property,
including therefore laws defining the conditions under which property
may be disposed of. If after Aunt Agatha’s funeral I walk out of her
house with a Sevre dinner service, I am stealing it unless she left it me;
and whether I had stolen it or not might turn on whether her will
leaving it me was valid or not—on whether she had signed it at the top,
instead of the bottom (cf. The Concept of Law, p. 12) or whether the
witnesses have seen her sign it only in a mirror, not being visibly
present to her (as in Dorothy Sayer’s Unnatural Death). These
examples are, I shall argue, typical. Every primary rule is correlative
with some secondary rules, and vice versa. For rules imposing duties
apply only in certain circumstances and subject to some exceptions,
and since I have control of circumstances, I can take steps to put
myself beyond the scope of the rule or within the ambit of some
exception, and thus possess the power of making it lawful for me to do
what I want to do. To take the most favourable case to Hart’s thesis,
even the rule prohibiting homicide does not apply to soldiers in time of
war, persons defending themselves against violent attack, or
executioners carrying out judicial sentence of death. Jesebel was able
to compass Naboth’s death under due form of law. Laws can be
manipulated. Even straightforward prohibitions can be read as giving
guide-lines on how the desired action may be performed without
breaking the law, and thus as conferring powers to restyle the legal
position so as to accomplish one’s purposes. Of course, in some cases
such purposes can be achieved only by devious or dishonest
manoeuvres, and the rule is correctly seen as imposing a duty rather
than conferring a power, but in other cases the emphasis is reversed.
So the distinction is valuable. But just as rules conferring powers
would have no point unless somewhere down the line a person could
by their aid bring about some alteration on the bearing on someone of
some rule imposing a duty, so rules imposing duties create also a web
of interlinked rights and powers, which they implicitly confer on various
people. Any system, whether of law or of morality or of social custom,
which imposes duties will also confer rights and powers, and he cannot
90
reasonably regard the rules imposing the former as more primary than
those conferring the latter.”38
3. The distinction between rules and meta-rules
The criticism of the distinction between rules and meta-rules focus on the rule of
recognition. It is indeed a meta-rule; however, it need not have been formulated as
explicitly and clearly as Hart did. Social circumstances provide the background in
which the rule of recognition, which need not be formulated with perfect
explicitness and clarity, can be specified and identified.
“Hart is at his unhappiest in his controversy with Professor Fuller over

the legal status of wicked edicts enacted by an iniquitous regime. If
the defining property of a law is that it should satisfy the rule of
recognition of a regime whose laws are generally enforced, then the
decrees of the Nazis were valid law, and the only question left is
whether it should none the less be disobeyed. But this is to abstract
too much and put on the rule of recognition more weight than it can
properly bear. It is, after all, only a meta-rule, and, as Hart himself
points out, not always fully formulated, and sometimes itself an object
of controversy. The rule of recognition need be neither explicit nor
clear. It was only very gradually that it changed in England, and
shifted sovereignty from the monarch to parliament. It was not clear
for centuries what the fundamental law of the land was, but laws were
made, applied, and enforced none the less, because for the most part
king and parliament were working together rather than in opposition.
Legal disputes were much more about the scope or application of laws
rather than their validity. The rule of recognition is implicit, rather than
explicit, and rests upon a number of tacit understandings about the
way in which various functionaries will cooperate and will be guided by
public interests rather than private purposes. It may be a matter of
considerable difficulty to say exactly what the rule of recognition is. It
therefore may not be a very usable criterion of validity. Moreover,
meta-rules, however useful they may be in remedying the
inadequacies of a pre-legal regime, are ‘in some sense parasitic upon’
primary rules and cannot stand on their own. Hart makes the point
very clear by his analogy with a scorer in a game. We may, for good
reasons, have a scorer and may have the meta-rule that his decision is
final, but this meta-rule operates, and can operate, only against the
background that scoring is something which players by and large do
themselves, and that the scorer is trying to do the same thing as the
players are, and on the whole succeeding. Else the game is different,
and becomes what Hart illuminatingly describes as the game of
‘scorer’s discretion’. Exactly similar arguments apply to the rule of
recognition. It makes sense only as a development of a pre-legal
regime in which most people know most of the time what their legal
rights and duties are, and look to the civil authorities only to enforce
them if need be and to adjudicate the relatively rare cases of dispute.
Provided, but only provided, that the developed legal system satisfies
38
Id., pp. 88-89; italics Lucas’.
91
their requirements, its various meta-rules will serve a social function
and be generally accepted. But if we divorce the meta-rules from the
substantive rules of social intercourse, they no longer fulfil their role
and so cease being the rules they were. So long as rulers are trying to
do the same thing, in regulating social behaviour, as the ordinary
members of society are, and see a developed legal system as an
improved version of a pre-legal regime. But once the tacit
understandings which direct the rulers’ activities on lines congruous to
the aims of ordinary citizens are dissolved, and the sole requirement
for legal validity is that it should conform to the rule of recognition, the
game has been changed, and we are no longer faced with a legal
system, but a potentially disastrous analogue of scorer’s discretion.” 39
Hence, due to the social background renders, there is no compelling necessity that
a rule of recognition be formally and explicitly laid out in as precise and
sophisticated a manner which Hart expects.
“The distinction between rules and meta-rules is a distinction

between what is primary and what is secondary. It serves to
characterize what Hart counts as a legal, in contrast to a pre-legal,
system, but implies that the center of gravity of a legal system is not
in its meta-rules, but in the primary rules of the pre-legal system and
the social mores it embodies. If we concentrate too much on formal
meta-rules, we shall abstract too much. If we concern ourselves only
with formal criteria of legal validity, as we shall fail to see that these
arise out of the law’s fulfilling a social function, and presuppose that
the substantive law is in line with social customs and moral
principles. Moreover, even if we wanted to, we would not give an
entirely formal definition of law, for whereas rules imposing duties
and rules conferring powers may be explicitly formulated in words,
and usually become more fully formulated with the passage of time
and the growing sophistication of a legal system, meta-rules, and
especially the rule of recognition, are not, and cannot be, fully
formulated, but must always be at least in part, implicit in a diffused
recognition of what is rationally acceptable. They thus cannot
provide completely adequate criteria of legal validity, since the
crucial question—whether the meta-rules are operated in such a way
as to produce the results that are by and large rationally acceptable
—is one that no formal decision-procedure can always settle. The
concepts of law, therefore, cannot be given too tidy a definition. It
can be elucidated, but only as a social phenomenon that arises when
men, who are rational but not very rational, and moral but not very
moral, live their lives together.”40
B. The Rule of Recognition
Dworkin, in “The Model of Rules I,”41 characterized Hart’s positivism in terms of

three tenets, only the first of which is to be examined. The first tenet states:
39
Id., pp. 92-93.
40
Id., pp. 97-98, italics Lucas’.
92
“The law of a community is a set of special rules used by the
community directly or indirectly for the purpose of determining which
behavior will be punished or coerced by public power. These special
rules can be identified and distinguished by specific criteria, by tests
having to do not with their content but with their pedigree or the
manner in which they were adopted or developed. These tests of
pedigree can be used to distinguish valid legal rules from spurious
legal rules (rules which lawyers and litigants wrongly argue are rules of
law) and also from other sorts of social rules (generally lumped
together as ‘moral rules’) that the community follows but does not
enforce through public power.” 42
Dworkin engaged in a long argument in an attempt to disprove this tenet. He

provided examples of judicial opinions, Riggs v. Palmer43 and Henningsen v.
Bloomfield Motors, Inc.,44 wherein appeal to moral principles was used in a material
way to justify the decision in the case. He then argued that this was not uncommon
to judicial decision-making. These moral principles cannot be identified by means
of tests of pedigree; as such, they defy the rules of recognition. In other words,
some legal standards are part of law because of the role they play in judicial
decision-making in a manner in which they are capture-free by the rule of
recognition.
“Most rules of law, according to Hart, are valid because some

competent institution enacted them. Some were created by a
legislature, in the form of statutory enactments. Others were created
by judges who formulated them to decide particular cases, and thus
established them as precedents for the future. But this test of
pedigree will not work for the Riggs and Henningsen principles. The
origin of these as legal principles lies not in a particular decision of
some legislature or court, but in a sense of appropriateness developed
in the profession and the public over time. Their continued power
depends upon this sense of appropriateness being sustained. If it no
longer seemed unfair to allow people to profit from their wrongs, or fair
to place special burdens upon oligopolies that manufacture potentially
dangerous machines, these principles would no longer play much of a
role in new cases, even if they had never been overruled or repealed.
(Indeed, it hardly makes sense to speak of principles like these as
being ‘overruled’ or ‘repealed’. When they decline, they are eroded,
not torpedoed.)
True, if we were challenged to back up our claim that some principle is

a principle of law, we would mention any prior case in which the
principle was cited, or figured in the argument. We would also mention
any statute that seemed to exemplify that principle (even better if the
41
Ronald Dworkin, “The Model of Rules I,” Taking Rights Seriously, Cambridge: Harvard University Press, 1978,
pp. 14-45.
42
Id., p. 17.
43
115 N.Y. 506, 22 N.E. 188 (1889).
44
32 N.J. 358, 161 A. 2d 69 (1960).
93
principle was cited in the preamble of the statute, or in the committee
reports or other legislative documents that accompanied it). Unless we
could find some institutional support, we would probably fail to make
out our case, and the more support we found, the more weight we
could claim for the principle.
Yet we could not devise any formula for testing how much and what
kind of institutional support is necessary to make a principle a legal
principle, still less to fix its weight at a particular order of magnitude.
We argue for a particular principle by grappling with what a whole of
shifting, developing, and interacting standards (themselves principles
rather than rules) about institutional responsibility, statutory
interpretation, the persuasive force of various sorts of precedent, the
relation of all these to contemporary moral practices, and hosts of
other such standards. We could not bolt all of these together into a
single ‘rule’, even a complex one, and if we could the result would bear
little relation to Hart’s picture of a rule of recognition, which is the
picture of a fairly stable master rule specifying ‘some feature or
features possession of which by a suggested rule is taken as a
conclusive affirmative indication that it is a rule. . .’
Moreover, the techniques we apply in arguing for another principle do

not stand (as Hart’s rule of recognition is designed to) on an entirely
different level from the principles they support. Hart’s sharp
distinction between acceptance and validity does not hold. If we are
arguing for the principle that a man should not profit from his own
wrong, we could cite the acts of courts and legislatures that exemplify
it, but this speaks as much to the principle’s acceptance as its validity.
(It seems odd to speak of a principle as being valid at all, perhaps
because validity is an all-or-nothing concept, appropriate for rules, but
inconsistent with a principle’s dimension of weight.) If we are asked
(as we might well be) to defend the particular doctrine of precedent, or
the particular technique of statutory interpretation, that we used in this
argument, we should certainly cite the other general principles that we
believe support that practice, and this introduces a note of validity into
the chord of acceptance. We might argue, for example, that the use
we make of earlier cases and statutes is supported by a particular
analysis of the principles of democratic theory, or by a particular
position on the proper division of authority between national and local
institutions, or something else of that sort. Nor is this path of support
a one-way street leading to some ultimate principle resting on
acceptance alone. Our principles of legislation, precedent, democracy,
or federation might be challenged too; and if they were we should
argue for them, not only in terms of practice, but in terms of each
other and in terms of the implications of trends of judicial and
legislative decisions, even though this last would involve appealing to
those same doctrines of interpretation we justified through the
principles we are now trying to support. At this level of abstraction, in
other words, principles rather hang together than link together.
94
So even though principles draw support from the official acts of legal
institutions, they do not have a simple or direct enough connection
with these acts to frame the connection in terms of criteria specified by
some ultimate master rule of recognition.”45
Dworkin concluded:
“So we cannot adapt Hart’s version of positivism by modifying his rule

of recognition to embrace principles. No tests of pedigree, relating
principles to acts of legislation, can be formulated, nor can his concept
of customary law, itself an exception to the first tenet of positivism, be
made to serve without abandoning the tenet altogether. One more
possibility must be considered, however. If no rule of recognition can
provide a test for identifying principles, why not say that principles are
ultimate, and form the rule of recognition of our law? The answer to
the general question ‘What is the valid law in an American
jurisdiction?’ would then require us to state all the principles (as well as
ultimate constitutional rules) in force in that jurisdiction at the time,
together with appropriate assignments of weight. A positivist might
then regard the complete set of these standards as the rule of
recognition of the jurisdiction. This solution has the attraction of
paradox, but of course it is an unconditional surrender. If we simply
designate our rule of recognition by the phrase ‘the complete set of
principles in force’, we achieve only the tautology that law is law. If,
instead, we tried actually to list all the principles in force we would fail.
They are controversial, their weight is all important, they are
numberless, and they shift and change so fast that the start of our list
would be obsolete before we reached the middle. Even if we
succeeded, we would not have a key for law because there would be
nothing left for our key to unlock.
I conclude that if we treat principles as law we must reject the

positivists’ first tenet, that the law of a community is distinguished
from other social standards by some test in the form of a master
rule.”46
Prof. Dworkin summarizes his argument against the existence of a rule of

recognition in another article:
“I might summarize the argument I made in this way. I said that the
thesis that there exists some commonly recognized test for law is
plausible if we look only at simple legal rules of the sort that appear in
statutes or are set out in bold type in textbooks. But lawyers and
judges, in arguing and deciding lawsuits, appeals not only to such
black-letter rules, but also to other sorts of standards that I called legal
principles, like, for example, the principle that no man may profit from
his own wrong. This fact faces the positivist with the following difficult
choice. He might try to show that judges, when they appeal to
45
Dworkin, supra., n. 35, pp. 40-41.
46
Id., pp. 43-44.
95
principles of this sort, are not appealing to legal standards, but only
exercising their discretion. Or he might try to show that, contrary to
my doubts, some commonly-recognized test always does identify the
principles judges count as law, and distinguishes them from the
principles they do not. I argued that neither strategy could succeed.” 47
Prof. Joseph Raz took up the cudgels for the positivists in defending Hart against
Dworkin’s attack. He replied that principles become part of the law in roughly the
same way that rules do. Thus, if rules can be captured by some sort of master test,
so can principles.
“Legal principles, like other laws, can be enacted or repealed by

legislatures and administrative authorities. They can also become
legally binding through establishment by the courts. Many legal
systems recognize that both rules and principles can be made into law
or lose their status through precedent. Rules and principles differ in
this respect. A court can establish a new rule in a single judgment
which becomes a precedent. Principles are not made into law by a
single judgment; they evolve rather like a custom and are binding only
if they have considerable authoritative support in a line of judgments.
Like customary law, judicially adopted principles need not be
formulated very precisely in the judgments which count as authority
for their existence. All that has to be shown is that they underlie a
series of courts’ decisions, that they were in fact a reason operating in
a series of cases.”48
Dworkin formulates two arguments against the existence of a rule or recognition or

a master test in law. This is how Raz describes and responds to the first argument.
“‘Hart’s sharp distinction between acceptance and validity,’ the first

argument runs, ‘does not hold. If we are arguing for the principle that
a man should not profit from his own wrong, we could cite the acts of
courts and legislatures that exemplify it, but this speaks as much to
the principle’s acceptance as its validity. (It seems odd to speak of a
principle as being valid at all, perhaps because validity is an all-or-
nothing concept, appropriate for rules, but inconsistent with a
principle’s dimension of weight.’ The concept of validity is said to be
inconsistent with a principle’s dimension of weight on the ground that
one establishes a principle’s validity by showing that it has
‘institutional support’ but the amount of support a principle enjoys
determines its weight and is a matter of degree: ‘[T]he more support
we found, the more weight we could find from the principle.’ But this is
surely mistaken. A principle might have been referred to frequently by
the courts as binding, but have little weight. The degree of support
may sometimes be evidence for a principle’s weight, but it need not be
and the two notions are not logically related.
47
Ronald Dworkin, “The Model of Rules II,” Taking Rights Seriously, Cambridge: Harvard University Press, 1978,
p. 46.
48
Joseph Raz, “Legal Principles and the Limits of Law,” Ronald Dworkin and Contemporary Jurisprudence, ed. by
Marshall Cohen, New Jersey: Rowman and Allanheld, p. 77.
96
Legal principles may be valid in precisely the same way that rules are.
They may, for example, be enacted in a constitution or in a statute, as
some of Professor Dworkin’s own examples show. It is true, though,
that some legal principles are law because they are accepted by the
judiciary. It is, however, an important point which does necessitate a
modification of Hart’s criterion of identity. But here again Professor
Dworkin claims too much. He claims that if the master rule says
merely that whatever other rules the community accepts are legally
binding, then it fails to act as an identifying criterion distinguishing
between law and social norms. Had all social customs in all countries
been legally binding, this would have been a valid criticism. Some
countries, however, do not recognize custom as a source of law at all.
These legal systems which do regard customs as legally binding do so
only if they pass certain tests. These tests, if they are not set out in a
statute or some other law, are laid down by the rule of recognition,
which determines under what conditions social customs are binding as
law.
The rule of recognition, therefore, does serve to explain the legal

status of general community customs. It cannot, however, explain in
the same way the legal status of judicial customs. Since it is itself a
judicial custom it cannot confer any special status on other judicial
customs. Judicial rule-making, as I indicated above, differs in this
respect from the evolution of principles by the courts. A rule becomes
binding by being laid down in one case as a precedent. It does not
have to wait until it is accepted in a series of cases to be binding. It is
binding because of the doctrine of precedent which is part of our rule
of recognition. Principles evolved by the courts become binding by
becoming a judicial custom. They are part of the law because they are
accepted by the courts, not because they are valid according to the
rule of recognition.
Hart’s criterion of identity must be modified. A legal system consists

not only of one customary rule of the law-enforcing agencies and all
the laws recognized by it, but of all the customary rules and principles
of the law-enforcing agencies and all the laws recognized by them.
This is an important modification, but it preserves the fundamental
point underlying Hart’s criterion and shared by many: namely, that law
is an institutionalized normative system and that the fact that the
enforcement of its standards is a duty of special law-enforcing
agencies is one important feature which distinguishes it from many
other normative systems. The importance of this feature of law is
made manifest by distinguishing between legal and nonlegal standards
according to whether or not the courts have an obligation to apply
them, either because they are themselves judicial custom or because
judicial customs make their application obligatory.” 49
49
Id., pp. 79-80.
97
Professor Dworkin’s second argument is repeated and responded to by Raz in this
manner.
“Professor Dworkin has a second argument disputing the possibility of

formulating an adequate criterion of identity. ‘True,’ he says, ‘if we
were challenged to back our claim that some principle is a principle of
law, we would mention any prior cases in which that principle was
cited, or figured in the argument. . . . Unless we could find some such
institutional support, we would probably fail to make out our case . . . .
Yet we could not devise any formulate for testing how much and what
kind of institutional support is necessary to make a principle a legal
principle.’ In this passage Professor Dworkin is rejecting not merely
Hart’s version of the thesis of the limits of law but all versions of the
thesis. He agrees that if legal and nonlegal standards can be
distinguished, this could only be done by relying on the fact that only
legal standards have adequate institutional support in the practice of
the courts. He denies, however, the possibility of a general
explanation of what counts as adequate institutional support. It follows
that it is impossible to provide a general account of the difference
between legal and nonlegal standards and the thesis of the limits of
law must be abandoned. What is the force of the argument? If a legal
system consists, as I have suggested, of those standards which the
courts are bound to recognize, we must agree with Professor Dworkin
that we need a general explanation of what counts as adequate
institutional support. For laws are binding on the courts either because
judicial customs make their recognition obligatory or because they are
themselves judicial customs. Thus the acceptability of the thesis of the
limits of law depends on our ability to explain the concept of a judicial
custom. But judicial customs are but a special case of social customs.
What we need is an adequate explanation of the concept of a

customary norm. Once we have it we will know what judicial custom is
and will have a complete criterion of identity. Hart has provided such
an explanation. No doubt it is possible to improve on it, but there is no
reason to suppose that the concept of a customary norm defies
analysis. It is true that an analysis of the concept does not give us a
decision procedure determining for every principle or rule whether or
not it has sufficient support to be regarded as a judicial custom.
Borderline cases will remain; they must remain for customary norms
evolve gradually. But Dworkin’s is a very weak argument, which
rejects a distinction because it admits the existence of borderline
cases.”50
To put the same point in another way, the law has sources. That the law has
sources is a result of its social and institutional character. It is a truism that law tells
men how to behave or consists of standards that guide their actions. And it is part
of the function of adjudicative institutions to pick out these standards in applying
the law. This is as true for primitive societies as well as for sophisticated ones. In
primitive societies, the courts in resolving a dispute may conclude that these
50
Id., pp. 80-81.
98
standards are to be found in custom or a shared morality. In more sophisticated
ones where greater emphasis is placed on spelling out these standards for clearer
and more effective guidance, the courts turn to the pronouncements of law-making
and law-applying institutions, like themselves, for his guidance. Hence, however
primitive or sophisticated the law (so long as society is governed by it), the courts
know where to look to find law or know how to identify the sources from which the
legal standards applicable to the case are derived.
The sources are identified in terms of the practice of the courts. An outsider looking
in at the court's behavior can observe a judicial practice or custom. From an
external perspective therefore, a rule or rules can be formulated describing it. From
the internal perspective on the other hand, the attitude of the courts that the
practice ought to be complied with can be discerned. The ingredients that
transform a practice into a binding rule are thus in place. 51 Some sort of rule of
recognition can be formulated,52 one which may be used to identify by means of
social fact the sources with which the law's existence is determined and content
identified, as well as for determining the law's existence and for identifying the law's
content.
C. The Semantic Sting
The second criticism of Dworkin is more general in character. It challenges the very
methodology of legal positivism and claims that it falls prey to the semantic sting.
The semantic sting is a knockdown argument against all legal methodologies which
avail of semantic theories of law. Dworkin insists that semantic theories, by its
insistence on shared criteria of meaning, are unable to account for theoretical
disagreement in law.
Dworkin first introduced certain terminology in order to construct his argument. He

initially defined a proposition of law as “all the various statements and claims
people make about what the law allows or prohibits or entitles them to have.” 53
He then distinguished between empirical and theoretical disagreements about law.
“Now we can distinguish two ways in which lawyers and judges might
agree about the truth of a proposition of law. They might agree about
the grounds of law—about when the truth or falsity of other, more
familiar propositions makes a particular proposition of law true or false
—but disagree about whether those grounds are in fact satisfied in a
particular case. Lawyers and judges might agree, for example, that
the speed limit is 55 in California if the official California statute book
contains a law to that effect, but disagree about whether that is the
speed limit because they disagree about whether, in fact, the book
does contain such a law. We might call this an empirical disagreement
about law. Or they might disagree about the grounds of law, about
which other kinds of propositions, when true, make a particular
proposition of law true. They might agree, in the empirical way, about
51
See Hart, The Concept of Law, 79-88.
52
Ibid., 92
53
Ronald Dworkin, Law’s Empire, London: Fontana Press, p. 4.
99
what the statutes and past judicial decisions have to say about
compensation for fellow-servant injuries, but disagree about what the
law of compensation actually is because they disagree about whether
statute books and judicial decisions exhaust the pertinent grounds of
law. We might call that a ‘theoretical’ disagreement about law.” 54
Next, he maintained that legal positivists advocated or subscribed to a plain-fact

view of law, which, incidentally, he also provided a definition for.
“(Legal philosophers) say that theoretical disagreement is an illusion,

that lawyers and judges all actually agree about the grounds of law. I
shall call this the ‘plain fact’ view of the grounds of law; here is a
preliminary statement of its main claims. The law is only a matter of
what legal institutions, like legislatures and city councils and courts,
have decided in the past. If some body of that sort has decided that
workmen can recover compensation for injuries by fellow workmen,
then that is law. So questions of law can always be answered by
looking in the books where the records of institutional decisions are
kept. Of course it takes special training to know where to look and how
to understand the arcane vocabulary in which the decisions are
written. The layman does not have this training or vocabulary, but
lawyers do, and it therefore cannot be controversial among them
whether the law allows compensation for fellow-servant injuries, for
example, unless some of them have made an empirical mistake about
what actually was decided in the past. ‘Law exists as a plain fact, in
other words, and what the law is in no way depends on what it should
be. Why then do lawyers and judges sometimes appear to be having a
theoretical disagreement about the law? Because when they appear to
be disagreeing in the theoretical way about what the law is, they are
really disagreeing about what it should be. Their disagreement is
really over issues of morality and fidelity.” 55
Dworkin then went on to provide counter-examples to this view in the form of actual
judicial cases which tended to show that the arguments in these cases seemed to
be about law and not about morality or fidelity or repair. According to him, the
surprising response of some legal philosophers, presumably positivists, to these
counter-examples was to claim that “theoretical disagreement about the grounds of
law must be a pretense because the very meaning of the word ‘law’ makes law
depend on specific criteria, and that any lawyer who rejected or challenged those
criteria would be speaking self-contradictory nonsense.” 56
These specific criteria are in the form of shared rules: “We follow shared rules, they
say, in using any word: these rules set out criteria that supply the word’s meaning.
Our rules for using ‘law’ tie law to plain historical fact. It does not follow that all
lawyers are aware of these rules in the sense of being able to state them in some
54
Id., pp. 4-5.
55
Id., p. 7.
56
Id., p. 31.
100
crisp and comprehensive form. For we all follow rules given by our common
language of which we are fully aware.” 57
In other words, these positivists subscribed to a semantic theory. “Semantic

theories suppose that lawyers and judges use mainly the same criteria (though
these are hidden and unrecognized) in deciding when propositions of law are true or
false; they suppose that lawyers actually agree about the grounds of law. These
theories disagree about which criteria lawyers do share and which grounds these
criteria do stipulate.”58
Dworkin then went on to argue that the ‘pretense’ or ‘fingers-crossed’ argument

does not hold water. The pretense argument goes like this: “(J)udges pretend to be
disagreeing about what the law is because the public believes there is always law
and that judges should always follow it. On this view lawyers and judges
systematically connive to keep the truth from the people so as not to disillusion
them or arouse their ignorant anger.” 59
Then he proceeds with its refutation. If “lawyers all agree there is no decisive law in
cases like our sample cases, then why has this view not become part of our popular
political culture long ago? And if it has not—if most people still think there is always
law for judges to follow—why should the profession fear to correct their error in the
interests of a more honest judicial practice? In any case, how can the pretense
work? Would it not be easy for the disappointed party to demonstrate that there
really was no law according to the grounds everyone knows are the right grounds?
And if the pretense is so easily exposed, why bother with the charade? Nor is there
any evidence in our sample cases that any of the lawyers or judges actually
believed what this defense attributes to them. Many of their arguments would be
entirely inappropriate as arguments for either the repair or the improvement of law,
they make sense only as arguments about what judges must do in virtue of their
responsibility to enforce the law as it is.”60
He then concludes: “In fact there is no positive evidence of any kind that when
lawyers and judges seem to be disagreeing about the law they are really keeping
their fingers crossed. There is no argument for that view of the matter except the
question-begging argument that if the plain-fact thesis is sound they just must be
pretending.”61
The positivists have a stronger, more sophisticated argument:
“There is, however, a more sophisticated defense of positivism, which

concedes that lawyers and judges in our sample cases thought they
were disagreeing about the law but argues that for a somewhat
different reason this self-description should not be taken at face value.
This new argument stresses the importance of distinguishing between
standard or core uses of the word ‘law’ and borderline or penumbral
57
Id.
58
Id., p. 33.
59
Id., p. 37.
60
Id., pp. 37-38.
61
Id., p. 39.
101
uses of the word. It claims that lawyers and judges all follow what is
mainly the same rule for using ‘law’ and therefore all agree about, for
example, the legal speed limit in California and the basic rate of tax in
Britain. But because rules for using words are not precise and exact,
they permit penumbral or borderline cases in which people speak
somewhat differently from one another. So lawyers may use the word
‘law’ differently in marginal cases when some but not all of the grounds
specified in the main rule are satisfied. This explains, according to the
present argument, why they disagree in hard cases like our sample
cases. Each uses a slightly different version of the main rule, and the
differences become manifest in these special cases. In this respect,
the argument continues, our use of ‘law’ is no different from our use of
many other words we find unproblematical. We all agree about the
standard meaning of ‘house,’ for example. Someone who denies that
the detached one-family residences on ordinary suburban streets are
houses just does not understand the English language. Nevertheless
there are borderline cases. People do not all follow exactly the same
rule; some would say that Buckingham Palace is a house while others
would not.”62
The sophisticated positivist legal defense fails too, according to Dworkin.
“The new story is in one way like the fingers-crossed story, however: it
leaves wholly unexplained why the legal profession should have acted
for so long in the way the story claims it has. For sensible people do
not quarrel over whether Buckingham Palace is really a house; they
understand at once that this is not a genuine issue but only a matter of
how one chooses to use a word whose meaning is not fixed at its
boundaries. If ‘law’ is really like ‘house,’ why should lawyers argue for
so long about whether the law really gives the secretary of the interior
power to stop an almost finished dam to save a small fish, or whether
the law forbids racially segregated schools? How could they think they
had arguments for the essentially arbitrary decision to use the word
one way rather than another? How could they think that important
decisions about the use of state power should turn on a quibble? It
does not help to say that lawyers and judges are able to deceive
themselves because they are actually arguing about a different issue,
the political issue whether the secretary should have that power or
whether states should be forbidden to segregate their schools. We
have already noticed that many of the arguments judges make to
support their controversial claims of law are not appropriate to those
directly political issues. So the new defense of positivism is a more
radical critique of professional practice than it might seem. The
crossed-fingers defense shows judges as well-meaning liars; the
borderline-case defense shows them as simpletons instead.
The borderline defense is worse than insulting, moreover, because it

ignores an important distinction between two kinds of disagreements,
the distinction between borderline cases and testing or pivotal cases.
62
Id., pp. 39-40.
102
People sometimes do speak at cross-purposes in the way the
borderline defense describes. They agree about the correct tests for
applying some word in what they consider normal cases but use the
word somewhat differently in what they all recognize to be marginal
cases, like the case of a palace. Sometimes, however, they argue
about the appropriateness of some word or description because they
disagree about the correct tests for using the word or phrase on any
occasion.”63
It is an argument of the second type which lawyers and judges engage in in actual
judicial cases.
“You might think that the second argument I just described is silly, a
corruption of scholarship. But whatever you think, arguments of that
character occur, and they are different from arguments of the first
kind. It would be a serious misunderstanding to conflate the two or to
say that one is only a special case of the other. The ‘sophisticated’
defense of positivism misunderstands judicial practice in just that way.
The various judges and lawyers who argued our sample cases did not
think they were defending marginal or borderline claims. Their
disagreements about legislation and precedent were fundamental;
their arguments showed that they disagreed not only about whether
Elmer should have his inheritance, but about why any legislative act,
even traffic codes and rates of taxation, impose the rights and
obligations everyone agrees they do; not only about whether Mrs.
McLoughlin should have her damages, but about how and why past
judicial decisions change the law of the land. They disagreed about
what makes a proposition of law true not just at the margin but in the
core as well. Our sample cases were understood by those who argued
about them in courtrooms and classrooms and law reviews as pivotal
cases testing fundamental principles, not as borderline cases calling
for some more or less arbitrary line to be drawn.” 64
Dworkin then summarizes his criticism of positivism and semantic theories of law.
“If legal argument is mainly or even partly about pivotal cases, then
lawyers cannot all be using the same factual criteria for deciding when
propositions of law are true or false. Their arguments would be mainly
or partly about which criteria they should use. So the project of the
semantic theories, the project of digging out shared rules from a
careful study of what lawyers say and do, would be doomed to fail. . . .
...
. . . So our legal philosophers try to save what they can. They grasp at
straws: they say that judges in hard cases are only pretending to
disagree about what the law is, or that hard cases are only borderline
disputes at the margin of what is clear and shared. They think that
63
Id., pp. 40-41; italics Dworkin’s.
64
Id., pp. 42-43.
103
they must otherwise settle into some form of nihilism about law. The
logic that wreaks this havoc is the logic just described, the argument
that unless lawyers and judges share factual criteria about the grounds
of law there can be no significant thought or debate about what the
law is. We have no choice but to confront that argument. It is a
philosophical argument, so the next stage of our project must be
philosophical as well.”65
This, Dworkin calls, the semantic sting. “I shall call the argument I have just
described, which has caused such great mischief in legal philosophy, the semantic
sting.”66
To summarize Dworkin’s point, his conclusions are based on his criticism of

semantic theories of law. Semantic theories are committed to the view that law can
be characterized and the truth of legal propositions determined in terms of shared
linguistic criteria. As such, they all fall victim to the 'semantic sting', since they are
unable to account for theoretical disagreements involved in judging the truth or
falsity of individual propositions of law. Either they pretend that no disagreement
whatsoever exists, these shared linguistic criteria being adequate to determine the
truth of legal propositions, whatever disagreement there is being one outside law;
or they insist that disagreement exists only in the borderline cases of legal
concepts, there being agreement at the core, which disagreement is to be resolved
purely by verbal convention, by choosing to extend the concept one way or another.
The semantic sting has limited impact. None of the jurisprudential theories of note
are semantic in the way described. Hence Dworkin's argument may be valid
against semantic theories, but not all jurisprudential theories are so crude. The
more viable jurisprudential theories, of which the jurisprudential theory of this thesis
is one, are conceptual and not semantic. The semantic sting has no effect on them,
and indeed on other jurisprudential theories.
Indeed Dworkin appears to suggest that there are only two types of jurisprudential
theories, semantic theories and his recommended interpretive theories, which is a
jurisprudential theory based on adjudication. Since semantic theories are not
viable, then only interpretive theories remain. This is a false dichotomy. Dworkin
conveniently neglects conceptual and other theories, or he falsely categorizes all
these viable jurisprudential theories as semantic.
Dworkin's error therefore lies in assuming that since jurisprudential theories attempt
to characterize or define law, they must be, like semantic theories, purely verbal or
linguistic, a matter merely of reporting the shared linguistic criteria people employ
in using a word or of extending this shared or uncontroversial meaning to account
for borderline cases. This argument appears to have greater force when applied to
conceptual theories, since they are preoccupied with ordinary language and engage
in linguistic analysis. But the point of the conceptual theorist's open concern with
language and of his search for some sort of definition is not purely linguistic, the
point being "a sharpened awareness of words to sharpen our perception of
65
Id., pp. 43-44.
66
Id, p. 45.
104
phenomena," an insight of Austin which Hart openly endorsed, 67 or to understand
our self-understanding of the world, which is Raz's view. 68 Hence it is not words or
language per se, which conceptual theories are interested in, or even our thoughts,
but our concepts, how we understand the world, and what lies behind the concepts,
the world. For the key to understanding the world lies in understanding our own
concepts.
Indeed, ordinary language's verdict on the use of our words cannot be the be-all and
the end-all of the matter. For ordinary language itself may be vague, ambiguous or
inconsistent. Moreover, even when ordinary language provides uncontroversial
shared truths about the phenomena being investigated, the puzzlement remains. It
does not disappear with improved reporting on actual usage, or by an agreement on
a set of common characteristics so as to categorize borderline cases. For the
problems are deep and substantial. This is precisely Hart's point in his introduction
to The Concept of Law. The book's purpose was "not to provide a definition of law,
in the sense of a rule by reference to which the correctness of the use of a word can
be tested; it (was) to advance legal theory by providing an improved analysis of the
distinctive structure of a municipal legal system and a better understanding of the
resemblances and differences between law, coercion, and morality, as types of
social phenomena."69
Conceptual theories therefore are not linguistic. Hence they share with interpretive
theories an appreciation of the same problem, which is to arrive at a better
understanding of the phenomena under investigation, and in the inadequacy of
purely linguistic enterprises to provide this understanding, but disagree in the
solution and the method for arriving at the solution.
D. Inclusive Positivism
The third criticism against Hart is an internal one. Professor Raz argues that Hart
subscribes to a version of positivism which endorses a weak social thesis, which
fails to comply with the basic positivist insight that what is law or not law is purely a
matter of social fact and not of moral considerations. Raz, on the other hand,
advocates the strong social thesis, which is more precisely defined as follows: “A
jurisprudential theory is acceptable only if its tests for identifying the content of the
law and determining its existence depend exclusively on facts of human behaviour
capable of being described in value-neutral terms without resort to moral
argument.”70
To elaborate, Raz maintains that the tests for the identity and existence of a legal
system contain three basic elements: efficacy, institutional character, and sources.
The weak social thesis subscribes only to the first two, while the strong social thesis
advocates all three.
67
H. L. A. Hart, The Concept of Law, p. 14.
68
This was made clear by Raz in a set of lectures he delivered in the Michaelmas Term of 1992-93, entitled
"Coherence and Law".
69
Dworkin, supra., pp. 65-68.
70
Joseph Raz, “Legal Positivism and the Sources of Law,” The Authority of Law, Oxford: Clarendon Press, 1979,
pp. 41-42.
105
“Most positivists are ambiguous concerning one interesting point.
While their general terms suggest an endorsement of the strong social
thesis, their actual doctrines rest on efficacy and institutionality as the
only conditions concerning the social foundation of law. Let the
combination of these two conditions be called the weak social thesis.
It is easy to show that the weak and strong theses are not equivalent.
Suppose that the law requires that unregulated disputes (i.e. those
with respect to which the law is unsettled) be determined on the basis
of moral considerations (or a certain subclass of them, such as
considerations of justice or moral considerations fundamentally at odds
with social morality). Suppose further that it is argued that in virtue of
this law moral considerations have become part of the law of the land
(and hence the law is never unsettled unless morality is). This
contention runs directly counter to the strong thesis. If it is accepted,
the determination of what is the law in certain cases turns on moral
considerations, since one has to resort to moral arguments to identify
the law. To conform to the strong thesis we will have to say that while
the rule referring to morality is indeed law (it is determined by its
sources) the morality to which it refers is not thereby incorporated into
law. The rule is analogous to a ‘conflict of law’ rule imposing a duty to
apply a foreign system which remains independent of and outside the
municipal law.”71
Raz’ proceeded to show the inadequacy or undesirability of the weak social thesis
under positivist principles. His argument proceeds as follows:
“The difference between the weak social thesis and the strong social
thesis is that the strong one insists, whereas the weak one does not,
that the existence and content of every law is fully determined by
social sources. On the other hand, the weak thesis, but not the strong
one, builds into law the conditions of efficacy and institutionality. The
two theses are logically independent. The weak thesis though true is
insufficient to characterize legal positivism. It is compatible with—
(a) Sometimes the identification of some laws turns on moral

argument,
but also with—
(b) In all legal systems the identification of some laws turns on moral
argument.
The first view is on the borderline of positivism and may or may not be
thought consistent with it. But whereas the first view depends on the
contingent existence of source-based law making moral considerations
of validity in certain cases (as in the example above), the second view
asserts a conceptual necessity of testing law by moral argument and is
71
Id., pp. 45-46.
106
clearly on the natural law side of the historical positivist/natural law
divide.”72
Unfortunately, this debate internal to positivism must await the next chapter, when I
discuss more fully the legal theory of Professor Joseph Raz. To Raz’s theory
therefore, I now turn.
0–0–0–0–0
REFERENCES
Bix, Brian, Jurisprudence: Theory and Context, London: Sweet and Maxwell, 2003
Dworkin, Ronald, “The Model of Rules I,” Taking Rights Seriously, Cambridge:
Harvard University Press, 1977.
----------“The Model of Rules II,” Taking Rights Seriously, Cambridge: Harvard
University Press, 1977.
----------Law’s Empire, London: Fontana Press, 1986, Chs. 1-3.
----------“Legal Theory and the Problem of Sense,” Issues in Contemporary Legal
Philosophy, ed. by Ruth Gavison, Oxford: Clarendon Press, 1987.
Fernando, Emmanuel Q., Legal Theory, Legal Reasoning and Philippine
Jurisprudence, to be published by the U.P. Law Center.
Gavison, Ruth, “Comment,” Issues in Contemporary Legal Philosophy, ed. by Ruth
Gavison, Oxford: Clarendon Press, 1987, pp. 21-34.
Hart, H.L.A., The Concept of Law, Oxford: Clarendon Press, 1961.
----------Essays in Jurisprudence and Philosophy, Oxford: Clarendon Press, 1983.
----------“Comment,” Issues in Contemporary Legal Philosophy, ed. by Ruth Gavison,
Oxford: Clarendon Press, 1987, pp. 35-42.
Lucas, J.R., “The Phenomenon of Law,” Law, Morality and Society, ed. by PM.S.
Hacker and Joseph Raz, Oxford: Clarendon Press, 1977.
Raz, Joseph, “Legal Positivism and the Sources of Law,” The Authority of Law,
Oxford: Clarendon Press, 1979.
----------“Legal Principles and the Limits of Law,” Ronald Dworkin and Contemporary
Jurisprudence, New Jersey: Rowman and Allenheld, pp. 73-87.
72
Id., pp. 46-47.
107

4 Law As The Union of Primary and Secondary Rules

Uploaded by

Document Information

Original Title

Copyright

Available Formats

Share this document

Share or Embed Document

Sharing Options

Did you find this document useful?

Is this content inappropriate?

Copyright:

Available Formats

4 Law As The Union of Primary and Secondary Rules

Uploaded by

Copyright:

Available Formats

CHAPTER FOUR:

LAW AS THE UNION OF PRIMARY AND SECONDARY RULES

“Questions such as those I have mentioned, ‘What is a State?’, ‘What is

“Hence, though theory is to be welcomed, the growth of theory on the

His definition of a ‘legal right’ is therefore as follows:

“I would, therefore, tender the following as an elucidation of the

(1) A statement of the form ‘X has a right’ is true if the following

(a) There is in existence a legal system.

(2) A statement of the form ‘X has a right’ is used to draw a conclusion

Hart’s methodology may be considered as a combination of conceptual analysis and

Notwithstanding its concern with analysis the book may also be

“When we reflect on the quite general ability of people to recognize

There is of course no doubt that a legal system often presents this

“Moral rules impose obligations and withdraw certain areas of conduct

There are then, to Hart, three recurrent issues of law.

But it is not a definition a legal theorist is looking for.

If not a definition, what then is the theorist after?

Consequently, I now turn to the substantive aspects of Hart’s theory, wherein he

II. SOCIAL RULES

The third feature distinguishing social rules from habits is implicit in

Hart elaborates on this internal aspect. It is not simply a matter of feelings, of

“The internal aspect of rules is often misrepresented as a mere matter

By means of the internal aspect of rules, Hart introduces a method of scientific

Hart’s argument is that whatever advantage a ‘scientific’ approach

III. THE KEY TO THE SCIENCE OF JURISPRUDENCE

“If a society is to live by primary rules alone, there are certain

There is a second defect, that of staticity.

The third defect is that of inefficiency.

“(Secondary) rules have important features in common and are

“The simplest form of remedy for the uncertainty of the regime of

In a developed legal system the rules of recognition are of course more

The second remedy is that of rules of change.

To the third defect of inefficiency, there correspond the rules of adjudication as a

“The third supplement to the simple regime of primary rules, intended

IV. THE RULE OF RECOGNITION

This is how Hart defines the ultimacy of a rule of recognition.

“The sense in which the rule of recognition is the ultimate rule of a

One who makes an internal statement concerning the validity of a

“(The rule of recognition), if it is to exist at all, must be regarded from

“There are therefore two minimum conditions necessary and sufficient

The assertion that a legal system exists is therefore a Janus-faced

V. COMMENTS AND CRITICISM

A. The Conflation of the Distinction between Primary and Secondary Rules

“Hart lays great emphasis on another distinction among rules. In

1. The distinction between rules concerning actions involving physical movements

Laws seldom characterize actions by reference to mere physical

3. The distinction between rules and meta-rules

“Hart is at his unhappiest in his controversy with Professor Fuller over

“The distinction between rules and meta-rules is a distinction

B. The Rule of Recognition

Dworkin, in “The Model of Rules I,”41 characterized Hart’s positivism in terms of

Dworkin engaged in a long argument in an attempt to disprove this tenet. He

“Most rules of law, according to Hart, are valid because some

True, if we were challenged to back up our claim that some principle is

Moreover, the techniques we apply in arguing for another principle do

“So we cannot adapt Hart’s version of positivism by modifying his rule

I conclude that if we treat principles as law we must reject the

Prof. Dworkin summarizes his argument against the existence of a rule of

“Legal principles, like other laws, can be enacted or repealed by

Dworkin formulates two arguments against the existence of a rule or recognition or

“‘Hart’s sharp distinction between acceptance and validity,’ the first

The rule of recognition, therefore, does serve to explain the legal

Hart’s criterion of identity must be modified. A legal system consists