A. Hart’s own account of a legal system: law as the union of primary and secondary rules
1. Why a legal system needs secondary rules
2. Three kinds of secondary rules
3. The defining features of a legal system
4. On his view, is law different from the gunman situation writ large?
B. Dworkin: Law’s Ambitions for itself
1. The three mysteries
2. Dworkin’s model of law
3. Dworkin’s interpretive model of judging (as illustrated by Dronenburg case)
Last class we looked at the negative side of Harts project. He tried to rehabilitate legal positivism by
discarding what he thought was dispensable (command theory) and explaining what he thought was
The command theory of law in Austin’s view gave a positive explanation of how law could be “law”.
Hart wants to reject this rule. We now need some other account of what makes law “law like”.
In Harts opinion, law is the union of primary and secondary rules. Why does a legal system need
secondary rules? But does his version really counter the “gunman situation”?
A. What makes a rule into a rule of law? How can we fill this gap that (taking the command theory
out) we are left with?
Begins with a thought experiment in order to show why a legal system needs secondary rules. He
asks us to imagine a society with the following features: they only have primary rules (don’t murder
other people, don’t injure other people’s property, these are rules telling others how they should
behave, lay down a series of dos and don’ts). There is no legislature, courts, officials, or other
authority. Nor are there rules about how these rules should be enforced or applied. No authority
vested in interpretations or applications of these rules. Could such a society even exist? What kinds
of problems would a society like this have?
1) There would be no way of deliberately changing the rules in a way that everyone in the society
would have to recognize and accept this change. The static character of rules. It’s not that it could
never evolve in terms of beliefs or moral practices but that there would be no deliberate way of doing
this (at a particular point in time) or ensuring that everyone would recognize the change (and take it
into account).The change would happen slowly and hazardously.
2) Problem of inefficiency. When disputes arise about what rules say there would be no authority to
resolve these sorts of disputes (or exert pressure for conformity to said rules). There will always be
disputes of these sorts, but there would be no authority to resolve them.
3) The problem of uncertainty. Related to number two but comes prior to it. When doubts arise as
to what a particular rule is in a particular case. What things count under certain rules and what does
These problems suggest that we need secondary rules. They are on different level from the primary
rules but they are about the primary rules. They specify how primary rules can be changed,
implemented, ascertained, and who is to do this (and how they are to do this and enforce it). How
they come in and out of existence and how they should be interpreted.
The solve these three issues there are three remedies:
(1)The rule of recognition: one of the most important. Group of rules, or set of rules that lays
down particular features (of a group of rules) of primary rules in order for them to be “laws”. It tells us what has to be true of primary rules in order for them to be valid laws. May require that laws have to
be properly enacted, laid down by a judge in a fair trial or that they conform to the constitution
(including a charter of rights or the like), etc. In Canada our rule of recognition includes: the law
has to conform the Canadian Charter of Rights and Freedoms. It divides powers between
provincial and federal governments.
How does this fix the problem of uncertainty? It sets some standards against which we may
measure what a law is and is not. It tells us the sorts of procedures that have to be followed if
something is to count as a law or not. * The rule of recognition is key because this is what we look at
to see what a valid law is in the legal system. *
(2)The rules of change: Empower us to introduce new primary and secondary rules and repeal old
ones. Gives us the right to change old rules and make new ones. They would include rules for judicial
decisions (when you can and cannot change laws). Rules of legislative enactment for example. They
are designed to empower people in authority.
(3)The rules of adjudication: It allows people to make authoritative announcements in particular
cases in regards to when a particular primary or secondary rule was broken.
Addresses problem of inefficiency.
Identify who is to adjudicate and what procedure they are to follow.
To sum up, primary rules are rules about behaviour and confer certain rights. While secondary
rules are about:
primary rules and tell us how to interpret them,
who is to interpret them and give us the right to change them.
An example of a Canadian secondary rule:
Law: a rule in the marriage act which says that a judge may solemnize marriages under the authority
of a license. It is a secondary rule of change and recognition. Someone might question this law. We
then look to a further secondary rule that says; whatever a provincial government legislates on
subject matters that are assignment to them under the constitution is law provided that they follow the
correct procedures for legislative enactment and provided that the rule conforms to the constitution
(including the charter of rights).
Secondary rules are not really “classifications”. We should think about why secondary rules help
us understand what law is? We should understand the function secondary rules have as a whole in
Hart’s project and what distinctive force the law has over us?
He uses these secondary rules to give an account of what a legal system is.
What are the defining features of a legal system?
There are three defining features in Hart’s view.
Firstly a legal system consists of a set of primary rules all of which are valid according to the
systems rule of recognition. Our primary rule of “marriage is between a man and a woman” is no
longer recognized as a valid rule and has been changed to include homosexual marriage.
Secondly, these primary set of rules are obeyed by the bulk of the population. Most people obey
these rules and the reasons for obedience are not an important aspect, what matters is simply that
they do. They may obey due to fear of punishment, because they believe it is just or etc. Does this
not sound very similar to the command theory (except for the “secondary rules” aspect)? How is
Hart’s version different from the command theory other than in these secondary rules? The third
defining feature, in Hart’s view, is what makes this difference. It is that the secondary rules are all accepted by officials as correct. He says that the officials take the internal point of view on these
rules. This is a key difference on the command theory of law since the command theory of law has no
internal point of view. Some understanding of this internal point of view and the claim that officials
must take this view is necessary if we are to understand what a legal system is and in particular if we
are to account for laws claim to legitimacy. The internal point of view is of an insider. Of someone
who accepts the rule and treats it as binding onto themselves and others. Assumes that there is an
obligation to act in accordance with the rule. Page 64 of the textbook explains this very well: “the rule
of recognition must be regarded from an internal point of view as a common standard of correct
judicial decision and not just as something each judge obeys for his part only”, “officials must regard
as common standards of official behaviour and appraise critically their own and each other’s
deviations as lapses”. So, failing to do things as they correspond to secondary rules is a mistake for
which you can be held accountable. The rule of recognition must be accepted by officials as the
standard of correctness in the judicial system. They are supposed to use it to criticise their own and
others behaviour. This is partly why that the rule of recognition is so important. Primary concern is in
the way that officials treat other officials.
In Hart’s opinion, Austin and Bentham fail to give an account laws claim to legitimacy. Hart
thinks he has achieved two things here. He has given an account of law that does not appeal to
morality (we can identify the rule of recognition, the other secondary rules and the primary rules
without engaging in controversial moral arguments.) This is really important because when Hart says
that we can give an account of law without turning to morality, this is what he means; that we can look
to what the rule of recognition is, we can look up the body of primary rules that are valid according to
that rule recognition and we can assess whether that official has taken the internal point of view on
this rule, and we can do all of that without having to argue about whether these rules are “morally
good rules” or whether or not the officials are behaving in a just way. Simply because the internal
point of view is only taken by officials of the society on the rule of recognition, and we ourselves do
not have to take it. He thinks he has given an
account of a system that uses ordinary sociological facts and doesn’t require us to take a position on
sorts of laws are just or unjust.
We can still account for the laws normative force in its claim to legitimacy by appealing to the
internal point of view the officials take. Because the officials of the system themselves treat the rule of
recognition as a standard of correctness; as the correct way to ascertain what a valid law is in the
Hart takes it to be one of the virtues of his account, that he can explain both the nature of wicked
systems and morally just ones. That it gives an account of all types of legal systems. The point is that
of them will have primary rules, a set of secondary rules and officials taking the internal point of view.
They are completely agnostic as to whether the legal system in question is moral or not (since we do
have to get into those questions at all).
Has Hart captured the essential feature of a legal system or not? Has he nevertheless succeeded in
accounting for the main features of law that he talked about last time. Or does something seem to be
missing from his account?
It can be somewhat different from the gunman writ large since the gunman isn’t required to take an
internal point of view. This is exactly why Hart objects the command theory. Officials in Hart’s view
take on the internal view and regard as what they are doing as correct and that they believe they are
following the correct procedures.
But are secondary rules and the internal view enough to allow us to capture everything we want to
capture about what we mean when we talk about something as being a legal system or a law?
There seems to be a problem here in explaining how a law can claim to have legitimacy. The notion of the internal point of view explains how officials think these procedures and laws are valid. It
doesn’t actually explain how law can have legitimacy. What could Hart say to this objection? Simply
his account is not supposed to show how it can have legitimacy. The whole point of his account is to
explain how laws claim to have legitimacy. And that his account attempts to explain all legal systems,
even those that do not have legitimate laws.
It seems that we have only explained how officials think the laws are correct. How does it provide
a legal obligation for a citizen to obey? His account only tells us how officials think they do, but not
How is the government accountable to the people in this view? What would Hart reply to this?
Maybe in a democracy this is a necessary feature of a legal system but this is not necessary in other
systems. Without the notion of accountability or sense that officials have some moral obligation to
citizens, are we not back to something like the command theory and the gunman situation writ large?
is the bulk of population still obeying the laws?
Do people not ask what makes a legal system legal or what makes a law an actual law (or give it
validity) in order to understand the justification for obeying the law or legal system. Hart only wishes
give an account of what all legal systems are like. While Dworkin wants to give an account of what a
morally just legal system is.
B. Dworkin’s aim is justificatory. While Hart’s was merely descriptive, Dworkin is trying to
explain how law can have and claim to have legitimacy.
What are the three mysterious of law and why is it beneficial to law to explain these mysteries?
Dworkin and Hart give us two different things. One is a model of law as law and the other is a
model of adjudication. For a test, map out what Hart’s model of law and adjudication is and then what
Dworkin’s model of law and adjudication is in comparison.
There are mysteries of law and they are as follows:
Law works itself pure. Somehow mysteriously when judges appear to change the law, the
changes are guided by the law itself trying to work towards a clearer statement of itself. It’s trying to
render itself more consistent.
The second mystery is that there is a hig