. 19
( 155 .)


Let us now put to one side the unlikely prospect of legislation by the United
Kingdom Parliament purporting to alter the law of a state such as Zimbabwe or
Canada to which it has ostensibly made an unquali¬ed transfer of legislative
53 The ideas of the constitution

The wider question remains, which has important practical implications,
whether Parliament can bind itself (including succeeding Parliaments) either as
to the content of future legislation or as to the manner and form in which future
legislation must be passed. As we shall see, consideration of this question takes
us into the treacherous waters of what the legal basis of the doctrine of parlia-
mentary sovereignty is. The sovereignty of the United Kingdom Parliament has
traditionally been held to be of that transcendent kind that cannot be limited
even by Parliament itself.

Godden v Hales (1686) 11 St Tr 1165 (KB)
Herbert CJ: . . . [I]f an act of parliament had a clause in it that it should never be repealed,
yet without question, the same power that made it, may repeal it.

Professor HLA Hart holds that the rule of parliamentary sovereignty is part
of what he calls the ˜rule of recognition™ of our legal system. This is the funda-
mental or ultimate rule of the system which states the criteria for identifying
valid rules of law: unlike all the other rules the rule of recognition is binding
simply because it is accepted by the community, in particular by its judges and
o¬cials. The rule of recognition sustaining our constitutional system is said to
include the proposition that Parliament cannot bind itself.

HLA Hart, The Concept of Law (2nd edn 1994), pp 149“50

Under the influence of the Austinian doctrine that law is essentially the product of a legally
untrammelled will, older constitutional theorists wrote as if it was a logical necessity that
there should be a legislature which was sovereign, in the sense that it is free, at every
moment of its existence as a continuing body, not only from legal limitations imposed ab
extra, but also from its own prior legislation. That Parliament is sovereign in this sense may
now be regarded as established, and the principle that no earlier Parliament can preclude
its ˜successors™ from repealing its legislation constitutes part of the ultimate rule of recogni-
tion used by the courts in identifying valid rules of law. It is, however, important to see that
no necessity of logic, still less of nature, dictates that there should be such a Parliament; it
is only one arrangement among others, equally conceivable, which has come to be accepted
with us as the criterion of legal validity. Among these others is another principle which might
equally well, perhaps better, deserve the name of ˜sovereignty™. This is the principle that
Parliament should not be incapable of limiting irrevocably the legislative competence of its
successors but, on the contrary, should have this wider self-limiting power. Parliament would
then at least once in its history be capable of exercising an even larger sphere of legislative
competence than the accepted established doctrine allows to it. The requirement that at
every moment of its existence Parliament should be free from legal limitations including
even those imposed by itself is, after all, only one interpretation of the ambiguous idea of
legal omnipotence. It in effect makes a choice between a continuing omnipotence in all
54 British Government and the Constitution

matters not affecting the legislative competence of successive parliaments, and an unre-
stricted self-embracing omnipotence the exercise of which can only be enjoyed once. These
two conceptions of omnipotence have their parallel in two conceptions of an omnipotent
God: on the one hand, a God who at every moment of his existence enjoys the same powers
and so is incapable of cutting down those powers, and, on the other, a God whose powers
include the power to destroy for the future his omnipotence. Which form of omnipotence “
continuing or self-embracing “ our Parliament enjoys is an empirical question concerning the
form of rule which is accepted as the ultimate criterion in identifying the law. Though it is a
question about a rule lying at the base of a legal system, it is still a question of fact to which
at any given moment of time, on some points at least, there may be a quite determinate
answer. Thus it is clear that the presently accepted rule is one of continuing sovereignty, so
that Parliament cannot protect its statutes from repeal.

The rule of recognition, which a¬rms the continuing sovereignty of
Parliament, may change over time; political developments may eventually “ or
even suddenly “ cause the courts to give obedience to a modi¬ed or new rule of
recognition. But while it stands it has, as Hart says, a ˜unique authoritative
status™. The rule of parliamentary sovereignty is not alterable by Parliament
acting alone.

HWR Wade, ˜The basis of legal sovereignty™ [1955] CLJ 172, 187“9

But to deny that Parliament can alter this particular rule [that the courts will enforce statutes]
is not so daring as it may seem at first sight; for the sacrosanctity of the rule is an inexorable
corollary of Parliament™s continuing sovereignty. If the one proposition is asserted, the other
must be conceded. Nevertheless some further justification is called for, since there must be
something peculiar about a rule of common law which can stand against a statute.
The peculiarity lies in this, that the rule enjoining judicial obedience to statutes is one of
the fundamental rules upon which the legal system depends. That there are such rules, and
that they are in a very special class, is explained with great clarity by Salmond [ Jurisprudence
(10th edn 1947), p 155]:
˜All rules of law have historical sources. As a matter of fact and history they have their
origin somewhere, though we may not know what it is. But not all of them have legal
sources. Were this so, it would be necessary for the law to proceed ad infinitum in tracing
the descent of its principles. It is requisite that the law should postulate one or more
first causes, whose operation is ultimate and whose authority is underived. . . . The rule
that a man may not ride a bicycle on the footpath may have its source in the by-laws
of a municipal council; the rule that these by-laws have the force of law has its source
in an Act of Parliament. But whence comes the rule that Acts of Parliament have the
force of law? This is legally ultimate; its source is historical only, not legal. . . . It is the law
because it is the law, and for no other reason that it is possible for the law itself to take
notice of. No statute can confer this power upon Parliament, for this would be to assume
and act on the very power that is to be conferred.™
55 The ideas of the constitution

Once this truth is grasped, the dilemma is solved. For if no statute can establish the rule
that the courts obey Acts of Parliament, similarly no statute can alter or abolish that rule.
The rule is above and beyond the reach of statute, as Salmond so well explains, because
it is itself the source of the authority of statute. This puts it into a class by itself among
rules of common law, and the apparent paradox that it is unalterable by Parliament turns
out to be a truism. The rule of judicial obedience is in one sense a rule of common law,
but in another sense “ which applies to no other rule of common law “ it is the ultimate
political fact upon which the whole system of legislation hangs. Legislation owes its author-
ity to the rule: the rule does not owe its authority to legislation. To say that Parliament
can change the rule, merely because it can change any other rule, is to put the cart before
the horse.
For the relationship between the courts of law and Parliament is first and foremost
a political reality. Historical illustrations of this are plentiful. When Charles I was executed
in 1649 the courts continued to enforce the Acts of the Long Parliament, the Rump,
Barebones™ Parliament, and the other Commonwealth legislatures. For a revolution took
place, and the courts (without any authority from the previous sovereign legislature)
spontaneously transferred their allegiance from the King in Parliament to the king-
less Parliaments. In other words, the courts altered their definition of ˜an Act of Parliament™
and recognised that the seat of sovereignty had shifted. This was a political fact from
which legal consequences flowed. But in 1660 there was a counter-revolution: Charles II
was restored, and it was suddenly discovered that all Acts passed by the Common-
wealth Parliaments were void for want of the royal assent. The courts, again without
any prior authority, shifted their allegiance back to the King in Parliament, and all the
Commonwealth legislation was expunged from the statute book. The ˜glorious revolution™
of 1688 was, in its legal aspect if in no other, much like the revolution of 1649, for
the courts, recognising political realities but without any legal justification, transferred
their obedience from James II to William and Mary. Had the Jacobite rebellions of 1715
and 1745 succeeded, the courts might once again have held all intervening legislation “
including the Bill of Rights and Act of Settlement “ void for lack of the assent of the proper
monarch. The fact that William and Mary™s Parliament had passed Acts confirming their
title to the Crown and its own legislative authority would obviously not have availed in
the least.
What Salmond calls the ˜ultimate legal principle™ is therefore a rule which is unique in
being unchangeable by Parliament “ it is changed by revolution, not by legislation; it lies
in the keeping of the courts, and no Act of Parliament can take it from them. This is only
another way of saying that it is always for the courts, in the last resort, to say what is
a valid Act of Parliament; and that the decision of this question is not determined by any
rule of law which can be laid down or altered by any authority outside the courts. It is
simply a political fact.

The ˜revolution™ which, in the view of Sir William Wade, is required for any
change in the ultimate legal principle of parliamentary sovereignty, may be
a gradual event rather than a sudden political convulsion.
56 British Government and the Constitution

Sir William Wade, Constitutional Fundamentals (rev edn 1989), p 37

I have never suggested that no shift in judicial loyalty is possible. One has only to look at
the shifts which took place in seventeenth-century England, in eighteenth-century America
and in the twentieth-century dissolution of the British Empire, latterly in particular in
Zimbabwe. These shifts are revolutions, breaks in continuity and in the legal pedigree of
legislative power. Even without such discontinuity there might be a shift of judicial loyalty
if we take into account the dimension of time. Suppose that Parliament were to enact a Bill
of Rights entrenched by a clause saying that it was to be amended or repealed only by Acts
certified to be passed by two-thirds majorities in both Houses. Suppose also that Parliament
scrupulously observed this rule for 50 or 100 years, so that no conflicting legislation came
before the courts. Meanwhile new generations of judges might come to accept that there
had been a new constitutional settlement based on common consent and long usage, and
that the old doctrine of sovereignty was ancient history, to be classed with the story of the
Witenagemot, Bonham™s case, the Rump, Barebones™ Parliament and the Jacobite pretenders.
The judges would then be adjusting their doctrine to the facts of constitutional life, as they
have done throughout history.

If a Bill of Rights with entrenched provisions such as Sir William Wade supposes
(and unlike the Human Rights Act 1998 which, not being entrenched, is recon-
cilable with the orthodox doctrine of parliamentary sovereignty) were to be
introduced with general political and popular support, it might perhaps in
a much shorter time than ¬fty or a hundred years establish itself in the British
political culture and be recognised by the judges as having worked a change in
the ground rules of the constitution.
It is argued by some constitutional theorists (whose position derives an
arguably gratuitous advantage from being sometimes described as the ˜new
view™ of parliamentary sovereignty) that the orthodox doctrine of sovereignty
does not prevent Parliament from binding itself as to the ˜manner and form™
(as opposed to the content) of future legislation. (See eg, RFV Heuston, Essays
in Constitutional Law (2nd edn 1964), ch 1.) According to this view, Parliament
could e¬ectively provide that an Act might be repealed or amended only by
a speci¬ed majority in both Houses, or only with the approval of the electorate
in a referendum, or only by the use of a prescribed verbal formula in the amend-
ing Act. The ˜new view™ that such self-imposed procedural limitations would
be binding on Parliament relies largely on Commonwealth cases, of which the
following is an example.

Harris v Minister of the Interior 1952 (2) SA 428 (Appellate Division
of the Supreme Court of South Africa)
The South Africa Act 1909, an Act of the United Kingdom Parliament, joined
together four colonies as the Union of South Africa, and created the Parliament
of the Union. This was initially a non-sovereign legislature which by reason of
57 The ideas of the constitution

the Colonial Laws Validity Act 1865 had no general power to legislate inconsis-
tently with United Kingdom statutes extending to South Africa. But section 152
of the South Africa Act empowered the Union Parliament to ˜repeal or alter any
of the provisions of this Act™.
Those who framed the South Africa Act were concerned to protect and
entrench existing voting rights, in particular the rights of the ˜Cape Coloured™
voters of the former Cape Colony. Accordingly section 35(1) of the Act provided
that no Act of the Union Parliament should disqualify any person in the Cape
Province as a voter by reason of his race or colour, unless the bill:

be passed by both Houses of Parliament sitting together, and at the third reading be agreed
to by not less than two-thirds of the total number of members of both Houses.

Section 35 was itself entrenched in a proviso to section 152, by which any repeal
or alteration of section 35, or indeed of section 152, could be e¬ected only by
the same method of a bill passed by a two-thirds majority in a joint sitting of
both Houses.
In 1948 a National Government came into power and initiated an intensi¬ed
policy of white supremacy under the name of apartheid. By that date South Africa
had, as a result of constitutional convention and the Statute of Westminster 1931,
shed its colonial status and was acknowledged to be an independent and sover-
eign state within the Commonwealth. In 1951 the Union Parliament passed by
a simple majority, the two Houses sitting separately, a Separate Representation
of Voters Act (Act 46 of 1951) which deprived Cape Coloured voters of their
existing voting rights, by providing for their registration on a separate voters™
roll. Some of the disquali¬ed voters brought proceedings to challenge the valid-
ity of the Act.
The argument of counsel for the Government was that the Union Parliament,
having acquired full legislative sovereignty as a result of the Statute of West-
minster, was free to disregard the limitations contained in sections 35 and 152
of the South Africa Act 1909.
A unanimous Appellate Division held that the Separate Representation of
Voters Act was null and void.

Centlivres CJ: . . . It is common cause that Act 46 of 1951 was passed by the House of
Assembly and the Senate sitting separately and assented to by the Governor-General and
that it was not passed in conformity with the provisions of sec. 35(1) and sec. 152 of the
South Africa Act. . . .
If Act 46 of 1951 had been passed before the Statute of Westminster, it is clear . . . that
that Act would not have been a valid Act, as it was not passed in accordance with the
procedure prescribed by secs 35(1) and 152. . . .
The effect of sub-sec (1) of sec 2 [of the Statute of Westminster] is that the Colonial Laws
Validity Act no longer applies to any law made . . . by the Union Parliament. Consequently
58 British Government and the Constitution

the Union Parliament can now make a law repugnant to a British Act of Parliament in so far
as that Act extends to the Union. . . . [I]t is clear that when [the Statute of Westminster]
refers to a law made by a Dominion, such law means in relation to South Africa a law made
by the Union Parliament functioning either bicamerally or unicamerally in accordance with
the requirements of the South Africa Act.
[The judge referred to the argument of counsel for the Government that the effect of the
Statute of Westminster was that the Union was a sovereign state and that all fetters binding
the Union Parliament had fallen away, and continued:]
A State can be unquestionably sovereign although it has no legislature which is completely
sovereign. As Bryce points out in his Studies in History and Jurisprudence legal sovereignty
may be divided between two authorities. In the case of the Union, legal sovereignty is or
may be divided between Parliament as ordinarily constituted and Parliament as constituted
under . . . the proviso to sec 152. Such a division of legislative powers is no derogation from
the sovereignty of the Union and the mere fact that that division was enacted in a British
Statute (viz, the South Africa Act) which is still in force in the Union cannot affect the question
in issue.
. . . The South Africa Act, the terms and conditions of which were, as its preamble shows,
agreed to by the respective Parliaments of the four original Colonies, created the Parliament
of the Union. It is that Act and not the Statute of Westminster which prescribes the manner
in which the constituent elements of Parliament must function for the purpose of passing
legislation. While the Statute of Westminster confers further powers on the Parliament of the
Union, it in no way prescribes how that Parliament must function in exercising those powers.
. . . [T]he Statute of Westminster has left the entrenched clauses of the South Africa Act
intact, and, that being so, it follows that . . . courts of law have the power to declare Act 46
of 1951 invalid on the ground that it was not passed in conformity with the provisions of
secs 35 and 152 of the South Africa Act. . . . To hold otherwise would mean that courts of
law would be powerless to protect the rights of individuals which were specially protected
in the constitution of this country.


. 19
( 155 .)