When Does Grundnorm Change

DATE DOWNLOADED: Sat Aug 20 06:01:41 2022
SOURCE: Content Downloaded from HeinOnline
Citations:
Bluebook 21st ed.

J. W. Harris, When and Why Does the Grundnorm Change, 29 CAMBRIDGE L.J. 103 (1971).
ALWD 7th ed.

J. W. Harris, When and Why Does the Grundnorm Change, 29 Cambridge L.J. 103 (1971).
APA 7th ed.

Harris, J. J. (1971). When and why does the grundnorm change. Cambridge Law Journal,
29(1), 103-133.
Chicago 17th ed.

J. W. Harris, "When and Why Does the Grundnorm Change," Cambridge Law Journal 29, no.
1 (April 1971): 103-133
McGill Guide 9th ed.

J. W. Harris, "When and Why Does the Grundnorm Change" (1971) 29:1 Cambridge LJ 103.
AGLC 4th ed.

J. W. Harris, 'When and Why Does the Grundnorm Change' (1971) 29(1) Cambridge Law
Journal 103
MLA 9th ed.

Harris, J. W. "When and Why Does the Grundnorm Change." Cambridge Law Journal, vol.
29, no. 1, April 1971, pp. 103-133. HeinOnline.
OSCOLA 4th ed.

J. W. Harris, 'When and Why Does the Grundnorm Change' (1971) 29 Cambridge LJ 103
-- Your use of this HeinOnline PDF indicates your acceptance of HeinOnline's Terms and
Conditions of the license agreement available at
https://heinonline.org/HOL/License
-- The search text of this PDF is generated from uncorrected OCR text.
-- To obtain permission to use this article beyond the scope of your license, please use:
Copyright Information
Cambridge Law Journal, 29 (1), April 1971, pp. 103-133.
Printed in Great Britain.
WHEN AND WHY DOES THE GRUNDNORM

CHANGE?
J. W. HARRIS
S. (1). INTRODUCTION
IN Pakistan, Uganda and Southern Rhodesia, courts have held

themselves entitled to declare that the effect of a successful, revolution
has been to change the law in their respective jurisdictions.' Judges
appointed under one constitution have held themselves to be bound
to recognise the validity of laws promulgated under a different
constitution; the judge's own political opinions have been said to
be irrelevant. 2
The primary authority upon which the courts in these cases-
hereinafter referred to as "the revolution cases "-have relied has
been Hans Kelsen's theory of the change in the grundnorm brought
about by a revolution.
In all three jurisdictions the following passage from Kelsen's
General Theory of Law and State has been cited with approval 1:
(D). Change of the Basic Norm: It is just the phenomenon
of revolution which clearly shows the significance of the Basic
Norm. Suppose that a group of individuals attempt :to seize
power by force, in order to remove the legitimate government
in a hitherto monarchic state, and to introduce a republican
form of government. If they succeed, if the old order ceases,
and the 'new order begins to be efficacious, because the
I The State v. Dosso (1958) 2 Pakistan S.C.R. 180; Uganda v. Commissioner of

Prisons, ex p. Matovu (1966) E.A. 514; Madzimbamuto v. Lardner-Burke N.O.
JD/CIV/23/66, 9 September 1966, Government Printer, Salisbury (Lewis and
Goldin JJ.), (1968) 2 S.A. 284 (Appellate Division), where the judges at first
instance and the majority of the Appellate Division of the High Court of
Rhodesia found that the revolution was not yet successful, but on various
grounds held that partial recognition could be given to the legislative and
administrative acts of the rebel rdgime; and R. v. Ndhlovu (1968) 4 S.A. 515,
where, eight months after Madzimbamuto's case, the Appellate Division held the
revolution to be successful, so that all the new regime's laws had internal
validity.
2 Dosso, n. 1, supra, 184-185 (Muhammad Munir C.J.), 195 (Shahabuddin J.), 222
(Amiruddin Ahmad J.); Matovu, n. 1, supra, 530, 535 (Sir Udo Udoma C.J.);
Madzimbamuto v. Lardner-Burke N.O. (1968) 2 S.A. 284, 326-328 (Beadle C.J.),
364-365 (Qu~net J.P.), 384-386 (Macdonald J.A.); R. v. Ndhlovu (1968) 4 S.A.
515, 520-522, 528-535 (Beadle C.J.), 538-542 (Qu~net J.P.).
3 General Theory of Law and State, 'Harvard ed., 1945 (hereinafter cited as
"G.T.L.S."), p. 118, cited in Dosso, n. I, supra, 185-186 (Muhammad Munir
C.J.), 195 (Shahabuddin J.); in Matovu, n. 1, supra, 535-536 (Sir Udo Udoma
C.J.);, and in Madzimbamuto v. Lardner-Burke N.O., JD/CIV123/,66, 10-11
(Lewis J.) (1968) 2 S.A. 284, 315 (Beadle C.J.).
103
The Cambridge Law Journal [1971]
individuals whose behavior the new order regulates actually

behave, by and large, in conformity with the new order, then
this order is considered as a valid order. It is now according
to this new order that the actual behavior of individuals is
interpreted as legal or illegal. But this means that a new basic
norm is presupposed. It is no longer the norm according to
which the old monarchical constitution is valid, but a norm
according to which the new republican constitution is valid, a
norm endowing the revolutionary government with legal
authority. If the revolutionaries fail, if the order they have tried
to establish remains inefficacious, then, on the other hand,
their undertaking is interpreted, not as a legal, a law-creating
act, as the establishment of a constitution, but as an illegal act,
as the crime of treason, and this according to the old monarchic
constitution and its specific basic norm.4
Kelsen had said: "No jurist would maintain that even after a
successful revolution the old constitution and the laws based there-
upon remain in force, on the ground that they have not been nullified
in a manner anticipated by the old order itself." 5
Kelsen's theory was positivistic (indeed, "pure "), and therefore
apolitical. The theory, therefore, licensed the judges in the revolu-
tion cases-or so they held-to accept the legality of the successful
revolutions occurring in their respective countries, without entering
the political arena.
Even the Privy Council, in holding that the laws and adminis-
trative acts of the Smith regime were completely without legal effect
in Southern Rhodesia, does not appear to have dissented from the
general principle that successful revolutions create new legal orders
whose validity may be 'adjudged by courts within the territory subject
to the revolution. It cited the Pakistani and Ugandan cases without
disapproval, but distinguished the position in Rhodesia, on the
ground that there the revolution could not be said to be successful,
since the old constitutional authority was still trying to regain
control.'
Discussion in legal periodicals of the decisions of the courts in
the revolution cases has been generally hostile. In particular, it has
been asserted that, supposing Kelsen's theory of the change in the
grundnorm to be accepted, that theory does not entitle courts to
4 " Thus, the basic norm, in a certain sense, means the transformation of power
into law "-G.T.L.S., p. 437. Cf. The Pure Theory of Law, Berkeley ed., 1967
(hereinafter cited as " P.T.L."), pp. 208-211.
5 G.T.L.S., p. 118. Cf. G.T.L.S., p. 437.
6 Madzimbamuto v. Lardner-Burke [1969] A.C. 645, 724-725. In Ndhlovu, supra,
n, 1, at 520 and 538, Beadle C.J. and Qunet J.P. interpreted the ruling of the
Privy Council to mean that, as soon as the Rhodesian revolution became
successful in fact, Rhodesian courts were entitled to regard the new r6gime as
lawful.
C.L.J. Change of Grundnorm
draw from it the sort of conclusion which was drawn. TIhe reasons
given for this assertion may be divided under four heads:
Criticism (i)
Even supposing a revolution to be successful, in Kelsen's
terminology "by and large effective," this does not entitle a court
to declare the constitution established by the revolutionaries to be
"valid," because Kelsen stipulates that efficacy is only a necessary
condition of validity and not identical with validity. This criticism
is advanced by J. M. Eekelaar.7
Criticism (ii)
When a revolution has recently occurred, no purely factual test
of efficacy can be applied, so that any decision that laws enacted in
accordance with the revolutionary constitution are valid cannot,
merely by applying Kelsen's theory, be made free of political con-
siderations. This criticism has been advanced by A. Mvi. Honord,1
S. A. de Smith 9 and R. W. M. Dias. 1°
Criticism (iii)
In particular, a court cannot make a factual judgment about the
efficacy of a recent revolution as a prior step to its decision whether
or not to recognise the validity of revolutionary laws, because its
decision on the latter question will be one of the factors upon which
the ultimate success of the revolution depends. This criticism has
been advanced by the same authors in conjunction with Criticism
(ii) above."
Criticism (iv)
Kelsen's theory, being purely descriptive of legal science, can
only indicate the role of the jurist (legal scientist) and has no applica-
tion to the role of the judge. This criticism has been cogently
advanced by F. M. Brookfield."
The purpose of this article is to ask the questions when and why,
7 "Splitting the Grundnorm" (1967) 30 M.L.R. 156, 171-172; " Rhodesia: the
Abdication of Constitutionalism " (1969) 32 M.L.R. 19, 22-23.
8 " Reflections on Revolutions ' [1967] Irish Jurist 268, 271-273. Cf. Annual
Survey of Commonwealth Law (1967), pp. 92-93.
9 "Constitutional Lawyers in Revolutionary Situations" (1968) 7 Western Ontario
L.R. 93, 106-107.
10 "Legal Politics: Norms behind the Grundnorm" (1968) 26 C.L.J. 233, 253-254.
11 Supra, nn. 8, 9, 10. The same criticism of the revolution cases was made by
counsel for the appellant before the Privy Council in Madzimbamuto v.
Lardner-Burke [1969] A.C. 645, 669, 671.
12 "The Courts, Kelsen, and the Rhodesian Revolution" (1969) 19 University of
Toronto L.J. 326, 342-344.
according to Kelsen's theory, the grimdnorm changes upon the

occurrence of a successful revolution. To make it possible to answer
these questions, it will be necessary to re-examine some of Kelsen's
leading concepts-especially those of "validity," "efficacy" and
"legal science "-and to evaluate the functions he attributes to the
grundnorm. The inquiry will show, it is submitted, that the first of
the above criticisms of the courts' application ,of the theory is
unfounded, and that the other criticisms are largely so.
Finally, something will be said of the merits of the theory as such.
S. (2). THE FUNCT[ONS OF THE GRUNDNORM

To understand the nature of the Grundnorm in change, we must recall
its nature when at rest.
Grundnorm, a Hypothesis
Kelsen says that the grundnorm is a "presupposition" or
"hypothesis" in juristic thinking."3 This means that its nature is
exclusively revealed by its function.
A physicist is reported to have answered the Emperor Louis
Napoleon, when asked why his works made no reference to the
existence of God: "I have no need of that hypothesis." There
may be other reasons for assuming the existence of God, apart from
the utility of the assumption, but this is not so of the grundnorm.
Kelsen asserts that the grundnorm is presupposed in juristic thinking
to serve certain logical purposes. If these purposes are in fact fully
served without presupposing the grundnorm, it is a hypothesis
without function and therefore a worthless piece of theorising.
The Grundnorm's Two Functions

Kelsen specifies two functions for the grundnorm: first, to enable
anyone to interpret a command, permission or authorisation 14 as an
objectively valid legal norm 11; secondly, to enable the legal scientist
to interpret ,all valid legal norms as. a non-contradictory field of
meaning."'
13 G.T.L.S., pp. 117, 396, 437; P.T.L., p. 204. Cf. Kelsen: "Professor Stone and
the Pure Theory of Law " (1965) 17 Stanford L.R. -1128 (hereinafter cited as
"Stanford "), 1144: " My characterisation cf the basic norm as 'hypothesis'
means exactly the same as its characterisation as ' presupposition,' which is the
literal translation of this Greek term." Cf. ibid. 1149.
1" ,Authorise" = " confer the power to create law "-P.T.L., p. 118.
Is P.T.L., pp..30, 50-54, 195, 202, 234. Stanford, p. 1143.
16 " With the postulate of a meaningful, that is, non-contradictory order, juridical
science oversteps the boundary of pure posi:ivism. To abandon this postulate
would at the same time entail the self-abandonment of juridical science. The
basic norm has here been described as the essential presupposition of any
positivistic legal cognition." G.T.L.S., p. 437. " Thus the science of law
C.L.J. Change 'of Grundnorm
He gives, as an example of the former function, the different inter-

pretation which, " we" (sc. people in general) give to the order by
a gangster to pay a sum of money, and the order of a tax officer to
pay the same amount. Both have the same "subjective meaning,"
Kelsen says, in the sense that in each case someone issues a command
in words equivalent to: "I direct you to pay me X pounds"; but
we interpret the " subjective meaning" of the tax officer's command
as its "objective meaning," that is, we interpret his command as a
valid legal norm. We do not interpret the gangster's command in
the same way and, according to Kelsen, this contrast can only be
ultimately explained by reference to the logical hypothesis! of the
grundnorm."' This first function of the grundno'm is further con-
sidered in S. (3).
The second function of the grundnorm is exemplified by Kelsen
only in general schematic form. To make his meaning clearer, it
seems desirable to fit a concrete example from our own legal system
into his schema. Such an example is set out later in this section.
Legal Science and Logic

The field of meaning which is the logical concern of the legal
scientist consists of the contents of legal norms valid within a system,
usually a national legal system. These contents are given directly
in the legal source materials---statutes, regulations and reported
judgments, etc. The legal scientist, when describing the law on
any particular topic, at least recites or paraphrases some of these
materials-he may also explain, interpret or criticise ithem. This
reciting and paraphrasing almost invariably accepts two canons of
logical arrangements-for Kelsen, it does not count as legal science
if it does not: first, that valid norms on the same level of authority-
for instance, two sections in the same statute-should :not conflict;
secondly, that valid norms on a lower level of authority-for instance,
delegated legislation or judicial decisions applying statutes-should
be subsumed under norms on a higher level of authority. "Two
legal norms are contradictory and can therefore not both be valid
at the same time, if the two rules of law that describe them are
conceives of its object as a logical unit: a system of non-contradictory norms."

Kelsen: "What is the pure theory of law? " (1959-60) 34 Tulane L.R. 269,
271. " (Epistemologically, the science of law) 'creates' its object in so far as
it comprehends the object as a meaningful whole." P.T.L., p. 72. Cf. P.T.L.,
pp. 74, 205-208; Stanford, p. 1143. For discussion of the conditions which
must be satisfied for a consistent set of norms to exist, see G. El. Von Wright,
Norm and Action, 1963, pp. 141-144. For discussion of the disciplinary status
of legal science, see A. Wedberg, " Some Problems in the Logical Analysis of
Legal Science" (1951) 17 Theoria 246.
17 G.T.L'.S., pp. 31-32; P.T.L., pp. 44--50. Stanford, p. 1144.
The Cambridge Law Journal [1971}
contradictory; and one legal norm may be deduced from another if

the rules of law that describe them can form a logical syllogism." 18
Kelsen's model of a legal system provides a logical justification
for both these canons. The non-contradiction point is met by
postulating a basic norm presupposed in legal science, which gives
unity to the system of valid legal norms under consideration." The
subsumption point is met by providing a schema, which depicts the
legal system as a primarily dynamic one in which lower norms appear
as individualisations and concretisations of higher norms." A higher
norm must "determine" 21 a lower norm at least to the extent of
authorising someone to create it; it may also determine the procedure
for its creation and, to a greater or less degree, its contents.
Content of the Grundnorm

Besides postulating a model which will justify these two logical
canons of legal science, Kelsen specifically adapts the model to meet
the one point of content which, in his view, is common to "law"
as the subject-matter of legal science. "Law," he says, is always
concerned with coercion.22 To build this universal into the model,
he postulates that all norms (including the grundnorm) have the
same form: they all command, permit or authorise the application
of force to human beings against their will (sanctions).2 "According
to the Pure Theory of Law the internal characteristic of the legal
19 P.T.L., p. 74. ". . . since the cognition of law, like any cognition, seeks to
understand its subject as a meaningful whole and to describe it in non-
contradictory statements, it starts from the assumption that conflicts of norms
within the normative order which is the object of this cognition can and must
be solved by interpretation "-P.T.L., p. 20'6. (But cf. n. 45, inira.) For
discussion of the value of the logical canons of legal science, see F. Castberg,
Problems of Legal Philosophy, 1958, pp. 56-72. Ci. G. Gottlieb, The Logic
of Choice, 1968, p. 14.
19 " A norm forms part of a dynamic system if it has been created in a way that
is-in the last analysis-determined by the basic norm." G.T.L.S., p. 113.
"The fact that a certain norm belongs to a certain order is based on the
circumstance that its last reason of validity is the basic norm of this order."
P.T.L., p. 195. 20 G.T.L.S., pp. 124-162. P.T.L., pp. 221-278.
21 Kelsen's use of the concept of " determination " in this context is ambiguous.
It is the legal scientist who, comparing the higher and the lower norms ex post,
places them in a relation of subsumption. The extent to which the norm-
authority who creates the lower norm is actually controlled in so doing by the
terms of the higher norm is a matter for dispute.
22 G.T.L.S., pp. 15-29. P.T.L., pp. 33-42, 54, 62.
23 G.T.L.S., p. 45. P.T.L., pp. 33-42, 50-54, 1(8-114. In the latter work Kelsen
distinguishes between "sanction," in the broader sense, as meaning any coercive
act stipulated by law, and " sanction " in the narrower sense, as meaning only
coercive acts stipulated as reactions to legally ascertained behaviour. This
refinement of his earlier views meets in part, but by no means completely,
criticisms which have pointed out that his insistence on slotting the concepts of
duty, delict and sanctien into his rigid model results in a severe distortion of
these concepts, when ccmparison is made with their use in ordinary language.
Cf. H. L. A. Hart, " Kelsen visited " [19631 U.C.L.A.L.R. 709, 717-722; A. D.
Woozley, "Legal Duties, Offences and Sanctions " (1968) 77 Mind 461.
norm is that it stipulates a coercive act." 24 Rules of law which do

not have this form are merely " secondary" or "dependent" legal
norms.2" "Only the coercive act, functioning as a sanction, ought
to be." 26
In any national legal system the constitution contains, at least,

authorisations to create all the general norms that can be created
within the system, and these in turn authorise the creation of
individual norms. But the different authorisations in the constitution
can only themselves be brought into a system if a higher norm--the
grundnorm-is presupposed which authorises their creation.
In the United Kingdom there are at least two norms which
authorise the creation of general norms. These are, giving them the
form stipulated by Kelsen: (i) coercive acts ought to be applied in
accordance with statutes enacted by the Queen in Parliament; (ii)
coercive acts ought to be applied in accordance with judge-made
rules established in conformity to the doctrine of binding precedents.
A third such authorising constitutional norm would have to be
postulated if custom is still regarded as an independent source of
27
law.
If these norms and those whose creation is authorised by them
are to be treated as a unified field of meaning, a grundriorm has to
be presupposed which authorises their creation. If it were not, it
would be impossible to express the two or three ultimate authorising
norms of the constitution as conclusions in syllogisms, each of which
had the same major premise, and hence the delegated norms
authorised by each of the ultimate constitutional norms would form
separate systems.
Form of the Grundnorm

But what is the form of the grundnorm which is presupposed in
the case of a national legal order whose constitutional norms have
been established by custom? Kelsen himself does not answer this
question.
Where the constitution is contained in a written document
promulgated at some date in the past, Kelsen tells us, in "General
Theory of Law and State," that the schematic form of the grundnorm
of a national legal order is as follows: "Coercive acts ought to be
24 Stanford, p. 1131. It is this internal characteristic which dikstinguishes an

individual legal norm from an individual moral norm, for, although the latter
may provide for "sanctions," these will only be the sanctions of approval or
disapproval-ibid.
25 G.T.L.S., pp. 60-61,,137-138. P.T.L., pp. 54-60, 238.
26 P.T.L., p. 119.
27 Cross, Precedent in English Law, 2nd ed.; pp. 155-162.
The Cambridge Law Journal [197Q]
carried out only under the conditions and in the way determined
by .the 'fathers' of the constitution or the organs delegated by
them." I8 In the case of any particular such national system, words
referring to the actual individuals responsible for promulgating the
constitution-either their names or a generic description of them-
are apparently to be substituted in this formulation for the word
"fathers." 29 Hence, in contemporary France, the "fathers" are
General De Gaulle and those who assisted him in the preparation
and promulgation of the Constitution of the Fifth Republic.
In "The Pure Theory of Law," the formula expressing the
grundnorm is stated thus: "Coercion of man against man ought to
be exercised in the manner and under the conditions determined by
the historically first constitution." 20 The grundnorm of any given
national order must apparently be found, in terms of this formulation,
by substituting for the words "the h:istorically first constitution"
some words by which the written constitution now in force can be
identified, such as "the constitution of the Fifth Republic" in the
case of contemporary France. But that there is no shift in theory
on this point between the two works is made clear when, later in
The Pure Theory of Law, we read that, where a basic norm is
presupposed in the case of a consciously created constitution, the
individuals who created it "are looked upon as norm-creating
authorities." 31
Where there is no written constitutiop, as in the United Kingdom,
Kelsen tells us only that the grundnorm authorises the norm-creating
effect of custom. 2 (It may do this even where there is a written
constitution which does not stipulate custom as a source of law; in
which case, the "constitution in a material sense "-which includes
all ultimate sources authorised by the grundnorm-will not be
confined to .the written constitution. 3)
In the latest version of his theory, Kelsen tells us that custom
is created by an act of will "individual or collective." "' What he
means by this is not clear, but the following interpretation is
suggested: the norms of an entirely or partially customary con-
stitution could not have attained their constitutional normative status
if, at various times in the past, influential individuals had not
28 G.T.L.S., pp. 115-116.
29 In his Stanford Law Review article of 1965, Kelsen, in answer to a question
raised by Professor Stone, tells us that the content of the grundnorm of each
national legal order is different, though in every case the grundnorm refers to
the ultimate constitutional sources of law for that order-Stanford, pp. 1148-1149.
30 P.T.L., p. 50.
31 P.T.L., p. 199. Cf. Stanford, p. 1145. 32 P.T.L., pp. 198-199.
33 G.T.L.S., p. 126. P.T.L., pp. 221-224.
3' P.T.L., pp. 225-226, 229. (Contr2, G.T.L.S., pp. 34-35.) Kelsen now believes
that all positive norms are created by acts of will-Stanford, pp. 1138-1139.
specifically wanted such norms to operate on the constitutional level

and if their wishes had not received general acquiescence: Accord-
ingly, the grundnorm Of the United Kingdom legal system, authorises
the population at large from time to time to fashion the ultimate
norms of the constitution in any way which meets with general
approval.
The following example is suggested to illustrate Kelsen's schema
for the cognition of law, as applied to the national legall. system of
the United Kingdom.
(A) Grundnorm of United Coercive acts ought to be applied only
Kingdom legal order: under the conditions and in the ways
customarily recognised as constitu-
tional from time to time by the
population at large.
(B) One of the ultimate Coercive acts ought to be applied in
constitutional norms.' accordance with statutes enacted by
the Queen in Parliament.
(C) General norm:. Where, pursuant to a contract of sale of
goods, the seller neglects or refuses to
deliver the goods to the buyer, and
the buyer elects to sue, the seller ought
to be condemned to pay damages."
(D) Particular norm If Smith neglects orrefuses to deliver his
created by contract horse (Dobbin) to Brown, and. Brown
between Smith and elects to sue, Smith ought to be pon-
Brown: demned to pay damages.
(E) Particular concret- If Smith fails to pay damages of £100
ised norm created, by within 28 days of this order, and Brown
a county court Judge issues a warrant of execution, the
on Brown's. suing, bailiff ought to levy execution upon
Smith for breach of the goods of Smith.
contract:
(F) Particular fully con- "Bailiff, levy execution upon the goods
cretised norm created of Smith."
by Brown's issuing .a
warrant of execution:
Aims of Legal Science

At any one time the. law of a country, conceived (as it is by legal
science) as a non-contradictory field of meaning, consists of some
35 Sale of Goods Act 1893, s. 51(1).

36 A more complete example than that presented in the text would illustrate the
application of the process of subsurnption to rules of evidence, procedure and
jurisdiction-GT.L.S., pp. 128-130.' P.T.L., pp. 130-132.
general rules which are available for particular application, and some
particular rules representing applications of these general rules. The
legal scientist who writes the traditional kind of legal textbook will
usually describe "the law" in terms of statements whose generality
will be parallel to norm (C) above. The legal scientist who describes
what is "the law" to a client-before giving advice or making
predictions on the basis of that law-will, in addition, be concerned
to make statements whose generality is parallel to norm (D) above.
So long as logical consistency is one of the aims of legal science,
some logical model justifying non-contradiction and subsumption is
an essential part of legal theory. Kelsen's model, it is submitted, is
the best which theory has so far achieved. However, there are other
aims which legal scientists commonly set themselves, such as the
historical and sociological explanation of the origin and application
of rules of law, the moral or political justification or criticism of such
rules, and policy-oriented suggestions for the interpretation of such
rules in doubtful cases and for the enactment or repeal of such rules.
Kelsen's theory, by insisting on "pure " cognition-that is, the
logically consistent arrangement of legal source materials without
reference to history, purpose or policy-as the hallmark of legal
science, impliedly downgrades these other aims.
S. (3). THE CONCEPT OF VALIDITY
First function of grundnorm rejected

On the basis of the above discussion it is submitted that something
like the hypothesis of a grundnorm is necessary to serve the second
of the two functions which Kelsen specifies for it; to enable the legal
scientist to interpret all valid legal norms as a non-contradictory field
of meaning. It is submitted, however, that this hypothesis is quite
unnecessary to serve the first of the two functions: to enable anyone
to interpret a command, permission or authorisation as an objectively
valid legal norm.
The citizen or lawyer who distinguishes the tax officer's demand
from the gangster's demand as "valid " or "lawful" does not need
to presuppose a basic norm, unless he is a legal scientist seeking to
show that the contents of the tax officer's directive form part of a
unified field of meaning constituted by it and all other valid legal
norms. For one not engaged in the specialist activity of legal science,
that the tax officer's demand is legally valid, while the gangster's is
not, means that it is in conformity with a particular higher general
norm which authorises the tax officer to issue it.
Nor can we accept Kelsen's view that the subjective meaning of
the tax officer's and the gangster's commands are the same. It is
true that they may both be expressed in the imperative mood, but
the semantic function of an utterance as imperative rather than
37
descriptive depends, not on grammatical form, but on context.
The subjective meaning of "command " is attributed to the utterance
of the tax officer or the gangster because of their contexts, and the
context of each enables their subjective meanings to be clearly
differentiated into "tax Officer's command" or "gangster's com-
mand." Context will always include as part of the subjective meaning
of the tax officer's command: "'I, in the exercise of legal powers
conferred on me, demand payment... ," '.
Four Senses of" Validity "

To assert that the hypothesis of the grundnorm is needed to
perform the first function as well as the second is an uncharacteristic
logical error. Why should Kelsen have made it?
The answer lies, it is submitted, in the much-confused concept
of validity. In the case both of the individual tax officer's command
and of a unified system of norms, Kelsen says that the grundnorm
supplies "the reason for validity."" He assumes that validity is
the same concept in both cases. There are, however, ail least four
different senses in which the adjective "valid" may be used to
qualify a legal norm or rule.
Valid/I = conforms to a higher norm (" is not ultra vires," "is
not void ").
Valid/2 = is a consistent part of a normative field of meaning.
Valid/3 = corresponds with social reality (" is in force ").
Valid/4 = has an inherent claim to fulfilment. 0
Valid/1 is the commonest sense of the term in English legal
discourse. Validity/1 enables the tax officer's and the gangster's
demands to be distinguished.
Valid/2 is an occasional term of English legal discourse, as when
it is said: "these two contradictory statutory provisions cannot both
37 Von Wright, n. 16, supra, pp. 96-98, 100-103. Alf Ross, Directives and Norms,
1966, pp. 69-74. Kelsen, " Derogation," Essays in Jurisprudence in Honour of
Roscoe Pound (R. Newman, Ed.), 1962, pp. 339, 342.
38 Kelsen's distinction between " subjective " and " objective " normative meaning
is not clear, but the following interpretation is suggested: where the complete
articulation of a normative expression would require the use of a verb-subject
other than the subject of the normative verb, its meaning is "subjective "; if it
does not require such an additional verb-subject, its meaning is " objective "-
contrast: "I (you, he) say(s) you (he, all men) ought to do such-and-such,"
and "I (you, he, all men) ought to do such-and-such."
39 G.T.L.S., pp. 10-11, 117. P.T.L., pp. 30, 46, 193 et seq. Stanford, pp. 1142,
1145, 1146. Cf. n. 19, supra.
40 For discussion of different senses of " validity," see Von Wright, n. 16, supra,
pp. 194-200. Ross, n. 37, supra, pp. 104-105, 177-182. G. C. Christie, "The
Notion of Validity in Modem Jurisprudence " (1964) 48 Minn.L.R. 1049.
The Cambridge Law Journal [19711
be valid." But Validity/2 is primarily a concept of legal theory.

Validity/2 is what legal theory, "looking down" upon the activity
of legal science, 4' points to as the logical reference-point Of the
assumptions of legal science.
Kelsen himself uses "valid" primarily as Valid/2, but sometimes,
apparently without recognising the distinction, uses it as Valid/i.
When the tax officer's demand is said to be interpreted as " valid"
rather than as a subjective command, Validity /1 is meant and there
is no need in such a context for the presupposition of the grundnorm.
In his earlier writings, Kelsen stated that the quality of "law-
fulness" could not be ascribed to any rule or to any human action
without presupposing the grundnorm. 42 He has now explained that
by -"lawful" in such contexts he meant "valid. ' 4 But this is
merely to replace one ambiguity by another. It is Kelsen's failure
to distinguish Validity/1 and Validity/2 which leads him to say,
in his Stanford Law Review article, that, on the one hand, we can
identify a legislator as legislator without presupposing the grund-
norm," but that, on the other hand, the revenue officer's command
can only be interpreted differently from the gangster's if we
presuppose the grundnorm. 5
Kelsen recognises and specifically rejects for his purposes

Validity/3 and Validity/4. Validity/3 equates the "validity." of a
rule with its effectiveness, and Kelsen insists that they are to be
distinguished.4' Validity/4 suggests that a rule has value, in a
political or moral sense. Validity/4 is, therefore, rejected as incon-
sistent with the pure theory of law. "The pure theory desires to
present the law as it is, not as it ought to be; it seeks to know the
real and possible, not the 'ideal,' the ' right' law." ',
41 Ross, On Law and Justice, 1958, pp. 25-26.

42 "It is impossible to grasp the nature of law if we limit our attention to the
single isolated rule. The relations which link together the particular rules of a
legal order are also essential to the nature of law." G.T.L.S., p. 1. "(The
basic norm) is presupposed to be valid because without this presupposition no
human act could be interpreted as a legal, especially as a norm-creating, act."
G.T.L.S., p. 116. Cf. G.T.L.S., pp. 118, 121,
43 Stanford, pp. 1144-1146. Cf. P.T.L., pp. 2-3. See also n. 64, infra.
44 Stanford, p. 1145.
'5 Stanford, p. 1144. In his article entitled " Derogation " (n. 37, supra) at pp.
351-355, Kelsen apparently denies the concept of Validity/2 altogether. He
there asserts that the " logical canons " referred to in section 2, supra, are really
express or implied positive norms. The implication of this passage would
seem to be that the grundnorm has no function at all.
46 G.T.L.S., pp. 40, 119, 169-171. P.T.L., pjp. 10, 78, 211-214. Stanford, pp.
1139-1140. Kelsen, "On the Pure Theory of Law" (1966) 1 Israel L.R. 1, 2.
47 P.T.L., p. 106. " Legal norms are not valid because they themselves or the
basic norm have a content the binding force of which is self-evident. They are
not valid because of their inherent appeal. Legal norms may have any kind of
(sc. moral or political) content. There is no kind of human behaviour that
because of its nature, could not be made into a legal duty corresponding to a
legal right. 'The validity of a legal norm cannot be questioned on the ground
C.L.J , Change of Grundnorm 115
Metaphysical Validity/5
The distinction between these various senses of "validity"
throws light on, but does not completely explain, two of Kelsen's
most puzzling assertions, namely: (i) no one who accepts the validity
of both a national legal order and the international legal order can
recognise any-conflict between them.' 8 If "valid" here is understood
in the sense of Validity/2, and it is further assumed that' there can
be only one field of normative meaning, Kelsen's assertion becomes
a tautology. But if Validity/1 isascribed to an individual rule of
national law and to an individual rule of international law, there is
nothing impossible about a simultaneous recognition that they
conflict.
(ii) No one can ascribe "validity" simultaneously to conflicting
rules of law and morality. 4" This also is a tautology if ",valid " is
understood as Valid/2, and it is assumed that there can be only one
field of normative meaning. But if a rule of law is asserted to be
Valid/l, Valid/2 or Valid/3, it is no contradiction to assert that a
conflicting rule of morality is Valid/4.
However, why should it be assumed that there is only one field
of normative meaning to which scientific discourse can refer when
it uses the concept of Validity/2? The fact that Kelsen :makes this
assumption is indicative of an unfortunate metaphysical gloss on his
theory. He Says: " By the word ' validity' we designate the specific
existence of a norm." 11 And he refers to Law as a "realm" which
"stands over against reality," I or "juxtaposes itself to reality." 52
By such expressions he evokes a picture of normative space, in which
norms stand as objects each occupying its segment of non-reality.
that its contents are incompatible with some moral or political value." G.T.L.S.,
p. 113. "Law is always positive law, and its positivity lies in the fact that it is
created and annulled by acts of human beings, thus being independent of
morality and similar norm systems." G.T.L.S., p. 114. "The validity of a
positive legal order does not depend on its conformity with some moral system."
P.T.L., p. 67. (At P.T.L., pp. 17-23, Kelsen refers to a judgment that behaviour
accords with a valid norm as a " value judgment." This terminology is mis-
leading; but his description of the value as " relative " to the human source of
the norm makes it plain that legal norms are not being described as Valid/4.
".. . the statement 'law is moral by nature' does not mean that law has a
certain (sc: moral) content, but that it is norm-namely a social norm that men
ought to behave in a certain way. Then, in this relative sense, every law is
moral; every law constitutes a-relative-moral value." P.T.L., p. 65. Cf.
Kelsen, "Norm and value" (1966) 54 Cali.L.R. 1624.
48"GT.L.S., p. 373. P.T.L., p. 332.
49 G.T.L.S., p. 374. P.T.L., p. 329. Cf. Hart, n. 23, supra, pp. 772--778.
50 P.T.L., p. 10.
51 G.T.L.S., p. 121.
52 P.T.L., p. 213. "... value and reality belonging to two different spheres, just
as the ' ought' and the 'is ' "-P.T.L., p. 19. " Norms have a reality different
from natural reality-a social reality "-P.T.L., p. 86. " (Law is to be
contrasted with) causally determined reality "-P.T.L., p. 104.
If a norm is Valid/5, in the sense that it actually occupies a segment

of normative-meaning-space, it excludes any contradictory Valid/5
norm.
Criticism (i) rejected

In the light of this discussion of the concept of validity, the first
of the above criticisms of the judgments in the revolution cases may
be seen to be unfounded. The criticism was that, even if the courts
found the revolutionary change efficacious, they were not entitled to
declare revolutionary law "valid" by Kelsen's theory since, as Kelsen
says, efficacy is only a precondition of validity.5
This criticism suggests that Kelsen requires something other than
efficacy before a legal system can be said to be Valid/4. But his
view is that Validity/4 can never be justified -by legal science."
What Kelsen does say is that efficacy is not enough for Validity/2.
However, all that is required in order to confer Validity/2 upon an
efficacious legal system is a decision to interpret the laws of the system
as a logically consistent field of meaning. This is a decision which any
legal scientist will normally take (and, according to Kelsen, must take
if what he engages upon is to count as legal science). It is a decision
which entails no value commitment. As Kelsen points out: "even
an anarchist, if he were a professor of law, could describe positive law
as a system of valid norms, without having to approve of this law."
S. (4). WHEN DOES THE GRUNDNORM CHANGE?

Grundnorm not a fact
The persistent tendency to identify the grundnorm with a socio-
political phenomenon must be resisted, if an appraisal of Kelsen's
theory is to be on target.
In many societies there exists a widespread, more or less inarticu-
late, attitude of commitment towards a certain form and style of
53 n. 46, supra.
54 " From the standpoint of scientific cognition of positive law, its justification
by a moral order different from the legal order, is irrelevant, because the task
of the science of law is not to approve or disapprove its subject, but to know
and describe it. True legal norms, as prescriptions of what ought to be,
constitute values; yet the function of the science of law is not the evaluation
of its subject but its value-free description. The legal scientist does not
identify himself with any value, not even the legal value he describes." P.T.L.,
p. 68. Cf. G.T.L.S., pp. 48-50, and n. 47, supra.
55P.T.L., p. 218n. At G.T.L.S., p. 413, however, Kelsen says that an anarchist
would not ascribe "validity" to the positive legal order, and at Stanford,
p. 1144, he says that a communist would deny that a legal order of a capitalist
society is "objectively valid." These statements can only be reconciled with
the citation in the text if it is assumed: (a) that the anarchist of G.T.L.S. and
the communist of Stanford are not supposed to be acting as legal scientists; and
(b) " validity " in these contexts is Validity/4.
government, which is not inaccurately described in Austinian langu-

age as a "habit of obedience "- although Austin made tae mistake
of assuming that the attitude was directed towards a personal
sovereign. 56 In many societies there may also exist a more explicit
rule-idea, functioning as a motive for social action. For instance, the
rule that "whatever the Queen in Parliament enacts is law" is learned
and communicated in this or equivalent explicit verbal formulations
and, as a consciously articulated norm-formula, motivates official and
other social behaviour. H. L. A. Hart's "rule of recognition" is a
generalised notion of such rule-ideas functioning as motives, for social
action.5"
The grundnorm, however, is riot a social phenomenon like either
of these. It is not a phenomenon at all, but a hypothesis.57
Grundnormchanges when lawyers make new logical assumptions

Being merely a hypothesis of juristic thinking, the grundnorm does
not change the moment the revolutionaries shoot the King. It cannot
change until jurists change their thinking, that is, until lawyers begin
to make post-revolutionary assertions to the effect: "The law in the
country now is .... " where "now" refers to some revolutionary
established source of law.
While attitudes of commitment and even rule-ideas may gradually
56 "If a determinate human superior, not in a habit of obedience to a like

superior, receive habitual obedience from the bulk of a given society, that
determinate superior is a sovereign in that society, and the society (including
that superior) is a society political and independent." J. Austin, The Province
of JurisprudenceDetermined, Hart ed., 1954, p. 194.
57 " 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 ultimate criteria of validity must be generally
obeyed, and, on the other hand, its rules of recognition specifying the criteria
of legal validity and its rules of change and adjudication must be effectively
accepted as common public standards of official behaviour by its officials."
H. L. A. Hart, The Concept of Law, 1961, p. 113. Hart here uses " valid " in
the sense of valid/I. He is right (it is submitted) when he asserts (p. 107) that
the rule of recognition cannot be valid/i. But his criticism of Kelsen's state-
ment that the validity of the basic norm is presupposed as a " clark saying"
(p. 106) is misplaced, if Kelsen is here understood to be speaking in terms of
validity/2.
57a In his Israel Law Review article (n. 46, supra) at p. 6,Kelsen says that those
who consider an order as an objectively valid order "in their thinking,
presuppose the basic norm as the meaning of an act of will. Since, however,
an act of will does not exist in reality, but only in the juristic thinking of men
who interpret the coeicive order as an objectively valid legal order, the
presupposition of the basic norm is a typical case of a fiction in the sense of
Vaihinger's Philosophie des Als-ob." The grundnorm is postulated by Kelsen
as something logically essential to explain the practice of legal scientific
discourse. If the logical bases of this discourse can be explained in some
better way, the grundnorm is useless. But there can be no question of its
being true or false; i.e., no amount of empirical inquiry can establish that it
does or does not exist as a psychological or sociological phenomenon. Cf.
H. Vaihinger, Philosophy of " As If," 1924, pp. 85-90.
The Cambridge Law Journal 1[1971]
develop, the grundnorm will not change at all so long as legal scientists
give primacy to the same ultimate sources in their logical arrangement
of legal material.
Kelsen generally distinguishes ultimate sources into two kinds
only: written constitutions and customary constitutions." This
entails that there can be only two kinds of grundnorm: one authoris-
ing the "fathers" to promulgate the written constitution at some past
date, and one which authorises the population at large from time to
time to create a material constitution by custom.
Hence, where -the constitution is written, there will be a change in
the grundnorm (a revolution) if jurists begin to deduce laws from
some newly promulgated constitution. The new basic norm will differ
formally from the old only in that ihe personnel constituting the
"fathers" will be different, for they will be the promulgators of the
new written constitution. There will also be a change in the grund-
norm if the written constitution is abandoned and jurists begin to
refer to as "laws" acts of legislation made in some new way sanc-
tioned by custom.
IIf the constitution is customary to begin with, a change in the
grundnorm will only occur if a written constitution is substituted.
Thus, in Kelsen's terms, the decision of the House of Lords in 1966
not to be bound by its own decisions did not constitute a change in
the grundnorm (revolution), and neither did the "revolution" of 1688.
In the terminology suggested above, however, the 1966 decision did
represent a change in an important "rule-idea," and the 1688 revolu-
tion did represent a change in a basic "attitude of commitment."
To achieve a more discriminating logic for revolutions, it would
be necessary to make a more detailed taxonomy of ultimate sources
and grundnorms than Kelsen's formal distinction between written and
customary constitutions. For instance, customary constitutions could
be subdivided according to the kind of social group which the grund-
norm effectively authorised to dictate constitutional developments; and
written constitutions, by reference to ihe kind of group which pro-
mulgated them-were they, for instance, socialist or non-socialist
"fathers "?
Legal science and efficacy

No change in the grundnorm can occur until legal scientists make
a fresh presupposition. But this, according to Kelsen, they do not
58 If there is a written constitution which does not specify custom as a source of
law, and yet custom is actually treated as such (n. 33, supra), the " material
constitution " of the society in question will presumably count as a "customary
constitution" for the purposes of the above distinction only if custom is
regarded as an effective means of varying the written constitution.
C.L.J. CChange of Grundnorm
and (if they are to fulfil the role appropriate to legal scientists) can-
not do without the pre-condition of efficacy. 59 The legal scientist
does not have an arbitrary choice."
But why does efficacy dictate the legal scientist's choice of grund-
norm? A true evaluation of the judgments in the revolution cases
requires an answer to this question. An answer to it will reveal the
philosophic core of Kelsen's theory.
Kelsen assumes that legal science-which, for him, consists in the
logically consistent arrangement of normative legal material emanating
from human legislators-is a socially useful activity. "The possibility
and necessity of such a discipline directed toward the law as a norma-
tive meaning is proved by the fact that the science of law has been in
existence for millennia-a science which, as dogmatic jurisprudence,
serves the intellectual needs of those who deal with the law." 61
If legal science is to be a socially useful activity, it must purport to
describe "law" which is both positive (laid down by human beings)
and by and large effective. Hence, "... the basic norm can only
establish a law-making authority whose norms are, by and large,
observed, so that social life broadly conforms to the legal order based
on the hypothetical norm." 2 "That means: we presuppose the basic
norm only if there exists a coercive social order by and large
effective." e
Thus, the grundnorm changes when legal scientists make a new
basic presupposition; but, as legal scientists, they must do, this when
the legal norms which are by and large effective within a territory can
only be interpreted as a consistent field of meaning if a new grund-
norm (authorising a new ultimate source of law) is presupposed.
Efficacy, a practical test?

Is it possible and practicable to make efficacy the test of a change
in the grundnorm? The second criticism of the revolution cases
asserts that, whatever may be the position for a settled legal order,
59 G.T.L.S., p. 437; P.T.L., pp. 46-48, 86-87. Stanford, p. 1142.

60 P.T.L., p. 201 ; Stanford, p. 1143.
61 P.T.L., p. 105. " ... the aim of this general theory of law is 1o enable the
jurist concerned with a particular legal order, the lawyer, the judge, the
legislator or the law-teacher, to understand and to describe as exactly as possible
his own positive law . . ."-G.T.L.S. xiii. Glendon Schubert, a protagonist of
"behavioural jurisprudence" (the successor movement to the school of American
realists), has written: " . . candor compels the admission that the older
mechanical jurisprudence remains the overwhelmingly dominant metaphor
among judges themselves, practising lawyers, journalists and the public."
Schubert, " Behavioral jurisprudence" (1968) 2 Law and Society Review 407,
409. By " mechanical jurisprudence" Schubert is referring (disparagingly) to
what Kelsen seeks to explain and justify under the approbative title of "legal
science."
62 G.T.L.S., p. 437.
63 Stanford, p. 1142.
The Cambridge Lw Journal [19711
no objective factual judgment as to efficacy can be made soon after

a revolution has taken place.
Kelsen, in his Stanford Law Review article, makes it plain that
the identification of effective legal norms does not itself require pre-
supposition of the grundnorm. Only the attribution to them of the
quality Validity/2 requires this presupposition." The discovery of
effective legal norms is a step logically prior to the presupposition of
the grundnorm. "The positivistic jurist., when he establishes the basic
norm, is guided by the tendency to recognise as objective law the
greatest possible number of empirically given acts the subjective
meaning of which is to be legal acts.""
If there are in a territory commands, permissions and authorisa-
tions which are issued by persons who purport to issue them as legis-
lators and which are effective, then the meaning-contents of these
commands, permissions and authorisafions are norms to which the
legal scientist can and must attribute Validity/2-he can and must
presuppose a basic norm which confers legislative authority upon
those who purport to enact them.
The effectiveness of each individual norm is to be measured,
according to Kelsen, by two criteria: first, by whether the norm is
"obeyed," that is, by whether the conduct (the opposite of which is
made the condition of the sanction stipulated by the norm) is per-
formed; secondly, by whether, if the noim is not obeyed, the sanction
is applied by the official whom the norm directs to apply it." These
are questions of fact. It is true that soon after the occurrence of a
revolution they may be future questions of fact, but that does not
make apolitical judgments about them impossible or impracticable,
only more subject to error.
What is meant by saying that norms, to be effective, must be
64 "The function of the basic norm is not 1:omake it possible to consider a

coercive order which is by and large effective as law for-according to the
definition presented by the pure theoy of law-a legal order is a coercive order
by and large effective; the function of the basic norm is to make it possible to
consider this coercive order as an objectively valid order. . . . The basic norm
does not answer the question as to whether the existing order is a legal order
but the question as to what is the reason for the validity of this legal order."
Stanford, pp. 1144-1146. Earlier in the same article, Kelsen criticises Stone
for suggesting that Kelsen's theory stipulated as one of the conditions for the
validity of a single norm the membership by that norm of a system of norms.
It is submitted that it is Kelsen who is at fault here, in that he slides from one
sense of "validity " to another. In the passage quoted in this footnote,
"validity" = Validity/2. In the earlier passage (pp. 1139-1141) "validity"
sometimes means Validity/l, as where Kelsen states that a statute is valid as
soon as it is enacted and before it can become effective; and sometimes
"validity " means Validity/3, as where Kelsen states that a norm loses its
validity if it fails to become by and large effective.
65 G.T.L.S., p. 437.
66 G.T.L.S., p. 62, P.T.L., pp. 11, 116.
obeyed or applied "by and large? " " It would be impracticable to

say, in the case of many norms, whether the sanction was applied
more often than not when disobedience to the norm occurred, because
statistical evidence of the number of cases of disobedience could not
be obtained. It would be impossible to say whether the norm was
obeyed more often than not, without first postulating what was to
count as obedience. Kelsen says: "Law is observed by that
behaviour to whose opposite is attached the coercive act of the sanc-
tion" "sand that motive is irrelevant. 9 Then laws which prohibit the
performing of acts which most people do not in any case perform
(such as laws against witchcraft) must always be generally "obeyed ";
in other words, one of Kelsen's criteria of obsolescence can never
apply to them.
It is submitted that the "by and large" test should be reformu-
lated in the following way. The first criterion (obedience) should 'be
eliminated, and the test should run: "a norm is to be judged effective
if the official acts of application of sanctions bear a socially significant
ratio to the recorded acts of disobedience."
The test need be no more precise in order to choose between the
effectiveness of competing norms issued by rival legislative authorities.
In the revolution cases it was manifestly possible to predict, at the
relevant dates, that official acts of sanction would occur in accordance
with revolutionary norms, and that these official acts would be in a
socially significant ratio to acts of disobedience to the norms.
Kelsen indicates that there must be some degree of permanence to
the effectiveness of legal norms before the legal scientist is justified in
ascribing Validity/ 2 to them."0
The question of how much turns, presumably, on whether there
seems to be sufficient permanence about the new norms to make works
of legal science-which, by presupposing a new grundnorm, confer on
them Validity/ 2-worthwhile (in a practical, not a moral, sense). Was
it, at the dates when the Madzimbamuto and Ndhlovu cases came
before the Rhodesian Appellate Division," worthwhile to write a text-
book of, say, Rhodesian criminal law which presupposed a grundnorm
that conferred legislative authority on the rebel legislature? An
affirmative answer seems inescapable.
If a -revolution is literally in balance so that it is an even bet
whether the norms of the old order or those of the rival revolutionary
order will be effective during, say, the next decade, efficacy alone will
67 G.T.L.S., pp. 118, 437. P.T.L., pp. 46, 86, 202. Stanford, pp. 1139-1142.
68 P.T.L., p. 236.
69 G.T.L.S., p. 40. P.T.L., p. 24. 70 P.T.L., pp. 47-48.
71 Judgments were delivered on 29 January 1968 and 13 September 1968 respectively.
The Rhodesian unilateral declaration of independence was made on 11 November
1965.
The Cambridge Law. Journal [19711
clearly not answer the question: which grundnorm ought a legal

scientist to presuppose? He might then either fold his hands and
refuse to make statements about what the law is " or he might make
statements in the alternative: "if the revolution succeeds, the law on
this point is ... , " "but if it fails, the law is....." There can be no
question of burden or degree of proof, or presumptions, on this issue
of fact, unless these are based on some general philosophical principle
of legal conservatism.
If legal scientists honestly differ as to whether the revolution will
succeed, the question whether the grundnorm has or has not changed
is nonetheless "objective" in the same way that any judgment about
future matters of fact is objective. No such judgment can be made
with certainty and many such judgments turn out to have been wrong.
Nevertheless, the presupposing of a grundnorm, which turns on such
a judgment, does not entail any decision of political commitment.
It is accordingly submitted that the second criticism of the
revolution cases is mistaken.
Efficacy and the judge

The third criticism of the revolution cases asserts that, whatever
may be the position of a legal scientist, a judge cannot make an
objective determination of the efficacy of revolutionary norms soon
after the revolution has occurred, because his own decision is an
element in the very efficacy which is to be determined upon.
It is true, as we have seen, that official action is an important
element in the efficacy of norms. It is submitted, however, that
whether such a criticism of a judge is well founded depends on the
circumstances. It depends on the relative importance of his decision
as against other present and future elements in the efficacy of the revo-
lutionary norms. If a judge believes that the success of the revolution
may turn on what decision he gives in a case before him, then clearly
he cannot decide as to the efficacy of the change without first making
a political choice whether or not to join the revolution. If he believes,
however, that whatever he decides, the revolution is likely to succeed
(if need be, 'by his dismissal and the appointment of an acquiescent
judge), then his decision that the revolution will be efficacious is not
necessarily politically motivated.
No one has suggested that the success of the revolutions in
Pakistan or Uganda turned on judicial approval, but some writers do
appear to suggest that this was the case in Rhodesia.73
72 The course recommended in the Annual Survey of Commonwealth Law (1967),
p. 93.
73.R. W. M. Dias, n. 10, supra, p. 257, has suggested that if the Rhodesian judges
had stood out against the 1965 constitution-which was set up by the rebel
C.L.J. -Change of Grundnorm
This was not the view of the judges in the two leading cases in
which the legality of the revolutionary r6gime was considered by the
Rhodesian Appellate Division. Beadle C.J., in Madzimbamuto. v.
Lardner-Burke N.O., stated that the decision of the court would
make little difference to the issue of success or failure. 4 He and
Jarvis A.J.A. concluded on the facts that the revolution seemed likely
to succeed.7" Qu~net J.P. and Macdonald J.A. concluded that the
revolution had already succeeded."6 Fieldsend A.J.A. expressed the
view that, whether or not the revolution had succeeded, no judge
appointed under the pre-revolutionary constitution ought to accord
legality to the new constitution.7 In R. v. Ndhlovu the court found,
as a conclusion of fact, that the revolution had succeeded.78
It is difficult to say whether the opinions expressed by the
Rhodesian judges in these and other constitutional cases " were honest.
and apolitical.8 0 But that there was plenty of evidence in 1968,
suggesting that the Rhodesian rebellion would succeed can scarcely be
denied.
On the basis of the foregoing discussion it is submitted that the
question: when does the grundnorrn change? can, in principle, receive
an apolitical, fact-based reply from a legal scientist, even when it is
asked soon after the occurrence of a rebellion; and, in practice, the
question can receive the same kind of reply from a judge in loco,
because his answer to it will not affect the success of the rebellion. The
grundnorm changes when the legal norms which are by and large
effective within a territory change in such a way that a legal scientist
rdgime-on grounds of morality or justice, they might well have forced the
"new grundnorm to incorporate the principle for which they stood out. Cf.
F. M. Brookfield, n. 12, supra, p. 344, who impliedly criticises the view that
the decision of the Rhodesian judges would make little difference to the success
of the revolution. The same point was made by counsel for the appellant when
Madzimbamuto's case came before the Privy Council, [1969] A.C. 545, 680.
T4.(1968) 2 S.A. 284, 321.
T5 Ibid. 326, 418.
76 Ibid. 369, 415.
77 Ibid. 430, 432; cf. Beadle C.J. in R. v. Ndhlovu (1968) 4 S.A. 51S, 522. For
discussion of the argument that judges cannot, or ought not to, pronounce upon
the validity of the constitution under which they were appointed, see R. S.
Welsh, "The constitutional case in Southern Rhodesia " (1967) 83 L.Q.R. 64,
73, 75, 81, 87-88; Palley, n. 80, infra, pp. 280-284; Annual Survey of Common-
wealth Law, 1968, pp. 108-113; Honor6, n. 8, supra, pp. 273-278: De Smith,
n. 9, supra, pp. 93-110, L. J. Macfarlane, "Pronouncing on Rebellion: The
Rhodesian Courts and U.D.I." [19681 P.L. 325, 355-361; Dias, n. 10, supra,
pp. 238-239; Brookfield, n. 12, supra, pp. 330-340. The argument isconsidered
in the text at the end of s. 5, infra.
78 (1968) 4 S.A. 515, 531 (Beadle C.J.); 542 (Qu~net J.P.); 554 (MacDonald J.A.).
19 Dhlamini v. Carter N.O. (1968) 2 S.A. 445; Madzimbamuto v. Lardner-Burke
N.O. (No. 2) (1968) 2 S.A. 457; Dhlamini v. Carter N.O. (No. 2) (1968) 2 S.A.
474; Dhlamini v. Carter N.O. (No. 3) (1968) 2 S.A. 467.
80 For the personal and political backgrounds of the Rhodesian judges, see the
article by Mrs. Claire Palley, " The Judicial Process: U.D.I. and the Southern
Rhodesian Judiciary" (1967) 30 M.L.R. 263.
The Cambridge Law Journal [19711
can only interpret the content of these by and large effective norms
as a logically consistent field of meaning by presupposing a new
grundnorm.
What legal norms are by and large effective at any given time may
be discovered in the following way: first, by recording what com-
mands, permissions and authorisations (stipulating sanctions) have
been issued (and not repealed) by a person or body purporting to act
as legislator; secondly, by recording (or predicting) occasions on which
the stipulated sanctions have been (or are likely to be) applied by
persons purporting to act as state officials; thirdly, by recording (or
predicting) acts of disobedience, that is, acts specified as conditions for
the application of the stipulated sanctions to the actor. If there is
a socially significant ratio between the official acts and the acts of dis-
obedience, and it can be predicted that this ratio will continue to
obtain for a reasonable length of time, the meaning-contents of the
commands, permissions and authorisations are by and large effective
norms.
S. (5). WHY DOES THE GRUNDNORM CHANGE?
As appears from what has been said above, the reason why the grund-
norm changes when it does is as follows: only by presupposing the new
grundnorm can the legal scientist fullil his socially useful role of
describing the law actually in force as a logically consistent field of
meaning. If anyone were to write a textbook of Rhodesian law today
presupposing the old grundnorm and the:refore assuming the Validity/2
of the United Kingdom legislation of 1965 (which nullifies all laws
passed and all administrative actions undertaken by the rebel
l'gime), s1 he might be considered to be performing a praiseworthy
81 The Southern Rhodesia (Constititution) Order 1965 (1965 S.I. 1952), was passed
under the Southern Rhodesia Act 1965. S. 2 (1) of this order provides: " It is
hereby declared for the avoidance of doubt that any instrument made or other
act done in purported promulgation of any constitution for Southern Rhodesia
except as authorised by act of Parliament is void and of no effect." S. 3 (1)
provides: " No laws may be made by the legislature of Southern Rhodesia, no
business may be transacted by the legislative assembly and no steps may be
taken by any person or authority for the pu:rposes of or otherwise in relation to
the constitution or reconstitution of the legislative assembly or the election of
any person to be a member thereof .. " S. 4 (1) (b) suspends various provisions
of the 1961 Constitution, relating to the functioning of the executive in Southern
Rhodesia. S. 6 provides: " It is hereby declared for the avoidance of doubt
that any law made, business transacted, step taken or function exercised in
contravention of any prohibition or restriction imposed by or under this order
is void and of no effect." The majority of the Privy Council in Madzimbamuto's
case interpreted these provisions to mean that no body in Southern Rhodesia
can make law, no matter how necessary for preserving law and order or for any
other purpose, [1969] A.C. 645, 729-731. Lord Pearce, in his dissenting judg-
ment, said that the order should be interpreted in the light of the directive
issued by the Governor soon after U.D.I., which required officials to continue
with their " normal tasks." On this basis he concluded that the order was
intended to nullify only acts done in purported exercise of powers under the
piece of propaganda against a racist regime, but he would not be

acting in the role of a legal scientist, simply because he would not be
collating the laws actually in force in Rhodesia.
Implications of the pure theory for a judge

However, granted that the legal scientist, if he is to fulfil his
traditional role, must presuppose the effective grundnorm, what
warrant does Kelsen's theory give the judge qua judge to presuppose
it? If we leave aside Kelsen's dogmatic views concerning the primacy
of international law,8" the answer is: directly, it gives none, and to
this extent the fourth criticism of the revolution cases is well founded.
The pure theory of law is intended to describe the science of law-
its aims, its a priori formal assumptions, and the logical status of the
judgments contained in it. Kelsen says that every act of law-
application is an act of law-creation, in the process of concretisation
and individualisation of norms.8" The judgment of a court is there-
fore simultaneously an act applying the law and an act creating law
and, to the extent that it creates law, it is an act of the will.8 4 The
act of presupposing the grundnorrn, on the other hand, he describes as
an act of cognition. "The science of law has to know the law--as
it were from the outside-and to describe it. The legal organs, as
legal authorities, have to create the law so that afterwards it may be
known and described by the science of law." 88
Thus the theory does not directly warrant the judge in presup-
posing any grundnorm; for jurisprudence, says Kelsen, is not a source
of law."
Nevertheless, it is submitted, the pure theory does have indirect
social suggestive force for judges. The theory assumes that legal
1961 Constitution, and consequently that it was possible for the courts to
recognise some of the laws enacted by the rebel legislature otherwise; than under
that constitution-ibid. 742-745.
82 G.T.L.S., pp. 121, 220-221. P.T.L., pp. 214-217. Kelsen believes: (a) that the
principle of effectiveness, according to which only the law effective within a
territory is valid law in that territory, is a rule of positive international law--an
extreme version of the declaratory (rather than constitutive) effect of recogni-
tion of states; (b) that systems of municipal law are subordinate to, and cannot
contradict, positive international law. If these opinions were correct it would
follow that judges in a municipal court would be directly authorised to pre-
suppose new grundnorms following successful revolutions. However, the first
of these opinions is highly controversial and the second is true of some juris-
dictions but not of others. Kelsen does not base these opinions on the evidence
of positive law. He deduces them from a metaphysical interpretation of the
concept of validity-see supra, p. 115.
83 G.T.L.S., pp. 132-133; P.T.L., p. 234.
84 P.T.L., p. 10; Stanford, pp. 1136-1137.
85 P.T.L., p. 204n.
86 P.T.L., p. 72.
87 G.T.L.S. xiv, p. 163. At Stanford, p. 1134, Kelsen rejects Stone's suggestion
that he (Kelsen) had expressed the " foolish opinion" that propositions of the
pure theory of law bind the judge in the way in which legal norms bind him.
science is socially useful, which is precisely the reason for insisting

that legal science should follow efficacy. Clearly legal science of the
sort described by the pure theory can only continue to be socially use-
ful so long as judges also indulge in it. If judges ceased altogether to
think of their decisions as subsumptions under general rules, if 'they
persistently ignored the hierarchy of constitution, statute, contract, etc.,
this sort of legal science would become pointless.
To assert that at the present time judges in developed legal systems
do act as legal scientists is a correct description of their verbal be-
haviour, although to what extent their actual decisions are determined
by factors other than what they say about "the law" is uncertain and
controversial. The pure theory assumes that this legal-science verbal
behaviour is not only what judges do indulge in, but also what they
ought to indulge in. Kelsen's assertion that: "What sociological
jurisprudence predicts that the courts will decide, normative jurispru-
dence maintains that they ought to decide," "8is true if we understand
"say " for "decide."
Thus, in so far as judgments are acts of law-application, the judges
are assumed to act as legal scientists, and in that capacity they do and
ought to presuppose the grundnorm.
A further implication of the theory, as of all positivist theories of
law, is that, to the extent that the solution of a particular case given
by the science of law is clear (that is, to the extent that the judge has
no discretion within the meaning of the relevant higher norms), the
judge ought to apply that solution. If judges acted as legal scientists
up to the point at which they found that solution X was given, and
then regularly opted for solution Y, again legal science would become
pointless. Hence, that it would be inconsistent with the role of a
judge to do this, is implied by a theory which 'assumes that. legal
science is useful.
The pure theory of law therefore has three relevant implications
for the' role of a judge: (a) judges do act as legal scientists; (b) they
ought so to act; (c) to the extent that legal science gives .a clear
solution to a particular case, the judge ought to accept that solution
as the basis for his decision.
Thus, where a revolution is, or is predicted to be going to be,
successful, Kelsen's theory (directly) recuires the legal scientist, acting
in his role as legal scientist, to presuppose a new grundnorm, and
(indirectly) suggests that a judge, acting in his role as judge, ought to
do the same. It does therefore seem that the theory warrants a judge,
who wishes to avoid express political commitment (" joining the
88 G.T.L.S., p. 172.
revolution "), in accepting the laws of a successful revolutionary

regime.
Loyalty to constitutions
It can, however, be argued that it is not consonant with the
role of a judge who has been appointed under one constitution
to accept the authority of any other constitution. It is not that
there is anything illogical in such a step. The rules conferring
adjudicative power on a judge are legal rules which can be the subject
of judicial determination like any other legal rules. Courts frequently
pronounce upon their jurisdiction in particular cases, and there is no
reason, in logic, why a court should not embark on an inquiry as to
whether or not there has been a change in the grundnorm, even
though, until the inquiry is completed, it is uncertain whether the
court's jurisdiction rests on old grundnorm rules or new grundnorm
rules.
. The point is that judges are normally thought of as upholders of
constitutions, not co-operators in rebellions. Ought not loyalty to the
constitution which made him a judge to outweigh his ordinary duty to
accept the clear ruling of legal science?
This is a general question touching the ethics of judgeship to which
there is no easy answer. It seems likely that a positivistic philosophy
of law would answer it in the negative, for the following reason: if the
revolution is successful, a loyal judge can only resign, whereas a legal-
science judge can continue with his useful role.
However, even if one admits, that a judge qua judge ought to
accept the laws of a successful revolutionary r6gime, this legal duty
may, in particular cases, be outweighed by other extra-legal duties. It
may be outweighed by a political duty not to give support to an
immoral rdgime or by a personal moral duty to observe a judicial oath.
A revolutionary upheaval is just the sort of situation where being a
good judge may have to give way to being a good citizen or a good
man.
S. (6). OUGHT EFFECTIVENESS TO BE ENOUGH?
Apart from the four criticisms mentioned at the beginning of this

article of the judgments in the revolution cases alleging misapplication
of Kelsen's theory, the theory itself has been criticised for making
effectiveness the sole test of whether a revolution has changed the
law.8 9 This is a fundamental objection to all positivist theories of law-
for it is axiomatic to all such theories that there is no point in refusing
89 Palley, n. 80, supra, pp. 280-284. HonorS, n. 8, supra, pp. 273-278. Dias,
n. 10, supra, pp. 243-244, 256-258. Macfarlane,.n. 77, supra, pp. 331-335.
The Cambridge Law Journal [1971J
to use the word "law" to describe any and every relatively permanent
and exclusive organisation of force within a territory, however evil it
may be.
Positivists stoutly maintain that to call something "law" or
"legally valid" implies no sort of approval, but is indeed consistent
with outright condemnation on moral or political grounds. It is sub-
mitted that they do not thereby commit themselves to a contradiction.
It is perfectly meaningful to say "(such and such) is an atrocious legal
rule;" or "the legal order presently in force in (X) is an appalling
tyranny."
On the other hand, it cannot be denied that in practice words like
"law" and "legality" do function as "titles of honour." 10 Politi-
cally (perhaps even ethically) they do ring in people's ears with an
approbative association. If it were not so, the internal " legality" of
the Smith regime would not be a political issue, that rdgime would not
have submitted to the adjudication of the Rhodesian courts on the
question of its "legality," and the British Government would not
continue to insist that the rdgime is internally "illegal."
Morality and the definition of law

Positivists are undoubtedly right when they assert that the defini-
tion of "law" need contain no reference to substantive moral prin-
ciples. Are they also right when they assert that it ought not to?
The answer to this question depends on our judgment on two
general issues of legal philosophy: first, is it possible to spell out a
moral criterion which would give a meaningful guide as to the
difference between a morally acceptabte organisation of state force
(" law "), and one which was not morally acceptable (" non-law ")?
Secondly, if such a criterion can be found, would the political advan-
tages to be obtained by maintaining the distinction outweigh the
practical disadvantages of blurring the merely descriptive function
which positivism attributes to legal science?
These are the fundamental questions raised by the revolution
cases. If an affirmative answer can be given to them a fundamentally
anti-positivist position would be attained which would be far more
significant philosophically than the sapping operations mounted
against positivism in recent years. The burden of such criticisms has
been that positivists, by insisting on a value-free explication of the
concept of a legal system, have pointed theoretical discussion away
from areas in which the day-to-day operation of a legal system does in
fact include consideration of value issues. For instance, the positivist
ch'aracterisation of law in terms of prescriptions or rules suggests, it
90 Ross, On Law and Justice, 1958, p. 31.

has been said, that a judge's role consists either in applying law to
clear cases or in creating law through the exercise of an unfettered
discretion in unclear cases; whereas in fact judges, in unclear cases, do
actually invoke the purpose of the particular rule "I or more general
principles and policies.9 2 Or again, positivists' explanations of law in
terms of organised coercion neglect, it has been said, the incidental
values which "organisation" necessarily entails, even where the sub-
stantive aims of an organisation are immoral.9" Supposing these criti-
cisms to be just, they do not touch the citadel of legal positivism,
which is the assertion that there are relatively permanent territorial
organisations of force through the medium of rules which, whatever
their moral differences, have such amoral similarities that tlhey ought
for practical purposes to come under a common heading--" law."
Are these practical purposes outweighed by the political advan-
tages of including a value criterion in the definition of law, and, if
so, what is the criterion?
Supposing, for instance, that a value of being non-racially-
oppressive were built into the definition of "law" by those who have
regularly to do with the exposition of law either as legal educators or
legal administrators? Could this moral criterion be defined with
sufficient exactness to give a reasonably clear guide when applied to
individual legal systems? If it could, would the political advantages
of refusing to call what happens in, say, Southern Africa "law" be
worth the cost?
If university libraries refused to catalogue textbooks on South
African law under "law," but insisted on some new classification to
describe the category "racially oppressive, but relatively permanent,
organisation of force," and if we added to the conflict of laws a
general rule that no effect was to be given to such non-legal systems,
would it be worthwhile?
If it had the effect of weakening racialism, many might agree that
it would. Would it have this effect?
English courts and Rhodesian "law"

There are two reasons why English courts cannot attribute the
quality of "legality" to any of the legislative or administrative acts
of the Smith regime, both of which flow from the English constitu-
91 L. L. Fuller, " Positivism and Fidelity to Law-Reply to Professor Hart"

(1958) 71 H.L.R. 630, 661-669.
92 R. M. Dworkin, "The Model of Rules " (1967) 35 University of Chicago L.R. 14.
J. C. Smith, " Law, Language and Philosophy " (1968) 3 U.B.C.L.R. 59, 107-
112, W. L. McBride, "The Essential Role of Models and Analogies in the
Philosophy of Law " (1968) 43 N.Y.U.L.R. 53, 72-82.
93 Fuller, n. 91, supra, pp. 643-648; The Morality of Law, 1964, pp. 33-44, 46-95,
106-118, 145-151.
C.L .- 5
The Cambridge Lw Journal [1971]
tional doctrine of indivisible sovereignty and not from any general

principle that "laws" must satisfy a moral criterion. First, no acts
of a government to which our own government has not accorded at
least de facto recognition can be recognised.
In Carl Zeiss Stiftung v. Rayner and Keeler Ltd. (no. 2)9' Lord
Reid stated the general principle as follows:
In the normal case a law is made either by the sovereign
directly or by some body entitled under the constitution of the
country to make it or by some person or body to which the
sovereign has delegated authority to make it. On the other hand
there are many cases where laws have been made against the will
of the sovereign by persons engaged in a rebellion or revolution.
Then until such person or the government which they set up have
been granted de facto recognition by the government of this
country, their laws cannot be recognised by the courts of this
country, but after de facto recognition such laws will be
recognised."
In the same case, however, Lord Wilberforce said that the court
should shrink from the view that non-recognition of a foreign govern-
ment by our government entailed a legal vacuum, quoting Locke's
words: "A government without laws is, I suppose, a mystery in
politics, inconceivable to human capacity and inconsistent with human
society." Lord Wilberforce left open the question whether English
courts might not accord recognition to some of the law-creating and
law-applying acts of an unrecognised government.9"
However, even if this doctrine could be applied to Rhodesia, there
is a second reason why none of the acts of the Smith r6gime can be
considered legal by English courts. The legislative sovereignty of
Parliament is unlimited. It may enact that "laws" in force in any
part of the world are not to be treated as laws by English courts,97 and
94 [1967] 1 A.C. 853, where the House of Lords held that the acts of the govern-
ment of the German Democratic Republic could be recognised, as that govern-
ment had power delegated to it by the government of the U.S.S.R. (which our
government had recognised as the sovereign authority for East Germany).
95 [19671 1 A.C. 853, 901.
96 [1967] 1 A.C. 853, 953-954.
97 S. 4 of the Statute of Westminster 1931, which purports to restrict Parliament's
power to legislate for the dominions could, in theory, be repealed-British Coal
Corporation v. R. [1935] A.C. 500, 520 (Lord Sankey). In Madzinmbamuto's
case, Lord Pearce (in his dissenting judgment) added to his somewhat strained
construction of the 1965 legislation-n. 81, supra-a comment which entails
that parliamentary sovereignty could be l[mited in some circumstances. He
said [1969] A.C. 645, 745: " there is no indication in the Order in Council that
it intended to exclude the doctrine of necessity or implied mandate by enjoining
(inconsistently with the Governor's directive) continuing disobedience to every
act or command which had not the backing of lawful authority. Even had it
done so, I feel some doubt as to how far this is a possible conception when
over a prolonged period no steps are taken by the Sovereign himself to do any
acts of government and the results would produce a pure and continuous chaos
or vacuum." If this opinion were ever accepted by the courts it would represent
a unique departure in English constitutional law.
it has enacted that any legislative or administrative acts of' the Smith
r6gime are without legal effect.9 8
Hence it was recently decided in Adams v. Adams 11 that a woman
who obtained a decree of divorce in Rhodesia, pronounced by a judge
appointed after U.D.I., must be regarded as still married and unable
to remarry. It was said in that case that Rhodesia was being sub-
jected to a legal blockade parallel to the economic blockade of the
sanctions policy, designed to make the rdgime capitulate.' The general
issues under discussion here amount to asking whether it is worthwhile
opening a general legal blockade on all systems of "law" which share
some defined criterion of immorality. It is submitted that it would not
make anyone capitulate and that, even as a gesture, it would have
little effect if this step were taken in isolation from all others. (This
is, of course, not to deny that there are good political grounds for the
combined economic and legal gestures in the particular case of
Rhodesia.)
It is therefore submitted that the general attack on positivism
arising out of criticism of the revolution cases fails.
S. (7). CONCLUSION
Kelsen's, theory describes legal science as that practical discipline
which expounds as a logical unity any effective r6gime of force.
The grundnorm is the hypothesis which closes up the arch of legal
logic. It enables the legal scientist to interpret the legal norms
effective within a territory as a logically consistent field of meaning.
Its actual content in any particular territory must include a reference
to all the fundamental sources from which the validity (Validity/ 1) of
individual legal norms is derived. Some such hypothesis is needed to
justify the ascription of absolute validity within the system (Validity/2)
to the contents of the individual norms, which is made by the legal
98 n. 81, supra.
99 [1970] 3 W.L.R. 934. A superior court could, in theory, reach a different
conclusion if (a) it adopted the interpretation of the effect of non-recognition
suggested by Lord Wilberforce in the Carl Zeiss case (n. 96, supra); and (b)
it preferred the interpretation of the legislation of 1965 suggested by Lord Pearce
to the interpretation of the majority of the Privy Council in Madzimbamuto's
case, n. 81, supra. [And see note in Case and Comment, supra, p. 37, Ed.]
1 At p. 951. In view of his understanding of the purpose of the 1965 legislation to
be " blockade," it is difficult to understand why Sir Jocelyn Simon P.. did not
accept the submission that all decrees issued by judges purporting to act under
the rebel constitution should not be recognised. On 7 April 1970, the Senior
Registrar of the Probate Divorce and Admiralty Division issued a practice
direction by which no grant of representation made by the High Court of
Rhodesia after 13 September 1968 (the date of the Ndhlovu decision) will, now
be resealed by the High Court of Justice in England-1970] 1 W.L.R. 687.
Current United Kingdom legislation has conferred on the English courts juris-
diction to grant divorces to unfortunate women in the position of the petitioner
in Adams' case.
scientist when -he interprets these contents according to the logical

canons of subsumption and non-contradiction.
The grundnorm changes when the legal norms effective within a
territory change in such a way that a legal scientist can only interpret
them as a logically consistent field of meaning by presupposing a new
grundnorm which refers to new sources of law. It changes then
because, only if he describes effective legal norms, does a legal scientist
fulfil his role.
The judges in Pakistan, Uganda and Rhodesia were acting properly
in the role of legal scientists when they found that the revolutionary
r6gimes were legal, because they were confronted with very strong
evidence that, whatever decisions they reached, the revolutions would
be successful.
It is true that Kelsen's theory does not directly authorise a judge
to make any particular decision. But indirectly it suggests that, when
legal science gives a clear solution to a case, the judge ought to adopt
that solution, and this is true when, soon after the occurrence of a
revolution, the question arises: has the grundnorm changed? The
reason why it has this suggestive force for a judge is that the theory
assumes that legal science is a socially useful activity, which it could
not be if it were not an essential part of the role of a judge to act as a
legal scientist and to apply the conclusions of legal science.
The theory thus implies that a judge is under a legal duty-that is,
a duty consonant with judgeship-to accept successful revolutions. It
seems probable that this duty is not outweighed by any general legal
duty of constitutional loyalty. It may, however, like any other legal
duty, be outweighed in particular ins:ances by moral or political
duties.
No fundamentally anti-positivist theory, which would deny the
characterisation "legal" to systems lacking some specific moral
element, can be sustained. It would have practical disadvantages and
few, if any, political advantages.
Kelsen's theory (like any other positivistic theory of law) entails
the conclusion that the internal law of Rhodesia, for the purposes of
legal scientific works (textbooks on Rhodesian law), is that law which
is effective there today in accordance with the constitution promul-
gated by the rebel rdgime-although this is not so for the purposes
of English conflict of laws nor, except on an extreme version of
the declaratory theory of recognition, for the purposes of public
international law.
However, legal science is not the noblest activity of the human
spirit, but merely a practical discipline with useful, but limited, aims.
The history of natural law theory consists of attempts to overload it

with noble endeavours which it could not advance.
The fact that one is acting in the traditional role of a legal scientist
if one describes effective racist law as "valid law," or that one is
acting in the traditional role of a judge if one applies it according to its
content, is not the last word in political justification of one's activity.
Political opponents of the Smith rdgime will approve the decision on
the part of two of the Rhodesian judges to resign, rather than to con-
tinue to act as judges under that r6gime.2 If they are positivists, they
will have done so precisely because "acting as judges" entails carry-
ing out ministerial functions in the service of the regime of force
actually established.
If there is a single political maxim underlying legal positivism, it
is that: "propaganda and gestures apart-' legality' without force is
nothing worth." It is suggested that, in the context of the Rhodesian
rebellion, this maxim does present the issues more honestly than any
attempt to introduce a value element into the definition of law,
- Fieldsend J. resigned in March 1968, Dendy Young J. in August 1968.

When Does Grundnorm Change

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

When Does Grundnorm Change

Uploaded by

Copyright:

Available Formats

DATE DOWNLOADED: Sat Aug 20 06:01:41 2022

SOURCE: Content Downloaded from HeinOnline

Bluebook 21st ed.

ALWD 7th ed.

APA 7th ed.

Chicago 17th ed.

McGill Guide 9th ed.

AGLC 4th ed.

MLA 9th ed.

OSCOLA 4th ed.

WHEN AND WHY DOES THE GRUNDNORM

IN Pakistan, Uganda and Southern Rhodesia, courts have held

I The State v. Dosso (1958) 2 Pakistan S.C.R. 180; Uganda v. Commissioner of

individuals whose behavior the new order regulates actually

according to Kelsen's theory, the grimdnorm changes upon the

S. (2). THE FUNCT[ONS OF THE GRUNDNORM

The Grundnorm's Two Functions

He gives, as an example of the former function, the different inter-

Legal Science and Logic

conceives of its object as a logical unit: a system of non-contradictory norms."

contradictory; and one legal norm may be deduced from another if

Content of the Grundnorm

norm is that it stipulates a coercive act." 24 Rules of law which do

In any national legal system the constitution contains, at least,

Form of the Grundnorm

24 Stanford, p. 1131. It is this internal characteristic which dikstinguishes an

specifically wanted such norms to operate on the constitutional level

Aims of Legal Science

35 Sale of Goods Act 1893, s. 51(1).

S. (3). THE CONCEPT OF VALIDITY

First function of grundnorm rejected

Four Senses of" Validity "

be valid." But Validity/2 is primarily a concept of legal theory.

Kelsen recognises and specifically rejects for his purposes

41 Ross, On Law and Justice, 1958, pp. 25-26.

If a norm is Valid/5, in the sense that it actually occupies a segment

Criticism (i) rejected

S. (4). WHEN DOES THE GRUNDNORM CHANGE?

government, which is not inaccurately described in Austinian langu-

Grundnormchanges when lawyers make new logical assumptions

56 "If a determinate human superior, not in a habit of obedience to a like

Legal science and efficacy

Efficacy, a practical test?

59 G.T.L.S., p. 437; P.T.L., pp. 46-48, 86-87. Stanford, p. 1142.

no objective factual judgment as to efficacy can be made soon after

64 "The function of the basic norm is not 1:omake it possible to consider a

obeyed or applied "by and large? " " It would be impracticable to

clearly not answer the question: which grundnorm ought a legal

Efficacy and the judge

piece of propaganda against a racist regime, but he would not be

Implications of the pure theory for a judge

science is socially useful, which is precisely the reason for insisting

revolution "), in accepting the laws of a successful revolutionary

Apart from the four criticisms mentioned at the beginning of this

Morality and the definition of law

90 Ross, On Law and Justice, 1958, p. 31.

English courts and Rhodesian "law"

91 L. L. Fuller, " Positivism and Fidelity to Law-Reply to Professor Hart"

tional doctrine of indivisible sovereignty and not from any general

scientist when -he interprets these contents according to the logical

The history of natural law theory consists of attempts to overload it

- Fieldsend J. resigned in March 1968, Dendy Young J. in August 1968.

You might also like