The Creation of Jus Cogens – Making Sense of

Article 53 of the Vienna Convention

Ulf Linderfalk*

I. Introduction 359
II. The Creation of Customary International Law 363
III. Why Article 53 of the Vienna Convention Is Not Circular 370
IV. Further Consequences of the Jus Cogens Analysis 374

Abstract

This essay provides an analysis of the creation of jus cogens. The analysis
makes intelligible the definition of jus cogens provided in Art. 53 of the 1969
Vienna Convention on the Law of Treaties. It has been said repeatedly
about this definition that it is circular. If a jus cogens status is conferred on a
rule of law because the international community of states accepts and rec-
ognizes this rule as non-derogable and modifiable only by the creation of a
new norm of jus cogens, then the definition assumes what remains to be es-
tablished: the creation of jus cogens. As shown in this essay, this criticism
builds on wrongful assumptions. It assumes that Art. 53 explains the crea-
tion of jus cogens, which it does not; it explains only its existence. A full ex-
planation of the creation of jus cogens requires further elaboration. Accord-
ing to the proposition argued in this essay, jus cogens obligations derive
from the usual processes creating ordinary customary international law.

I. Introduction

Jus cogens is in vogue. While not so very long ago, peremptory interna-
tional law was regarded by international lawyers as an idea of little more
than academic interest, today it is part of the common rhetoric of the inter-
national legal profession. More than ever before international lawyers resort
to jus cogens for the construction and reinforcement of legal arguments.
Even more importantly, they do so in an exceptionally wide variety of dif-

* Associate Professor of International Law at the Faculty of Law, Lund University, Swe-
den.

ZaöRV 71 (2011), 359-378

http://www.zaoerv.de/
© 2011, Max-Planck-Institut für ausländisches öffentliches Recht und Völkerrecht
360 Linderfalk

ferent legal contexts. As I have argued extensively elsewhere, from the per-
spective of international legal science at least, this raises a need for a better
understanding of the relevant international law.1 This is a need I am intent
to meet. In this essay, I will focus on what appears to be one of the more
crucial issues for the accomplishment of this task, if not the most crucial
issue of them all: the rather peculiar wording chosen for Art. 53 of the 1969
Vienna Convention on the Law of Treaties (VCLT).2
Let me begin by setting this provision into its proper context. As we are
used to thinking, jus cogens is a legal concept closely associated with the ap-
plication of a particular set of norms in customary international law. In my
earlier writing, I have referred to this set of norms as the international jus
cogens regime.3 This is to stress the complexity of the jus cogens concept,
which obviously presupposes the existence of norms with rather distinct
functions. First of all, jus cogens presupposes the existence of norms com-
manding or prohibiting certain actions, such as for instance aggressive war-
fare:

If a state engages in aggressive warfare, then this shall be considered a

breach of an international obligation owed by that state to the international
community as a whole.4

In the international legal literature, these norms are often (rather bluntly)
referred to as peremptory norms of international law, or just jus cogens. To
add clarification, they will here be referred to as first order rules of jus co-
gens.5 Secondly, the concept of jus cogens presupposes the existence of
norms specifying the legal consequences that ensue from the postulated su-
periority of jus cogens over ordinary international law, as illustrated by the
following examples:

- If a treaty is in conflict with a first order rule of jus cogens created prior
to the conclusion of the treaty, then the treaty shall be considered void.6

1 See U. Linderfalk, Normative Conflict and the Fuzziness of the International ius cogens
Regime, ZaöRV 69 (2009), 961 et seq.
2 UNTS, Vol. 1155, 331.
3 See U. Linderfalk (note 1).
4 See e.g. the Commentaries to the Draft Articles on Responsibility of States for Interna-
tionally Wrongful Acts, adopted by the International Law Commission in 2001, ILC, Report
on the Work of its 53rd Session, Session 20, at 85 (Draft Art. 26).
5 See U. Linderfalk (note 1).
6 Compare Arts. 53 and 44 of the VCLT.

ZaöRV 71 (2011)

http://www.zaoerv.de/
© 2011, Max-Planck-Institut für ausländisches öffentliches Recht und Völkerrecht
 The Creation of Jus Cogens – Making Sense of Art. 53 of the Vienna Convention 361

- If the purport of a reservation to a treaty is in conflict with a first order

rule of jus cogens, then that reservation shall be considered void.7
- If a resolution adopted by an international organization is in conflict
with a first order rule of jus cogens, then that resolution shall be considered
void.8
- To the extent that a claim of sovereignty over a certain territory is based
on action violating a first order rule of jus cogens, that claim shall be consid-
ered invalid.9

To facilitate reference, tentatively I will refer to these norms as second or-

der rules of jus cogens.10
As can be seen from the way first and second order rules of jus cogens are
described, if we expect the international jus cogens regime to be effective,
then those rules cannot themselves be sufficient. Any application of the sec-
ond order rules of jus cogens presupposes that first order rules of jus cogens
(like the prohibition of aggressive warfare) can be identified. In other
words, first and second order rules of jus cogens need to be accompanied by
a norm that helps define the jus cogens concept for the overall purpose of
general international law. This is where Art. 53 of the 1969 VCLT enters the
picture. It reads:
For the purposes of the present Convention, a peremptory norm of general in-
ternational law is a norm accepted and recognized by the international commu-
nity of States as a whole as a norm from which no derogation is permitted and
which can be modified only by a subsequent norm of general international law
having the same character.

7 Compare the Dissenting Opinion expressed by Judge Tanaka, International Court of

Justice, North Sea Continental Shelf (Federal Republic of Germany v. Denmark; Federal Re-
public of Germany v. Netherlands), Judgment of 20.2.1969, ICJ Reports 1969, 3, at 182. In the
same vein, Alain Pellet, Special Rapporteur, Tenth Report on Reservations to Treaties, submit-
ted to the International Law Commission at the occasion of its 57th session (UN Doc.
A/CN.4/588/Add.1), 34-35, §§ 135-137. For a further discussion of this rule, see U. Linder-
falk, Reservations to Treaties and Norms of Jus Cogens – A Comment on Human Rights
Committee General Comment No. 24, in: I. Ziemele (ed.), Reservations to Human Rights
Treaties and the Vienna Convention Regime, 2004, 213 et seq.
8 See e.g. Separate Opinion of Judge Lauterpacht, International Court of Justice, Applica-
tion of the Convention on the Prevention and Punishment of the Crime of Genocide (Bosnia
and Herzegovina v. Yugoslavia (Serbia and Montenegro)), Further Requests for the Indica-
tion of Provisional Measures, Order of 13.9.1993, ICJ Reports 1993, 325, at 440.
9 See e.g. A. Orakhelashvili, Peremptory Norms in International Law, 2006, 218 et seq.
and the further references cited there.
10 See U. Linderfalk (note 1). With regard to this designation, see, however, the remark
below, section 4.

ZaöRV 71 (2011)

http://www.zaoerv.de/
© 2011, Max-Planck-Institut für ausländisches öffentliches Recht und Völkerrecht
362 Linderfalk

Some international legal scholars argue that Art. 53 cannot possibly be

the general rule required by the international jus cogens regime, since, ac-
cording to the wording of the VCLT, this provision applies only for the lim-
ited purposes of the Convention.11 Stated more specifically, the provision is
applied only to resolve conflicts between treaties and first order rules of jus
cogens,12 and then only in the relations between the parties to the VCLT and
with respect to treaties concluded by those parties after they entered the
Convention.13 This objection, however, builds on an overly narrow reading
of the Vienna Convention. It ignores the role played by Art. 53 in modern-
day international legal discourse. In judicial decision-making, as well as in
scholarly debate, the contents of Art. 53 are applied to situations with little
regard to whether they come within the personal, material or temporal
scope of application of the VCLT or not.14 Obviously, the great majority of
international lawyers look upon Art. 53 as a reflection of customary inter-
national law. They regard Art. 53 as the expression of a general definition
that reaches beyond the rather limited scope of application of VCLT, Arts.
53 and 64. The present essay will start off from this same assumption.
All since the beginning of the 1960s, when the provision finally adopted
as Art. 53 began to take shape, it has been criticized for its rather peculiar
wording. According to critics, the definition provided in Art. 53 is circu-
lar.15 Naturally, when states accept and recognize a rule (R) as a norm, from
which no derogation is permitted, and which can be modified only by the
creation of a new rule of jus cogens, this cannot be the constitutive act by
which a jus cogens status is conferred on R. That would commit us to the
conclusion that states wrongly assumed R to be a norm, from which no de-
rogation is permitted, etc. It has to be taken for granted that if states accept
and recognize R as jus cogens, they do so for a good reason. They do so be-
cause according to their judgment, international law does not permit dero-
gations from R, and international law accepts modifications only when ac-
complished by a new norm of jus cogens. Hence, by the way Art. 53 is
stated, the definition assumes what remains to be established: the creation of
jus cogens.

11See e.g. D. Dubois, The Authority of Peremptory Norms in International Law: State
Consent or Natural Law?, Nord. J. Int’l L. 78 (2009), 133, at 144.
12 See Arts. 53 and 64 of the VCLT.
13 Compare Art. 4 of the VCLT.
14 See U. Linderfalk (note 1), at 962, notes 3-7.
15 See e.g. D. Dubois (note 11), 155; G. Christenson, Jus Cogens: Guarding Interests Fun-
damental to International Society, Va. J. Int’l L. 28 (1987-1988), 585, at 594; C. L. Rozakis,
The Concept of Jus Cogens in the Law of Treaties, 1976, 45.

ZaöRV 71 (2011)

http://www.zaoerv.de/
© 2011, Max-Planck-Institut für ausländisches öffentliches Recht und Völkerrecht
 The Creation of Jus Cogens – Making Sense of Art. 53 of the Vienna Convention 363

This criticism inspires me. Indeed, it seems a rather fair requirement that
if international legal science wishes to uphold the idea of Art. 53 as a reflec-
tion of a valid rule of law, one way or another it has to come to grips with
the argument just reiterated. It has to provide an explanation of the creation
of jus cogens that renders Art. 53 intelligible. This – to put it in a short sen-
tence – is exactly what I now aim to do. As the investigation will show, if
international legal scholars consider Art. 53 deficient, it is not because of
what is actually stated in this provision. It is because of the wrongful as-
sumption that Art. 53 explains the creation of jus cogens. In fact, Art. 53 of-
fers no such explanation. The real explanation of the creation of jus cogens
has to be looked for elsewhere. According to the proposition that I wish to
establish, jus cogens is created by the same processes that create customary
international law.
I will argue this proposition in three steps. In section 2 of the essay, I will
briefly delineate the processes that create customary international law. For
the accomplishment of this task, I will draw on the analysis provided by the
British legal philosopher John Finnis. In section 3, I will show how Finnis’
analysis can be used for the resolution of the particular problem at hand.
Stated more specifically, I will provide an analysis of the creation of jus co-
gens. As section 3 will establish, contrary to what many international legal
scholars have argued, there is nothing circular about the definition laid
down in Art. 53 of the VCLT. In section 4, I will continue exploring the
consequences ensuing from my jus cogens analysis. As the investigation will
show, the analysis has many virtues. Not only does it take care of the circu-
larity that seems to inhere in the wording of Art. 53. It can also be used for
explaining a host of other related issues, such as for instance the meaning of
the expression “accepted and recognized by the international community of
States as a whole”, the modification of jus cogens rules, and the status of the
second order rules of jus cogens.

II. The Creation of Customary International Law

In the rich literature on the sources of international law, references to

Art. 38, para. 1 of the Statute of the International Court of Justice (ICJ) are
a commonplace.16 Although, by its wording, Art. 38, para. 1 remains a list
of the sources to be applied by the Court, it is the general assumption that
these same sources are applicable also outside ICJ proceedings, in the rela-

16 UNTS, Vol. 1, xvi.

ZaöRV 71 (2011)

http://www.zaoerv.de/
© 2011, Max-Planck-Institut für ausländisches öffentliches Recht und Völkerrecht
364 Linderfalk

tions between international legal subjects in general. The sources listed in

Art. 38, para. 1(b) include “international custom, as evidence of a general
practice accepted as law”. International lawyers infer from this wording a
general definition of the concept of customary international law. According
to this definition, in order for a rule of customary international law to exist,
two conditions have to be met: (i) there has to be a certain pattern of con-
duct, in the sense of a general, constant, and uniform usage; (ii) there has to
be a widespread belief among states (often referred to as the opinio juris or
the opinio juris generalis) that this same pattern of conduct is prescribed,
permitted, or prohibited (whatever might be the case) by customary inter-
national law.17
This definition forms a usual starting-point for scholarly analysis and
discussion of the question how customary international law is created. Since
obviously, the existence of a rule (R) presupposes the creation of that same
rule R, commentators assume that the criteria applied to establish the exis-
tence of customary international law must also be the criteria that explain its
creation. Hence, statements like that made by Rosalyn Higgins follow quite
naturally: “For the formation of customary international law, practice and
opinio juris is required.”18 Analytically, however, such descriptions of the
customary law-creating process remain problematic. If someone suggests
that a rule of customary international law is created by a conviction on the
part of states that some certain action is required by or consistent with the
existing customary international law, then this statement can be assessed in
two different ways. Either the statement is circular – it assumes what re-
mains to be established. Or, the statement assumes that a rule of customary
international law can come into existence by virtue of the erroneous belief
that it already exists. Both alternatives are equally unattractive.
The British legal philosopher John Finnis has written on this problem,
and as I have already stated, I will base my essay on his findings.19 It should
be realized that Finnis’ enquiry into the process creating customary interna-
tional law has a very particular purpose. In want of arguments for a secular
version of natural law theory, Finnis’ aim is to establish that legally authori-
tative rules can emerge although there is no one with authority to make
them. Separated from this context, however, Finnis’ analysis can be viewed
as theory-neutral. On further scrutiny, it does not commit us either to the

17 See e.g. I. Brownlie, Principles of International Law, 6th ed. 2003, 7et seq.
18 R. Higgins, Problems and Process: International Law and How We Use It, 1994, 19.
For other similar statements, see e.g. H. Thirlway, The Sources of International Law, in: M.
Evans (ed.), International Law, 2nd ed. 2006, 115, at 122; V. Lowe, International Law, 2007, 38.
19 J. Finnis, Natural Law and Natural Rights, 1980, 231 et seq.

ZaöRV 71 (2011)

http://www.zaoerv.de/
© 2011, Max-Planck-Institut für ausländisches öffentliches Recht und Völkerrecht
 The Creation of Jus Cogens – Making Sense of Art. 53 of the Vienna Convention 365

natural law position adopted by Finnis himself, or to the position of legal

positivism. This is why I find it useful.
To solve the paradox inherent in the opinio juris criterion, Finnis intro-
duces the distinction between practical, empirical, and juridical judgments.20
His important point is that a statement like “R is an authoritative legal rule”
can be analyzed in three different ways. First, it can be analyzed as a practi-
cal judgment. In this case the statement is viewed as asserted by an utterer,
who treats the rule R as authoritative not merely for others but also for
himself. According to the utterer, the rule R gives anyone, including the ut-
terer himself, an exclusionary reason for action in accordance with that
rule.21 Secondly, “R is an authoritative legal rule” can be analyzed as an em-
pirical judgment. In this case the statement is viewed as asserted by a person
who does not necessarily consider the rule R to have authority in relation to
himself. What the statement aims to describe is the fact that some other per-
son or persons consider the rule R authoritative. The utterer in this case
speaks in the capacity of an observer. Finally, “R is an authoritative legal
rule” can be analyzed as a juridical judgment. In this case the statement is
viewed as asserted by a person “from the legal point of view”.22 The state-
ment is made neither in recognition of the rule R’s authoritativeness in rela-
tion to the utterer, nor by way of report about other people’s attitude to the
rule R, but rather by way of stating what is the case from the viewpoint of
law, without either endorsing or rejecting that view.
Finnis’ observation is that these distinctions bear on the analysis of the
processes that create customary international law.23 At the root of the for-
mation of custom, Finnis explains,24 is a practical judgment:

PJ0: “In this domain of human affairs, it would be appropriate to have

some determinate, common, and stable pattern of conduct and, correspond-
ingly, an authoritative rule requiring that pattern of conduct; to have this is
more desirable than leaving conduct in this domain to the discretion of in-
dividual states; the particular pattern of conduct P is appropriate, or would

20 In using this distinction, Finnis draws on Raz. See J. Raz, Practical Reason and Norms,
1975, 170 et seq.
21 By an exclusionary reason, Finnis understands “a reason for judging or acting in the ab-
sence of understood reasons, or for disregarding at least some reasons which are understood
and relevant and would in the absence of the exclusionary reason have sufficed to justify pro-
ceeding in some other way”. J. Finnis (note 19), 34. The terminology derives from Raz. See J.
Raz (note 20), 35 et seq., 58 et seq.
22 This expression derives from Raz. See J. Raz (note 20), 177.
23 J. Finnis (note 19), 238 et seq.
24 J. Finnis (note 19), 239 et seq.

ZaöRV 71 (2011)

http://www.zaoerv.de/
© 2011, Max-Planck-Institut für ausländisches öffentliches Recht und Völkerrecht
366 Linderfalk

be if generally adopted and acquiesced in, for adoption as an authoritative

common rule of conduct.”

This practical judgment PJ0, however, is quite distinct from the empirical
judgment (EJ2) that many states in fact subscribe to it:

EJ2: “The opinion PJ0 is widely subscribed to by states.”

Also, the practical judgment PJ0 must be distinguished from the empirical
judgment (EJ1) that the practice of states is convergent with the pattern re-
ferred to in PJ0:

EJ1: “There is widespread concurrence and acquiescence in this pattern of

conduct P by states.”

The empirical judgments EJ1 and EJ2 in turn are pertinent for the making
of a new practical judgment. This judgment (PJ1) expresses the utterer’s un-
derstanding of the empirical judgments EJ1 and EJ2 relative to the prevailing
doctrine of sources:

PJ1: “The widespread subscription to PJ0, and the widespread concur-

rence and acquiescence in the pattern of conduct P, are sufficient to warrant
the judgment (PJ2) that there is now an authoritative rule of customary in-
ternational law (R) requiring or permitting P (as the case may be).”

The practical judgment PJ2, obviously, corresponds to the opinio juris as

traditionally defined:

PJ2: “There is now an authoritative rule of customary international law

(R) requiring or permitting P (as the case may be).”

This is not the end of the matter, however. As Finnis’ analysis goes, the
practical judgments PJ1 and PJ2 have to be distinguished from yet another
empirical judgment. According to what this judgment (EJ3) asserts, the rule
R is effective, in the sense that, as a matter of fact, states generally recognize
R as an authoritative rule of customary international law:

EJ3: “States generally accept the rule (R) that P is to be done, or may be
done (as the case may be).”

ZaöRV 71 (2011)

http://www.zaoerv.de/
© 2011, Max-Planck-Institut für ausländisches öffentliches Recht und Völkerrecht
 The Creation of Jus Cogens – Making Sense of Art. 53 of the Vienna Convention 367

Finally, to make the analysis complete, the practical judgments PJ1 and
PJ2 have to be distinguished from the juridical judgment (JJ1) made by any-
one speaking “from the point of view of law”:

JJ1: “According to international law, P is required or permitted (as the

case may be).”

This analysis has many virtues. Clearly, the distinctions suggested pro-
vide a basis for a more subtle analysis of the creation of customary interna-
tional law than those usually attempted in the international legal literature.
Finnis himself, however, rests content with the observation that there is a
conceptual difference between the opinio juris (i.e. the practical judgment
PJ2) and the other beliefs that create customary international law (i.e. the
practical judgments PJ0 and PJ1); while the former is clearly dependent upon
the latter, analytically, it is quite distinct.25 He does not bring out explicitly
the full consequences ensuing from his analysis for the understanding of
Art. 38, para. 1(b) of the ICJ Statute and the relevance of this provision for
an explanation of the creation of customary international law. I will now try
to fill this void.
To make my account concrete, I will use Finnis’ analysis relative to a spe-
cific rule of customary international law. I will assume that according to
customary international law, a receiving state is prohibited from instituting
criminal proceedings against the diplomatic agents of a sending state.26 For
the purpose of easy reference, henceforth, I will call this the rule of diplo-
matic immunity. Applied to this rule, the important distinctions entailed by
Finnis’ analysis can be specified as follows:

PJ0: “For every relation between a receiving and a sending state, it would
be desirable and appropriate if a rule of customary international law prohib-
ited the former from instituting criminal proceedings against the diplomatic
agents of the latter.”
EJ1: “Receiving states generally do not institute criminal proceedings
against the diplomatic agents of a sending state.”
EJ2: “States widely subscribe to the opinion that for every relation be-
tween a receiving and a sending state, it would be desirable and appropriate
if a rule of customary international law prohibited the former from institut-
ing criminal proceedings against the diplomatic agents of the latter.”

25J. Finnis (note 19), 231 et seq.

26Compare Art. 31, para. 1 of the 1961 Vienna Convention on Diplomatic Relations.
(UNTS, Vol. 500, 95.)

ZaöRV 71 (2011)

http://www.zaoerv.de/
© 2011, Max-Planck-Institut für ausländisches öffentliches Recht und Völkerrecht
368 Linderfalk

PJ1: “The widespread subscription to the practical judgment PJ0, and the
widespread concurrence and acquiescence in the pattern of conduct de-
scribed by the empirical judgment EJ1, are sufficient to warrant the conclu-
sion that there is now an authoritative rule of customary international law,
according to which, in every relation between a receiving and a sending
state, the former is prohibited from instituting criminal proceedings against
the diplomatic agents of the latter.”
PJ2: “There is now an authoritative rule of customary international law,
according to which, in every relation between a receiving and a sending
state, the former is prohibited from instituting criminal proceedings against
the diplomatic agents of the latter.”
EJ3: “States generally accept the rule of customary international law, ac-
cording to which, in every relation between a receiving and a sending state,
the former is prohibited from instituting criminal proceedings against the
diplomatic agents of the latter.”
JJ1: “According to customary international law, in every relation between
a receiving and a sending state, the former is prohibited from instituting
criminal proceedings against the diplomatic agents of the latter.”

This scheme helps to explain why the criteria applied to establish the ex-
istence of a rule of customary international law, as reflected in Art. 38, para.
1(b) of the ICJ Statute, do not easily serve as an explanation of the creation
of that same rule. Let us assume, for instance, that we wish to establish the
existence of the rule of diplomatic immunity. Obviously, this amounts to
showing that the juridical judgment JJ1 is supported by good reasons. Fol-
lowing the definition enshrined in Art. 38, para. 1(b), we need to show (i)
that receiving states generally do not institute criminal proceedings against
the diplomatic agents of a sending state, and (ii) that there is a widespread
belief among states that according to customary international law, in every
relation between a receiving and a sending state, the former is prohibited
from instituting criminal proceedings against the diplomatic agents of the
latter. Obviously, this is tantamount to showing that the empirical judgment
EJ3 is correct. Stated somewhat differently, Art. 38, para. 1(b) makes the
state of affairs described by the empirical judgment EJ3 a sufficient reason
for the juridical judgment JJ1. The state of affairs described by the empirical
judgment EJ3 entails the existence of the rule of diplomatic immunity as a
matter of fact.
Now, this same state of affairs does not itself explain how the rule of dip-
lomatic immunity was created. As Finnis’ analysis makes fairly plain, how-
ever, it is a basis on which such an explanation can be founded. The empiri-

ZaöRV 71 (2011)

http://www.zaoerv.de/
© 2011, Max-Planck-Institut für ausländisches öffentliches Recht und Völkerrecht
 The Creation of Jus Cogens – Making Sense of Art. 53 of the Vienna Convention 369

cal judgment EJ3 entails the empirical judgment EJ1 and the widespread sub-
scription to the practical judgment PJ2. Stated in clear language, the correct-
ness of the empirical judgment EJ3 requires that it can be shown (i) that re-
ceiving states generally do not institute criminal proceedings against the
diplomatic agents of a sending state, and (ii) that there is a widespread belief
among states that according to customary international law, in every rela-
tion between a receiving and a sending state, the former is prohibited from
instituting criminal proceedings against the diplomatic agents of the latter. If
the practical judgment PJ2 is not to lead to the fallacy indicated in the be-
ginning of this section, it too requires further explanation, however. Hence,
in an explanation of the creation of customary international law, obviously,
the practical judgment PJ2 presupposes the empirical judgments EJ1 and EJ2
and the widespread subscription to the practical judgment PJ1. For the es-
tablishment of the existence of the rule of diplomatic immunity, given what
is provided in Art. 38, para. 1(b) of the ICJ Statute, this dependency of the
practical judgment PJ2 on the empirical judgments EJ1 and EJ2 and the wide-
spread subscription to the practical judgment PJ1 is of no relevance. For the
explanation of the creation of this same rule, however, it is. Consequently, as
revealed by Finnis’ analysis, the real explanation of the creation of the rule
of diplomatic immunity lies in a combination of the states of affairs de-
scribed in the empirical judgments EJ1, EJ2, and EJ3, together with the wide-
spread subscription to the practical judgments PJ1 and PJ2. Contrary to
what is assumed by international legal scholars, it does not lie solely in state
practice and opinio juris, i.e. the factual state of affairs described in the em-
pirical judgment EJ1 and the widespread subscription to the practical judg-
ment PJ2.
Secondly, and perhaps most importantly, Finnis’ analysis renders Art. 38,
para. 1(b) of the ICJ Statute intelligible. According to Art. 38, para. 1(b), if
we wish to establish the existence of the rule of diplomatic immunity, we
are required to show the existence of a certain belief: an opinio juris. Stated
more specifically, we are required to show that among states there is a wide-
spread belief that according to customary international law, in every relation
between a receiving and a sending state, the former is prohibited from insti-
tuting criminal proceedings against the diplomatic agents of the latter. There
is nothing paradoxical or circular about this requirement. In this respect
Art. 38, para. 1(b) makes perfectly sense. The opinio juris criterion becomes
paradoxical only when it is applied to explain the creation of a customary
international legal norm, such as the rule of diplomatic immunity, and when
in this explanation it is treated as self-sufficient.

ZaöRV 71 (2011)

http://www.zaoerv.de/
© 2011, Max-Planck-Institut für ausländisches öffentliches Recht und Völkerrecht
370 Linderfalk

Certainly, beliefs are important for the creation of customary interna-

tional law, but those beliefs include more than just the opinio juris as tradi-
tionally defined. As stated earlier, the creation of the rule of diplomatic im-
munity requires not only the belief that according to customary interna-
tional law, in every relation between a receiving and a sending state, the
former is prohibited from instituting criminal proceedings against the dip-
lomatic agents of the latter. Furthermore, it requires, first, the belief that for
every relation between a receiving and a sending state, it would be desirable
and appropriate if a rule of customary international law prohibited the for-
mer from instituting criminal proceedings against the diplomatic agents of
the latter. Secondly, the creation of the rule of diplomatic immunity requires
the belief that the widespread subscription to the practical judgment PJ0,
and the widespread concurrence and acquiescence in the pattern of conduct
described by the empirical judgment EJ1, are sufficient to warrant the prac-
tical judgment PJ2. The conclusion that remains to be drawn from this
analysis appears rather obvious. The problem with the opinio juris criterion
does not occur because of Art. 38, para. 1(b) of the ICJ Statute, which
clearly does not pretend to explain the creation of customary international
law. The problem, rather, occurs because of the wrongful assumption made
by international legal scholars that Art. 38, para. 1(b) can be used for this
purpose exactly.

III. Why Art. 53 of the Vienna Convention Is Not

Circular

Now, let us return to where we started: to the definition laid down in

Art. 53 of the 1969 Vienna Convention on the Law of Treaties. According
to this definition – it might be worth repeating – “a peremptory norm of
general international law is a norm accepted and recognized by the interna-
tional community of States as a whole as a norm from which no derogation
is permitted and which can be modified only by a subsequent norm of gen-
eral international law having the same character”. There is a striking similar-
ity between this provision and Art. 38, para. 1(b) of the ICJ Statute. Art. 53,
like Art. 38, para. 1(b), is interested, not in the law-creating process as such,
but in the existence of law as a matter of fact. Stated more specifically, for
Art. 53 the only relevant question is whether a rule of international law is
jus cogens or not. This notwithstanding, Art. 53, like Art. 38, para. 1(b), is
used by international legal scholars to explain the law-creating process ex-
actly. While Art. 38, para. 1(b) is used to explain the creation of customary

ZaöRV 71 (2011)

http://www.zaoerv.de/
© 2011, Max-Planck-Institut für ausländisches öffentliches Recht und Völkerrecht
 The Creation of Jus Cogens – Making Sense of Art. 53 of the Vienna Convention 371

international law, Art. 53 is used to explain the creation of a rule of law as

jus cogens. (It should be noted that by “the creation of a rule of law as jus
cogens” I mean, not the creation of a rule of law as law, but rather the eleva-
tion of a rule of law to a jus cogens status.) This raises the assumption that
the circularity alleged to inhere in the wordings of Art. 53 can be unraveled
in much the same way as the paradox traditionally associated with the
opinio juris criterion. This is the assumption now to be investigated.
To make the analysis of Art. 53 concrete, let us assume the existence of
some certain first order rule of jus cogens. Let us assume the existence of a
rule (R) prohibiting states from engaging in aggressive warfare:

R: “If a state engages in aggressive warfare, then this shall be considered a

breach of an international obligation owed by that state to the international
community as a whole.”

According to Art. 53, in order to establish the existence of the rule R as

jus cogens, we have to show that generally, states accept and recognize that
derogations from R are not permitted, and that modifications of R cannot
be made by means of ordinary international law. In line with what Art. 53
provides, the existence of R as jus cogens can be described by the following
empirical judgment:

EJ3: “The international community of states accepts and recognizes that

no derogations from the rule R are permitted, and that modifications of R
cannot be made by means of ordinary international law.”

Stated from the perspective of Art. 53, we may say that Art. 53 qualifies
the state of affairs described by the empirical judgment EJ3. It makes it a
sufficient reason for the following juridical judgment:

JJ1: “According to international law, the rule R is jus cogens.”

The empirical judgment EJ3 needs interpretation. As already stated, if the

international community of states accepts and recognizes that no deroga-
tions from the rule R are permitted, and that modifications of R cannot be
made by means of ordinary international law, then it has to be taken for
granted that they do so for good reason. They do so because according to
their judgment, international law does not permit derogations from R, and
it accepts modification only when accomplished by a new norm of jus co-
gens. Any other interpretation of Art. 53 would commit us to the assump-

ZaöRV 71 (2011)

http://www.zaoerv.de/
© 2011, Max-Planck-Institut für ausländisches öffentliches Recht und Völkerrecht
372 Linderfalk

tion that states wrongly assumed R to be a norm, from which no deroga-

tions are permitted, etc. If international law does not permit derogations
from R, and if it does not accept modification of R except by a new norm of
jus cogens, then clearly this is what ensues from the application of the set of
second order rules of jus cogens existing in international law. Hence, it
would seem the empirical judgment EJ3 can be restated as follows:

EJ3: “The international community of states accepts and recognizes as le-

gally authoritative the set of rules (in this essay referred to as second order
rules of jus cogens) permitting no derogations from the rule R and prohibit-
ing all modification of R by means of ordinary international law.”

Obviously, the existence of the rule R as jus cogens turns on whether the
empirical judgment EJ3 can be justified or not. According to what seems to
be the prevailing assumption among international lawyers, second order
rules of jus cogens are customary international law.27 If we accept this as-
sumption, justifying the empirical judgment EJ3 will be tantamount to sho-
wing the following two propositions to be true: (i) “States generally do not
derogate from the rule R, and they generally do not modify R by means of
ordinary international law”; (ii) “States widely subscribe to the opinion that
by virtue of an authoritative set of rules existing in customary international
law, no derogations from the rule R are permitted, and all modification of
the rule R by means of ordinary international law is prohibited”. In this re-
spect, Art. 53 of the Vienna Convention compares favorably with Art. 38,
para. 1(b) of the ICJ Statute. The criteria governing the existence of jus co-

27 Not only is this implicit in all writing on the effects of jus cogens beyond the personal,
material, and temporal scope of application of the VCLT. It is also implicit in the work of the
International Law Commission. See e.g. A. Cassese, International Law, 2nd ed. 2005, 205 et
seq.; L. Hannikainen, Peremptory Norms (Jus Cogens) in International Law, 1988; A.
Orakhelashvili (note 9); D. Shelton, Normative Hierarchy in International Law, AJIL 100
(2006), 291 et seq., and the further references cited there; C. Focarelli, Promotional Jus Co-
gens: A Critical Appraisal of Jus Cogens’ Legal Effects, Nord. J. Int’l L. 77 (2008), 429 et seq.,
and the further references cited there. For the work of the ILC, see particularly the Commen-
taries to the Draft Articles on Responsibility of States for Internationally Wrongful Acts,
adopted by the International Law Commission in 2001, Report of the International Law
Commission on the work of its 53rd session, 23.4.-1.6. and 2.7.-10.8.2001, UN Doc. A/56/10,
20, at 84et seq. (Draft Art. 26) and 132 et seq. (Draft Art. 50 § 1(d); Fragmentation of Interna-
tional Law: Difficulties Arising From the Diversification and Expansion of International Law,
Report of the Study Group of the International Law Commission, Chaired by Martti
Koskenniemi (UN Doc. A/CN.4/L.682), 184 et seq., Report of the International Law Com-
mission on the work of its 58th session, 1.5.-9.6. and 3.7.-11.8.2006. For substantial arguments
supporting this same position, see U. Linderfalk, The Source of Jus Cogens: How Legal Posi-
tivism Copes with Peremptory International Law, forthcoming.

ZaöRV 71 (2011)

http://www.zaoerv.de/
© 2011, Max-Planck-Institut für ausländisches öffentliches Recht und Völkerrecht
 The Creation of Jus Cogens – Making Sense of Art. 53 of the Vienna Convention 373

gens are very much the same as those governing the existence of customary
international law. First, there has to be a certain pattern of action: states
generally do not derogate from the rule R, and they generally do not mod-
ify R by means of ordinary international law. Secondly, there has to be a
certain belief: states widely subscribe to the opinion that by virtue of an au-
thoritative set of rules existing in customary international law, no deroga-
tions from the rule R are permitted, and all modification of the rule R by
means of ordinary international law is prohibited.
For the very same reason as Art. 38, para. 1(b) of the ICJ Statute does not
explain the creation of customary international law, Art. 53 of the Vienna
Convention does not itself explain the creation of jus cogens. It provides a
basis on which such an explanation can be founded, which is a different
thing. As already stated, the empirical judgment EJ3 entails the following
empirical judgment:

EJ1: “States generally do not derogate from the rule R, and they generally
do not modify R by means of ordinary international law.”

It also entails the widespread subscription to the following practical judg-

ment:

PJ2: “There is now in customary international law an authoritative set of

rules (in this essay referred to as second order rules of jus cogens), by virtue
of which no derogations from the rule R are permitted, and all modification
of the rule R by means of ordinary international law is prohibited.”

The practical judgment PJ2 presupposes in turn the following empirical

judgments:

EJ1: “States generally do not derogate from the rule R, and they generally
do not modify R by means of ordinary international law.”
EJ2: “The following opinion (PJ0) is widely subscribed to by states: it
would be desirable and appropriate if customary international law permit-
ted no derogations from the rule R and prohibited all modification of that
same rule by means of ordinary international law.”

In addition, the practical judgment PJ2 presupposes the widespread sub-

scription to the following practical judgment:

ZaöRV 71 (2011)

http://www.zaoerv.de/
© 2011, Max-Planck-Institut für ausländisches öffentliches Recht und Völkerrecht
374 Linderfalk

PJ1: “The widespread subscription to the opinion PJ0, together with the
fact that states generally do not derogate from R, and do not modify R by
ordinary international law, are sufficient to warrant the judgment that there
is now in customary international law an authoritative set of rules permit-
ting no derogations from the rule R and prohibiting all modification of R
by means of ordinary international law.”

As revealed by the analysis, the real explanation of the creation of R as jus

cogens lies in a combination of the states of affairs described in the empirical
judgments EJ1, EJ2, and EJ3, together with the widespread subscription to
the practical judgments PJ1 and PJ2. Contrary to what is assumed by inter-
national legal scholars, it does not lie in the factual state of affairs described
in the empirical judgment EJ3.
This analysis renders Art. 53 of the Vienna Convention intelligible. Ac-
cording to Art. 53, if we wish to establish the existence of a rule of law as jus
cogens, we are required to show the existence of a certain belief. Stated more
specifically, we are required to show that states widely subscribe to the
opinion that by virtue of an authoritative set of rules existing in customary
international law, no derogations from the rule R are permitted, and all
modification of the rule R by means of ordinary international law is prohib-
ited. There is nothing paradoxical or circular about this requirement. The
requirement becomes paradoxical only when it is offered as the sole expla-
nation of the creation of R as jus cogens. As shown in this section, however,
the real explanation of the creation of R as jus cogens requires further elabo-
ration. The conclusion ensuing from this analysis is almost a reproduction
of the conclusion that ended section 2. If international legal scholars regard
as deficient the definition of jus cogens laid down in Art. 53 of the Vienna
Convention, it is not because of what is actually stated in this provision. It
is because of the wrongful assumption that Art. 53 can be used to explain
something it does not pretend to explain: the creation of jus cogens.

IV. Further Consequences of the Jus Cogens Analysis

In order to appreciate the full value of the analysis just accomplished, we

have to realize just how productive it is. As shown in section 3 of this essay,
my jus cogens analysis helps us come to grips with the criticism claiming
that Art. 53 of the VCLT is circular. This is not its sole virtue. Above and
beyond this, as I will now argue, the analysis provides a tool for the resolu-
tion of a host of other, equally perplexing questions.

ZaöRV 71 (2011)

http://www.zaoerv.de/
© 2011, Max-Planck-Institut für ausländisches öffentliches Recht und Völkerrecht
 The Creation of Jus Cogens – Making Sense of Art. 53 of the Vienna Convention 375

First, my jus cogens analysis helps us understand the meaning of the ex-
pression “accepted and recognized by the international community of States
as a whole”. According to Art. 53 of the VCLT, in order for a jus cogens rule
to exist, it has to be shown that it is “a norm accepted and recognized by the
international community of States as a whole as a norm from which no
derogation is permitted and which can be modified only by a subsequent
norm of general international law having the same character”. Some interna-
tional legal scholars have remarked that this in effect amounts to a require-
ment for a special form of opinio juris.28 This statement needs modification.
As observed in the earlier section 3, the existence of a rule (R) as jus cogens
depends on whether for good reasons the following empirical judgment
(EJ3) can be made:

EJ3: “The international community of states accepts and recognizes as le-

gally authoritative the set of rules (in this essay referred to as second order
rules of jus cogens) permitting no derogations from the rule R and prohibit-
ing all modification of R by means of ordinary international law.”

In other words, the existence of R as jus cogens turns on the existence and
scope of application of the second order rules of jus cogens. Those rules are
customary international law.29 If we accept this analysis, then the expression
“accepted and recognized by the international community of States as a
whole” will entail no further requirements beyond those applied for the es-
tablishment of customary international law in general. Certainly, to estab-
lish the existence of the rule R as jus cogens, as commentators have argued,
there has to be a certain belief, an opinio juris. According to the analysis ac-
complished in section 3, we have to show that states widely subscribe to the
following practical judgment (PJ2):

PJ2: “There is now in customary international law an authoritative set of

rules (in this essay referred to as second order rules of jus cogens), by virtue
of which no derogations from the rule R are permitted and all modification
of R by means of ordinary international law is prohibited.”

Contrary to the claim, however, there is nothing special about this belief.
It is like any opinio juris establishing any rule of customary international
law. We cannot infer from the expression “accepted and recognized by the
international community of States as a whole” that the opinio juris establish-

28 See e.g. M. Ragazzi, The Concept of International Obligations Erga Omnes, 1997, 53.
29 This was the assumption adopted in the earlier section 3.

ZaöRV 71 (2011)

http://www.zaoerv.de/
© 2011, Max-Planck-Institut für ausländisches öffentliches Recht und Völkerrecht
376 Linderfalk

ing the existence a rule of law as jus cogens is either more express or more
widespread than the opinio juris establishing customary international law in
general.
Secondly, my jus cogens analysis helps explain the status of the second
order rules of jus cogens. It is a disturbing fact that although the 1969 Vi-
enna Convention presupposes the existence of a set of rules tailored to gov-
ern such things as possible conflicts between first order rules of jus cogens
and ordinary international law, the Convention is silent as to the particular
status of this set of rules. Some commentators have suggested that, clearly,
they must be jus cogens, too. It would be pointless “if a norm was endowed
with peremptory status, but its effects and legal consequences were gov-
erned by the criteria of ordinary rules” – this is the argument typically
used.30 It does not convince. According to the conclusions of section 3 – I
repeat – the classification of a rule R as jus cogens turns on the existence and
scope of application of the second order rules of jus cogens. The claim that
R is jus cogens is correct if it can be shown that R comes within the scope of
application of the set of second order rules of jus cogens existing in interna-
tional law. If this analysis is accepted, the second order rules of jus cogens
cannot possibly be classified as jus cogens. That would assume that second
order of jus cogens can be used to explain their own status. In the alterna-
tive, it would assume the existence of third order rules of jus cogens, which
assume in turn the existence of fourth order rules of jus cogens, and so on ad
infinitum. Like Occam, I prefer simpler solutions. Hence, it is my conclu-
sion that second order rules of jus cogens do not themselves have the status
as jus cogens. They are ordinary customary international law. For this rea-
son, the second order rules of jus cogens designation used in this essay may
seem misplaced. However convenient from the point of view of description,
in order not to cause unnecessary confusion, it should perhaps be avoided.
Thirdly, my jus cogens analysis helps explain how first order rules of jus
cogens can be modified and replaced. Although Art. 53 of the VCLT ex-
pressly envisages the possibility that jus cogens norms be changed, the Con-
vention says very little about this procedure exactly.31 The only indication
given is the requirement that jus cogens norms shall be modified by means
of jus cogens. However, if we accept that norms of jus cogens derive from
one of the usual norm-creating processes recognized by international law, it
appears pretty obvious that the modification of a first order rule of jus co-
gens (R) will always entail, at least partly, a derogation of that same rule.

30 A. Orakhelashvili (note 9), 80.

31 Presumably, it does so for the same reason as it does not explain the creation of jus co-
gens.

ZaöRV 71 (2011)

http://www.zaoerv.de/
© 2011, Max-Planck-Institut für ausländisches öffentliches Recht und Völkerrecht
 The Creation of Jus Cogens – Making Sense of Art. 53 of the Vienna Convention 377

The complicating fact is that according to Art. 53, jus cogens norms must
not be derogated from. This has led some international legal scholars to
conclude that Art. 53 is in fact self-contradictory.32 If a modification of a
first order rule of jus cogens (R) inevitably entails derogation from that same
rule R, and derogations from the rule R are prohibited, the modification of
R would seem to be logically impossible. This argument builds on a wrong-
ful assumption. It assumes that the modification of a first order rule of jus
cogens is achieved by the immediate change of that rule itself. As suggested
in this essay, this is not so. The jus cogens status of rules derives from the
second order rules of jus cogens. Since a first order rule of jus cogens cannot
be derogated from, consequently, the modification of such a rule requires,
first of all, a modification of the second order rules of jus cogens. The sec-
ond order rules are not themselves jus cogens, but ordinary customary in-
ternational law. They can be modified the same way as any rule of custom-
ary international law, through a process of desuetude accompanied by a new
opinio juris and a new practice. This would seem to solve our problem.
Contrary to what some international legal scholars assume, first order rules
of jus cogens can indeed be modified. Modification is accomplished by a two
step procedure: first, there is a modification of the second order rules of jus
cogens, and then, only secondly, is there a modification of the first order
rule itself.
On the negative side – it has to be admitted – my analysis still leaves the
jus cogens concept rather empty. As shown in section 3, the creation of jus
cogens builds on a practical judgment. Applied to the creation of the rule R,
this judgment was stated as follows:

PJ0: “It would be desirable and appropriate if customary international

law permitted no derogations from the rule R and prohibited all modifica-
tion of the rule R by means of ordinary international law.”

The practical judgment PJ0 entails the utterer’s opinion that a determi-
nate, common, and stable pattern of conduct would be desirable and appro-
priate. It does not indicate the utterer’s reason for holding this opinion,
however.33 Naturally, this affects the practical usefulness of my analysis.
While the analysis explains the definition of jus cogens laid down in Art. 53

32 See e.g. A. d’Amato, It’s a Bird, It’s a Plane, It’s Jus Cogens!, Conn. J. Int’l L. 6 (1990),
1, at 5 et seq.; M. J. Glennon, Peremptory Nonsense, Human Rights, Democracy and the Rule
of Law, in: S. Breitenmoser (ed.), Liber Amoricum Luzius Wildhaber, 2007, 1265, at 1269; T.
Meron, On a Hierarchy of International Human Rights, AJIL 80 (1986), 1, at 9.
33 Consequently, different people uttering the practical judgment PJ0 may have different
reasons for thinking said pattern of conduct desirable and appropriate.

ZaöRV 71 (2011)

http://www.zaoerv.de/
© 2011, Max-Planck-Institut für ausländisches öffentliches Recht und Völkerrecht
378 Linderfalk

of the VCLT, it does not help us identify rules of jus cogens in concrete
cases. As should be realized, however, this is more a reflexion of a political
reality than a flaw in the analysis itself. The fact remains that in the interna-
tional community of states there is very little agreement as to what makes
jus cogens desirable and appropriate. In this regard, jus cogens can be viewed
as a meddling link between the consequences ensuing from the application
of the second order rules of jus cogens and ideas that are highly controver-
sial. Politically and morally, jus cogens remains a contested concept.34 My
analysis cannot do anything to change this fact.

34 See U. Linderfalk, What’s so Special About Jus Cogens: On the Distinction between the
Ordinary and the Peremptory International Law, International Community Law Review
(2011), forthcoming.

ZaöRV 71 (2011)

http://www.zaoerv.de/
© 2011, Max-Planck-Institut für ausländisches öffentliches Recht und Völkerrecht