UvA-DARE (Digital Academic Repository)

Party autonomy and its limitations in the Rome II regulation

Published in:
Yearbook of Private International Law

DOI:
10.1515/9783866537200.1.19

Link to publication

Citation for published version (APA):

General rights
It is not permitted to download or to forward/distribute the text or part of it without the consent of the author(s) and/or copyright holder(s),
other than for strictly personal, individual use, unless the work is under an open content license (like Creative Commons).

Disclaimer/Complaints regulations
If you believe that digital publication of certain material infringes any of your rights or (privacy) interests, please let the Library know, stating
your reasons. In case of a legitimate complaint, the Library will make the material inaccessible and/or remove it from the website. Please Ask
the Library: http://uba.uva.nl/en/contact, or a letter to: Library of the University of Amsterdam, Secretariat, Singel 425, 1012 WP Amsterdam,
The Netherlands. You will be contacted as soon as possible.

UvA-DARE is a service provided by the library of the University of Amsterdam (http://dare.uva.nl)

Download date: 05 Jun 2018

Th.M. de Boer ∗

1. Choice-of-law principles

Party autonomy is one of the leading principles of contemporary choice of law. Long gone are the
days when conflicts scholars still argued that the applicable law could not be displaced by party
agreement, as private individuals, by their own volition, cannot rise above the applicable law. 1 Not
until the 1960’s did it dawn on them that this levitation trick might be supported by the adoption of a
conflicts rule allowing the parties to choose the applicable law. 2 Since then, party autonomy has
gradually gained ground in national and international conflicts legislation, and its scope has been ex-
tended from contracts to torts, succession, 3 matrimonial property, 4 and, in some jurisdictions, even
divorce and maintenance. 5 Given the acceptance of party autonomy in such a wide range of choice-of-
law categories, one is bound to assume that it is supported by a common denominator. It has been
argued that the parties’ freedom of choice is justified by the need for legal certainty and predictabil-
ity, 6 by a ‘subjective theory of the proper law,’ 7 or ‘for want of a better solution’. 8 That does not ex-
plain, however, why there are areas of conflicts law in which the parties’ freedom of choice is either
limited or non-existent, while there are other areas in which the need for predictable solutions is met

∗ Professor of private international law and comparative law at the University of Amsterdam. The text of this article was
concluded in December 2007.
1. In this vein, e.g.: J.P.Niboyet, La théorie de l’autonomie de la volonté, Recueil des Cours 1927 (I), p. 53 ff.; M. Caleb,
Essai sur le principe de l’autonomie de la volonté en droit international privé, Sirey, Paris, 1927, p. 167 ff.; I Henri
Hijmans, Algemeene Problemen van Internationaal Privaatrecht, Tjeenk Willink, Zwolle, 1937, p. 169 ff.; P. Scholten,
C. Asser’s Handleiding tot de beoefening van het Nederlandsch Burgerlijk Recht, Algemeen Deel, Tjeenk Willink,
Zwolle, 1954, p. 211.
2. Frank Vischer, Internationales Vertragsrecht, Die kollisonsrechtlichen Reglen der Anknüpfung bei internationalen
Verträgen, Verlag Stämpfli |& Cie, Bern, 1962, p. 23, citing W. Niederer, Einführung in die allgemeinen Lehren des in-
ternationalen Privatrechts, 1954, p. 193, and a decision by the Swiss Bundesgericht, BGE 79 II 299.
3. Cf. Article 5 Hague Convention of 1 August 1989 on the Law Applicable to Succession to the Estates of Deceased
Persons; Articles 90(2) and 91(2) Swiss Bundesgesetz über das Internationale Privatrecht (IPRG); Article 79 Belgian
Wetboek van Internationaal Privaatrecht.
4. Cf. Articles 3 and 6 Hague Convention of 14 March 1978 on the Law Applicable to Matrimonial Property Regimes;
Article 52 Swiss IPRG; Article 15 German Einführungsgesetz zum Bürgerlichen Gesetzbuche (EGBGB); Article 49
Belgian Wetboek van Internationaal Privaatrecht.
5. A limited freedom of choice already exists in several EC member states and is also endorsed in the various drafts of the
so-called ‘Rome III Regulation’ (more properly: the ‘Brussels II-ter Regulation’) and in the drafts of a Regulation on in-
ternational maintenance obligations.
6. This used to be a popular proposition in the 1960’s and 1970’s. Surprisingly, it is still the main argument supporting
party autonomy in the Rome II Regulation; infra § 2.
7. F.A. Mann, The Proper Law of a Contract, 3 ICLQ p. 60 ff. (1950): ‘subjective theory of the proper law: that legal
system which, by their express or implied intention, the parties intended to apply’. Contra: J.H.C. Morris, The Proper
Law of a Contract: A Reply, 3 ICLQ, p. 197 ff., at p. 207 (1950): ‘the subjective theory of the proper law is no more
than a fiction ...’.
8. In a rather pedantic article – my first attempt at academic writing – I took the view that, in a choice-of-law method
which is geared to the designation of the law most closely connected, party autonomy is an anomaly to be tolerated only
– for want of a better solution – if objective choice-of-law rules prove to be inconclusive: Th.M. de Boer, Subjectieve
verwijzing: anomalie of grondbeginsel?, in: H.U. Jessurun d’Oliveira e.a., Partij-invloed in het internationaal privaat-
recht, Kluwer, Deventer, 1974, p. 47-72, at p. 68. My contribution to this collection of essays is best forgotten.

1
by clear-cut uniform conflicts rules. To me, therefore, freedom of choice in conflicts law can only be
explained by linking it to freedom of disposition in substantive law. Where party autonomy is given
free reign in the forum’s domestic law, its national conflicts law is likely to allow a choice of law by
the parties. Conversely, freedom of choice is likely to be denied in areas in which the corresponding
substantive law is meant to protect third parties or public interests and therefore is couched in manda-
tory language. That is why the law governing marriage, parentage, adoption, parental responsibility, or
rights in rem can generally not be chosen by the parties. On the other hand, where a dispute does not
concern anyone but the litigants – which usually implies that they are free to resort to alternative
means of dispute resolution – there is no reason to curb their freedom to choose the applicable law.
That is why party autonomy is now generally accepted as a leading principle in contracts choice of
law.

As a tenet of the choice-of-law method adopted in most legal systems in the world today, party auton-
omy is an anomaly. At any rate, a method that is supposed to designate the law of the country with
which the legal relationship at issue is most closely connected should not leave room for the parties to
choose a law that does not qualify as such. Yet, we have come to accept that the closest connection is
not the only criterion by which the applicable law may be selected. Apart from conflicts rules based on
the principle of the closest connection and the principle of party autonomy, there are at least two dif-
ferent kinds of rules that are not meant to determine the ‘center of gravity’ or the geographical ‘seat’
of the legal relationship. Instead, they give expression to a social policy underlying the forum’s sub-
stantive law, which is either translated into a connecting factor that focuses on the party in need of
protection, or into a number of alternative references, with a preference for the one by which a specific
goal can be achieved. 9 Choice-of-law rules focusing on consumers, employees, children, maintenance
creditors, etc. could be said to reflect the function of the corresponding substantive law, which is pro-
tection of the weaker party. They are based, therefore, on a choice-of-law principle that is known as
the ‘principle of functional allocation,’ or the ‘protection principle’. 10 Alternative reference rules are
based on the so-called ‘favor principle,’ as they allow a choice of the one law among several alterna-
tives that achieves a favored result, such as the formal validity of a legal transaction, 11 the legitimation
of a child born out of wedlock, 12 or the obligation to pay maintenance. 13

9. A more extensive treatment of the principles underlying choice of law can be found in my dissertation, Beyond Lex Loci
Delicti, Conflicts Methodology and Multistate Torts in American Case Law, Kluwer Law and Taxation Publishers, De-
venter/New York, 1987, p. 42 ff., and in my Hague lectures, Facultative Choice of Law, The Procedural Status of
Choice-of-Law Rules and Foreign Law, 257 Recueil des Cours (1996), p. 277 ff.
10. It should be noted that protection is only afforded at the choice-of-law level. A rule referring to the law of the con-
sumer’s habitual residence (cf. Article 5(3) Rome Convention 1980) may leave the consumer empty-handed at the level
of substantive law. Conflicts rules that are based on the principle of functional allocation do achieve equal treatment of
the weaker party, regardless of whether they have entered into a domestic or international legal relationship.
11. Article 1 Hague Convention of 5 October 1961 on the Conflicts of Law Relating to the Form of Testamentary Disposi-
tion; Article 9(1) Rome Convention of 1980 on the Law Applicable to Contractual Obligations.
12. Article 4 Dutch International Parentage Act (Wet conflictenrecht afstamming); Articles 72-74 Swiss IPRG; Articles 20-
21 German EGBGB.
13. Articles 5 and 6 Hague Convention of 2 October 1973 on the Law Applicable to Maintenance Obligations.

2
2. Party autonomy as defined in the Rome II Regulation

According to the preamble of the Rome II Regulation, a rule allowing the parties to choose the law
applicable to a non-contractual obligation is based on two objectives: ‘to respect the principle of party
autonomy and to enhance legal certainty’. This rationale is not very convincing, as it does not answer
the questions of why party autonomy should be respected, or why a choice by the parties would lead to
greater certainty than the one to be achieved by the Regulation’s uniform conflicts rules. At any rate,
the preamble does not explain why the parties should be allowed to choose the law of a country that is
not in any way connected with the non-contractual obligation at issue. Such freedom of choice can
only be understood if it is viewed as a transposition of the parties’ power to dispose of their rights
under substantive law to the level of conflicts law.

It would seem that the main provision on party autonomy in Rome II, Article 14 in its final version,
has not given rise to much debate, as all the elements of the Commission’s original proposal have re-
mained basically intact. The parties’ freedom to choose the applicable law is still subject to a number
of restrictions: (1) the choice-of-law agreement must be entered into after the event giving rise to the
damage occurred; (2) the choice must be express, or demonstrated with reasonable certainty by the
circumstances of the case; (3) it may not affect the rights of third parties; (4) in domestic cases, a
choice of foreign law cannot displace the mandatory provisions of the law that would apply without
the choice; (5) in intra-community cases, the choice of the law of a non-member state cannot displace
mandatory provisions of Community law; (6) the choice of a foreign law cannot displace overriding
mandatory rules of the forum, and (8) a choice-of-law agreement does not affect the law applicable to
non-contractual obligations arising from an infringement of intellectual property rights, as designated
by Article 8. In the final text of the Regulation, a similar restriction has been included with respect to
unfair competition and restrictive trade practices, both covered by Article 6. Another novelty is the
exception to the first restriction if all parties concerned are engaged in commerce. For the rest, Article
14 is essentially the same as Article 10 in the Commission’s first proposal. 14

14. There are some minor differences in wording (‘shall’ instead of ‘must’, ‘where’ instead of ‘if’, etc.), as well as some
legal differences. In the Commission’s proposal, the agreement would only be valid if the parties entered into it ‘after
their dispute arose.’ Article 14 now provides: ‘after the event giving rise to the damage occurred.’ The time factor in Ar-
ticle 10(2) and 10(3) – all elements located in one state or in one or more member states – has been changed from ‘at the
time when the loss was sustained’ into ‘when the event giving rise to the damage occurs.’

3
The restrictions on the parties’ freedom of choice in domestic 15 and intra-community cases 16 can also
be found in the Rome I Regulation. 17 The same can be said about the priority of the forum’s overrid-
ing mandatory provisions, 18 the criteria for a valid choice, 19 and the rights of third parties. 20 That
leaves us with three limitations that are specific to Rome II and, therefore, deserve separate treatment.
I will first deal with the bar on party autonomy in the fields of unfair competition (§ 3) and intellectual
property (§ 4), and then with the requirement that a choice-of-law agreement between private parties
can only be made post factum (§ 5). As I am fully aware of the fact that in the area of non-contractual
obligations parties seldom exercise their freedom of choice, the following remarks are primarily meant
to stir up academic debate, not to illuminate the limits of Article 14 in actual practice.

3. Article 6: no freedom of choice in cases relating to unfair competition or restriction of trade

Under Article 6, non-contractual obligations arising out of an act of unfair competition are subject to
the law of the country where competitive relations or the collective interests of consumers are (or are
likely to be) affected. If the damage results from a restriction of competition, the law to be applied is
the law of the country where the market is (or is likely to be) affected. In either situation, the applica-
ble law cannot be displaced by a choice-of-law agreement: Article 6(4). This restriction is explained,
more or less, in recital 21, where it says that ‘the conflict-of-law rule should protect competitors, con-
sumers and the general public and ensure that the market economy functions properly.’ Read in con-
junction with the reference to ‘competitive relations and collective interests of consumers’ in Article
6(1), this statement suggests that unfair competition and restrictive trade practices affect public rather

15 . Article 14(2) referring to situations in which a choice of foreign law is the only foreign element. The occurrence, the
resulting loss, the habitual residence of the parties, etc. are all located in one and the same country, but not the country
whose law was chosen.
16. Article 14(3), now referring to ‘elements ... located in one or more Member States.’ In the Commission’s first proposal,
Article 10(3) referred to ‘elements ... located in one of the Member States.’ Strictly speaking, this phrasing would make
Article 10(3) redundant next to Article 10 (2), while it would not have the effect of turning the EU member states into
one ‘territoire juridique.’
17. In the Rome Convention of 1980, there was a restriction on party autonomy only in domestic cases: Article 3(3). At the
time of this writing (December 2007), a political agreement was reached on the text of the Rome I Regulation. Article
3(3) and Article 3(4) now address the same issues as those covered by Article 14(2) and 14(3) Rome II.
18. Article 16 Rome II, only referring to overriding mandatory provisions of the law of the forum; cf. Article 9(2) Rome I.
There is no reference to overriding mandatory provisions of the law of other countries – cf. Article 9(3) Rome I, how-
ever limited in scope – probably because it is difficult to imagine what interests other countries could have, in the field
of non-contractual obligations, in the application of their law. Even if the parties have chosen a different law, under Ar-
ticle 17 rules of safety and conduct in force in the country where the harmful event occurred must be taken into account
anyhow.
19. Under Rome II, ‘the choice shall be expressed or demonstrated with reasonable certainty by the circumstances of the
case.’ Article 3(1) Rome I is phrased differently: ‘The choice shall be made expressly or clearly demonstrated by the
terms of the contract or the circumstances of the case.’ Hopefully, both provisions have the same meaning. Curiously,
the issue of whether the parties actually reached a choice-of-law agreement is not subjected – as in Rome I – to the law
that would apply if the agreement were valid, but should be resolved by ‘respect[ing] the intentions of the parties’ (pre-
amble, recital 31).
20. In Rome I, Article 3(2) only refers to a change in the law chosen by the parties: such a change ‘shall not .... adversely
affect the rights of third parties.’ In Rome II, there is no provision on the possibility of changing the chosen law: ‘The
choice shall not prejudice the rights of third parties.’ I assume that Rome I does not allow the rights of third parties to be
prejudiced by any choice-of-law agreement, and that under Rome II the parties are free to change their choice-of-law
agreement, without prejudice to the rights of third parties, of course.

4
than private interests. Or, to put it differently, in the area of unfair competition and restrictive trade
practices, interests of a higher order than those of individual competitors are at stake. On the level of
substantive law, those interests are generally protected by mandatory provisions, 21 which may range
from standards of conduct to rules on the determination of the burden of proof, or on the nature and
measure of damages, etc. As explained in § 1, the less room there is for party agreement on this level
the less freedom the parties should be allowed on the choice-of-law level. In this respect, the ban on
choice-of-law agreements in Article 6(4) is quite justified.

Unfortunately, there are several provisions in Rome II that are difficult to square with this line of rea-
soning. If the parties’ freedom of choice corresponds with their freedom of disposition under substan-
tive law, it is hard to see why the drafters of Rome II have chosen to prohibit choice-of-law agree-
ments in all unfair competition cases, even those in which the defendant’s conduct does not affect
collective interests – the situation covered by Article 6(2) – while the parties’ freedom of choice is not
curbed in any way by Article 7, the provision on environmental damage, or Article 9 covering indus-
trial actions. ‘Where an act of unfair competition affects exclusively the interests of a specific com-
petitor,’ Article 6(2) calls for the application of the general conflict-of-law rules laid down in Article
4. Yet, a choice-of-law agreement is still blocked by Article 6(4). If the law of the market may be dis-
placed in this situation by the law of the parties’ common habitual residence or the law of another
more closely connected country, it is hard to see why it cannot be displaced by the law designated by
the parties in accordance with Article 14. 22 In my view, this is a mistake that should be corrected in
the next version of Rome II.

On that occasion, the relation between party autonomy and environmental law should be reconsidered
as well. Article 7, the provision on the law applicable to environmental damage, calls for the applica-
tion of the law of the place of injury, or – at the victim’s option – the (probably more favorable) law of
the place of conduct wrong. Since protection of the environment would seem to be no less a public
interest of the state concerned than protection against unfair competition or restrictive trade practices, I
fail to see why Article 7 does not also contain a restriction on a choice by both parties similar to the
one laid down in Article 6(4). The same could be said with regard to the provision covering industrial
actions: Article 9 refers to the lex loci delicti, subject to the common habitual residence exception of

21. Laws that are meant to protect trade and commerce from restraints, monopolies, price-fixing, price discrimination, etc.
(antitrust laws) qualify as ‘overriding mandatory rules’ within the meaning of Article 9(1) of the latest draft of the
Rome I Regulation, as they are crucial to the ‘political, social, or economic organisation’ of the country in which they
are in force. However, Rome II only gives precedence to overriding mandatory provisions of the forum state (Article
16). If choice-of-law agreements were allowed under Article 6, the parties could evade the rules of foreign competition
law, whether or not they would qualify as overriding mandatory provisions.
22. Cf. Karl Kreuzer, Tort Liability in General, in: The Unification of Choice of Law Rules on Torts and Other Non Con-
tractual Obligations in Europe, The ‘Rome II’ Proposal (ed. Alberto Malatesta), Studi e Pubblicazioni della Rivista di
Diritto Internazionale Privato e Processuale no. 63, Cedam, Padova, 2006, p. 45-58, at p. 55: ‘Where an act of unfair
competition exclusively affects the interests of an individual competitor, there is no reason why the parties should not
be allowed to select, by agreement, the law applicable to their relations.’

5
Article 4(2), but it does not rule out choice-of-law agreements. Still, it would seem that industrial ac-
tions affect social and economic interests of a higher order than those of the parties concerned, justify-
ing a restriction on party autonomy similar to the one included in Article 6(4).

4. No freedom of choice in cases relating to infringement of intellectual property rights

It is generally assumed that the territorial nature of intellectual property rights does not support appli-
cation of any other law but the law of the country for which protection is claimed (lex loci protec-
tionis). 23 Hence, the Rome II Regulation neither leaves room for the law of the parties’ common ha-
bitual residence, nor provides for a proper law exception. Choice-of-law agreements are expressly
prohibited in Article 8(3). The exclusive reign of lex loci protectionis is justified in recital 23 by the
invocation of a ‘universally acknowledged principle’ that ‘should be preserved’. A more compelling
explanation is given by Kreuzer: ‘The exclusion of an agreement on the law applicable to infringe-
ments of intellectual property rights is due to the fact that in these cases the rights are existing only on
a territorial basis so that the parties possibly could select a legal system under which no intellectual
property right may exist.’ 24

However, if infringements of intellectual property rights are compared to other ways of causing dam-
age to another’s property, it can hardly be maintained that the law governing the tort should also be
applicable to incidental questions such as the existence or extent of property rights. Those issues be-
long to a different choice-of-law category and should be treated accordingly. A similar approach could
be taken with regard to intellectual property rights. I agree with Kreuzer that the existence and extent
of such rights should be measured against the lex loci protectionis. The question of whether a particu-
lar act constitutes an infringement – in other words, the unlawfulness issue – cannot be separated from
questions pertaining to the existence, scope, and duration of the intellectual property right at issue. 25 I
see no reason, however, why other issues – such as: who can be held liable, the limitation or division
of liability, or the nature and measure of damages – could not be subjected to a different law. In the
Swiss statute on private international law, Article 110(2) does allow the parties to choose Swiss law
with regard to ‘claims resulting from an infringement of intellectual property rights,’ a phrase which is

23. Cf. Mireille M.M. van Eechoud, Choice of Law in Copyright and Related Rights, Alternatives to the Lex Protectionis,
Kluwer Law International, The Hague/London/New York, 2003, p. 95 (with numerous citations in footnote 263):
‘There is however quite widespread agreement that the general conflict rule for copyright (and other intellectual prop-
erty) is the one referring to the law of the country for which protection is claimed (Schutzland principle; lex protec-
tionis). Some infer this conflict rule from the territorial nature of copyright. Others derive it from the national treatment
principle as enshrined in the Berne Convention and other treaties, or see the lex protectionis expressed directly in Arti-
cle 5(2) BC and similar clauses.’ See also: James J. Fawcett & Paul Torremans, Intellectual Property and Private Inter-
national Law, Clarendon Press, Oxford, 1998, p. 462 ff.
24. Kreuzer, supra note 21, p. 55/56.
25. Cf. Van Eechoud, supra note 22, p. 206 ff., p. 229; id., Overleefde teritorialiteit: grensoverschrijdende auteursrechtin-
breuken onder de ‘Rome II’-ontwerpverordening, AMI/Informatierecht 2005, p. 45-56.

6
meant to exclude the unlawfulness issue from the scope of choice-of-law agreements. 26 While I would
not limit the parties’ options to a choice of lex fori, there is much to be said for an approach in which
the legal consequences of infringement could be subjected either to the law of the country with which
they are most closely connected (lex domicilii communis, proper law, or accessory choice of law),
rather than the lex loci protectionis, or to the law chosen by the parties.

5. Freedom of choice before or after the event

The last restriction on party autonomy I should like to discuss is the one laid down in Article 14(1),
requiring that choice-of-law agreements between non-professional parties are entered into after the
event giving rise to the damage occurred. Similar restrictions can be found in, e.g., the German, Bel-
gian, and Swiss statutes on private international law. 27 It would seem that they all rest on the same
rationale, which is: preventing abuse of choice-of-law agreements where the parties do not have equal
bargaining power. 28 Whether or not such fear of abuse is justified, it should be borne in mind that most
choice-of-law agreements covering an anticipated tort are made in connection with a contractual rela-
tionship between the parties. They – or at least one of them – want to make sure that the law of their
choice will be applied to any dispute that may arise as a result of their relationship, regardless of
whether the action sounds in contract or in tort. Even if their agreement would be invalid under Article
14(1), an action sounding in tort would still be governed by the law of their choice, as there is likely to
be a close connection between their contractual relationship and the tort at issue, 29 the situation ex-
pressly mentioned in Article 4(3). In other words, even if a choice-of-law agreement would be invalid
because it was entered into before the event giving rise to the damage occurred, it would be replaced
by an accessory choice of the (chosen) lex contractus. 30 The remaining cases – those in which the
parties have entered into an agreement on the law applicable to a future tort while they are not bound

26. IPRG Kommentar, Kommentar zum Bundesgesetz über das Internationale Privatrrecht (IPRG) vom 1. Januar 1989
(eds. Anton Heini e.a), Schulthess Polygraphischer Verlag, Zürich, 1993, p. 865 (Frank Vischer): ‘Nach dem Wortlaut
von Art. 110 Abs. 2 ist die Rechtswahl nur für Ansprüche aus der Verletzung, nicht für die Frage der Verletzung selbst,
zugelassen.’ The German text of Article 110(2) refers to ‘Ansprüche aus Verletzung von Immaterialgüterrechten’, the
French text to ‘les prétentions consécutives à un acte illicite’, the Italian text to ‘pretese derivanti dalla violazione di
diritti immateriali’, all of them suggesting that unlawfulness as such is a separate issue.
27. Germany: Article 42 EGBGB; Belgium: Article 101 Wetboek van Internationaal Privaatrecht; Switzerland: Article 132
IPRG. In the Netherlands, on the other hand, the parties are free to choose the applicable law both before and after the
tortious event: Article 6 Wet conflictenrecht onrechtmatige daad.
28. Cf. the reason given in the Swiss Message concernant une loi fédérale sur le droit international privé (loi de DIP) du 10
novembre 1982/Botschaft des Bundesrats zum Bundesgesetz über das internationale Privatrecht (IPR-Gesetz) vom 10.
November 1982, BBl. 1983 I p. 263-519, at no. 284.222: ‘Dans le droit délictuel international, on a affaire de plus en
plus à des parties qui ne bénéficient pas d’une capacité de négocier égale. Adopter une réglementation trop libérale re-
viendrait à exposer la partie la plus faible au danger d’être exploitée.’
29. Cf. Jan Kropholler, Internationales Privatrecht einschließlich der Grundbegriffe des Internationalen Zivilverfahrens-
rechts, 6. Aufl., Mohr Siebeck, Tübingen 2006, 53 IV 5 at p. 531/532; Kurt Siehr, Internationales Privatrecht, Deut-
sches und europäisches Kollisionsrecht für Studium und Praxis, C.F. Müller Verlag, Heidelberg, 2001, p. 241.
30. With regard to their contractual relationship, the parties’ freedom of choice could be restricted as well. The protection of
the weaker party then no longer depends on Article 14(1) Rome II, but on, e.g., Article 6(2) or Article 8(1) Rome I in its
version of 29 November 2007 (Article 5(2) and Article 6(1) of the Rome Convention of 1980).

7
by a pre-existing relationship – will be so rare that they hardly warrant the elaborate distinction be-
tween anterior and posterior agreements formulated in Article 14(1)(a) and (b).

Still, this concern for the plight of the weaker party reveals another flaw in the methodological under-
pinnings of the regulation. If an imbalance in the parties’ bargaining power should be remedied by a
provision like Article 14(1) – which would seem to be based on the same rationale as the one underly-
ing the validity of forum selection clauses in insurance contracts, consumer contracts, and contracts of
employment under the Brussels I Regulation 31 – it is hard to see why ‘non-commercial parties’ should
not be given the opportunity to rely on an anterior choice-of-law agreement if that would benefit their
case, just like the Brussels I Regulation allows the weaker party to rely on an anterior choice-of-forum
agreement.

On a more fundamental level, it could be asked whether the desire to protect the weaker party, as ex-
pressed in Article 14(1), should not be mirrored in the rules that apply in the absence of a choice-of-
law agreement, notably the main rule of Article 4. In an Explanatory Memorandum accompanying the
Commission’s original proposal, the lex loci damni rule is said to reflect ‘the modern concept of the
law of civil liability, which is no longer ... oriented towards punishing for fault-based conduct: nowa-
days, it is the compensation function that dominates.’ 32 This view is upheld in recital 16 of the pream-
ble: ‘A connection with the country where the direct damage occurred strikes a fair balance between
the interests of the person claimed to be liable and the person sustaining the damage, and also reflects
the modern approach to civil liability and the development of systems of strict liability.’ These state-
ments suggest that Article 4(1) is based on the principle of functional allocation, translating the com-
pensatory function of modern tort law into a connecting factor that is focused on the tort victim as the
party to be protected. This suggestion is reinforced by the assumption that the place of injury coincides
with the victim’s habitual residence ‘in most cases.’ 33

Yet, it can hardly be maintained that the lex loci damni rule reflects the compensatory function of
modern tort law. Statistically, I would say, the place of injury might as well coincide with the tortfea-
sor’s place of residence, if there is any coincidence at all. A functional approach would assure that
victims are afforded compensation in accordance with their own law, possibly subject to a foreseeabil-
ity exception. 34 In my view, the lex loci damni rule is based on the principle of the closest connection,
reflecting no other choice-of-law objective than the achievement of predictable and uniform results.

31. Articles 13(1), 17(1), and 21(1), respectively, all of them referring to choice-of-forum agreements entered into after the
dispute arose.
32. Explanatory Memorandum, COM(2003) 427 final, at p. 12.
33 . Ibid., at p. 11: ‘In most cases [the lex loci damni] corresponds to the law of the injured party’s country of residence.’
34. Cf. the provision on product liability, allowing for an exception if the person claimed to be liable could not foresee that
his product would be marketed in one of the countries listed in Article 5(1). In its original version, Article 5 only re-

8
There is a methodological discrepancy, therefore, between provisions in Rome II that are meant to
further substantive policies – such as the protection of public interests in Article 6, or the protection of
the weaker party in Article 14(1) – and those that remain totally blind to the result they achieve, de-
spite the token remarks on the function of modern tort law in the Explanatory Memorandum and the
preamble.

6. Conclusion

The practical import of a provision on choice-of-law agreements in the area of non-contractual obliga-
tions is undoubtedly limited. Where disputes arise between parties that are not bound by a pre-existing
relationship, they have little cause to agree on a law that could put one of them at a disadvantage. Still,
the Rome II Regulation would be incomplete without a provision on party autonomy and its limita-
tions. Article 14 allows the parties to choose any law they want – not necessarily the lex fori or the law
of one of the countries with which the case is in some way connected. Unless both parties are engaged
in commercial activities, a valid agreement can only be made after the event giving rise to the damage
occurred, a restriction meant to protect the interests of the weaker party. Collective interests are
deemed to be affected by infringements of national or Community competition law, hence the ban on
choice-of-law agreements in Article 6(4). A similar prohibition can be found in Article 8(3). The latter
restriction does not seem to be warranted by a concern for public interests but – more likely – by an
outdated choice-of-law approach to disputes over the infringement of intellectual property rights.

Studying the various drafts and the final version of the Rome II Regulation, one can hardly escape the
impression that it has come to be based on rather disparate choice-of-law considerations, ranging from
the need for uniformity, predictability, and legal certainty to ‘the function of modern tort law, the pro-
tection of ‘public interests,’ and a concern for ‘weaker parties’. But even if methodological eclecticism
is a characteristic of contemporary choice of law, consistency is required in assessing the choice-of-
law objective to be achieved in a particular area, and in choosing the choice-of-law technique best
suited to achieve it. Thus, if collective interests call for a ban on choice-of-law agreements, party
autonomy should be ruled out in all areas in which such interests are at stake. If the exclusion of ex
ante agreements is deemed to be justified by the interests of the weaker party, an exception should be
made for cases in which the weaker party wants to rely on such an agreement. And, generally, if the
main rule in Rome II is said to be supported by ‘the compensatory function of modern tort law’, it
should not ‘strike a fair balance between the interests of the person claimed to be liable and the person
sustaining damage,’ as it says in recital 16; rather, it should focus on the person to be compensated. To
me, these inconsistencies are illustrative of a general lack of awareness of the bedrock of modern

ferred to the law of the victim’s habitual residence, a clear example of functional allocation. The current version is
based on a ‘cascade system of connecting factors, together with a foreseeability clause’ (recital 20).

9
European conflicts law among those involved in drafting EU conflict-of-law instruments. Ever opti-
mistic, I trust that their understanding of choice-of-law methodology will have improved before the
Rome II Regulation is up for revision.

10