A. Common Law

Common law courts have long endorsed concurrent claims in private international law. Sometimes, this occurs implicitly, when courts simply assume that Henderson applies in the international context, as in Australia, Canada and New Zealand. By contrast, English law – while still untouched by EU Regulations – explicitly endorsed concurrent claims in private international law, at least in the choice-of-law context. In Coupland v Arabian Gulf Oil Co.,Footnote ⁴⁰ a Libyan company's British employee suffered a workplace injury at the company's Libyan premises and sued the company in negligence under English law. The company argued that its duties were limited to those under the Libyan law employment contract.Footnote ⁴¹ Goff L.J. disagreed: “[t]he plaintiff can advance his claim, as he wishes, either in contract or in tort”,Footnote ⁴² and unless the contract itself contained a “clause which restricts or limits the plaintiff's right to claim damages in tort”, the tort claim can be advanced.Footnote ⁴³ A similar conclusion was reached in Base Metal Trading v Shamurin.Footnote ⁴⁴ A Guernsey company sued its director for breaches of contract and tortious duties of care for alleged speculative trades he had made on the company's behalf on the London Metal Exchange. The director's unwritten employment contract was governed by Russian law, under which the contract claim failed,Footnote ⁴⁵ but the company argued that the breaches of tortious duties of care were governed by English law instead. The director countered that the tort claim should be characterised as contractual as well, since (citing Briggs) there was “only one duty which ha[d] been broken”, and the accumulation of claims would disproportionately disadvantage him.Footnote ⁴⁶ Tuckey L.J. agreed with the company, reasoning that “[d]omestic law allows concurrent claims in contract and tort and it has always been assumed that English private international law does so also”.Footnote ⁴⁷ In support, he cited Coupland, which he read as “echo[ing] precisely in a public [sic] international law context what Lord Goff of Chieveley was later to say in the context of the domestic law about concurrent claims in contract and tort in the Henderson case”.Footnote ⁴⁸

The common law's clearest endorsement of concurrent claims in choice-of-law is the Singapore Court of Appeal's decision of Rickshaw Investments Ltd. v Nicolai Baron von Uexkull.Footnote ⁴⁹ Seabed, a German company, employed a marketing agent under a German law agreement to find buyers in Singapore for artefacts it had salvaged. Subsequently, Seabed's successor, Rickshaw, terminated the employment agreement and sold the artefacts to Singapore's tourism authorities. The agent sued Rickshaw in Germany for breach of the employment agreement, alleging that he was owed salary and expenses. Rickshaw then counterclaimed in Singapore in, inter alia, tort, alleging that the agent had withheld information from Seabed and divulged information to the authorities, jeopardising Rickshaw's position in negotiations with the authorities and forcing them to sell the artefacts at an undervalue. The agent applied to stay the Singapore proceedings on the grounds that Germany, not Singapore, was the natural forum, and argued that Rickshaw should be allowed to sue him only for breach of the employment agreement and not in tort. The Court of Appeal disagreed; citing Henderson, it held that “absent bad faith”, there is “no reason why [claimants] should be denied the freedom of choice to frame their causes of action” in whichever way was “more advantageous” to them.Footnote ⁵⁰ The agent's alleged tort had a “separate legal existence from his contractual existence and breaches thereof”, and since “the latter does not expressly limit or exclude the former”, Rickshaw could sue him “concurrently in tort and contract”.Footnote ⁵¹

Common law courts have also recognised the claimant's ability to bring concurrent tort and contract claims with different applicable jurisdictional rules, albeit less explicitly than they have in the choice-of-law context. The UK Supreme Court's two decisions on jurisdictional issues arising from Lady Brownlie's claims, regarding the accident Sir Ian Brownlie and his family met while holidaying at the Four Seasons hotel in Cairo, are examples. In Brownlie v Four Seasons Holdings Inc., Lady Brownlie sued the hotel's Canadian holding company in contract and in tort. Lord Sumption, with whom the entire Court agreed on this point, denied her permission to serve the holding company out of jurisdiction, because she lacked a good arguable case in contract, because Lady Brownlie had evidently not contracted with the holding company, and in tort, because the holding company had not operated the Cairo hotel.Footnote ⁵² At no point, however, was Lady Brownlie's ability to bring concurrent tort and contract claims on identical facts, with different applicable jurisdictional gateways, ever questioned.

Subsequently, in Brownlie v FS Cairo (Nile Plaza) L.L.C.,Footnote ⁵³ Lady Brownlie brought similar claims against the Cairo hotel's Egyptian operating company. Parties agreed that both contract and tort claims were governed by Egyptian law,Footnote ⁵⁴ but the operating company resisted service on grounds that Lady Brownlie again lacked a good arguable case against it in contract, because that would be time-barred under Egyptian law, and in tort, because the Egyptian law doctrine of non-cumul precluded concurrent liability in contract and tort.Footnote ⁵⁵ Lord Leggatt, with whom the entire Court agreed on this point, disagreed with the company on both points. On the contract claim, it was arguable that the Egyptian limitation period only ran from the date Lady Brownlie learnt the operating company's identity.Footnote ⁵⁶ On the tort claim, although “the doctrine of non cumul” was “a basic principle of civil law”,Footnote ⁵⁷ it was unclear on the Egyptian law evidence whether Lady Brownlie's tort claim would be negatived by it, so the claim was arguable.Footnote ⁵⁸ Lord Leggatt's judgment affirms that English law recognises concurrent claims in private international law even if the applicable law subsequently denies concurrent liability. The latter is a concern for trial, or at most for interlocutory determinations on whether claims should be summarily dismissed,Footnote ⁵⁹ but it does not affect the claimant's ability to rely on the tort gateway at the jurisdictional stage.

Thus, common law courts generally endorse the claimant's ability to bring concurrent contract and tort claims on one set of facts, despite this having different consequences for jurisdiction and choice-of-law. However, they have also recognised that certain limits should be placed on the claimant's freedom to choose between concurrent claims. These limits, though, are nevertheless only tangential limits based on separate concerns from those which would preclude concurrent claims outright.

The first of such limits is Rickshaw's “bad faith” limit, under which concurrent tort claims may be excluded by parties’ express agreement. This reflects the point raised earlier, that parties may, by agreeing to subject tort claims to the same forum or governing law as concurrent contract claims, effectively eliminate the effects of recognising concurrent claims. Jurisdiction and choice-of-law agreements, however, cannot comprehensively eliminate those effects. This is because the question of whether parties have subjected their tort claims to the same forum or governing law as their contract claim involves two questions: whether as a matter of contractual interpretation the jurisdiction or choice-of-law agreement's scope covers the tort claim; and whether as a matter of legal policy that agreement should be enforceable.Footnote ⁶⁰ Both these questions are conceptually anterior to the question of whether concurrent claims in contract and tort should be recognised in the first place. They also involve different principled concerns (i.e. whether parties have and may validly exercise an actual choice over the forum and law which their tort claims will be subject to), and so cannot be invoked in every situation where concurrent claims arise.

For jurisdiction agreements, the main question will be the first one of contractual interpretation, on whether parties actually intended their jurisdiction agreement to cover a related tort claim as well. Such an intention, importantly, can never be taken for granted. Though since Premium Nafta Products Ltd. v Fili Shipping Co. Ltd. (“Fiona Trust”)Footnote ⁶¹ English courts generally “start from the assumption that the parties, as rational businessmen” intend all disputes “arising out” of their commercial relationship “to be decided by the same tribunal”,Footnote ⁶² there is doubt even in English law whether this is a true evidential “presumption”, or a mere restatement of the common law's contextual approach to contractual interpretation.Footnote ⁶³ Other apex common law courts in Australia and Singapore have explicitly favoured the latter position.Footnote ⁶⁴ Moreover, whatever the Fiona Trust rule's effect, it is a rule of substantive contract law rather than private international law, and so applies only to jurisdiction agreements governed by English law.Footnote ⁶⁵ For choice-of-law agreements, the second question of whether such an intention to subject tort claims to the contractual governing law ex ante, even if clear, should be enforceable, further prevents such agreements from limiting concurrent claims. This is because most common law jurisdictions – like Australia, Canada and the UK prior to Rome II – do not generally consider choice-of-law agreements for tort claims enforceable.Footnote ⁶⁶ Thus, jurisdiction and choice-of-law agreements can at best only place tangential direct limits on the effects of concurrent claims.

Another potential limit on concurrent claims would be tort choice-of-law exceptions, like the common law's flexible exception and PILA s. 12. These, however, also place only tangential rather than comprehensive limits on concurrent claims. The common law's flexible exception, for example, does not operate simply because the tort claim arises concurrently with a contract claim. For a start, it is unclear whether the exception may select any law other than the lex fori or the lex loci delicti.Footnote ⁶⁷ And even if it can, the exception will only do so when both those laws are, as the Court in Rickshaw put it, “purely fortuitous”.Footnote ⁶⁸ In particular, for the lex loci delicti to be fortuitous, the tort claim must involve such a transnational array of facts that the very idea of territorial connections becomes arbitrary. This will not be the case in most disputes where contract and tort claims arise concurrently: in Rickshaw, for instance, the agent's alleged torts caused harm in Singapore, so the flexible exception could not be invoked to apply German law, which governed his employment contract.Footnote ⁶⁹

PILA s. 12, by contrast, is more capaciously worded and thus apparently broader than the common law's flexible exception. In Trafigura Beheer B.V. v Kookmin Bank,Footnote ⁷⁰ a South Korean issuing bank sued Dutch sellers of oils for failing to provide it with a full set of bills of lading. The bank was out of pocket since the buyers, to whom the goods were released, had become insolvent. The sellers had done this because they discovered errors in the original bills of lading issued by its Singapore office and went about correcting them there for reissuance. The bank sued the sellers in tort under South Korean law, argued that it had a legitimate expectation to receive valid bills of lading.Footnote ⁷¹ Aikens J. accepted that the bank's claim, which essentially involved the seller's acts at its Singapore office, was properly characterised as tortious, so Singapore law applied under the default lex loci delicti rule in section 11.Footnote ⁷² However, under section 12, “the law that the parties have expressly or impliedly chosen to govern their pre-existing contractual relationship” which “g[a]ve rise to events constituting the alleged tort” could apply.Footnote ⁷³ This was English law, since the sale contract, the letter of credit and other related contracts were all governed by English law.Footnote ⁷⁴

Trafigura establishes three significant propositions about PILA s. 12. First, the relevant enquiry is a multifactorial one that necessitates a “value-judgment as to the significance of factors that point to an applicable law other than that identified under [section 11]”; the lex loci delicti is displaced only when the former factors are “substantially” more significant.Footnote ⁷⁵ Second, a contract between parties could be one such factor to displace the lex loci delicti.Footnote ⁷⁶ Third, there is no pre-requisite that the tort's connection with lex loci delicti be insignificant or “fortuitous”.Footnote ⁷⁷ However, it goes too far to suggest, as Briggs did, that after Trafigura section 12 practically excludes the effects of recognising concurrent claims.Footnote ⁷⁸ After all, Aikens J. said that the governing law of parties’ contract is simply “a factor” under the broader multifactorial enquiry, and applied it because, on the facts, it was the most significant factor by far.Footnote ⁷⁹ Properly understood, then, while PILA s. 12 is broader than the flexible exception, it does not limit the effects of concurrent claims outright. Instead, it only places a tangential limit on the effects of recognising concurrent claims, triggered when relevant factors, including but not limited to the parties’ contractual relationship, substantially point toward a law other than the lex loci delicti.

B. European Union Regulations

The recognition of concurrent claims has often arisen for consideration within the EU jurisdictional context, when courts determine whether proceedings should be fitted under Article 7(1)(a) or 7(2) of Brussels Ia. In Athanasios Kalfelis v Bankhaus Schröder, Münchmeyer, Hengst und Co.,Footnote ⁸⁰ involving similar provisions under the Brussels Convention, the CJEU recognised concurrent claims in contract and tort, giving claimants a choice between both Articles. An investor made unprofitable investments with a Luxembourgian bank through a German intermediary on the latter's advice. The investor sued the bank in tort and contract, and the bank argued that the tort forum should also have ancillary jurisdiction over the contract claim. The CJEU disagreed: under the Brussels Convention's “autonomous” taxonomy of legal categories, “a court which has jurisdiction … over an action in so far as it is based on tort or delict does not have jurisdiction over that action in so far as it is not so based”.Footnote ⁸¹ Thus, the investor's different claims could legitimately lead to “different aspects of the same dispute being adjudicated upon by different courts”.Footnote ⁸²

Kalfelis, however, was qualified by Marc Brogsitter v Fabrication de Montres Normandes EURL.Footnote ⁸³ There, the CJEU laid out a highly ambiguous ratio which it did not apply to the facts (which are thus irrelevant for our purposes), which would later be read by subsequent courts in three different ways, qualifying a claimant's ability to bring concurrent contract and tort claims to different extents. The first reading of Brogsitter recognises and endorses concurrent claims in general subject to a minor limit, which may be described as the “indispensability” principle: claimants can choose between concurrent contract and tort claims unless the latter requires proof of the contract's existence and contents to succeed. This arises from paragraph 25 of Brogsitter: tortious conduct would raise “matters relating to a contract” only “where the interpretation of the contract which links the defendant to the applicant is indispensable to establish the lawful or … unlawful nature of the conduct complained of”.Footnote ⁸⁴ In Source Ltd. v TUV Rheinland,Footnote ⁸⁵ Staughton L.J. held that tort claims should be characterised as contractual under the Brussels Convention if they could not be brought “independently of a contract” – that is, if the contract had to be “established” for the claim to succeed.Footnote ⁸⁶ A second reading of Brogsitter places a broader limit on the recognition of concurrent claims, which may be described as the “reasonableness” principle: a tort claim will be precluded if, on the facts, a contract claim could reasonably have been brought instead. Paragraph 26 of Brogsitter appears to support this: claims would fall within the Brussels Regulation's contract jurisdictional rule when their “purpose … is to seek damages, the legal basis for which can reasonably be regarded as a breach of the rights and obligations set out in the contract which binds the parties”.Footnote ⁸⁷ In Bosworth v Arcadia Ltd, Gross L.J. held that Brogsitter's “true ratio” was that a tort claim should be characterised as a “matter relating to a contract” if “the legal basis of the claim can reasonably be regarded as a breach of contract”.Footnote ⁸⁸ A third reading of Brogsitter rejects concurrent claims entirely. Part of paragraph 24 of Brogsitter supports this: a tort claim is contractual under Brussels if “the conduct complained of may be considered a breach of contract”.Footnote ⁸⁹ In Holterman Ferho Exploitatie BV v Spies von Büllesheim, the CJEU read Brogsitter as authority that tort claims would be characterised as raising “matters relating to a contract” if “the conduct complained of may be considered a breach of … contract”.Footnote ⁹⁰ If any set of facts which “may” give rise to a contract claim will do so to the exclusion of possible tort claims, concurrent claims are impossible.

Recently, in Wikingerhof, the CJEU confirmed the first “indispensability” principle reading of Brogsitter as correct.Footnote ⁹¹ A German hotel agreed to be listed on a Dutch company's online booking platform, subject to the latter's unilateral ability to redefine the terms of their agreement. The hotel later sued the booking platform for abusing its dominant position in contravention of German competition law. The hotel said that this raised “matters relating to tort” under Brussels Ia Article 7(2), giving German courts jurisdiction; the booking platform countered that the claim raised “matters relating to a contract” under Article 7(1)(a), giving Dutch courts jurisdiction. The CJEU agreed with the hotel. Although the “contract” and “tort” jurisdictional rules were “mutually exclusive”, Brussels Ia “is characterised by the possibility which it confers on the applicant of relying on one of the rules of special jurisdiction laid down by that regulation”.Footnote ⁹² The court had to apply a two-step analysis: first, follow “the applicant's choice whether or not to rely on one of those rules of special jurisdiction”;Footnote ⁹³ second, assess whether the applicant's claim involved matters falling within that rule, “by reference to the obligation … which constitutes the cause of action”.Footnote ⁹⁴ On the latter point, a “claim”Footnote ⁹⁵ would fall under Article 7(2) where it relies on a “breach of an obligation imposed by law, and where it does not appear indispensable to examine the content of the contract concluded with the defendant in order to assess whether the conduct of which the latter is accused is lawful or unlawful”.Footnote ⁹⁶ Since the hotel's claim was that the booking platform used its dominant position to force it to contract, that contract was irrelevant to establishing the lawfulness of the booking platform's conduct, and so the claim was tortious.Footnote ⁹⁷

After Wikingerhof, then, Brussels Ia generally recognises concurrent claims subject only to the “indispensability” principle. This position is mirrored in the choice-of-law context under the Rome Regulations. In ERGO Insurance SE v Gjensidige Baltic AAS, the CJEU held that, in interpreting the concepts of “contractual” and “non-contractual” obligations within the Rome Regulations, it was important to uphold “the aim of consistency … in the application of the Brussels I Regulation which … draws a distinction … between matters relating to contract and matters relating to tort, delict and quasi-delict”.Footnote ⁹⁸ Echoing Kalfelis, the Court then defined “contractual” obligations as “legal obligation[s] freely consented to by one person towards another”, and “non-contractual” obligations as “all actions which seek to establish the liability of a defendant and are not related to a contract”.Footnote ⁹⁹ The Brussels case law, which now endorses concurrent contract and tort claims subject to Wikingerhof's indispensability principle, should apply to the Rome Regulations as well.

As with the common law, EU Regulations also contain limits on concurrent claims. But again, these tend to be tangential limits based on separate concerns, which thus only qualify concurrent claims in select situations. The first of these, Wikingerhof's indispensability principle, is already discussed. The second, the use of jurisdiction and choice-of-law agreements, is similar to the position under common law. The positive rules are admittedly different, at least for the second question of enforceability rather than the first question of scope, the latter being a question of national law allocated to the governing law of the relevant jurisdiction or choice-of-law agreement.Footnote ¹⁰⁰ Brussels Ia is stricter about the enforceability of jurisdiction agreements for non-contractual claims than the common law: Article 25(1) requires those claims to be “connected” with or “proximate” to parties’ legal relationship,Footnote ¹⁰¹ though this is unlikely to bar enforcement of any agreements covering concurrent contract and tort claims,Footnote ¹⁰² since those tort claims are likely to be considered proximate to the relevant contracts. Conversely, for choice-of-law agreements, Rome II is more permissive than the common law: under Article 14(1)(b), parties “pursuing a commercial activity” can agree to subject tort claims to a chosen law if that agreement is “freely negotiated”, expressed with “reasonable certainty” and does “not prejudice the rights of third parties”. This, however, is still less permissive than Brussels Ia Article 25(1): as Carr J. noted in Pan Oceanic Chartering Inc. v UNIPEC UK Co. Ltd., Article 14(1)(b) demands that defendants must have had a “genuine opportunity to influence [the agreement's] content … for [their] own benefit”.Footnote ¹⁰³ On the whole, however, under EU Regulations the same conclusion reached under common law can also be drawn: jurisdiction and choice-of-law agreements at best place only tangential direct limits on the effects of concurrent claims, since they involve different principled concerns from and thus do not comprehensively preclude the latter.

The third limit on the effects of concurrent claims is the indirect limit imposed by the exception to the default tort choice-of-law rule, Rome II Article 4(3). Significantly, the only enquiry under Article 4(3) is whether a “clear preponderance of factors” connects the tort to a law other than the lex loci damni; there is no additional requirement that the lex loci damni's connection with the tort be “fortuitous” or lack genuineness.Footnote ¹⁰⁴ Many commentators have thus suggested that Article 4(3) eliminates the effects of concurrent claims: a tort claim brought on the same facts as a claim under a contract between parties “closely connected” with it will be governed by the latter's law.Footnote ¹⁰⁵

However, it is unclear whether under Article 4(3) the fact that a closely connected contract between parties is governed by a law other than the law of the place of damage conclusively evinces a manifestly closer connection with that law, or whether, as with PILA s. 12, that fact is only one factor (albeit a potentially weighty one) that suggests a manifestly closer connection with that law. Importantly, while commentators have generally assumed the former,Footnote ¹⁰⁶ the case law actually suggests the latter: as Miles J. noted in Lyle and Scott Ltd. v American Eagle Outfitters,Footnote ¹⁰⁷ “the overall question under Article 4(3) is not merely whether the tort claim is connected with the contract … [but] whether the tort is manifestly more closely connected with the country other than that indicated in Article 4(1)”.Footnote ¹⁰⁸ This, as Linden J. emphasised in Owen v Galgey,Footnote ¹⁰⁹ means that the connected contractual relationship is “but an example of the circumstances which may bear on identifying the country with the most manifest connection”; courts must still consider “all the circumstances of the case”.Footnote ¹¹⁰ Significantly, the lex loci damni was not displaced in those decisions.Footnote ¹¹¹ In Lyle and Scott, the English lex loci damni of a passing off claim was not displaced by Pennsylvanian law, which governed the parties’ agreement on how their goods should be branded and marketed.Footnote ¹¹² In Owen, the French lex loci damni of an occupier's liability claim could not be displaced by the English governing law of the agreement for the claimant to work the property.Footnote ¹¹³

Moreover, while courts have occasionally invoked Article 4(3) to replace the lex loci damni with the law governing a closely related contract between parties, they have never rested their reasoning on that fact alone. Instead, as Picken J. observed in Avonwick Holdings Ltd. v Azitio Holdings Ltd.,Footnote ¹¹⁴ in those cases courts reason that a significant majority of all factors (including other territorial connecting factors like the location where the “alleged wrongdoing was planned, orchestrated and implemented”; where “direct or indirect” damage was suffered; and where assets “at the heart of the alleged wrongdoing” wereFootnote ¹¹⁵) also point to that law which governs the parties’ contract.Footnote ¹¹⁶

Thus, Article 4(3) requires a “clear preponderance of factors” that connect the tort to a law other than the lex loci damni, and the fact that parties’ closely connected contract is governed by that law is only one (albeit potentially weighty) factor that points towards the other law, to be considered and weighed alongside other factors. So understood, Article 4(3) is merely a tangential indirect limit on the effects of recognising concurrent claims similar to PILA s. 12: the governing law of a concurrent contract claim is only one factor to consider alongside others when invoking the exception.

A. Principled Classification and Concurrent Liability in Domestic Law

Private law has a well-established tradition of classifying legal obligations on the basis of principle. Aristotle, for instance, adopted a division between voluntary and involuntary transactions which resembles the contemporary division between contract and tort law.Footnote ¹¹⁷ Kant likewise developed a taxonomy which divides private law into contracts, property and fiduciary obligations.Footnote ¹¹⁸ Both accounts proceed from first principles: voluntariness for Aristotle, rights for Kant.Footnote ¹¹⁹ Legal philosophy has since consistently recognised the significance of classification to private law:Footnote ¹²⁰ classification helps the law maintain internal coherence and enhances the values of predictability and certainty.Footnote ¹²¹ As Burrows notes, the classification of legal obligations is fundamental to “principled reasoning” and the “essential requirement of the rule of law that like cases should be treated alike”.Footnote ¹²²

Of course, no consensus has been reached, even within single legal traditions, on how private law classification should be carried out.Footnote ¹²³ However, regardless of how a legal system defines its categories of contract and tort, overlaps between categories on particular facts does not threaten, but is instead the inevitable and acceptable consequence of, a principled classification of private law. This argument holds regardless of the theory of private law one adopts. It flows naturally from a pluralist theory of private law, where different legal categories exist to vindicate different social values with different normative underpinnings. As Hanoch Dagan notes, to a pluralist “some overlaps are perfectly acceptable, or even desirable” since “life is messy and … different contexts, while distinct in some senses, often raise overlapping normative concerns”.Footnote ¹²⁴ Overlaps between legal categories on particular facts are also consistent with a monist theory of private law. There, as Jacob Weinrib argues, the legal classificatory exercise performs three functions:Footnote ¹²⁵ it posits a unified and coherent normative basis of private law, divides that unified basis into particular legal categories according to its internal logic, and finally orders specific facts and events into each of those legal categories. To monists, then, an overlap between legal categories does not betray a lack of unified normative bases: “an adequate legal classification must articulate the conceptual interrelation between, for example, property and obligation”, rather than assiduously assign “each legal issue to one concept alone”.Footnote ¹²⁶ Thus, as Paul Davies acknowledges, “[n]o matter what taxonomy is adopted when dividing the law up into discrete areas for exposition, it seems inevitable that there will be disputes which can be analysed as falling within one or more taxonomies”.Footnote ¹²⁷

The above discussion may be summed up in two propositions: first, that the principled classification of claims into legal categories is inherent to the very idea of private law adjudication; and second, that such legal categories may legitimately overlap on particular facts if demanded by their underlying principles. These two propositions undergirded Lord Goff's endorsement of concurrent liability in Henderson, which rested on the identification of different contract and tort duties with different normative justifications.Footnote ¹²⁸ The important point there, however, is not so much the identification of duties but the basis for doing so – even if concurrent liability is a manner of concurrent “obligations”, “claims” or even “rights”, what matters is that the concurrence of these legal concepts necessarily rests on principle rather than pragmatism. For once overlapping legal concepts are identified and separately justified, the general availability of concurrent liability is inevitable. So well-established is this proposition in domestic private law that the debate among courts and commentators on concurrent liability no longer centres around its existence in general, but instead focuses on the appropriate limits that should be placed thereon. As mentioned, however, these limits on concurrent liability have heretofore been narrow and tangential: they do not effectively eliminate, but in fact implicitly support the existence of, concurrent liability.Footnote ¹²⁹

B. Principled Characterisation and Concurrent Claims in Private International Law

Characterisation is as central to private international law as it is to domestic private law, being the prerequisite to the application of its jurisdictional and choice-of-law rules. However, characterisation in private international law cannot use the substantive private law taxonomy of any one system of domestic private law, because private international law's function is to allocate disputes to various different legal systems on an even-handed basis. Thus, if private international law chooses a given legal system's legal taxonomy (say, the forum's), it will inevitably be faced with the problem of having to justify that choice in the first place. Moreover, it is impossible to maintain that characterisation can be performed without any reference to law,Footnote ¹³⁰ since characterisation for legal processes is an inherently legal exercise.Footnote ¹³¹ A solution to this chicken-and-egg problem requires that characterisation in private international law be performed by reference to an external, international, “autonomous” legal taxonomy of private law.

The crucial question, then, is: how is this international legal taxonomy of private law to be constructed? The typical answer to this question is given in the form of a negative proposition: that characterisation should not take place in accordance with the strict legal taxonomy of any one state, especially the forum. But this is unhelpful, because it only begs the positive question: how should courts construct the applicable international taxonomy instead?

One might suggest that courts must construct that taxonomy using a different methodology from that which they use in domestic private law: while domestic private law looks to substantive theories of private justice to approach classification in a principled manner, private international law should be agnostic about theories of private justice and approach classification in a pragmatic and ad hoc manner. Thereunder, courts would look at comparative law datum on how the various legal systems of the world, or at least those systems pleaded as applicable, have characterised claims similar to those before them, and determine whether parties’ claims are “contractual” or “tortious” on that basis. The pioneer of this approach was Ernst Rabel, who argued that the interpretation of choice-of-law rules required courts to rely on comparative law.Footnote ¹³² Shades of this pragmatic comparative approach are also evident in John Falconbridge's “via media” theory of characterisation, which posits that courts should first “engage in a process of provisional or tentative characterisation … informed [by] the content of each of the two laws which may be the proper law”;Footnote ¹³³ and Otto Kahn-Freund's “enlightened lex fori” theory of characterisation, which refers courts to a “system of internationally accepted analytical concepts”.Footnote ¹³⁴

However, the problem with this pragmatic comparative law approach to characterisation is that it attempts to convert an “is” into an “ought”, and thus cannot justify how private international law should operate as a field of law. As Mathias Reimann notes, although comparative law and private international law share an “intimate relationship”, the two fields share very different goals: comparative law seeks to determine law as fact, but private international law must justify what law should do in practice.Footnote ¹³⁵ If private international law abdicated its function to comparative law, it would surrender law's essential normative function of ordering legal relationships to achieve justice.

Thus, private international law must adopt the same methodology in characterisation as domestic private law does: it must ask itself how, in principle, private disputes should be characterised for the purposes of jurisdiction and choice-of-law. Of course, the principled taxonomy private international law constructs cannot be modelled on any one domestic law taxonomy – but it remains the case that private international law must still ascribe to some theory of private law in constructing that taxonomy. This proposition, that characterisation in private international law remains a principled affair, is as old as the discipline itself: Savigny, for instance, believed that courts should characterise claims according to the nature of parties’ “legal relations”, which would reflect various substantive ideas of private interpersonal justice.Footnote ¹³⁶ It also has support from recent high common law authorities: in Raiffeisen Zentralbank v Five Star Trading LLC, Mance L.J. noted that characterisation in private international law remains “a search for appropriate principles to meet particular situations”, using legal categories which are defined and re-defined in line with “the purpose for which they were conceived”.Footnote ¹³⁷ This search must be undertaken in a “broad internationalist spirit”, but it remains a search for “appropriate” principles rather than mere practice.Footnote ¹³⁸ More recently, in Solomon Lew v Kaikhushru Shiavax Nargolwala,Footnote ¹³⁹ Lord Mance, now sitting as an International Judge on Singapore's International Commercial Court, re-emphasised that private international law as a whole demanded “a principled, not simply a pragmatic, approach”, under which “[t]he law is shaped to serve and give effect to the reasonable expectations of its users”.Footnote ¹⁴⁰ This principled approach is also endorsed by the authors of Dicey, Morris & Collins, who see characterisation as requiring courts to “consider the rationale of the English conflict rule and the purpose of the rule of substantive law to be characterised”, and “[o]n this basis … decide whether the conflict rule should be regarded as covering the rule of substantive law”.Footnote ¹⁴¹

Characterisation in private international law, just as in domestic private law, thus requires courts to ask: what normative concern(s) are implicated by the facts of parties’ dispute, and which legal category (or categories) reflects them? This has tremendous ramifications for the recognition of concurrent claims in private international law: if the principled classification of disputes and issues into legal categories is central to private law, and if in domestic law such categories may legitimately overlap when doing so would vindicate their underlying principles, it is hard to see why the same conclusion should not be reached in private international law. In both situations, the only pre-requisite for recognising concurrent liability/claims should be the same: that there are principled justifications for so defining the breadth of the legal categories that happen to overlap on particular facts.

Thus, while Briggs is right that concurrent claims should not exist in private international law disputes when there is “only one duty which has been broken, even though domestic law has a variety of ways of conceptualising it”, he is too hasty in concluding that this is “often” the case, and that therefore concurrent claims should not be recognised in general.Footnote ¹⁴² Instead, the true position is that, in private international law just as in domestic private law, the question of the existence of multiple duties is a separate and logically anterior question of private law theory. As the Singapore Court of Appeal noted in Rickshaw, a claimant's tort claim need not necessarily have the same principled basis as a concurrent contract claim, because they may have “separate legal existence[s]”.Footnote ¹⁴³ Likewise, once the CJEU defined “tort” obligations in Wikingerhof, as “obligation[s] imposed by law”,Footnote ¹⁴⁴ and thus a category of legal obligations with independent normative underpinnings,Footnote ¹⁴⁵ this led inexorably to its confirmation that concurrent tort and contract claims were generally permissible.Footnote ¹⁴⁶ We should emphasise, however, that our argument is not that an identifiable international substantive theory of private law exists, either in EU law or across different national legal systems, under which certain fact patterns should give rise to concurrent claims in contract and tort; we take no position on this matter. Our argument is simply that concurrent claims should be recognised whenever, in light of whatever theory of private law the forum's courts adopt for international disputes, there are indeed two separate principled bases for imposing liability on the defendant. Accordingly, the court's task is nothing more and nothing less than to determine, according to its international legal taxonomy for private international law disputes, whether such separate bases exist on the facts.

C. The Limits on Concurrent Claims in Private International Law

Of course, just as in domestic private law, the fact that concurrent claims in private international law should exist does not mean that there should be no limits on their recognition and enforcement. But the conclusions reached above allow us to go further, in explaining how the current limits on the recognition of concurrent claims can be justified. Properly understood, the only real limit on concurrent claims in private international law (i.e. the only reason why courts should subordinate one claim to another despite there being separate principled grounds for both on the facts) is the uncontentious proposition that parties may, via a genuine expression of intent, agree to exclude tort jurisdictional or choice-of-law rules. Every other ostensible limit on the recognition of concurrent claims may be understood as simple reiterations of the pre-requisite for recognising concurrent claims: that both claims exist on the basis of independent normative concerns implicated on a single set of facts.

The first limitation, the “bad faith” principle recognised in Rickshaw, should be uncontroversial, because it is private international law's analogue of Burrows’ “exclusion principle”, that parties can use express exclusion clauses to limit concurrent liability. As mentioned,Footnote ¹⁴⁷ today both the common law and EU Regulations allow parties to subject tort claims to the contractual forum under jurisdiction agreements, and, to a lesser extent, to contractual governing laws under choice-of-law agreements. Commentators also recognise that, at least when the tort claim arises from parties’ commercial dealings – which will always be the case when it arises concurrently with a contract claim – there is no real reason why such agreements should not be enforceable.Footnote ¹⁴⁸ The only concern here is to protect weaker or third parties, which is met by requirements like those contained in Brussels Ia Article 25 and Rome II Article 14(1)(b), which ensure that agreements reflect a genuine consensus between equally-situated commercial parties. But this limit on the effects of concurrent claims is uncontroversial, because it turns on the intuitively-acceptable principle that parties can, via a genuine display of consent, voluntarily agree to relinquish their rights. And importantly, this recognition of concurrent claims would be the logical pre-requisite of this limit, because it is only by first recognising concurrent claims in contract and tort that it becomes meaningful for courts to subsequently ask whether parties have genuinely agreed to exclude tort claims. Enforcing jurisdiction and choice-of-law agreements thus cannot be said to threaten concurrent claims in private international law, any more than enforcing exclusion clauses threatens concurrent liability in domestic private law.

The second limitation, the “indispensability” principle recognised in Wikingerhof, is more controversial. Since what must be “indispensable” to the tort claim is a need to “examine the content of the contract”,Footnote ¹⁴⁹ the idea seems to be that if a sufficiently-particularised tort claim requires the claimant to plead and prove a contractual provision's content, the latter should prevail over the former.Footnote ¹⁵⁰ However, it is unclear why a tort claim requiring proof of a given contract provision should be excluded in favour of the contract claim – after all, here there could still be two different normative grounds for imposing two different heads of liability. Conversely, it is unclear why the very fact that a contract provision need not be interpreted to determine the scope of tort liability should allow the latter to be asserted independently of a concurrent contract claim – just because one claim's formal provenance differs from another's does not mean the two are normatively independent. At face value, then, Wikingerhof's indispensability principle seems both over-broad and under-inclusive, which renders its application somewhat arbitrary.Footnote ¹⁵¹ This is probably unsurprising; like the doctrine of illegality's traditional “reliance” rule, rules which apply depending on whether the claimant has pleaded a particular fact tend in many situations to be too rigid to accommodate the concerns said to justify them.Footnote ¹⁵²

A better (re)interpretation of Wikingerhof's indispensability principle, then, is one that ignores the technicalities of such formal reliance and instead goes straight to the CJEU's underlying concern in enunciating it:Footnote ¹⁵³ the need to ensure that tort claims exist “independently of [a] contract”.Footnote ¹⁵⁴ This would then appear to align Wikingerhof's indispensability principle with Burrows’ “independence principle”, that only a tort claim which “in essence … is one to enforce a contract” should be precluded.Footnote ¹⁵⁵ But if this is so, Wikingerhof's indispensability principle may be no limit on recognising concurrent claims at all. In illustrating how the independence principle operates, Burrows noted that concurrent liability should only arise when “the thrust of the plaintiff's claim[s]” are different:Footnote ¹⁵⁶ if you hire a workman to clean your chandelier but he destroys it instead, you may sue him in negligence rather than breach of contract because you are aggrieved about the interference with property rather than the failure to perform.Footnote ¹⁵⁷ But if all the independence/indispensability principle requires is that there be different normative grounds for a claimant's separate complaints, then it is not really a limit on recognising concurrent claims, but a mere restatement of its essential pre-requisite. The only claims precluded by Wikingerhof's “indispensability” principle, then, are tort claims founded on, and only on, the breach of a contractual term between parties – and there, there is no separate basis for liability other than the breach of the agreement, and thus no basis for recognising concurrent claims at all.

The third limitation on the effects of recognising concurrent claims, namely tort choice-of-law exceptions, is perhaps the most controversial – if concurrent claims should be recognised when separate principled bases exist for each, how can a rule that may potentially collapse both claims into the contract claim be justified? As mentioned, these exceptions are triggered either by the fact that the default applicable law under the default rule was “fortuitous” (common law) or overshadowed by a “preponderance” of factors pointing to another law, including but not limited to the fact that a closely related contract is governed by that law (PILA, s. 12 and Rome II, Article 4(3)). Properly understood, then, neither of these exceptions are truly designed to preclude the effects of concurrent claims. Instead, they simply ensure that tort choice-of-law rules do not favour one territorial factor (the location of the act complained of or the location where the damage is suffered) over other territorial or non-territorial factors which are more significant given the nature of the tort alleged.Footnote ¹⁵⁸ Thus, the common law's flexible exception operates only when the connection between the tort and the initial applicable law is insignificant in absolute terms, because that connection is “fortuitous”. Similarly, PILA, s. 12 and Rome II, Article 4(3) demand that the connection between the tort and the default applicable law be overshadowed by the connections that a “clear preponderance” of other factors have with the tort; this logically imposes a requirement that the first connection be insignificant, and the latter connections be very significant, in relative terms.Footnote ¹⁵⁹

Thus, while differences exist between them, tort choice-of-law exceptions all share the same function of ensuring that torts are significantly connected with their applicable laws. Importantly, we can also see this significant connection requirement not as a limit on recognising concurrent claims, but as another reflection of the essential pre-requisite for concurrent claims: that a tort claim reflects genuine normative concerns independent of those which support a contract claim. After all, in identifying connecting factors, just as in the prior act of characterising issues, private international law seeks to give effect to the various substantive private justice ideals reflecting parties’ reasonable expectations.Footnote ¹⁶⁰ Default tort connecting factors thus reflect value judgments about the law most closely connected to a tort claim, assuming that the claim implicates certain normative substantive justice concerns (that, given the kind of wrongdoing alleged, the law of the territory where it occurs/causes damage should govern).Footnote ¹⁶¹ However, when the claim does not substantially raise those concerns, the default connecting factors are deemed insignificant and set aside. In their place, tort choice-of-law exceptions invoke alternative connecting factors to establish a connection between the tort and the new applicable law. And these alternative connecting factors, like the default connecting factor, are also selected on the assumption that the tort claim implicates certain other normative substantive justice concerns (e.g. that, given the kind of wrongdoing alleged, the governing law should be that of which various territorial and non-territorial connecting factors leans significantly in favour).

The selection of (default or alternative) connecting factors under (default or alternative) tort choice-of-law rules is therefore a principled affair much like the characterisation of issues: courts recognise connecting factors as relevant, and others as irrelevant, on the basis of the normative concern(s) implicated by the facts of parties’ dispute. It follows that the only situation in which tort choice-of-law exceptions could choose the governing law of a closely related contract as the sole connecting factor is when the facts implicate only one normative concern: party autonomy. If other normative concerns apply, the exceptions would dilute that factor with others.Footnote ¹⁶² Only in that situation are the effects of recognising concurrent claims eliminated, because only then is the law governing parties’ closely related contract selected to govern the tort claim solely because it is the law governing parties’ closely related contract. However, that result in that situation is uncontroversial, because the essential pre-requisite of recognising such claims (i.e. that a claimant's separate claims be grounded on different normative concerns) is absent. Thus, properly understood, tort choice-of-law exceptions, like Wikingerhof's indispensability principle, are not really limits on the recognition of concurrent claims in private international law; the only situation in which they would have that effect is when the essential pre-requisite for recognising such claims does not exist.

D. Defending Concurrent Claims

In sum, courts should recognise concurrent claims in private international law as they do concurrent liability in domestic private law, because, in both, principled legal categories are necessary, and their occasional overlap is desirable to vindicate their underlying principles. This means that whenever a dispute implicates separate normative concerns in accordance with the forum's courts substantive theory of private (international) law, those courts should recognise concurrent claims in contract and tort. The only justifiable limit on concurrent claims is the parties’ intent that jurisdiction or choice-of-law clauses in a related contract should cover tort claims as well; every other ostensible limit simply re-iterates the essential pre-requisite for recognising concurrent claims, that tort claims be based on normative concerns separate from those which support contract claims.

But as noted in Section II, the opposition to concurrent claims in private international law does not stop here – commentators further argue that recognising concurrent claims would offend values and goals specific to private international law: general goals like preventing disproportionate defendant disadvantage; choice-of-law values like international uniformity; and jurisdictional values like predictability and anti-fragmentation. The argument here is that even if the existence of separate normative concerns could by itself justify concurrent claims, the need to uphold these values defeats that initial justification. Yet, properly understood, the need to uphold private international law's values requires no such thing.

First, recognising concurrent claims would not disproportionately disadvantage defendants. In the choice-of-law context, Briggs's expressed fear, that recognising concurrent claims would compromise “equality of arms” because it would “allow … the claimant, more or less unilaterally, to choose the law which will govern the substance of the dispute in a way most favourable to him”,Footnote ¹⁶³ is exaggerated. It is admittedly true that recognising concurrent claims gives claimants the power to choose between claims, and thus choice-of-law rules. But, as Briggs himself recognises, private international law's concern with equality is to prevent the claimant from unilaterally choosing the applicable law.Footnote ¹⁶⁴ The availability of concurrent claims would not grant claimants this power, since the available claims and choice-of-law rules are still dependent on private international law's judgment of which legal categories, formulated on principled grounds by the law of the forum, actually relate to the facts of parties’ dispute. To use Briggs’ terminology,Footnote ¹⁶⁵ while the claimant does have a “menu” of choice-of-law rules available, that “menu” is itself catered for by the forum's legal system, on the basis of the normative concerns implicated by the facts of parties’ dispute.

The same point can be made in the jurisdictional context, where recognising concurrent claims also would not give claimants an unequal and unilateral choice of forum. Private international law's jurisdictional rules represent normative connections between legal categories and various judicial fora.Footnote ¹⁶⁶ Thus, just as claimants may justifiably exercise free choice within the closed universe of choice-of-law rules corresponding to the legal categories to which parties’ dispute can be allocated, so too may they justifiably exercise free choice within the closed universe of jurisdictional rules corresponding to the same. Of course, a jurisdiction clause may limit these choices, but that, as mentioned, is an uncontroversial party-choice-based limit on the effects of recognising concurrent claims, not a denial of such recognition in general.Footnote ¹⁶⁷

Second, recognising concurrent claims would not threaten choice-of-law specific values like international uniformity. For starters, it is not clear why choice-of-law should seek to achieve international uniformity by denying concurrent claims. This is because choice-of-law doctrine does not seek uniformity for uniformity's sake, because the identity of the grounds on which legal systems converge is as important as the fact that they do converge; as Alex Mills notes, “it is not enough to avoid potential inconsistent legal treatment by selecting any single source of regulatory authority – it matters which source is chosen”.Footnote ¹⁶⁸ Since uniformity must be achieved not at the expense of, but in tandem with, substantive justice,Footnote ¹⁶⁹ we should reject an approach to choice-of-law which denies concurrent claims even when there are principled reasons for both claims to be recognised on the facts, because that approach achieves uniformity only by unjustifiably sacrificing principle. But even if international uniformity is choice-of-law's paramount goal, recognising concurrent claims would not at all frustrate it. Of course, the existence of a single non-overlapping taxonomy of private law categories – accepting for the moment that utopia is attainable – would certainly ensure uniformity and decisional harmony.Footnote ¹⁷⁰ Yet, properly understood, it is the fact that a single taxonomy is adopted, rather than the fact that the taxonomy's categories do not overlap on particular facts, which is essential for uniformity. If one single taxonomy exists, a dispute which gives rise to concurrent claims will never be subject to inconsistent treatment: the claimant may have a choice between claims, but the breadth of that choice would be uniform no matter which court she proceeds in.

Third, recognising concurrent claims also would not threaten jurisdictional values like predictability and anti-fragmentation. On anti-fragmentation first, it is admittedly true that allocating different claims arising out of one factual matrix to different courts would fragment proceedings. However, to argue that concurrent claims should not be recognised so as to prevent such fragmentation is unhelpful, because this does not tell courts how they should go about denying concurrent claims to achieve that end, in particular which of the concurrent claims should be subordinated to the other. This puts courts in a bind: after all, if concurrent claims in contract and tort are justified by the existence of two equal and separate normative concerns arising on a single factual matrix, neither concern can justifiably be subordinated. Instead, then, the problem of fragmentation requires a claim-neutral practical solution which does not choose between (and thus does not downplay) the normative concerns that justify recognising concurrent claims in the abstract. In the common law, the relevant claim-neutral method is forum non conveniens, a dominant enquiry in both disputes involving jurisdiction as of right and discretionary jurisdiction, which allows courts to stay proceedings in favour of any foreign court which they consider better placed to adjudicate upon facts underlying multiple claims. Under Brussels Ia, the relevant claim-neutral method involves rules like Articles 29 and 30, on the priority that the court first legitimately seised of jurisdiction enjoys over others, that determine which of several courts with subject-matter connections to the parties’ dispute should have jurisdiction. It is beyond this article's scope to discuss whether the common law's or Brussels Ia's claim-neutral method is superior; the only point made is that any claim-neutral method is a more appropriate solution to the problem of fragmentation than a denial of concurrent claims.

The need to ensure predictability for the defendant, which justifies the jurisdictional rules for matters relating to contract and tort in Brussels Ia being applied in a “mutually exclusive” manner,Footnote ¹⁷¹ also should not preclude claimants from choosing between contract and tort claims with independent normative underpinnings leading to different jurisdictional consequences. As Samuel Zogg notes, Articles 7(1)(a) and 7(2) may refer to one of two potential “objects of mutual exclusivity”: is the “matter” which relates to contract or tort the “entire dispute” between parties, comprised of “all causes of action contained in the proceedings/dispute”; or is it merely a “single cause of action/claim”?Footnote ¹⁷² The answer, as the CJEU recognised in Kalfelis, is the latter: the object of mutual exclusivity is merely a portion of the claimant's “action”, so “different aspects of the same dispute” could be “adjudicated upon by different courts”.Footnote ¹⁷³ The CJEU made the same point in Wikingerhof: it is the applicant's “claims” which the court must classify as “matters” relating to a contract or tort;Footnote ¹⁷⁴ the essential task is thus “an exercise in characterisation of the obligation … underlying the claim”.Footnote ¹⁷⁵

Thus, under the Brussels regime, the claimant's individual claims – more precisely, the obligations on which they rely – are the “matters” that must relate exclusively either to contract or tort. And this does not preclude, but rather encourages, the recognition of concurrent claims: if two obligations in contract and tort exist on the same set of facts, claimants may invoke either of the relevant jurisdictional rules. But what of the broader concern that this undermines predictability – should the defendant not know precisely the single forum it will be sued in as soon as the dispute arises?

The answer is no, for three reasons. First, as the AG's opinion in Wikingerhof states, if this is what “predictability” requires, the Brussels regime has never been able to achieve it, because it has always given the claimant the freedom to choose between suing in the state selected by its special jurisdictional rules or suing in the state of the defendant's domicile.Footnote ¹⁷⁶ Second, “predictability” under Brussels Ia in truth requires not that defendants know, but only that they be able to “reasonably foresee”, where they would be sued.Footnote ¹⁷⁷ Since concurrent claims exist only when a single dispute implicates separate normative concerns reflective of parties reasonable expectations, the courts seised under Articles 7(1)(a) and 7(2) should both be within the defendant's reasonable contemplation. Third and finally, if “predictability” did indeed require such knowledge rather than reasonable foreseeability, it would be harmed rather than furthered if courts took entire disputes rather than causes of action as the object of mutual exclusivity. After all, if courts must choose between two normative concerns raised on a single set of facts, they need a test to do so; and as the CJEU's experience since Brogsitter demonstrates, such tests either fail to produce certain resultsFootnote ¹⁷⁸ or collapse into a mere restatement of the essential pre-requisite of recognising concurrent claims.Footnote ¹⁷⁹ By contrast, while the recognition of concurrent claims would theoretically leave defendants unable to know which of the two jurisdictional rules claimants will pick from the outset, they will know that claimants are free to choose that which best suits their interests; and this, paradoxically, creates certainty in practice, since claimants will always choose the jurisdiction available to them which best suits their interests.Footnote ¹⁸⁰ Thus, the recognition of concurrent claims in contract and tort under the Brussels regime is no threat to the value of predictability, and surprisingly enough is the best way to ensure that defendants will be certain of the jurisdiction they will be sued in.