OUTER HOUSE, COURT OF SESSION
|
|
PD903/11
|
OPINION OF LORD UIST
in the cause
MARY ADAMSON LORIMER STEWART AND OTHERS
Pursuers;
against
(FIRST) TRAFALGAR HOUSE STEAMSHIP COMPANY LIMITED and BRUCE ALEXANDER MACKAY and DAVID PAUL HUDSON the Liquidators thereof; and (SECOND) BRITISH INDIA STEAM NAVIGATION COMPANY LIMITED
Defenders:
________________
|
Pursuers: A C Forsyth; HBM Sayers (for Charles Hennessy & Co, Solicitors, Glasgow)
Defenders: N McKenzie; Brodies LLP
5 March 2013
The issue
[1] X
(consisting of three pursuers) bring an action of damages in the Court of
Session in respect of a death caused by mesothelioma against three limited companies,
A, B and C. A is domiciled in Scotland. B and C are domiciled in England. Jurisdiction
is founded on the domicile of A and the fact that the claims against A, B and C
are so closely connected that it is expedient to hear and determine them together
to avoid the risk of irreconcilable judgments arising from separate proceedings
(Regulation (EC) 44/2001, Article 6.1 and the Civil Jurisdiction and
Judgments Act 1982, schedule 4, para 5). X subsequently abandon
their claim against A and wish to continue the action against B and C. Does
the court in these circumstances have jurisdiction in the action against B and
C? That is the question which arises for determination in this case.
Procedural history
[2] The
three pursuers raised the present action initially against a company known as D
P & L Group Limited (D P & L) as first defenders and the two present
defenders jointly and severally on 21 April 2011. The pursuers are
members of the family of the late James Mailer Stewart, who died of mesothelioma
on 25 April 2008. They averred that he contracted mesothelioma as a
result of negligent exposure to asbestos while in the employment of each of the
three defenders in the merchant navy. They subsequently discovered that D P
& L were not the correct company to sue and that the company which should
have been sued had been dissolved in 1989 and not restored to the
Companies Register. They therefore abandoned their claim against D P & L,
who were assoilzied by interlocutor of 11 January 2012. The remaining
defenders have now taken a plea to the jurisdiction of the court.
The relevant
statutory provisions
[3] Council
Regulation (EC) 44/2001 of 22 December 2000 on jurisdiction and the
recognition and enforcement of judgments in civil and commercial matters ("the
Regulation") provides in article 2.1 that, subject to the Regulation,
persons domiciled in a Member State shall, whatever their nationality, be sued
in the courts of that Member State. Article 6.1 provides that a person
domiciled in a Member State may also be sued, where he is one of a number of
defendants, in the courts for the place where any one of them is domiciled,
provided the claims are so closely connected that it is expedient to hear and
determine them together to avoid the risk of irreconcilable judgments resulting
from separate proceedings. The Regulation applies in civil and commercial
matters whatever the nature of the court or tribunal (article 1.1) and is
binding and directly applicable (recital 6).
[4] Part II
of the Civil Jurisdiction and Judgments Act 1982 deals with jurisdiction
(and recognition and enforcement of judgments) within the United Kingdom. Section 16,
which deals with allocation within the United Kingdom in certain civil
proceedings, provides:
"(1) The provisions set out in Schedule 4 (which contains a modified version of Chapter II of the Regulation) shall have effect for determining, for each part of the United Kingdom, whether the courts of law of that part, or any particular court of law in that part, have or has jurisdiction in proceedings where -
(a) the subject-matter of the proceedings is within the scope of the Regulation as determined by article 1 of the Regulation (whether or not the Regulation has effect in relation to the proceedings); and
(b) the defendant or defender is domiciled in the United Kingdom or the proceedings are of a kind mentioned in article 22 of the Regulation (exclusive jurisdiction regardless of domicile)."
Section 20 deals with the rules as to jurisdiction in Scotland and provides:
"(1) Subject to the Regulation, to Parts I and II and to the following provisions of this Part, Schedule 8 has effect to determine in what circumstances a person may be sued in civil proceedings in the Court of Session or a sheriff court."
Schedule 4 to the Act provides rules for the allocation of jurisdiction within the United Kingdom. Para 5(a) provides:
"A person domiciled in a part of the United Kingdom may, in another part of the United Kingdom, also be sued where he is one of a number of defendants, in the courts for the place where any one of them is domiciled, provided the claims are so closely connected that it is expedient to hear and determine them together to avoid the risk of irreconcilable judgments resulting from separate proceedings."
Para 2(o)(i) of Schedule 8 repeats the substance of para 5(a) of Schedule 4.
Submission for the
defenders
[5] The
broad proposition for the defenders was that the pursuers, in order to bring
themselves within article 6.1 of the Regulation, had to satisfy the
requirement that the claims were so closely connected that it was expedient to
hear and determine them together to avoid the risk of irreconcilable judgments
resulting from separate proceedings. Article 6.1 could not apply where
the claim against the purported "anchor defenders" (D P & L) was an
impossible one. The claim against D P & L was not closely connected to the
claim against the defenders because (even on the pursuers' own pleadings) the
former was impossible, having been directed against the wrong party. There was
a fundamental distinction between, on the one hand, the court being seised of
an action and, on the other hand, the court having jurisdiction to deal with
the case. D P & L had been sued in error. The company which should have
been sued no longer existed when the action was raised, so there was no actual
risk of irreconcilable judgments. As D P & L had never employed the
deceased recovery against them was impossible. It was a fundamental
misconception to equate the court being seised of an action with the court
having jurisdiction to hear the action. The provisions of the Regulation had
to be interpreted independently by reference to their scheme and purpose: the
general principle was that it was the courts of the Member State in which a
defender was domiciled which had jurisdiction over him. It was only by way of
derogation from that principle that the Regulation provided special rules of
jurisdiction, which were exhaustively listed. These special rules required to
be strictly interpreted, and it was for the national courts to interpret them
having regard to the principle of legal certainty and so that a normally
well-informed defender could reasonably foresee in which courts he might be
sued. While bad faith was not an issue in the present case, it was for the
court to examine whether Article 6.1 was being misused. It was accepted
that the application of Article 6.1, and thus para 5(a) of
Schedule 4 to the 1982 Act, was not dependent on domestic rules. It
was also accepted that jurisdiction did not depend on a material examination of
the substance of the case. The risk to be avoided under Article 6.1 was
that of irreconcilable judgments, which would not promote legal certainty. Nonetheless,
the court required to be satisfied that the risk of irreconcilable judgments
was actually present when the action was raised or subsequently. Taking a
broad common sense approach in this case, there had never been any risk of
irreconcilable judgments where D P & L had been sued in error, the correct
party which should have been sued in their place no longer existed and the only
two defenders in existence were domiciled in England. In these circumstances
there was no basis for an exception to the general rule that jurisdiction was
based on the defenders' domicile, which resulted in only the English courts
having jurisdiction against the two defenders in this case.
[6] The
submission for the defenders then turned to examine the reported cases bearing
on the issue.
[7] In Kalfelis
v Bankhaus Schroder [1988] ECR 5565 the Fifth Chamber of the
European Court of Justice (ECJ) held that for Article 6.1 to apply a
connection must exist between the various actions brought by the same plaintiff
against different defendants and that that connection, whose nature had to be
determined independently, had to be of such a kind that that it was expedient
to hear the actions together in order to avoid the risk of irreconcilable
judgments resulting from separate proceedings. In its judgment the court
stated as follows:
"8. The principle laid down in the Convention is that jurisdiction is vested in the courts of the State of the defendant's domicile and that the jurisdiction provided for in article 6.1 is an exception to that principle. It follows that an exception of that kind must be treated in such a manner that there is no possibility of the very existence of that principle being called in question.
12. The rule laid down in article 6.1 therefore applies where the actions brought against the various defendants are related when the proceedings are instituted, that is to say, where it is expedient to hear and determine them together in order to avoid the risk of irreconcilable judgments resulting from separate proceedings. It is for the national court to verify in each individual case whether that condition is satisfied.
13. It must therefore be stated ... that for article 6.1 of the Regulation to apply there must exist between various actions brought by the same plaintiff against different defendants a connection of such a kind that it is expedient to determine those actions together in order to avoid the risk of irreconcilable judgments resulting from separate proceedings."
[8] Reisch
Montage AG v Kiesel Baumaschinen Handels GmbH [2007] I L Pr 10
dealt with how the Regulation should be interpreted and affirmed the approach
taken in Kalfelis (paras 21-27). Since the factual context was
different from that of the present case it would be unhelpful to read the
decision as directly applicable to the present case, in which the action was
brought against D P & L on the basis of an error of fact and it would never
have been possible for the pursuers to succeed against the true employer of the
deceased at the material time. There was a particular misuse of the Regulation
in this case because the pursuers had no cause of action against any party
domiciled in Scotland at the time the action was raised.
[9] In Compagnie
Commercial Andre SA v Artibell Shipping Co Ltd 1999 SLT 1051 Lord Macfadyen
dealt with the question of the risk of irreconcilable judgments between
p 1057G and p 1058K. He stated at p 1058E "that in most cases
the presence or absence of the risk of irreconcilable judgements will be
obvious". As the present action was live only against two defenders domiciled
in England there was no risk of irreconcilable judgments. That was something
which could be discovered only now, but the risk was not present and never had
been.
[10] In Painer
v Standard Verlags GmbH [2012] ECDR 6 the Third Chamber of the ECJ
affirmed (at pps 104-5, paras 72-84) the approach taken in Reisch
Montage.
[11] In FKI
Engineering Ltd v Stribog Ltd [2011] 1 WLR 3264 Rix LJ
stated at p 3295D, para 128, that "the claimant has to find a court
with jurisdiction sanctioned by the legislation, without which he cannot get to
first base". The pursuers in the present action had failed to get past first
base because they had not found a court with jurisdiction sanctioned by the
Regulation, the exception in Article 6.1 therefore could not apply and the
action should be dismissed for want of jurisdiction.
Submission for the
pursuers
[11] The
broad proposition on behalf of the pursuers was that the jurisdiction of the
Court of Session had been established when the action had been raised and that
that jurisdiction remained. It was accepted that there was now no risk of
irreconcilable judgments, but there had been a genuine basis for jurisdiction
when the action had been raised. Jurisdiction fell to be determined at the
time when the action was raised, and only then. For example, if the action in
so far as directed against D P & L had been dismissed as irrelevant after
debate, the court would not then have considered anew the issue of jurisdiction
against the remaining defenders. Jurisdiction had always to be determined at
the time of the raising of the action: otherwise the system would be
unworkable.
[12] In MacDonald
v FIFA 1999 SLT 1129 it was held by the then temporary judge,
founding on Kalfelis, that the appropriate time to test jurisdiction was
at the raising of the action. At p 1130D-E he quoted a passage from
p 5570 in Kalfelis in the following terms:
"... the parties, the various governments and the Commission all agree ... that the rules of jurisdiction should be considered with reference to the position when the matter was brought before the Landgericht ... so that the outcome of the various claims made ... at each stage of the procedure is irrelevant to the application of the rules on jurisdiction."
The temporary judge went on at p 1130H-I to state:
"I hold that I ought to decide the issue of jurisdiction when the action was raised and free from any further procedural steps which may arise after that."
That decision was a fairly clear application of what was stated in para 12 of the judgment in Kalfelis.
[13] In Freeport
plc v Arnoldsson [2008] QB 634 the Third Chamber of the ECJ
answered a question referred to it by stating in its judgement at p 658D-E,
para 54, as follows:
"In those circumstances, the answer to the question referred must be that article 6.1 of Regulation No 44/2001 applies where claims brought against different defendants are connected when the proceedings are instituted, that is to say, where it is expedient to hear and determine them together to avoid the risk of irreconcilable judgments resulting from separate proceedings, without there being any further need to establish separately that the claims were not brought with the sole object of ousting the jurisdiction of the courts of the member state where one of the defendants is domiciled."
[14] In Reisch
Montage the Second Chamber of the ECJ stated at para 33 of its
judgment:
"Article 6.1 of Regulation 44/2001 must be interpreted as meaning that ... that provision may be relied on in the context of an action brought in a Member State against a defendant domiciled in that state and a co-defendant domiciled in another Member State even when that action is regarded under a national provision as inadmissible from the time it is brought in relation to the first defendant."
It was clear from that passage that there was no distinction to be drawn between the court being seised of an action and having jurisdiction to try the action. In that case, as in the present case, the action against the first defender as pleaded could never have succeeded: there were unfinished bankruptcy proceedings against him, which meant that under Austrian law the action against him could never have been competently raised and was always bound to fail. The effect of the ECJ judgment on the present case was that it was irrelevant whether with hindsight the pursuers could ever have succeeded in the case which they pleaded against D P & L, who were genuinely taken to be the correct company to sue as defenders, under the correct designation. The only relevant consideration was whether the terms of Article 6.1 were satisfied as at the raising of the action, taking the averments in the summons pro veritate. Once jurisdiction had been established on the basis of Article 6.1 taking the averments pro veritate at the commencement of proceedings it could not be elided by the later departure of D P & L from the action for whatever reason, whether that be by the fact that it was never competent to sue them, or by dismissal through irrelevancy following debate, or dismissal following a preliminary proof on time-bar, or through abandonment by the pursuers after discovery by them during the proceedings that there were no prospects of success against those defenders, or as in the present circumstances. Once jurisdiction was established in the court first seised of the proceedings it was retained thereafter. The continuance of jurisdiction in Scotland in this case was not compromised by the subsequent absence of D P & L from the proceedings. Such an approach was consistent with European decisions where there was a loss during the action of the original basis of jurisdiction.
[15] In Staubitz-Schreiber
[2006] I L Pr 30 the Grand Chamber of the ECJ held that there was no transfer
from the original court if a debtor's assets in insolvency proceedings were
transferred to another jurisdiction during the action, jurisdiction having been
established in the original court at the commencement of proceedings. The
court stated at para 29 of its judgment that Article 3.1 of the EC
Insolvency Proceedings Regulation "must be interpreted as meaning that the
court of the Member State within the territory of which the centre of the
debtor's main interests is situated at the time when the debtor lodges the
request to open insolvency proceedings retains jurisdiction to open those
proceedings if the debtor moves the centre of his main interests to the
territory of another Member State after lodging the request but before the
proceedings are opened".
[16] In Re
Jurisdiction based on the Domicile of the Defendant [2012] I L Pr 12
the German Federal Supreme Court held that the principle of German procedural
law under Article 261.3 of the Code of Civil Procedure, to the effect that
the jurisdiction of a court, once established, was maintained even if the
circumstances which justified the initial assumption of jurisdiction ceased to
apply during the course of the litigation (perpetuatio fori), was
applicable to international jurisdiction as well, and in particular to the
jurisdiction established under Article 2.1 of the Brussels I Regulation. It
held (at para 26) that the Appeal Court was correct to accept that
international jurisdiction, once established, did not cease to operate
subsequently because the defendant moved back to France during the litigation
and (at para 27) that the principle of German procedural law to the effect
that the jurisdiction of a court, once established, is maintained even if the
circumstances which justified the initial assumption of jurisdiction cease to
apply during the course of the litigation is considered by the vast majority of
courts and academics to be applicable to international jurisdiction as well. At
para 29 it stated:
"In line with the jurisprudence of the ECJ, it is to be assumed that the principle applies in relation to Community law provisions on jurisdiction if their objectives of foreseeability, efficiency and legal certainty would otherwise - i.e. in the case of a change of jurisdiction from the court first seised to a court in another Member State - be defeated."
[17] In the
recent case of Sibir Energy Ltd v Tchigirinski [2012] I L
Pr 52 in the English Commercial Court the law on Article 6.1 was
conveniently summarised by Judge Mackie QC. He concluded as follows at para 27:
"From Kalfelis onwards the Court repeats regularly that 6.1 cannot be applied to allow an applicant to claim against a number of defendants with the sole object of ousting one defendant's domicile. That risk is however seen to be avoided by ensuring that the irreconcilability test is met - this is implicit in Kalfelis and Reisch and explicit in Freeport and also in the Advocate General's Opinion in Painer. The starkness of the proposition is illustrated by Reisch, a decision I find difficult to follow unless looked at in a very narrow sense, where the Court permitted article 6.1 to be invoked against the foreign defendant even though the case could not proceed against the anchor defendant - a situation in which there could be no possibility whatsoever of irreconcilable judgments."
At para 31 he went on to say that the court must be able to refuse an application which otherwise meets the requirements of Article 6.1 if there is clear evidence of collusion or abuse.
Discussion
[18] It
was never disputed that, assuming that the pursuers had sued the correct party
instead of D P & L, the claims against the three defenders would have been
so closely connected that it was expedient to hear and determine them together
to avoid the risk of irreconcilable judgments. The action is one based on
joint and several liability against three successive employers (or their
successors) of the deceased for causing his mesothelioma and in these
circumstances the claims are self-evidently so closely connected that it is
expedient to hear and determine them together in order to avoid the risk of
irreconcilable judgments resulting from separate proceedings. The only basis
for challenging the applicability of Article 6.1 to the present action was
that it could not apply when the claim against D P & L was impossible from
the time of the raising of the action, with the consequence that there never
was any risk of irreconcilable judgments.
[19] I have no
difficulty in reaching the conclusion that the submission for the defenders is
an erroneous one. At its heart was the proposition that there was a
fundamental distinction between the court being seised of an action and the
court having jurisdiction to deal with the case. No authority was referred to
which supports such a distinction and I think that is because no such authority
exists. In my opinion the fact that the claim against D P & L could never
have succeeded is irrelevant to the question of jurisdiction. Such a view is
supported by the decisions in Kalfelis, MacDonald v FIFA, Reisch
Montage, Staubitz-Schreiber, Freeport plc v Arnoldsson and
Re Jurisdiction based on the Domicile of the Defendant. These decisions
clearly lay down that jurisdiction is established on the basis of the pleadings
in the case at the time that the action is raised, and not at any later time. Moreover,
once jurisdiction is established it cannot be lost, whatever may happen in the
action. That is known as the principle of perpetuatio fori. It would
be quite impracticable for the court to monitor or keep under review the
question of jurisdiction throughout the course of an action. It is well
established that a plea of no jurisdiction is one which must be taken in
limine. The pleadings in this case averred jurisdiction against D
P & L. There is no suggestion that the relevant averments were made in bad
faith or based on collusion: it is plain that they were made on the basis of a
genuine misapprehension on the part of the pursuers. That being so, the court
had jurisdiction when the action was raised and that jurisdiction remains
although the action was subsequently abandoned against D P & L.
Decision
[24] I
shall repel the defenders' plea of no jurisdiction.