1.

Introduction

As we have seen, conflict of laws deals with cases which involve a foreign element in a
number of ways. First, there are rules which decide whether the English court can and
should hear a particular dispute. Secondly, there are rules by which the English court will
recognise or enforce a foreign judgment. Thirdly, there are choice of law rules by which
the English court identifies a system of applicable rules under which the substantive
result of the case is determined. These rules may be ones of English law or of a foreign
law. In the American case of Loucks v Standard Oil Co. of New Jersey, Cardozo CJ
eloquently put it ‘We are not so provincial as to say that every solution of a problem is
wrong because we deal with it otherwise at home’. 1 Choice of law rules are not
generally determinative of themselves, they point to substantive rules which are so
determinative. Some domestic English rules are unilaterally applied because their
territorial scope requires their application to particular circumstances, even if there is a
foreign element to the facts or the parties. These unilateral rules are not generally
recognised as choice of law rules but the interpretation of the territorial scope of English
rules often requires rather similar techniques to multilateral choice of law rules. It is clear
that choice of law rules themselves are rules of English law and the English court in
applying foreign law does so in the context of English procedure.

2. Analysis
The conflict of laws, in so far as it is concerned with the choice of applicable law,
consists only of a relatively small number of rules; the problems tend to arise, as in most
areas of law, with the exceptions. But, for the moment we will stay with the general rules.
These can all be stated in the same simple form, for example:
(a) the formal validity of a marriage is governed by the law of the place of celebration;
(b) capacity to marry is governed by the law of the parties’ domiciles;
(c) title to movable property is governed by the lex situs2 at the time of the transaction;
(d) succession to movable property is governed by the law of the domicile of the
deceased at the date of death;
(e) succession to immovable property is governed by the lex situs;
(f) procedure is governed by the lex fori; 3 (
g) liability in contract is governed by the applicable law of the contract;4
(h) liability in tort is generally governed by the law of the country in which the damage
occurred.5
It is possible for some issues to be governed by more than one system of law. These may
be applied cumulatively as, for example, liability for defamation committed abroad arises
only if liability is established both by the law of the place where the events took place
and by English law as the lex fori. 6 Or the laws may be applied alternatively as, for
example, a contract is formally valid if it is valid either by the law applicable to the
contract or by the law of the place where the parties were when it was concluded.7 These
rules can all be analysed in the same manner. First, the legal category or issue is to be
identified. For example, in (c) the legal category is a question of title to movable
 property. This is a pigeon-hole into which the legal issue disclosed by the facts of cases
may be placed. Secondly, the connecting factor is to be identified, so-called as they
connect legal categories to the applicable law. In (c) it is the situs of the property, i.e., its
location, similarly there are not many connecting factors.
This may all seem very straightforward, but there are disguised difficulties about it.
Problems may occur in a number of ways. (1) The case may fall into one legal category
in the view of the lex fori (English law) but into another in the view of the foreign law
which is alleged to be the lex causae (applicable law). Thus, English law may regard the
case as being concerned with formalities of marriage, but French law may regard it as
raising the question of capacity to marry. To put it another way, how is the issue to be
characterised?
(2) English law and the foreign law may agree on the legal category but disagree on the
connecting factor to be employed, as where English law regards succession to movables
to be governed by the law of the domicile, whereas by the foreign law it is governed by
the law of nationality. This raises a question of renvoi. What is meant by a reference to
the ‘applicable law’? Is it the law as would be decided exactly by a judge of the foreign
court, including some reference to the international elements of the case, i.e., together
with the private international law of the foreign law? Or is it merely a reference to the
domestic rules of the foreign court?

(3) English law and the foreign law may agree on the legal category and even on the
connecting factor. However, this conceals a latent conflict, because the two laws mean
something different by the connecting factor. Thus, under both laws succession to
movables is governed by the law of the last domicile of the deceased. But, by the English
law of domicile the deceased died domiciled in France; under French law the deceased
died domiciled in England. This is also a question of renvoi.

(4) The issues might fall into two different legal categories, each having a different
choice of law rule. So a wife may claim a right of intestate succession to her husband’s
immovable property. The property is situated in Italy, therefore the choice of law rule
says that Italian law applies. However, if there is also a question whether the wife is
married to the husband, a separate choice of law issue also arises. This is termed an
‘incidental question’.

3. Connecting factors
The connecting factors employed by the conflict of laws are not especially numerous.
They include the personal law (domicile, habitual residence and, rarely, nationality); the
place where the transaction takes place (as place of celebration of marriage); the intention
of the parties (as expressed by a choice of law clause in a contract); the place of the
seller’s residence (as in contracts for the sale of goods); the place of the damage (as in
torts); the situs (the location of property); and the place where the court is hearing the
case. Since the conflict of laws forms part of English law, English law alone can
determine when a foreign law is to be applied. It follows from this that English law must
 not only select the connecting factor, it must also say what it means. This is clear in
respect of domicile and, for jurisdictional purposes, the place of contracting.8 Thus, if
both English and French law use domicile as a connecting factor, but by English law a
person is domiciled in France and by French law in England, he will be regarded by an
English court as domiciled in France. In Re Annesley: Mrs A died domiciled in France
according to English law. By French law she had never acquired a domicile there. Russell
J held that Mrs A died domiciled in France.
This question also arose in relation to the situs of property, which is the connecting factor
for proprietary interests. In Dornoch Ltd v Westminster International BV: 10 The matter
concerned the situs of the ship; in this case its physical location was Thailand. However,
the ownership of ships and aircraft is generally registered on a registry recording sales
and some other proprietary interests; this ship was registered in the Netherlands. There
was evidence that the law of Thailand would refer to Dutch law to determine title.
Tomlinson J determined the situs of the ship for the choice of law rule at the physical
location of the ship not the place of the registry. There can be no doubt that if it should be
necessary to determine the situs of intangible property, for example, a bank account at a
New York bank’s English branch, English law would apply, and the situs would be
England, even if by New York law it would be New York.

4. Characterisation
We have seen that the process known as ‘characterisation’ is sometimes necessary
because English law may regard the case as falling into one legal category but the
relevant foreign law believes it to belong to a different one. This process is undertaken
also in cases which do not contain a foreign element; a court may be called upon to
determine whether the issue sounds
in contract or tort, or whether property is realty or personalty, but it is obviously more
difficult where a foreign element is present. Examples of characterisation are to be found
throughout this book, where they are discussed in some detail. They include questions
such as whether a rule of English law which required that in order for an action to be
brought on a contract there must be written evidence thereof, was a rule of evidence and
thus procedural, so that it applied by virtue of the lex fori to a contract governed by
French law. It was held that it was procedural and applied.15 In several cases the
question was whether a rule of a foreign law requiring an action to be brought within a
certain period of time was substantive and applicable or procedural and irrelevant. The
courts held that it was procedural16 although that has now been overturned by statute.17
The question has arisen whether a sum awarded as part of a judgment by a foreign court
and described by the foreign law as a penalty, though it was not so regarded by English
law, should be characterised as a penalty or not. It was held that English law governed the
matter and it was not a penalty.18 It will be observed from these examples that the
English courts have generally characterised the issue before them according to their own
notions; this will be illustrated further.19 The problem of characterisation is one of the
most difficult in the conflict of laws, and it has generated an enormous amount of writing
in many languages. It might well be thought that its difficulties and obscurities increase
in direct proportion to the amount of juristic discussion of it.20 There has been
considerable difference of opinion as to how the problem should be solved. The courts
are usually criticised for solving it the wrong way and nearly all the cases referred to
above have been the subject of severe criticism. It is true that the solutions arrived at have
caused, or are capable of causing, considerable difficulties. This is so much so that in
some areas, legislation has been used to turn the law around.2

Before proceeding further it has to be said that there has been very great debate and
confusion right at the start of the inquiry as to what it is that is characterised. Is it a ‘legal
issue’, a ‘legal relation’, ‘a legal claim’, ‘a legal question’, ‘a factual situation’, the ‘facts
of a case’, or ‘the rule of English (or foreign) laws’? The real question is whether it is the
facts or factual situation or a legal question. Since some aspects of characterisation
clearly do not involve the facts this leads to the conclusion that it is a legal question. It is
generally accepted that legal issues are characterised for the purposes of the choice of law
rule,22 rather than particular legal rules. There is one exception to this rule: the case of
classifying foreign penal, revenue and other public laws. Here, it is the foreign rule which
is classified for the very particular reason that such rules are not applied even if they are
part of the governing law. English law as the lex fori appears to adopt its own
classification of the foreign law.23 Legal issues are characterised rather than facts
because although any case involves the facts, of course, what the choice of law rule
points to is the legal rules of some system. The facts are those data which enable the
judge to formulate, as he must always do, a legal issue which leads to the application of
the legal rule. A judge or lawyer is not interested in the facts in some sort of vacuum and
they cannot be characterised in the abstract, but only by formulating legal categories;
these are categories of legal questions. In addition, the manner in which the claim is put
to the court for decision inevitably involves some prior classification of the facts into
legal issues by the parties. Various solutions to the problem of characterisation have been
put forward; five will be mentioned.

a. The lex fori theory

This was proposed by the German and French writers, Kahn24 and Bartin,25 who
‘discovered’ the problem in the 1890s. It has been the prevailing theory in continental
Europe, and by and large, has been adopted in practice by the English courts.
According to this theory the court should characterise the issue in accordance with the
categories of its own domestic law, and foreign rules of law in accordance with their
nearest analogy in the court’s domestic law. Thus, a French rule requiring parental
consent to marriage should be characterised as pertaining to formalities, since English
law so regards its own rules regarding parental consent. Objections raised to the lex
fori theory are that its application may result in a distortion of the foreign rule and
render it inapplicable in cases in which the foreign law would apply it, and vice versa.
Moreover, if there is no close analogy in the domestic law (as there is no analogy in
English law to the matrimonial property regime known to foreign laws) the theory
does not work. Lastly, proponents sometimes seem to suggest that it is facts alone
 which have to be classified, but this is not so; it is facts which are presented as issues
in the light of a foreign law.

b. The lex causae theory

According to this theory, characterisation should be effected by adopting the
categories of the governing law. It is sometimes suggested that at least one English
decision is based on this method, though this is perhaps doubtful.26 There are two
serious objections to this theory. First, the whole purpose of characterisation is to
discover what law governs the issue. To say that the governing law dictates the
process of characterisation is to argue in a circle, for how can we know what the
governing law is until the process of characterisation is completed.27 Secondly, if
there are two possible foreign laws to govern the matter, and they characterise the
issue differently, which is to be adopted by the English court? It may be added that
the adoption of this theory could compel the adoption of idiosyncratic foreign
characterisation, such as a rule of Utopian law that a Utopian person can only be
validly married if he or she goes through the ceremony before a Utopian shaman,
wherever in the world the ceremony takes place. A court may have to resort to public
policy to avoid the consequence of an invalid marriage if a Utopian person marries in
England without a shaman. As public policy should be very sparingly used in conflict
of laws,28 the theory can be criticised for requiring such an exception.
c. Analytical jurisprudence and comparative law
This theory was espoused by the author of the encyclopaedia of comparative conflict
of laws, Ernst Rabel,29 and views similar to his were advanced in England by W. E.
Beckett,30 who said that conflicts rules should use ‘conceptions of absolutely general
character’, and that These conceptions are borrowed from analytical jurisprudence,
that general science of law, based on the results of the study of comparative law,
which extracts from this study essential general principles of professedly universal
application – not principles based on, or applicable to, the legal system of one country
only. This is at first sight attractive, but it has its drawbacks. First, few universal
principles are disclosed by analytical jurisprudence and comparative law which would
be of assistance in this area. Secondly, though comparative law may disclose
similarities between legal systems, it may also disclose differences, which it is hardly
capable of resolving; thus, it may show that requirements of parental consent to
marriage pertain to formalities in some systems or to capacity in others, but this does
not tell us how in the case before us these differences are to be settled. Thirdly, it is
rather impractical; it would be asking too much of legal advisors and judges to
undertake the exercise involved and the results would be unpredictable.31 This
method would certainly add to the length and cost of litigation. However, a variant of
this theory has found some favour with the English court which has said that
characterisation must be done with a ‘broad internationalist spirit’. 32

d. Falconbridge’s views
 The Canadian lawyer, Falconbridge, proposed a two-stage process.33 The first stage,
a task for the lex fori, is to define the scope of the legal category, the categories not
being those of the domestic law system but of its private international law; and the
second to examine the relevant legal rule in its own context to see whether it can be
fitted into the legal category in question.

e. English courts and characterisation

The English courts have not consciously adopted any one doctrine or theory. Indeed,
the question of characterisation has been referred to in the cases in the most general
form. For example, in Wight v Eckhardt Marine GmbH, 34 Lord Hoffmann approved
of the argument made by the defendants, noting ‘the purpose of the conflicts
taxonomy is to identify the most appropriate law. This meant that one has to look at
the substance of the issue rather than the formal clothes in which it may be dressed.’
It is to be remembered that characterisation is done for the purpose of identifying the
choice of law rule. Choice of law rules are developed in order to achieve just results
in particular cases in which a foreign element is present. Justice is a broad concept but
in this area it has been focused largely on fulfilling parties’ expectations. Also, choice
of law rules limit the role of domestic law because the foreign elements make
unrestricted application of that law unjust.35 Looked at in this way, characterisation
is necessarily part of a process of the lex fori. Indeed, the lex fori theory, modified in
some cases so as to approximate to Falconbridge’s view, represents the actual method
employed by the courts. The foreign classification is not adopted as such except in
deciding whether foreign property is immovable or movable.36 This is not to say that
the foreign law is completely ignored as irrelevant, as it may be considered both to
identify the issue and to determine the choice of law rule ‘in a broad internationalist
spirit’. 37 So the concept embraced in the choice of law rule is given a wider
interpretation than it would be for domestic law. In De Nichols v Curlier, 38 the
English court accepted a claim based upon the French community property regime
despite there being no exact equivalent in England.39 Likewise at common law the
‘contract’ choice of law rules cover agreements not supported by consideration.40 In
practice, the English courts formulate the issue and define the ambit of the legal
category for themselves, and then they determine whether a question posed by a
foreign rule comes into that category. An example of this is Re Cohn: 41 A mother
and her daughter, domiciled in Germany, were killed by the same bomb in an air-raid
on London. It could not be shown which died first and which survived the other. The
choice of law rule for succession issues is that of the law of the domicile, here
German law. By that law the daughter’s estate could only succeed under her mother’s
will if she survived the mother. There is an English rule by which a younger person is
presumed to have survived the older where they die together.42 Uthwatt J held that
this rule was not one of evidence but of substance and so did not apply as English law
was not the applicable law. The corresponding provision of German law, by which
the two were deemed to have died simultaneously, he held part of the law of
inheritance and so applicable by the choice of law rule. Thus, he categorised the issue
 as one of succession and then held that the question presented by German law came
within that category

A special case: substance and procedure

Matters of procedure are governed by the lex fori (i.e., English law) whatever be the lex
causae, for example, the Utopian applicable law of a contract. Whether a question is
procedural or substantive has presented particular difficulties of characterisation, as has
the question of whether a foreign rule of law affects procedure or substance. It is easy
enough to state that the substantive issues are those which concern the existence of a
right, whereas procedural issues are those which concern the method and means of
enforcement of a right; but acute difficulties may be encountered in deciding whether
even an English rule is procedural or substantive. Thus, in Harding v Wealands51 the
English court was faced with an issue concerning damages for an accident in New South
Wales, Australia. The applicable law determining liability was that of New South Wales
and that law contained various rules calculating the sums payable as a result of accidents
on the road. However, their Lordships followed the earlier case of Boys v Chaplin52 to
hold that the issue of quantification of damages was procedural and thus for English law
as the lex fori. Therefore, these rules of New South Wales law did not apply to limit the
claimant’s damages. Whether the events gave rise to liability was a question for the
applicable law, but the extent of the remedy in terms of money damages was procedural.
If one is to look at the purpose for the rule that procedural matters are decided by the lex
fori, it can be concluded that procedural matters should be limited to those relating to the
administration of the court proceedings, such as what language is spoken, which party is
to make their case first and what advocates are permitted. In addition, as the purpose of a
substantive choice of law rule is often to conform to the parties’ expectations, the
applicable law should be given a broad scope which could include the manner of
assessment of damages. Although this is the better view, it has not found complete
acceptance in England. An example of a pragmatic approach which appears to take
account of both the lex fori and the substantive law can be seen in the more recent case of
Maher v Groupama Grand Est: The claimant had been injured in a road accident in
France by a French driver insured by the French defendant. Liability was accepted but
quantum was disputed. The claimant brought an action in damages directly against the
insurer in England.53 The dispute centred on the issue whether the award of pre-
judgment interest on the damages was to be determined under English or French law.
Was the liability to pay interest before judgment substantive or procedural? Senior Courts
Act 1981, s. 35A, gives the court power to award such interest ‘as it thinks fit’. A further
complication was that the insurer’s liability sounded in contract but the tort victim’s
claim was in tort. Fortunately, the applicable law of both tort and contract was French
law. The lex fori, obviously, was English law.54 The Court of Appeal decided that
individual issues were characterised rather than whole claims. In this case, therefore, the
issue to be characterised was what damages the claimant should receive to compensate
the claimant for the injury suffered by the action of the tortfeasor. This sounded in tort
and was substantive. It was referred to French law. The insurer was liable for damages
under that law. The liability to pay interest was also to be determined by French law.
However, Moore-Bick LJ held that Senior Courts Act 1981, s. 35A (which is a rule of
English law) creates a remedy which was to be characterised as procedural. The section
therefore applied so long as French law imposed liability for interest. As it is a
discretionary remedy, the exercise of discretion should take into account the relevant
provisions of French law on the recovery of interest. Note that this case was decided
before the Rome II Regulation55 was applicable.56 After the entry into force of that
Regulation, the assessment of damages is generally for the applicable law.57

The incidental question

The problem of the so-called incidental question arises when, in the course of deciding a
case, an issue which is subsidiary to the actual issue to be decided occurs. Thus,
entitlement to share in the estate of a deceased person may depend on whether the person
in question is legitimate or illegitimate, and this in turn depends on the validity of his
parent’s marriage as in Shaw v Gould. 71 For the problem to arise there must be (1) a
principal or main question governed by English conflict rules by the law of country A;
and (2) a subsidiary or incidental question in the same case, which could arise on its own
and is governed by the law of country B. Also (3) the application of the law of A must
produce a result different from that which would follow from the application of the law
of B. In only a very few cases has this situation arisen. It did not in the event arise in
Shaw v Gould, since all the issues involved were governed by the same law.72 Writers
express different views on whether the answer to the subsidiary issue should, when the
problem arises, decide the case (in which event the law governing the principal issue
should not be given its normal effect), or whether the latter law should be applied so that
international harmony as to the result might be more easily achieved. The problem arose
in one Canadian case and in two English cases. In all three the principal question
concerned capacity to marry or remarry and in all three the subsidiary issue was the
recognition of a foreign divorce decree. In the Canadian case and the first of the English
cases, the court applied the law governing the principal issue and effectively excluded the
law which would have governed the subsidiary issue had it arisen on its own. In the
second English case, the court allowed the main question to be determined by the answer
to the subsidiary question. In the Canadian case of Schwebel v Ungar: 73 H1 and W were
Jews domiciled in Hungary. They both left Hungary for Israel but en route obtained a
divorce in Italy. They both arrived in Israel where W acquired a domicile. She then went
to Ontario and there married H2 who was domiciled in Ontario. Under Ontario conflict
rules W had capacity to marry H2 since her capacity to marry was governed by Israeli
law at the time of the marriage. Since Israeli law recognised the Italian divorce, it
regarded her as a single woman. But by Ontario conflict rules that divorce was not
recognised since at the time it was obtained the parties were still domiciled in Hungary,
whose courts had not granted it and did not recognise it. Therefore, in the eyes of the
Ontario court W was still married to H1 and H2 under the law of his domicile, Ontario,
had no capacity to marry her. The court, applying Israeli law (governing the principal
issue) and ignoring Ontario law (governing the subsidiary issue) held the marriage valid.
The converse situation arose in England in the Brentwood Marriage case: 74 H and W
were domiciled in Switzerland, where a divorce was obtained. This was recognised in
England. H was an Italian national and by Swiss law his capacity to marry was governed
by Italian law as his national law. By Italian law the Swiss divorce was not recognised so
that under Italian and Swiss law he could not marry. The English court, like the Canadian
court, concentrated on the issue of H’s capacity to marry to the exclusion of the
recognition of the foreign divorce, applied the law of his domicile75 and held that he
could not remarry in England, though in the eyes of English law he was an unmarried
man. In one sense Schwebel v Ungar might be regarded as satisfactory, but the
Brentwood Marriage case as unsatisfactory, since the former promoted freedom to marry
and the latter denied it. Indeed the Brentwood Marriage case has since been reversed by
statute, in that H’s incapacity would now be disregarded.76 From another point of view
Schwebel v Ungar is equally unsatisfactory. If it represented English law it would mean
that a person who is domiciled here could marry someone who, in the eyes of English
law, is a married person, that is to say, contract a bigamous union and in so doing commit
a crime here. It is not clear that Schwebel v Ungar would be followed in England, but it is
thought it would not. In the second English case Lawrence v Lawrence, 77 the facts were
basically the same as the Brentwood Marriage case except that the remarriage had taken
place abroad. This being so, and the legislative provision which reversed the latter case
being inapplicable where the marriage takes place outside the United Kingdom, one
would have expected the decision to have been the same and the remarriage to have been
held invalid. But the Court of Appeal, without saying that the Brentwood Marriage case
was wrong, held that because the court must recognise the foreign divorce the wife was
free to remarry, though she had no capacity to do so under the law of her domicile.78 It
thus made the incidental question effectively determine the main question. Indeed, the
majority specifically said that the question of capacity to marry in the usual sense did not
arise. It is submitted that this was highly dubious at common law.79 However, it was
enacted in law by the Family Law Act 1986, s. 50. Another version of the incidental
question can arise when a defendant claims that a contract between the defendant and the
claimant contains a defence to what would otherwise amount to tortious liability; for
example, the tort claim is governed by one law, say Utopian, and the contractual defence
is governed by another law, say Ruritanian. Ruritanian law allows the defence even
though the claimant is a consumer, whereas Utopian law would invalidate this contractual
provision. Which law is to govern whether the defendant is liable? The tort issue might
be argued to be the principal one and the contractual issue subsidiary, so that although the
contractual defence is valid by its own applicable law, it is ineffective as a defence to the
tortious liability. This argument suggests that it is the law of the principal issue which
governs the matter, but the law of the incidental issue must also be inspected. Where
Ruritanian law would invalidate the contractual defence, it should not be given effect
even where Utopian law applicable to the tort would – if the contract had been governed
by Utopian law – have denied liability.80