This page uses so called "cookies" to improve its service (i.e. "tracking"). Learn more and opt out of tracking
I agree

Lowenfeld, Andreas F., Lex Mercatoria: An Arbitrator's View, Arb.Int’l 1990, at 133 et seq.

Lowenfeld, Andreas F., Lex Mercatoria: An Arbitrator's View, Arb.Int’l 1990, at 133 et seq.
Permission Text
This document is included in TransLex by kind permission of Kluwer Law International.
Table of Contents

Lex Mercatoria: An Arbitrator's View



In going through the surprisingly large literature on the topic of lex mercatoria and international arbitration, I was struck most by comment by Mr Justice Rogers of the Supreme Court of New South Wales. In discussing the debates surrounding the adoption of the UNCITRAL Model Arbitration Law,1 the judge refers to the 'clash in philosophy ... between the civil law countries and the United States, on the one hand, and the common law countries on the other ...'2 I have for some time thought that in their attitude to litigation, the United States and France have more in common than the United States and England, or say, France and Germany;3 I had not thought that England's devotion to law and the former colonies' and dominions' devotion to the mother country's law were so passionate as to read the United States out of the common law world.

In fact, the issue that Mr Justice Rogers described as dividing the delegates to the UNCITRAL Conference was not precisely the issue on which we focus here, but it is closely related. The English view -notwithstanding its rather grudging, market-driven relaxation of review of arbitration in international cases4 - on the 134 whole is founded on the belief in judicial supervision of the work of arbitrators, whereas the American and majority continental view, at least among the arbitration community, is that if Parties to a contract have agreed to submit their disputes to arbitration, the decision of the arbitrators ought to be final, subject to challenge only for misconduct of the arbitrators or (possibly) excess of jurisdiction by the arbitral tribunal.5

The continental European and American view, if one can speak of it as such with the understanding that there is in fact a range of views,6 is that arbitrators -even if left free of judicial control- must and do decide according to law. The debate is over what law the arbitrators should apply: the law designated by the parties, the law of the situs of the arbitration, the law that would be applied by the courts at the situs, the law designated by the choice of law rules selected by the arbitrators or -our topic- lex mercatoria.7

A subsidiary aspect of the debate is whether parties may elect that any dispute between them be governed by lex mercatoria, and whether or not such an election is the same as an election of arbitration under amiable composition, i.e., not necessarily governed by legal rules at all.

The English view, hostile to freedom from judicial control, regards lex mercatoria as a slightly wicked misnomer (not to say contradiction in terms), on the ground that lex mercatoria is not law at all.If I understand the exponents of the135 English view, led by Lord Justice Mustill,8 (i) arbitrators may not apply lex mercatoria on their own; (ii) an agreement by Parties to a contract to submit their claims to arbitration (or adjudication) pursuant to lex mercatoria or its English equivalents is void; and (iii) an award rendered by arbitrators according to lex mercatoria ought not to be enforced in English courts, at least if the award is not made in a foreign country that makes a different view and the award is required to be enforced by the New York Convention.9

To my mind [Mustill writes], the whole purpose of a contract being regarded as enforceable in law is that the Parties can ascertain their rights and duties by reference to some external objective standard, before any dispute has risen, and can be confident that if a dispute does arise those rights duties will be enforced as they stand. There cannot be a contract which is governed by no law at all.10

Lord Justice Mustill, on the whole a proponent of arbitration11 - so long as he and his colleagues on the bench can supervise- presses his criticism in a way that one might have thought uncharacteristic for an English jurist. 'Is the lex a law,' he asks, 'or is it a body of rules which the parties choose (expressly or implicitly) to apply to their individual contract?'12 One can guess at his answer from the tone of the question. But just in case a reader or listener with some Latin might he misled into thinking 'lex' means 'law', Mustill puts another question: 'If it is a law, from where does it draw its force?'13 Could not that question be asked about the common law as a whole, not to speak of equity in the English sense?

Mustill is conscious that his attitude might be seen as 'dangerous parochialism'. But he insists on asking the question anyway, with a good deal more than a hint of his own answers. '... for otherwise there is a risk that a tide of joyous enthusiasm for the cutting of the ropes which tie international commercial arbitration to established juridical orders may lead it to float so freely into the theoretical stratosphere that it loses contact with the practical commercial men whom everyone concerned in arbitration must wish to serve.'14

I do not think Lord Justice Mustill's questions are so perplexing. Indeed I do not understand very well why the discussion of lex mercatoria has been focused 136 almost exclusively on international arbitration. To use a well known case as illustration, it will be recalled that in The Bremen v. Zapata Offshore Co.,15 a German towing company and an American off-shore oil drilling company had entered into a contract stating that 'Any dispute arising must he treated before the London Court of justice', but containing no choice of law clause. The U.S. Supreme Court, in upholding the forum selection clause, assumed that English law would apply.16 In fact, since the dispute concerned a casualty at sea, it would have been more accurate to say that the English court would apply maritime law as understood in England. And while that law varies in detail from country to country and has been significantly modified by Statutes and international conventions, essentially it is international commercial law,17 traced by some back to the Island of Rhodes but in any event largely derived from the codification in the Ordonnance de la Marine issued by Louis XIV in 1681.18 Suppose that instead of a accident at sea, the dispute between the same American and German parties concerned an alleged defect in a product sold by one party to the other for installation in a third country, and a controversy arose about the scope and effectiveness of implied warranties and disclaimers of liability. Suppose further (a) that the parties, as in Zapata, had made no choice of law; or (b) that they had provided in their contract that it was to be governed 'according to the law understood to govern relations between merchants.' Why shouldn't the result be the same in both cases? (Mustill might say, 'yes, indeed, the result would be the same in both cases -in (a) we apply English law , implied from the choice of an English forum;19 and since (b) is an invalid choice we apply English law as well.' But of course this is not what I had in mind).

Coming back to arbitration , I am content with the definition, for instance, of professor Ole Lando of Copenhagen:20

By choosing the lex mercatoria the parties oust the technicalities of national legal systems and they avoid rules which are unfit for international contracts. Thus they escape peculiar formalities, brief cut-off periods, and some of the difficulties created by domestic laws which are unknown in other countries such as the common law rules on consideration and privity of contract. Furthermore, those involved in the proceedings -parties, counsel and arbitrators- plead and argue on an equal footing; nobody has the advantage of having the case pleaded and decided by his own law and nobody has the handicap of seeing it governed by a foreign law.


To Mustill's plea for certainty and predictability, Lando might say, that the general principles of the law merchant offer at least as much predictability as the (often unexpected) law of a given country, particularly a law not selected by the parties. To Mustill's question about the source of the law, set forth also in F. A. Mann's forceful insistence that ‘Lex [by which he does not mean lex mercatoria] facit arbitrum,‘21 Lando replies:22

... the binding force of the lex mercatoria does not depend on the fact that it is made and promulgated by State authorities but that it is recognised as an autonomous norm system by the business community and by State authorities.


My own view of lex mercatoria is somewhat different from those both of its critics and of its proponents, though closer to the latter. It may be useful to begin with some illustrations, both from my own experience and from that ofothers, and then attempt to resume the doctrinal debate.

First, Lando gives the illustration of the Danish seller who has sold goods to a German buyer but delivered them after the last date specified.23 Under Article 27 of the Scandinavian Sale of Goods Act as in force in Denmark,24  a buyer who seeks to make a claim arising out of late delivery must give notice immediately at arrival of the goods. No such rule applies in Germany or (so far as appears) in other nations. An arbitral tribunal faced with a claim based an late delivery and no law designated in the contract should not devote its energies to the question of whether seller's law (Danish) or buyer's law (German) applies to the controversy: The Scandinavian rule is an internal one, not fit for international sales. Whether Danish or German law applies to other aspects of the contract, the German buyer should not be defeated by failure to give immediate notice, as long as he has complied with the general commercial rule requiring notice within a reasonable period of time. Use of lex mercatoria yields the correct solution: the German buyer prevails.

Confidence in the solution is here strengthened by the fact that the Vienna Convention on Contracts for the International Sale of Goods25 provides in Article 49(2) (a), that notice must be given within a reasonable time after the buyer has become aware that delivery has been made. But the case as posed by Lando is not one of the application of a treaty. At the time Lando was writing the138 Vienna Convention was not in effect for any country, and even today, when the Convention has entered into effect, it is not clear that it would be binding on the contract in question.26 Nevertheless jurists coming from different countries and sitting as arbitrators in Paris or Zurich or Geneva -or indeed in Copenhagen- can draw comfort and support from the fact that their understanding of international commercial law/ practice is consistent with the United Nations Convention designed to reflect international consensus.27

Secondly, a similar case, in which I served as one of the arbitrators, may serve further to illustrate the uses of lex mercatoria, though the case is more complicated and perhaps more controversial. The claimant was a French company which asserted that it had entered into a contract with an Austrian company for the long-term supply of a commodity produced in a developing country. It was clear that there had been no performance, but the Austrian company, the respondent, maintained that the contract had not been validly concluded, and that even if it had been, the formalities required to constitute a valid agreement to arbitrate had not been complied with. The document asserted to embody the contract -some twenty pages of detailed clauses negotiated oven several months- had been signed at a hotel in New York by the two principal negotiators, Mr A for the claimant and Mr B for the respondent. The respondent, however, pointed out that under Austrian law (at least Austrian internal law) an agent's power in bind a corporation depends on the corporate resolutions inscribed in the Commercial Register, and that Mr B was listed in the Register as having only collective and not individual signing authority. The document stated that it was to be governed by New York law, and that any disputes that could not be resolved by negotiation were to be submitted to arbitration in Geneva under the rules of the ICC.

The claimant argued for application of New York law as the place of execution and the law chosen by the parties. Pointing out that New York not only had no double signature requirement but also had a well developed doctrine of apparent authority. The respondent had two main arguments: first that the139 issue was one of capacity and that capacity of a Corporation is determined by the law of the place of incorporation-here Austria; and second, that if the Austrian law were not applied directly by the arbitrators, it should be applied on the basis of the conflict of laws rules applicable at the seat of the arbitration, i.e., Switzerland. The claimant and the respondent each produced an expert opinion on Austrian law, one showing the importance of Austria's rule on authority to bind a corporation, especially in the context of derogating from jurisdiction of the courts (i.e., agreements to arbitrate), the other suggesting that the role might not apply to international contracts entered into abroad. Both Parties also produced expert opinions on Swiss law, one stating that under Swiss law the authority of a person to bind a corporation depended on the law of the corporation's siège social, the other stating that the rule with respect to inscriptions in the commercial register applied to domestic transactions -i.e., between different cantons- but probably not to international transactions or to agreements to arbitrate governed by the New York Convention, which requires only that an agreement to arbitrate be in writing.28 It was also unclear from the expert reports whether Swiss law would regard the validity of the agreement to arbitrate as governed by the law of the chosen forum, here Geneva,29 or by the proper law of the contract, whatever that might be. Nor, of course, was it clear that the arbitrators should look to Swiss law for choice of law purposes -the ICC rules suggested the reverse.30

Thus the dispute in its preliminary phase -long before issues such as excuse for non-performance, interpretation of the obligations undertaken, or duty to mitigate damages were addressed- could be governed by the law of three different states: New York, Austria and Switzerland, plus the conflict of laws rules of Switzerland or New York and a question of interpretation of the Austrian law.

As it happened, I do not know whether by coincidence or by design, two of the arbitrators were experts in conflict of laws, and we managed to find our way to what I believe was a principled and technically correct solution. I cannot be certain however, whether we decided first on the solution and then found a way to achieve it,31 or whether we set out on a truly unguided journey and ended up with the right results.32 It is fair to add that the arbitrators had heard testimony140 and had studied the prior relations of the parties, so that they had some feeling for the transaction beyond the abstract contract and conflict of laws issues they were required to decide.

My feeling is that wholly neutral principles of conflict of laws are an illusion, but that an understanding of reasonable behaviour and expectations of major commercial enterprises can be defined with a fair degree of precision, focusing on such issues as the customs of the particular trade,33 justified expectations of the parties in the light of the prior and current communications between them, the obligation of good faith dealings, and evidence of reliance. Of course these concepts are not precise in the same way as the number of grams in a kilogram, but they are everywhere the stuff of contracts, the kinds of issues an which arbitrators and judges are competent to pass. The suggestion is that whether or not the two parties before the tribunal had concluded a contract could and probably should have been decided by reference to lex mercatoria, rather than by elaborate exercises in comparative conflict of laws.34

Together with Goldman, Lando, and most of its other proponents, I do not view lex mercatoria as some arcane mystery, open only to anointed guardians of an ambiguous flame. It is perfectly appropriate, in my view, for counsel to submit argument to the tribunal about the content of the lex mercatoria, as well as about the usages of the particular trade and the circumstances on which the parties had, or fairly could have, relied. In fact, in my experience, counsel nearly always do present such evidence and argument, in one guise or another. I do not recall anyone arguing, in a commercial case,35 'The result we seek is plainly unfair, but it's just too bad, that's the law.' Even on issues of time bar, perhaps the most common technical defence to contract claims the respondent usually makes some argument to the effect that the claimant knew or should have known, or was negligent in not giving notice, and so on.

It may be argued, and indeed it had been forcefully argued by Mustill and others, that in receiving submissions or deciding on the basis of lex mercatoria the arbitrators cannot assure that their decision will be the same as it would have been had the controversy been submitted to a court -to follow my illustration in Vienna, Geneva, or New York. But (quite apart from not knowing with which of these courts the arbitral tribunal should identify) the parties did not contract to deliver goods from Vienna to Innsbruck: they entered into (or at least con-141 templated) an international transaction; and they did not agree to submit the controversy to one of the courts mentioned: if they agreed at all, they agreed to submit their differences to international arbitration.36

It is important to emphasize that lex mercatoria is not amiable composition. Nor is it a search for abstract justice. Dr. Mann, an enemy of lex mercatoria, gives the example of English law, which permits a buyer six years to pursue a claim for breach of warranty against the seller,37 and German law,38 which (absent fraud) allows the buyer only six months to claim a defect.39 'Which system,' he asks, 'is fairer?' making clear that he would not trust any arbitrator to give a answer. He is probably correct on that point, though whether it follows that an arbitrator sitting in Paris can accurately decide, without more, whether English or German law should govern the dispute is more doubtful.40 It is to solve such problems that the Vienna Sales Convention seeks to establish a governing rule that would apply regardless of the forum, provided the two states in question were both parties to the Convention.41 But lex mercatoria does not just invite one or three individuals to cogitate about what is fair. Lando, a friend of lex mercatoria, cites the same provision in the German Civil Code but in a somewhat different example.42 Suppose, he writes, a Scandinavian seller had sold goods to a German buyer which are supposed to have a long life, and the buyer notices the defect after both the German (6-month) and the Scandinavian (l-year) periods of limitation have run.43 An arbitrator might say, 'I need not decide between the German and the Danish law, because under either one the claim is barred.' Alternatively, the arbitrator might determine that both the German and the Scandinavian statutes were designed for internal transactions only, and might look to the Vienna Sales Convention for guidance in applying lex mercatoria. That Convention provides (in Article 39) that the buyer loses the142 right to rely on a lack of conformity of the goods if he does not give notice to the seller within a reasonable time, but in any event not more than two years from handover to buyer. An amiable composition might decide that if the defect were not discovered until three years after delivery it would be unfair to deprive the buyer of his right to claim against the seller. An arbitrator deciding according to lex mercatoria, i.e., according to law, could not permit the claim if it were brought more than two years after delivery, even if he thought it would be fairer to do so.

Similarly, to go back to the case of the dispute between the French and the Austrian parties, arbitrators prepared to employ lex mercatoria but not authorized to act as amiables compositeurs could not determine that the Parties had made an agreement to arbitrate by telephone without any signature, because the international usage, reflected in the New York Convention, calls for an agreement in writing, signed by the parties.44

I want to give just one more example, again from my own experience, where I believe lex mercatoria was applied, though the arbitrators saw no need to advertise that fact. The dispute grew out of a large construction project in a Middle Eastern country, and in particular out of a sub-contract between the prime contractor, the European subsidiary of a major American corporation and a Scandinavian sub-contractor. The sub-contract provided for arbitration in Geneva, but provided that it was to be governed by the substantive and procedural law of New York. There were some 32 separate disputes relating to alleged failure by the prime contractor to make timely payments, and counterclaims that the sub-contractor had failed to perform a number of specified tasks on schedule. There were also three general questions of law applicable to all or most of the individual disputes, and the Tribunal, with consent of the parties, decided to take these up first, with a view to issuing a preliminary award. Two of the questions concerned interpretation of the contract; the third question, and the one relevant to the present discussion, concerned the payment of interest. Some of the 32 disputes might well be decided in favour of the claimant subcontractor, and liability for some other claims was conceded but payment had been withheld, as is so often the case, once the dispute flared up. The subcontract contained no provision concerning interest.

The respondent's first position was that failure to provide for interest in the sub-contract was consistent with the principal contract, which also had no provision for interest, and with the (Islamic) law of the country where the project was being built. Accordingly, the respondent requested a ruling that no interest at all should be awarded for such sums as the Tribunal found to be payable. As a second position, the respondent argued that if any interest was to he awarded, it should be at New York's statutory rate, which at the relevant 143 time was 6 per cent per annum.45 The claimant, on the other hand, argued (i) that it was unheard of that in a construction contract between two western parties calling for payments at specified times, delay or default in making payments should bear no cost; and (ii) that interest should be at the London Interbank Offered Rate (LIBOR), which reflected the claimant's cost of funds and which at the time stood at about 15 per cent per annum.

The arbitrators, an Englishman, a Canadian, and an American, quickly agreed that interest should be awarded: if payments were found to have been due in Year I, to order them made in Year III or IV without interest would be wrong - commercially unreasonable and therefore contrary to law. Dr Mann and Lord justice Mustill might well ask 'contrary to what law?', and if pressed the arbitrators would have said 'contrary to the law reflecting the normal expectations and usages of enterprises engaged in international construction projects.'

The more difficult question for the arbitrators concerned the appropriate rate of interest. After a good deal of discussion, the arbitrators decreed in their interim award that neither the statutory New York rate, nor the London Interbank rate would be applicable: if it became necessary to render a final award after hearing the individual disputes, the arbitrators would choose a figure in between, possibly after taking further evidence; meanwhile, leaving some flexibility within the guideline stated might make it easier for the parties to reach a global settlement. In fact the Parties did so even as the arbitrators were packing their bags for a summer of hearings in Geneva. The arbitrators never found out what the final terms were, and whether a precise interest rate was fixed; they were satisfied, however, that they had made a correct decision -correct at law, i.e., the law applicable to international transactions of this kind. Of course, if the arbitrators had found that under New York law, the 6 percent rate was mandatory, the decision would have been wrong. But the arbitrators found that the statutory interest rate was mandatory only for judgments of New York courts,46 and thus that there was a gap for them to fill. Perhaps, like Molière's hero, who did not realize that for forty years he had been talking prose,47 the three arbitrator s like hundreds of international arbitrators before them, did not realize that they were applying lex mercatoria. But I believe the decision was correct, legally and mercantilely, as was confirmed by its acceptance by both parties.48



(a) Is Lex Mercatoria a Complete System?

Mustill writes:

... [T]he rules of the lex mercatoria have a normative value which is independent of any one national legal system. The lex mercatoria constitutes an autonomous legal order.

49 With all respect, my view, derived from the cases mentioned and numerous similar ones,50 is quite different.

In each of the cases I have mentioned, lex mercatoria supplies the solution to a particular issue, without necessarily governing all aspects of the transaction. In the case of the Danish seller and the German buyer, Danish law may well govern the scope of the warranty, the effect of any disclaimer of liability, and the measure of damages51 only the rule not fit for international trade, i.e., the immediate notice rule, is replaced by lex mercatoria. In the case of the contract that may or may not have been effectively signed , if it were held that a contract had been formed, the scope of the obligations, excuse for non-performance, waiver, damages, and any other issues that might arise would be determined according to the law chosen by the parties, there the law of New York. In the case of the construction sub-contract, where formation was not an issue, the other legal questions, including the respective liabilities of prime- and subcontractor for defaults of the employer, were decided upon briefs and arguments under the law of New York, and if evidentiary hearings had been held, New York's rules of evidence would have been applied.

Thus, in my view, the criticism that lex mercatoria purports to be a complete system and is unable to sustain the claim falls wide of the mark. My concept of lex mercatoria, and here perhaps I stray from Professor Goldman, at least in his more recent writing,52 is not that of a self-contained system covering all aspects of international commercial law to the exclusion of national law, bot rather as a source of law made up of custom, practice, convention, precedent, and many national laws. Thus lex mercatoria as I see it can furnish an alternative to a conflict of laws search which is often artificial and inconclusive, and a way out of applying rules that are inconsistent with the needs and usages of international145 commerce and that were adopted by individual states with internal, not international, transactions in mind.

Mustill writes further:

... the purpose of the conflict sit laws is to enable the tribunal accurately to identify the national law which governs the contract. This is precisely antithetical to the premise of the lex mercatoria, which is that the arbitrator's first step is to reject any national law as the governing law?

53 My experience is that it is not the arbitrator's first step to reject any national law as the governing law. In most cases, even where legal questions are or may be decisive, the choice of law is clean, either because the parties have unequivocally supplied it or because one State plainly has the closest and most real connection (to use the English formulation54 ) with a particular state.55 Lex mercatoria comes into play when no state clearly has the most significant relation with the contract or the issue in controversy (to shift to the American formulation56 ), where the state to which the conflict of laws principles point has no clear answer to the question in controversy ,57 or where the answer is clearly out of line with the reasonable expectations of the parties and the business community as a whole -where, to repeat Lando's phrase,58 the rule is not fit for international trade. To put the same point differently, I reject the suggestion that the first step of the arbitrator prepared to accept lex mercatoria is to cast aside all national law. My view is that lex mercatoria, derived from the law of many states and international conventions (in both senses of that word), is a useful tool for arbitrators (and I would add for judges) when faced with unsatisfying answers from their initial inquiries. It is, in other words, an additional option in the search for the applicable law, not an alternative to that search.59

(b) Is Lex Mercatoria Universal Law?

'Is there,' Mustill asks, 'a single useful generalization that can be made about the commercial law if, say the member states of the United Nations?' 'If not,' he goes on, 'then where do the mercatorists identify the fountain-head of this new international juristic order?'60 With all respect, I submit that the criticism that lex mercatoria cannot in the late twentieth century claim to be a universal system,146 as it may have been in ancient Rome,61 falls wide of the mark. In a transaction between a Swedish and an American firm, or between an Austrian and a French firm, it seems to me quite irrelevant whether or not the custom or understanding asserted is limited to the western industrial countries. That question may well be relevant in relation to a transaction between parties from two non-industrial countries, or even with one party from a developed, the other from a non-western country or from a less developed country whose legal system cannot fairly be traced to that of one of the colonial powers. In such circumstances a given principle may well not Support the justified expectation of the parties, and thus not meet the definition of lex mercatoria. Universality, in other words, depends on the universe. I find nothing illogical in saying that the German buyer in Professor Lando's case who will not be bound by a quirky and unexpectedly severe time limit under Danish law will not be similarly protected if he contracts with a Soviet Foreign Trade Organization or a Nigerian state enterprise. The ordentlicher Kaufmann (reasonable merchant) simply undertakes different burdens when he trades outside his universe (at least if he does so for the first time), and lex mercatoria recognizes this truth. Of course as the world, having split apart, draws together again through UNCITRAL conventions and model laws, widely accepted codifications such as the Uniform Customs and Practice for Documentary Credits,62 and treaties such as the New York Convention an Recognition and Enforcement of Arbitral Awards,63 international commercial law may not be as balkanized as Mustill suggests. Still, the point is that lex mercatoria does not depend on proof of universality, but on the other hand is itself a factor (not to say a force) in the direction of a common law of international private trade, using the work 'common' in its original sense, before it became a naturalized English citizen.

(c) Is Lex Mercatoria Dangerous?

Unlike some of my continental colleagues, I have never sat on a case in which the parties chose lex mercatoria to govern their contract. But agreements without a choice of law clause are common in my experience -and there is usually no way of knowing the reason for the omission. It might have been because the parties did not think about choice of law when they made their contract, or it might have been because they (or their lawyers) discussed the topic but could not agree, but the disagreement was not sufficiently important to kill the deal. Because we do not know the reason for the omission, I would not draw an inference that the parties intended to withdraw their relationship from national147 law,64 or that they meant that lex mercatoria should apply.65 The fact that they provided for arbitration under the auspices of an international institution such as the I CC does not, in my view, lead to a different conclusion. Here I agree with Mustill. But if such a contract comes before arbitrators and the controversy meets the other condition, i.e., no national law is clearly the lex solutionis, it may well be a good candidate for application of lex mercatoria in whole or in part.

I do not disagree with Mustill in his denunciation of abuses. He puts the question of a contract that expressly stipulates a choice of governing law and asks (or rather has a hypothetical client ask; whether the arbitrator not acting as amiable compositeur can properly apply the lex mercatoria in preference to the chosen law. 'The answer', he writes, 'must surely'66 I agree. The arbitrators cannot rewrite the contract, and if the contract directs them to apply a given law, they must do so. Even then, however, lex mercatoria intelligently used, can, I submit, be useful. Faced with an unclear statute or judicial precedent, an arbitral tribunal would be justified, I believe, in interpreting the national source so as to be consistent, and not inconsistent, with generally accepted international understanding and usage, or so as to be applicable only to internal commerce.67

A good illustration is suggested by one of the expert opinions in the case earlier discussed concerning the formalities needed to conclude an agreement to arbitrate. Suppose in that case it had been clear that the law of Patria was the law chosen by the parties; the respondent shows that Patrian judicial decisions have upheld rules imposing particular formalities an agreements to arbitrate; the claimant replies that these decisions all pre-date the New York Convention, that they all concerned disputes between citizens of different provinces of Patria, and that when Patria ratified the New York Convention there was no discussion of special local requirements. It would seem to me not unreasonable for an arbitral tribunal passing on its own jurisdiction in these circumstances to uphold an agreement meeting the general, but not the special formal requirements, and I would hope that if the resulting award came before Lord Justice Mustill he would uphold it as well. 68 148

(d) What Does Lex Mercatoria Say?

Lord justice Mustill asks with some justification what are the rules of lex mercatoria. He proposes a sample of twenty such rules, drawn directly or indirectly from past arbitral awards.69 As I have not read most of the awards he builds on, and I am not aware that any of the cases in which I have participated as arbitrator have been published.70 I think, however, the statement of principles is fair, and I cannot do better than to reproduce some excerpts (slightly edited) from Mustill's abstract:71


Contracts should prima facie be enforced according to their terms (pacta sunt servanda).


A contract should be performed in good faith.


A State entity cannot be permitted to evade the enforcement of its obligations by denying its own capacity to make a binding agreement to arbitrate , or by asserting that the agreement is unenforceable for want of procedural formalities to which the entity is subject.


If unforeseen difficulties intervene in the performance of a contract, the parties should negotiate in good faith to overcome them, even if the contract contains no revision clause.


One party is entitled to treat itself as discharged from its obligations if the other hat committed a breach, but only if the breach is substantial.


No party can be allowed by its own act to bring about a non-performance of a condition precedent to its own obligation.


A party which has suffered a breach of contract must take reasonable steps to mitigate its loss.


A party must act promptly to enforce its rights, on pain of losing them by waiver.


Contracts should be construed according to the principle ut res magis valeat quam pereat.


Failure by one party to respond to a letter written to it by the other is regarded as evidence of assent to its terms.

Mustill regards his list of twenty principles, of which ten are here reproduced, as a rather modest haul. I suppose that could be said of the Ten Commandments or of the American Bill of Rights as well. But the excerpts here quoted give a quite clear picture of the thrust of a modern international commercial law, and at least three of' the principles quoted bring to mind other cases with which I am familiar. No case in which I have participated went against these principles, though of course some cases raised issues not covered by the ten principles quoted or the twenty on Mustill's list.72 I would add just one other principle, stated by Goldman73 and reflected in many standard form contracts and some149 arbitral awards. Classical contract law seems to hold that in the face of an unforeseen event, a party is either completely discharged from its obligation or is fully liable.74 Lex mercatoria seems inclined to postpone performance for the time of an interruption while preserving the underlying obligations, or to apportion losses from an occurrence that neither party caused or could have foreseen. That too, I think, substitutes commercial reasonableness for rigidities more appropriate to other times and places.75

If it is objected that the principles reflect the common (i.e., Anglo-American) law generally, or reflect the civil law generally, or both, that seems to me to be support, not criticism. Lex mercatoria is not, in other words, supposed to be revolutionary. What it does do, if properly used, is to clarify, to fill gaps, and to reduce the impact of peculiarities ofindividual countries' laws, often not designed for international transactions at all.76

At the end of the day, one must ask whether those participants in the international economy who choose arbitration as their mode of dispute settlement have simply made a choice of forum, with the expectation that that choice has no effect on the substance of any dispute that may arise, or whether some difference in the source of law as well as in the decision-making process is part of the understanding of those who choose arbitration. My view is that a decision to include an agreement to arbitrate in an international contract is not analogous to a decision in the United States to bring a state-created civil action in federal rather than in state court, knowing that (at least in theory) the federal court sitting in Boston and the Massachusetts state court are to apply the same body of law in the same way.77  To pursue the imperfect analogy one step further, whereas the United States since 1938 hat opted for vertical conformity,78 international arbitration and lex mercatoria aim for horizontal conformity. Whatever errors and distortions followed from justice Story's famous opinion in Swift v. Tyson,79 I believe he was right in his view of the 'general principles and 150 doctrines of commercial jurisprudence'.80 'The law respecting negotiable instruments' he wrote 'may be truly declared in the language of Cicero, adopted by Lord Mansfield ... to be in a great measure, not the law of a single country only, but of the commercial world. Non erit alia lex Romae, alia Athenis, alia nunc, alia posthac; sed et apud omnes gentes, et omni tempore, alia eademque lex obtinebit.'81

Of course much has changed since Story's time; indeed when the financial world thinks today of Swift it thinks not of the great Scholar/ Justice, but of electronic funds transfers.82 But the vision of 'general principles and doctrines of commercial jurisprudence', I would hope, retains its compelling character.

Within the United States, it is not an exaggeration to suggest that the Uniform Commercial Code saved internal (i.e., what Americans call interstate) commerce from the balkanizing tendencies of Erie R. Co. v. Tompkins.83 For international commerce, we have only fragments of a written uniform commercial code as the century comes to an end. But an unwritten law understood among merchants does exist, and goes a long way to filling a comparable need.

Arbitrators do not make up the law as they go along, and contrary to popular view, most arbitrators do not look first of all for compromise.84 International arbitrators do seek to achieve just results within a legal framework, and that framework is by definition wider than the frontiers of any state. To me, this is the vision, the promise, and the usefulness of lex mercatoria.

Charles L. Denison Professor of Law, New York University School of Law. This article is adapted from a paper delivered to the Colloquium on "The New Law Merchant From a Legal and Arbitral Perspective" at the Eason Weinmann Center for Comparative Law, Tulane University School of Law, November 10, 1989.
For the text of the Model Law, see Report on UN Comm. on International Trade Law on Work of its 18th Sess. (3-21 June 1985) and Annex thereto, 40 GAOR Supp. No. 17, U.N. Doc. A/40, 17 (1985). For a summary of the debates, see U.N. Doc. A/CN. 9/SR. 305-333 (one volume 18 Dec. 1985).
Rogers, Contemporary Problems in International Commercial Arbitration, 17. Int. Bus. L. 154, 156. (1989).
To spell this thought out would be a diversion from the present topic. But if one looks, for instance, at the percentage of road accidents that result in litigation, the American and French figures are quite close, while the English and German figures are substantially lower. For some numbers, see A. Lowenfeld, Aviation Law, ch. 7, pp. 7-17 and sources there cited (2nd ed. 19851).
The change achieved by the Arbitration Act 1979 has two principal aspects. First, the 'case stated' procedure is abolished and appeal to the courts requires leave of the court (s. 1.); second, parties to an international arbitration agreement may by agreement exclude all judicial review (s. 3). Exclusion agreements may not, however, be made with respect to claims in admirals or relating to insurance or commodity contracts unless these are expressly governed by law other than English law (s. 4), and they may not be made with respect to domestic arbitration agreements. A domestic arbitration agreement, on this purpose, I think, means an agreement providing for arbitration in the United Kingdom between parties all of whom are nationals or residents of the United Kingdom (I say 'I think" because s. 3.7) which contains the definition of 'domestic arbitration agreement', contains five consecutive negatives and defies confident interpretation); an arbitration agreement between a British party and a foreign party or between two or more foreign parties may, it appears, be subject to a exclusion agreement if it passes the other tests. For confirmation by one of its architects that the reasons for the 1979 reforms 'were purely logistical and commercial,' see Lord Justice Kerr, 'Commercial Dispute Resolution: The Changing Scene," in Liber Amicorum for Lord Wilberforce 111, at p. 134 (1987).
Switzerland, a favorite situs for international arbitration, until recently provided substantial scope for judicial actions for annulment of arbitral awards, particularly on questions of jurisdiction. See International Arbitration Convention of August 27, 1969 (The Concordat), Art. 3, 9, 35. Under Articles 190-192 of the new Swiss Private International Law Code, effective Jan. 1, 1989, the scope of judicial review is substantially reduced, and may be eliminated entirely by an exclusion agreement, provided none of the parties is domiciled or established in Switzerland.
The leading proponent of lex mercatoria in the past quarter century has been Berthold Goldman, notably in his Hague Lectures of 1963, 'Les Conflits de Lois en Matière d'Arbitrage international de Droit Privé. 109 Hague Academy International Law, Recueil des Cours, 347 (1952-II), and in two major articles, Frontières du Droit et lex mercatoria, 9 Archives de Philosophie du Droit 177 (1964) [hereafter 'Frontières du Droit']; and 'La Lex Mercatoria dans les Contrats et l'Arbitrage Internationaux: Réalité et Perspectives,' 106 J. du Droit International [Clunet] 475 (1969) [hereafter 'Réalité et Perspectives']. For a critical view of lex mercatoria by one of France's leading conflict of laws scholars, see Lagarde, Approche Critique de la Lex Mercatoria, in Le Droit des Relations Économiques Internationales: Études Offertes à Berthold Goldman, at p. 125.
There is of course still another possibility - deciding on the choice of law only when it is clear that the law of all the possible jurisdictions in question is different on a point critical to the outcome of the arbitration. For the present author's discussion of this option - not inconsistent, I believe, with the position taken in this paper, see Lowenfeld, 'The Two-Way Mirror: International Arbitration as Comparative Procedure' 7 Mich. J. Legal Studies 163, 180-81 (1985), reproduced in Craig, Park and Paulsson, International Chamber of Commerce Arbitration, App. VII, p. 187 at pp. 204-05 (1986).
For a brief statement of his views, see Mustill, Contemporary Problems in International Commercial Arbitration: A Response [to Mr. Justice Roberts, note 2, supra], 17 Int. Bus. L. 161 (1989) [Hereafter 'Contemporary Problems']. For a full-length exposition by the same author, see Mustill, The New Lex Mercatoria in M. Bos and I. Brownlie, eds. Liber Amicorum for Lord Wilberforce 149 (1987) [Hereafter 'The New Lex Mercatoria], reproduced in 4 Arb. Int. 86 (1988).
See, notably, Deutsche Schachtbau - und Tiefbohrgesellschaft m.b.H. v. The Government of R'as Khaimah National Oil Co. [1987] 3 W.L.R. 1023, 1031-35 [1987] 2 All E.R. 769 (C.A.), reversed in part (but not affecting enforceability of award rendered in Switzerland), [1988] 3 W.L.R. 230 (H.L.(E.)).
Mustill, Contemporary Problems, note 8, supra, at p. 162.
See, generally, Mustill and Boyd on Arbitration (1982).
Mustill, Contemporary Problems, note 8, supra, at p. 163; to the same effect, The New Lex Mercatoria, at p. 160, 4 Arb. Int. at 96.
Contemporary Problems at p. 163. The topic is given extended discussion in The New Lex Mercatoria, at pp. 155-59, 4 Arb. Int., at pp. 92-96.
407 U.S.A. (1972).
Indeed that was the main basis for the dissent by Justice William O. Douglas, who expected the English court to uphold exculpatory clauses in the towing contract (almost certainly correct), and assumed that such clauses would be disallowed in the United States, which is not so clear. See 407 U.S. at p. 13, n. 15, majority opinion: pp. 23-24 (Douglas, J. dissenting).
See, to this effect, Lord Mansfield's statement in Luke v. Lyde, 2 Burr. 883 at p. 887, 97 Eng. Rep. 614 at p. 617 (1759), followed by the famous Latin refrain quoted by Justice Story in Swift v. Tyson. See text at note 81, infra.
See, e.g. G Gilmore and C. Black, The Law of Admiralty, pp. 3-11 (2d ed. 1975), and sources there cited.
See. e.g. Cie. Tunissenne de Navigation, S.A. v. Cie d'Armement Maritime S.A. [1971] A.C. 572, modifying the absolute reliance on the maxim 'qui elegit judicem elegit jus' but confirming the maxim as a strong presumption.
Lando, The Lex Mercatoria in International Commercial Arbitration, 34 I.C.L.Q 747 at p. 748 (1985).
Mann 'Lex Facit Arbitrum,' in P. Sanders, ed., Arbitration: Liber Amicorum for Martin Domke , at p. 157 (167).
Lando, note 20, supra, at p. 752.
Ibid, at p. 753.
For English version, see 1961 Y.B. of International Institute for Unification of Private Law (Unidroit) 203, at p. 208 (1962).
Signed in vienna April 11, 1980. Entered into Force Jan. 1, 1988.
Both the Federal Republic of Germany and Denmark signed the Vienna Convention . Denmark deposited its instrument of ratification on February 14, 1989, as at September 1989, the Federal Republic had not yet done so. Under Article 1 (1) (b), if the places of business of the contract partners are not both parties to the Convention, application of its provisions depends on whether the rules of private international law lead to application of a law of the Contracting State. Article 95 of the Convention permits states to opt out of Article 1 (1) (b); the United States and the People's Republic of China have done so.
A fair question, put to me by my colleague Professor Linda Silberman of New York University School of Law, is whether lex mercatoria assists in solving the reverse situation - e.g., where the seller is German and the buyer is Danish. My answer (and I presume Lando's) is that the same considerations apply, that is that the German seller, as defendant, should not be able to take advantage of a law out of line with the general expectation of merchants, even if for other purposes Danish law governed some or all of the contract. Of course if the parties expressly agreed on Danish law, including specifically the rule concerning the short notice period, that rule would no longer be unexpected and should be applied, just as if set out verbatim in the contract of sale. If the sale is between two countries with the same rule - say a Danish seller, Norwegian buyer - there is no reason to resort to a mercatoria, as a uniform law known in both parties and intended to govern trade between them is in place.
Convention on the Recognition and Enforcement of Foreign Arbitral Awards
The relation between Cantonal law, in general governing procedure, and federal law, in general governing substantive law, and in which category conflict of laws falls, is just another problem that might have had to be faced. This subject, quite difficult for an outsider to grasp, has now been removed by passage of the (federal) Law on Private International Law of Dec. 18, 1987, eff. Jan. 1, 1989. For an English translation with commentary, see e.g. P. Lalive, 'The New Swiss Law on International Arbitration,' 4 Arb. Int. 1 (1988).
Article 13 (3) of the ICC Rules states: '...In the absence of any indication by the parties as to the applicable law, the arbitrator shall apply the law designated as the proper law by the rule of conflict which he deems appropriate.'
Compare J. Lookofsky, Consequential Damages in Comparative Context, p. 192 and note 553 (1989).
I exclude for present purposes the possibility that we reached the wrong result.
For instance, whether deals are normally made by telephone, or by exchange of particularized documents, or by references to standard forms.
Accord, I believe, Goldman, The Applicable Law: General Principles of Law - The Lex Mercatoria, in J. Lew, ed., Contemporary Problems in International Arbitration 113, at p. 116 (1986). In his earlier writings, Frontières du Droit, supra, note 6, at p. 189, Réalité et Perspectives, supra, note 6, at pp. 479-80, Goldman seems to exclude questions of capacity and consent to be bound from his collection of principles. For the view that (except within the arbitral world itself) questions of contract formation belong exclusively to jurisdiction of courts, se Lagarde, supra, note 6, at pp. 147-48.
I am not here talking, for instance, about aircraft accidents, in which it is common for carriers to rely on low limits of liability, regardless of the loss of life and income involved.
I do not discuss here the problem much discussed in the literature, of 'compétence de la compétence'. In the case I have described, the parties agreed that in the first instance, the arbitrators would have to decide whether they had jurisdiction, which here depended on whether a contract had been formed. There was, of course, no guarantee that the unsatisfied party might not have appealed to the courts. In the event, the case was selected before the award was transmitted to the parties.
Presently Limitation Act 1980, s.5.
Fed. Rep. Germany BGB (Civil Code) § 477.
Mann, Lex Facit Arbitrum, supra, note 21.
It is worth a footnote to point out that the United Kingdom recently adopted the Foreign Limitation Periods Act 1984, the effect of which is to apply the limitation period of the proper law governing the transaction. Thus an English judge would be required to make a choice of law decision before determining whose limitation period to apply.
See note 26, supra, for discussion of the provision for application of the Convention when only one of the states in question is a party.
Lando, supra, note 20, at p. 755.
I am grateful to my colleague and former student, Professor Joseph Lookofsky of the University of Copenhagen, for pointing out that only Denmark of the Scandinavian countries retains the one-year rule from the original Scandinavian Sale of Goods Act drafted in the first decade of this century. Both Sweden and Norway have recently adopted a two-year rule for notice of defects, and Finland has no special rule on the subject. Of course the more unique Denmark's rule is, the more persuasive is Lando's argument for not applying is to an international transaction.
Article II (1) (2). Whether an agreement to arbitrate signed by one part, for instance in an order confirmation, but not countersigned by the other party, is sufficient is a matter of considerable doubt, and there is probably insufficient consensus to form the basis of a rule of lex mercatoria.
N.Y. Civil Practice Law and Rules §5004. The rate was changed to 9 per cent by N.Y. L. 1981, ch. 258, effective June 25, 1981.
See 5 J. Weinstein, H. Korn, A. Miller, New York Civil Practice, §5001.06 (1988).
Molière, Le Bourgeois Gentilhomme, Act II, sc. 4.
For a recent article illustrating some of the pitfalls and arbirtariness of a different approach, see Hunter and Triebel, Awarding Interest in International Arbitration, 6. J. Int. Arb. 7 (1989) .
Mustill, the New Lex Mercatoria, supra , note 8, at p. 151, 4 Arb. Int. at 38.
For other illustrations from my own experience, see The Two-Way Mirror, supra, note 7.
That would be the solution under article 3 of the Hague Convention of 1955 on the Law Applicable to International Sales of Goods, 510 U.N.T.S. 147. As it happens, Denmark is a party to the convention, but the Federal Republic of Germany is not, at January 1989.
Compare Goldman, 'Frontières du Droit' (1961), supra, note 6 at p. 189, speaking about constitutive elements of lex mercatoria 'although it does not form an entirely autonomous system,' which Goldman 'Realité et Perspectives' (1979), supra, note 6, at pp. 499 and 502, undertaking to show that 'lex mercatoria carries out well the function of a system [assemble] of rules of law.' The contrast is pointed out by Lagarde, supra, note 6, at p. 130. However, Goldman also writes that the difficulty in establishing a hierarchy between national law and lex mercatoria 'does not prevent the existence of lex mercatoria and national law, nor their combined application, when this seems appropriate': Ibid, at p. 498.
The New Lex Mercatoria, supra, note 8, at p. 154, 4 Arb. Int. at p. 90.
See, e.g. Dicey and Morris on the Conflict of Laws, Rule 180 (11th ed. 1987).
Professor von Mehren suggests that in time a lex mercatoria of conflict of laws may develop. Von Mehren, To What Extent is International Commercial Arbitration Autonomous? In Le Droit des Relations Économiques Internationales: Études offertes à Berthold Goldman, 217 at p. 226-27.
See, e.g., Restatement (Second) of the Conflict of Laws, §188 (1971).
I had a case, for example, which turned on whether a particular event did or did not amount to force majeure; the arbitrators and parties were agreed that the law of seller's place of business was applicable, but try as we might we could not find any modern authoritative guidance in that state's statutes, appellate decisions, or authoritative commentary.
Note 20, supra.
Accord, Goldman, Réalité et Perspectives, note 6, supra, 106 J. du Droit International, at p. 482.
Mustill, Contemporary Problems, supra, note 8, at p. 163. See also The New Lex Mercatoria, supra, note 8, pp. 155-56, 4 Arb. Int. at pp. 92-93.
For a concise description, with sources, of the development by the early Romans of law more flexible and functional that the jus civile applicable to relations among citizens of Rome, see the discussion of the praetor peregrinus in F. Juenger, General Course in Private International Law, 1985-IV).
ICC Publication No. 400 8th Rev., in force as from October 1, 1984. As at 1987, banks from 144 countries have adhered to the UCP.
See note 28, supra.
Compare Goldman Frontières du Droit, note 6, supra, 1964. Archives de Philosophie du Droit, at pp. 184-85.
Again here I do not go quite as far as Professor Goldman. See Goldman, Réalité et Perspectives, note 6, supra, 106 J. du Droit International, at p. 481.
Mustill, the New Lex Mercatoria, at pp. 167-68, 4 Arb. Int., at p. 104.
Compare, at least by way of analogy, §114 of the Restatement (Third) of the Foreign Relations Law of the United States: 'Where fairly possible, a United States statute is to be construed so as not to conflict with international law or with an international agreement of the United States.' Such an issue could come before the English court on the basis of Article V (1) of the New York Convention, which permits, but does not require, refusal to enforce an award if 'the parties to the agreement [to arbitrate] were, under the law applicable to them, under some incapacity, or the said agreement is not valid under the law to which the parties have subjected it ...'
Such an issue could come before the English court on the basis of Art. V (1) of the New York Convention, which permits, but does not require, refusal to enforce an award if 'the parties to the agreement [to arbitrate] were, under the law applicable to them, under some incapacity, or the said agreement is not valid under the law to which the parties have subjected it ...'
It may well be, as suggested to me by Prof. Goldman, that Mustill is not as absolute a foe of lex mercatoria as is indicated by the excerpts of his articles here quoted. He did after all engage in a careful search which yielded twenty principles, and there is, surely, a substantial kinship between lex mercatoria and English common law, as Lord Mansfield would certainly recognize.
It is worth adding that with a few notable exceptions, such as the Libyan Nationalization cases, it is usually difficult from the denatured versions of arbitral awards as published to get the proper feel for a case. In part, of course, this is true as well of published opinions of judicial decisions, but the tradition of confidentiality in arbitration exacerbates the problem.
The New Lex Mercatoria, at pp. 174-177, 4 Arb. Int., at pp. 110-14.
For another statement of principles drawn from ICC awards, see Craig, Park, and Paulsson, International Chamber of Commerce Arbitration , ch. 35 (1984).
Réalité et Perspectives, note 6, supra, 106 J. du Droit International, at p. 488.
See, e.g., ALI Restatement (Second) of contracts, §§ 261, 264, 265 (1981); German Civil Code  §§275, 285; Tsakiroglou & Co. Ltd. V. Noblee Thorl GmbH [1962] A.C. 93.
This principle is, of course, akin to Mustill's principle no. 9, in recognizing the problem of unforeseen difficulties, but is, I think, different in that it seeks to preserve but postpone the contractual relationship, or to allocate the loss.
Keith Highet, not a supporter of lex mercatoria, suggests that principia mercatoria would be a better term, Highet, The Enigma of the Lex Mercatoria, 63 Tul. L. Rev. 613, 616, 628 (1989). There is some appeal to his point, though I believe the use of the word 'lex' is valuable. In any event, it is clearly too late to change the name of the concept so widely described if not always understood.
Compare the following excerpt from the U.S. Supreme Court's opinion in Mitsubishi Motors Crop. V. Soler Chrysler Plymouth, Inc., 473 U.S. 614, 629 (1985): '... [W]e conclude that ... respect for he capacities of foreign and transnational tribunals, and sensitivity to the need of the international commercial system for predictability in the resolution of disputes require that we enforce the parties' agreement [to arbitrate], even assuming that a contrary result would be forthcoming in a domestic context.'
I refer of course to the landmark decision of Erie R. Co. V. Tompkins, 304 U.S. 64 (1938). See note 83 infra, for the briefest of explanations.
41 U.S. (16 Pet.) 1 (1842).
Ibid, at p. 19.
Swift is the acronym for the Society for Worldwide Interbank Financial Transfers, the mechanism for computer and satellite-transmitted transfers of funds among the world's major financial institutions.
304 U.S. 64 (1938). For readers not familiar with American federalism, the case may be condensed into the holding that, absent a federal statute, federal courts hearing controversies between citizens of different states (or between a citizen and an alien) must apply the law that would be applied by the courts of the state where they sit, rather than any general federal common law. The Uniform Commercial code, completed in the mid 1950's, has now been adopted (with some omissions in Louisiana) in every state and the district of Columbia.
I do not mean to suggest that three arbitrators may not try to achieve a blend of three individual views, as do juries, as well as courts with more than one judge, and committees commissions of all kinds. My point is that they do not say 'if we hold for claimant on issue a, we have to hold for respondent on issue b, or we have to reduce damages...'.

Referring Principles
A project of CENTRAL, University of Cologne.