HANS SMIT*

International commercial arbitration is growing by leaps and bounds. Both in institutional and ad hoc arbitrations are proliferating. In these arbitrations, the tribunals are most frequently composed of arbitrators from different countries with different legal training, background, and experience. The lawyers handling the cases before these international tribunals are also different from those appearing before domestic tribunals. They not only practice before the courts of their own countries, but, increasingly, they are likely to possess experience in processing cases before international, arbitral tribunals.

Indeed, arbitrators are likely to be selected from a rather limited group, especially in the more important cases. The institutions appoint them because, quite properly and understandably, they wish to appoint persons with proven track records. Similarly, the parties refer to appoint arbitrators about whose performance information can be gained. The usual selection process, under which each party appoints an arbitrator and the1310two party-appointed arbitrators or an institution select the chairman,1 permits the institutions and the parties to take these factors into account.

These developments have had a marked effect on the law practiced in international arbitral tribunals, both on the procedural and the substantive law level. Unfortunately, since most proceedings before, and awards rendered by, international arbitral tribunals remain unreported, much of what has been happening is neither generally known nor accessible to the legal profession at large.2 As a consequence, this Article reflects in large measure my personal experiences as an international arbitrator.

Existing rules on much of the writings about arbitration are premised largely on the notion that in international arbitration the normal choice of law rules apply: The tribunal must determine the applicable law by reference to appropriate conflict of laws rules, and the apply the law to which the conflict of laws rules refer fit. This is true as to both procedure and substance. On the procedural level, the tribunal must apply the rules chosen by the parties, whether they be institutional rules, rules laid down in the contract by the parties, or the procedural law of any given state. In the absence of a choice of law by the parties, the tribunal must apply the rules referred to by the appropriate conflict of laws rules.3 On the substantive level, the tribunal again must apply the rules chosen by the parties or the rules indicated by the appropriate conflict of laws rules.

Qualifications to the traditional approach are recognized sparingly. Some of the institutional rules, such as those of the International Chamber of Commerce International Court of Arbitration, free the arbitrators from application of the choice of law rules of a particular national system, including those of the1311place of arbitration, and authorize them to apply such choice of law rules as they may deem appropriate.4 However, institutional rules and parties, in their contracts, generally do not permit arbitrators to apply rules, of substantive law they deem appropriate.5

Regarding procedure, arbitrators enjoy considerably greater flexibility. Institution rules of procedure regulate procedure to a certain extent, but generally leave the regulation of procedure to the parties or the tribunal. Even rules of institutions such as the ICC International Court of Arbitration, which involve the institution as far as possible in the actual arbitration process, leave the tribunal a very large measure of freedom. Experienced arbitrators normally seek the parties consent to having the tribunal determine how the case is to be processed. National arbitration laws are of little help, as they typically refrain from regulating the procedural aspects of arbitration beyond some basic rules, such as those providing for proper notice and a reasonable opportunity to be heard. It is therefore fair to conclude that procedure is left largely to the tribunal and the parties.

On the whole, little attention has been given to the manner in which international arbitral tribunals use this freedom. The confidential character of most international arbitrations is in large measure responsible for this phenomenon. However, as might be expected in light of the composition of international arbitral tribunals and the bar practicing before them, the rules of procedure applied in international tribunals frequently do not follow the pattern of any international law system. Instead, international arbitrators are developing a procedural system of their own.

Perhaps surprisingly to some, the same development is occurring on the substantive level. Existing rules generally require arbitral tribunals to apply, on matters of substance, the law of a particular nation. In practice, however, the rules of substantive law applied by such tribunals may very well not be identical to those applied by the ordinary courts in the country whose laws are, theoretically, applicable.

The explanation for these developments is both simple and obvious. When lawyers with different professional training, educational background, and professional experience are brought1312together to litigate or adjudicate on international dispute, they bring to the exercise a desire to reach a solution acceptable to the persons that select them, rather than a dispassionate zeal to follow the dictates of the presumably applicable law. Furthermore, they are likely to approach the issues they confront in the manner in which they customarily approach legal problems. That manner is, of course, heavily influenced by their educational background and professional experience, which is rooted in their own legal system. The necessary consequence is that such lawyers will seek legal solutions that are compatible with their legal systems and are developed by recourse to legal techniques and approaches they have been trained to use. Even if the participating lawyers diligently try to understand and apply the applicable foreign law, they are likely to come up with a foreign law that is heavily colored by approaches and techniques that are not those of the lawyers practicing in the country whose law they apply.

In addition, international arbitrators must contribute to the proper operation of international business. Application of ideosyncratic provisions of foreign law, or of foreign law provisions addressed to purely domestic conditions, does not serve the purposes of international business.6 International arbitrators are therefore likely to seek ways of avoiding the application of such provisions. Furthermore, international arbitrators are under the same pressure as all judges to mold and shape the law to fit the purposes of the community they serve. Particularly important in this context is that many international transactions, in their size, their financing, the parties involved, the nature of the transaction, and the procurement involved, are different from those that domestic law is designed to regulate. International arbitrators, therefore, confront the same task that judges traditionally undertake when they must evolve the law to meet changed conditions. This new lex mercatoria is receiving increased attention. Where it fits in the traditional scheme of things, however, is not always clear.7

What is clear is that in international arbitration a new legal order is in the process of taking shape. It is an order that draws strong inspiration from, but is at the same time different from, national legal systems. Because of the absence of consistent1313reporting of the decisions of international arbitral tribunals, much of the creative efforts of these tribunals is not generally known. Some of its elements, however, can be gleaned from incidental reports of decided cases and the writings of authors involved in actual international arbitrations. The following is a brief sketch.

Common and civil-law procedure differ in essential respects. Typically, in a civil-law system, much of the case is developed in the pleadings or other submissions, which deal with allegations of ultimate facts, evidentiary materials, and legal theories. The prevalence of the requirement that denials must be reasoned or risk being disregarded makes it impossible to practice the skeletal pleading that is the hallmark of American procedure. In a real sense, the whole case, in all of its aspects, is developed in the pleadings. Pretrial discovery in the American sense is unknown.

The pleadings identify the issues on which testimonial proof may be needed (documentary proof is normally submitted with the pleadings). Pursuant to an interlocutory decision specifying such proof, witnesses are examined by the court, which dictates a summary of the testimony into the record.

Under Anglo-American practice, the pleadings are skeletal in nature; their principal purpose is to define ultimate issues. Information on facts and law is gathered in pretrial discovery, utilizing a large number of procedural devices, including interrogatories, examinations of witnesses before trial, document production, physical and mental examinations, and pretrial conferences. The trial is a complete presentation of all testimonial, documentary, and other proof, and also accommodates legal arguments. The trial is conducted by counsel of the parties, the examination and cross-examination of witnesses is recorded verbatim by a court reporter, and the judge's role is much more limited than it is in a civilian court.

When arbitrators and practitioners conduct an international arbitration, the are likely to follow a procedure that combines elements from their own procedural systems. This is especially probable when the arbitrators come from both civil and common-law countries. When they all come from the same1314sort of legal system, they are more likely to follow the procedures of that system.

In my experience, when the arbitrators come from civil and common-law systems, the procedure most likely to be followed is a hybrid of civil and common-law systems. It combines the information gathering functions of the civil-law systems with reliance on discovery devices that avoid the pitfalls of the unrestrained discovery that typifies American procedure.

For example, in a number of recent cases, the tribunal prescribed the following procedure: The claimant's complaint makes, in appropriate detail, all allegations of fast, including evidentiary. facts, and arguments of law on which the claims are based. Attached to this complaint are copies of the documentary evidence and statements of witnesses on which the claimant relies. The respondents answer must respond in similar fashion to the complaint. Mere denials are insufficient; they must provide the grounds on which they are based. The requirement of a reasoned denial is of crucial importance. It requires the respondent to put its cards on the table. Following the same rules, the claimant will be afforded an opportunity to reply and the respondent to rebut.

At some appropriate time, all parties are directed to produce copies of all documents and other tangible evidence relevant to the subject matter of the, dispute, or to afford all other parties appropriate access to such evidence. Unless good cause is shown, the parties are put on notice that failure to produce appropriate evidence may result in either preclusion of reliance on this evidence or the drawing of negative inferences.

At the conclusion of the pleading stage, any party may move for discovery of information not produced at that stage. The application must specify the precise information sought and indicate the manner in which it is to be obtained. Discovery will then proceed in accordance with the. directions of the tribunal.

When a hearing is deemed necessary by the tribunal, witnesses are requested to affirm their previously given statements, are permitted to make such additional statements as are, deemed proper by the tribunal, and are examined by both adverse parties and the tribunal. In major cases heard in the United States and the United Kingdom, the testimony may be recorded verbatim. In other countries, the arbitrators may make their own transcriptions or may rely on the chairman to prepare summaries.

These procedural rules have worked extremely well in prac-1315tice. They are probably most prevalent in international arbitrations conducted in the United Kingdom. In a recent case heard before me as a sole arbitrator, with American counsel for a French claimant and an American respondent, these rules were followed.8 The case presented a rather complicated dispute with many factual and legal issues. It was adjudicated, with separate decisions on liability and damages, in somewhat over a year. There appeared to be a consensus among counsel that the adjudication of the dispute in a federal court would have taken from six to eight years. Most importantly, counsel expressed their satisfaction that they had obtained at least as much information as they could have obtained under the federal discovery rules. Indeed, these rules work so well that they deserve to be emulated by municipal law systems.

The development of new rules of procedure attuned to the needs of the international commercial community has been facilitated by the large measure of freedom that institutional rules and the parties have left to the arbitrators. As to matters of substantive law, the arbitrators are normally not given similar latitude. While some international contracts, particularly concession agreements, direct arbitrators to apply general principles of law and equity or some similarly designated set of rules, most international agreements either provide for the application of a particular foreign law or leave the designation of the governing law to applicable choice of law rules. As a consequence, arbitrators are normally required to apply the law of a particular country. Nevertheless, arbitrators do not appear to feel unduly constrained by what they regard as undesirable rules of applicable law.9

One explanation may be that arbitrators normally apply rules of contract law, which do not differ dramatically from country to country, and that they are readily prepared to accept that the applicable foreign law rules are the rules that they regard as reasonable and compatible with their own law. Another explanation may be a readiness to assume that the applicable law accommodates special rules for international1316transactions that appear desirable to the arbitrators. While these may be the theoretical justifications for their applying the rules they regard as appropriate, a more straightforward explanation is that international arbitrators simply evolve and apply the rules they regard as the most appropriate in the circumstances.

Unfortunately, what international arbitrators do in this regard is mostly escaping notice. The failure to publish arbitral awards is largely responsible, as arbitrators have an obligation to preserve the confidential nature of proceedings. However, when arbitral awards are sought to be enforced or vacated in the courts, awards must be disclosed and evidence of what is happening in international arbitration is revealed. In addition, arbitrators may disclose some of their efforts without breaching their obligation to preserve confidentiality. Some instances of the creativity of international arbitrators are discussed here.

International contracts for the supply of goods or services are frequently concluded for a long term. Changes in market conditions, not foreseen or taken into account at the time the contract was concluded, are increasingly disturbing the harmony of these contractual relationships. International arbitrators, when called upon to address these problems, have often found domestic law doctrines too inflexible. For example, frustration doctrines have developed in the United States at a much more timid pace than elsewhere, particularly Germany and Switzerland.10 French authorities commonly state that the doctrine is recognized by the Conseil d'Etat, but not by the Cour de Cassation. How is an arbitral tribunal, composed of a Swiss, a French, and an American arbitrator, likely to address a frustration problem? First, the arbitrators are likely to choose the choice of law rule that points them to the most flexible law. But even if that is not possible-for example, because the contract provides for the application of a particular law - the arbitrators, for obvious reasons, wish to have reasonable flexibility and will construe the applicable law as allowing this.

A case in which I sat as an arbitrator provides a good example. The price of a commodity sold under a very long-term contract increased due to a combination of unexpected market price1317fluctuations (allegedy resulting from anticompetitive manipulation of the market) and a price-escalation mechanism that drove the contract price significantly higher than the already unexpectedly high market price. The price eventually reached a level that was likely to bankrupt the purchasers and produce exceedingly large profits for the seller. New York law had been declared applicable in the co tract. The arbitrators wished to find an equitable solution that was in accordance with the principle of objective good faith underlying the contract laws of both parties.

The black letter law provided an opening. The UCC and the Restatement of Contracts (Second) both recognize that a contract may be terminated if the basic assumptions of the parties are frustrated.11 But they appear to contemplate only termination, not modification, of the contract. They also give little indication of whether and how contract terms are to be adjusted to changed conditions. The arbitrators found that, because of the supervening unanticipated conditions, the contract as written could not be construed to address the new situation, and that the arbitrators had to supply reasonable terms. The parties' presumed intention, often regarded as the guiding principle, was of no great help because the conditions that had arisen were so unusual, that trying to construct a fictitious presumed intention made little sense. Instead, the arbitrators opted for a different approach. Evidence had been introduced indicating that the seller had reaped substantial profits from the contract and that it already had recouped its investment in the plant used to manufacture the product. The purchaser, on the other hand, because of unforeseen circumstances no longer had any use for the product, and would suffer ruinous losses if forced to resell it at a market price that had severely slumped. The arbitrators decided to continue the contract at a reduced price for a sufficiently long period to ensure that the seller would recover its investment plus a very liberal return to its investment, and to terminate the contract thereafter.

There appeared to be no American authority directly on point supporting this result. The cases found supported only a termination of the contract on the ground of frustration. And while there was authority for the view that frustration could excuse performance that would be ruinous for the party asked to1318perform, no existing authority had espoused an evaluation of the benefit to one contract partner in the light of the loss to the other. The decision of the arbitrators, therefore, added significantly to developing a doctrine of frustration in an international context.

Closely related to the problem of frustration is that of contractual price adjustment. Many long term supply contracts contain these clauses that seek to ensure that the contractual price level conforms some measure to prices charged in the market place. A usual approach is to provide for escalation by reference to stated criteria, with the proviso that the escalated contract price will be adjusted if it deviates more than a stated percentage from the market price.

Frequently, the concept of market price is not defined, or is defined insufficiently. Economists and market specialists are often asked to present heir view as to the relevant market price. Although the usual interpretative techniques have been urged upon arbitrators in an effort to encourage them to adopt a particular definition, arbitrators typically seek to construe the concept so as to assure that the parties are left in a position in which both can continue to live with the contract. The principle of reasonableness is often said not to permit a judge to modify the terms of the contract. But black letter law is not particularly helpful in this context. When the market price, in the ordinary sense of the price paid y a willing buyer to a willing seller, produces an unacceptable result, a more flexible construction that leaves greater leeway is sought.

Applicable law often prescribes specific interest rates. A statutorily prescribed interest rate may apply both before and after judgment. International arbitrators are likely to seek to escape the constraints of such statutory schemes. Often, the schemes may be interpreted to apply only to domestic relationships and not to international transactions. Furthermore, if the market rate in one country is significantly lower than in another and the parties have elected the law of the latter, it is improbable that they intended the statutory rate of the latter country to apply to obligations to e performed in the former. But even if such a divergence in interest rates does not exist, international1319arbitrators are likely to desire greater flexibility. One approach that can make a reasonable claim to recognition is to award the rate of interest that the creditor would have to pay if it were to borrow the amount owed to it.12 This solution more nearly puts the creditor in the position in which it would have been if the contract had been duly performed. If, however, the creditor could command an usually low rate, there might appear insufficient reason to allow the defaulting debtor the benefit of the creditor's position on the credit market, and therefore to grant the market rate or the rate the debtor would have to pay. Granting the latter rate would offer the advantage of depriving the debtor of any benefit in interest rate that he would achieve by defaulting.

These problems are likely to be particularly acute in international transactions that have relationships with countries that have quite divergent interest rates. Unfortunately, some courts have improperly sought to compel international arbitrators to award the statutory rate prescribed for local judgments.13 The courts would be well advised to leave international arbitrators the leeway they need to achieve just results.

International arbitral tribunals have also been in the forefront of solving the problem of providing appropriate remedies in cases of obligations incurred or owed in a foreign currency. The usual domestic law rule has been that a judgment will be rendered only in the currency of the country in which the judgment is rendered. When an obligation or arbitral award is expressed in a foreign currency, the judgment enforcing the obligation or award therefore converts the foreign currency into the domestic currency. It is only the exceptional statute, such as that enacted recently in New York, that, in specified conditions, permits a judgement to be rendered in a foreign currency.14

The tradition approach to currency conversion by judg-1320ment has been to fix a date at which the conversion is to be made. The principal approaches vying for prevalence have been those of making the conversion at the time to breach and of making it at the time of judgment. However, neither approach assures the judgment, debtor of what may be his reasonable due15 In the recent GTE v. Thomson16 case, the tribunal,17 while acknowledging hat New York law was controlling, developed a more flexible approach. Under this approach, the conversion is made at the time that assures to the creditor the amount in its currency that it would have had if the debtor had properly performed its obligation. Depending on the circumstances of the particular case, under this approach the conversion may have to be made at the date of breach, the date of judgment, or, at the option of the creditor, the date of payment.18 Hopefully, American courts will not, through an entirely inappropriate insistence upon what is (mis)perceived to be the local rule, deprive international arbitrators of this flexibility. On the contrary, American courts would do well to follow the example set by international arbitration in this regard.

Contracts procured through corrupt practices pose a recurring problem in cases involving countries in which such practices are endemic. On the one hand, international arbitrators do not wish to condone corrupt practices; on the other hand, the reality may be that no business can be done without the customary bribe or pay-off. How are international arbitrators to resolve this dilemma? One thing seems certain: International arbitrators will not enforce agreement to pay a bribe. However, this form of determination is possible only if the parties to the bribe are bound by the arbitration clause. The more important question is whether corruption can be a ground for invalidating the contract achieved through it. The question assumes particular importance because the political prominence of those involved in the corruption frequently would make it impossible to obtain annulment of the contract in the courts of the country whose officials were involved in the corruption. Therefore, in an action1321for invalidation, international arbitral tribunals are asked to assure to a foreign state or state company the benefit of enforcing laws that the country itself is not prepared to enforce.

International arbitrators must fashion a solution that does not offend international public policy and, at the same time, does not unduly punish businessmen or unduly reward foreign governments that do not keep their own houses in order. Several solutions appear possible. The tribunal may maintain the contract but reduce the contract price by the amount of the bribe. Supposedly, the person giving the bribe would have been prepared to lower the price if no bribe had been given. The theoretical underpinning for this solution would be that giving the bribe was a breach of the duty of good faith to be observed in contractual relations, thus entitling the other party to damages but not to annulment or rescission of the entire contract. Another solution would be to annul or rescind the contract, but to rule that the purchaser is entitled on a quantum meruit basis to recover the benefit bestowed upon the other party and that the contract price minus the bribe provides rebuttable proof of the benefit bestowed. A third solution would be to accept that bribery is a way of life in man countries and that, as long as the foreign countries themselves do not make effective efforts to combat it, they are ill-situated to call upon international arbitral tribunals to do what they fail to do. After all, a foreign country that permits corruption cannot, compatibly with the principle that nemo suam turpitudinem allegans audietur (no one shall be heard to allege his own turpitude) be heard to complain of its consequences.

In mergers and acquisitions, reliance is placed on balance sheets and profit and loss statements. These are normally guaranteed to have been prepared in accordance with proper accounting principles. These guarantees create special problems when the accounting principles customarily observed differ from country to country. Unless the contract unambiguously provides otherwise, international arbitrators, in a dispute involving parties from countries in which accounting principles differ, need not necessarily choose the principles accepted in either of these countries. Taking the purpose of these accounting principles as a guide, the arbitrator may formulate the principles that should have been observed. For example, when a balance sheet1322contains figures which reflect increased values in accordance with rising market prices, the fact that proper accounting principles may require that the assets be carried at original cost is not necessarily dispositive, even if both countries endorse the principle that these assets must be carried at cost. The crucial question in that case is to what extent the principles observed provide proper guides for assessing the value of the assets transferred. In answering that question, international arbitrators may formulate accounting principles particularly adapted to international transactions.

No doubt the examples discussed here may be multiplied by many others. As international arbitrators recount their experiences, the new international legal order will be further fleshed out.

An obvious disadvantage of the development of this order by international arbitrators appointed on an ad hoc basis is that the law developed is created by arbitrators coming from many parts of the world who do not perform as part of a structured judiciary. There can be no doubt that a systemic reporting of international arbitral awards would contribute significantly to consistency and harmony in these developments. However, several factors will help in guiding these developments into a rational structure. They include: The fact that the groups from which international arbitrators are selected are rather limited and regularly meet at international conferences; the recent emergence of periodicals devoted exclusively to international arbitration;19 and the publication of the World Arbitration Reporter, which will publish all International awards it is able to obtain.

*Stanley H. Fuld Professor of Law and Director; Parker School of Foreign and Comparative Law, Columbia University.
1See, e.g., International Chamber of Commerce [ICC], International Court of Arbitration Rule art. 2(4), reprinted in 3 World Arb. Rep. (Butterworth Legal Pubs.) 3649 (1987); see also AMERICAN ARBITRATION ASSOCIATION [AAA] INTERNATIONAL ARBITRATION RULES art. 6 (1991).
2On the lack of reporting of international arbitral awards, see Smit, The Future of International Commercial Arbitration: A Single Transnational Institution?, 25 COLUM. J. TRANSNAT'L L. 9, 31-32 (1986).
3Normally, the rules are indicated by the law of the forum. However, most national laws fail to provide detailed rules of procedure. An arbitral tribunal would appear to possess inherent power to regulate its procedure.
4ICC International Court of Arbitration Rule art. 13(3), supra note 1, at 3652.
5See Smit, supra note 2, at 22-24. But see AAA INTERNATIONAL ARBITRATION RULES art. 29, supra note 1.
6See, e.g., infra notes 12-18 and accompanying text.
7See Highett, The Enigma Of the Lex Mercatoria, in LEX MERCATORIA AND ARBITRATION 99, 100-08 (T. Carbonneau ed. 1990).
8See GTE Communication Sys. v. Thomson CSF Tel., S.A., 1 AM. REV. INT'L ARB. 315, 315-16 (1990) (Smit, Arb.).
9See id. at 315-16.
10See Smit, Frustration of Contract: A Comparative Attempt at Consolidation, 58 COLUM. L. REV. 287, 307-08 (1958).
11See E. FARNSWORTH, CONTRACTS § 9.7, at 721 (2d ed.1990). See generally Smit, supra note 10.
12See Smit, Elements of International Arbitration in the United States, 1 AM. REV. INT'L ARB. 64, 74-75 (1990).
13See, e.g., Carte Blanche (Singapore) PTE, Ltd., v. Carte Blanche Int'l, Ltd., 1 AM. REV. INT'L ARB. 172 (1990) (Smit, Arb.) (discussing Carte Blanche (Singapore) PTE, Ltd. v. Carte Blanche Int'l, Ltd., 888 F.2d 260 (2d Cir. 1989)). A subsequent district court decision in the same circuit rejected the Second Circuit's approach (without mentioning it). International Standard Elec. Corp. v. Bridas Sociedad Anonima Petrolera, Industrial y Comercial, 745 F. Supp. 172 (S.D.N.Y. 1990).
14N.Y. JUD. LAW. § 27(b) (McKinney Supp. 1991).
15See Smit, supra note 12, at 72-74.
16See supra note B.
17I was the Sole arbitrator.
18See GTE Communication Sys. v. Thompson CSF Tel., S.A., 1 AM. R.EV. INT'L ARB. 315, 316-17 (1990) (Smit, Arb.).
19Two such periodicals are Arbitration International, published in London, and Journal of International Arbitration, published in Geneva.