The decision earlier this week of the UK Supreme Court, that facts disclosed in confidential settlement discussions are admissible to interpret the settlement agreement achieved (Oceanbulk Shipping & Trading SA v. TMT Asia Ltd. & Ors, [2010] UKSC 44 (27 October 2010)), may contribute in a useful way to current discussions about ethical rules of conduct for counsel in international arbitrations.

The decision is useful to the debate over arbitration counsel ethics, because it re-focuses our attention on the reasons for what is known in the UK as the “without prejudice” rule. That rule, as the decision reminds us, is mainly a rule of public policy that encourages settlement of disputes by promoting candor in negotiation. It is secondly a rule of contract law, that an agreement between litigants to hold discussions confidentially is generally as deserving of enforcement as other contracts (though this is closely related to the public policy rationale, which provides the justification for enforcing the contract through an exclusionary rule of evidence rather than merely a claim for damages).

The rule becomes freighted with ethical overtones, because counsel is invariably responsible for the decision to disclose elements of the “without prejudice” discussion in furtherance of a litigation objective, and is usually responsible for the making of the “without prejudice” agreement if not also for the ensuing discussions. When counsel seeks to use information obtained in a “without prejudice” discussion, opposing counsel will issue sanctimonious condemnations of the behavior, often seeking some sanction.

When this happens in an international arbitration what is the arbitrator to do?

As an initial matter, a quest for the applicable law may be fruitless. The relevant rules are likely to be part of the law of evidence or civil procedure in the respective jurisdictions of the disputants. These rules do not apply unless the parties have agreed to them specifically. Arbitration rules of procedure do not ordinarily regulate the matter.  Complicating matters, it is debatable whether the without prejudice rule is a rule of international public policy that warrants vigilant protection from the international arbitrator. The public policy element of the rule is essentially a domestic objective to reduce the burdens that private disputes place on public courts. So if the parties have no agreement on confidentiality, the arbitrator is hard pressed to regulate.

But a without prejudice agreement between the parties, directly or through counsel, providing for the confidentiality of a settlement discussion or any other discussion among the parties or counsel, is properly to be viewed by the arbitrator as a rule of procedure governing the arbitration that she is bound to enforce. Further, when viewed as a rule of procedure, and not merely as a contract, it is quite arguable that the arbitrator has less discretion than a judge might have to give effect to many of the numerous exceptions to the UK version of the without  prejudice rule identified by the UK Supreme Court. Except for non-enforcement that might be said to result from international public policy (e.g. to permit disclosure of actual or threatened criminal conduct), the arbitrator may find little room to interpret the rule adopted by the parties to permit, for example, the disclosure permitted in the UK decision of facts pertinent to interpretation of a settlement agreement.  

Still, there exists the sense at least among common law lawyers that there is something quasi-ethical about respecting an agreement made on the confidentiality of communications between the disputants during the course of the arbitration.

If the arbitrator can establish that there is a shared expectation in this regard, the inclusion of a without prejudice rule in the initial procedural order may save controversy later on. As the debate over codifying ethical rules for arbitration counsel advances, we may expect to see this course being adopted by tribunals more frequently.

Just when you thought the day would never return when a U.S. court would actually show deference to a foreign arbitral tribunal on an issue of that tribunal’s jurisdiction, along comes a new decision of the U.S. Third Circuit Court of Appeals in Philadelphia, holding that an appeal before it on such a jurisdiction issue was moot – moot – because the ICC arbitral tribunal in Paris had ruled upon the issue in a partial award during the pendency of the appeal.  (Invista S.A.R.L. v. Rhodia SA, 2010 U.S. App. LEXIS 21950 (3d Cir. Oct, 25, 2010)).

The Invista case had its origins in a joint venture in the 1970s between affiliates of the chemical giants DuPont and Rhone-Poulenc, for the manufacture of an intermediate chemical essential to the production of nylon.  DuPont licensed the technology for the manufacturing process to the joint venture. The joint venture agreement strictly prohibited disclosure of the technology or related proprietary information for a period of 15 years, and it provided for disputes to be resolved by ICC arbitration in Paris.

By 2007, when the arbitration began, the DuPont and Rhone Poulenc interests in the joint venture had been transferred to entities in the Invista and Rhodia groups, respectively.  Each group had an interest in exploiting the licensed technology to build and operate a chemical plant in Asia, and the Invista side claimed the Rhodia side could not do so without disclosing the licensed technology in violation of the non-disclosure provisions of the joint venture agreement.

The crux of the arbitrability problem was that the entities, on both sides, that sought to exploit the technology in Asia were not the direct successors to the joint venture signatories  Rhodia commenced the ICC arbitration, naming as Claimants both the successor entity to the Rhone Poulenc signatory, and its non-signatory parent Rhodia SA. Invista commenced litigation in the U.S., naming Rhodia SA as defendant and contending it was not obligated to arbitrate with Rhodia SA .  Consistent with that position, the Invista parties in the arbitration sought dismissal of Rhodia SA as a claimant party.

The U.S. litigation made its way from a Texas state court, to a Texas federal court, to a New York federal court, to the Delaware Chancery Court, and at last to the federal district court in Delaware upon removal (as a case falling under the New York Convention) under Section 205 of the Federal Arbitration Act. (Readers seeking a more detailed account of this meandering itinerary will study the case.) In Delaware, the federal judge concluded that Rhodia SA, movant on the motion to compel arbitration, was not a significant participant with its affiliate in the running of the joint venture, and therefore as a non-signatory could not invoke the arbitration agreement against Invista under estoppel, assumption, or agency principles.

Several months after the Delaware District Court’s decision, in 2009, the ICC Tribunal in a partial award held that it lacked jurisdiction over Rhodia SA for substantially the same reasons given by the federal judge for denying Rhodia SA’s motion to compel, i.e. “documents produced on the record do not show that Rhodia SA was directly involved in the performance of the Joint Venture Agreement.” (Partial Award as quoted in the Third Circuit’s opinion).

The Third Circuit, on Rhodia SA’s appeal from the District Court’s denial of its motion to compel arbitration, held that the appeal was moot because the very arbitral tribunal to which Rhodia sought to have the claims referred had decided, in the meantime, that it lacked jurisdiction to hear those claims.

While the result here might be hailed, somewhat reflexively, as a vindication of the competence vested in the arbitral tribunal to determine its own jurisdiction, one may wonder as a matter of American arbitration law whether it is convincing for the Third Circuit to justify a finding of mootness solely on the basis of the Tribunal’s partial award.  The award had not been brought to the United States for confirmation (or to the court of any other nation, so far as appears in the opinion). Nor was this a case where it could be said that the parties to the arbitration had clearly and unmistakably agreed to submit the issue of arbitral jurisdiction to the arbitrators. The Invista parties in the arbitration moved to dismiss Rhodia SA as a party, but they had little choice but to do so, Rhodia SA having intervened as a party claimant without any agreement from the Invista side. Further, it is unclear that Rhodia SA could not have had access to any ICC arbitral forum if the Third Circuit had proceeded to determine that it was eligible to enforce the arbitration clause in the joint venture agreement. Certainly the existing tribunal could not vacate or modify its own award. But what if Rhodia then filed a separate arbitration and sought appointment of a different tribunal? Would the ICC Court not feel constrained to avoid choosing between the existing partial award on jurisdiction and the contrary judgment of a U. S. Court of Appeals? Would not the ICC Court in all likelihood would launch the case based on a prima facie assessment under Article 6(2) that an arbitration agreement might exist, and leave to the new tribunal the challenge of reconciling the partial award and the U.S. judgment?    

This is not to suggest that the Third Circuit outcome is wrong, but only that mootness was perhaps not the most persuasive basis for the decision.  Some form of estoppel would seem to be more convincing. Rhodia SA had intervened in the ICC arbitration with full knowledge of the tribunal’s powers to decide upon its own jurisdiction, and presumably with full knowledge that Invista would raise that issue before the tribunal. Rhodia could have treated Invista’s motion in the tribunal to dismiss Rhodia from the case as a refusal to proceed to arbitration, and on that basis might have purported to withdraw without prejudice from the arbitration while insisting that it was entitled to a final judicial determination of arbitrability from the Third Circuit. The heart of the matter seems to be Rhodia’s unqualified willingness to litigate arbitrability in the arbitral tribunal, which made its appeal to the Third Circuit a proverbial second (or third, counting the District Court in Delaware) bite at the apple.

One important question the Third Circuit did fail to address was whether Invista’s Delaware litigation involved any of the same issues already raised in the ICC arbitration between proper parties to that arbitration.  FAA Section 3 provides for a stay of proceedings if the lawsuit is brought “upon any issue referable to arbitration under an agreement in writing for such arbitration.”  It was not necessary to a Section 3 stay for the Court to find that Rhodia SA had the right to participate in the arbitration. The Third Circuit barely brushes up against this issue, merely noting that Invista in its Complaint alleged that the issues and claims presented in the litigation were distinct from those presented in the arbitration. The issue is not analyzed further. Examination of the appellate briefs to see if the issue was raised is beyond the scope of this writer’s present ambition. And so this criticism of the Third Circuit’s decision is made subject to that large qualification.

Under compétence-compétence, American style, nothing may be arbitrated if there never was a contract. We were reminded of this last week, when a federal judge in Manhattan (i) denied a motion to compel arbitration, and (ii) refused to stay its own proceedings pending the decision of an ICC arbitral tribunal in London on the contested issue of the existence of a contract containing an agreement to arbitrate.

(Dedon Gmbh v. Janus Et Cie, 2010 U.S. Dist. LEXIS 112131 (S.D.N.Y. Oct. 19, 2010).

In Dedon, the parties either did or did not enter into an exclusive distribution agreement. If they did, it unquestionably provided for arbitration of disputes — including disputes over the existence of the agreement — before an ICC arbitral tribunal in London. But it was common ground that the purported agreement had never been signed; the disputed issue was whether it had been adopted by performance.

The aggrieved terminated U.S. distributor, Janus, filed the ICC case, and Dedon responded by asking the ICC Court for a determination prima facie under ICC Rule 6(2) that an agreement to arbitrate did not exist (while expressly reserving the right to obtain a judicial determination of that issue) . The ICC Court instead found prima facie that an agreement to arbitrate might exist, and launched the case.

The U.S. court concluded that the text of Article II(3) of the New York Convention and U.S. Supreme Court precedent compel the federal judge to resolve the question of arbitrability presented when one party moves to compel arbitration and the other opposes that motion by contesting whether any contract between the parties ever came into being.

On the question of whether to stay proceedings and await the arbitral tribunal’s arbitrability ruling, the court concluded — with evident reluctance and hesitation — that awaiting the ICC tribunal’s ruling on that issue would only complicate matters, because the Court’s duty to decide the arbitrability issue de novo might then collide with its obligation under the New York Convention to recognize the award and any decision of a competent English court that might uphold that award. The court further justified its election not to defer to the arbitral tribunal on the ground that the plaintiff distributor in the New York action had sought urgent injunctive relief.

Should we regard this decision as a proper exercise of judicial control over the fundamental contractual basis for arbitration? Or was it a premature “intervention” in an ongoing ICC case?

As an initial matter, the notion that this was an “intervention” or “interference” by the judiciary in the arbitral process should be resisted. Within a three-week time frame, one party commenced proceedings ar the ICC, and the other filed the New York lawsuit. In the lawsuit, the ICC-filing party moved to compel arbitration. The New York Convention required the court to decide if the parties had made an arbitration agreement. Otherwise the motion to compel could not be addressed. The Court was not “interfering” or “intervening,” it was addressing a motion presented by the party seeking to pursue the ICC case.

Further, as no arbitral tribunal had yet been formed, in practical terms an arbitral decision on the arbitrability issue was several months off into the future. From a “comity” perspective, the intrusion into the province of an international tribunal was far less than if the tribunal had been seized of the case, and had already held a hearing on the jurisdiction question.

The fact that the ICC Rules make arbitrators competent to rule on their own jurisdiction should be of no consequence to a court asked to consider the arbitrability issue, when the arbitrability issue is whether the parties ever made a contract.  If they did not, no arbitral tribunal could legitimately claim power to apply the ICC Rules to a dispute. If the arbitral tribunal is the first to adjudicate this arbitrability issue, everything it decides, and all proceedings before that tribunal, have only provisional legitimacy until there is a judicial affirmation of the correctness of the arbitral decision on jurisdiction.

A stay of proceedings would be justified if the matter were ripe for arbitral decision, prompt judicial review of that decision could be had in the courts at the seat of arbitration, and the applicant for judicial relief on the merits would not suffer serious prejudice if required to await those rulings.

It remains to be seen how the ICC Tribunal will respond if the New York court rules that no contract ever existed. The Court’s denial of the motion to compel arbitration would be only an interlocutory order with no necessary binding effect on the arbitral tribunal. Of course the tribunal might elect to adhere to the New York court’s decision under collateral estoppel principles. Or the New York court might possibly accept an invitation to enter an anti-arbitration injunction — an order in personam against the party, directing the party not to pursue the arbitration, that could made into a judgment and enforced by a court at the seat of the arbitration.

Notably absent from the Court’s analysis is any discussion of whether an accelerated judicial determination of the arbitrability issue could be obtained from an English court. Indeed there is a provision  for such relief in Article 32 of the English Arbitration Act of 1996. It provides that the court may “determine any question as to the substantive jurisdiction of the tribunal.”  But it goes on to state that an application shall not be considered unless it is made with the agreement of all parties to the proceedings, or it is made with permission of the tribunal and the court is satisfied that judicial determination will likely produce substantial cost savings and there is otherwise “good reason why the matter should be decided by the court.” This may be the best solution for the New York court — which is scheduled to hold a conference October 27 to set procedure for a trial of the arbitrability issue. Should the court succeed in securing the mutual consent of the parties to submit the issue to an English court on an expedited basis, considerable savings in costs and avoidance of potentially inconsistent outcomes could be achieved.

International arbitrators are regularly called upon to resolve disputes over the application attorney-client privilege between parties from different countries that have fundamentally different rules concerning the existence and scope of the privilege. When one of the parties is a corporation from a civil law jurisdiction in Europe, or from an Asian nation, the legal function within the corporation may be carried out by lawyers or non-lawyers, or persons with legal training but who have a different professional status and are subject to different regulations than lawyers in private practice.

As a result, when international arbitrators are called upon to resolve privilege issues, the first question to be considered often will be what law governs the particular privilege issue presented.  The law chosen by the parties to govern their contract may bear no particular relationship to the jurisdictions implicated in the process of obtaining legal advice. Nor does the privilege law of the place of arbitration appear to be a uniformly attractive candidate, as at least one of the parties will usually have its internal legal department and its regular external counsel elsewhere.

Recently such issues were the subject of an elaborate analysis by a United States Magistrate Judge sitting in the Southern District of New York, in a litigated trademark infringement case between Gucci and Guess?, the denim purveyor. (Gucci America, Inc. v. Guess?, Inc., 2010 U.S. Dist. LEXIS 101219 (S.D.N.Y. Sept. 23, 2010)).    At issue were documents prepared by an in-house legal counsel for Gucci in Italy, an individual who was not trained or admitted as a lawyer but who was an expert in intellectual property and was responsible for Gucci trademark investigations that might lead to litigation.   

The U.S. law approach to this choice of law question, dating back at least to the mid-1970s and evidently developed mainly in cases involving the work of foreign patent agents, has been to ask whether the attorney-client communications of the foreign lawyer “touch base” with the United States — in the sense that the communications concern prosecution (actual or prospective) of a U.S. judicial or administrative proceeding, or the issuance of a U.S. patent, trademark or copyright. Under this approach, if an in-house counsel in Paris for the European affiliate of a U.S.-headquartered multi-national discusses by e mail with the affiliate’s CEO issues germane to the commencement of legal proceedings in a U.S. court, then American privilege law should apply.

This was the approach followed in the recent Gucci v Guess? case. The Italian non-lawyer working as an intellectual property counsel for Gucci’s affiliate in Italy gathered information for and communicated with the affiliate’s general counsel in Italy, an admitted to practice lawyer, and also sent messages to the Gucci group general counsel in London, also an admitted lawyer. The communications contained information and legal analysis for the purpose of evaluating= prospective trademark infringement actions in the courts of both Italy and the United States. His communications were considered to “touch base” with the United States in view of the contemplated American legal proceedings relating to alleged trademark infringement in the U.S., and so the Court applied U.S. privilege law.

What relevance might such case law have to an arbitrator sitting in the United States to decide a dispute between two foreign parties? An initial instinct might be to discount it, on the basis that the parties’ choice of New York as the place of arbitration does not imply the embrace of rules of evidence including rules of privilege that are features of litigation in U.S. courts. On the other hand, if the communications claimed to be privileged were had with a specific view toward the prosecution or defense of a New York-seated arbitration, especially if the contract is governed by New York law, then there is a respectable argument that the attorney or client who made the communication did so with a reasonable expectation that New York privilege rules would apply.

Another scenario in which the “touch base” choice-of-privilege-law formulation might come into play in a New York-seated international arbitration is if the communications were held for the purpose of evaluating claims or defenses that might eventually be asserted in wither an arbitral or a judicial forum.  It would be very common for an in-house counsel to have several communications about a brewing commercial dispute before checking to see if the contract contains an arbitration clause. If a mistaken but innocent assumption was made that there could be judicial proceedings in New York, but the dispute then goes to arbitration, there is good ground for protecting the expectation that confidentiality of attorney-client communications would be governed by U.S. principles. One could also imagine that communications might occur without certainty about the scope of arbitrable disputes, or about who the parties might be and whether some of them could be non-signatories who could not be brought into the arbitration.

Still, all the scenarios described above, in which the “touch base” formula could lead to applying U.S. privilege rules to foreign counsel’s entirely foreign-based communications, involve a common thread: the perceived possibility of litigation in U.S. courts on the part of the legal professional involved in the communications. If the communicators fully expected that forthcoming proceedings would be entirely in an arbitral forum, the fact that the arbitral tribunal would have a New York seat does not give rise to a justifiable expectation that the tribunal would apply New York privilege rules, as those rules are an attribute of the US judicial process and not such an important aspect of US public policy more generally that they could be expected routinely to trump less prophylactic privilege rules prevailing in the jurisdiction where the involved legal professional plies her trade.

When the foreign lawyer-client communication occurs before the contract is made, or before the dispute arises, and its subject matter is not a prospective litigation or other US proceeding, the arbitrator may find little justification for applying U.S. privilege rules to those communications — at least if choice-of-privilege-law is based on the same “grouping of contacts” principles that would apply to other choice-of-law issues.  But there remains the important question of equality of the parties. If the adverse party has U.S. counsel, and the tribunal applies U.S. privilege law to its communications, then there is a strong argument that fairness requires equal protection for the other side even though this may grant greater coverage than the foreign party and its counsel would enjoy in their own courts.