As the compétence-compétence stare-down continues between the U.S. courts and the drafters of the American Law Institute’s Restatement of the Law of International Commercial Arbitration, rather few seem to have taken notice that the U.S. Second Circuit Court of Appeals may have blinked.

The reflexive twitch happened on Halloween (boo!), when the Second Circuit in NASDAQ OMX Group, Inc. v. UBS Securities, LLC, 770 F.3d 1010 (2d Cir. Oct. 31, 2014) affirmed a preliminary injunction — based upon a judicial first instance determination of non-arbitrability — that prevents UBS from arbitrating claims against the NASDAQ for damages allegedly caused by the exchange’s handling of Facebook’s initial public offering in 2012.

NASDAQ and UBS had a “Services Agreement” whose dispute resolution clause provided for arbitration of all but a carved-out category of disputes, under the AAA Commercial Arbitration Rules. And those AAA Rules provided that the arbitrator “shall have the power to rule on his or her own jurisdiction….”

The Second Circuit has led the way for other federal courts in treating such arbitration rule provisions as “clear and unmistakable evidence” that the parties have agreed to permit arbitrators not courts to decide most issues of “arbitrability,” i.e. whether the arbitrator has jurisdiction — other than the threshold question of whether any agreement to arbitrate was made. But that position, at least per this latest decision of the Second Circuit, is based upon the presence of a “broad” arbitration clause that commits “all disputes” to arbitration, presumably including disputes over the scope of arbitrable issues.

In this instance the clause was not unqualifiedly broad. The carve-out mentioned above made non-arbitrable any disputes relating to (in simplified terms) alleged NASDAQ violations of its self-regulatory obligations imposed by SEC rules and regulations. And the Second Circuit declined to treat the agreement to use AAA Rules in any arbitration as clear and unmistakable evidence of an intent to have arbitrators decide whether UBS’s claims were within the carve-out, because, under the disputes clause as construed by the Court, the AAA Rules only apply to disputes not covered by the carve-out, and the scope of the carve-out is therefore an issue for preliminary judicial determination if raised in court before the arbitration unfolds.

It is probably premature to forecast a complete about-face by the Second Circuit on the compétence-compétence question. But one can readily derive a major limitation here on the allocation of “scope” arbitrability issues to arbitrators: If the scope issue presents the question of whether the entire dispute sought to be arbitrated is non-arbitrable, i.e. whether the arbitrators have any jurisdiction, then the jurisdiction issue is primarily for the court notwithstanding the agreement to arbitrate some but not all disputes under arbitration rules providing for arbitral competence to rule on objections to jurisdiction (meaning it may be resolved judicially on a motion to compel or enjoin arbitration, or it may be reviewed judicially without deference to the arbitrator’s ruling on the matter unless the parties clearly submitted the matter to the arbitrator without reserving the right to full judicial review).

This development in the Second Circuit’s jurisprudence does not, however, squarely confront the Restatement drafters’ position that compétence-compétence provisions typically found in institutional and UNCITRAL international arbitration rules are intended only to enable arbitrators to decide jurisdiction issues provisionally so that arbitration may proceed without the hiatus that would be necessary if only the courts had power to address arbitral jurisdiction objections. Nevertheless there seems to be as discernible shift, with the Second Circuit placing considerably more evidentiary weight on the scope of the agreement to arbitrate than on the selection of arbitration rules, as a measure of whether the parties agreed to have primarily arbitral determination of arbitral jurisdiction questions.

An occasionally encountered problem in international commercial arbitration is the Party Boycott. I will use that term here to refer to the situation where a Respondent in a pending arbitration registers its objection to arbitral jurisdiction systematically through a two-pronged strategy: (1) seeking an anti-arbitration injunction in a friendly court, and (2) refusing any participation in the arbitration itself.

Formation of the Tribunal

When the Tribunal is to be formed according to a list procedure by the administering institution, the Boycotting Party’s refusal to strike-and-rank the listed candidates typically entails that the institution will select the arbitrator or arbitrators. (As a preliminary matter the Boycotter may well argue to the institution that the arbitration should not proceed, but if the institution is satisfied prima facie of the existence of an arbitration agreement providing for arbitration administered by that institution, this effort ordinarily should fail). In the strike-and-rank scenario, the Boycotter succeeds in depriving the Claimant of input into the selection of the Tribunal. If the parties’ agreement provides for party-appointed arbitrators, then the Boycotter’s refusal to appoint entails that the appointing authority will make an appointment the Boycotter’s co-arbitrator, and that the Claimant’s party appointee and the co-arbitrator who has been administratively appointed will seek to reach agreement on a chair. Two of the three members of the Tribunal have been selected without input from the Boycotter.

Boycotter Requests Stay of the Arbitration

Invited to join a conference call to discuss a procedural timetable, the Boycotter may decline to confirm a date for such a proceeding, object to its “unilateral” scheduling, and request the Tribunal to stay proceedings at least for a time sufficient to allow the Boycotter to apply to a friendly court for an anti-arbitration injunction. The Boycotter proposes to invoke judicial authority of a court that is not at the seat of the arbitration, and whose arbitral procedural law does not apply to the arbitration. The Boycotter might for instance assert that it is a company in liquidation and that the liquidation court’s jurisdiction ousts that of the arbitral tribunal.  Whatever may be the merit, or lack of it, of that contention, from the perspective of arbitration procedure the question of the effect of the liquidation on arbitral jurisdiction should be addressed by the Tribunal and/or by a court at the seat of the arbitration unless the arbitration law of the seat, exceptionally, were to recognize the competence of the liquidation court to decide the issue.

Ordinarily, therefore, a motion to stay the arbitration in deference to a prospective jurisdiction-related decision from a court lacking competence on the jurisdiction issue ought to be denied.

Procedural Timetable and Time Limits for the Award

The first procedural conference is convened by telephone, with only the Claimant’s counsel appearing, the Boycotter and its litigation counsel (who requested the stay) having been duly notified. If the Claimant consents to bear the cost for a transcript, a court reporter records the proceedings, and Boycotter’s litigation counsel receives a copy.

Suppose a provision in the arbitration agreement requires a final award within a very stringent time limit measured from the formation of the Tribunal? Claimant cannot reasonably prepare complete written submissions on the merits and appear at a hearing within the time limit. Of course Claimant is willing to extend the time limit. But the Boycotter’s actual consent is not obtainable due to the Boycott, and neither the Tribunal nor the Claimant relishes the prospect that an Award made after the deadline might be denied enforcement on the ground that the Tribunal became functus officio at the time limit.

One solution is for the Tribunal to construe the time limit provision in the form of an Interim Award, deciding the issue of whether strict compliance with the time limit is required. If the Tribunal decides that strict compliance is not required, and the Boycotter has declined to take a position in the proceedings on this question, the Boycotter may find it difficult later on to deny the effectiveness of a Final Award on the basis that the time limit should be enforced strictly and the Tribunal ceased to have power before the Final Award was delivered.

Boycotter Obtains Anti-Arbitration Injunction From Friendly Court

The liquidation court in Boycotter’s home jurisdiction (not the arbitral seat) finds for the Boycotter on its application to enjoin the arbitration. The order operates in personam against Claimant, directing Claimant to proceed no further with the arbitration save to ask the Tribunal to stay the proceedings while Claimant pursues appellate remedies to vacate the injunction.

Now the arbitration is at a crossroads, and several factors are in play:

1) The Claimant submits to the Tribunal that while it disagrees with the injunction on the merits and as the Boycotter-friendly non-seat court’s power to impose it, the Claimant is loathe to risk a contempt judgment and therefore seeks to proceed no further with the arbitration on the merits until it can secure an order vacating the injunction.

2) A court with no supervisory power over the arbitration derived from the arbitration agreement or any national arbitration law has determined in an impactful way — via its own contempt powers exercisable over the Claimant — the competence of the Arbitral Tribunal,

3) Claimant assuredly would ask the Tribunal to rule on the jurisdiction issue were it not constrained by the injunction and by the fear of contempt penalties,

4) If the Tribunal accedes to Claimant’s request for a stay, without doing anything more, it gives legitimacy to the illegitimate (even if it were substantively correct) injunction of the liquidation court, in effect allowing contempt powers to trump all applicable arbitration rules and law.

5) Yet if the Tribunal insists that the case go forward — on the agreed accelerated timetable — it forces Claimant to a difficult  election: risk contempt penalties and trust in the appellate process to reverse the injunction and the contempt, violate the Tribunal’s order to proceed, or withdraw the arbitration.

Shall The Tribunal Retreat Into Cold Storage?

What is a Tribunal to do? In the case that inspires this post, the Tribunal (seated in London) elected to make an award on the issue of whether the liquidation proceeding divested the arbitral tribunal of jurisdiction. Having in the first procedural order directed both parties to provide copies to the Tribunal of all submissions made the Boycotter-friendly non-seat court relating to the arbitration, and all orders of that court relating to the arbitration, the Tribunal was fortunate to secure Claimant’s compliance with that direction notwithstanding the anti-arbitration injunction.

Armed with the same factual and legal record made by the parties in the liquidation court, the fact that Respondent and Claimant were for different reasons (derived from the same injunction) unwilling to brief the issue to the Tribunal was not seen as a deterrent to a decision. The Tribunal proceeded to a Partial Award finding that its jurisdiction was intact. And having given that award, the Tribunal rather than grant the Claimant’s request for a stay, instead amended the procedural timetable to defer the merits hearing and pre-hearing submissions for a time sufficient for Claimant to pursue a first-level appeal against the injunction.

The Tribunal in such circumstances is rightfully uneasy at the prospect of granting either a proceedings stay of indefinite duration, or, what would eventually amount to the same thing, seriatim adjustments of the hearing timetable as each previously-fixed set of submission and hearing dates approaches and the illegitimate judicial anti-arbitration injunction remains in force while appellate challenge proceeds at a pace that seems to reflect the appellate court’s indifference to the Tribunal’s desire to proceed on the timetable it has established.

It would seem that a decent respect for the arbitration law that the Tribunal has a mandate to apply should bring into play a self-imposed obligation of the Tribunal to express in every appropriate way its refusal to be governed by the illegitimate injunction. But to insist that the enjoined Claimant defy the injunction and risk contempt of the enjoining court seems untenable: It would be an abuse of arbitral power to insist that a party bear this risk or, if it should refuse to bear it, have its claim determined against it on the merits based on its refusal to proceed with the hearing.

The solution decided upon in the case that inspires this post was for the Tribunal to end is mandate without reaching the merits. Claimant was not put to the choice of proceeding to the merits or not, but instead was given the choice of proceeding to the merits or withdrawing the arbitration without prejudice. The Claimant’s claim and its right to arbitrate that claim were preserved.

It will not be in every case where the Claimant desires a stay pending judicial appeals of the injunction that the Tribunal would opt for this solution. But the judgment made here was that the mission of an Arbitral Tribunal is to render an Award resolving the disputes presented by the parties, and if it has no reasonable prospect of being able to do so on a reasonable time horizon because of a judicial injunction that the Tribunal has determined to be procedurally and substantively illegitimate, then discontinuing the arbitration without impairment of the Claimant’s right to arbitrate the claim in a newly-filed case may usefully address several concerns. Already mentioned is the systemic reason: that the Tribunal should not be seen to be under the thumb of a judiciary that has no legitimate authority to supervise or regulate the arbitration. Another reason is that if the injunction is eventually dissolved, the arbitration should proceed as closely as possible as it would have proceeded had the injunction not be issued. The seated Tribunal cannot help but have been influenced, and certainly will be seen to have been influenced, unfavorably to the Boyoctter.  A new Tribunal should in principle have no such disposition, and, of course, if the agreement to arbitrate provides for party-appointees, and the appointees’ participation in selecting the chair, a Tribunal constituted with the full voluntary participation of both parties is desirable. An Award rendered by such a new Tribunal should face a smoother path to enforcement if it must be presented for enforcement in the courts of the same State that provided the injunction. Termination of the proceedings also eliminates the possibility that the fact of the pendency of the arbitration serves some secondary business purpose for the Claimant, making the Tribunal an unwitting facilitator. That is an inevitable by-product when a case is proceeding along a normal course, but is better avoided if the Tribunal cannot accomplish its core mission. Further, the monitoring of the judicial proceedings may affect the Tribunal’s disposition toward the Claimant and its counsel, as the Tribunal makes its own private assessments of the efficacy Claimant’s efforts to have the injunction lifted.

Of course there can be countervailing factors that militate in favor of the Tribunal being more patient in awaiting possible rectification in the enjoining State’s courts. One would be if the Tribunal has worked on the merits at significant cost to the parties before the anti-arbitration injunction is obtained. But it will be more usual that the Boycotting Party will invoke judicial authority early in the case, and that the Tribunal will not invest much time.

* * *

One potential avenue of recourse for the Claimant aggrieved by the Boycott is to seat an anti-suit injunction against the Boycotter. For a reminder that where UK arbitration law applies, a UK court may enjoin foreign litigation that would interfere with the arbitration, see “Antisuit injunctions: No arbitration? No worries” from the International Arbitration Newsletter of DLA Piper, 26 September 2013 (www.dlapiper.com/en/japan/insights/publications/2013/09/antisuit-injunctions-no-arbitration-no-worries).

Query how shall a Tribunal when asked by the enjoined Claimant for a stay of indefinite duration, pending appeals of the illegitimate injunction, take into account that party’s failure at an early stage to have availed itself of this opportunity in a UK court to enjoin the adverse party from pursuing an anti-arbitration injunction?

***

Query also, if the Tribunal makes a Partial Award in favor of its own jurisdiction, how shall the Tribunal take into account, when asked by the enjoined Claimant for an indefinite stay, the fact that the Claimant undertakes no proceedings for enforcement of that Award?  Is it a satisfactory answer that pursuing a proceeding for enforcement of the Award might motivate the Boycotter to apply for contempt sanctions?

***

Also, what degree of proof shall a Tribunal require that the enjoined Claimant will be subjected to a contempt sanction if it takes particular steps in the arbitration or related to the arbitration (such as award-enforcement or anti-suit injunction proceedings in a court at the seat)? Shall the Tribunal require some proof as to the severity of the potential sanction, the likelihood of its imposition, and its irreversibility even if the injunction is ultimately declared to have been unlawful? Not to be overlooked in this regard is the possibility for the Tribunal as a provisional measure to direct the Boycotter to refrain from taking contempt proceedings, as was done in the well-known SGS v Pakistan ICSID arbitration. (See E. Gaillard, Reflections on the Use of Anti-Suit Injunctions in International Arbitration, in L. Mistelis & J. Lew eds., Pervasive Problems in International Arbitration, pp. 203 et seq at p. 205 n.6 (2006)). But such an order against a non-State party in commercial arbitration may be toothless unless it can be enforced judicially in a jurisdiction where the Boycotter has assets.

Among the common law theories in American law that may permit enforcement of an arbitration clause against a non-signatory, equitable estoppel is perhaps the most elusive. Its application is intensely fact-dependent, and different sets of equitable considerations apply depending on whether the party seeking to invoke arbitration is the non-signatory or the signatory. And when the matter comes before an American court, this is essentially a question of state law, and different states have different refinements of the conduct standards that may trigger estoppel as well as variations in the evidentiary burden that the party invoking estoppel must satisfy.

These factors can be seen in operation in a recent decision from the U.S. Third Circuit Court of Appeals, in which the Court rejected the contention that a U.K. insurer by joining in a mediation under Contract A which contained an arbitration clause but which it had not signed, waived an express reservation of the right to litigate under Contract B which it had signed. Flintkote Co. v. Aviva PLC, 2014 WL 50033218 (3d Cir. Oct. 9, 2014).

The foundations of this dispute were two agreements concerning dispute resolution over asbestos insurance coverage claims. The first, an agreement with multiple insurers known as the Wellington Agreement and dating from 1985, is an early example of a multi-tiered ADR clause, providing for mediation, binding arbitration, and an appeal process. But Aviva, the respondent here, did not sign the Wellington. It made a separate contract with Flintkote in 1989, expressly providing for litigation of coverage disputes and expressly disavowing any ADR obligations under the Wellington.

Aviva nevertheless opted to join a multi-insurer mediation with Flintkote, based on the Wellington, and Flintkote’s estoppel theory of Aviva’s obligation to arbitrate turned largely but not entirely on that opt-in. Adding fuel to the eventual estoppel fire, Aviva joined with other London insurers who were members of the Wellington group in negotiating terms of a potential arbitration agreement. But Aviva eventually parted ways with the group and insisted on its separate contractual right to litigate. Flinkote responded with a petition in U.S. District Court to compel arbitration.

Here the applicable state law (of Delaware) required “clear and convincing” proof of the estoppel, and while the District Court had been satisfied on this test by Aviva’s voluntary opt-in to the Wellington multi-insurer mediation, the Third Circuit saw the matter differently. Aviva had not “embraced” the Wellington’s entire ADR structure by opting into the mediation, the Court held,  as it had not been required to affirm the entire Wellington Agreement or to disavow its separate litigation rights as a condition for joining the mediation. The Court also rejected the notion that Flintkote had been misled to believe Aviva was accepting to arbitrate by opting into the group mediation —  as Flintkote knew the separate contract with its litigation clause existed, and had no reasonable basis to believe Aviva had waived the benefit of it.

It is tempting merely to classify this case as fact-dependent and offering no transcendent particular guidance.  But it is valuable to note that arbitration law as opposed to contract law really plays no part: when the question is whether to extend the obligation to arbitrate to a nonsignatory who is unrelated to any signatory, the issue is purely one of state contract law and U.S. courts quite properly analyze the matter without the playing field-tilting weight of federal pro-arbitration policy and presumptions.

Before the author of Arbitration Commentaries was deployed to the trenches and thus temporarily lost to his readers (some would say mercifully), it was written in this c-space that the “Next Cool Thing” in U.S. arbitration jurisprudence, after BG Group v. Argentina, would be the question of who decides — court or arbitrator — whether an arbitration clause permits class arbitration, when the parties have no agreement on the “who decides” question itself. See “Brush Up Your Bazzle,” Arbitration Commentaries, July 1, 2014.  A four-judge plurality of the Supreme Court in Bazzle was prepared to hold that the question of whether an arbitration clause permits class arbitration is ordinarily a procedural issue for the arbitrator to decide. But the Court in later class arbitration cases has taken pains to note that this was the position of four justices not five.

Behold. In the middle of my combat assignment — ended skillfully by a mediator just in time for Labor Day Weekend to assume its traditional Blog-N-BBQ format — a Third Circuit panel in Philadelphia, spurning the Jersey Shore, held that the “class arbitration” question involves “arbitrability” and thus must be decided by courts not arbitrators unless the parties clearly (and unmistakably) have otherwise agreed. Opalinski v. Robert Half Int’l, 2014 WL 3733685 (3d Cir. July 30, 2014).

The first premise of the Third Circuit panel’s decision is that “[t]he availability of class arbitration implicates whose claims the arbitrator may resolve.” And the decision cites the Supreme Court’s 1964 John Wiley decision in support of the assertion that “[t]he Supreme Court has long recognized that a district court must determine whose claims an arbitrator is authorized to decide.” The Court in John Wiley decided, precisely, that the court not the arbitrator should decide whether a successor by merger to a company that had a collective bargaining agreement with a union is bound by that agreement. For the Third Circuit panel to generalize that holding into one that “a district court must determine whose claims an arbitrator is authorized to decide” — and thus to suggest that it has bearing on who decides the class action question, amounts to putting apples and oranges together in a crate marked “fruit.”  The other cases cited by this Third Circuit panel involved whether non-signatories were bound by an agreement signed by someone connected to them. The obvious difference is that each class member in the proposed class arbitration has an identical agreement to arbitrate with the same Respondent. The “who decides class arbitration” question does not present an issue of “whose claims may be arbitrated (as opposed to litigated)” but only an issue of “whose claims may be arbitrated (in one arbitration rather than several).” The Third Circuit panel ignores rather than addresses this obvious distinction.

The other major premise of the Third Circuit panel decision is that the differences between class and individual arbitration are more “substantive” than procedural, so that the question deserves to be regarded as a “gateway” issue for judicial determination just as would an issue of whether the subject matter of the dispute was within the scope of the arbitration clause. Certainly there is merit to the notion that whether a clause permits class arbitration has great importance to the defendant, who may face huge potential damages if class arbitration is allowed, and negligible exposure if class arbitration is disallowed because individuals may find it un- economic to pursue only their own claims. But the Supreme Court’s jurisprudence has until now classified as “gateway” issues only issue that entail whether there was consent (1) to arbitrate at all, or (2) to arbitrate the subject matter of the proposed arbitration — and not whether an admitted consent to arbitrate embraced consent to arbitrate in a particular format. Class arbitration surely has larger economic implications than many other arbitration format issues ordinarily do, but it would seem to remain in the category of format questions, not consent questions, within existing Supreme Court jurisprudence, so long as the defendant has an agreement to arbitrate the subject matter with each member of the proposed class.

This critical view of the Third Circuit panel’s decision is surely not a prediction of what the Supreme Court might decide. The panel was clearly attuned to the discomfort expressed by several members of the Supreme Court, especially in Stolt-Nielsen and Concepcion, based evidently on the tendency of class proceedings to shift economic power away from corporations and toward individually powerless individuals joined together opportunistically to enlarge the litigation risks — and consequently the business behavioral risks — borne by corporations. There is an undercurrent in these cases that such power-shifting decisions are unsuitable for private-sector decision-makers appointed only by the parties or the arbitral institutions whose rules the parties have embraced. But if limits are to be imposed for such reasons on the availability of arbitration, perhaps it is a matter for Congress to decide, and not for the Supreme Court to resolve by stretching the notion of “gateway dispute” beyond its established borders.

The use of subpoenas by arbitrators pursuant to Section 7 of the Federal Arbitration Act remains an evolving area of arbitral practice. There are several sources of difficulty. One is how to adapt the language of a 1925 statute to complex and multinational disputes. Another is that arbitral subpoenas shall be judicially enforced with reference to judicial rules of procedure governing compulsion of the attendance of witnesses. A third issue is how technology and especially video technology should affect the ability to secure evidence from an individual who resides very far from the seat of arbitration and sometimes overseas.

Suppose the Arbitral Tribunal, sitting in New York in an international case, issues a subpoena for a pre-merits hearing to receive evidence including oral testimony from a multinational firm with its global headquarters in New York, although it is understood that the individuals with hands-on knowledge work in an Asian office of the firm and are foreign nationals.

Proper service of the subpoena is not a difficulty, assuming the firm is a “person” under FAA Section 7 who may be summoned — and there is no reason to think that it is not (see the definition of “person” in the Dictionary Act. 1 U.S.C. § 1). Section 7 provides that the subpoena (technically, a “summons”) “shall be served in the same manner as subpoenas to appear and testify before the court….” And under FRCP 45 as amended December 1, 2013, a federal judicial subpoena may be served anywhere in the United States. Territorial limitations in Rule 45 inhibit how far from home an individual may be required to travel for a witness appearance, but this is no longer implemented by limiting the geographic range of effective service.

If the corporate non-party witness so served resists making its knowledgeable foreign employee available to testify, what level of judicial compulsion does Section 7 permit? It is clear at least that the foreign employee cannot be required to appear in New York. But the relevant knowledge legally is possessed by the firm. May a federal court under Section 7 compel the summoned firm to identify and educate a person who could have been individually compelled to appear, offering the firm the option of producing the already-knowledgeable foreign employee by video conference if desired to avoid the burden of educating a witness?

The question just posed triggers two separate lines of analysis. And neither is well-developed in precedent. The clear judicial analogue for requiring a corporate witness to identify (and sometimes educate) an individual to testify is Rule 30(b)(6). It is settled in federal discovery practice that Rule 30(b)(6) applies to the deposition of a non-party corporate witness. It is controversial, on the other hand, whether under Rule 45 a trial subpoena to a corporation may be enforced by an order that compels the firm to designate a knowledgeable individual.

FAA Section 7 permits the federal district court where the arbitrators “are sitting” (an issue for another day and another post) to “compel the attendance of such person…in the same manner provided by law for securing the attendance of witnesses…in the courts of the United States.” If that language is held to mean an arbitral summons may only be enforced in the same manner as a judicial trial subpoena, then a court lacks power to enforce an arbitral summons to a corporation if the court considers that a Rule 45 trial subpoena to a corporation, calling for designation of an testifying representative with knowledge, is not enforceable.

And indeed this was the holding by a federal district judge in the Southern District of New York two weeks ago in Progenics Pharmaceuticals, Inc. v. IMS Consulting Group,  No. 14 Misc. 00245  (S.D.N.Y., Aug. 14, 2014) (unpublished order, on file with this writer). The arbitral summons, to the extent it required testimony of a corporate-designee witness with substantive knowledge of the subject matter identified in the summons, was held to be unenforceable because the Court lacked power under FAA Section 7 to require a 30(b)(6)-type witness designation by the summoned firm.

I make no claim of objectivity, having acted as counsel for the proponent of the arbitral subpoena in this case. And so I merely report here the line of argument that left this particular judge unconvinced: (1) that the enforcement provision of Section 7 should be read liberally, not restrictively, as its main purpose is to provide support for the evidence-gathering efforts of arbitrators not to inhibit them, (2) read liberally, the words of the statute do not appear to limit enforcement to those powers associated with a judicial subpoena to testify at a trial, as “attendance of witnesses…in the courts of the United States” can easily be read to include all proceedings involving the attendance of witnesses in cases before the courts, (3) the reading that limits enforcement to judicial powers to enforce a trial subpoena is flawed because the text FAA Section 7 clearly provides that there may be a pre-merits arbitral hearing before just one of three arbitrators, and such proceedings are more akin to depositions for the perpetuation of testimony than they are to trials, (4) at the time FAA Section 7 was enacted (1925) the conducting of non-party depositions for the perpetuation of trial testimony from witnesses who would not or might not be available to testify in person at trial was common, and specifically provided for in sections of the Judiciary Act (28 U.S.C.),  (5) Rule 30(b)(6) is not only a rule for taking discovery from entity (“legal person”) witnesses, but is also the method of perpetuating trial testimony of witnesses who would not or might not be available at trial, and thus is the post-1938 counterpart to the former Judiciary Act provisions just mentioned, and (6) given this longstanding non-discovery application of Rule 30(b)(6), cases holding that Section 7 does not empower arbitrators to issue subpoenas for discovery depositions do not foreclose the conclusion that an arbitral subpoena may require a legal person to identify a natural person to testify as its representative.

None of this was convincing, mainly I believe because there are no reported cases deciding this question one way or the other, and in that sense the enforcement sought was “unprecedented.” (But perhaps there were 10 unpublished orders reaching the opposite result, and in reality the denial of enforcement was unprecedented. This Order, published only in the hearing transcript, will be found mainly due to the loquacious nature of the author of Arbitration Commentaries).

What was not included in my argument, of course, was the obvious: FAA Section 7 is broken and antiquated. It is a crooked unpaved 1925 country road desperately in need of replacement by a modern superhighway. Lawyers trying to make Section 7 work sensibly in modern complex arbitration should not have to engage in the legal gymnastics of proving that a procedure that makes sense for gathering evidence in 21st Century arbitration is akin to a judicial litigation practice that existed in 1925.

But there is no reasonable prospect for legislative clarification, and so we count on judges to have some sense of the arbitral process and to recognize its distinctive character and to appreciate the needs of litigants within that process. Unfortunately from the perspective of arbitration, there are perhaps just as many federal district judges who have never struggled with FAA interpretive issues as there are those who have done so on many occasions. The organized arbitration bar needs to do a better job of providing our judges with authoritative guidance.

If you don’t remember Bazzle, you had best put it on the beach-and-boat-reading list. Come September, you will need to know it well for survival at every luncheon and cocktail reception on the arbitration circuit. Why? Because the Next Cool Thing in US arbitration law, now that BG Group v Argentina is just . . . So Last Term, is the question whether interpretation of the arbitration agreement to determine if it permits class arbitration presents a “gateway” issue of (or akin to) “arbitrability” that a court not an arbitrator should, presumptively and thus quite often,  decide.

In Bazzle (Green Tree Financial Corp. v. Bazzle, 539 U.S. 444 (2003)), a four-Justice plurality of the United States Supreme Court was of the view that this question was one for the arbitrator to decide, and whereas no arbitrator had yet decided it, but only the courts of South Carolina, the case was remanded to the arbitrators. A fifth Justice (Mr. Justice Stevens, now retired to the literary life) concurred in the result, saying that the plurality’s reasoning was “close to his own” but that he did not agree with the plurality that a remand to the arbitrators was necessary here, as he would have found that South Carolina’s highest court had correctly determined as a matter of State law (not contract interpretation) that class arbitration was permitted, and because Petitioner did not advance the contention that the wrong decisionmaker had decided the class arbitration question.

That leaves Bazzle as a tenuous precedent, one to be eventually perhaps ratified by a majority decision of the Court, or to be relegated to obsolescence by a different-minded majority.

And we now have a decision from a US District Court that possibly will in due course find its way to the Supreme Court. In In re A2P SMS Antitrust Litigation, 2014 WL 2445756 (S.D.N.Y. May 29, 2014), the Court found the Bazzle plurality’s position persuasive and held — in the context of an AAA international arbitration spawned by a federal judicial antitrust class action — that the task of interpreting the arbitration agreement to decide if it allows class arbitration is a non-gateway procedural issue for arbitral decision.

The posture of this case at the time of the decision is itself worthy of a practitioners’ note. Plaintiffs filed the case as a class action in the court and when defendant moved to compel arbitration, Plaintiffs argued that the arbitration clause was unenforceable as a matter of public policy because, if construed to require individual arbitration, it precluded “effective vindication” of the Plaintiffs’ rights under the antitrust laws. After supplemental briefing of that issue following the Supreme Court’s 2013 decision in the Amex-Italian Colors case, the District Court granted the Defendant’s motion to compel arbitration — without deciding if the clause required only individual arbitration. Plaintiffs then filed a Demand for Arbitration seeking class arbitration. And Defendants responded by asking the Court to decide that class arbitration was not permissible. Presumably (we do not know), Claimants asked the arbitral tribunal to decide the question, and presumably (we do not know) the arbitral tribunal decided to wait for the court to decide who decides. Uncertainty about the law concerning arbitral power prompts arbitrators to defer to judges, and the efficiency of arbitration suffers if judges cannot act swiftly (swift action being painfully difficult when the legal issue is unsettled, as this one is in the Second Circuit, and the parties wish to have ample time for briefing and argument).

Here the District Court voraciously devoured not only the Supreme Court’s arbitration class action jurisprudence, but also the relevant decisions of lower federal courts nationwide, and concluded that the position of the plurality in Bazzle remained the most persuasive: whether arbitration will proceed on an individual or class basis despite the fundamental changes in the process if class arbitration is ordered, remains a question of procedure not a question of consent and therefore does not qualify for “gateway” status (i.e. Questions ordinarily to be decided by courts) under the Supreme Court’s arbitration jurisprudence. The District Court found the question presented to be “a close one,” mainly because the Supreme Court in the Stolt-Nielsen case (2010), and again last year in the Oxford case, took pains to draw attention to the fact that they were not deciding (and that Bazzle as a plurality decision had not decided) if the class arbitration clause construction issue is a “gateway” question for courts to handle. The question was not presented in either case because the parties had agreed to submit the clause construction issue to the arbitrators, which left for the courts only questions about the scope of review and (retrospective) limits on arbitral power.

“Gateway” issues are, broadly speaking, issues of consent: Did you agree to arbitrate at all? Did you agree to arbitrate this type of issue? Argentina argued in BG Group that when it agreed to arbitrate only with investors who first litigated pointlessly for 18 months in Argentine courts, an investor’s by-pass of that requirement presented an issue of consent. The Chief Justice and one of his brethren bought in (the sistren, refusing to cry for Argentina, all gave this pitch the thumbs down).

But BG Group is … just So Last Term.

So what is the proper view of the question “Did you agree to arbitrate in one arbitration with a nameless faceless mass of persons ostensibly in the same boat as the signatory claimant who filed the case, all of the said boatmates having signed arbitration agreements with you but none of them save one having asserted any claims?” Is the question only about process, or (as five Justices suggested in Stolt-Nielsen) is class arbitration such a transformation from the arbitration archetype of Halcyon Days that it deserves to be classified with pure questions of consent as a matter presumptively for determination by judges?

Brush up your Bazzle, readers. This is So Next Term.