Marc J. Goldstein Arbitrator & Mediator NYC
April 03, 2017

Parsing Protective Orders

Party autonomy and American litigation custom sometimes collide in disconcerting fashion in arbitrations involving American counsel, whether international or domestic. One such collision involves the establishment early in the case of an agreed or imposed order concerning the confidentiality of exchanged information (“Protective Order”).  The parties have an understandable desire for formal confidentiality restrictions applicable to the data that they will be required to share with adverse parties who may be, or may be aligned with, actual or potential business competitors. But the templates for Protective Orders that many US counsel will retrieve as drafting models for their arbitral confidential orders will come from prior litigations handled by their firms in federal and state courts, and counsel are far more likely to focus on having the most ironclad protections against misuse of their clients’ data than they are to process systematically how the arbitral forum and arbitral process changes the dynamics of confidentiality and in turn the transferability of the litigation Protective Order template to the arbitral forum.

While it is generally understood that an arbitration seated in the United States is not inherently confidential — as regards the ability of the parties to make public reference to the proceedings — there is no public docket for the arbitration as there is for a judicial proceeding, and pleadings, orders and awards in the arbitration will not reach a public judicial docket save insofar as judicial proceedings relating to the arbitration are maintained. This may seem obvious, but counsel sourcing model Protective Orders from their archives, for adaptation to an arbitration, may not pause to reflect on the fact that a great deal of the restrictive regulation of data usage that appears in typical Protective Orders  is motivated by the fact that litigation dockets and proceedings are public (subject to case-specific, episode-specific discretionary exceptions for filing under seal of particular documents or portions thereof, as approved by judges on a case by case basis) and that American courts operate on a principle of the public’s right of access to the proceedings.

  1. The “Use Restriction”

Arbitrators will have this principle in mind when they see in a party-drafted confidentiality order a ubiquitous provision of such orders — the so called “use restriction.” Typically couched in language reciting that the purpose of the Protective Order is to avoid use of disclosed information for business purposes and to avoid public disclosure, the typical use restriction literally reads more broadly: that disclosed information designated as confidential “shall not be used for any purpose except the prosecution and defense of this Action.

Taken literally and apart from context, this language might be held to prohibit any litigation-related “use” of disclosed information. Suppose the arbitrating party obtains from the adverse party information designated as “highly confidential, attorneys eyes only” under, and subject to, the Protective Order (see Section 2 below concerning tiers of confidentiality), and the information so obtained might support causes of action in court against a third party with whom the receiving party has no agreement to arbitrate. Suppose further that the receiving party applies to the Tribunal to remove the “attorneys eyes only” designation so that she might discuss with her client the documents’ significance in regard to potential new litigation against third party? Should the arbitral tribunal adopt the producing party’s objection, supported by an intervention from the third party who has been notified of the initiative, that even such a pre-litigation exploratory discussion is a “use” for a purpose other than “this Action” and is prohibited by the Protective Order?

One can see that quite rapidly a seemingly standard (for US civil litigation) Protective Order provision could embroil the Tribunal in a controversy that has nothing to do with protection of business secrets against competitive use, nothing to do with maintaining a non-public profile of the arbitral proceeding, but everything to do with regulation of substantive rights between a party to the arbitration and a stranger to it. Would the Tribunal be well advised to insert in the parties’ version of the use restriction “except as the Tribunal may permit for good cause shown“? This arguably captures the intent of the parties, which was not — at the time of drafting — to squelch potential new claims derived from the content of produced documents but simply to avoid the drafting difficulties and potential for circumvention that would be involved in stating in the Order a more particularized list of prohibited uses.

  1. Tiers of Confidentiality

The typical litigation model Protective Order submitted to arbitrators by the parties, or one of them, via US counsel will contain a so-called “two-tier” confidentiality regimen, permitting a party producing data to designate its data as “confidential” (generally, subject only to the use restriction) or “highly confidential” (generally, restricted to viewing and handling by outside counsel and consulting/testifying experts). In the US court system, the two-tier model often results in over-designation of documents as “highly confidential” by the producing party, with the task of deciding on requests to downgrade the tier designation delegated to a US Magistrate Judge. In the court system an application by the receiving party to downgrade the tier designation typically would be made during the discovery phase of the case, before the setting of a trial date or a schedule for briefing on proposed summary disposition. Therefore such disputes in a court litigation do not disrupt the trial schedule expectations of the trial judge or the parties. The dynamics of an arbitration in regard to this question are self-evidently quite different. The Tribunal has no institutionalized subordinate to whom tier-designation disputes may be delegated and efficiently resolved. The parties may well not have factored into their vision for the arbitration procedure and timetable the enlistment of a three-person Tribunal to the urgent task of deciding which persons other than counsel may consider the producing party’s documents for purposes of assisting in the prosecution or defense of the arbitration. Further, the full procedural timetable culminating in the merits hearing often will be fixed before there is information exchange, and a time-consuming dispute about which persons other than outside counsel may view produced documents threatens to disrupt set schedules for the submission of pre-hearing witness and expert statements and in turn the agreed merits hearing dates.

Will it be an advantage to the arbitral process for the Tribunal to insist upon or at least express its strong preference for a presumptive one-tier confidentiality protocol (“confidential” but not “highly confidential”)? Placing the initial burden to justify a higher tier of confidentiality on the producing party should ordinarily discourage over-designation of documents at the “highly confidential attorneys’ eyes only ” tier, by forcing the party to justify the designations before making them and to do so in such a time that the schedule fixed for information exchange can be met. The Tribunal’s burden of deciding designation disputes should be reduced, and more of the initially set procedural timetables will be able to be met without alteration.

  1. Production Inadvertently of Privileged Communications

One further provision that has become part of the “boilerplate” of stipulated confidentiality orders in complex US federal civil litigation, and may warrant different handling by arbitrators, concerns the consequences of allegedly inadvertent disclosure of data that is subject to a claim of attorney-client privilege. US Federal Rule of Civil Procedure 26(b)(5)(b) specifies how the party in receipt of such data shall act upon being notified by a producing party of a claim of inadvertent disclosure, and how that party may contest the claim before the Court. Generally the recipient must embargo all use of the data, apply to the Court if it wishes to contest the privilege claim, and maintain the embargo until and unless the challenge to the privilege claim is resolved. Some parties represented by US counsel will be wont submit a draft Protective Order that, without thoughtful adaptation to the arbitral context, essentially incorporates Rule 26(b)(5)(b), and if the draft is approved as a ministerial act of So-Ordering by the Tribunal there are at least two not necessarily desirable consequences: (1) a US rule of civil procedure designed for judicial application is adopted into the arbitration without serious consideration for its suitability in the arbitral context, and (2) a foundation is laid for the party asserting a claim of privilege to maintain that US attorney-client privilege law has been enshrined as the applicable law in the arbitration. These consequences would be regrettable in an international arbitration where the applicable privilege law is an open and seriously debatable issue, and where the privilege law that the Tribunal may find to be applicable to a particular communication may treat inadvertent disclosure in a fashion quite different from the American model.

The US law of inadvertent disclosure of privileged communications is constructed around the concept of “claw back” — a restitutionary equitable remedy that seeks to put the disclosing party in the same position as if the communication had never been revealed (although such restitution is necessarily imperfect, as the recipients’ recollections of the documents’ contents, absorbed without knowledge of the possibility of a claim of privilege, cannot be erased). Rule 26(b)(5)(b) is designed to preserve the effectiveness of the “claw back”, for the duration of any dispute over the right to that remedy, by imposing a stringent moratorium on use of the communication by the receiving party, who must return, destroy or sequester all copies of the communication. But foreign law of attorney-client privilege that the Tribunal might determine to be applicable might not recognize the “claw back” remedy.  The foreign law might simply provide that factual information in the communication may be used as evidence but legal advice reflected in the communication may not be used as evidence and the inadvertent disclosure entails no subject-matter privilege waiver. Even in a US domestic arbitration where the applicable privilege law is not in doubt, the Tribunal might prefer a different method of adjudication, for example shifting the burden to initiate a claw back claim to the disclosing party and addressing the question of interim restriction of the recipient’s use of the communication on a case-by-case basis. Thus in lieu of the boilerplate Rule 26(b)(5)(b) clause in the submitted Protective Order the Tribunal for example might prefer: “A party aggrieved by an alleged inadvertent disclosure of a communication subject to a claim of privilege shall prompt apply to the Tribunal for relief, making full disclosure of all relevant circumstances, and specifying the remedies sought and the law applicable.”

  1. Compliance Obligations of Service Providers for the Arbitration

Another typical provision of litigation-based Protective Orders addresses the relationship to the Court of a non-party service provider, such as an expert witness. When the arbitral tribunal receives from counsel a proposed Protective Order addressing this question, it will often include a subscription form to be signed by the non-party service provider, whereby she agrees to be bound by and to comply with the terms of the Protective Order and to submit to the jurisdiction of the Court/Tribunal for its enforcement. Whereas the power of arbitrators to impose disciplinary measures upon counsel who appear in the case remains a controversial and unsettled domain, arbitrators should be reluctant to assert by approving such an Order that they might in appropriate circumstances impose a sanction upon a consulting firm or a document management firm, engaged by a party to the arbitration or by the parties jointly, for a violation of the Protective Order. Would it not be more sensible for an arbitral Protective Order to provide that a party by contracting with third parties for support services that entail exposure to confidential information bears full responsibility for the service providers’ compliance and accepts that the tribunal may impose sanctions upon the party for the provider’s violation?  The service contract between the party and the provider can in turn be drafted/modified to provide for shifting of responsibility to the provider in certain circumstances, with disputes in this regard to be resolved in such fashion as the parties to this contract might agree. In this fashion substantially effective compliance of service providers with the Protective Order should be achieved without the prospect of a controversial extension of arbitral disciplinary powers to non-parties.


These are but a few of the more prominent provisions in arbitral Protective Orders that have their origins in US litigation Protective Orders adapted as drafting templates. Arbitrators who recognize the adaptability issues will be well positioned to deal with them, and to made suitable changes even when a proposed Order has been submitted as a stipulation between the parties.

Leave a Reply