REUTERS | Jean-Paul Pelissier

Holding back the tide: the rise of the machines in arbitration

In the era of big data, parties need to ensure their digital approach to arbitration is diligent.

The ever-increasing volume of electronically stored information (ESI), which is potentially discoverable in modern disputes, affect arbitration no less than traditional document heavy High Court litigation. It is not uncommon for a party to be allowed to conduct a broad-ranging fishing expedition, resulting in the obligation for his or her counterpart to review hundreds of thousands, even millions of documents.

At the same time, there is a proliferation of new technology which is available to assist with the investigation and review of ESI. This can be broadly described as technology assisted review (TAR). The legal industry appears to be reluctant to commit fully to new solutions, particularly those which clearly have the potential to increase efficiency and reduce costs. This may be due to an abundance of caution, fear of being the “first mover”, and a simple lack of understanding of the potential applications of technology on the part of decision makers.

Disclosure model

The Disclosure Pilot Scheme in the Business and Property Courts is now more than half a year in. It is too early to say how enthusiastically the parties responded to the court’s strong encouragement of the use of TAR during the disclosure process. As things stand, the pilot requires that, where the disclosure model to be used requires searches to be performed, the parties must discuss and seek to agree, and the court may give directions on, amongst other things, the use of software or analytical tools. This includes:

  • TAR software and techniques.
  • Coding strategies, including to reduce duplication.
  • Prioritisation and workflows.

The court may also make orders requiring the use of specified software or analytical tools in the disclosure process.

Significantly, the disclosure review document requires that the parties explain why they have decided not to use some form of technology/computer assisted review. This is a significant change of position on the part of the courts. It is based on the assumption that technology has advanced to the point at which it is presumed to be effective.

However, it is not clear that arbitrators and arbitral parties in evidence heavy disputes have fully engaged with the logistical difficulties of the review and disclosure process in the age of big data. There is no detailed and up-to-date procedure, protocol or methodology as to the most efficient and effective use of new technology, although the Sedona Principles, little known outside of the United States, are ahead of the curve. The IBA Rules on the Taking of Evidence in International Arbitration contain no specific guidance in this regard and have not been updated since 2010. It must be hoped that the next iteration will contain more detail in this area. In the meantime, discovery remains an area in which tribunals appear to be loath to take a proscriptive approach.

Reasonable positions

Reaching a reasonable, agreed position in relation to the disclosure process in arbitral proceedings often appears to be very low in the list of the parties’ priorities. Experience shows that tribunals do not tend to encourage parties to adopt forward-thinking, cost-effective approaches to the management, assessment and review of ever-increasing volumes of electronic data.

A determined approach is required from one or both parties, rather than paying lip service to the use of new technology. In practice, the party which is likely to bear the major burden of discovery should take the front foot in this regard and make efforts early in the proceedings to build the case for the use of TAR. This is so that ample time is allowed for the issue to be debated before the discovery process begins.

Stage one: Scoping of available documents

As soon as possible after receiving the statement of case, the respondent should work with its lawyers and, if necessary, an e-discovery specialist to scope the nature and extent of potentially relevant ESI, and begin the collection process. In practice, this is often left too late, potentially even after document requests have been received. It may seem like unnecessary front-loading of costs, but the information gleaned at this stage is critical for the purposes of timetabling and assessment of the likely future workload. It is usually helpful for the legal team to work alongside an e-discovery provider not just for technical support, but also to explain the usefulness of this stage to the client, who is likely to be reluctant to front the cost without an understanding of the benefits of a proactive approach.

Stage two: Apply early case assessment (ECA) software

Once the collection process is complete, the data gathered can be run through analytics software. This will assist in identifying data volumes, custodian names, relationships, date ranges and key concepts, all of which can be displayed in a detailed report. Some version of this report can also be deployed in the proceedings at the next stage, in order that any proposal regarding the deployment of TAR can be explained and quantified in an easily digestible way.

Stage three: Build a case for the user of TAR

The TAR technology which is currently available cannot replace human review. However, it is likely to assist greatly in the prioritisation of review and largely eliminate the need for extensive review of documents, which are unlikely to be relevant. The TAR strategy should include key elements of transparency and accountability. It is likely to be something about which the other party and the tribunal will want to be regularly updated. However, if the focus of the strategy is saving time and cost whilst maintaining quality, a tribunal will be wary of dismissing it.

A party which takes this approach should focus on examples of decisions in the UK courts in which TAR has been encouraged, and in which its benefits have been recognised, for example Pyrrho Investments Ltd, David Brown v BCA Trading Ltd and Triumph Controls UK Ltd and another v Primus International Holding and others. The following principles should be noted:

  • TAR has been shown to lead to more accurate disclosure than manual review and keyword searches.
  • When TAR software is effectively “trained” by a senior lawyer, it can provide a more consistent approach than numerous individual, less experienced, document reviewers.
  • Cooperation is recommended between the parties as to the criteria to be applied to the TAR process. Unilateral and un-agreed use of TAR is generally not advised and may lead to poorer quality results than manual review alone.
  • The party or parties using TAR should be prepared to provide regular updates regarding the TAR process.
  • A full audit trail of the use of the TAR software is necessary in order to demonstrate that the software has been used correctly.

Leave a Reply

Your email address will not be published. Required fields are marked *

Share this post on: