By Elizabeth Adams
Hajime Sato’s #ATA62 presentation on eDiscovery was a valuable opportunity to learn more about an industry that is a source of interesting and challenging work for linguists. While eDiscovery is usually a pretrial procedure for preserving and analyzing electronically stored information (ESI), it can also be a tool for responding to a government investigation or ensuring regulatory compliance. As a Japanese translator, Hajime has been involved in discovery since the days when clients used to lock him in a room with boxes of documents stacked to the ceiling. Today, he is Manager of Linguistic Services at International Litigation Services, where he helps plaintiffs overcome language barriers and lower the cost of discovery through the intelligent use of technology. After giving an overview of the industry and its size ($12 billion in 2019 and expected to reach $24 billion by 2026), Hajime walked attendees through the stages of eDiscovery and highlighted all the potential roles for linguists beyond translation.
Hajime explained that the stages of eDiscovery fall into two categories: up to production, where the producing party is busy collecting the information requested, and after production, where the receiving party has to make sense of it all. Once the producing side expects litigation, its attorneys will issue a litigation hold, which basically tells employees not to delete their emails. Then, the two sides will negotiate an ESI protocol that covers important issues like the format in which information will be produced and the search terms used. Sato explained that search terms are controversial because the producing party only has to turn over documents responsive for those terms. This is an area where case law is evolving rapidly. After the producing party loads its data to a review platform, runs the search terms, and filters the results for responsiveness, attorney-client privilege, and personally identifiable information, the ESI is produced as agreed, usually as a hard drive or flash drive.
At this point, the action moves to the receiving party, which loads the data to its own platform for review. The plaintiff’s attorneys look for what they call “hot documents” that may help them answer litigation-related questions and draw up lists of people they want to depose. Hajime explained that an eDiscovery vendor can help plaintiffs make the most of their budget by using technological advances such as technology-assisted review (TAR) and real-time predictive coding to cut the human hours involved in wading through terabytes of information. Artificial intelligence learns from the work of human reviewers to automatically classify and retrieve relevant information. Another strategy is to build a latent semantic index showing words and phrases that occur in proximity and may be useful for searching the database.
Machine translation is useful in early stages after production when attorneys are homing in on a strategy and want to scroll through lots of documents. However, Hajime shared that because of data security concerns, companies like ILS prefer to keep data inside their firewall and avoid cloud-based solutions.
Human translation jobs in eDiscovery range from the gist translations used to code documents to the certified translations eventually filed with the court. Other roles for linguists include interpreting, whether for depositions or at trial, and foreign language document review. During document review, linguists help attorneys identify key terms and how they are used at the company in question (including jargon and acronyms), identify the cast of characters and potential deponents, construct a timeline, code documents, and pull materials from the database to prepare for depositions. Hajime gave the example of the term “airbag,” which has multiple renditions in Japanese, all of which a linguist would record, as well as any slang terms for airbag that they found in the ESI. The names of individuals and departments are also tricky because they are often abbreviated in internal discussions. The review linguist has to also be a researcher.
Hajime advised linguists wanting to do more eDiscovery work to acquire a narrow specialization or learn how to use relational databases and review platforms. He also warned that eDiscovery work is unpredictable: you can be working full steam ahead and get shut down by a settlement. Here, as in any contract work, flexibility is key.
If you missed Hajime’s presentation, watch it on the conference portal at https://ata62.hubilo.com/community/#/session-stream/100587.
And mark your calendars: December 3 is eDiscovery Day, an industry event featuring free webinars and live talks where you can deepen your knowledge of the topics Hajime touched on and plan out your CPE focus for the coming year. Follow the Association of Certified eDiscovery Specialists (ACEDS) on LinkedIn for live talks and check their website for upcoming webinars.
About the Author:
Elizabeth Adams is a Russian to English translator of law and fiction who lives near Seattle. She studied Russian at The College of William and Mary and is ATA-certified for Russian to English.