June 2023 - Modern E-Discovery: Using Linguistics to Improve Language Analysis

Anyone who has seen the movie “The Princess Bride” undoubtedly remembers the character Vizzini, who often exclaimed “inconceivable!” when events did not go as expected. Eventually the swordsman Inigo Montoya replied to his boss, “You keep using that word. I do not think it means what you think it means.”

Words, and their meanings, matter a great deal to lawyers, who tend to have a strong command of language. While collectively they’ve earned a reputation for being slow adopters of new technology, many lawyers are recognizing how modern tools can enhance language analysis during the e-discovery process.

AI and machine learning tools are highly sophisticated, and as they continue to advance, the complexity of how they work can be challenging to understand for even the brightest attorneys. But they are incredibly useful. These tools apply models to cull collections of documents, divide them into ranked sets from the most likely responsive to the least likely and make coding decisions that require quality control, but not necessarily a full-scale review.

Lawyers might better appreciate how these models work by understanding they’ve been built by some fellow masters of language – professional linguists. As experts in the scientific study of language, linguists have played an integral part in the development of machine learning for data analytics and e-discovery. Among other things, they ensure that computer models capture the essence of language and the rules we follow when we speak or write.

Skeptics are often more willing to consider the use of machine learning tools once they recognize there is more than just computer code involved and thoughtful experts who understand the nuances of language are the ones creating these models. This is what often initiates dialogue about the advantages and accuracy of current and future software platforms.

The combined expertise of lawyers and linguists is what makes the application of these tools successful. Linguists understand language analysis, in terms of the what and the how of the words, while lawyers understand the why – why certain content would be important for an agreement, for producing documents pursuant to a discovery request and for identifying significant content that could help or potentially hurt a client for a given matter.

The legal industry has come a long way since simple keyword search tools were state of the art for accessing crucial content in a large set of documents. But data sets keep getting larger, and current tools, while already highly sophisticated, must continue to advance to keep up with the ever-growing volume. Coupling this with the constant changes in the way we use language, the words we choose and the definitions we apply, lawyers will need technology partners to continue to isolate key content buried in a sea of unimportant emails and documents. Linguists will be vital participants in those partnerships, ensuring that legal teams have a true expert voice translating their needs.

QuisLex’s David Olener wrote about this topic in more depth recently in an article published by Legaltech News. Read the full text here: The Art and Science of Language: Understanding the Role of Linguistics in Modern E-Discovery.