Posted Mon Nov 30th 2020 12:37 pm by Tom Goldstein
Long-time readers of SCOTUSblog are now familiar with the legal search tool from Casetext. It solves a pervasive need on our team: to find opinions at every level of the court system for our articles and case pages. Practitioners reading this blog, on the other hand, face a different need in their daily work with the law. Instead of searching cases by name, attorneys need a way to search the case law to find support for specific suggestions. This task is challenging not only because the common law is very extensive, but because the judges use different articulations for the same sentence or principle.
Casetext meets this huge challenge with its new tool: parallel search. In contrast to simple keyword search, the limitations of which most of us are very familiar with, Parallel Search uses machine learning technology to match complete phrases and sentences with those with similar meanings in the case law, even if the results and the query are almost nonexistent Words contain common.
It’s so powerful that users have described the technology as “pure witchcraft” and “almost … like a fraud” (although it certainly isn’t).
Do you need a federal appeals court to rule on a protest case? If you search for an argument like “The right to protest is fundamental to American democracy,” you get court-specific results that a traditional search would never find: opinions that say, “Citizens’ rights, by theirs.” Filing a motion to redress complaints is fundamental to government ”or“ A fundamental principle of the First Amendment is that citizens have the right to disapprove of government policies. “
The secret is to apply the latest machine learning technology through casetext to the law. Transformers are artificial neural networks that “learn” by being trained on a preselected data set. The parallel search uses two layers of transformers, each trained in advance of all US case law. The result is not just accurate. It’s quick, and most importantly, it’s affordable.
Parallel Search was introduced as part of Casetext’s Compose shorthand solution and was so popular with users that it is now available as a standalone solution. Interested? A free trial is available, as well as full pricing information and a practicing attorney demo.
Tom Goldstein, legal research no longer limited to keywords,
SCOTUSblog (November 30, 2020, 12:37 pm), https://www.scotusblog.com/2020/11/legal-research-no-longer-limited-to-keywords/