Inventus is producing a webinar on 16 July called Accelerating International eDiscovery: The Challenges of Multilingual Litigation. The speakers are Dominic Piernot, eDiscovery Consultant Germany/France at Inventus, John Tinsley, CEO of Iconic Translation Machines, and Jérôme Torres-Lozano, Director of Professional Services at Inventus. The moderator is Sarah Brown, eDiscovery expert at Inventus.

There is information about this, and a registration form here.

To supplement the webinar, Inventus has published a two-part article on multilingual edisclosure by Jérôme Torres-Lozano called It’s all Belgian fries to me: the art of multilingual disclosure. Part 1 is here and Part 2 is here.

Jérôme Torres-Lozano was brought up bilingual, and his first job was working with a major translation corporation. He now brings specialist language skills to electronic disclosure / eDiscovery at Inventus.

Let’s pause first on the expression “multilingual eDisclosure” in Jérôme Torres-Lozano’s headings. Leaving aside the tiresome complication that the Civil Procedure Rules of England and Wales introduced the term “disclosure” to a world already comfortable with “discovery”, it is good to see that “multilingual” has displaced “foreign” in this context. To US eDiscovery people, everything not in English used to be described as “foreign”. Leaving aside (again) the fact that the US has appropriated the term “English” for its own variant of the tongue, more than 40 million people in the US have Spanish as their first or main language. Furthermore, however insular many American citizens are, its commercial market is a multinational one, and “foreign” seems a paradoxically imperialistic expression to use. We should welcome the term “multilingual”.

Looked at in process terms, the presence of multiple languages in discovery sources has become a largely mechanised affair as software has become capable first of detecting multiple languages and identifying them, and secondly of translating them.

Jérôme Torres-Lozano does a good job of making this sound interesting as well is important. Here is an example:

Then it all got a bit clever, using dictionaries, but as always with languages, nothing is that straightforward. Little use is a dictionary when you need to differentiate the Spanish word fresco from the Portuguese, Italian, Dutch, or English fresco (meaning fresh, insolent, wall painting). Languages within the same family tree share many common roots and attributes, which made automated differentiation and detection a challenge.

The first article sets the scene with points like this. The second one describes the technology developments, and covers the increasingly important area of audio discovery. I recall, as an articled clerk (trainee solicitor), listening to a recording of a meeting on one of those old piano key cassette recorders, stopping and starting the tape in chunks small enough to transcribe. As Jérôme Torres-Lozano points out, regulated industries like finance and insurance require the capture of audio data, and some pretty sophisticated technology is needed to search and review the volumes which result. A call to almost any organisation these days begins with a drab-voiced warning that calls may be recorded “for training purposes”; what they really mean is that if a dispute arises about any ensuing contract, they may transcribe the call to produce evidence equivalent to written communication. Audio translation and transcription matters.

Machine translation can take you only so far. Human reviewers – often lawyers with language and industry-specific skills as well as legal ones – may be needed for those documents which are or may be significant. The use of machine translation in this context parallels the use of technology-assisted review and similar technology – it does the grunt work, at the least weeding out documents of no relevance, leaving a smaller pool of documents or recordings which must be examined manually.

If the worst thing you can do is to ignore documents or recordings in languages other than your own, the next worst is to ignore the availability of tools and services which can cut through the dross and point to the information which matters. Jérôme Torres-Lozano’s articles are a helpful primer on the subject. The webinar is likely to be similarly useful.

Home