Facing Multilingual Litigation in e-Commerce

Prepare for International Document Handling

Despite an increasingly globalized business environment, very few organizations have a strategy in place to deal with non-English documents, and that can have serious litigation consequences, particularly in the U.S. Without the inclusion of foreign language documents in their search, companies can never be sure they are producing the right documents at the production stage, which can lead to unpleasant surprises later on in the process. However, the execution of including non-English documents in searches can be difficult since proper translation takes time and accuracy and defensibility of the information are crucial elements in the process. In a high-stakes regulatory investigation or e-discovery case, there is often little room for a time-consuming translation process.

The last decade has seen e-discovery regulations that allow sanctioning for even the smallest accidental omissions, particularly where there is other perceived “bad behavior” like stonewalling, lack of legal hold, non-cooperation and other behavior such as incomplete productions. Companies are starting to acknowledge that without robust text analytics solutions that help identify all relevant documents that are needed for a case, legal departments and law firms will have great difficulty discharging their Federal Rules of Civil Procedure (FRCP) 26(g) “reasonable inquiry” obligations.

Multilingual e-Discovery Technology

Every law firm or enterprise that has faced a large-scale international dispute or a regulatory investigation understands that technology is indispensable when you need to collect all possible relevant information. This information then needs to be further researched, analyzed and structured in a transparent and defensible way so that the results can be shared and used with other parties.

While solutions exist that can automatically classify and analyze large collections of data to drastically decrease time and cost, many legal professionals may not be aware of how this technology can also help address the challenge of translating documents and communications in multiple languages easily and quickly. The following example explores how a U.S.-based law firm used machine translation technology to help address its multilingual challenges.

A Case Study

A U.S.-based law firm provided legal counsel to a multinational chemical company based in Spain. As a national U.S. firm without local Spanish resources, the firm was struggling with its translation needs. Its collection partner, a large accounting firm, was sending tens of thousands of documents that needed to be translated before being processed and reviewed by the firm. The client team needed a way to quickly process and translate the foreign language documents to ensure that all relevant data was identified and produced.

The Solution

The law firm utilized the ZyLAB e-Discovery solution, integrating SDL automated machine translation technology, SDL BeGLobal. To help process the foreign language documents, ZyLAB's software detects and recognizes 400 languages, including all Western European, Eastern European, Russian, Chinese, Japanese, Arabic, Persian, and most regional indigenous languages. The solution then quickly translated all the information immediately, similar to the way many firms pre-tag based on known search terms. The translations then were reviewed so that legal teams could quickly uncover relevant data and route critical information for a specialized human translation if needed.

The Results

Using this combined solution, the law firm was able to process the required documents within the strict time parameters of the investigation. The firm was also able to save time and money by using functionality that allows for the exclusion of documents from the review by keyword. This way, key documents were identified sooner, giving counsel time to include them in the case strategy. Another key benefit was that the process was highly defensible because the tagging was logged in the same database as the translations, the remaining data reduction and review calls. In addition to delivering on the case requirements at a quality level they could validate, the firm was also able to keep more work in house, rather than having to turn the majority of the review over to another law firm. In doing so, the firm also reduced the cost of human translations by 90%.

In this case, the option to use human translation only for the handful of documents required ' rather than the tens of thousands delivered ' was made available through the technology. The case was able to proceed without any delay on the part of the client, and without incurring a six-figure human translation bill.

In the integrated language recognition, automated translation and text analytics solution described above, content is first ingested into the e-discovery platform, where it can be stored for search and analysis. Following this, the option to detect a language and automatically translate it is chosen. The content is then processed through the integrated machine translation so it can be reviewed in English or any other selected language. For example, in the ZyLAB and SDL BeGLobal integrated solution used in the case study above, translated text is treated in the same way as text extracted from OCR (optical character recognition) software: as part of the metadata of the original document. This means that a search in English or the other selected language(s) will bring up the original document, no matter the original language.

An alternative approach would be to machine translate selected documents during the review stage. This can be particularly helpful when it is not known beforehand what multi-lingual documents exist in the collection.

Upon production, in the case study above, it was up to the case team to decide whether to provide the translations with the extracted text. For case exhibits, the team used human translation for the documents, and had the translators provide an affidavit or declaration before introducing them into evidence.

Multilingual Content from Social Media

e-Commerce entities should make sure the e-discovery solution they use is able to translate multilingual content such as social media, Web content and enterprise data in real-time. This empowers businesses to rapidly identify and respond to critical global trends and customer insights, allowing these organizations to improve sales, increase ROI and even enhance customer experience.

Mary Mack, Esq.'is Enterprise Technology Counsel for ZyLAB. Formerly Corporate Technology Counsel for Fios, Inc., she is a graduate of Northwestern University School of Law and a member of the Illinois bar. Mack is the author of A Process of Illumination: The Practical Guide to Electronic Discovery, published by Fios.

International borders are no longer a barrier in e-commerce business. Enterprises often have multinational corporate parents, subsidiaries, suppliers and customers. Even smaller companies can develop e-commerce websites and reach out to prospects in all corners of the world.

However, while the “world” has become the accessible market, business requirements have also changed. With global communication, offices and customers spread across different countries around the world, companies have a much greater need to access and analyze data in multiple languages. This alone is a challenging task for most, but when it comes to a company's legal requirements, the uncertainty over whether the content of a document in a foreign language is critical or unimportant can drive costs up and trigger timeline overruns.

With the growth of global business and the accessibility of online information, law firms and corporate legal professionals are faced with increasing challenges when handling international, multilingual legal cases. When an external counsel or a corporate lawyer becomes involved in cross-border disputes or regulatory investigations, it is hard enough to fully grasp the complete spectrum of international litigation issues and implications. But when these cases come in multiple languages, things only get more complicated.

Prepare for International Document Handling

Despite an increasingly globalized business environment, very few organizations have a strategy in place to deal with non-English documents, and that can have serious litigation consequences, particularly in the U.S. Without the inclusion of foreign language documents in their search, companies can never be sure they are producing the right documents at the production stage, which can lead to unpleasant surprises later on in the process. However, the execution of including non-English documents in searches can be difficult since proper translation takes time and accuracy and defensibility of the information are crucial elements in the process. In a high-stakes regulatory investigation or e-discovery case, there is often little room for a time-consuming translation process.

The last decade has seen e-discovery regulations that allow sanctioning for even the smallest accidental omissions, particularly where there is other perceived “bad behavior” like stonewalling, lack of legal hold, non-cooperation and other behavior such as incomplete productions. Companies are starting to acknowledge that without robust text analytics solutions that help identify all relevant documents that are needed for a case, legal departments and law firms will have great difficulty discharging their Federal Rules of Civil Procedure (FRCP) 26(g) “reasonable inquiry” obligations.

Multilingual e-Discovery Technology

Every law firm or enterprise that has faced a large-scale international dispute or a regulatory investigation understands that technology is indispensable when you need to collect all possible relevant information. This information then needs to be further researched, analyzed and structured in a transparent and defensible way so that the results can be shared and used with other parties.

While solutions exist that can automatically classify and analyze large collections of data to drastically decrease time and cost, many legal professionals may not be aware of how this technology can also help address the challenge of translating documents and communications in multiple languages easily and quickly. The following example explores how a U.S.-based law firm used machine translation technology to help address its multilingual challenges.

A Case Study

A U.S.-based law firm provided legal counsel to a multinational chemical company based in Spain. As a national U.S. firm without local Spanish resources, the firm was struggling with its translation needs. Its collection partner, a large accounting firm, was sending tens of thousands of documents that needed to be translated before being processed and reviewed by the firm. The client team needed a way to quickly process and translate the foreign language documents to ensure that all relevant data was identified and produced.

The Solution

The law firm utilized the ZyLAB e-Discovery solution, integrating SDL automated machine translation technology, SDL BeGLobal. To help process the foreign language documents, ZyLAB's software detects and recognizes 400 languages, including all Western European, Eastern European, Russian, Chinese, Japanese, Arabic, Persian, and most regional indigenous languages. The solution then quickly translated all the information immediately, similar to the way many firms pre-tag based on known search terms. The translations then were reviewed so that legal teams could quickly uncover relevant data and route critical information for a specialized human translation if needed.

The Results

Using this combined solution, the law firm was able to process the required documents within the strict time parameters of the investigation. The firm was also able to save time and money by using functionality that allows for the exclusion of documents from the review by keyword. This way, key documents were identified sooner, giving counsel time to include them in the case strategy. Another key benefit was that the process was highly defensible because the tagging was logged in the same database as the translations, the remaining data reduction and review calls. In addition to delivering on the case requirements at a quality level they could validate, the firm was also able to keep more work in house, rather than having to turn the majority of the review over to another law firm. In doing so, the firm also reduced the cost of human translations by 90%.

In this case, the option to use human translation only for the handful of documents required ' rather than the tens of thousands delivered ' was made available through the technology. The case was able to proceed without any delay on the part of the client, and without incurring a six-figure human translation bill.

In the integrated language recognition, automated translation and text analytics solution described above, content is first ingested into the e-discovery platform, where it can be stored for search and analysis. Following this, the option to detect a language and automatically translate it is chosen. The content is then processed through the integrated machine translation so it can be reviewed in English or any other selected language. For example, in the ZyLAB and SDL BeGLobal integrated solution used in the case study above, translated text is treated in the same way as text extracted from OCR (optical character recognition) software: as part of the metadata of the original document. This means that a search in English or the other selected language(s) will bring up the original document, no matter the original language.

An alternative approach would be to machine translate selected documents during the review stage. This can be particularly helpful when it is not known beforehand what multi-lingual documents exist in the collection.

Upon production, in the case study above, it was up to the case team to decide whether to provide the translations with the extracted text. For case exhibits, the team used human translation for the documents, and had the translators provide an affidavit or declaration before introducing them into evidence.

Multilingual Content from Social Media

e-Commerce entities should make sure the e-discovery solution they use is able to translate multilingual content such as social media, Web content and enterprise data in real-time. This empowers businesses to rapidly identify and respond to critical global trends and customer insights, allowing these organizations to improve sales, increase ROI and even enhance customer experience.

Mary Mack, Esq.'is Enterprise Technology Counsel for ZyLAB. Formerly Corporate Technology Counsel for Fios, Inc., she is a graduate of Northwestern University School of Law and a member of the Illinois bar. Mack is the author of A Process of Illumination: The Practical Guide to Electronic Discovery, published by Fios.