As Asia-based corporations continue to expand their footprint across the globe, these businesses are increasingly involved in legal disputes staged in foreign jurisdictions where litigants’ legal duties and obligations may differ substantially from duties litigants face in their home jurisdictions. The differences in parties’ legal duties are especially acute when legal disputes are brought in US courts. Corporations and their counsel often struggle to understand the revised U.S. Federal Rules of Civil Procedure relating to e-discovery and the growing body of case law interpreting those rules. Counsel also has a pressing need to stay abreast of the growing body of both domestic and U.S. federal and state laws concerning data privacy and international data transfers in addition to the vocabulary used when discussing the discovery of electronic evidence. Corporate executives who are adequately informed of the legal complexities surrounding the collection of electronic data in jurisdictions where their company maintains a presence are best equipped to manage legal and business-related considerations involved in global e-discovery. Today’s corporate leaders need practical guidance about technology and best practices to enable in the efficient and accurate filtering, processing, review, and production of global and multilingual discovery.

Recent amendments to the U.S. Federal Rules of Civil Procedure include changes intended to facilitate disclosure and discovery of electronically stored information, including emails, documents, and spreadsheets. The complexity of issues involved in the collection, review, and discovery of this information is compounded when the sought-after documents are stored in multiple jurisdictions or contain multiple languages.

Data privacy laws and concerns come to the fore in such circumstances. Within Asia, data privacy laws vary widely from country to country and great care must be exercised when preserving, collection, and transferring any data that may fall under the perview of one or more of the many data privacy regulatory frameworks in the region. Notably, Japan and South Korea have among the most strict data privacy laws. While the People’s Republic of China has the draft Personal Information Protection Measures consultative document, little centralized legislation controls the collection, processing, and transfer of personal data. Of special interest is Hong Kong’s Personal Data (Privacy) Ordinance, which protects the privacy interests of living individuals in relation to personal data. However, Section 33 of the Ordinance, “Prohibition against transfer of personal data to place outside Hong Kong” is officially unenforced at the present time and no date has yet been appointed for the commencement of its operation.

The European Union has a complex network of legal rules to protect individual privacy rights. Because European Union data privacy laws are more stringent than those of many other jurisdictions, the European Commission’s Directive on Data Protection prohibits the transfer of personal data to non-European Union nations that, like the U.S., do not meet European Union standards for privacy protection. Because notions of data privacy are different depending on the country of origin of a document,issues relating to the collection, review, and transfer of data across borders are increasingly fraught with legal peril.

The complex legal issues involved in international discovery are illustrated by a March 2008 case from the U.S. Eastern District of New York, Strauss v. Credit Lyonnais, in which Magistrate Judge Kiyo Matsumoto rejected a more stringent test for the discovery of foreign documents and compelled a French bank to produce documents to victims of a terrorist attack in Israel. In ordering the disclosure of the foreign documents, the court carefully weighed numerous factors, including the importance of the documents to the litigation, the degree of specificity of the request, the availability of alternative means of securing the information and the extent to which noncompliance would undermine the interests of the United States. Additionally, the court weighed each country’s privacy interest in the sought-after documents, each Defendant’s hardship in complying with Plaintiffs’ requests, and whether Defendant acted in good faith.

As the Strauss case illustrates, issues of discoverability and transferability of documents kept in other jurisdictions are often convoluted and difficult to discern—even for a Federal District Court Magistrate. However, the analysis does not end once documents are deemed discoverable. Corporations and outside counsel must still grapple with questions such as preferred methods of document collection, filtering, processing, review and production.

Collection of Data from Countries Across the Globe

Once documents are identified and deemed discoverable, various collection techniques must be considered. Keeping in mind the technical aspects of international data collection, parties must also adhere to specific legal requirements of a particular country or countries. Collection of foreign or multilingual data also involves tactical consideration of a variety of other issues, including the location of the data, who will be involved in its collection, tools to be employed in foreign data collection, and regulations concerning its collection and transfer.

Filtering and Processing Data in Multiple Languages

Once data is collected, it must be filtered and processed into reviewable form. For example, collected data must be culled to remove documents that are not relevant to the lawsuit (for example, operating system and program files necessary to run applications like Microsoft Word or Excel). Further, paper documents may be scanned and loaded into a database also containing electronic documents.

The filtering and processing of multilingual data (i.e., documents containing two or more languages) is a more complex task. As a threshold issue, languages must be identified and organized before text may be searched and culled. The highly technical nature of these tasks is best appreciated when realizing that there are literally hundreds of languages that the technological tool must be able to recognize and that many languages use a right-to-left format (such as Hebrew and Arabic) rather than a left-to-right format (such as English and most Western European languages). In addition, linguistic blocking of written text varies substantially for different language families. For example, East Asian languages differ from Western European languages in the quantity of characters and script, use of spaces, and lack of differentiation between singular and plural, gender, and verb tenses.

Reviewing Multilingual Data

Once the collection and processing of documents is accomplished, other challenges await in the review phase. Before the first document is examined, one must be determined whether and when to translate the multilingual documents into English or another language for review.

Review teams have two options: review documents in the native language(s) using native language reviewers, or translate the documents into the language of review for review by reviewers proficient in the review language. A hybrid option also exists, combining a native speaking review and a translated review on various data subsets.

Human translation tends to be the most accurate but also the most costly, so human translation is best when a small number of documents are at issue. However, it is important to consider the subject matter expertise of the translators. A professor of Korean literature may be of minimal assistance translating engineering specifications, for example.

Machine translation involves higher percentages of error, usually dependent on the language being reviewed and the level of formal versus informal speech in the document set. But because it is less expensive, machine translation is often used on larger volumes of data and in cases where a search system cannot support non-English character sets or handle accented characters.

An optimal balance of accuracy and economy may be struck with a hybrid system, where documents are first translated by computer, after which important documents and documents likely to be produced are identified and marked for human translation.

Documents may be reviewed before translation by employing a review team that is fluent in the language of the data set. This option likely entails significant expenditures of both time – to find qualified, experienced lawyers who are fluent in the relevant language(s) – and money. The cost of hiring capable and fluent review lawyers may be staggering when there are large quantities of documents to be reviewed. However, the cost may be justified when there are a relatively small number of documents to review, the documents have significant importance, or when the stakes are high.

Producing Multilingual Data

Fortunately, multilingual data does not present many new issues once a party reaches the point of document production, although special attention should be paid to the format in which documents are produced.

The U.S. Federal Rules of Civil Procedure require data to be produced in a format that is reasonably usable by the opposing party. The format options for producing multilingual documents are the same as for other electronic data: in their native format (i.e., the files themselves), in a load file (i.e., images of the documents with associated information), in an online repository (generally the most cost-effective option), and in print format. However, pros and cons accompany each format option. For example, if producing documents as native files, the receiving party must have the software necessary to open the files and then recognize and render the language characters contained within those files. If producing printed documents, the printers must have the proper print drivers installed so that all characters may be represented correctly on paper.

Staying abreast of all of the new challenges associated with modern corporate litigation is difficult, especially when that litigation is in another jurisdiction such as the U.S. The law and technology concerning the collection, filtering, processing, review, and production of foreign and multilingual data is in its infancy. Over time, law in jurisdictions will be better defined, technology will advance, and corporate executives and outside counsel will be better equipped to make informed decisions regarding the legal and technical aspects of global electronic discovery. In the meantime, the ability to make tactical business decisions and control litigation costs require an understanding of relevant terminology and obstacles attendant to international discovery, along with an appreciation of technological opportunities and best practices involved in the exchange of foreign and multilingual data.

Kenneth V. Nourse is Managing Director, Legal Technologies, Asia Pacific at Kroll Ontrack. A former U.S. litigator, he consults with corporate and law firm clients on all aspects of electronic evidence-related issues. He may be reached at knourse@krollontrack.com.