Acquired Foreign Flavor versus Natural Local Taste

Over the last decade, public reaction in Japan to foreign writing (ranging from literature through business books to technical manuals) translated into Japanese has changed from acceptance of hard-to-understand "translatese" to demand for easy-to-digest, natural-sounding text. For a long time awkward-sounding text was accepted as a reflection of the "foreignness" of the original writing and readers put up with it. The tide has changed, however, and readers now want foreign text to read as naturally as any Japanese book, demanding a shift from the source text-oriented to the target text-oriented approach.

Achieving this requires that differing degrees of cultural adaptation take place, which in turn can lead to quirky ethical issues, particularly with literary works. In the world of product localization, however, such considerations will be absolutely essential prerequisites to successful sales. In fact, those who are involved in Asian localization seem to be rapidly realizing the serious damage cultural pitfalls can cause. Cultural considerations are part and parcel of translation practice in general, but technical translation has been regarded as largely immune from such factors, and thus the ideal target for machine translation (MT), even with its highly literal translations.

However, while computer manuals come under "technical translation", the localization process as a whole really goes beyond this. Until recently, the main concern for Asian localization appears to have been how to tackle double byte character enabling (DBCE) and treat character sets, rather than the cultural aspect. But, as the industry becomes familiar with the procedure (this is not to say that DBCE is now a non-issue for everybody), attention may be shifting toward more intangible aspects of how to adapt the product to a given cultural paradigm. A survey of recent articles on localization issues reveals highlighting of the importance of the cultural aspect in localization practice ("Cultural Assumptions in Software" (Language International 8.5 1996); "Japan's Expanding Localization Market" (Computing Japan, Feb./March 1996); "New Age of Localization" (Computing Japan January 1997); "Culturalizing Software for Japan" (Multilingual Computing 8.2, 1997)"; "Helping Multimedia Go Multilingual" (Language International 9.2, 1997).

The cultural factor is harder to grapple with than technical problems such as DBCE, since it is often abstract and difficult to define exactly; there is no standard code for cultural parameters. And yet cultural incompatibility of products may lead to failure of the whole localization effort, almost as much as technical incompatibility would. Recent well-known faux pas include Microsoft's Spanish edition of Word 6.0 and Korean version of Encarta CD-ROM, which contained some disagreeable bits for the respective cultures of their target markets. This also applies to localization of Asian products, as in Matsushita Electric's last-minute withdrawal of "Touch Woody" (named after Woody Woodpecker) as the name of the touch-screen interface for its multimedia PC because of a certain connotation in some parts of the US.

As localization user markets mature, they will no longer readily accept mechanically localized products that lack cultural sensitivity, or that are hard to use because they lack familiar conventions. Cultural considerations will become a key element in maximizing the localization opportunities potent in Asian markets. In order for a bestseller in one country to become one in another country with different language and cultural values, localizers increasingly need to take cultural dimensions into consideration.

There is, of course, always a danger of overkill by excessive cultural adaptation through naive oversimplification, whereby stereotypical views are implemented. This is where expert advice is needed to keep a fine balance between the foreign flavor of the original product and the right amount of local familiarity. The value added by localization will in future increasingly involve "culturalization".

How to Make it Culturally Compatible

The "cultural" dimensions which need to be addressed in the localization process may range from the base color of the VDU screen and text and the position of buttons and icons to documentation style and content selection, as well as the name of the product. The requirements will obviously vary from product to product and depend on markets. However, there are some broad strategies localizers could take. For example, as a step towards producing localized products acceptable in the target market, some localizers are already carrying out "in-country" (target language country) translation and localization. As is well noted by the industry, this process will be particularly necessary for markets such as Japan, where resistance against "lesser quality" products is strong. While the technical "transcoding" process of localization may be done elsewhere, culturally critical components will need to be advised and checked primarily by in-country industry experts. The benefit of establishing a reliable relationship with in-country localizers or advisors therefore cannot be overemphasized. Any suggestions for major cultural adaptation should ideally occur right at the start of the project so that reasonable budget, time and resources can be allocated, and the expert consultation should be built-in for most localization projects.

In reality, however, the wide variety and the large volume of localization work as well as price competition often necessitate a wider talent pool outside of strictly in-country resources. Or, in some cases, there is no such thing as in-country experts, as pointed out by Microsoft's Mr. Haynes (LISA Forum Asia, December 1996) in reference to the firm's efforts in developing the Vietnamese market. In my personal experience, having organized middle- to large-scale Japanese translation projects in New Zealand, the best results often came from combining local talents (expat Japanese) who understand all facets of the source text (produced by a New Zealand company) with in-country Japanese editors who have currency in the target language in a given field. Today's efficient communications technologies mean that culturally compatible products using "in-country" expertise can be achieved not only by large localization companies with extensive overseas resources, but also by small-scale localizers able to attain these goals in a cost-effective way with flexible resource usage.

Some localizers are already making terminology databases and TM (Translation Memory) facilities accessible via telecommunications to teleworking translators. The ability to make translation tools available wherever translators are will become increasingly important. Reliable communications channels and procedures mean that the localization team members do not have to be gathered in one physical place. While maintaining the basic "in-country" principle, the translator, editor, engineer and tester may still be in geographically diverse locations. Such a distributed production process in turn will require suitable QA/QC, as well as project management models which will differ from the one based on one-location production. While human talents play a major role in contributing to the cultural dimensions in localization, it is also becoming increasingly essential to apply technology to meet the requirements.

Localization and Cyberspace

The growth of the Internet and the Web in particular is significantly impacting on the translation industry as a whole. For example, translation operators are taking advantage of the Web both to offer a wide range of language services to worldwide customers and to recruit translators on a global scale. Translation demand is also arising from the Web itself, which will eventually represent a significant portion of the work. While monolingual Web sites are pushing some individual users to use Web MT (desktop products or on-line services), an increasing number of business Web sites are being localized into a multitude of languages. When this task also includes maintenance of the sites with regular updates and simultaneous changes in multiple languages, it will clearly exceed a single translator's capability-it will require a team of localization parties working together on different components of the task with sophisticated IT tools. Added interactivity and extensive use of multimedia on the Web, and widespread adoption of the hypertext format in all kinds of documentation in the future will mean new challenges for translators and localizers.

In this environment, the time and volume pressure for language service providers will further intensify, and the use of productivity tools for translators will become essential. Teleworking translators are gradually equipping themselves with IT tools to respond to opportunities which are opening up. The Internet is allowing the job to go where the talent is, irrespective of location. Future freelance translators may all have today's high-end translator's workstation, which integrates all necessary CAT (computer-aided translation) components plus MT (machine translation) functions. Translators' skills in using technology will have a progressively greater correlation with producing quality work.

While the future of fully automatic MT seems as yet uncertain, the option to use some kind of CAT systems is on the rise, particularly in the localization sector. Having said that, there is a strong sign of a developing market for a low cost real-time automatic translation mechanism that matches today's browser functionality. Skin-deep localization by MT in the immediate future will be restricted to rudimentary information assimilation rather than to the dissemination of information. Nevertheless, MT may still have a far-reaching impact on the localization business by offering alternative "quick and dirty" solutions that will force lower prices and shorter time frames on localization jobs.

Returning to the cultural issue, the real edge localizers will be able to offer over instant MT localization will be their ability to craft products which are both technically and culturally acceptable; localizers who do not go beyond mechanical localization may risk their livelihood in the long run. Designer localization calls for creative human talents assisted by appropriate technologies. Localization work is far from cut and dried; it requires a great deal of human creativity. When a good job is done, there is no reason why localized products cannot be better than the originals or competing domestic products. The current market share race between JustSystem's Ichitaro 7 (the Windows 95 version of Japan's bestseller word processing software) and Microsoft's Japanese version of Word 95 may offer some interesting insights into how localized products can gain acceptance from even the most discerning consumers.

Into the New Millennium

My current research is about the future of cyberspace and its impact on the translation industry. I am examining a technological possibility called HyperReality (a prototype of which has been developed in Japan) to merge virtual reality (VR) into physical reality. While VR cuts off the relationship with the physical environment, this new capability will link VR into real reality. The most primitive form of this concept is the telephone, through which we hear the virtual voice of a person who is elsewhere, while at the same time maintaining awareness of our local environment. But today's telephone lacks the ability to convey visual information and tactile senses. One cannot literally "reach out and touch someone", unlike a certain telephone company's ad.

HyperReality, on the other hand, will allow the telepresence of people and objects with whom and which one can interact. The core engine of this technology is a database of pre-recorded images of people, objects, animals and other environmental elements, and a broadband communications superhighway to transmit information of interactions (speech and movements) in real time to superimpose on the pre-recorded images accordingly. The end result is that your communicating partner will be projected in 3-D right in front of you, and both of you can manipulate virtual objects interactively.

Potential applications of this technology include telemedicine, virtual resort and virtual university. The implication of this concept for the localization industry is that the pre-recorded databases containing people, objects, or other artificial habitats will need to be localized to make them available in any requested language and cultural context. This entails not only verbal communication in text and speech, but images and non-verbal communication. This kind of need will test the intricate understanding of the more deep-seated cultural dimensions on the part of localizers. The Japanese virtual pet "Tamagotchi" dies if it's neglected, whereas its localized American version just goes into cyberspace! Some parts of the cultural psyche are certainly more baffling than double byte encoding.