August 03, 2003

Katakana Revolution

The Japanese language is in trouble. The arrival of the Digital Age finds it increasingly at the mercy of the media and the marketplace, each better equipped today than at any other time in history to shape society, culture, and the modern vernacular. The rush towards globalization and eager pursuit of the technological tools that facilitate it have created in Japan an environment of indiscriminate assimilation, where the foreign appellations for emerging technologies are cut-and-pasted from English directly into the various Japanese media. The language of Nippon is being subtly transformed through a reckless frenzy of linguistic borrowing, and rather than enrich the language, this katakana revolution will ultimately only dilute and pollute it.

The source of most of the recent imports is the computer industry at large and the proliferating media that report on it. With the lion's share of computer manufacturing, R&D, and software production being conducted in the United States, it follows that a correspondingly large portion of the latest computer terminology is English. Magneto-Optical drive. Reduced Instruction Set Computing. Web browser. America is the birthplace of the majority of the coinages, and consequently English has become the lingua technotica of the Digital Age.

This propagation of modern language across national borders is inevitable, and various countries respond to it in uniquely different ways. In France, legislation has been enacted to check the introduction of foreign words by requiring their translation into French. Exporting similar English terms to Chinese, on the other hand, faces no government opposition, but does require the creation of corresponding new terms using hanzi , or Chinese characters. In either case, the process involves veritable translation from English to the respective languages. The Japanese language, however, is equipped with a script designed specifically for rendering foreign words, or gairaigo, called katakana.

The function of gairaigo in Japanese might be compared to that of the cache in a computer CPU or disk drive, where "data" is stored until the system has the time or resources to process it. For much gairaigo enough time has passed (decades or even hundreds of years) since its introduction that it has simply become part of the language. However, the technological vocabulary flooding into Japan today is unknown to the vast majority of Japanese speakers. Held indefinitely in the cache these terms exist in a kind of linguistic limbo, belonging to neither the original language nor Japanese.

As we hurtle toward the 21st century the world by degrees grows smaller, and the borders that divide us are crumbling. The primary agents of change in this transformation--mass media, information technology, and the Internet--have arrived in Japan, and they come laden with more than the promise of increased productivity and CRT-driven entertainment. This modern Black Ship will initiate sweeping changes in the Japanese language, the repercussions of which may not be realized until too late. In the course of this article I will highlight some of the characteristics of what I call the Katakana Revolution, and explain why it behooves us to pay close attention to the changes it is causing in both Japan and Japanese.

Background

Gairaigo (loan words) have been a part of the Japanese language for centuries. The first standardized characters used by the Japanese were kanji imported from China in the 6th and 7th century. Even the first native script, the phonetic kana syllabary that the Japanese created to address the unsuitability of the Chinese characters to express highly-inflected Japanese, is based on the simplification of selected kanji. The transfer of language, science and culture from China to a long-isolated Japan continued for hundreds of years. As a result of this prolonged influence from the Middle Kingdom, Japanese came to contain thousands of words of Chinese origin. Much like the vestiges of Latin and Greek that exist as prefixes and suffixes in English, these words (called kango) predominate in the early sciences and medicine because the transfer of knowledge in these fields coincided with the transfer of their lexicon . Centuries later but in the same fashion, the period of rapid Westernization known as the Meiji Restoration was characterized by the enthusiastic importation of European language. The flood of words borrowed from the West at that time was encouraged by the newly-formed Meiji government as a means of accelerating the internationalization process, and received little popular opposition. Even putting aside the considerable borrowings from China and the West there still remains gairaigo imported from Korean, Sanskrit, and the language of the aboriginal Ainu.

Considering Japan's history of linguistic borrowing, it stands to reason that little impetus would exist among most observers, Japanese and otherwise, to criticize or oppose the modern growth of gairaigo, disregarding it as only the most recent and conspicuous phase in the natural evolution of the language. Ignoring the trend on the basis of this logic, however, may not necessarily be in our best interests. Nor should we assume that we are somehow relegated to passive observation, for since the changes I will discuss are taking place today rather than in some remote historical niche we are afforded the opportunity to evaluate, criticize, and even influence the way in which the Japanese language evolves, or does not.

Some Problems with Gairaigo

Gairaigo is, by definition, "language from outside," and distinguished in the written language by the use of katakana. Like hiragana, katakana is a phonetic syllabary capable of expressing every phoneme in Japanese. Where hiragana characters are cursive and flowing, katakana are sharp and angular. This appearance lends them an aesthetic quality that many Japanese, especially the younger generations, find appealing. Words written with katakana are imbued with a visual quality that is at once modern, foreign, hip and cutting-edge.

According to research conducted a decade or so ago, over 10 percent of Japanese is gairaigo, and the numbers are growing. New words are created all the time, the vast majority of them culled from English. Certainly the mass media are in part responsible for propagating gairaigo, but especially influential is Japan's monolithic advertising industry which, in its zeal to entice consumers, employs English or English-based constructions--often with little regard for English grammar or context--in eye-catching slogans that enjoy vast popular exposure, plastered as they are on TV screens, billboards and subway car interiors like societal wallpaper.

The creation and dissemination of gairaigo leads to a number of problems for speakers of Japanese, native and otherwise. Were it always simply window dressing or advertising copy there might be little reason for concern. However, language is by nature a communication medium, and this is where the fundamental problems with gairaigo lie.

  • Gairaigo is without intrinsic meaning

Importing a word from another language into Japanese katakanafication, for lack of a better term) is a simple process. One need not even know the meaning of the word that is being imported, merely the sounds. The phonemes (units of sound) of the original word are modified, slightly or a great deal, to correspond to the closest equivalents found in Japanese. As Japanese is a language with fewer distinct sounds than English, for example, this usually results in a word that resembles the original only somewhat or not at all.

It is important to note that the purpose of this process, though, is not to strictly adhere to the original English, but rather to create a word that can be vocalized using Japanese sounds and represented in written Japanese. For example, If the imported word comes from English, the Roman characters are discarded and replaced with katakana that represent the sound of the word in Japanese syllables. So, using this system to convert, say, "multimedia" into gairaigo results in マルチメデイア , or maruchimedeia (mah-roo-chee-mee-deh-i-uh).

One obvious problem with this example is that the meaning intrinsic to the English word multimedia, namely multi (from the Latin multus, meaning much or many) and media (plural of media), has been completely eliminated in the conversion process. Where once the composition of the word provided a measure of insight into its meaning, all that remains for the Japanese reader is a random collection of sounds. Unfortunately this loss of meaning is a regular byproduct of katakanafication, and thereby interferes with the transmission of meaning between languages in the importation process.

  • Gairaigo often displaces standard Japanese

It is much easier to remember and reproduce katakana than kanji, which can be extremely complicated and number in the thousands. Accordingly, many speakers and writers of Japanese develop a predisposition for using foreign words in place of established yet more difficult Japanese terms. Younger generations especially, with many more hours in English classrooms or in front of the television, are more likely to possess a lopsided, gairaigo-rich vocabulary and greater difficulty reproducing and reading uncommon kanji. Because of this trend many gairaigo words are replacing standard Japanese, and in general terms this amounts to a shift away from kanji to the less expressive katakana. This is a problem for the same reason cited in the previous example: loss of intrinsic meaning. Just as prefixes and suffixes allow English speakers to surmise the meaning of many English words, Japanese has the corresponding benefit of kanji.

Most kanji impart meaning to the reader either ideographically or pictographically, and are thus an excellent medium for conveying new concepts. It is for this reason that a Japanese (or Chinese, or Korean) reader can make an educated guess about the meaning of a word rendered in kanji without having ever seen it before, provider he is familiar with the characters that comprise it. Katakana, by contrast, simply represent (foreign) sounds, and so initial comprehension of gairaigo requires an appended explanation, a dictionary, or sufficient familiarity with the source language.

In both scenarios, when English is imported as gairaigo or when gairaigo displaces existing Japanese, a loss of the intrinsic meaning occurs that reduces much of the words' usefulness as tools for communication.

  • Gairaigo is an obstacle to language learning

Gairaigo undermines and impedes the difficult process of language learning for both Japanese students and foreigners studying Japanese. For Japanese students, English is compulsory from middle through high school, a total of six years. The spoken English taught in the classroom (usually by Japanese instructors) has been subjected to katakanafication for the benefit of Japanese speakers, and thereby stripped of many of the very sounds necessary for aural comprehension by native speakers. As ESL Professor Joseph Sheperd writes in The Internet TESL Journal,"Stressing every syllable and adding a vowel at the end of the word, [Japanese students of English] often sound as if they are reading Katakana placed alongside of the words."

Add to that the problem of misinterpreting the huge number of English words already in use as gairaigo and it is no mystery why so many Japanese share a common dilemma: the inability to communicate orally in English. The scope of this problem might best be measured by the flourishing presence of eikaiwa (English conversation) schools, a booming industry in Japan which markets courses that teach students how to correctly pronounce the English they have been studying for all these years.

Gairaigo can be a pitfall for the foreign student of Japanese as well. Native English speakers are probably confronted with the greatest number of loanwords from their own language, and tend to lean toward the original pronunciation and interpretation of these words rather than that used by Japanese. Equally vexing is the fact that although a word may have numerous meanings distinguished by context in English, often only one is used in Japanese. Additionally, since grammar is often ignored in the importation process it is not uncommon for nouns to become verbs, prepositions to become nouns, etc. Taken together, these characteristics contribute significantly to the difficulty of learning Japanese.

  • Gairaigo is renegade

The growth of gairaigo is out of control, and the creation of new words goes completely unchecked and unregulated. Many of the words used today by youth and pop authors are unknown to well-educated, older Japanese. Aside from guidelines for converting foreign sounds to those found in Japanese, there are no rules that limit or regulate the creation of new gairaigo.

This is indeed a problem for the reasons outlined above and others, but more important is the scale of change taking place today. Certainly, the process of importing and assimilating foreign words and language is the natural result of inter-cultural communication and not worrisome in and of itself. However, the rate of change and the sheer number of new words arriving from abroad today has reached crisis proportions, and has only been made possible through the technological advances of recent decades.

Gairaigo and the Information Age

The global diffusion of communications technology in the latter half of this century has ushered in an Information Age, where digitized data, culture, and language streak continuously across obsolete borders via satellite broadcast or the ubiquitous Internet, and instantaneous communication through a variety of media is simply taken for granted.

The Internet is especially relevant in any discussion of Japan, a country which has experienced 84% growth in the number of Internet hosts since January of this year (Source: Network Wizards InterNIC survey, July 1996). Japan is the fastest growing large-scale Internet market in the world, and at the current growth rate will surpass both the United Kingdom and Germany to occupy the number two position after the United States in only two years. To be sure, Japan is getting connected. But what does this portend for the Japanese language?

The Net was born in America, as was much of most recent and most significant computing technology. Although once upon a time the language of science and technology in Japan was based on Chinese, the vocabulary of the Digital Age, spawned in the computer rooms of Bell Laboratories and UC Berkeley, is overwhelmingly English. Virtually any Japanese computer magazine today (and there are many) is filled with technical jargon that has been imported part and parcel from English. Detailed discussions of video interlacing techniques or parallel processing are rife with specialized English terms and acronyms whose meaning would be lost on most native English speakers, yet are present nonetheless in katakana form.

The use of katakana and this method of dealing with gairaigo reveal a growing trend for choosing transliteration over translation in the fields of engineering and computer technology. The practice, however, is a modern contrivance that seems curious when you consider similar technological advances earlier this century and the way in which their jargon was adapted to Japanese.

Take, for example, RF electronics. When the technology fundamental to modern cellular telephony, RADAR, and radio was first developed early in this century, new vocabulary were created for concepts such as electromagnetic waves, signal modulation, RF propagation, etc. At that time the translation of RF terminology into Japanese was done mostly conceptually, using existing words or characters to render the new terms. The use of kanji facilitated this process by conveying the ideas associated with the words rather than the sounds. This method allowed Japanese, or anyone with a knowledge of kanji, to understand to a certain degree the meaning of the new terms without having been exposed to them previously. Consider the examples of fundamental RF terminology as they exist in English and Japanese in the left-hand portion of Table 1:

Table 1. Translation Past and Present
RF Terminology
Computer Terminology
frequency
周波数
network
ナットワーク
wavelength
波長
peer-to-peer
ピアツーピア
amplification
増幅
multimedia
マルチメデイア
bipolar
正反対
microprocessor
マイクロプロセッサ
wireless
無線
modem
モデム
reception
受信
throughput
スループット
transmission
転送
memory
メモリ
phase
移送
fixed disk
ハードヂスク
FM
周波変調
protocol
プロトコル

Electromagnetic energy is described in terms of "waves" because as it travels through space is exhibits many properties associated with liquid waves including oscillation, peaks and troughs, repetition frequency, and wavelength to name a few. Just as in English, the wave concept for RF energy exists in Japanese terminology in the form of the character 波 (nami, or wave). Fundamental as it is to our perception of electromagnetic energy, it naturally appears often in the list of examples above.

Next, if we turn our attention to the right-hand portion of Table 1 and study some the words that are being created to express basic concepts related to computing and networking, it is clear that no effort has been made to translate the English words into corresponding Japanese. Instead, merely the sounds of the original English have been imported. Although Japanese characters could have been employed to translate the concepts above, thereby making comprehension of the words far simpler for literate Japanese, but transliteration was chosen instead.

The obvious problem with this pattern is that the Japanese public, confronted with the current flood of new technological ideas and terminology, are being forced to absorb them in what is essentially a foreign language. In the excited dash towards globalization and the economic potential it holds for a resource-weak Japan, the Japanese will be making the trip on linguistic crutches. Whose idea was this, anyway?

Agents of Change

The long-awaited ascension of the personal computer in Japan has begun, and what for so long has been exclusively a tool of the workplace is becoming increasingly common in living rooms from Sapporo to Kumamoto. It has finally become a toy, a surfboard, and a desk reference. In short, the computer has become, well, personal. PC fever rages in Japan and the Internet is red hot. The computer industry here and abroad is frantically striving to meet the demands of the burgeoning Japanese PC market, and competition among vendors and manufacturers is fierce.

Microsoft has entered the fray with typical determination and adroitness, and through a combination of shrewd partnerships and savvy marketing has slowly but effectively wrested the Japanese software market from the grip of industry leaders such as Just Systems and NEC. Windows 3.1 and 95 are now the industry standard operating systems for Intel-based PC's, and Microsoft's flagship applications in the three categories that count--word processors, databases, and spreadsheets-- have secured the number one slot in each. Microsoft Office commands an impressive 51% of the office suite market as of May 1996 (Source: Business Computer News, Computer News, Inc.). None of this phenomenal success in the foreign Japanese market, however, would have been possible without localization.

Localization, according to Ken Lunde's Understanding Japanese Information Processing, is "the process of adapting software (or hardware) such that it conforms to the expectations of a specific country. This often includes rewriting menus and dialogs into the target language, but sometimes involves more complex changes, such as handling special character encoding methods." Mr. Lunde also employs the term Japanization in the same book to refer to the localization of software for the Japanese market. However, japanization and my own coinage katakanafication are not interchangeable terms. Here is why.

Windows95 and the new 32-bit versions of MS applications clearly reflect the current trend of exploiting katakana to merely transliterate, not translate, from English to Japanese in the course of program localization. The programs are rife with examples of words that aren't English but rather Microsoft copyrighted coinages that have been converted to Japanese through katakanafication. Not surprisingly, the resulting ambiguities often lead to frustration on the part of the mystified Japanese user. In a recent issue of the Japanese computer magazine Interface, Osaka University Professor Satoshi Kawata recounts his frustration at trying to use the "Japanese" version of MS Excel. "It's so difficult to use I can hardly stand it, " he laments. Commands like "Add-in Manager" and "Pivot Table" that have simply been rendered in katakana are, according to Professor Kawata, "hard to read and difficult to understand."

And it's not just Microsoft. If anything, some of the other software giants, Novell Inc. and Lotus Corp. among them, are just as guilty of this disservice to the Japanese consumer. User manuals for the Japanese version of cc:Mail, for example, are notoriously difficult to read, filled as they are with mysterious constructions like ピアツーピア (Peer-to-Peer) and スタンドアロン (Stand Alone). Are we to assume that the localization team for this project was incapable of producing corresponding Japanese terms from the canonical expanse of the language? Is "stand alone" such an arcane, inscrutable concept that these translation professionals eventually threw up their hands in defeat and settled on katakana?

This practice would not present a significant problem were it not for the fact that these products, notably those coming out of Redmond, Washington, will almost certainly come to dominate the Japanese software market. The natural consequence of this is that the terminology chosen by them and employed in their application software and user manuals will become the industry standard and, eventually, part of Japanese itself through simple mass exposure. The reason for this is that application software is the birthplace of the computer argot. A decade ago in the US the words "cut and paste " or "double-click" would have fallen on mostly deaf ears, but today they help comprise the foundation of a fast-growing digital idiom. This pattern is inevitable, and gives monolithic corporate entities like Microsoft an unsettling degree of power to influence and define the development of language.

Realistically speaking, software manufacturers will always place their own interests first when choosing the verbiage used in their programs. The expectations and background of the user, for example, must be considered in order to produce a product that is easy to use yet powerful, and therefore marketable. In some cases this may mean standardizing the terminology and placement of often used menu commands (such as "Open File" or "Select All"), but in other cases the motivation may be less altruistic. By defining the lexicon of the user interface at the OS level and then dictating that other software producers conform, industry leader Microsoft has been able to foist its own copyrighted terminology on the industry as a whole. A brief perusal of Windows 95 environment yields a handful of examples: Properties, Shortcut, Toolbar, Wizard. The dictatorial standardization of these terms by Microsoft is designed to condition the computing masses to be familiar with and comfortable in the Windows environment. Ultimately, this is simply another means for assuring their current and future dominance of the computer industry. In Japan the objective is no different, and the extremely popular "Japanese" version of Windows 95 comes shipped with all of the important MS terms predictably intact: プロパテイ , ショートカット , ツールバー , ウイザード , among others.

Unfortunately, however, the interests of the computer industry obviously have very little to do with considering the long-term impact on language or the evolution of translation methodologies. By endorsing transliteration as a viable alternative to translation in their localization efforts, the products of which will enjoy singular prominence on desktops in homes and offices throughout Japan, these titans are burdening Japanese consumers with unnecessary ambiguity and corrupting the language, all the while telling translators and anyone else paying attention that it's okay to do so.

What is perhaps most perplexing about the current flood of gairaigo and the computing sector's active role in it is the apparent lack of criticism of the matter. One among the few voices of concern in the Japanese media is that of Professor Kawata. In his Interface article, he advocates the use of a system called Ruby to address the various problems associated with the use of gairaigo in Japanese. The system basically involves using katakana in superscript form above foreign terminology (much like furigana), be it a computer term or a Chinese person's name, in order to both preserve the components of the original word(s) and provide the correct pronunciation in Japanese. The idea is a good one, and certainly addresses some important issues, but I would argue that we need to go further.

Attempting to reduce the number of copyrighted terms and coinages in software programs would be futile because these terms often distinguish the various vendors from one another and are thus essential for market competitiveness. In other words, quibbling over the transliteration of a Microsoft-ism like Wizard into the Japanese version of Word© would be a waste of time. The area that more deserves our attention is general terminology that isn't trademarked and exists abundantly in the industry vernacular. These are words that seem to have mystified the translation community and as such are in copious use as simple gairaigo. These are words like digital, hypertext, browser, client-server, macro, peer-to-peer, display, PIM, OCR, chip, refresh rate, memory, controller, and about a zillion others. Omitting laziness, incompetence, or a particular love for katakana there is no good reason why these terms cannot be rendered in authentic Japanese.

It is possible, of course, that the problem is simply a lack of consensus. Since no governing body exists for handling the creation of new words, a lack of direction or guidelines in the area of localization is hardly surprising. For many in the translation field, for example, the use of katakana as a kind of interim translation may be less intimidating that actually coining a new word and offering it for public scrutiny and evaluation. With this in mind and in hope of contributing to positive change, I offer the following as rough guidelines for those who work in the areas of localization and translation:

  • The serious work of creating language should not be taken lightly, and the decision whether to translate or transliterate is an important one whose consequences should be considered well. In terms of ease of comprehension, which is better for introduction to native Japanese speakers, an unknown (katakana) gairaigo term, or an unknown Japanese word rendered in kanji or hiragana?

  • Since countries such as China and Taiwan use Chinese characters as well and do not have the convenience of a phonetic script for handling foreign terms, an occasional look at Chinese translations could prove instructive.

  • Translators working on the bleeding edge of technology, the source of most of the latest terminology, should be willing to consult with those who specialize in language itself when considering new terms. Linguists and educators are well-placed to contribute positively in the translation process, and their input should be considered invaluable.

  • Consensus among all of the parties involved--engineers, translators, journalists, and linguists alike--is the best means at our disposal for ensuring that the changes we make to Japanese today will be in the best interests of the speakers and users of Japanese.

In conclusion, let me say that what motivates me in this area is nothing less than a love for Japanese and the profound dismay I feel seeing it casually discarded more and more often in favor of gairaigo. Japanese is an amazingly rich language with great expressive power, and there is no good reason why it should be supplanted by English in the 21st century.

It is my hope that the people more in a position to make a difference will share my concerns and help restore translation as the standard. Should they choose to do otherwise, to leave the fate of Japanese to Bill Gates or the staff writers of computer magazines, to sit idly by as the myriad problems related to gairaigo multiply, and to wring their hands in dismay when the dictionary fails to produce a corresponding term, they will ultimately count themselves as accomplices in the dumbing-down of Japanese and all of Japan.

Posted by denbushi at August 3, 2003 11:11 PM
Comments