TECHNICAL JAPANESE TRANSLATION

PrevIndexNext

No. 20 — October 10, 1984

The newsletter is published primarily for the far-flung and obscure community of Japanese technical translators but is also aimed at a broader group of readers: all those who have scholarly and professional interests in the Japanese language and in Japanese technical and scientific information. The newsletter is now in its second year of publication and has readers on three continents. The editor's main goals are twofold: to use the newsletter to focus attention on the importance of Japanese scientific and technical information and to win greater public recognition for the crucial role which technical translators have to play in making this information accessible.

Next issue will be the last issue of this six-month's period. The editor will announce in next month's issue whether or not the newsletter will be continued in its present form for another six months. (For more on this, see pp. 47–48, this issue.)

HOW TO SUBSCRIBE

Your subscriptions are now for a 6-month period from June (no. 16) to November, 1984 (no. 21). The 6-months' subscription is $20. Overseas readers should contact the overseas distributor for their area.

A complete set of the first year's issues (no. 1 through no. 15) is available from the editor for $20 (U.S.) or $25 (Japan and Europe).

ASSOCIATE EDITORS

Readers' letters and contributions are very welcome. Readers who wish to help out can volunteer to be associate editors. They will be asked to write one or two articles a month. The editor will supply them with the raw material (Japanese-language press clippings, reference materials, etc.). Volunteers should write me giving details of when they will be available.

Associate editor for this issue is Frederik L. Schodt. Many thanks to him and to the other contributors writing in this issue, including Edward E. Daub, Hisashi Kubota and John Bukacek.

OVERSEAS DISTRIBUTORS

The following readers are reproducing the newsletter and distributing it to others in their countries:

Mr. John M. Shields, BEC Service, C.P.O. 42, Nagoya 450, Japan

Ms. Rosemary J. Yates, Japan Business Services Unit, The University of Sheffield, Sheffield S10 2TN, England

* * *

FUJITSU MARKETS ATLAS SYSTEM

Fujitsu Corporation announced on September 20 that it has developed and will market its ATLAS system for English-Japanese and Japanese-English machine translation. This is the first commercialized MT system for translating in both directions: E-J and J-E. The maximum speed is said to be around 60,000 words an hour with the FACOM M-380 mainframe computer. Thus, the Fujitsu system is faster than a system developed by Hitachi, Ltd. (see the article below), but the translations are not perfect and still require human editing.

Fujitsu claims that a page of about 300 English words would take a human translator 40 minutes to translate into Japanese, and it would take another 15 minutes to prepare the final draft of the translation on a word processor, adding up to a total of 55 minutes. If the ATLAS system is used, inputting and editing will take on an average 22 minutes, and the machine will take only 2 minutes to do the translation. All in all, the machine system will allow a time saving of 60%. However, the system cannot operate by itself and requires an operator who is proficient both in English and in the technical vocabulary. A less skilled operator would take longer to edit and input the text.

The ATLAS system is designed to run on Fujitsu's FACOM M series of mainframe computers and also on its FACOM S-3000 series of super-minicomputers. The software has two parts: ATLAS-I for E-J, and ATLAS-II for J-E. The monthly lease is 350,000 ($1,434) for ATLAS-I and 550,000 ($2,254) for ATLAS-II. ATLAS has a basic vocabulary of about 50,000 words, but specialized words and new words can be added by the user. Optional dictionaries of scientific and technical terms in various fields such as electricity and chemistry are also available for a monthly fee of 150,000 ($614) each.

Deliveries of ATLAS-I will begin on September 20 for some systems, and deliveries of ATLAS-II will begin in June, 1985. (Nikkan kōgyō shimbun, Nikkei sangyō shimbun and Nihon keizai shimbun for September 21, 1984)

Incidentally, Fujitsu says that ATLAS is an abbreviation for Automatic Translation System. Fujitsu will demonstrate ATLAS-I at the Data Show '84 exhibition to be held at Harumi in Tokyo beginning on September 26, according to a one-page advertisement I saw in the Nihon keizai shimbun of September 22, 1984.

* * *

HITACHI AND QUICK DEVELOP E-J MACHINE TRANSLATION SYSTEM

Hitachi, Ltd. has announced that it has developed an English-Japanese machine translation support system for economic articles. It developed the system jointly with the Quotation Information Center, K.K., a database which supplies an on-line service called QUICK for economic and market information.

The system has a dictionary of about 70,000 words and is designed to translate English articles about economics into Japanese at a speed of about 20,000 words an hour after inputting. It is now being used internally by the database company and will later be applied in the QUICK service to supply foreign economic news in Japanese translation.

The system for QUICK will use a Hitachi mainframe computer called HITAC M-240H and will have keyboard inputting. The operator will be able to make corrections of the resulting translations on a multi-window screen displaying both the English and the Japanese.

Hitachi says that it will also complete development of a Japanese-to-English translation system for science and technology sometime this year. (Nikkan kōgyō shimbun and Nikkei sangyō shimbun September 19, 1984)

* * *

IBM JAPAN DEVELOPS PROTOTYPE E-J TRANSLATING SYSTEM

IBM Japan's main research institute, the Science Institute, has developed the prototype of a machine translation system for translating English into Japanese. The Institute claims that the system has a rapid translating speed and can iron out some grammatical errors in the original. The system uses a large-size mainframe computer, the IBM 3081. IBM Japan plans to refine the system further and to put it into practical application sometime this year. However, it will take at least another two years before the system can be marketed commercially.

In a test, the prototype machine translated, 1,500 computer messages (the messages which a computer addresses to the user) with a 93% accuracy. IBM Japan says that it can translate an English sentence of 20 words in 2 seconds.

The main feature of this system is that the three steps which a machine translation system goes through to translate (grammatical analysis, word substitution, and creation of the translation) are performed simultaneously, rather than consecutively. Thus, the semantic analysis which was performed twice in the past systems (when analysing the original and when creating the translation) is here performed only once. Moreover, analysis starts with the parts which are easiest to analyze. This gives the system a speed some 3–5 times faster than previous systems. (Nikkei sangyō shimbun, August 30, 1984)

* * *

JPO INTENDS TO USE CAS ONLINE SERVICE; JICST'S PARTICIPATION IN CAS INTERNATIONAL NETWORK NOW IN DOUBT

According to the August 30, 1984 issue of the Nikkan kōgyō shimbun, the Japanese Patent Office has decided to "utilize fully" the on-line network of Chemical Abstract Service (CAS) of Columbus, Ohio beginning in October. The U.S. PTO has proposed adopting the CAS system of classification for chemicals substances in the joint studies to be carried out by the Japanese, American and European patent offices about establishing a tripartite "paperless" patent system. The JPO decided to introduce the CAS online service in order to keep in step with the PTO. The Japan Association for International Chemical Information (JAICI), which is CAS's representative in Japan, will add an additional communication line between Tokyo and Columbus for exclusive use by the JPO.

As reported in a previous issue (no. 18, p. 1), the Japan Information Center of Science and Technology (JICST) previously decided to participate in STN International, an international network being organized by CAS. If JICST joins it, the new online network will link American, West German and Japanese databases. However, JICST has been dragging its feet. Moreover, now the JPO will become the largest Japanese user of the CAS online services, and this move is sure to have an effect on JICST's participation in the international network. While agreeing to participate in the network, JICST and its parent agency, the Science and Technology Agency, have failed to come up with any clear-cut attitude because of their fear that, if JICST participates in STN International, it will have to change its current JOIS software over completely to the CAS software. (See issue no. 17, p. 6–9.)

JAICI is strongly urging JICST to participate in STN International and has been hinting that if the STA and JICST continue to drag their feet on the matter, it may appeal to the Ministry of International Trade and Industry, which is the parent agency of the Patent Office.

This bickering raises an entirely new possibility: a CAS-MITI collaboration. When JPO joins the CAS network, Japan may participate in STN International in a quite different way from what was originally assumed. The JPO has a budget of approximately ¥130 billion (roughly $537 million at a recent exchange rate) to spend for building its "paperless" patent system (see issue no. 18, p. 3), and it intends to spend one-tenth of this amount on software to be developed for it by the Japan Patent Information Center (JAPATIC). Thus, there is a quite sufficient groundwork for building an international network with CAS under the hegemony of MITI, centering in JAPATIC.

Under these circumstances, says the Nikkan kōgyō shimbun, if the Science and Technology Agency really intends to continue fostering JICST in the future as Japan's one and only comprehensive center of scientific and technical information, it will have to provide JICST with some resolute support so that JICST will be fully able to tangle with CAS.

According to the Nikkei sangyō shimbun of September 11, 1984, even now the on-line service of the STN International network is available to users in Japan through JAICI, but this is a one-way arrangement lacking reciprocity (foreign information flowing into Japan is freely available, but there is no outgoing information from Japan to foreign countries). If this situation continues, Japan is sure to attract overseas criticism for making full use of overseas databases while supplying none of its own information to overseas users. Japanese industries depend on the CAS databases, which currently contain 6,500,000 items about chemistry alone, while JICST's databases altogether contain about 1,200,000 items including subject areas other than chemistry. If the supply of information from CAS were to be shut off, Japan's pharmaceutical and chemical industries would be crippled.

According to the Nikkei sangyō shimbun, the following difficulties arise in including the JICST online service in the STN International network. First, the software and the command terminology must be unified so that the databases can be accessed by the same method in Japan, the U.S. and Germany. JICST would have to change over to the CAS software, and it is unclear whether the shift can be made smoothly. Second, it is unclear whether it will be possible to use the Japanese language to access the databases. CAS says that it wants a method for using katakana in addition to the CAS software. A third problem has to do with rates. The NTS International network will charge the same rates to users anywhere in the world; the American CAS databases will be available to a user in Japan for the same rates as those paid by a user in America or Germany, and American users may object to paying higher world-wide rates because of the participation of overseas users, as if subsidizing the network for the sake of foreign users. The American databases will surely be used more frequently than the Japanese ones. It will be necessary to re-examine the system of rates. Finally, the JICST databases must be translated from Japanese into English, and this will take time and money. JICST seems to be pinning its hopes on the perfection of a machine translation system by the end of FY 1985, but its databases continue to grow by some 470,000 items a year, and it will be difficult to catch up no matter what happens about MT.

Once again, the question of translation seems to be the biggest bottle-neck, concludes the article in the Nikkei sangyō shimbun.

* * *

AGENCY OF INDUSTRIAL SCIENCE AND TECHNOLOGY WILL SUPPLY REPORTS TO NTIS

The Nikkei sangyō shimbun of September 14, 1984 reports that the Agency of Industrial Science and Technology (Koogyoo Gijutsuin) an agency belonging to the Ministry of International Trade and Industry, has decided to supply research reports originating from its research institutes and and regional departments to the National Technical Information Service (NTIS), an agency of the U.S. Department of Commerce. This is said to be the "first time" that the most up-to-date governmental reports about high technology have ever been "exported" systematically. The move is expected to deflect overseas criticism to the effect that Japan exports its industrial products but not its technical information. (I know it's confusing, but please do not mistake the Agency of Industrial Science and Technology (AIST), called Koogyoo Gijutsuin, belonging to MITI, for the Science and Technology Agency (STA), called Kagaku Gijutsuchoo, which belongs to the Prime Minister's Office. The STA is the parent organization of JICST. — Editor.)

The reports which the Agency will send to NTIS are monthly and quarterly reports issued by the Agency's 16 research institutes and laboratories (including the Electrotechnical Laboratory) and regional industrial research institutes. They include 20 different publications, and 150–160 issues of them are published annually.

The research reports will be collected first by the Japan Industrial Technology Association (JITA), an auxiliary organization of MITI, and then forwarded to the Mitsubishi Research Institute, Inc., which acts as the Japanese representative for NTIS. MRI will then send them to NTIS by airmail at least once a month.

Although almost all of the published research reports have English-language abstracts, only a few of them are in English. But this should be no problem for the NTIS, says the Nikkei sangyō shimbun because "NTIS's staff of experts who are both well-versed in science and technology and also have a good knowledge of Japanese will make English translations of the full texts." (I just can't wait. Concerning NTIS's "enthusiasm" for Japanese laboratory reports, see TJT no. 19, pp. 12–13. — Editor)

The Nikkei sangyō shimbun says that a survey by MITI in fiscal year 1983 showed that, of the 679 databases which are currently usable in Japan, 522 are supplied from overseas and only 157 are Japanese-made. Almost none of the Japanese databases are supplied overseas. It is inevitable that overseas criticism should be aimed at Japan for this totally lopsided situation. Recently, the situation is starting to change. That is, overseas information suppliers are beginning to require Japanese customers to reciprocate by supplying Japanese information in exchange. Japanese are now starting to realize that information is not something which can be acquired in any amount by just paying money for it. The current move by the Agency of Industrial Science and Technology is an attempt to reply to this criticism.

The increasing willingness in Japan to supply information overseas is based on a sober assessment of Japan's own self-interests. Hitches in information flow might well be disastrous to Japanese industry. There is a growing fear in Japan that overseas information may be cut off at some time in the future. The Japanese word for that is joodan (short for joohoo danzetsu, "information cut-off"). The president of a Japanese database called Keiei Joohoogen Sentaa says in a newspaper interview that "information cutoff is not a phantom" (joohoo danzetsu wa maboroshi de wa nai). Mr. Kumiyata Shimamoto is quoted as saying that Japanese companies utilizing the online services of important American database agencies have recently been complaining that they are not being served adequately when they try to access information in certain high-technology areas. He fears that the Americans may possibly be trying to pick on users in rival countries such as Japan and West Germany. The best way to deal with this, says Mr. Shimamoto, is to create powerful Japanese information agencies in order to correct the one-way flow of information and also to deepen human relations with the Americans. That is his strategy for surviving the age of information cut-off. (Nikkan kōgyō shimbun, September 14, 1984)

* * *

JAPANESE COMPANIES ESTABLISH RESEARCH FACILITIES IN U.S.

Japanese newspapers have recently been mentioning a growing trend for large Japanese manufacturers to establish research institutes and R&D-type factories in the U.S. Some of the companies doing so are Kyocera, Sumitomo Electric Industries, Ltd., TDK, Nakamichi and Nippondenso Co., Ltd., and still others are expected to follow suit in the future. MITI is planning to offer low-interest loans to Japanese companies wishing to establish research institutes overseas.

According to the September 1, 1984 issue of the Nihon keizai shimbun, Kyocera will establish a research center for fine ceramics and other new materials in Vancouver, Washington by the summer of 1986. The research facility will be attached to a plant manufacturing electronic parts such as ceramic IC packages and ceramic chip capacitors which will go into production in the autumn of 1985. The research center will initially employ 20 to 30 scientists and researchers, including some dispatched from Kyocera's Japanese laboratories. Additional engineers will be recruited from U.S. institutions such as the Massachusetts Institute of Technology. When fully staffed in eight years, the research center will have around 160 researchers.

Sumitomo Electric Industries, Ltd. plans to build a plant at the Research Triangle Park in Raleigh, North Carolina to produce fiber optic cables. The plant will initially have a payroll of about 200, of whom 180 will be recruited locally. In the same building, Sumitomo Electric Industries will establish a research institute devoted to optical fibers and gallium arsenide. There will be some 30 to 40 researchers at the beginning, and 90 percent of them will be recruited from American universities and companies. The company expects to hire graduates of North Carolina State University, Duke University and specialists now working at the Research Triangle Park. (Nikkei sangyō shimbun, Sept. 7, 1984)

Another Japanese company is planning to establish a research center at the Research Triangle Park. Kobe Steel, Ltd., will open an office there in November and begin activities with a staff of five, but it plans to increase the number of staff members and to develop the office into an independent research institution in the near future. The center will be for research, investigation, and information-collection in fields of high technology in general. It will be operated jointly by Kobe Steel and a company called Midorex (?) in Charlotte, North Carolina which it purchased last year. (Nihon keizai shimbun, Sept. 27, 1984)

TDK established in March of this year a small-scale trial plant for electrical parts for communication equipment in the U.S. Currently it has only six researchers, five Japanese and one American, but by March 1985 20 American researchers will be recruited, and the number of Japanese researchers from TDK will be increased to 20, making up a total of about 40 researchers.

Nippondenso Co., Ltd., a manufacturer of automobile parts, is also planning to establish a "technical center" in the U.S. for research in electronics and new materials. It has decided to locate its research center at Southfield, Michigan. Construction of a large, 4540 m² building will start sometime this year, and one-third of the floorspace will be occupied by the Technical Center, which will begin activities in spring of 1986. The rest of the building will be occupied by Nippondenso Sales, Inc. Initially, the staff of the Technical Center will number 30 persons, many of them Americans. The company has also decided to build a largescale factory at Battle Creek, Michigan. It will also go into operation in spring of 1986. (Nikkan kōgyō shimbun, Sept. 5, 1984)

Nakamichi, a manufacturer of audio equipment, has also established a research facility in California. It will be staffed by American specialists. The company's American subsidiary, Nakamichi U.S.A. Corporation, has been engaged in sales of the company's cassette decks and other audio equipment. The headquarters recently moved from Santa Monica to Torrance, California (address: 19701 S. Vermont, Torrance, CA 90502). The company has recently begun to expand into the non-audio field, and the new research facility will engage in research and development aimed at expanding the company's range of products in fields such as computer peripherals, optical equipment, etc. The new research facility will occupy a floorspace of about 1,200 m². Six American researchers have already been hired and are at work; still more will be hired from time to time. It is believed that the research will be aimed at developing software for use with optical disk devices, magnetic heads, and other products. (Nikkei sangyō shimbun, September 7, 1984)

Other Japanese companies such as Honda Motor Co., Ltd. and Toshiba are also said to be thinking of building research facilities in the U.S.

The article in the September 1, 1984 issue of the Nihon keizai shimbun says that this trend reflects the desire of Japanese companies to expand research to establish their own independent technologies which will help them overcome the increasing tendency for the U.S. to mount "technological blockades" against Japan (see no. 14, p. 15–16). This tendency towards technological protectionism is making it more and more difficult for Japanese companies to import new technologies from the U.S. and Europe. As a result, Hitachi, Ltd. and Matsushita Electric Industrial Co., have both decided to establish new research institutes in Japan next year in an effort to develop their own independent technologies. However, the speed of technological development will be slowed down if Japanese companies rely exclusively on the limited personnel and information resources available to them in Japan. To overcome this problem, many Japanese companies have established joint ventures together with American high-tech companies and venture businesses. The establishment of new research and development facilities is to be seen in the context of their overall technology strategy, and building Japanese "brain bases" (zunoo kichi) in the U.S. is the latest move in this field.

The Ministry of International Trade and Industry has recently adopted a policy of offering lowinterest financing through the Export-Import Bank of Japan to Japanese companies wishing to establish overseas research facilities. MITI made a survey of about 130 large Japanese businesses as of June, 1984. It found that six of these companies already had research facilities in America or Europe and that 12 companies were currently planning to establish them. (Nikkei sangyō shimbun, Sept. 4, 1984)

* * *

MATSUSHITA ESTABLISHES FOUNDATION TO AID EDUCATION IN U.S.; LANGUAGE EDUCATION MAY BE FUNDED

According to articles in the wall street Journal of September 18 and the Nikkei sangy6 shimbun of September 19, Matsushita Electric Corporation of America (1 Panasonic Way, Secaucus, New Jersey), the American corporation of Matsushita Electric Industrial Co.. has established a $10 million foundation to support educational programs in the U.S. This is to be the largest such foundation established by a Japanese company in the U.S. The Matsushita Foundation was establish to commemorate the 25th anniversary of the establishment of Matsushita Electric Corporation of America, which was set up in 1959.

Robert S. Ingersoll, a former U.S. ambassador to Japan, was appointed the foundation's chairman. Ingersoll also heads the Japan Society of New York and is a regent of the University of Chicago. He said that funding, which will begin in spring of 1985, would likely support curriculum and teaching in secondary schools and universities, leadership training and international education and language study.

"Americans are notoriously weak in foreign-language skills," Mr. Ingersoll was quoted as saying. "I would expect that experimentation with total language immersion and other intensive language techniques might be supported."

* * *

JAPANESE TRANSLATION AGENCIES PREPARE FOR AUTOMATION; SENSE OF CRISIS ABOUT EXPECTED SHAKE-OUT LEADS TO ATTEMPTS AT DIVERSIFICATION

An article by reporter Maruyama in the Nikkei sangyō shimbun (Sept. 5, 1984) reports that Japanese technical translation agencies are implementing strategies aimed at survival when the "machine translation" era arrives. Even without the advent of computerized translation, competition is already excessive in the Japanese technical translation industry, and if machine translation systems really do attain widespread acceptance the blow to the industry will be far-reaching. There is a high sense of crisis (kikikan mo takai) among the various translation agencies, reports the article. Translation agencies are therefore embarking on a number of strategies. Some are making efforts aimed at developing sophisticated translating know-how which cannot be duplicated by machines. Others are seeking out overseas technical information which they translate and publish themselves to avoid total dependence on translation jobs. Still others are diversifying into fields such as "visual translation" using video tape recorders and slides.

For example, Maruyama mentions Tectran, a Small translation company based in Nara. It has only 10 registered translators in its files. This autumn it plans to translate into Japanese and publish in Japan quality-control manuals for medical equipment used in the U.S. After this, it is studying the possibility of publishing overseas a newsletter in English containing information about current developments in the Japanese medical-equipment industry and papers delivered at meetings of Japanese learned societies.

Mr. Makino, president of Tectran, predicts that only one-third to one-fifth of the companies currently doing technical translation will be able to survive the full-scale dissemination of machine translation. Deterioration of quality is inevitable, he says, if one small translation agency having only about 10 translators attempts to translate documents in all technical fields ranging from medical systems to computers and measuring instruments.

Tectran is planning to publish quality-control manuals and newsletters because of its "two-birds-with-one-stone" strategy. That is, it hopes to enhance its translating ability by concentrating solely on one single area: that of medical equipment. At the same time, it wants to establish a business department which will be able to carry on stably even if the company cannot rely on continuing translation jobs from manufacturers.

But an agency desiring to specialize in one technical field needs to do its homework in order to master the required technical knowledge in the given field. That is why Mr. Makino, who had foreseen at an early stage that machine translation systems would be introduced, decided four years ago to begin a Japanese-language edition of Biomedical Business International (BBI), a newsletter published by an American market-research agency for American manufacturers of medical equipment. For the past four years, his company has been working to build up its translating know-how in the field of medical technology.

A quaint and hopefully antiquated Japanese term for translating from Japanese into a foreign language, implying that translation involves merely transposing the way the letters are arranged on the page, shifting them from a vertical direction up and down the page to a horizontal direction going across the page.

A different approach is followed by Tom, a Tokyo company headed by Mihoko Katsuta. Katsuta's company has departed significantly from the old tradition of taking Japanese technical documents and simply converting the "vertical to horizontal" (tate kara yoko ni naosu)* Instead, it dispatches technical translators to client companies, where they participate actively in the process of creating descriptions and specifications for new products from the planning stage. (A similar practice, called euphemistically FM (for "facility management") is widely adopted by Japanese software companies. See the article "Software Crisis in Japan (Part 1)" elsewhere in this issue.)

Even the most competent specialist in translation will be unable to turn out an easily understandable, reader-friendly translation if the original Japanese text is difficult to understand. Even worse results would be obtained if such a difficult text were merely fed into a computer for machine translation. To ensure accurate copy, Tom makes a practice of dispatching translators with specialized technical knowledge to the client companies for periods of at least three months, and sometimes as long as one year. The aim is to begin at the very beginning and to create first an original which will be written specifically for the purpose of translation into the foreign language(s), and then to create explanation booklets or manuals for the new products which will have a quality entirely different from something which could be turned out by machine translation.

At the same time, Tom is working with visuals. That is, it is preparing foreign-language slides and video tapes giving accurate technical explanations which would be difficult to communicate by means of written documents alone. The aim is to give the client a comprehensive range of technical translation and communicating services making full use of visual technologies, not merely to supply technical explanations by means of printed words on a page.

While both Tectran and Tom aim at establishing techniques which will differentiate them from machine translation, other translation agencies are looking for ways of coexisting with machine translation. For example, Intergroup of Osaka, which recently added a course for training technical translators to its school for training interpreters and translators, is planning to add a training course for operators of translating machines. Mr. Maeda, who is in charge of the training courses, says that "knowledge of specialized technology will still be needed in the world of inputting and editing even after machine translation has been introduced." The school's course for technical translators includes training in nuclear power and computer technologies.

The struggle for survival by Japanese translation agencies after the awaited introduction of machine translation systems is expected to intensify even further as time goes on, concludes Maruyaina in his article.

I couldn't help noticing that the translation agencies mentioned in the article seem to accept it as a foregone conclusion that the "age of machine translation" is about to arrive. No trace of skepticism is alluded to. Are there no maverick translation agencies in-Japan which adopt the contrary view that machine translation is not going to have an appreciable impact on J-E technical translation? And what about those agencies which adopt a more moderate view of things and prefer to explore the possibilities of machine-assisted human translation (MABT), rather than human-assisted machine translations (BAMT), possibly along the lines suggested by Preston Maxwell of Tokyo in our no. 18 (p. 13)?

* * *

LIBRARY OF CONGRESS IMPLEMENTS RLG SYSTEM FOR CATALOGING

An article in the September 3, 1984 issue of Computerworld says that the U.S. Library of Congress is about to implement fully the Chinese-Japanese-Korean (CJK) system developed by the Research Library Group, Inc. (RLG) of Stanford, Calif. (See TJT no. 16, pp. 32–34, for a description of this system.)

LC will move this fall from the test phase to full-scale implementation of the system. By the end of 1984, all East Asian books received there will be cataloged in a national library database using 24 CJK terminals, according to Dr. Harriet Zais-Gabbert, coordinator of LC's CJK project. The system is operating at LC and 18 other North American libraries and has already cataloged more than 20,000 volumes.

The CJK system uses a terminal made by the Transtech International Corp. of Natick, Mass., that has a total of 179 keys on its keyboard — three times as many as the standard keyboard. The operator enters individual components of kanji, and then the computer combines the components to form the actual character. "We break a character into word roots," said Peter Kang, general manager of Transtech. "Some characters require only one keystroke, while others combine two, three or more word roots to compose a character."

133 of the 179 keys on the keyboard are character keys, each of which contains two to five components of Chinese, Japanese and Korean scripts as well as Roman letters for Western languages. There are four function keys, one for each language, by which the operator can switch from one language to another.

The terminal comes with a dot matrix printer and a cluster controller (a DEC LSI-11/23 microprocessor) which can support four terminals and communicate with RLG's Research Libraries Information Network, a computerized bibliographic utility at Stanford University.

Zais-Gabbert said that some libraries may be slow to make full use of the system's vernacular search capability. "Oriental libraries are so traditional; I think it'll take a generation for them to get used to it," she said.

Thanks to Anthony Meadow for sending me the article from Computerworld,

* * *

NEC WILL MARKET I-TRON TYPE OPERATING SYSTEM

According to an article which appeared in the Nihon keizai shimbun on September 8, Nippon Electric Co., Ltd. announced on Sept. 7 that it will soon market a computer operating system which is "essentially the same" as I-TRON, which was recently developed by a group at Tokyo University headed by Ken Sakainura (see TJT, no. 18, p. 8–9). The OS will be used to control the V20 and V30 microprocessors which NEC has designed independently.

I-TRON is designed specifically to meet the requirements of the Japanese market, such as processing of kanji. NEC's move signifies that it is beginning to place major emphasis on the sales of Japanese-manufactured basic software in an effort to attain independence from American companies such as Intel, Motorola, and Microsoft, which currently control the microprocessors and operating systems of computers.

* * *

SAKAMURA DESCRIBES TRON PROJECT

Ken Sakamura of Tokyo University has written an interesting article in the September 1984 issue of Bit (pp. 4–11) explaining his personal viewpoint and describing the TRON project which he heads. (See our article in TJT no. 18, p. 8–9 and the article above.)

The first part of the article, giving Sakamura's personal viewpoint, is by far the more interesting part. He begins his article by discussing why it is that so few original microprocessors, operating systems, and other basic software have been designed by Japanese. Japan has relied on imports in these areas thus far.

First, the Japanese market is too small. The American market is ten times larger and much easier to enter than the Japanese market.

Second, Japan is less inclined to spend money on research and development. This conservative tendency is bound to change in the future but still applies today. R&D on computer architecture is very expensive, and the results look quite puny when compared with the vast expenses. From now on, Japan will have to pay large sums for what it has not developed itself, as we can see from the IBM incident and the question of copyrights for microprocessors.

Third, the Japanese are too insular (shimaguni konjoo). They easily become jealous and disparage those of their countrymen who try to do something novel. They do not feel sincerely happy about the success of other Japanese. On the contrary, the Americans tend to reward success, respect individual differences and like variety, perhaps even to an excessive degree.

Fourth, the Japanese lack seriousness about developing unique technologies. Too many Japanese think vaguely that a new technology is something which just springs up somewhere else, not something which they must create by themselves. Sakamura says that Japan ought to develop self-sufficiency in both computer hardware and software. Hardware and software should be considered as an inseparable whole and should not be compartmentalized.

Next, Sakamura says that in essence a computer is something invisible. Software is already invisible, but even the hardware itself will become invisible in time. Japan must learn how to evaluate properly such intangibles and to spend money on them. The Americans seem to be more willing to fund basic research all the way up from the conceptual stage, while the Japanese demand to be shown immediate, visible results from their research investments.

Another problem about Japan, he says, is that collaboration between business and universities is very difficult to achieve in Japan. Sakamura mentions the cases of close cooperation between American university professors and private industry. American academics are allowed far more freedom, he says, in deciding how to spend their research funds than is the case in Japan. This tends to impoverish innovative thinking in Japan (hassoo ga nan to naku bimboo-jimita mono to naru)

Now Sakamura begins to talk about personal computers. The 16-bit microprocessors used in personal computers are American made (Intel or Motorola), and so are their operating systems (MS-DOS, CP/M-86 or UNIX). They are all based on English and do not have sufficient power to handle Japanese. English has 26 letters in its alphabet; if we add together the upper and lower case letters, plus the numbers and punctuation marks, we obtain a total less than 100 characters. Japanese has more than 10,000 characters, he says. In the font memory, a display of 5x7 dots, or at most 10x16 dots is quite sufficient for English, but Japanese characters require a minimum of 16x16 dots, or preferably 24x24 or even 32x32 dots. Several kilobytes are sufficient for English, but Japanese requires nearly 1 megabytes of font memory. A CRT display for English will be adequate if it has a resolution of about 480x200 pixels for displaying 20 lines of 80 characters each, but Japanese will require 640x400 pixels for displaying 20 lines of 40 characters. An adequate frame memory will be needed if the bit-mapping system is adopted. Sakamura thinks the Macintosh is very well designed for Americans at the current level of technology, but its display has a resolution of only 512x342 pixels. This would be sufficient for English but would not give an adequate Japanese display. To use the Macintosh for Japanese, he says, you would need twice its resolution and several times both its computing power and memory capacity.

This is the reason why Sakamura thinks the Japanese have never created a personal computer entirely adequate for the Japanese language: because it was simply too difficult. However, the overall technical level has risen, and the time is now ripe to develop a computer able to process the Japanese language on a computer.

Now Sakamura begins to describe the TRON Project. The aim of TRON (The Realtime Operating system Nucleus) is to create in Japan an entirely unique total architecture. It will take shape as a 32-bit VLSI microprocessor. The instruction sets in use today, he says, are too "dirty." They were aimed at compatibility with existing hardware, and that hardware is now outdated. He discusses the VAX, the 68000 series, and the NS32000 series and says that performance-wise they are not adequate. His own goal is to develop a processor with a performance 100 times better than the 68000 (8 MHz) in real-time applications.

The TRON project does aim at a von Neumann type architecture, but it will be the ultimate, the decisive variety of von Neumann system. Sakamura wants to incorporate the achievements of adaptive type architecture and even artificial intelligence technologies in the TRON project. Current plans are to produce a TRON chip (a 32-bit VLSI microprocessor) about three years from now. The operating system will center in the TRON chip kernel, which will be built into ROM together with I-TRON (Industrial TRON), a real-time multitask OS. B-TRON (Business TRON) will be an OS suitable for business applications and for controlling large-size machinery. H-TRON (Home TRON) will be a user-transparent OS based on B-TRON but designed so as to be usable by complete novices (on a home computer).

Sakamura says that I-TRON, the real-time multi-task operating system for industrial use, is now near completion. Until the TRON chip is perfected, it will be supplied in two forms: I-TRON/86 (for the 8086 microprocessor) and I-TRON/68K (for the 68000).

* * *

SOFTWARE CRISIS IN JAPAN (PART 1)

Donald L. Philippi

The situation in the past used to be that computer manufacturers would spend 80% of their development costs on the hardware, but now the situation apparently has reversed itself; and the software has assumed a far greater importance than the hardware. Cries of alarm in Japan are being raised about the "software crisis." Another closely related problem in Japan is the question of legal protection. Is software covered by the copyright laws, or is some other system of legal protection required?

Statistics for the software industry in Japan are included under the "information services industries." The latter industries taken as a whole, according to a MITI survey, had sales of some ¥1,095 billion (roughly $4.5 billion) in 1983. The American industry is predicted to be a $25 billion industry in 1985. Therefore, the Japanese industry is regarded as being roughly one-sixth the size of the American industry. In spite of this gap, or rather because of it, the Japanese industry is expected to grow vigorously in the future.

The Japanese market for software makes up about one-third of the entire market of the "information services industries," amounting to some

364.4 billion (roughly $1.5 billion). In the past eight years, the market of the Japanese software industry has increased by slightly less than 9 times.

In recent years, the emphasis in the Japanese software industry has been shifting away from development towards maintenance. Much upgrading and debugging has to be done manually. Today, with the current hardware systems coming into their mature period, there is less emphasis on development of new software, and the maintenance services provided by software houses are of the greatest importance, In 1970 the cost ratio between development and maintenance was 4:6, but in 1985 it is estimated that it will be 3:7. In the future, when and if the Fifth Generation computer is introduced, it will be necessary for the Japanese software industry to develop generations of completely new software programs.

The Japanese market for personal computer software is still relatively small, although it is expected to grow rapidly. A survey of Japanese and American personel-computer software houses was made recently by Nikkei pasokon magazine. The results showed that the market for personal computer software in Japan was worth about ¥30 billion (roughly $122.5 million) in 1983. The U.S. market for it during the same period was estimated at some $960 million. Since a survey for 1982 had shown the Japanese market to be about ¥20 billion ($81.6 million), there has thus been a growth of about 1.5 times in one year. By 1988, five years from now, the Japanese market for PC software is expected to grow to ¥230 billion (roughly $939 million), and the American market will be $7.5 billion. When 32-bit personal computers make their appearance in Japan, personal computer software may become the mainstream of the Japanese software market. (Nihon keizai shimbun, September 21, 1984)

Scope of the Japanese Software Industry

According to a MITI survey, the number of companies in the information service industry totaled some 2,148 in 1983. This amounts to an increase of 184 in one year, since the number for 1982 was 1,864. Many or most of these companies are software companies. If we include everything down to the so-called "mansion soft houses" (software houses occupying one room in an apartment building), the total number of software houses all over Japan is estimated to be some 7,000 or 8,000. They are in a state of constant flux and cannot even be estimated accurately.

Some 70% of Japan's software houses were established in the decade from 1965 to 1974. Most of them were "spin-offs," i.e., start-ups founded by former employees of large companies who resigned to form their own venture businesses. A famous example of such a venture company is Cosmo 80 Co., established in 1979 by President Yutaka Usui and some 80 colleagues who were working at Ishikawajima-Harima Heavy Industries Co. It has since teamed up with Mitsubishi Corp and IBM Japan to develop software for the INS digital network planned by NTT, and a 20% interest in the company was acquired in July 1984 by Schlumberger Ltd. Another example is Argo 21, formed by about 40 former employees of Nippon Univac.

Personnel Is Biggest Problem

The biggest problem for Japanese software houses is the chronically insufficient number of system engineers (SE) and programmers. The Nikkei pasokon magazine made a survey of American and Japanese software houses and asked them what their biggest problem was in developing and marketing software for personal computers. 51.7% of the American software houses answered that it was "financing difficulties," reflecting the drying-up of venture capital. Only 19% of the Japanese software houses gave this answer. For them the biggest problem was the "shortage of manpower" (40.8%). This was the lowest ranking problem in the American respondents (6.9%). (Nihon keizai shimbun, September 21, 1984)

The term "software crisis" refers mainly to this chronic personnel crisis, and also to the delays in mechanizing or computerizing software development. Although the "crisis" has come to be spoken of in Japan only recently, because the Japanese software industry has been such a late starter, the "crisis" is a longstanding one in the countries which are more advanced in software. (The term "software crisis" seems to have originated at a NATO-sponsored meeting of software specialists in October, 1968.) Even though some progress is being made in mechanization of the industry, it is still largely manpower-oriented rather than technology-oriented, and the biggest task for software houses is to train personnel.

How many software specialists does Japan have? The 2,148 companies in the information industry surveyed by MITI in 1983 had 127,978 employees. The same survey for 1982 showed 1,864 companies with 113,414 employees. Of the 1982 total, 3,026 were researchers, 19,661 system engineers, 27,371 programmers, 13,433 operators, and 23,252 keypunchers. The total number of software specialists (including the categories from "researchers" down through "programmers") in Japanese software houses during 1984 would probably be a little less than 90,000, according to a rough estimate made by the Nikkan kōgyō shimbun (July 13, 1984).

On the other hand, another survey of some 5,700 computer-using companies revealed that the number of their software specialists during 1981 totaled a little less than 105,000. These surveys do not, of course, include the software specialists jn tiny, one-room operations, such as those who are developing game software in one room of a "mansion" (an apartment building). The same article in the Nikkan kōgyō shimbun estimates, on the basis of these two surveys, that the total number of software specialists everywhere in Japan at the present time is somewhere around 120,000 or 130,000. However, this is no nearly enough to meet the need for such specialists. The number of software specialists which will be required by the software industry in 1985 is estimated to be 500,000. The personnel gap is truly desperate.

What is the result of this extreme imbalance between the requirements and the actual supply of personnel? For one thing, highly qualified system engineers and programmers are overworked and required to stay late at night to do overtime work. Some 50% of all programmers surveyed recently complained of being "too busy." I suspect that plain old "overwork" may account for some of the defects in Japanese-made software.

What are the companies doing to remedy the gap? For one thing, some of them are establishing training schools. In April, 1978, Computer Services Corporation (CSK), a publicly traded company, established Computer Business Academy to train its universitylevel students. Naturally, CSK tries to hire the graduates itself. Nihon Denshi Kaihatsu and NJK, two Tokyo software houses, also operate their own specialized computer schools. Nihon Denshi Kaihatsu operates six schools in various parts of the country which have already graduated more than 5,000 students, and some of them have gone to work for the company. NJK operates five "information sciences" schools and tries to absorb the graduates. Many other companies are expending large sums on training their employees, viewing these expenses as a long-term investment.

However, the need for trained personnel is still acute, and a variety of methods are being tried to cope with the problem. One method is to use the talents of women. Some software houses staffed entirely by women, and women who quit their companies after marriage are recalled to work at home or in the provincial software affiliates of the large computer manufacturers. (See the article entitled "Is the Foundation Shifting?" by Fred Schodt elsewhere in this issue.) Another disadvantaged group, prison inmates, are even being given a role to play. (See "Software in Prisons" in TJT, no. 19, p. 32)

Still, there is much mobility among the personnel of Japanese software houses, who often complain of lack of satisfaction in their work. Software houses are notorious for trying to recruit each other's employees, and software writers are more jealous of their independence and tend td float from job to job more easily than most Japanese.

Expansion into the Provinces

The large computer manufacturers like Fujitsu, NEC and Hitachi are actively expanding their software departments by establishing bases in the smaller provincial cities. The larger software houses are also starting to expand into the provincial cities, and there are also many independent software houses established in the provincial cities. This move reflects a notable demographic trend in Japan: what the Japanese like to call the "U-turn phenomenon," meaning the flow of population, especially youthful, well-educated Japanese, from the congested metropolises back towards the medium-sized provincial cities with populations of 100,000 to 1,000,000.

Fujitsu has a total of 31 affiliate software companies located in the provinces. It aims at establishing ten of them every year and hopes eventually to have 100 — more than one in every prefecture. In March, 1984 it established one in Nagoya and another in Matsue. It also has a network of more than 40 software subcontractors organized in the FACOM Software Association (FSA). In locating its provincial affiliates, Fujitsu aims at hiring local personnel and tries to find areas where it might attract graduates of local technical universities and high schools. Companies in areas which are technically less advanced are assigned less demanding tasks, such as developing package software, while those in more advanced areas develop all-purpose software for mainframes.

NEC established its Hokkaido NEC Software in May, 1984, thus completing its plan to divide Japan into nine blocks and to establish a software subsidiary in each. The blocks are then to be subdivided into networks of subsidiary software houses. Thus, Kyushu NEC Software at Fukuoka will set up its own subsidiary software houses in Kumamoto and Kagoshima. NEC will come to be, not only the parent company, but the grandfather company of a whole network of sub-subsidiaries.

Hitachi, Ltd. also is adopting the same approach. It has a network of directly owned software houses in Kansai, Chubu, Tohoku. Hitachi Software Engineering at Yokohama, which was established in 1969, now has set up its own subsidiary plant in Kanazawa and is planning others in Kurume and Chitose.

The independent provincial software houses feel threatened by the incursions of these growing networks of the major companies. They will bring intensified competition for the local personnel, and some independent businesses will surely be absorbed into the larger company networks.

What is the "Average" Software House?

The Information-Technology Promotion Agency (IPA) made a survey of 1,290 companies as of July, 1983 and obtained replies from 413, including 233 large and small software companies. It found that software companies were overwhelmingly in the small business category. 41% of the software houses surveyed had 50 or fewer employees, 21% had 51100 employees, 18.5% had 101–200, and only slightly less than 10% had 300 or more. About half of the companies (117) had annual sales of 100 million (roughly $408,000) to 500 million yen (roughly $2 million). 40% of them (91 companies) had capital of less than 10 million yen (roughly $40,816). Interestingly, there would seem to be parallels between the software industry and the translation industry, at least in the small size of the businesses.

The average picture which emerges from this survey shows a software house with 159 employees, sales of ¥1.4 billion (roughly $5,714,000), and a capital of ¥43 million (roughly $175,510). Thus, one concludes that most Japanese software houses have 50–100 employees, annual sales of around ¥500 million (roughly $2 million), and capital ranging from 10 to 20 million.

Since these companies are labor-intensive, personnel expenses occupy a large ratio of sales (the average is 38.5%, but 45.7% among the independents and 30.3% for the affiliates of large manufacturers). The sales per employee are lower than in ordinary manufacturing industries, and the ratio of personnel costs to total sales is extremely high. The Japanese software houses appear to be sitting on a time bomb since, like other industries, they will have to face the problem of ageing of their work force. The average age of their employees today is under thirty, and there is even some talk about "retirement at 35." On the surface this seems to mean that emphasis in software is on flexible-minded youthful people, but the hidden meaning is that a company with large numbers of employees over 35 years of age will have personnel expenses which will place a burden on its balance sheet. Either there must be massive firings, or software development will have to be computerized or rationalized in some other way.

I wonder why Japanese software houses do not approach the problem in the same way that translation companies do: by hiring freelancers working at their own homes.

Labor-Saving Measures

With the desperate shortage of personnel, an important problem is to find ways of developing software more efficiently. The increasing demand for software cannot possibly be satisfied by simply continuing to pour in more and more expensive manpower, as is being done today. Some measure of mechanization of software development would appear to be the most logical solution. Attention has lately been focused on CASE (computer-aided software engineering) in which "reusable parts" are used. General-purpose software which has already been developed and which might possibly be used in other applications is stored in module or skeleton form in a computer database and tapped when needed in developing other programs. The large companies such as IBM, Hitachi, Fujitsu, and NEC have each elaborated their own methodologies and are constructing automatic production systems, and this approach has penetrated down to the larger software houses as well. The future of the software industry will be governed to a large degree by the degree of success achieved in mechanization and computerization of software creation.

The FM (Facility Management) Problem

A peculiar problem of the Japanese software industry is something called "facility management" (FM) (in Japanese called yooin haken) FM is a system whereby software houses send out key punchers, operators, programmers or system engineers to work at computer manufacturers or companies using computers for a fee paid to the software houses. Software companies would prefer to give more attention to software development on higher levels and want to get out of FM, but the demand for it remains strong and it is profitable. This type of service amounts to a market of roughly 100 billion, or some 9.4% of the entire information service industry. They must do it because they develop the software and then feel that they are obligated to provide personnel to service it. Actually, once suspects that the system verges on the abusive. It provides services resembling those provided by a temporary employment agency or a building maintenance firm. The practice is, strictly speaking, illegal under Article 44 of the Employment Security Act, but the letter of the law apparently is not being enforced, and if it were, we are told, there would be an immediate breakdown of the entire Japanese software industry. The practice causes innumerable problems for the software houses because of the differences in working conditions between the home company and the client company, and it makes it difficult for software firms to attract young graduates to their employ. The word yooin haken is a taboo for recruiters in most software houses, who when answering questions prefer to use euphemisms such as gyoomu ukeoi or FM. Almost all prospective graduates who visit recruiters ask them pointedly whether the software company engages in the practice of yooin haken. They are afraid that, after they have been hired by the software house, they will be sent out on loan to the plant of a computer maker or to another company's office, where they may be looked down upon and given discriminatory treatment as compared with the host company's full-time employees. Some translation companies also follow a similar practice, although I have seen no mention of complaints against it by the translators (see article elsewhere in this issue).

[To be continued]

Additional Reading

Bro Uttal, "Japan's persistent software gap," Fortune, October 15, 1984, pp. 150–160.

Konpyuutaa sofuto kigyoo meikan [Directory of computer software companies], compiled by Yano Research Institute, July 1984. ¥85,500.

Joohoo shori, sofutouea kaisharoku '85 [Directory of information-processing and software companies '85], 13th rev. ed., compiled by Shii Sangyoo Kenkyusho, August 1984. ¥22,000.

"Women are rising force in program development jobs," Japan Economic Journal, August 14, 1984.

* * *

IS THE FOUNDATION SHIFTING?

Frederik L. Schodt

With a low birth rate and virtually no immigration, Japan has few options when an expanded workforce is needed; longer hours, deferred retirements, and more robots may not be enough. There is, however, a long ignored, highly educated and motivated section of the population that has thus far been largely underemployed — women. Recent articles in Japanese newspapers during the last year suggest that changes in the job market may be in the offing.

Women have been working hard — perhaps harder than men — since time immemorial, and in Japan today most hold jobs outside the home at some point in their lives. But all too often they are excluded from positions of responsibility in business and government. Many college graduates can look forward to secretarial, decorative jobs (as tea-serving "O.L.", or office ladies) until around 24, then a forced retreat into marriage and child-rearing, followed by small, part-time jobs when the children have grown. At present 0.4 percent of management positions in government are held by women; in the corporate world only 0.1 percent are women [1]. Women interested in careers in government at least have the advantage of an examination system that allows them a foot in the employment door; over 79 percent of the top Japanese corporations interviewed in a 1983 poll reported that they were not even hiring any female university graduates [2].

Women Company Presidents Outnumber Women Managers

Ironically, there may be more Japanese women company presidents than there are managers. On September 14, 1983, an article in the Nikkei sangyō shimbun on women in business noted that around 3.2 percent of companies capitalized with over 1 million have women presidents today. Many of these women clearly inherited their companies from husbands or parents, but some have gone on to become quite successful. The Nikkei article, with a rather conservative orientation, showcased five women, most in their seventies, who currently head companies ranging from a construction firm to a sembei manufacturer. All are widows. One of the women, Kunieda Ushijima (73), currently heads Saga prefecture's biggest supermarket chain, the Saga Shufu no Mise, with thirty stores and sales of over 13 billion yen. Although she took over the business from her husband, she runs it today with a woman's orientation. Of 15 executives in the company, 12 are women, as are ninety- five out of one hundred stockholders.

If you can't join a company or inherit one, you can always form your own, and a new generation of women is doing precisely that. On November 10, 1983, the same newspaper ran an article titled "Softnomics — The Emergence of Women in the Industrial World." It noted that women comprise approximately fifty percent of the service industry today, an increase of over 22 percent since 1975, and that current "softnomics" (a buzzword for the information and service industries) could not exist without them. Like the September article, it featured several examples of successful women presidents. As befits the new age, they had identified unique needs in the society, and moved to satisfy them. Among their companies are a "wake-up call" service for single young salarymen, a service that rents temporary office space for meetings, etc., a tour escort service, a software company, and a firm that designs and produces technical documents for overseas (Tom). Most of the women founders featured were in their thirties; the youngest was twenty-three.

Women Managers On the Increase

New opportunities for women are also appearing in management, in traditionally male bastions. This spring the Nikkan kōgyō shimbun (7/26/84) reported that Kitazawa Valve had shocked the normally staid valve industry in Japan by hiring two women college graduates as sales managers. The two were selected from over 700 applicants, and will primarily be in charge of telephone marketing, yet will also make the rounds of wholesalers and make presentations at board meetings. Their appointment is mainly a strategic move, to help smooth over the company's planned price increase with a softer, feminine touch. But as the firm's director states, "We have a lot more use for competent women than incompetent men."

Women managers are also increasing in the distribution industry, particularly in the food business. On September 7, 1984, the Nikkan kōgyō shimbun ran an article titled "The Quietly, Slowly Expanding World of Women Managers." It noted that women are particularly suited for much retail work because they are "close to the consumer," who in the food stores more often than not is female. With this in mind, the Nichii supermarket chain has adopted a policy of promoting women managers in retail. The giant Daiei will not only make them managers, but make them eligible for transfers throughout Japan (ineligibility would kill a woman's chances for advancement). At Marui, on the other hand, 51 percent of the 8,700 employees are already women; the company has elected to entrust management of its 300 in-store women's "shops" nationwide to women.

In the same article, the Nikkan kōgyō decided to enlighten its readers — most of whom are males — about the merits and demerits of women managers. Their strengths were deemed to be consideration and attention to detail, womanly observations and "inspirations," emotional sensitivity, communication skills and thoroughness in routine work. Negatives listed were weakness in making independent decisions, lack of physical strength, obligations to home and family, and a tendency toward self-centeredness and over-dependence.

For aspiring women managers who cannot make it into Japanese firms, there are always foreign companies. Whereas working for foreign firms tends to carry a subtle social stigma for men, for women it provides new opportunities. Accordingly, data from the Nippon Recruit Center shows that IBM Japan ranks among the top five most attractive corporations for female university seniors [3].

Opportunities in Computer-land

Along with IBM Japan, the computer industry may actually prove to be the biggest equal opportunity employer in Japan. At NEC, for example, there are now over 600 women working in software development; this spring alone the company hired 150 women college graduates. One of the NEC employees, Mitsuyo Yoshino, has become something of a media phenomenon. She is head of one of NEC's software development departments, the first female department head in the history of the firm, if not the entire industry. At Toshiba, moreover, the Computers and Office Automation Systems Division now has a woman team leader [4].

Why the increased opportunity in computer-land? As the Nikkan kōgyō shimbun noted in a July 13 article on the industry subtitled "Software Crisis," there are currently around 120,000–130,000 software specialists at work in Japan, but by 1985 it is estimated that half-a-million will be needed. Current employees are overworked, and some firms have become so desperate that they have started their own schools. Thirty-four percent of those working in the field are women.

The current crunch in supply and demand is a golden opportunity for women. An increasing percentage of Japanese women are college graduates, effectively shut out of meaningful employment in traditional industry because of late-night labor laws supposedly designed to "protect" women, bias against the "weaker sex" when physical work is required, and the sheer weight of tradition. The software industry may give those who want to work a chance to show their true mettle. Because of the chronic shortage of programmers and encoders, qualified women are actively being solicited by firms, at least as a stop-gap measure until enough men can be trained. And systems engineering and programming, in the final analysis, is gender blind; it makes no difference who creates a good program, as long as it works, and no physical strength is required. New technologies, furthermore, now make it possible for more and more people to work out of the home, or at least on hours of their own choosing. . This helps put women with families on a more equal footing with men in the job market. Finally, as NEC's Yoshino and others have suggested, women may actually be better suited than men for software creation, by temperament.

"Soft-Houses" with a Dark Side?

On August 3 of this year, the Nikkan kōgyō shimbun featured an article titled "A Hard Battle in the Soft(ware) Industry" with the subheading "Woman Power." It noted that several "softhoues" or software groups composed solely of women have recently appeared. Core Group (phonetic spelling), for example, formed Ladies Core in 1982 to handle software development (and also word processing assignment). Pasol (a 100% subsidiary of Fuji Kosan) has formed Pasol Ladies, a group of two hundred women, developing educational software. Other groups of mostly women software specialists include the Matsuyama Seisakujo in Matsuyama, Aichi prefecture; O.B.C. in Nishinomiya, Hyogo prefecture; and the Gendai Joohoo Kenkyuujo, an all-woman software firm owned by a women in Tokyo. Many of the women in these groups work out of their homes and send their work to the main offices over modems. Nippon Denki Software has even arranged for its regional subsidiaries to contract with former women employees who quit upon marriage and moved to other areas.

Naturally, there is a darker side to this new trend. Despite the hoopla in the media about a new "woman power," many of the women seem to be employed on the low end of the so-called "software industry," as simple encoders of data. Some of the "software" firms also handle word processing. Pasol Ladies of the above article, for example, contracts with a well-known fashion model agency and hires ten models as "waapuro redii" (word processor ladies). For the women, who are then sent out to corporations than need their services, the work is an opportunity to supplement their incomes in slack times. According to the Nikkan kōgyō, the women are "'relatively beautiful,' and at the corporations they are sent to, increasingly popular because of their 'added value.'"

From the standpoint of the firms, hiring women also can represent considerable cost-savings. On September 17, 1983, the Nihon keizai shimbun ran an article "Tug-of-War Over Women Accountants: Helpers During Peak Periods / Labor Cost Savers." It pointed out how popular it has become to employ women in accounting and finance in Japanese firms. Many of them, however, are hired part time or on contract, so that the firms do not have to pay bonuses, pensions, or other fringe benefits. Trading firms and banks, in particular, have taken in greater numbers to a "project system," where professional women are hired through contracting agencies in teams, according to the needs of the moment. In part, this trend is stimulated by the growth of new technologies in the office, such as computers and-word processors, which enable relatively few women to accomplish a great deal of work in a short time. Fort the women, it also is an opportunity to work on a more flexible schedule, and earn substantial wages. But it still excludes them from the company hierarchy.

Are Working Women Just Playing?

With new technologies and changes in the labor demands, Japanese women are well positioned to make a major contribution to Japanese business and industrial worlds. Married women, now that they live nearly eighty years, need devote an increasingly smaller percentage of their lives to child-raising and, because of mechanization, to housework.

Will women take advantage of the new opportunities? Japanese government statistics for 1960–1978 show that the percentage of women working actually declined for all age groups up to 45 (by nearly ten percent among the 25–29 year olds) [5]. Given tradition, the low level of employment evailable to women at the time and increased wealth in Japan, it is perhaps not surprising that more elected to concentrate on the home. More recently, however, Kei Sahashi, the president of Consarun, made a dismal comment on women in the job market. Her firm counsels and trains women who wish to re-enter the work force, but she claims that of the one thousand graduates of her program over the last fifteen years, only five percent still hold jobs. Worse, many wives, she said, were only "playing at being working women." [6]

All evidence is not in yet, but it suggests that Japanese women are eager to commit themselves to careers if given the chance to do something interesting and meaningful, especially if they can do so without putting their families second. Should women be able to enter Japanese corporate and government worlds in force, with positions of true responsibility, they would bring with them new perspectives and patterns of communication. As Japan moves into the 21st century, this can only enrich the society as a whole.

NOTES

1) Wendy Cole, "Climbing the Bureaucratic Ladder," PHP, May 1983, p. 74.

2) Sharon Noguchi, "Career-Minded Co-eds Don't Have It Easy," PHP, Feb. 1984, p. 30.

3) Cole, p. 30.

4) "Haado na Tatakai, Sofuto Sangyo: Sofutouea Kuraishisu." Nikkan kōgyō shimbun, July 13, 1984.

5) Kittredge Cherry, "Breaking New Ground," PHP, Feb. 1984, p.15–16.

* * *

APPLE LAUNCHES ADVERTISING CAMPAIGN FOR MACINTOSH IN JAPAN, IS DEVELOPING JAPANESE-LANGUAGE SOFTWARE

According to an article in the Sept. 11, 1984 issue of the Wall Street Journal, Apple Computer Inc. has introduced a version of its Macintosh computer which has a memory capacity of 512K. The new model, which has a suggested retail price of $3,195, is identical with the earlier model except for its augmented memory capacity.

Apple is also trying to make inroads in the Japanese market. Apple's Japan unit, Apple Computer Japan Inc., was created in June, 1983 and was recently said to be employing 18 people. Apple has recently been engaged in a gigantic advertising campaign for its Macintosh computers. It engaged the Japanese branch of Arthur D. Little to carry out market surveys and during August placed huge, full-page advertisements in Japanese newspapers. The advertisements, featuring "tasteful" Japanese calligraphy and paintings, were on a level of refinement totally different from any of the Apple advertising material ever seen in American newspapers, and the advertising budget spent in Japan must have been enormous.

Japanese-language software is of crucial importance for Apple if it hopes to break into the Japanese market. Japanese newspapers reported in June that Apple had organized a consortium of about 20 Japanese software companies to develop software for the Japanese market. According to an article in the Nikkei sangyō shimbun, (August 8, 1984), at least six more months will still be necessary before the Macintosh operating system will be equipped completely with Japanese-language processing functions, but programs such as JL wordprocessing, spreadsheets and database software probably will be available in Japan before then. Canon Sales, which is acting as Apple's key distributor in Japan, says that it expects to sell 1,000 units a month of all Apple-brand personal computers this year (Nihon keizai shimbun, August 14, 1984).

An article in the Nikkan kōgyō shimbun on September 26 says that the first Japanese-language word-processing software for use with the Macintosh and the Lisa has now been developed and will go on sale in late October. The software, called "Easy Word," was developed by a company called Ergo Soft (Motoakasaka 1-2-5, Minato-ku, Tokyo, 03-478-2234). It has a dictionary of 30,000 words. The software will be demonstrated in the Apple Computer Japan booth at the Harumi Data Show beginning on September 26. Ergo and Apple are currently negotiating about the sales price and sales system for the software. Interesting features include a multi-window function allowing the user to display a number of documents at the same time; documents can be moved or copied from window to window. In addition to the dictionary, a "virtual" kanji tablet can be displayed on the screen; the mouse is used in inputting to point to single kanji or symbols from the tablet.

Apple is currently producing about 40,000 units of Macintosh computers a month at its factory at Fremont, California, but it plans to double its manufacturing capacity for the Macintosh by the end of 1984.

In the meantime, other printers are now becoming available for use with the Macintosh. I saw an advertisement for the Toshiba P1340, a high-resolution printer with a 24-pin dot matrix head, which sells for $995 in the U.S. It can work with many different computers, including both the IBM PC and the Macintosh. (Wall Street Journal, Sept. 12, 1984) Besides, Steven P. Jobs, Apple Computer's chairman, says that-Apple plans to introduce early next year a laser printer that will represent "a dramatic breakthrough" in laser printers, and also to introduce hard disks for the Macintosh, perhaps within 18 months. (Wall Street Journal, Sept. 13, 1984)

* * *

ANALYTICAL CHEMISTRY SOCIETY WILL PUBLISH JOURNAL IN ENGLISH

The Japan Society for Analytical Chemistry, which publishes the Japanese-language journal Bunseki kagaku (Japan Analyst) has decided to publish another journal in English beginning in April, 1985. The new journal will be titled Analytical Chemistry Journal. It is considered very unusual for a Japanese learned society with a membership of about 8,200 to embark on publication of an English-language journal. (Nikkei sangyō shimbun, Sept. 17, 1984)

* * *

COMPREHENDING TECHNICAL JAPANESE

Jay Kilpatrick, a reader in Dazaifu-shi, contributes the following idea:

So far the newsletter has had two reviews of Comprehending Technical Japanese by Edward E. Daub, Byron Bird, and Nobuo Inoue and both were favorable. In his review in No. 19, John Bukacek points out that the Final Translation Test, for which the authors provide no example translation, "is not really a test if there is no way to check success or failure." I agree. How about printing one of these tests in TJT each month (there are 25 tests in all, one at the end of each chapter)? Readers could send in their translations, two or three of which could be run in the following issue, possibly along with comments or corrections by you or another translator at Peter Lim's "veteran" level. This would set a one-chapter-a-month pace for beginners like me currently using the text, and would also provide a means for other more experienced translators to compare their work with that of others. After 25 months a very nice collection of annotated translations would be completed and could be made available to future users of the text.

It sounds like a good idea. I am asking John Bukacek to coordinate this and will be glad to publish the trial translations if readers will submit them. See the following article by one of the textbook's authors. DLP

* * *

COMMENTS ON "COMPREHENDING TECHNICAL JAPANESE"

Edward E. Daub

I want to share with readers of TJT some of the background to how Comprehending Technical Japanese took the form that it did and also respond to some comments about the book that have appeared in these pages. I certainly welcome the interest you have shown in our book because, when we wrote it, the importance of learning to read Japanese technical literature was not widely recognized.

However, the past decade has made it quite clear that Japan could become the world center for new scientific information and engineering invention. Bob Bird and I recognized that possibility when we began to pool our ideas for a book to teach Western scientists how to read technical Japanese. We both had directly experienced the high level of Japanese science and engineering, Bob as a Fulbright lecturer in chemical engineering and I as an assistant professor of chemical engineering at Doshisha University in Kyoto. Since we came to learn the language along different routes, we emphasized different strategies and these differences helped shape the format of the book.

Bob is a very talented linguist as well as a world-renowned chemical engineer and a member of the National Academy of Engineering. He is highly skilled in the major European languages and has attained such a competence in Dutch that he has co-authored a textbook on Dutch grammar and a book of Dutch readings. He learned Dutch while he was a Fulbright scientist in the Netherlands years ago, and it was as a Fulbright lecturer in Japan that he took on Japanese as another linguistic challenge.

After working through the basic levels of conversational grammar and drills, he turned to developing his reading skills. However, he found the traditional readers frustrating because he wanted to learn technical Japanese and the readers focused on social customs and literature.

Taking his strategy from his experience in learning Dutch where he searched out the most frequent grammatical constructions and idioms, he now sought to determine the most frequent kanji in physics. To do so, he read an excellent high school physics text and painstakingly counted each character as it appeared. He published this study, and it became the basis for our selection of kanji in the lessons on physics.

Thus, Bob's experience with Japanese became the basis for our decision to base the book on the 300 most frequent kanji in physics, chemistry, and biology. The order of presentation, of course, came from the fact that physics is the most fundamental science, followed by chemistry and then biology.

I learned Japanese along a different path. As a newly appointed Presbyterian missionary in 1950, I studied the old World War II conversation course at Yale and then studied the Naganuma readers in Japan during my first five year term. I worked with pastors in industrial areas of Osaka and gained a deep appreciation for the strong rebirth of industrial productivity in Japan.

In my second five year term, I taught chemical engineering at Doshisha University and there encountered technical Japanese for the first time. I had very little difficulty because of my background in the sciences and my preparation through studying the Naganuma readers. I did find the new uses of kanji that I had learned in reading the Bible, theology and sociology fascinating and admired the genius of kanji to express technical language.

I judged that my ability to read technical Japanese fairly fluently came about because I had already learned those kanji in assorted contexts. Thus, my experience gave birth to the supplementary readings designed to give the student practice in recognizing the new kanji in a variety of texts.

A recent comment in TJT suggested that it was a weakness in our book that we did not provide the reader with translations of the supplementary readings. I can understand the reason for that criticism from the point of view of a non-scientist seeking to attain competence as a technical translator.

However, as the title of our book suggests, we sought to help students comprehend technical Japanese, and comprehension does not require translation into polished English. We made Western scientists the target for our book and expected that they would want the skill to read rapidly through Japanese technical papers in their fields in order to find the new information in them. Much of a technical paper consists of information already known to scientists in that field. So we wanted students of our book to work through these readings for content and not for practice in translating.

The other criticism, namely, that we should have provided translations for the Final Translation Tests, is well-taken. I still cannot see a way to provide those translations and yet expect students to hammer one out first on their own. In class I first have the students offer their own versions before I introduce my own.

The translation tests that appear in the first half of the book were written by the third member of the haiku that appears at the close of our introduction, Nobuo Inoue. Nob, of course, came to Japanese the ideal way, from the cradle on. He learned his science in another ideal way, as a student at Sanko and Kyodai in the venerable Imperial University tradition. After World War II he spent a year in Great Britain, and he also taught for several years at Purdue University and did research at the Franklin Institute in Philadelphia in the 1960's. I came to know Nob as a delightful colleague and friend at Doshisha, and he contributed to our book the authenticity that only a Japanese scientist could bring. He wrote all the Construction Examples and helped us find suitable books for selecting the Supplementary Readings.

In this issue of TJT a proposal is made to have shiroto submit their versions of the final translations to be followed by comments from kuroto. I look forward to this dialogue. Moreover, when TJT takes up a given chapter in that manner, I hope that readers who have questions about particular phrases and constructions in the Supplementary Readings for that chapter will bring them up for discussion as well.

Thanks again for the positive things you have said about Comprehending Technical Japanese and for the thoughtful criticisms you have raised. I hope that the book will serve well the network of freelancing technical Japanese translators.

* * *

CHINESE EDITION OF INTER PRESS DICTIONARY WILL BE PUBLISHED

The Inter Press Dictionary of science and engineering (Kagaku gijutsu 25-mango daijiten) (published April, 1983 and reviewed in June 1983 in TJT no. 2, pp. 35) is going to be published in a trilingual Chinese edition. A contract was signed in Peking on July 28, 1984 between Inter Press and a Peking research institute on metallurgy.

The metallurgy research institute will organize a committee consisting of 50 leading Chinese specialists who will assign Chinese equivalents. Another 50 experts will supply Pinyin pronunciations for them. The project is scheduled to be completed in two years; work will start in September, 1984 and will be completed at the end of August, 1986. (Kōgyō eigo, October 1984, p. 104–105)

* * *

SPLITTING THE TWO VOLUMES OF THE INTER PRESS DICTIONARY

A number of translators have thus far refrained from buying the two-volume Inter Press Dictionary of Science and Engineering because it is so expensive and they felt that the E-J volume would be less useful to them than the J-E volume. For example, Jay Kilpatrick of Dazaifu-shi writes:

A question for experienced translators: For a translator working solely from Japanese to English, is the English-Japanese volume of the big Interpress dictionary useful? If so, for what? As I am planning to work only from Japanese to English, it seems to me I won't need it. I am aware that the dictionary is sold only as a two-volume set, but I think some way could be found to purchase just the volume I need.

To answer the question, the J-E is of course more useful to anyone working in that direction, but both volumes are useful. The kana arrangement of the entries in the J-E volume is sometimes so frustrating and time-consuming that you may be able to find what you want more easily in the E-J volume, if you know approximately where to look. The E-J volume also contains abbreviations which are omitted from the J-E volume.

About purchasing just one volume, Cliff Bender of Kyoto recently wrote me and said that readers in Kyoto recently held a meeting. One of the persons who attended the meeting teaches translation to Japanese E-J translators and students. According to Cliff, it might work out that some of these translators or students would be interested in buying the E-J half of the set if someone else wanted just the J-E half. Splitting the cost down the middle would be logical, says Cliff, and might make the dictionary affordable for many who need only half the set. Ron Granich and Cliff Bender would both be willing to help coordinate this. Their addresses are in TJT no. 15, p. 35 and in no. 16, p. 29.

And also let's not forget about the Taiwan edition of the Inter Press dictionary, which is available in Tokyo and sounds like quite a bargain (see TJT, no. 16, p. 29).

* * *

MORE ABOUT DICTIONARIES

The publisher Baifuukan has published a new dictionary of physics titled Butsurigaku jiten. More than 760 contributors wrote the articles in the new dictionary, which has 2,538 pages and some 13,000 articles. Head of the editorial committee is Tetsuji Nishikawa, director of the National Laboratory for High Energy Physics. The dictionary is illustrated with numerous figures and photographs and has English, German, French and Russian indices. Price is ¥28,000 until the end of March, 1985. After that the price goes to ¥32,000. (Nihon keizai shimbun, August 30, 1984)

A three-volume biographical dictionary of Japanese active in various fields is scheduled to be published on September 20. The title is '84 nenkan Jinbutsu joohoo jiten. It lists biographical information and recent activities of some 11,000 prominent persons and also lists articles about them in magazines and newspapers. Volume I deals with leaders in politics, economics and industry; Volume II with leaders in learning, literature and "society"; and Volume III with leaders in the arts, sports and various social activities (sesoo). The volumes cost ¥13,000 each and can be bought separately. The publisher is Nichigai Associates (Ohmori Kita 1-23-8-918, Ohta-ku, Tokyo).

The September 1984 issue of the SWET newsletter has an article about "Japanese-English dictionaries." It discusses Kenkyusha's New Japanese English Dictionary (4th ed.), the Kodansha Japanese-English Dictionary, Sanseido's New Concise Japanese-English Dictionary, New Standard Japanese-English Dictionary, Shogakukan Japanese-English Dictionary of Current Terms, Wa-Ei Hon'yaku Handobukku, and Gendai-yoogo no Kiso Chishiki.

* * *

PUZZLE

This issue's bonus is the following word puzzle which I am reproducing from issue no. 1 of GS (June 1984). GS is an abbreviation for "la gaya scienza" and is a new Japanese magazine published by Tojusha. You should allow yourself about a half hour. Hints: Start with hajimari and end with Daedalus. There are a couple of misprints but don't let them bother you.

[Scanned Image No.1]

* * *

YELLOW PAGES JAPAN IN USA LISTS TRANSLATORS

Yellow Pages Japan in USA 84 is an intriguing directory published in (July?) 1984 which lists addresses and phone numbers of Japanese firms and businesses in six U.S. cities: New York/New Jersey, Chicago, Houston, San Francisco, Los Angeles, and San Diego. Unfortunately, it has no index of the names of the companies listed. It was published by Apcon International, Inc., 548 So. Spring St., Suite 1039, Los Angeles CA 90013. It costs $18.50 at Kinokuniya in San Francisco and is much less expensive than the PressAid Directory of affiliates and offices of Japanese firms in the U.S.A. (see TJT, no. 18, p. 32). The latter costs $88 and is, I suspect, only slightly more useful. The Yellow Pages lists the following "translators" under the various cities:

New York: None listed

New Jersey:

Japan Desk

312 5th St., Palisade Park, NJ 07650

Japanese Interpreting Associates

10 Edwards Point Rd., Rumson, NJ 07760

Chicago:

Trans-Lingual Communications, Inc.

233 E. Wacker, Chicago IL 60601

Houston: None listed

San Francisco:

Japanese Language Workshop

16 California, San Francisco, CA 94111

Japanese Linguistic Service

61 Renato Ct., Suite 1, Redwood City, CA 94061

Japanese Translation Consultant

707 Stockton St., San Francisco, CA 94108

Sanae International

253 Grant Ave., San Francisco, CA 94108

Los Angeles:

Asian Translation Center

250 E. 1st St., Suite 910, Los Angeles, CA 90012

Arai & Evans Translation service

869½ N. Alexandria Ave., Los Angeles, CA 90029

Logo's International Service

184 E. Casuda Canyon Dr., Monterey Park, CA 91754

Fuji Translation & Interpretation Service

3514 S. Garnsey St., Santa Ana, CA 92707

Niijima & Mitani Translation Service

321 E. 2nd St., Suite 605, Los Angeles, CA 90012

Syn Tech

1365 Riviera Dr., Pasadena, CA 91107

San Diego: None listed

* * *

REPORT ON THE AMERICAN TRANSLATORS ASSOCIATION CONVENTION New York, September 19–23

by John Bukacek

I would like to highlight the events of the ATA convention which I consider of most significance to translators of technical Japanese.

Carl Kay chaired a meeting of the Japanese Translators' Specialized Interest Group. Plans for an ATA accreditation exam for Japanese to English were discussed. Ronald K. Jones of Yokohama proposed sending out an anonymous questionnaire about rates for J-E technical translation.

Hisashi Kubota held a workshop on Translating Scientific Japanese. Mr. Kubota distributed a list of general and technical dictionaries drawn up by Carl Kay. He also discussed some of the fine points of patent translation.

Don Gorhazn discussed the need for J-E interpreters and translators in government. His workshop included a talk by Leroy F. Butkus of the Joint Publications Research Services, who stressed the great need for J-E technical translators.

The presence of J-E translators at the ATA convention was greater this year than ever before. Our voices are being heard at ATA, and our influence is increasing. We not only need to organize our own J-E interest group, but we must also be part of a larger professional organization such as the American Translators Association. This will increase our visibility and influence in the translation profession as a whole.

* * *

TRANSLATING SCIENTIFIC JAPANESE

Hisashi Kubota

[Editor's Note: This is a slightly abridged version of a paper delivered at the New York convention of the ATA. I am omitting the handouts which were passed out at the meeting. Many thanks to Dr. Kubota for sharing the paper with TJT readers.)

It is the speaker's opinion that translating scientific Japanese is a different world from translating literary Japanese.

Probably the first feature of a scientific paper in Japanese that one becomes aware of is the more direct nature of the statements. As a result, a scientific paper mechanically is easier to translate than a literary one. However, knowledge of the subject matter assumes greater importance. Thus it seems wise when starting out on a career in scientific translation from the Japanese, that one start off with a subject that he is most familiar with before branching out into other areas.

The question of who is better qualified to do scientific translations — the subject specialist or the language specialist — is a matter of considerable debate. I do not propose to take a stand for the subject specialist except to say that a knowledge of the subject often enables the translator to figure out some technically complicate passages.

The first preparation for doing translations is the acquisition of dictionaries. We are indebted to Mr. Carl Kay for the handout which is a list of dictionaries he has compiled with his assessments of their utility. [See TJT no. 16, pp. 25–28. — Ed] The only dictionaries that will be mentioned here are the basic ones which everyone should have. The first is Kenkyusha's New Japanese-English dictionary. I don't know what the latest edition is, but I have a Fourth Edition which has given me very good service over the years. Because one must be able to give a phonetic rendition of any term before he can look it up in this dictionary, there is also need for a character dictionary. Probably the one best suited to translators whose English is the stronger language is Nelson. A person more versed in Japanese would probably do well with a straight Japanese character dictionary. On the other hand, it would be wise for everyone to obtain a Japanese character dictionary because it is useful for looking up characters which sometimes do not appear in Nelson, gives various phonetic renditions of a character which are useful in determining some odd sounding names, and allows one to look up a character from its phonetic rendition.

I want to add here that there is a monthly publication titled Technical Japanese Translation edited by Mr. D.L. Philippi of San Francisco. New dictionaries are mentioned, which, along with the many interesting articles it contains, makes it well worth reading. It is very difficult to judge a dictionary sight unseen, and if you know of anybody who has one such dictionary you desire, it would be a good idea to find how good it is before spending the all too often considerable sum for its acquisition.

I want to add a few dictionaries of my own that I have found useful from time to time. The first is Agne's Dictionary of Metallurgy. It covers the basic metallurgical area very well. Many of the papers one sees in metallurgy are more concerned with actual production, and the terms used are unique to the trade and are not found in most dictionaries but have to be derived from the text. The Monbusho puts out a series of small dictionaries in the various disciplines, and some are rather useful. I have in mind the ones on chemistry and mathematics. In the latter, there is a special statistics section which, though very incomplete, was the only one I was able to come across 20 years ago.

A good English dictionary is a must, and Webster's New International Dictionary is a handy one to have. Other useful English books are glossaries and dictionaries of technical terms. I do a lot of work in chemistry, and chemical terminoloy, particularly organic terms, can be a problem. A copy of Merck has often proved helpful as well as a catalog put out by a chemical supplier such as Aldrich.

Now what does one do when a term appears which cannot be found in any of the dictionaries? This is a very good possibility if a paper is a very recent one in a newly developing area. A term may appear for the first time, at least in Japanese, and even the dictionaries have not kept pace. I hope to point out a few areas where these terms sometimes appear, and if one is on the alert he may be able to decipher the new terms.

The following discussion will describe a typical scientific paper in Japanese and some of my experiences in dealing with the different sections and the figures and tables.

Looking now at a typical scientific paper, we find that it generally follows the formats used by our writers.

First of all, there is the title. More often than not, there is a title in Japanese and one in English. This is often very helpful. Many Japanese terms can have so many different meanings that one can be stuck with the proper translation. When this equivalent is given in the English title, one is that much ahead.

On the other hand, the title given in English often is quite different from what you would translate. The author of the title knows his subject and what he wants to say and is probably more correct in his rendition. Let me give you an example. At the top of one of the handouts is a Japanese title under which is a title in English. [The Japanese title romanized is: Aruminiumu oyobi doo funmatsu atsumitsutai no nekkan oshidashi ni okeru kakooryoku to henkei. The "supplied" English title is: "Load and material flow in hot extrusion of aluminium and copper powder compacts" Ed.] Now if I were to translate the Japanese title, I would have said: "Working force and deformation in hot extrusion ...." On going through the paper, I find that the author was indeed talking about the material flow which is part of the deformation process and the load that causes such changes. On the other hand, I had an example in which another author made what I considered a serious mistake in his choice of the English equivalent.

Next is the author(s). This brings up the problem of transliteration. There are the old Hepburn school of transliteration and the more recent, "official" Japanese mode. One must know both systems in order to be able to use the dictionaries. Old standbys such as Kenkyusha use the Hepburn system while the newer technical dictionaries are more likely to use the latter system. Now the transliteration of the authors' names for the customers' benefit is the question. I prefer the Hepburn system for my customers because it is easier for them to pronounce the names. I do like to depart from strict Hepburn in the case of a long o by adding an h. As you may have heard Pat Newman say, one usually need not translate journal names but simply transliterate them.

Next is the abstract. This may come at the head of the paper or at the end. If the abstract is in Japanese only, then it has to be translated. There are many papers which have abstracts in both English and Japanese. It is wise to compare these abstracts because they often do not say the same things. Many times the abstract gives a lead on what was done which is not given in the text. An English abstract often serves as a valuable source of technical terms. By comparing the terms used in the English abstract with what appears in the Japanese text, new terms often come to light.

The introduction usually discusses the history of the subject. The next part is usually the experimental section. This can be a very easy section if you are working in your own discipline. On the other hand, if one is in a subject which has a terminology of its own (say biochemistry compared to chemistry, or medicine compared to metallurgy), this section can prove very trying.

Probably the same things can be said for the experimental results and discussion. The discussion often proves to be the most trying section because here the authors often speculate or introduce rather nebulous concepts which may be difficult to follow.

Most papers come with an array of figures and tables. Here again, there is a strong tendency. to present the figures and tables in English, in which case once more they can prove to be a valuable source of terminology.

When a table is in Japanese, abbreviations are often introduced to conserve space. Just what the abbreviation stands for may prove to be rather formidable. Close look has to be taken on the data that follows to get a lead on what the heading over the data may be. This is another situation where a specialist may have some advantage.

Let me introduce here some of my experiences in metallurgy. Many of the papers I handle discuss mill processes, and some of the terms do not appear in my metallurgical dictionary. On the other hand, these terms may appear in the figures and tables in English and by closely checking the text with these terms, a considerable vocabulary can be built up.

The bibliography that comes last is often useful. If a paper by an English writing author is referred to in a subject that is very vague, reading it may make possible better understanding of the current subject and may even give a clue to the terminology.

We next consider some of the care that needs to be exercised in translating scientific documents.

The Japanese language has tense, although it is not as evident as in English. There is a tendency to stick mainly to the present tense, particularly in scientific papers, but there is often a rapid switch in tenses which is not too bad. In fact, one can either follow their tense or stick to either the present or the past and still come out in good shape.

The more serious problem is number. There is no singular or plural, particularly in the verbs, and one must exercise great care in deciding whether the singular or plural case applies in any given situation.

When I was new in the game, the art of translation was explained to me in the following way. Read the text. Make sure you understand what is being said. Next take your eyes off the original text and state in English just what was said. Then go back to the text and check, making sure you have not left out anything. I bring this up because one of the greatest handicaps in translating from Japanese that I have observed in many people is the tendency to let the Japanese structure control your translation. You can check for yourself on this matter. Where a Japanese paragraph has sentences all starting off with introductory clauses, a faithful translation will look rather awkward as the final copy. It is often prudent to take some such sentence and turn it around to give a cause and effect relationship to make for better reading. This brings us to the question of whether it is permissible to break up long awkward sentences in the interest of better translation. There is nothing wrong with this practice, although some customers do not like it, but as long as nothing is omitted, there should be no cause for concern. In fact, more often than not, it makes for a better translation. One note of caution here is that this does not apply to patents. The legal aspects of patent translation require faithful adherence to the original where structure is concerned, and even very clumsy and lengthy paragraph long sentences should, wherever possible, be translated intact.

Now to mention some problems associated with the Japanese written language. The first is the rendition of proper names. There are some dictionaries with readings of personal names, place names, and other proper names, but they are not the final answer. That is because some names can be read in a number of ways. There is no pat answer for this problem. Where an author has had several publications previously, reference to abstracts often proves useful in determining his name.

The second and perhaps the more irritating problem, which in my experience is becoming worse, is the random use of kana. I can recall the old days when a scholar would not be caught using kana in place of a kanji when one was available. The old newspapers and magazines all used kanji which had furigana. With the advent of Toyo kanji and the trend to simplify the more difficult kanji, there has been a trend toward slipping in kana in place of a kanji. This is usually the case when hiragana are used together with a kanji indicative of a compound word. Then there are the katakana terms which usually are direct transliterations from some foreign language. To be sure, we are rather well off in that most of these transliterations are from English terms. On the other hand, homonyms again become a problem. The transliteration of th is a particularly vexing problem. The first time I saw the term ground (arsu) some 40 or so years ago, none of the dictionaries carried it. It was a frustrating experience trying to figure what it might be. Fortunately, I stumbled across a circuit diagram which showed it, and it dawned on me what it was. Th is often transliterated su but sometimes chi or ti. Load and road both have the same kana. So do Bessel and vessel. One other sore point on the use of kana is the often very poor reproduction of the nigori and maru designations. (Example: Halite versus pearlite)

At this point I would like to introduce a positive trend that I have noticed in the Japanese journals during the past decade. Twenty years ago, the average technical paper was not well edited, and there were many obvious errors not just limited to typography. The presence of errors in the original text makes the translator's life miserable. The situation has changed dramatically over the past few years.

We will now discuss the subject of patent translations. I believe that many of you derive the greater source of your incomes from patent translation. The Japanese are prolific in their patent literature as they put out more patent publications per year than any other country. Two years ago, Paul Ishimoto discussed patents and made the distinction between the different types of patent publications. Up to a few years ago, nearly all the patent publications were of the Kokoku type, or examined patent publication. There has been a sharp shift in the past years to the Kokai (early disclosure) or unexamined patent because of a change in the Japanese patent laws. The main difference between the two was that the Claims usually came at the end in case of the Kokoku and at the start for the Kokai. This has changed, and now the Claims appear first even in Kokoku. Since the Claims is the single most important section, it needs to be done accurately. Thus by reading the entire text and coming to the Claims at the end, one is in a good position to make a good rendition. It is often wise to translate the rest of the patent first and leave the Claims for the very last. I do not favor any particular format in doing translations, but an exception can be made for the Claims. There is a definite format which is followed by at least 90% of the patents. The same applies to the first statement under the heading Detailed Description of Invention.

Where style is concerned, I would hesitate to follow the style adopted in writing an American patent. I believe it is the translator's role to make a faithful rendition of what is being said in Japanese to English, and if there is any further revamping of the rendition, then it is up to a technician versed in writing patents the American way to make the change.

Finally I want to discuss some of my pet gripes. The first, and to me the most serious one, is the poor copy one often gets. This seems to apply more frequently to patents. Newspaper articles are pretty bad as is, but when reproduced can be very bad. The second is the time factor. I feel that I can do the best job if, after I finish a certain paper, I lay it aside for a couple of weeks before proofing it and making the final copy. Of course, this problem can be eliminated for the most part if you have someone else edit your work. Most of us freelancers are lone wolves, however, and the time factor requires us to proof the work and send it off post haste. A third is the situation that often crops up when you are asked to do only a certain section of a paper. This is not bad if the whole paper is available so you can read through to make some sense of what the particular section covers, but it is a problem when all you get is just the section desired.

The above represents some of my impressions regarding translations from scientific Japanese. It is possible that some of my impressions and ideas may not agree with yours, and if such is the case, please let me know. I hope that some of the thoughts presented may prove useful to some of or as it would be said in Japanese, Saiwai to omou.

* * *

JOB OPPORTUNITIES

The September 1984 issue of the Swet newsletter contains the following classified ad for a job opening in Tokyo:

Japanese computer manufacturer seeks an editor/rewriter for English computer manuals, brochures, and technical documents. Must be native speaker with technical writing experience and background in computer hardware. Send resume to Mr. Nitta, Overseas Operations Department, Computer Division, Hitachi, Ltd., 6-27-18 Minami Ohi, Shinagawa-ku, Tokyo 140.

The September 7 1984 issue of OCS News had the following ad for a technical translator:

The Xerox Corporation Technical Information Center has an immediate opening for an experienced Technical Translator (Japanese). Responsibilities will include translating Japanese patents and journal articles and translation and indexing of internal technical reports. Must be fluent in Japanese and English and possess a strong technical background in engineering or science. Send resume to: Herbert I. James, Ph.D., XEROX CORPORATION, Joseph C. Wilson Center for Technology, Building 128, Webster, N.Y. 14580.

* * *

LETTERS FROM READERS

James D. White of Sayama-cho, Osaka-fu, writes the following to clear up a few of the mysteries about words for "rice":

I'm writing in response to your article on "Foreign Loan-Words Abound" (p. 23, TJT No. 19). Meshi does not belong to the dark ages and the preference for raisu over gohan is not simply based on some internal psychological need to "appear modern." Which is used when depends on how it is served, i.e.: If on a plate — you are being served raisu, if in a chawan (small rice bowl) — you got gohan and, if in a donburi (large bowl) — then enjoy it as meshi. Don't ask me "why," but considering that it is kome before being cooked and ine before being harvested... then why not?

He also adds:

Most of the translation I do, strictly small scale and parttime, involves audiovisual education and education technology but I still run into the same problems of katakana words, acronyms invented yesterday and known only to the author, etc... In closing, anyone else out there struggling with the "soft sciences"?

Jay Kilpatrick of Dazaifu-shi in Kyushu, writes the following (September 8) about the newsletter:

A comment: Some readers have suggested that you organize each issue around a theme. This probably would be a lot of extra work for you and would not, in my opinion, improve the newsletter. In fact, I suspect it would make it less interesting. I like your current approach. Every issue is full of surprises, and going through them is kind of like opening a grab bag. "What will I find on the next page?" I wonder.

This issue's "surprise" is the puzzle on p. 31. It will be both fun and educational for translators and non-translators alike.

Cliff Bender of Kyoto writes (September 3) to say that the new, third edition of Kōjien (published in 1983) has a definition of ima hitotsu in the sense ("not quite") discussed in no. 18, but the second, revised edition (published in 1976) does not. This might indicate that the word came into use, or at least become more accepted, between the years 1976 and 1983. He also found karaoke in the 1983 edition but not in the 1976 edition. Dasai was in neither edition. He continues:

Considering these findings, it occurred to me that many translators in the States might not have a regular kokugo jiten. Many technical dictionaries have been recommended in the newsletter but I do not recall any Japanese dictionaries. I have found the Kōjien though there are others both bigger and better, quite handy for identifying nontechnical words not in Kenkyusha's. It is, as you undoubtedly know, a fairly respected dictionary and can be quite helpful in learning nuances of several similar terms. Of course, if someone is looking for English equivalents a kokugo jiten is not usually the place to go.

Last week Ron Granich organized a meeting.. . one thing we were struck by was the apparent groping for fairly common words and phrasings some translators in the States must go through. It was mentioned, too, if there might be some way a word exchange or definition service between here and there might be set up. Perhaps a regular column or feature in the newsletter? What might you or other readers think about the practicality and merits, or demerits, of this or some other system?

The new third edition of Kōjien was indeed mentioned in the newsletter in December, 1983. See no. 10, p. 17. I have the older edition but seldom find it useful. Groping? Yes, we do grope quite a lot, but so do translators in Japan. One's precise physical location on a map does not make that much of a difference, I imagine. What counts is access to dictionaries and other resources.

A definition service would be useful, I think. I have long advocated the establishment of a sort of resource center where translators working at home could phone in their reference questions to be answered by a reference librarian with a large library of technical dictionaries. In line with my proposal in TJT no. 16, pp. 10–11, I am going to propose the establishment of such a center at the University of Wisconsin when I visit there in November. Such a service would be useful only if it provided information of guaranteed accuracy based on a large library or, even better, on a well-organized database and gave out the information without delay (in real time). Translators who need information usually need it urgently and certainly can't wait for the post office to deliver letters. Perhaps the readers in Kyoto would like to establish a telephone hot-line which translators from all over the world could call up on a 24- hour basis.

* * *

INTRIGUING QUOTATIONS

It is fair to say that Japan has one of the most active trade presses in the world, but it is inaccessible to probably 98 per cent of American businessmen who would be interested in it.

John P. Stern, Senior Representative, American Electronics Association, head of the AEA representative office in Tokyo. Quoted in Japan Economic Journal, September 11, 1984.

Our language is profoundly dualistic. It is a terrible drag. I have already said once before, what a drag language is on our thought, making it impossible to express the relationships of the existing universe... if we understand to what extent our language hides from us the true picture of the world, we shall see that it is not only difficult but absolutely impossible to express in our language the true relationships of things of the real world.

P.D. Ouspensky, Tertium organum, Rev. tr., 1982, p. 156–157.

機械翻訳が今後どのように進展していくかを予瀕することは困難であるが、いず れの方向に進むにしても、記号化された情報を基にして自然言語を処理していく限 りにおいては、.処理しなければならない情報は無限に近くあり、現在の進歩したハ ードウェアを用いてもそのごく一部しか処理しきれないであろう。真の自動翻訳を 実現するには、人間の言語処理のメカニズムに対する基礎的な研究から再開し、そ の言語学習等を解明していくことが必須である。その意味で機械翻訳も人間が自分 自身の秘密を探究するという永遠のテーマの一つであると言えるものだと思われる

内田裕士(富士通研究所)「bit」1984年9月号 p. 50

言語は定義のできない対象である。定義のできないものは明確なモデル化ができ ない。ある物事を工学の対象にしようとすれば、必然的にその物事に対してある種 の枠をはめることになる。言語を機械翻訳しようとするならば、必然的に言語を工 学の対象として、ある種の制限のもとに使うということを覚悟せざるをえない。文 学的文章は当然対象外としても、−般の科学技術文章でも、表現が過度に複雑であ ったり、文として不適切な表現であることも多い。人間が不自由なく書けて、あい まいさのない正確な表現であり、なおかつ機械が解析可能な文章とはどのようなも のであるかを明らかにし、そのような範囲で機械翻訳を実射ヒすることが大切であ る。

言語は自然現象とちがって、少数の基本原理ですべてを十分に説明することは不 可能な対象である。言語は無限に近い組合せ問題であるともいえ、これを取り扱う ためには、この組合せに関する知識を文法や辞書情報の形ですべて記憶するという 膨大な根気のいる仕事をしなければならない。そういったときに最も大切なことは、 機械翻訳システムの構造を能力の高いものとし、無限に出てくる種々の言語現象を その都度解決できるようにすることが必要で、そのような機械翻訳のモデルを伴っ ておかねばならない。

長尾真(京都大学工学部電気工学第二教室)「科学」 1984年9月号 p. 531−2

* * *

DIRECTORY

Many useful contacts have been made through the directory, not only by persons who are looking for translators, but also by translators who want to contact each other. For your convenience, I am including in this issue again a composite directory which embraces all the names submitted so far.

Readers should let me know promptly when they move so that I can change their directory entries.

APPS, F.R.D. Owner, Ingatestone Translations, 57 High St., Ingatestone, Essex, CM4 OAT, England. Japanese to English Technical Translation. Specialties electronics, organic chemistry.

BRAND, Russell. Director, Seijo Language Center. 9-21-15 Seijo, Setagayaku, Tokyo 157, Japan. Tel. (03) 484-0257.

CHANDLER, Brian. 10, Simmondley Grove, Glossop, Derbyshire SK13 9NQ, England. Tel. Glossop (04574) 61906. J-E Technical Translation. Specialties computers, electronics and patents.

DAVISON, Lucile S. 50 Tomlinson Rd., Woodbury, CT 06798. Tel. (203) 2634207. Japanese, Russian, German, and French to English. MS in Chemistry, Univ. Oklahoma 1940; technical librarian for U.S. Rubber Co. 1942–45; abstractor for Am. Petrol. Inst., Chemical Abstracts, Metals Abstracts; associate editor Rubber Chemistry and Technology 1967–80. Specializes in chemical literature and patents.

FIELD, Rikko. Director, JLS, 61 Renato Court, Suite 1, Redwood City, CA 94061. Tel.: (415) 321-9832. Technical translation and interpreting; IBM Displaywriter-processed copy; Japanese typesetting and typewriting. All technical fields including patents and legal documents.

FRASER, D.A., BA, MIL, 22 Gresley Road, London N19 3JZ, England. Japanese to English. Chemistry, medicine, life sciences. Freelance translator.

GLEASON, Alan. 1280 4th Ave., No. 3, San Francisco CA 94122. Tel. (415) 753-2702. Member NCTA. J-E translation and J-E/E-J consecutive interpreting. Fields: electronics, chemistry, computers, business, economics.

GORHAM, Don Cyril. 12900 Old Columbia Pike, Silver Spring, MD 20904. Tel. (301) 3849512. Jpn/Eng Eng/Jpn interpretation (consecutive) and translation. Liberal arts background; graduate of pre-WWII Jpn schools. Ten years experience as freelancer: economic, legal, political, and some technical (electronics, fishing, nuclear energy).

HAZELRIGG, Meredith. 468-1 Mizuno Sayama-shi, Saitama-ken, Japan 350-13. Tel. 0181-429-59-7959 (direct dial from U.S.), 0429-59-7959 (Japan). Freelance translation (general technical work with emphasis in three areas: medical, electronics, financial) and consultant in cross-cultural communication. Member, SWET. Listed, Contemporary Authors Trustee, Allegan Education Foundation (Box 155, Allegan, Michigan 49010 USA).

HENRY, Motoko. 1-21-3 Kamiogi, Suginami-ku, Tokyo 167, Japan. Tel. 03-392-4970. Simultaneous interpreter. Pharmacist with knowledge of chemical and medical terms.

HOWE, Josephine. 158 Wood Street, Doylestown, PA 18901, USA (Philadelphia Area). Tel.: (215) 348-7884. Japanese and French to English; pharmaceutical, medical, biochemical and industrial chemical research and patents. M.A. Japanese, Columbia U., New York, N.Y. B.A. English (French and Japanese minors, biosciences course work) Indiana U., Bloomington, IN.

HUNG, John Y. Owner, SANTRAN, 141 Zengel Dr., Centerville, OH 45459. Tel. (513) 4349288. Chinese, Japanese and German into English, and English into Chinese. ES, MA, MS. College instructor 1960–69, computer systems engineers 1969–73, and avionics and flight systems digital controls since 1973. Freelance translator since 1980. Specialties include patents, nuclear sciences, avionics, computers, aeronautics, electronics, software and aerodynamics.

ISHIMOTO, Paul. Japan Patent Service, 2025 Eye Street, N.W. #307, Washington, D.C. 20006. Specializes in translation of Japanese patents. Worked as translator in U.S. Patent Office for more than 15 years.

JENSON, W.L. 1-24 Tashiro Hondori, Chikusa-ku, Nagoya, Japan 464. Tel.: (052)762-6532. Freelance Technical Translation (J-E), proofreading & rewriting. Field: Mechanical engineering specializing in automobiles. (Seven years' experience with Toyota Motor Corp. and affiliates.)

JONES, David G. 3013 Yellowpine Terrace, Austin, TX 78757. Tel.: (512) 451-4381. Japanese to English. Patents covering chemical synthesis, plastics, photography, electrophotography, magnetic recording media, etc.

KAY, Carl. Owner, Japanese Language Services, 45 Putnam Ave., Cambridge, Massachusetts 02139. Tel.: (617)661-9784. Offers Japanese translation, typesetting, interpreting and research services.

KURODA, Keiko. 1655 ½A Mason Street, San Francisco, CA 94133. Tel.: (415) 7718409. Member ATA and NCTA. J-E interpreting and translation, abstracting and editing. Specialties: Legal translation and court interpreting. Technical translation in chemistry, electronics and engineering. Patents in all fields.

LAMB, John D. 5-32-10, Nagasaki, Toshima-ku, Tokyo 17l,Japan. Tel.: (03) 957-7390. Japanese and German to English; general technical work, mainly computers and electronics. B.Sc. Computer Science, London. Research in AI, Sussex University ('72-'73) and London University ('73-'75).

LIM, Peter, Ph.D. Chemistry, Syracuse University. 5220 N.E. 51st St., Vancouver, WA 98661. Tel.: (206) 694-5353 (home); (206) 695-4477, Ext. 349 (business). Areas of experience: Electronics, semiconductors, nuclear engineering, chemistry, mathematics, mechanical engineering, pulp and paper, patents, contracts.

McRAE, John R. 632 Colfax Court, Goleta, CA 93117. Tel.: (805) 964-9601. Ph.D. from Yale University (Chinese Buddhism). Computer Systems Manager, Berlitz/Agnew TechTran, 6415 Independence Ave., Woodland Hills, CA 93167.

McWILLIAMS, John J. President, The Word Shop, Ltd. 71 Karasawa, Minamiku, Yokohama 232. Tel.: (045) 261-2304. J-E translations in the fields of computers, robotics and data communications. Custom-made microcomputers with word processing software and an Epson MP-80 dot printer.

MOFFETT, Linda. 749 Ash Street, Denver, Colorado 80220. Tel. (303) 388-0284. BA, University of Colorado Japanese Studies, background in chemistry and pharmacy. Specializes in chemistry, batteries, textiles, medical.

MUNOZ, Roey. 7401 Cannon Mt. Pl., Austin, TX 78749. Tel.: (512)288-1897. Japanese-English and Spanish-English translation. BS in Japanese language from Georgetown Univ. with 6 years of translation experience. Special interest in medicine, pharmaceuticals and chemistry.

NISHIMURA, Keiichi. 461 Grizzly Peak Blvd., Berkeley, CA 94708. Patent attorney. Translates technical and scientific documents including Japanese patents into English.

OSBORN, David K. 2495 Santa Rosa Ave., Altadena, CA 91001. Tel. (818) 797-8956. Japanese-English translator, editor, business correspondent and technical writer. Experience in medical, cultural, legal and business fields. Abstracts, typing and revisions.

OZAWA, Saeko. 2140 Roosevelt Ave., #203, Berkeley, CA 94703. Tel. (415) 849-9362. Interpreter and translator, English to Japanese B.A. Waseda, M.A. from U.C. Berkeley. Experience in legal, business, telecommunications, computers and chemistry. Patents in all fields.

PACIFIC INTERFACE, Inc. 60 West 76th St., New York, NY 10023. Tel. (212) 877-9159. Tokyo office: #401 Tagami Building, 4-1-14 Nishishinjuku, Shinjuku-ku, Tokyo 160, tel. (03) 376-3908. Laurin Herr, president, Betsy Kuga Schneider, associate. Research, consulting, and J-to-E translation, specializing in computer graphics. Sixteen years combined experience in technical translation.

PATNER, Richard. 14226 61st Place West, Edmonds, WA 98020. Tel. (206) 745-8089. Japanese to English translation of patents and technical documents with over ten years experience.

PEARCY, Ralph. 1900 S. Eads St., Arlington, VA 22202. Tel. (703) 892-1081. J-to-E technical and scientific translation, much patent experience. Biochemistry, electronics, textiles, physics, math.

PHILIPPI, Donald L. 715 Tenth Ave., San Francisco, CA 94118. Tel. (415) 752-7735. J-E technical translation in all fields. 23 years experience, Tokyo 1961–1971, S.F. since 1971. IBM Displaywriter. Editor of Technical Japanese Translation since May, 1983.

RIDLEY, Charles P. 2780 Ross Road, Palo Alto, CA 94303. Tel.: (415) 328-7344. Japanese-English and Chinese-English. BA in biology from Bates College; MA and PhD in Chinese from Stanford University. 20 years experience. Specialties include medicine, pharmaceuticals, chemistry, computer technology.

RUSOFF, Arnold F. 209 Hudson St., Ithaca, N.Y. 14850. Tel.: (607) 277-2292. Japanese to English translation of computer documentation. Eight years of experience. IBM PC with Wordstar.

SATOH, Tomoyuki. 2714 E. 22nd St., Austin TX 78722. Tel. (512)479-8460. M.A. Physics, The Johns Hopkins University. Japan Information Service, Austin. J-E translation and abstracting. Excellent in physics, mathematics, chemistry, elec. and mech. engineering, and patents in all fields; adequate in medical, pharmaceutical, electronics and computer science, law, economics, etc.

SCHODT, Frederik L. 144 Parnassus Ave. Apt. 14, San Francisco, CA 94117. Tel. (415) 681-9803. Member A.T.A. and N.C.T.A. Japanese-English interpreting, translation. Specialties computers, electronics. Epson QX-10 word processor.

SHIELDS, John N. CPO 42, Nagoya Japan 450 (Fax & business phone in Japan: (052) 7222-7181) and 1380 Lone Oak Road, Eagan, MN 55121 (Fax & business phone: 612-221-9088; RICOH GIII). J-E technical translation in many fields. 23 years in Japan, 15 in translation.

SHKOLNIK, Alexander. 4434 Fulton St., Apt. #2, San Francisco, CA 94121. Tel. (415) 3870290. Member A.T.A. and N.C.T.A. Japanese-English-Russian technical translations. Specialization in patent matters, mechanical engineering, metallurgy and chemistry. Worked as patent agent since 1963. Translating since 1959 from English to Russian, since 1965 from Japanese to Russian, and since 1970 into English from both.

SORDEAN, Jay, C.A. 2637 Regent St., #306, Berkeley, CA 94704. Tel.: (415) 841-9167. Languages: Japanese. Subjects: Medical, biological, patents, pharmaceuticals, chemistry, electronics, mechanical engineering, Oriental medicine, computers, legal, etc. 6 years translation & abstracting experience. Epson QX-10 word processor.

TALBOT, BRUCE G. Maisonette Daita #E, 5-13-17 Daita, Setagaya-ku, Tokyo 155. About 2 years experience in business, economics and other non-technical areas. Presently shifting to computers, electronics, patents and other technical areas. Wants to correspond with U.S.based translators.

UYAMA, Hiroshi. Director, Japan Technical Information Center, 706 Seventh Street, S.E., Washington, D.C. 20003. Tel.: (202)544-6314 (days), (703) 979-0054 (evenings). B.A., Chemistry, International Christian University in Tokyo, M.S. Inorganic Chemistry, Tokyo Institute of Technology. Since 1975 Japanese-English interpreting, translation, and consulting services in Washington, D.C. Science, technology, politics, social sciences.

* * *

EDITOR'S POSTSCRIPT

Japan Recruit Center reported the results of a survey it made of the most popular companies among university students seeking employment. 12,659 male university seniors planning to graduate in March, 1985 were surveyed. Among the humanities graduates, IBM Japan moved up from 19th place to 8th place. Among science graduates, 7 of the 10 most popular companies were electronics thanufacturers. The highest ranking company was NEC, followed by Hitachi, Ltd. This ranking has continued for the past 4 years. IBM Japan also moved up from 7th place to 4th place among the science graduates. (Nikkan kōgyō shimbun, Sept. 5, 1984) Another survey was conducted by Nikkei Computer of about 500 university seniors specializing in computer sciences. The results showed that IBM Japan is the company which was chosen in first place by the largest number of these students as the one where they would like to work. (Nihon keizai shimbun, Sept. 17, 1984) The rising popularity of IBM Japan among graduates specializing in computer sciences, and also among humanities and science graduates, would seem to disprove Dr. Umano's statement (see TJT, no. 19, p. 28) that superior-quality Japanese university graduates would not want to work for IBM Japan.

IBM recently launched a big sales compaign for the Japanese-made 5550 work station in China, according to an article in The Asian Wall Street Journal Weekly of Sept. 17, 1984. IBM executive Don Stanger, director of staff services in Peking, is quoted as saying that IBM designed new software "specifically for the Chinese-language model." "IBM is selling the absolute latest technology in China," he said. The IBM Chinese-language software uses a non-phonetic inputting system invented by Taiwan-born Samuel C. Tseng, a naturalized U.S. citizen. The CL 5550 work station uses a keyboard with 26 keys representing "the most common basic strokes in the Chinese script." After 18 months' research, Tseng contrived a way to group words under these basic strokes. "When a particular key is hit, the computer screen projects a number of similar characters, and the user selects the one required." (The system seems to be similar in its basic idea to the Transtech system used by the Research Library Group, as described in an article elsewhere in this issue.)

The sales efforts appear to be having considerable success. The Nihon keizai shimbun reports that the Chinese government appears to be about to place a large order with IBM Japan for its 5550 work stations. The number ordered is believed to exceed 1,000. If true, this will mean a considerable victory for IBM in its efforts to penetrate the Chinese market and will be a blow for Japanese manufacturers such as Hitachi and NEC who are vying for the Chinese computer market. NEC has been making vigorous efforts in China but nevertheless has been able to sell only a total of about 400 units so far. (Nihon keizai shimbun, Sept. 6, 1984)

Incidentally, the Chinese government is giving top priority, in opening its computer market, to foreign companies that are willing to transfer their technology to China. Japanese companies are lagging far behind their U.S. counterparts in facilitating the transfer of technology to China. IBM, Hewlett-Packard and other U.S. manufacturers are already taking a clear lead ahead of the Japanese companies. IBM is currently contemplating the establishment of an assembly plant for personal computers in China, and Honeywell, Inc. set up a technology center in Chongqing, Sichuan province. (The Japan Economic Journal, Sept. 18, 1984)

Dr. Umano Shuji, whose book Gijutsu bunmei no hosoku was reviewed in TJT no. 19, has just published his Japanese translation of The downwave, a 1983 book by Robert C. Beckman, an American economist currently working in London as an investment analyst. The translation was published by Tokuma Shoten on August 31 and contains an interesting introduction by Dr. Umano. Beckman argues that an economic downtrend is inevitable in the West and that Japan will be the world's leader during the 21st century.

The May 1984 issue of Design Graphics World has an article by Brian Shaw under the title "Japan's Automation Pioneer Tackles Records Management." The article descibes the work of a Japanese-speaking American management consultant, Thomas Wilds, who went to Japan to advise Okamura Corporation, an international supplier of total office environments, industrial machinery, and automation systems. Thomas Wilds is former president of the International Records Management Federation (1980) and of the American Translators Association (1971–73). He now heads his own consulting firm, Thomas Wilds Associates (P.O. Box 11120, Greenwich CT 06830).

The editor has been invited by Professor Edward E. Daub to give a talk entitled "Important Entrepreneurs — America's Freelance Technical Japanese Translators" at the College of Engineering, University of Wisconsin. The tentative date for the talk is Wednesday, November 28. A luncheon meeting has also been scheduled with the College of Engineering Library Committee. During his visit, the editor will propose the establishment of a center for technical Japanese translation at the University of Wisconsin (in line with the proposal in no. 16, p. 10–11). The editor also hopes to meet with TJT readers in Madison, Wisconsin and thereabouts.

Constructions with hodo (again). I still think that this type of construction is used far more frequently in Japanese, and that it is awkward or maybe impossible to translate into English. Do readers agree with my opinion? For example, I saw an advertisement in the Nihon keizai shimbun which had a picture of two charming little girls wearing tutus and ballet slippers. They are holding bouquets and beaming happily at the camera in what appears to be a ballet studio, as if they had just made their debuts as little ballerinas. The caption was:

小さな記憶ほど、鮮やかに残る。

Now how would that be translated? Why are the recollections "small"? The event — a debut as a ballerina — could hardly have been a very "small" one for the little girls or for their parents. Is it because the little girls are "small" and their youthful recollections are therefore regarded as somehow being "small"? Here are some possible translations:

All of these translations sound pretty awkward, don't they? Is this because constructions of this type, which come so naturally to the Japanese in everyday speech, are alien to the mentality of English speakers? Do readers have any insights?

Future of the Newsletter

I will announce next month whether or not the newsletter will be continued. On the one hand, I know that the newsletter has been fulfilling a real need and has been addressing issues of great importance. But on the other hand, I think that many or most of the questions which interest translators have already been addressed thoroughly enough in the newsletter, and we may now be approaching a sort of saturation point. Perhaps we have already said just about everything that needs to be said.

Another thing I wonder about is whether putting out a newsletter for J-E technical translators is really the thing for me to do. Being very well aware of the difficulties which my fellow-translators are facing every day, I am extremely tolerant of their little ways, and I cannot really say that I am disappointed by any lack of response from them. They have, as a matter of fact, responded to the best of their ability. In fact, given the formidable difficulties, it does seem almost miraculous that this newsletter has continued already for well over a year as a one-man venture with no external support. Nevertheless, I really cannot escape the depressing conclusion that, as a whole, J-E technical translators really do lack the sense of solidarity or cohesion which they would need if they ever wished to gain the necessary visibility as a group. I have often pointed out the need for concerted action to improve the professional status of J-E translators as a whole, but I am reluctantly beginning to conclude that there is some fatal flaw in this professional community which really makes it difficult for them to make a concerted effort on their own behalf. I think probably the basic reason for this is the fact that most technical translators have fallen into this career by accident, as it were. I know of almost no translator who has started out with the intention of adopting translating as a career. The resulting looseness, the lack of professional commitment is only too obvious. Some translators do not really care about being career translators, and if you ask them, they will tell you that they really would prefer to be doing something else. Since at heart they don't really want to be translators at all and have just landed in this profession accidentally because of a need to scrape together a livelihood temporarily, before moving on to something better, well, we can hardly be surprised if they have a low self-image. We can't really blame them for being a raggedy, frustrated lot of misfits or criticize them if they are not willing to invest large amounts of their money in dictionaries and word processors, to spend a lot of their time doing research in connection with their assignments, and generally to work hard at raising their professional standards. But, on the other hand, if they basically don't care, what do they need a newsletter for?

For some time now, the newsletter has been dealing with issues which go far beyond the narrowly-defined world of the technical translator. It is not really very surprising that the most enthusiastic response to the newsletter has come, not from the community of J-E technical translators, but from well-informed and intelligent non-translators who have professional or scholarly interests in the Japanese language and/or in Japanese scientific and technical information. If that is the case, perhaps the best thing for the editor to do would be to bid a reluctant adieu to the community of J-E technical translators and begin to address himself to non-translators who have a really informed interest in the type of information which the editor can supply. The editor's needs to communicate might be better served by discontinuing this newsletter and starting another one devoted primarily to broader issues connected with Japanese scientific and technical information in general, and only secondarily with the narrowly defined questions of linguistics which are of interest mainly to technical translators.

If you want to express yourself about this one way or the other, write me before the end of October, but please try to contribute something new to the discussion. I will announce my decision next month.

* * *

[Scanned Image No. 2]

Received at the Last Minute Department

Corporate Data Sciences of Santa Clara, California has announced an ultra-high resolution (1024 x 1024 pixels) IBM PC-compatible terminal for text processing, CAD and Chinese word processing. Called the Video Scroller Terminal (VST), it has a 17-inch screen and "uses an AM29116 16-bit bipolar microprocessor to relieve the IBM PC of all major i/o processing and screen writing chores." CDS provides two software packages: VTEXT, a text processing software allowing the user to work on an IBM PC keyboard, and CTEXT, a Chinese software package which includes a 7,310-character dictionary standardized by the People's Republic. It requires an additional bit pad and 50 megabytes of additional disk storage. The system retails for under $12,000 with an Epson LQ1500 (24-pin) dot matrix printer and under $18,000 with an electronic (laser) printer that will be available during the fourth quarter of this year. The system does not include the IBM PC XT.

Linda Anne Kazares, Western Sales Manager, told me that CDS is planfling to develop and introduce Japanese-language software at some time in the future and has promised to provide further information in the next issue.

The address is:

Corporate Data Sciences
2560 Mission College Blvd. #102
Santa Clara, CA 95050
Phone: (408)980-9747

****

Send me your tidbits, and I will incorporate them in the next issue. The current subscriptions are for SIX MONTHS (from June through November, 1984). If you paid only for the first year (ended in May 1984), your subscription has now expired. Don't forget to send in your $20 if you want to continue receiving the newsletter for the current six-month period. And also write if you want your name included in the Directory. Those who send in lists should provide the kanji for the Japanese words. I will type them in Japanese if necessary.

October 10, 1984

Donald L. Philippi
715 Tenth Ave.
San Francisco, CA 94118
(415) 752-7735

* * *