No. 14 — April 10, 1984

This is, probably, the only newsletter published anywhere in the world by and for our obscure community of Japanese technical translators. It is a completely grass-roots endeavor with no subsidies or support except that given by its readers. I hope that the newsletter will lead to winning greater recognition for the important role which we Japanese technical translators have to play, and that it will eventually evolve into a trailblazing publication which will shed light on the entire area of Japanese scientific and technical information, a strategically important field which is attracting increasing attention in the U.S. today.

Currently, I am interested in targeting the following audiences of readers: (1) Japanese technical translators or novice translators; (2) librarians, data base workers and others engaged in gathering, processing and disseminating Japanese technical and scientific information; (3) business people with a professional interest in such information; and (4) non-translators with a specialized interest in Japan and the Japanese language.

I plan to continue publishing the newsletter regularly as a one-man venture until June, 1984 (one year after the first issue was published), and then to reassess plans. Readers should send in their comments or suggestions about this. Currently I am thinking of continuing publication of the newsletter in its present format for periods of 6 months, and then reassessing plans after each 6-month period. Readers would pay for 6-month periods instead of a whole year. I doubt whether I will be able, physically, to continue doing so much work for an indefinite period, but perhaps breaking it up into 6-month periods would make it easier for me to escape from the editorship if health or financial reasons were to make that necessary.

It would be very nice if someone else were to volunteer to take over the editing and publishing work, or at least to collaborate in the work of editing and publishing future issues. One possibility would be for readers to volunteer to write one or two articles a month, with the editor supplying them with the raw material (Japanese-language press clippings, reference material, etc.) Volunteers should write me giving details of when they will be available and what they can do to help.

Readers who cannot write something for publication should at least send in a donation of money to help pay for the expenses. Suggested minimum amount of donation for a non-contributor is $20 for the year May 1983–May 1984. If the newsletter is published after June 1984 we will begin a new period, possibly a 6-month period, as mentioned above. Therefore, your current "subscriptions" are for the one year through May 1984.

Thanks to those who have sent in items for publication and/or money for the expenses, and dire warnings to those who send neither: your names will be removed from the mailing list. If there is an asterisk by your name on the envelope, that means that I am not sure whether you wish to continue to receive the newsletter. You must write to me indicating that you want to receive it, or your name will be dropped from the mailing list. Those who need specific back issues should write me (enclosing a check sufficient to cover the postage) and specify which issues they want. Those who want only the current issues should send in a donation commensurate with the issues they intend to receive. (Please, do not ask me to find articles on subjects of interest to you and send you xeroxed copies of them. I don't have time for that. If you want the back issues, send me $20 and you will receive a full file of them, plus the future issues until the end of the current year.)

The following two readers have kindly volunteered to reproduce the newsletter and distribute it to other readers in their countries:

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

Mr. F.R.D. Apps, 57 High Street, Ingatestone, Essex CM4 OAT, England

* * *


by Donald L. Philippi

Q: My guest for this interview is James Unger, Associate Professor at the Department of East Asian Languages and Literatures, University of Hawaii at Manoa in Honolulu. He took his BA at the University of Chicago and obtained his Ph.D. in linguistics from Yale. He first taught at the University of Canterbury in New Zealand. Then in 1977 he went to Hawaii, where he teaches courses in the Japanese language and history of the Japanese language and a seminar on artificial intelligence for linguists. When did you begin studying the Japanese language?

A: I started studying it just before I entered college in 1965. Evanston Township High School was offering a summer intensive course in Japanese and Chinese for high school students. I took the Japanese course. When I entered the University of Chicago, I was, I think, the first freshman to ever sign up for first year Japanese. I was regarded as an enfant terrible by some of the professors.

Q: Tell me how you first got started with computers.

A: When I was in New Zealand, there was a Burroughs 6700 mainframe with CANDE timesharing, and I taught myself how to program ALGOL so that I could write concordancing programs for Early Japanese linguistic materials. I transcribed the parts of the Kojiki written in Man'yōgana and input them and created a concordance to facilitate my analysis of the kō-otsu syllable distinctions. I also did some work on comparative Japanese-Korean analysis. So I was self-taught. Then I went to Hawaii.

At that time, the University of Hawaii was operating a connection to a computer network with headquarters at the University of Illinois in Champaign-Urbana. That system is called PLATO and is dedicated to educational purposes. It's an ongoing project operated by an organization called CERL (Computer-based Education Research Laboratory). PLATO runs on equipment of Control Data Corp. (CDC), which has its headquarters in Minneapolis. At that time, the chairman of East Asian Languages at Hawaii was Dr. Agnes Niyekawa. She was interested in finding out to what extent PLATO or any other computer system could be used in teaching Japanese as a foreign language. At Hawaii we teach more than 1,000 undergraduates every semester at Manoa Campus alone in Japanese language skill courses. We have a severe student-teacher ratio problem; we are living through a series of budget cuts; and we have no prospect for improving student-teacher ratios in sight. We have fundamental motivation problems with many students because of the demographics of Hawaii, and our students come from extremely heterogeneous backgrounds. These are the reasons why Dr. Niyekawa was so anxious to find out whether or not we could, in fact, do something with PLATO. I started off very skeptical, from what I knew of computers and of the mentality of College of Education type people and university administrators. I tended to believe that the claims being made for PLATO were mostly hype. I started using the system in the sunmer of 1977 and taught myself how to program it in about half a year. At that time I reported back to the Chairman. I told her that, in my opinion, it was technically possible to do kanji-kana majiri-bun on PLATO and that the supporting software for educational purposes in general on PLATO was excellent, much better than anything I had seen on other systems. As a matter of fact, at that time the problem was: What would happen if the University, instead of investing in a system like PLATO, would take the cheap way out and buy a gross of micros. In that case, it looked difficult or impossible to handle standard kanji-kana majiri-bun. So we established an informal departmental policy to go for PLATO, or else to ask that we have the money to invest in something more traditional such as library books or more positions.

Anyway, I continued to work on PLATO in my spare time till the end of 1979. At that point Control Data offered me a grant. They wanted to add Japanese orthography software to PLATO so that they could market PLATO in Japan, and they saw that my work in developing materials to teach Japanese as a second language, and my expertise as a linguist, as their stepping-stone to get that. During calendar years 1980, 1981 and 1982, we worked under grants from Control Data. Some graduate students worked with me, and we did it. We designed vector forms for all the JIS Level 1 kanji, kana and other punctuation marks, and I programmed an interface so that all of that was accessible through the pre-existing PLATO software. At the end of the project, some finishing touches were put on by Control Data Japan and the Advanced Development Group in Minnesota, and we transported all of the Japanese orthography software from network operations to off-line operations on what is called MicroPLATO. Now, Control Data Japan is marketing Micro-PLATO stations and using a software package called PCD1. It's a standard Micro-PLATO workstation, a terminal with a keyset and two floppy disk drives, plus an off-the-shelf Japanese tablet. You can do Japanese or English. I understand that they have two Japanese airlines as their initial clients.

The problem in Hawaii is that Control Data has this product that I developed for them, and we are still waiting for the State to grant funds to acquire their own PLATO system! That's the bitter irony of the situation.

Q: You said you were teaching a seminar in Artificial Intelligence. What books form the basis for your course, and what do you tell linguists about AI?

A: Anybody who is interested in problems of language and computers should read two books. The first one is called What Computers Can't Do (Harper & Row Colophon Books, 2nd edition 1979), by Hubert L. Dreyfus, a professor of philosophy at Berkeley. The other book is called Computer Power and Human Reason (W.H. Freeman & Co., 1976), by Joseph Weizenbaum, a professor of computer science at MIT. Weizenbaum is known as the developer of the program known as ELIZA. Weizenbaum thinks that computers can be made to simulate human intelligence but that they shouldn't be, while Dreyfus argues that, even in principle, machines cannot be made to simulate human intelligence in the general terms which are being claimed by workers in AI. The moral question becomes whether or not we should keep dumping money into AI research. In both books it's clear that what lies at the heart of the matter is the interrelationship between natural language and the nature of digital computers. I generally side with Dreyfus.

Q: How does what you said apply to translation?

A: Translation is a good example of a task that human beings with experience do very well, and that machines have a very hard time doing at all. But there's more to it than that. A human translator must have a lot of experience in terms of bilingual abilities, general education and practice in reading and writing skills but when the actual translation is done, it's almost an unconscious or seemingly effortless transformation of information. It's very much like playing chess in that respect. There are good computer programs for playing chess, but glaringly, the ones that play well do so because they use techniques that no human being could ever make use of, such as exhaustive searching of very large databases and the repeated application of so-called heuristic rules. The other side of the coin is that there are companies working on machine-aided translation from and into Japanese; that kind of work focuses one's attention on the limits of AI. To the extent that these machine-aided translation techniques are successful, it is because the sub-tasks are tightly circumscribed. In other words, the computer is effective in a well-defined situation precisely because the programmer could count on the boundary conditions of the problem to make his end of the task easier. But anybody who translates knows that the really difficult problems are exactly the unexpected, often seemingly trivial difficulties that are usually only solved in a rather haphazard or intuitive sort of way.

Q: What is your view of the Japanese Fifth Generation project, one of the aims of which is machine translation?

A: First of all, I think that any time you put a lot of money and a lot of bright people together, you will get results. The question is whether you get the results you anticipated or not. I think there are both overt and covert reasons behind the Fifth Generation project. The overt reasons, nobody could argue with. These are very obvious: things like trying to get a jump on the world computer market, trying to catch up in software, where there is a big lag between Japan and the U.S., and so forth. I suspect the main covert reason for this project is the realization that current technology for kanji input is hitting a dead end. If they don't develop an AI machine, and if, as would seem to be the case, the current efficiency of kanji input cannot be improved upon with current technology, then either Japan would be at a permanent disadvantage with respect to the U.S. and Europe in computer applications, or else they would have to modify their national educational policies. Specifically, they would have to admit the use of rōmaji as a significant, if limited, segment of Japanese writing, and I don't think they want to do that.

Q: How did you reach this conclusion?

A: First of all, take a look at the jōyō kanji documents. They show a definite swing to the "right," in my opinion, a definitely more conservative outlook on the part of the Japanese government officials responsible for these matters. Just compare them with the earlier tōyō kanji documents. For example, the tōyō kanji reform called for strict observance of the list. When it was implemented, there was great reluctance to expand the number of kanji or the number of permissible readings, whereas in the jōyō kanji documents, the tone is just the opposite. It permits people to use any gaiji they want to. Rules about okurigana usage have been relaxed somewhat. Use of non-standard readings is now tolerated to a greater extent than it was before. This is independent evidence that there is a conservative tendency among those responsible in Japan for setting these policies.

(There is a good book by Toodoo Akiyasu called Kanji no kako to mirai (Iwanami Shinsho). The last section describes his indignation and frustration with the Committee that decided on the jōyō kanji document.)

The second point is that voice input is high on the priorities of the Fifth Generation project. Why do you suppose that is so? It seems clear to me that this is a response to the problems of Chinese character input. It reflects a mistaken belief that by eliminating the typewriter keyset one can somehow make kanji input simple, accurate and rapid.

Q: It seems to me that this matter was not dealt with adequately by Feigenbaum and McCorduck in their book on the Fifth Generation.

A: You're right. In fact, if you read only their book, you would hardly know that there is an input problem in Japan. I think there is a total of maybe two or three sentences about it in the entire book, and they are vague and uninformative.

I'll tell you an interesting story. I was in Tokyo in October, 1983 at a conference called ICTP (International Conference on Text Processing with a Large Character Set). Most of the people there were computer engineers. I was one of the few linguists. I asked whether or not the success of the Fifth Generation project was a necessary condition for a really good solution of the kanji input problem. The majority of the Japanese engineers said that in their opinion the input problem was already satisfactorily handled by word processors, and that therefore the question of the Fifth Generation project was irrelevant. A significant minority of the engineers, however, felt that indeed an AI machine was a necessary part of any comprehensive solution of the kanji input problem. I asked Dr. Moto-oka Toru, who is actually head of the Fifth Generation project, for his opinion. His response was that he didn't think the kanji problem would be completely solved even after an AI machine had been built. The significant thing is that some of the same people who insisted that an AI machine was irrelevant to the question of input also belittled the Fifth Generation's goal of voice input. I think this shows that beneath the surface there is a recognition of the importance of improving the input situation.

Q: As a linguist, what do you think about voice input?

A: It is clear to me, as a linguist, that voice input is a chimera. Eliminating the typewriter keyset will not significantly alter the difficulties of input in any way. In fact, it only makes it more difficult. But it is very significant that this appears to be something that the Japanese are either very enthusiastic about or cynically derisive of.

Q: You seem to be very critical of Feigenbaum. Why?

A: Feigenbaum and all of these AI people - what Weizenbaum so eloquently calls the "artificial intelligentsia" - have been working on AI for years, making small successes on well-defined problems, and consistently failing to generalize their results in the manner which they have always claimed they could do if only they had more time, and presumably more research money. Once you do some reading in AI, you realize that Feigenbaum is green with envy of the Japanese. Their government has just given them what he has been dreaming of for years. Feigenbaum has a very clear and dogmatic vision of what artificial intelligence is, what it can do, its significance for world history, and so forth. I think he is totally uncritical about it. His book is basically a California blend of "Yellow Peril" and Reagonomics.

Q: What translators want to know is, not necessarily the ulterior motives of Dr. Feigenbaum, but whether machine translation and artificial intelligence will, 15 or 20 years from now, have an appreciable effect on their livelihood. For instance, will computers put them out of work in the same way that robots have put some auto plant workers out of a job?

A: The answer basically is No. What will happen is this: Various systems will become commercially available at reasonable prices for providing automatic dictionary look-ups and a variety of other computer-assisted services to speed up through-put for translators. But as far as automatic translation is concerned, I do not believe it will ever get to the point where anything less than a skilled translator will be able to operate the machine and check and document the translation. The kinds of errors that the machine makes are due to fundamental incompatibilities between the way the machine works and the way natural language works. Think about it from the programmer's viewpoint. The programmer is dealing in a situation where he can't be sure of 100% accuracy. From the programmer's point of view, the problem is not just reducing the number of errors, but rather to make sure that any error that goes through is detectable, which is a much more difficult proposition. To put it another way, the big saving for the businessman is to get rid of the translator, or to replace him with somebody much less skilled. Unless you had a machine that was so good that you could turn the editing task over to a monolingual, then there is no big saving from the viewpoint of the businessman. How can a monolingual person tell whether or not the computer has made a mistake? There are (a) mistakes a computer will make but not recognize or mark. There are (b) correct translations which the computer will think are erroneous when in fact they are not. Then there are (c) cases where, presumably, both the computer and the editor think that the translation is correct. The programmer wants to eliminate type (a) errors, but on the other hand he wants to minimize the number of type (b) errors. He wants to make sure that if an error gets through, it will be obvious even to a monolingual speaker of the target language. On the other hand, the programmer doesn't want to create unnecessary extra work for the monolingual editor of a text.

Q: To change the subject, we have been having a debate about Millerism, or more specifically Miller's perception that the Japanese cultivate a myth that their language is more or less unlearnable by foreigners.

A: There's more to Miller's book than that. First of all, in the opening part Miller stresses that his "myth" is characteristic of the postwar period. He specifically says it is the last surviving remnant of nationalistic ideology and that the myth originated out of the devastation and humiliation of defeat, the pressure of the Occupation, etc. I would submit that anyone who has lived in postwar Japan knows the factual basis for Miller's statements and that he is correct in saying that this is something characteristic of the post-1945 period. Before that, the Japanese expected other Asians to learn Japanese, as is shown by their policies in occupied territories from Meiji onwards. 1945 is indeed a turning point, but Miller's claim that there is a cultural myth, a Jungian collective unconscious, is a needless generalization. A much better explanation of these phenomena, in my view, is simply that the Occupation abolished the elite higher schools. Until now there has been almost no research on the level of literacy achieved by average Japanese before 1945. We know that there was compulsory education from Meiji, and in terms of minimum literacy we know that a large percentage of the Japanese population was literate; but I strongly suspect that before 1945 the average Japanese (not the handful of elite students) probably did not have to learn kanji on anything like the level that a Japanese high school student of today must. If I am right, this means that the generation of students who went through the Japanese school system right after the war represent the first time in Japanese history when the vast majority of young adults had been subjected to the rigorous demands of Chinese writing. This would certainly explain many of the things Miller talks about without resort to some kind of psychosocial explanation of the kind that Miller thinks is necessary.

Miller compares the U.S. and Japan quite a bit, and does not discuss the situation in other countries. Is the situation in Japan remarkably different from, say, the situation in Finland or Czechoslovakia? I strongly suspect that these attitudes are not as unique to Japan as Miller seems to imply they are. Miller criticizes English-language education in Japan, and to some extent his criticisms are obviously based on facts; but he ignores some important things. For example, in Japan you can find a person who speaks some English in almost every small town, but you can certainly not expect to find a person who speaks some Japanese in a small town in the U.S. I don't think that Americans are particularly prejudiced against foreigners, although they are often only dimly aware of the importance of foreign languages in a shrinking world. Miller is saying that the situation in Japan is somehow worse than that. I see no basis for it.

Q: Reading his book, I felt that it may have been based on some deep-felt but purposely obscured sense of personal rancor for some slights he had experienced in the past.

A: No, I think it's a sort of love-hate thing. I think Miller really loves Japan. He really wants to be Japanese or at least to be regarded as a dai-sensei in Japan, and it's not happening. He feels that he is really doing something important in scholarship but it's not being appreciated. What kills him is that he is not getting back what he feels is his due, and he is letting out his frustration by lashing out. He is trying to rationalize in public his own personal disappointment and frustration.

I think there is a connection between Miller and the Fifth Generation.

Q: I beg your pardon?

A: Miller's view distorts the problems that do exist for using the Japanese language in modern society, whereas Feigenbaum just ignores the whole set of problems. What's needed is some kind of rational middle ground where one takes due cognizance of the cultural and linguistic problems that the Japanese are now facing because of the advent of computers, but is neither foolishly optimistic about the solution of these problems nor is cynically pessimistic about the chance of their ever being solved.

I am going to Japan to do a book in which I will try to stress that as my basic theme. I just received a Japan Foundation grant for calendar year 1985. The project proposed deals with language, computers, and contemporary Japanese society.

Q: What will be the thrust of your research in Japan?

A: I already mentioned a number of my hypotheses. Through discussions with people actually doing the AI development work, interviews with appropriate government personnel and visits to laboratories, I must either verify or refute my hypotheses. One sub-category is the issue of literacy in pre-1945 Japan. A second important area is the input problem itself. Very exciting work is being done on that by a Tokyo University professor called Yamada Hisao. I will take a look at the state of projects like machine-aided translation, the nature of office automation — for instance, complaints from workers using the products. I will try to be on the scene, talk to the people involved, and find out what their orientation is towards the AI problem and the kanji input problem.

Q: What can you say about the problems we technical translators in the U.S. face? For example, how are novice technical translators to be trained?

A: I teach Japanese as a foreign language, and I find that teaching languages for special purposes now is a trendy topic among certain educators, but it usually means teaching something other than a well-rounded language program. I would hope that people who want to see the number of technical translators increase would not jump on that particular bandwagon.

Q: Since we are on that subject, what is your impression of the Sheffield University course taught by J. Jelinek?

A: All I know about it is what I read in your newsletter. It sounds to me as if the idea is to train people who know nothing about the Japanese language to start with, to translate Japanese. Is that a good thing to do to someone who wants to build a long-term career? First of all, emphasis on vocationally-oriented training programs leaves me a little cold. I am unhappy with the idea that somehow one can pick and choose specific items and specific skills to teach somebody on the theory that they are going to use the language in only one specific way. Even if they are going to do that, that is a specialization that can and should be looked at after a solid foundation has been laid in acquiring the second language. It is unclear, from what I read in the newsletter, at whom this course is aimed. Are they trying to turn out a bunch of specialized clerks, or are they giving an auxiliary skill to already securely established professional people to enable them to cope with Japanese language materials that would otherwise be unusable to them? If the idea is to create a group of professionals who really can't do anything else but translate from Japanese into English, I find it a little scary. On the other hand, if this is a special service course aimed at people who are already secure in their professions and simply need the ability to translate from Japanese to fill the gaps in their professional qualifications, then I can see considerable merit in it. But I wouldn't want to take undergraduate students and run them through that sort of course, give them a diploma at the end, and send them out into the world to earn a living. The methodology struck me as a little bizarre. Their concept of how to train people to read and write and translate is 100-degrees diametrically opposed to the teaching philosophy set forth in Jorden's textbook "Reading Japanese." Even more important, I noticed at the end of the article that the method would be taught through computer-assisted instruction. In fact, if I were asked to program something like that, I would want to move the whole thing onto the computer and get rid of the books altogether.

Q: What do you think are some of the most useful ways in which computers can be used to assist translators?

A: The real value of computers lies in their networking capabilities. Technical translators ought to have a central mainframe computer that could provide them with on-line dictionaries and other types of real-time resources. PLATO is an example of such a system. The main reason that stand-alone micros will continue to be a strong part of the market, in my view, will be telephone deregulation in the U.S., which I think will keep communication costs sufficiently high to discourage network use except by large universities and businesses.

In Japanese technical translation, what is probably needed is a professional organization that publicizes the personnel available, makes some attempt to standardize rates, is capable of lining up grants to support computer networks and other types of profession-wide assistance. I was just at the American Oriental Society's 194th annual meeting in Seattle, and I was a participant on a panel dealing with the use of computers in Oriental studies. The people were mostly archeologists, philologists, and other scholars, but in some ways the problems they are confronting are the same as the ones that professional translators are confronting. Without proper organization, there are a lot of wheels being invented every day. If I compile a file of dictionary terms, how do you know what I have done and get access to it? What is a reasonable charge in a case like that? In that respect, the problems are similar.

Q: What could we Japanese technical translators now do to obtain greater visibility?

A: To gain visibility, you would have to do what other professionals do: organize a society, publish a professional newsletter, gain some influence with people who are deciding government policy, handing out contracts, etc. But the key is that you must show that you really represent the people who are competent and are really doing the work. You must really be able to deliver the goods, so to speak. My experience in teaching Japanese as a foreign language is that there are a lot of people who are not really very good at teaching it but that is all they can do professionally. This includes, by the way, many native speakers of Japanese. It goes without saying that translations, generally speaking, should be done by a native speaker of the target language. In the same way, elementary language teaching should be done by a speaker of the student's native language who has him or herself learned the foreign language as a foreign language. (Advanced language students, of course, can benefit from native speaker teachers because, at that level, interference from their own native language becomes minimal.) There is an analogy there, I think. Just having a professional society alone isn't enough; there has to be some kind of quality control. Generally, that is difficult to achieve because of the personal and political issues which become involved, and that is why it is difficult to organize a society of this type. There are always lots of mediocre people who want to be protected by the aegis of such a professional society, and that makes it very difficult to obtain credibility in the minds of those who would give you the visibility you want. The organization must represent the competent people who are actually doing the job, and you have to make sure that you are delivering the goods. That's tricky. At some point the issue of quality control comes up.

* * *


Congressional Testimony on Japanese Information

According to the Nihon keizai shimbun of March 7, 1984, a U.S. House of Representatives Subcommittee on Science, Research and Technology met on March 6 and heard testimony about the utilization of Japanese scientific and technical information. At the meeting, David B. Shonyo, Director of the National Technical Information Service, Department of Commerce, testified to the effect that "Japanese scientific and technical information is fairly open as far as its institutional aspects are concerned, and there are few limitations on its utilization by Americans. However, in practical utilization the Japanese language poses an important obstacle."

According to Shonyo, during the past 3½ years, the demand in the U.S. for foreign technical documents has increased fivefold, and the demand for Japanese technical information has been especially high. "There has been a demand for about five copies of foreign documents other than the Japanese, while the demand for copies of Japanese documents has averaged 12, and in some cases as many as 50 copies have been requested."

Other witnesses appearing at the committee included a Mr. Landow (phonetic), president of a company called Engineering Information (New York), and representatives of the academic world. While indicating that there is a high interest in Japanese high technology, they all expressed concern about the fact that useful technical information cannot be utilized on account of the obstacle posed by the Japanese language.

Bruce D. Merrifield, assistant secretary of commerce for productivity, technology and innovation, testified that "progress of Japan's scientific and technology has been remarkable, and in some fields the U.S. cannot afford to reject Japan's superior knowhow out of a sense of pride. Ten years ago, the U.S. was responsible for creating more than 70% of the world's new technologies, but within the next ten years this will decline to about 30%. This is because of the advances made by Japan." (Quotations translated from Japanese-language news dispatch filed in Washington by Jiji.)

Ray Connolly's Article in Electronics

A commentator writing in the March 22, 1984 issue of Electronics has some harsh words to say about collection of Japanese scientific and technical information in the U.S. Ray Connolly, writing under the title "Washington commentary," reports that the Congress' Office of Technology Assessment (OTA) will in a few weeks issue a new report entitled "Development and Diffusion of Commercial Technologies: Should the Federal Government Redefine Its Role?" In the report, he says, the OTA states that "the Government must work harder to make more Japanese science and engineering data available to the U.S. community." Except for the AEA's establishment of a Tokyo office to gather Japanese information (see our newsletter, no. 13, p. 14), says Connolly, "U.S. corporate and Government efforts to collect, abstract, and translate Japanese technical papers and patent filings are pathetic at best."

A previous report, published by the OTA last November, was entitled "International Competitiveness in Electronics." That paper discussed Japan, Western Europe and some other countries. John Alic, policy analyst at OTA, is quoted as saying that "too many Americans still look for simple explanations of Japan's successes: cheap labor, cheap capital, government subsidies, and favorable exchange rates." Alic's opinion about the Japanese is that they are excellent technologists, highly creative, and getting better fast. He continues: "It has always seemed to me a sign of how widespread technological illiteracy has become within the U.S. that Japanese management can be in vogue while the excellence of Japanese engineering goes unrecognized."

Connolly says that the "U.S. must quickly do more to disseminate Japanese technical documentation, just as Japan does with U.S. papers." He quotes Clark E. Johnson, Jr., founder and chairman of Vertimag Systems Corp., to the effect that fewer and fewer English translations of Japanese research on magnetics technology are available even as Japan improves its capability. Japan is publishing fewer of its magnetics papers in English, and Johnson says that the U.S. is "in imminent danger of losing both our technological leadership and a large portion of the computer peripheral disk-drive business to the Japanese."

Connolly's recommendations are that a superpower which wishes to obtain access to engineering technology that is equal or superior in quality to its own must "recognize and take advantage of the adage that it is people, not papers, that provide the best access to new technology." Thus, some American companies have concluded agreements with Japanese manufacturers to buy Japanese technology. He might also have mentioned the fact that some U.S. companies have set up R&D and manufacturing operations in Japan and are buying into Japanese firms in order to introduce their products into the Japanese market. Examples are Merck, which recently acquired 50.5% of Banyu Pharmaceutical Co., and BOC Group, which purchased 30% of Osaka Oxygen Industries. (On this point see the article "Making R&D Dollars Work Harder" by William P. Dunk and Courtenay W. Beirihorn, in High Technology, April 1984, pp. 67–69.) Efforts are also being made to publish at least one journal of translations from Japanese. The Magnetics Society of the Institute for Electrical and Electronics Engineers plans to begin in 1985 a journal of translated Japanese papers. The journal will have some 1,500 pages annually. This development should be of interest to American J-E translators. Connolly mentions also the Massachusetts Institute of Technology's MIT-Japan Science and Technology Program and projects at Northwestern and North Carolina State universities.

Connolly concludes, however, that "better Federal support and coordination is needed," possibly by the White House Office of Science and Technology Policy run by the Science Advisor to the President. OTA's Alic suggests that Federal aid should be given to EE students who agree to become proficient in the Japanese language. Connolly suggests a national indexing and abstracting service for Japanese technical and economic data for nation-wide distribution to American schools and libraries.

Wil Lepkowski's Article in Chemical and Engineering News

Another article about this subject, entitled "Technical Data Policy on Japan Urged" (by Wil Lepkowski) appeared in the March 19, 1984 issue of Chemical and Engineering News. The article begins:

The people who follow worldwide scientific and technological trends are now insisting with a single chorus that Japan is winning the global information war. It is winning because it has a well-organized global information strategy, and the U.S. has nothing to match it.

John A. Alic of OTA is quoted as saying in testimony during hearings "on the Japanese information threat" held by the House Science & Technology Committee: "A surprising number of U.S. corporations, even among the larger multinationals, have neither offices in Japan nor information-gathering networks there... Sometimes "they realize too late how good the competition has become."

Lepkowski says that, whatever the reasons may be why the U.S. has become such an information laggard, the key to catching up with Japan is a language and information policy towards Japan. Witnesses at the Congressional hearing agreed that the U.S. must develop a total information strategy towards Japan. One witness, Sam Coleman, an associate director of the North Carolina-Japan Center in Raleigh, N.C., said that training scientists and engineers in the Japanese language and culture is the only way the U.S. can tap into Japanese innovations in science, technology, and management.

James V. Seals, Jr. director of international programs for Chemical Abstracts Service, argued that the government should do much more to help the U.S. science information community establish a true national policy. "U.S. federal policy makers are not sufficiently aware of the policy issues concerning information transfer and information technology... This virtual absence of national policies in the U.S. is in sharp contrast to the situation in Japan and the industrial European nations." CAS is working actively to create a worldwide computer network of scientific and technical information arid to extend this network to Japan. "This would improve western access to some valuable Japanese scientific and technical databases as well as Japanese access to U.S. and European databases."

George Mu, a deputy director of the U.S. Department of Commerce, is quoted in the C&EN article as saying that his office and the National Science Foundation are co-operating in making a pilot study of Japanese advances in computer science, "mechanotronics" (that should be "mechatronics," no doubt), nonsilicon-based microelectronics, and biotechnology. Mu says that the Japanese have a competitive information advantage over the U.S. because vastly more of their technical personnel know both Japanese and English.

The article by Lepkowski concludes with a quote from Mu: "We hope... to be able to comment on the quality of R&D in Japan, assess how much of the significant developments in Japan are translated into English, and suggest ways and means to increase the utilization of this information to meet American competitive requirements."

* * *

It is encouraging, at least, that Congress is hearing testimony on this subject, and that articles about it are beginning to appear in periodicals like Business Week, Electronics and Chemical and Engineering News. The first step towards the solution is always to identify the problem clearly. Richard L Samuels, Program Director of the MIT-Japan Science and Technology Program, has sent us an "Open Letter" on this subject:

* * *


Richard J. Samuels
Associate Professor and Program Director
MIT-Japan Science and Technology Program

26 March 1984

I read with much interest Mr. Philippi's opening essay in Number 13 in which he reported on the growing concern in the United States about the insufficient and uncoordinated approaches to Japanese Scientific and Technical Information (JTSI). The Newsletter has been doing an excellent job of covering this issue, and given the interests of the readership, this is entirely appropriate.

Having been on leave in Japan since August 1983, I cannot claim to be able to provide intimate and up-to-the-minute coverage of the current American effort to come to grips with generations of ignoring Japanese science and technology. Inertia is a mighty force, and I suspect that some of the hyperbole one hears can be excused for its potential contribution to making something happen in Washington and in industry.

The recent hearings of the House Subcommittee on Science, Technology, and Research as well as the Business Week piece are cases in point. It is hard to be hyperbolic about the imbalance in the flow of scientific and technical information between the United States and Japan. Yet, as we discovered when we convened our Workshop on "Japanese Scientific and Technical Information in the United States" at MIT in January 1983 (and as Mr. Philippi pointed out) there is indeed an enormous volume of publicly available technical information in Japan. The problem is most decidedly not a Japanese one. It is an American one. Too few in government and industry have taken JSTI at all seriously. It is not a problem of Japan being closed, but one of America not being receptive (at least until very, very recently). It is abundantly clear, moreover, that tatewari gyōsei no heigai is as much an American as a Japanese affliction.

Yet, all things considered, things are moving along. That no one is arguing we should do less or continue to do little can be taken as a positive sign of progress. I encourage the readers of the Newsletter to take Mr. Philippi's call for a "politics of translation" seriously. His suggestion that a census be taken of qualified technical Japanese translators is an excellent one. I should think that the Newsletter's mailing list is a good place to begin. We at MIT will continue to strive to increase that population, through the education of our scientists and engineers in Japanese studies and through their placements in Japanese laboratories and research institutions. I would also like to encourage our readership to write directly to Congress, NTIS, and NSF with ideas and suggestions about how to improve the supply, care, and feeding of technical Japanese translators. One place to start would be with letters to Dr. John Holmfeld, the senior staffer of the House Subcommittee on Science, Technology, and Research who did so much to move the issue along on Capitol Hill.

Richard J. Samuels
Associate Professor and Program Director

Until July 1984:
1-29-3 Takaido Higashi
Tokyo 168, Japan

* * *


"Few Americans are able to read or speak Japanese."

"Useful technical information cannot be utilized on account of the obstacle posed by the Japanese language."

"We are not doing an adequate job of translating, much less monitoring."

These are some of the statements describing the current perception that there is an urgent need for translating and disseminating Japanese scientific and technical documents. The discussions until now have proposed that one way of overcoming the problem might be to train engineers and specialists to read the Japanese language. Machine-translation systems are also mentioned in passing, for reasons which we know well. Oddly, few of the discussions thus far have paid attention to the existence of a professional community already engaged in the translation of Japanese scientific and technical literature. It would seem to me far more practical to pay translators, who are already working in the area of Japanese technology, to translate the documents, rather than starting from scratch and training an army of American scientists and engineers fluent in reading Japanese documents in their field of specialization.

I feel strongly that the current upsurge of interest in Japanese scientific and technical information may offer us translators a unique opportunity to enhance our status and to gain recognition for the role which we are uniquely equipped to fulfill. But just when we are being presented with our Golden Opportunity, we are invisible and unknown! Why are we translators so obscure? In order to increase our visibility, a "politics of translating" is necessary. I mean by this that we Japanese technical translators as a professional group (forgetting for a moment the interpreters, translators of literary books, and other persons interested in the Japanese language) must stake out our own claim as a recognizable interest group. First of all, we must do something to increase our visibility.

I would like to ask readers to respond to the following questions:

  1. What are the reasons for our obscurity? Is it because we have no institutionalized professional framework? No advocates? Do we translators have a negative self-image, and if so why? Is translating something that we are doing for want of anything better or because we stumbling into this career by accident?
  2. What are the obstacles which will stand in our way if we attempt to win greater social recognition for our professional role? I can think of one possible obstacle: the Japanese-language teachers who refuse to entertain the idea of including technical Japanese in their curricula. This deprives their students of preparation for what might become a promising career opportunity for them. Their negative attitude towards science and technology ought to be combatted.
  3. What can our profession do now and in the future to bring greater awareness of the importance of our role as Japanese-English translators to business, government, education and the general public?
  4. Should we take the initiative into our own hands, such as by starting up our own J-E technical translators' organization, or should we rely on others (the ATA, the translation agencies, etc.) to enhance our status for us? Should we ally ourselves with other groups with similar interests? Or should we just wait here silently and hope that someone will eventually discover that we exist?
  5. How could we make a survey of the number of qualified Japanese technical translators? Unless we establish first how large our group is, we cannot argue very credibly that it is we who have the key to solving the problem.

* * *


As we mentioned in previous issues, a number of articles in the Japanese press have reported that the U.S. government is using various methods to limit the flow of information about advanced technologies from the U.S. to Japan and also to block exports of U.S. high-tech products to Japan. (See no. 10, p. 5; no. 13, p. 9)

Now the Mitsubishi Research Institute, Inc. (Mitsubishi Sōgō Kenkyūsho) has issued a survey report entitled "Seeking the scenario for America's industrial revitalization" [Beikoku sangyō saisei no shinario wo saguru] in which it claims that American industry is focusing on Japan as its "technorival" and has begun to show signs of mounting an "information blockade" [jōhō fūsa] of Japan in the most advanced areas of technology. (Nikkei sangyō shimbun, March 26, 1984)

The results of the survey show that most of the leading American companies have in the past laid primary emphasis on management, marketing and finances, but now they are making fundamental changes in their management strategies which are aimed at overtaking Japan. The new strategies involve re-examining manufacturing systems and laying more emphasis on technological development.

In the high-tech industries (computers, communications, optics, etc.) moves are being made to exclude Japanese from scientific meetings and to mount an "information blockade" against Japan.

Another article in the Nikkei sangyō shimbun of March 1, 1984 says that the situation has come to resemble an era of "U.S.-Japan technological blockades." It predicts the occurrence of a second and a third IBM incident.

The article says that at international scientific meetings such as the recent meeting of the ISSCC, an international semiconductor society, in San Francisco, the number of American papers presented has been dropping dramatically. For example, at the ISSCC meeting in San Francisco, a total of 115 scientific papers were delivered: 44 from Japan, 60 from the U.S., and 11 from Europe. Japanese participants were surprised at the small number of papers from the U.S., which in the past used to dominate the proceedings with more than 80% of the papers. At a robot meeting held in September, 1983 in Tokyo, only a few of the papers were delivered by Americans, and none of them contained anything new.

Japanese participants have been prevented from attending meetings of American scientific societies, such as one dealing with fiber-reinforced plastics (FRP).

The article goes on to say that this secretiveness ("Nihon ni dasanai, misenai, shirasenai") among American businesses, universities, and research institutes has become quite noticeable within the past year or two. In December, 1983 Japanese were amazed at a news report which stated that the U.S. government was considering measures which would (1) forbid Japanese participation in scientific groups dealing with high technology and (2) forbid giving technical papers to the Japanese.

On March 23, 1984 the Nihon keizai shimbun reported on its front page that the U.S. government had proposed to the COCOM member countries that all computers, including even personal computers, be included on the COCOM list of items which cannot be exported to Communist countries (excluding China). Exports of U.S. computers and other hightechnology items to Western countries would also require approval by the Defense Department. The newspaper reported that these measures, while greatly strengthening restrictions on outflow of U.S. technology to the Soviet Union, might also hinder technology transfers from the U.S. to other Western countries and would probably make it difficult for Japanese businesses to import high technology from the U.S. Many Western European countries, says the newspaper, are likely to argue that the proposed restrictions would make it impossible for them to obtain technologies they need from the U.S. and would, in fact, be restrictions aimed at the West, rather than at the East.

* * *


by Kevin Dwan

Since our company, Dwan Typography, has been specializing in designing, typesetting, and producing books for press from customer-supplied disks for three years, I thought describing the process in detail might interest translators who in the future may be using a similar process.

We produce computer manuals and technical material for most of our clients, though we also have done poetry books and scholarly books for such clients as University of California Press and North Point Press.

In the early days of the microcomputer software industry, most publishers used word processors to write their documentation. ("Documentation" is a term held over from electronic engineering terminology, and is used because the manuals contained, and often still do, formal specifications, technical detail, and legal notices on the limitations of hardware or software.)

Once a manual was keyboarded onto disk, early publishers often simply printed out manuals on their office printers, quick-printed a limited number of copies, and inserted them in three-ring binders or stapled them before mailing.

As the market grew in volume and sophistication, more and more publishers explored ways to typeset their material from the floppy disks they were accustomed to prepare. Not only do typeset manuals improve sales because of their increased attractiveness and legibility, but since typeset material commonly is more compressed than printer-output material, paper, printing, and mailing costs were lowered. In addition, typesetting manuals have become a matter of corporate prestige.

Manuals that are not typeset are now becoming the exception.

Typesetters attracted to the growing market began researching how to convert floppy disk material as quickly and cheaply as possible into typeset form. They were hampered because most conventional computerized typesetting equipment was not (and still is not) very efficient or flexible purely as computing equipment.

Not only do most front-end systems (the computers that drive sophisticated phototypesetting output units) commonly run only programs supplied by the manufacturer, and not those necessary to efficiently convert text files prepared on general-purpose micros into typeset form, but most typesetters, accustomed to viewing their equipment as black boxes, lacked the programming skill to write such programs.

In addition, typesetting equipment manufacturers widely adopted the practice of formatting the floppy disks for their equipment in such a way that commonly-available disks designed for personal and office computers could not be read by typesetting front-end systems. Manufacturers did this, among other reasons, to ensure that buyers of their equipment would be a captive market for supplies.

Fortunately, the demand for interfacing between common microcomputers and typesetting equipment has been so great that this philosophy has been changing, but the trend is still present in the typesetting trade.

As a result, most typesetters still are forced to accept text over the telephone, using what is termed "datacom," since the telephone represents a kind of Esperanto between computers of all types.

Apart from the problem of simply transferring text from a customer's floppy disks, keyboarded on a machine like a Kaypro, Osborne, or IBM-PC, typesetters still have to confront the programming problem: office printers and phototypesetting devices handle formatting commands in very different ways, and many typesetters find that even after the have the actual text on their screens, a very great deal of coding must be inserted manually to accomplish the desired typeset result.

A solution some typesetters use is to require their clients to enter what is called "pseudocode" at points in text in which special characters, display heads, tables, or other unusual material appears. Not only is this bothersome to non-typesetters, but errors in such code can result in increasing typesetting costs rather than reducing them.

Unhappy with the pseudocode approach and the various other limitations of dedicated equipment, our company decided in 1976 to follow an unusual strategy: we purchased general-purpose computers, rather than dedicated front-end equipment, and began a long programming effort on our own.

As a result, we now can accept and directly read floppy disks produced on over 50 common personal and office computers. While we can datacom, we prefer to avoid this approach and work from a disk copy, since for sizeable projects datacom ties up phone lines for considerable periods and because dirty phone lines can slow and corrupt transmission.

Our program allows us to automatically handle the printer control codes, if any are present in text files, and generally to accept — without any pseudocoding on the part of the client — text files produced with any text editor the client has chosen.

Certain design goals have thus been achieved. The client, using a machine and software of his choice, can write material, edit it, print it out for in-house or client review on an office printer, and then pass the disks to us for rapid and economical conversion to type.

The design of the piece to be printed is handled in two basic ways. Some clients give us quite explicit design in the form of typed specifications and sketches that have been prepared by their in-house designers, an advertising agency, or a free-lance book designer. In other cases, we develop a design ourselves, in consultation with the client, and submit sample pages until we have satisfied the client.


We are often asked by writers considering buying a word processing computer what our recommendations are; these are some guidelines.

Most of our own micros are IBM Personal Computers. This machine has been criticized for its conservative design, and for being more expensive than "IBM clones" — lookalike machines that do not have the magic three initials on them. In our heavy-use situation, the comfort of knowing repair parts and facilities are very widely available, and that the machine is extremely reliable, outweigh cost considerations. The price per machine, with the added equipment we require (two double-density disk drives, 512K RAM, clock/calendar and extra serial port) is about $4,000.

For those seeking a less expensive machine, we have heard good things about the Compaq and the Kaypro. I have a personal prejudice against the new Apple MacIntosh, and Apples in general present some data-transfer problems.

There is a computer folk-saying that goes, "Never buy anything with a low serial number." This refers to buying a machine that is less than a year old, or one produced by a company that may not be around in several years. Mail-order purchase is risky for the neophyte. We pay an independent dealer $200 per machine to handle purchase and set-up, despite our relative sophistication, and find that a reputable dealer is worth every penny, particularly when advice, rush supplies, and repair or maintenance is needed.

As a technician friend of mine once curtly reminded me as I was rhapsodizing about the reliability of a machine we had just purchased, "Everything breaks, eventually."

I have followed with interest references to the IBM 5550, but can contribute nothing about its use in the U.S. If the machine is disk-compatible with the IBM-PC, there are exciting possibilities pertaining to translation work. IBM machines, however, are never cheap.

In terms of word-processor software, I do not share the market's enthusiasm for WordStar, the most widely used text editor in this country, but it seems to have many enthusiastic users.

Over the years, we have become so impressed by a text editor named Omniwriter, which sells for $139, that we have become dealers. I cannot be completely objective about Omniwriter, but I write on a micro a good deal myself, and have found it simple and fast to use, with a minimum of commands to memorize.


We have done some research on translations for a client that wanted an American English manual produced, and was interested in French and Spanish versions as well. Perhaps outlining the process will illuminate how the process works in general, and how it may apply to Oriental language translation in the future.

(1) The client supplied us with the book text on floppy disks in files of about 20,000 characters each. We ran our usual program, and produced typeset galleys of type. Once the client had made his changes and approved these, we were ready to page out the book and send it to press.

(2) We ran a special program that compressed the typesetting code (which in its native state is very lengthy, and is intermixed with text in a way distracting to a non-typesetter) into short, easily-recognizable forms. A chapter head and the first line of a chapter then appeared in this form:

*Al*Chapter One: Preparing the Installation *A2*

*B*In this chapter, you will learn how to install...

(3) The translator, to whom we had given these disks with compressed code, then proceeded to work at his own microcomputer, ignoring and leaving in place the compressed code, and translating into French, erasing the English text as he went.

(4) We then ran another program that expanded the typesetting code around the French text, and again output the book.

The advantage of the process described above is that the typesetting coding only had to be done once, and our charges were dramatically lessened.

I think a microcomputer is an extremely useful tool for those doing J-E translations, particularly if the end product is to be typeset, but even if only a printer-output English translation is desired.

For some time after I had a computer available, I continued to write using an electric typewriter, and I sympathize with those who resist using a micro, if for no other reason than its cost.

The speed and flexibility by which a recalcitrant paragraph can be rewritten, however, eventually converted me, and I would now feel uncomfortable composing on a typewriter.

I would be happy to answer calls or letters from readers who would like to talk about how they might equip themselves with a word processor, or ask questions about the various aspects of book production I've touched on.

Our interest in Japan, and in J-E and E-J translation, is both professional and personal: more and more Japanese firms will be interested in the professional preparation of material for the American market, and it is useful for us to locate reliable translators and to advertise our company as a supplier of high-quality design, typesetting and printing brokering. For ourselves, Japanese culture has the fascination that must have initially attracted most of Japanese Technical Translation's readership to the field, and we look forward to drawing closer to that culture through our book work.

Kevin Dwan
Dwan Typography
427 Spring St.
Nevada City, CA 95959

* * *


As Dan Kanagy pointed out in his review of Marvin J. Wolf's The Japanese Conspiracy, one thing missing in that book was a single word of criticism of the United States. I found the book to be a very eloquent and well-researched criticism of Japanese business practices. As such, it is a valuable contribution to the growing literature of critical works about Japan, and it is well known that this point of view is very widespread in the U.S. However, I did not agree with the fundamental thesis of the book: that America's current ills are attributable to a "conspiracy" of a country which has embarked on a course of "economic totalitarianism." Other explanations would be much more convincing to me than a simple black-and-white picture of "conspirators" and their victims.

On the other hand, reading the book did not evoke in me any particular antipathy towards the author. I rather enjoyed familiarizing myself with the author's viewpoint, even though I did not agree with his basic thesis, and could hardly disagree with the parts which were obviously based on hard facts. Surely an author pointing out what he feels to be unfair economic competition deserves an attentive hearing.

The only other review of the book I have seen was a rather strident rebuttal by Yoshi Tsurumi in the Japan Economic Journal of March 20, 1984. Tsurumi is a professor of Marketing and International Business at Baruch College, the City University of New York, and president of the Pacific Basin Center Foundation in New York. He has written eight books, including one called The Japanese Are Coming, and frequently gives public lectures and appears on American television as a Japanese apologist; one gets the impression that he is the closest thing to a Japanese propagandist that one could find outside of the confines of MITI or JETRO. Recently Tsurumi appeared on a television talkshow with Marvin J. Wolf. The confrontation must have been lively, judging from the excerpts from it which Tsurumi publishes in his JEJ review.

Tsurumi begins his review by saying that several years ago, "when very questionable books like Theory Z began to tout euphoric pictures of all omnipotent miracle worker-corporations of Japan [sic] ," he (Tsurumi) began to warn that soon the pendulum would swing to the other extreme of "castigating Japan as the source of all the evils of America." This is, alas, what has happened. The latest fad, he argues, is a rash of quickie books which follow the familiar "big lie" tactics perfected by the Nazis with their "Jewish Conspiracy" theory. The Japanese conspiracy theorists, according to Tsurumi, base their arguments on their supposed expert knowledge of Japanese society.

Marvin J. Wolf is, according to Tsurumi, one of these half-baked Japan scholars. "He claimed that his 16 visits to Japan during the last 20 years eminently qualified him to write the book and uncover the Japanese conspiracy. Naturally, he does not understand the Japanese language. Nor is he bothered by the fact that he has not read or checked even a single credible reference on Japan available in English." Wolf's pet theory was already firmly established before he began his "research."

Further, Tsurumi claims that he (Tsurumi) has uncovered "many disturbing evidences" in Washington and New York linking prominent lawyers, — lobbyists and American executives to the financing and dissemination of Wolf's "terrible" book, "written by a hithertofore unknown freelance journalist like Marvin J. Wolf."

Tsurumi claims that the book turns out to be, on the contrary, the product of an "American conspiracy" which is "aided and abetted by businesses, politicians, and government officials blaming Japan, instead of their own failure, for the American industrial decline." That is a rather extreme statement coming from a university professor, but Tsurumi seems to mean it seriously. He states: "Mr. Marvin Wolf is a mere front man propped up by more insidious forces in the U.S. These forces are not only against Japanese imports, but also against Japanese investments and manufacturing operations in the U.S. Wolf and his sponsors are now running around the U.S. to stir up American grass-roots' opinions against Japanese firms."

Finally, Tsurumi concludes that fair-minded Americans must rise up and expose the Japanese conspiracy theory before both the U.S. and Japan, and the entire free world, suffer "irreparable damages." The last sentence of his review is phrased in terms of cold-war rhetoric: "The newly found ally of the Soviet Union is indeed the public relations machinery in the U.S. that is promoting the Japanese conspiracy theory."

Naturally, it is not inconceivable that Mr. Wolf may have been aided or financed by American business leaders or officials who share his views, but that would not be surprising or reprehensible. In fact, the value of his book consists precisely in the fact that it gives coherent expression to a viewpoint which is widespread among many influential Americans, including some Presidential candidates. Tsurumi is unfair when he accuses Wolf of not having read "even a single credible reference on Japan available in English." The book is well-documented, although I do not recall whether the bibliography contained many references in Japanese. Surely there is no objection to a person who does not know Japanese writing a book about Japan. Readers of the book can do some checking-up on their own if they want to track down additional facts; the reason why we read controversial books by authors such as Wolf or Miller is because we want to familiarize ourselves with a viewpoint, compare it with our own, and perhaps eventually revise our own ideas. Tsurumi is not making a very credible argument when he tries to turn the tables on the conspiracy theorists and accuses the "American public relations machinery" (presumably he is referring to Wolf's publisher) of engaging in a conspiracy of its own, founded on Nazi "big lie" tactics in the service of the Soviet Union. That accusation amounts to degrading the discussion to gutter-level invective. Hysteria on either side will not advance rational discourse.

There are definite reasons why Japanese and Americans act and think in certain ways under the concrete historical circumstances, and these underlying reasons are what really ought to occupy the minds of students of economic history. I do not believe that the actions of nations can be explained simplistically in terms of "conspiracies," and for that reason I cannot agree with the basic thesis of Wolf's book. He is trying to blame the Japanese for doing things which may well have been inevitable for them. Accusations of "unfairness" are predicated on commonly agreed notions of what is "fair," and obviously in cases of international economic friction such agreement is lacking. On the other hand, I think that the book does deserve to be read because it gives a comprehensible account of the perception on the part of some Americans that Japanese behavior is a conspiracy, and because we need to understand why that appears to them to be true. We also ought to make an effort to understand how things appear from the point of view of Japanese business and government leaders. The same things can look remarkably different to people with different geopolitical standpoints. Even more important, I think we ought to try to rise above the passionate accusations of partisan propagandists on both sides and attain a truly objective and non-judgmental view of the historical facts. I guess what I would like most of all to read would be a truly dispassionate, objective, scientific history of U.S.-Japanese economic relations.

* * *


Fred Uleman of Shibuya-ku, Tokyo, comments about Miller's contention that Japanese dislike Japanese-speaking foreigners:

I would like to comment on... the contention that Japanese do not like non-Japanese who speak the language too well. I am very uncomfortable with this assertion, and find it impossible to prove one way or the other. Of course, I can argue from experience that I do not notice Japanese disliking for me, but this could just as well be because (1) I am too dense to notice the dislike that is there or (2) my Japanese is still not up to dislike-provoking fluency. At the same time, I have no assurance that people who are disliked are disliked because of their language ability. There are, after all, all kinds of ways to be obnoxious (including the famous chip-on-shoulder syndrome), and it would be all too human to blame ostracism on the prejudices of others. However, even in the absence of empirical proof, logic dictates that Japanese would not dislike non-Japanese who have mastered the language, since mastery of the language would also imply mastery of the cultural mores and other language-associated traits which would make one acceptable to Japanese society. (Speaking American thought patterns in Japanese words is no more good Japanese than grammatically correct Japlish is good English.) So logically, if Miller and his ilk are really that good at the language, it is not the language that is putting other people off.

Miller's contention is that the Japanese have a widespread myth which holds (unscientifically, of course) that race and language are identical. "Japanese society," argues Miller, "is being neither cruel nor illogical when it chooses to make jokes about the odd foreigners who go to the trouble of learning to speak and understand the Japanese language. It is simply exercising its cultural prerogative; it is making an arbitrary choice about what it will and will not consider to be conduct worthy of approbation." (Japan's modern myth, p. 162–163) Miller mentions the case of a semi-legendary Reuter correspondent named Cox who was arrested by the Japanese secret police in 1939. Cox apparently understood and spoke Japanese. During his interrogation by the police, Cox had the bad sense to answer questions in Japanese, which so enraged his interrogators that they threw him out the window to his death. Comments Miller: "Everything that one learns and experiences about the contemporary Japanese response to the employment of their language by foreigners inclines one to believe that the Cox legend is no legend, only a plain and quite painfully accurate account of what actually happened." (ibid., p. 164)

Meredith Hazelrigg of Sayama-shi, Saitaina-ken, writes:

With Sato, I regret Miller's "laziness" in which he seems to become guilty of what he condemns in others. Yet, Sato's off-hand dismissal of Miller's "excuse" is hardly adequate. Miller is referring to popular, not specialist, interest ...i.e., the popular concept of Nihonteki, not philology... sources include everything from the discussion of language during television quiz shows to notes in English texts. Yet, even counting column inches in the popular press, computing viewer exposure time, etc. would remain virtually meaningless as raw percentage scores; qualitative analysis and cross-cultural comparison would be the only means to give objective substance to disproof of Miller's observation ....

Nevertheless, "the lady doth protest too much" for us not to suspect that Miller is probing at some sensitive truth. While Saint-Jacques' criticisms are equally valid on the same grounds as Sato's, I cannot conclude that there is nothing wrong with "linguistic mysticism." Such attitudes may lend themselves to nationalistic movements from Moldavia to Israel and the Punjab, but they do little to encourage cross-cultural understanding while doing much to foster chauvinism and cultural conflict... does Saint-Jacques approve of de Gaulle's anti-Anglicism?

As for his survey, it hardly confutes Miller's "inverse rule" shared by Seward among others. The greater competency achieved by most foreigners, the more they tend to challenge Japanese culturally preferred evasiveness, indirection or "illogical" (actually logic based on quite different — and undemocratic — assumptions) approach to a common problem. Further, his survey seems consistent with Miller's discussion of "Japanese for foreigners" or the concept of "good" Japanese. (Without proper research, I can only venture the hunch that even in superficial transactions, body language is differently interpreted when that body is 'known' to contain unpredictable understanding on the one hand or contrary to desired impressions on the other.)

Has Saint-Jacques dealt with the popularly discussed "gaijin complex?" Certainly the frank expression (or assumption) of democratic values is sure to cause irritation if unescapable or misunderstood... misunderstood on cultural rather than linguistic grounds. Precisely, how were Saint-Jacques' questions presented? What was their rationale?

I don't know how far to go with this discussion. I don't need to convince anyone who can see my point and proof for those who don't requires original research and elaborate theorizing. I'll simply conclude with these observations: Although I have more than mere quibbles with other points in Miller's analysis (he seems to see no important relation between literacy and linguistic competency, no possible shortcuts or remedial approaches to problems like pronunciation and few practical applications) I find it refreshing for someone to get out of the ivory tower and attempt to get at real problems in language; I find almost 90% consistency with my experience (which is disquieting since Miller does indeed "declaim," even overgeneralize... pandering to my prejudices?). At any rate, if his work is to appear in Japanese, he probably will have to so render it himself.

While Miller goes down with a sip of sake (or wine — use salt to your own taste) the xenophobic issue is very real. Otherwise, why should people whose native tongue is Japanese, whose birthplace is Japan and who have touched no other soil, be fingerprinted at a given age and required to possess alien registration certificates simply because their Korean ancestors (here by "invitation") are more recently arrived than, say, the Prime Minister's?

1984 is very much more realized in Japan through the laboriously manual but universal registry system than can be by, say, any computer hardware used by the FBI (or KBG). It must be produced for any important life decision: entering school, employment, etc. and, tell me, why is the Japanese term eta absent from Japanese dictionaries? And what nation's Ministry of Education can provoke an international issue with "recommendations" to revise historical descriptions of national conduct and have them carried out against the authors' wishes?

Evidently, there are no easy answers to the questions, and some have even suggested that a person's attitude towards Miller and Millerism is a matter of personality and attitude, such as whether one is an optimist or a pessimist, rather than a question of objective fact. However, I (DLP) think that we can all agree, generally speaking, on the following:

(1) Historically, the number of Westerners (Europeans and Americans) who speak Japanese has in the past been much smaller, proportionally speaking, than the number of Germans who speak French or the number of South Americans who speak Italian (to give just a few examples for the sake of making a comparison). Thus, French-speaking Germans have never created as great a sensation in France as Japanese-speaking Westerners have at times made in Japan. An unusual phenomenon always causes more stir, interest, or sensation than one which is more common, and knowledge of Japanese by foreign-looking persons has been very unusual in Japan. Until our days, the number of foreigners speaking Japanese may have been roughly the same as the number of foreigners speaking Welsh or Basque.

(2) A person — any person — who possesses a skill or trait which is considered to be unusual or surprising in a certain social situation will often be regarded with undisguised curiosity, or if events take a certain course, even with hostility or suspicion. If traits of my behavior mark me out as a clearly uncommon or out-of-the-ordinary person, I am not alarmed. I think that the people around me are entitled to a certain amount of explanation for my peculiarities, and it is my responsibility to explain myself to them, to make myself understood and accepted and (in some cases) even liked. At least I ought to try to make them feel comfortable about the things that bother them at first glance. I do not consider it to be my inalienable right to walk into a foreign environment, where my appearance or behavior is likely to be considered unusual or aberrant, and to demand to be immediately accepted and liked. There may be ineradicable prejudices against me or my ethnic group. Ethnic minorities have always had to struggle against such odds, but they would be very foolish if they did not make strong representations to appeal to the good will of the best elements in the majority population. I think some of the phenomena connected with "Millerism" may result from the fact that for a long time Europeans have taken their own languages too much for granted, have considered it to be natural for non-Europeans to learn European languages, rather than the opposite, in fact have insisted that English be adopted as a lingua franca throughout much of the world, and have generally neglected to learn Asian and African languages. Their attitude is rooted, I think, in a colonialistic view of the world, a view which is today utterly anachronistic and irrelevant. Against this background of universal insistence on the natives' using English, when a small handful of Europeans begin to speak languages which almost no "foreigners" have ever spoken before, their conduct cannot but evoke considerable surprise and curiosity, at the very least. At any rate, there is a definite historical basis for the predicament, and there is really no need to panic; a large dose of patient determination will, eventually, smooth over the difficulties.

(3) On the other hand, foreigners who have Japanese language skills are very lucky. They have so much going for them! They have every possibility to explain themselves, to make themselves understood, and to try to change the attitudes of those who surround them. If they have a certain minimum of social skills, I don't think that they will have any trouble in establishing good rapport with the Japanese and winning their good will. If there are some who have personality disorders (if they are generally "obnoxious"), they should not blame the Japanese, although I do not believe that many such unfortunates have mastered the Japanese language.

If there are really any Japanese at all who dislike Japanese-speaking foreigners to the point of throwing them out a window, this is probably just a case of simple ignorance or racial prejudice. If there is racial prejudice involved, it would be wise for the victims to work with patient determination to remove the misunderstandings of the ignorant. I personally do not believe that the Japanese as a nation can be said to have any such widespread prejudices, although there may be widespread ignorance and disbelief, if only because so few Japanese have ever come into contact with Japanese-speaking foreigners. The vulgar jokes about hen na gaijin are, just examples of very bad taste and should be quietly discarded in the same way that ethnic slurs are being discarded everywhere. They are boring.

I differ with Mr. Hazelrigg about "language mysticism." I do not consider it to be at all sinister or reprehensible. But when we talk about "linguistic mysticism," we ought to define clearly what we are talking about. Do we mean "sacred languages"? Some languages have been considered as "sacred" because divine revelations were believed to have been given through them. I can mention examples such as Sanskrit, Hebrew, and Arabic. I am not aware of any such belief among the Japanese about their own language.

Do we mean a belief that the national character resides in some mysterious way in each national language? There is nothing odd about that. It would seem to me to be demonstrably true. A moment's thought would suffice to convince anyone that this is true, except in languages like Esperanto and perhaps English which are not closely identified with any one ethnic group. Linguistic nationalism has been prevalent most frequently in historical cases where the disappearance of the national language was perceived as posing a real danger of loss of national identity. Persons like Miller who feel quite secure about the fate of their own native language are probably ill equipped to judge the feelings of Navajo or Okinawans or Ainu about the continued survival of their languages. Besides, who can blame the Welsh or Basques or Catalans for trying to preserve their languages from extinction? We ought to be grateful to them for their efforts. Language riots in Belgium or Sri Lanka are unfortunate, but they can hardly be blamed on "linguistic mysticism" alone. Economic and social factors have been mentioned more frequently.

Does "language mysticism" mean a belief in mystical powers inherent in language (the spoken or sung word)? Such a belief is almost universal among all branches of mankind, and this is certainly true of the early Japanese. The term kotodama in Shintoism, does not refer to any distinctive quality of the Japanese language in particular; it refers to a mysterious quality in words in general — the quality which gives man power over the non-human world and makes it possible for human beings to pray, to prophecy, to utter incantations, to communicate with spirits and to practice magic. It is well known that the ancient Japanese believed that spiritual force ("soul" or tama) lodged in language; when a word was uttered, it took on an independent existence of its own and was able to influence reality. Both good and evil results could ensue from words which had been spoken or sung. The belief has something in common with the superstitious prohibitions against uttering certain persons' names, which are frequently seen among "primitive" peoples. (Those few who are interested in this matter can consult the entry for kotodama in any good dictionary of Japanese. I recommend particularly the authoritative Jidai-betsu kokugo daijiten, Joodai-hen, p. 300.)

I have the impression that Miller has a mistaken understanding of this aspect of early Japanese religion, or perhaps he has been misled by vulgar contemporary writers who have distorted the facts for their own purposes. At any rate, he is mistaken if he argues that the contemporary Japanese believe that the Japanese language is a "sacred" language because it has mystical qualities (kotodama) lacking in any other language. They do not believe any such thing, and that is not what kotodama means at all. Linguists, I think, should be a little bit more careful when they write about the religious ideas of other nations.

* * *


It is customary in Japan to pay translators by the number of pages of manuscript paper (genkooyooshi) of the original Japanese text, each page containing spaces for 400 characters. This practice may now be changing, with payment being offered sometimes in terms of English typed pages (which may average 230 or 250 words each). In the U.S. the custom is to pay for translations from Japanese by the English word, rather than by the page. (Of course, those who are skilled at estimating can easily make conversions from numbers of Japanese characters to English word counts, or even more easily from Japanese manuscript sheets to typed English pages.) Some readers in Japan have found the U.S. practice to be difficult to understand and have written letters inquiring about it. Fred Ulernan of Shibuyaku, Tokyo, writes (March 4, 1984):

Like many earlier comments on rates, the section you Source: HONYAKU JITEN gives Japanese-to-English translation rates in terms of English pages. Why? If you are doing translation from the Japanese, why not charge in terms of Japanese pages? (400-character pages are standard.) Why should the translator be rewarded for flabby writing — or penalized for tightness? In a way, billing Japanese-to-English translation in terms of English pages is like the military contractor's cost-plus billing: it rewards the sloppy worker and penalizes the person who spends the extra time to tighten up the English and use the right word instead of several approximations.

Another reader in Japan, Clifford Bender of Kyoto, writes with even more far-going questions (March 6, 1984). I am reproducing parts of his letter:

I am a technical translator living in Kyoto, and would now like to move back to the States. However, though familiar with translation work here, I am not so with that in the States. For example, I read in your newsletter and other places that translations are charged by the word, but what is "one word"? Here they are billed by the page, usually 20–25, 60-space lines/page. I hope you, or some of your readers via the newsletter, will be able to answer this and the following questions. As you have asked for contributions and so this is not a onesided exchange, I have tried to answer most questions as they pertain to Japan for general interest and those who may be contemplating moving to this side of the ocean.

To repeat, I have read that translations in the States are paid for by the word, usually quoted in units of 100 or 1000, whereas we go by the page. What comprises "one word," and how are the words counted? Are they averaged out to X words per line or page? Surely armies of minions don't sit around counting hundreds or thousands of words every day. If one word consists of, say, five spaces, what happens to the empty spaces at the end of a line and between paragraphs? For comparison, the average here in Kansai is ¥2200–2700/page; this is generally after tax. Indeed, many companies quote freelance rates "after taxes" and pay the flat 10% levied on part-time work themselves. This tax is reported on the "W-2" form issued at the end of the year and might be received when one files their Japanese taxes.

What is the average load for a freelance and for a full-time translator in a translation company? I noted that Charles Kay of Massachusetts translated 1.5 million words in five years. Not accustomed to counting words, a friend and I were rather surprised at the amount this worked out to: in the last two years I have translated nearly one million, this friend has done about 2.8 million words in four years (when calculated at 240 5-space words/page). Of course, difficulty of content has much to bear on productivity, which raises the next question: What is the range of subjects handled in J-E translating, and what is their general level of difficulty? Is a technical background in the field required, or simply the ability to understand new technical materials with a little background reading? Here, one is expected to be a jack-of-all-trades, and master of none. The ability to understand rough Japanese and to ask questions of the client is usually sufficient. Reference materials are often provided for the translation in question.

Related to this question is the definition of "technical." Is most in the States "technical" as opposed to general simply by virtue of its being science- or industry-oriented and noncultural? I noted the discussion of karaoke, dasai (a real nerd), and other colloquial terms in the newsletter, terms which would not seem necessary for technical materials. Again for contrast, a large portion of translation here consists of operating manuals, narrations, teaching and training materials for salesmen, technicians, repairmen; product descriptions for use with patent applications, press releases, advertising, and so on.

What is the general level of Japanese language education and experience in Japan required? Extensive or hit-and-miss? For example, didn't you translate the "Kojiki," Mr. Philippi? (or are there two?) Is the Japanese original usually printed — and therefore usually well written — or are there a lot of handwritten "genko" (which here may be doodles and have thought notes)?

What, in the U.S., does one do when something is simply not understandable? This could be due to words made up by the author (e.g. in-house abbrev., names, part names) or to cryptic, excessively ambiguous grammar, new colloquialisms (dasai?), or many other reasons. Here, of course, one can easily call the author, customer, manufacturer, or even friend or spouse; there I should think this rather more difficult.

In Japan translations are often published, and translated for that purpose. What, however, are they for in the States? Research into developments in Japanese industry? Market research?

With this, I seem to have run out of questions, and I will be extremely appreciative of any answers you or your readers might give. If there are any questions, or any other way in which I might help, please let me know.

First, let me say that I am indeed the "Mr. Philippi" who translated the Kojiki, that I am flattered that you know my name, and that I also worked as a technical translator in Japan for about 10 years before moving to San Francisco at the end of 1970, since which time I have been plying my trade here. Therefore, I think I can answer some, perhaps all, of your questions, but I think that some of them may have been answered already in previous issues of the newsletter. Perhaps some other readers may be better able to answer some of them in future issues.

About the range of subjects of the jobs that are translated here, I would say that the mix depends to a large degree on the type of clients which a translation agency has (whether they are manufacturers of automobiles, electronics firms, camera makers, etc.), but that in general a large percentage of the jobs will be printed monographs (papers published in scientific journals or proceedings of learned societies) and handwritten manuscripts (yes, the kind with scrawled doodles on them, too). But one important difference between working in the U.S. and working in Japan is the fact that in the U.S. a large proportion of the work done seems to be translations of Japanese patents. Undoubtedly there are many J-E translators working in Japan who have never had any reason to translate a single Japanese patent. In this country, we have immense amounts of Japanese patents to translate.

Now about the question of word counts, which seems to interest readers in Japan.

To tell you the truth, the real story is that in the U.S. translation agencies like to pay translators by word counts — but the words of the original language, not of the English translation. There is a good reason for this. Since some languages, notably Russian, do not use articles, the word count of a Russian original is almost always less than the word count of the English translation. Therefore, the translation agencies in this country have, oh so wisely, hit on the idea of paying translators by the original-language word count instead of the English word count. Pages mean little to them here, since there is no uniform number of words on a given page, and manuscripts can be typed in 10 pitch or 12 pitch, double-spaced or single-spaced or even 1½-spaced. Anyway, the count is usually (in the case of the European languages, that is) by the original word count, and therefore the number of pages is of no importance.

However, when they look at a Japanese text, they can't see any breaks between words at all and can't even begin to make "word counts," either in the original language or in English. What is a Japanese "word," anyway? They know nothing about the convenient way of counting by 400-character manuscript sheets. Even if someone were to tell them about that, I doubt whether they would be very interested. Therefore, their solution is to say: "Since we don't know how many Japanese words there are in this text, why don't we just wait until we get the English translation, and then count the English words? We can then pay the translator the going rate according to an English word-count. Or even better, perhaps we can show the original to the translator and get the translator to make a rough estimate for us and tell us how many English words there will be in the translation after it has been translated." That is their reasoning. Remember, these people, unlike the people that you work for in Kyoto, are American translation agencies and can't read Japanese.

Incidentally, making estimates of English word-counts from Japanese texts is a skill which is much in demand here, as you can surmise from the foregoing. A number of conversion formulas are in vogue here (my favorite would be 2.4 Japanese characters per 1 English word, but I think I have heard something about "2 characters per word"), and some non-Japanese-speaking Americans working in translation agencies have become extremely skilled at estimating the final English word-count by looking at the Japanese originals.

Thus, translations from Japanese in the U.S. are paid by the English word-count. How do we define a word, and how do we count words? My method is the following. First, you make an estimate to determine the average number of words in a full line in your job. This can be done very easily and objectively. If you are using a 10-pitch typewriter, the number of words per line may be about 10, plus or minus one or two words, depending on the contents. Now, every page normally has two counts on it: (a) the count for the number of words in the full lines; and (b) the count for the number of words in "widow" lines. To obtain count (a), you simply multiply the number of full lines on the page by the average number of words. This is especially easy to do if you use a word processor with the lines justified on the right (that's what I am doing with this newsletter). Then you go through and count, manually, one by one, the number of words in the incomplete lines, and this gives you count (b). You add the two counts together, and this gives you the total number of English words on that particular page. Repeat this for every page, and you have the total word-count for the job. It is a bit tedious, but no big deal. You can always pay someone to count the words for you if you can't be bothered with it yourself. (I wish my word processor had a built-in word-counter.)

A good translator ought to be able to turn out at least about 2,000 words a day, and some average 3,000 words a day year-round. Even 4,000 words a day is possible, if you really want to work that hard. Anything over 5,000 words a day, sustained over a long period of time, would be almost miraculous, and there would probably be all kinds of problems with the quality. Naturally, using a word processor rather than a typewriter is the best way to increase both one's productivity and one's quality at the same time, and it is highly recommended although the equipment may be a bit expensive.

Now for my conclusion about the economics of the thing. If you work, say, 300 days a year (allowing yourself time off for vacations and sickness, etc.), and if your average output is 2,000 words a day, your annual total will be 600,000 words a year. If you are being paid $50 per thousand, or .05 per word, your annual pay will be $30,000. By the way, you receive the full amount; taxes are not deducted at the source in this country — for translations, that is. (A translator receives the full amount and figures out his own income taxes.) If you increase your output to 3,000 words a day, the annual total will be 900,000 words, giving you an income of $45,000 if you are paid $50. If you translate an average of 3,000 words every day of the year, your annual total will be 1,095,000, and your income will be $54,750, which is more than a university professor is paid in California. You can also increase your income if you get a raise. One million words a year is by no means an unusual feat. Charles Kay may have translated 1.5 million words in five years, but that was probably because he was running an agency at the same time and did not have time to devote himself full-time to his translating. If you, Cliff, translated nearly one million words in two years, I would say that your output is still not remarkably high. In fact, it is probably below the average. Perhaps an investment in a word processor, if you don't have one yet, might be a way to increase your productivity, or perhaps you have some other source of income and are not completely dependent on translating, as many of us are.

I asked some other readers to continue this discussion and supply some more answers to the questions from Fred Uleman and Cliff Bender. Here is what Fred Schodt of San Francisco wrote:

"As Don has pointed out, most Japanese-English translations in this country are billed from a count of the English words, for a very simple reason: Americans cannot make heads or tails of a page of Japanese characters. When I worked in Japan I became used to the system of counting everything according to the number of 400-character genkōyōshi so when I set up shop here in San Francisco I too found the American way of doing things confusing. There are, however, advantages and disadvantages to both systems.

"The most widely accepted way of calculating Indo-European language words on a page is called the International Standard Word Count. For English it is as follows:

1) Find the longest line on the page, and count the number of spaces in it (spaces, not characters).

2) Count the number of lines on the page (lines that go less than halfway across the page are ignored).

3) Multiply 1) x 2). 4) Divide by 5.

"This yields a fairly accurate estimate, but in itself is quite time-consuming. Technical documents I find are often easier to count word by word. Sometimes it is also possible to eyeball blocks of text and come up with a fairly accurate count. The easiest method of all, however, is with a word count function on a word processor. I am currently using a QX-10 (QC-10 in Japan), which comes with a Peach Tree Spelling Proofreader bundled in its software. The proofreader counts the number of words in the document as it goes through it, and displays the total on the screen in a matter of seconds. As far as I can determine, the count is fairly accurate, but it includes numbers, and will therefore inflate a technical translation somewhat. For a long document of hundreds of pages, however, the margin of error is probably not that much greater than the International Standard Word Count. There are other programs on the market which may be even more accurate, and according to Dan Kanagy it would not be hard to write one.

"As Fred Uleman has pointed out, charging on the basis of words in the target language rather than the source language would seem to penalize the succinct, careful translator, but this is not always the case. From my experience, some technical documents in Japanese (particularly those using a great number of kanji compounds) automatically become longer in English. Charging on the basis of Japanese characters could then be unfair to the translator. Japanese speeches, on the other hand, have different standards of redundancy than English ones; a document of ten pages in Japanese often has to be whittled down to eight in English. In this case the person paid on the basis of English words is being gyped.

"The world of translation has some arcane aspects to it, and as Jimmy Carter and Gautama have both pointed out, life ain't always fair.

"While most people in the U.S. charge according to the number of translated English words, there are exceptions. I have met people who charge according to the number of pages, although I can't imagine how on earth they do this for technical work. I have even talked to agencies that charge on the basis of hours of work, but the potential for abuse of such a system is enormous.

"The happiest compromise is available to translators who work directly with clients. Given that this is a free country, we can use any system we want: if the client agrees, we can charge according to the number of Japanese words or English words. The optimum method is to bid on a job, taking into account not only the number of words, but the degree of technical difficulty, legibility, time, and other problems that plague us. If the client is sharp, he will then collect estimates from other translators and agencies, and make an educated decision. I like this method for long jobs, where word counts are often only a rough approximation anyway.

"I would be interested in hearing from other translators how they charge for numbers in technical documents."

* * *


"Technopolis" is a word coined by bureaucrats in the Ministry of Trade and industry in line with MITI's industrial policy "vision" for the 1980's announced in march, 1980. The idea is to attract to local cities high-tech industries like electronics, electromechanical engineering (mekatoro) and biotechnology, at the same time bolstering universities, research laboratories and civic centers in the areas. A "technopolis," in other words, is "a concept of a new type of industrial city, as opposed to the past program for local city buildup that depended on heavy and chemical industries." Special tax incentives will be given for investments in machinery, equipment or building in these areas for five years after designation as "technopolises.". (Japan Economic Journal, March 6, 1984; Nihon keizai shimbun, February 15, 1984)

The following 14 areas are the ones which the government has decided to designate as "technopolises":

* * *


I noticed an advertisement in the March 12, 1984 issue of the Nikkan kōgyō shimbun announcing that a company called Nihon Jidoo Honyaku Kenkyuusho K.K. (Mishina Building 4F, Kanda Tacho 2-7, Chiyoda-ku 101, Tokyo, tel. 03 (254)9322/7330) is offering assistance to companies in Tokyo who wish to set up their own machine translation programs on word processors. The company has organized a seminar consisting of 24 lectures on the second Wednesday and Thursday of every month. The lectures, to be given by Sakae Okaji, are to begin on April 11 and 12 and are to be held at the Tokyo Green Hotel in Awaji-cho. The full course of lectures costs ¥l,000,000 (including text) in advance, or each lecture separately costs ¥50,000.

I don't suppose we will hear anything about the contents of the lectures, but it is interesting to know that such a service is being offered by a small private company in Tokyo.

* * *


Dataquest Incorporated is an American market research company founded in 1971. Located in San Jose, California, it specializes in researching the high-technology industries, with main emphasis on the computer industry (including office automation and CAD/CAM), computer peripherals (copying and duplicating, electronic printer, display terminal, and computer memory industries) and semiconductors. In 1978 Dataquest was acquired by A.C. Nielsen Company and is now a Nielsen subsidiary. It publishes newsletters and reports, holds annual conferences in the U.S., Europe and Japan, and offers consultation and data base services. The San Jose headquarters is said to accommodate almost 300 employees, and Dataquest claims that its various services have more than 2,000 clients.

In January 1982 Dataquest Japan Limited was established in Tokyo to strengthen information-gathering activities in Japan. In April 1983 Dataquest Japan was upgraded to a permanently-staffed organization headed by Osamu Ohtake, who formerly worked as Components Group Manager for Dempa Shimbun. The Japanese office now has 5 employees and serves 280 Japanese clients in addition to reporting on the Japanese industries to Dataquest in San Jose. Masakazu Nanseki joined the staff as Marketing Manager in January, 1984, and additional personnel are now being recruited in Japan.

With the establishment of the permanent personnel in Japan, Dataquest began to offer its Japanese Semiconductor Industry Service. This English-language service, which sells for a $10,500 annual fee, includes two 200-page database notebooks on Japan and its semiconductor industry, a fortnightly newsletter, and telephone inquiry privileges. There are said to be about 100 subscribers to this particular Dataquest service. Gene Norrett, based at the San Jose headquarters, is Manager of Dataquest's Japanese Semiconductor Industry Service. He had a 14-year career at Motorola. An analyst fluent in Japanese, Sheridan M. Tatsuno, is also working as an Associate with the JSIS in San Jose, and Patricia S. Cox serves there as Research Associate. Other analysts based in San Jose include Frederick L. Zieber and Lane Mason; Maureen Davies is Group Secretary.

Information collected is stored in databases in the company's computer in San Jose (a DEC VAX minicomputer). An on-line database service is currently being planned and may become available sometime this year.

Dataquest Japan will offer a "1984 Japanese Semiconductor Industry Conference" on June 21, 1984 at the American Club in Tokyo.

The company's addresses are:

1290 Ridder Park Drive
San Jose,
California 95131
Phone: (408) 971-9000

Azabu Heights, Suite 410 1-5-10
Roppongi, Minato-ku
Tokyo 106, Japan
Phone: (3)582-1441

[Information in this article is from Dataquest trends of Autumn 1983 and other Dataquest materials, an article about Dataquest Japan in the Nikkei sangyō shimbun (March 6, 1984), and an article from the San Jose Business Journal (May 29, 1983).]

* * *


We reported on the Fuji Xerox 8012-J Star system in our issue no. 11 (pp. 16–18). I sent information about the Fujitsu My OASYS, which is to be marketed in the U.S. this month, to Joseph D. Becker, manager, International Advanced Development, Xerox Office Systems Division. He sent me the following comments (March 13, 1984):

I'm sure the Fujitsu "My OASYS" will be a big hit in the US, at $4K including printer. It has always been my favorite among our competitors, especially since they had the guts to dispense with the terrible JIS kana keyboard layout. I'm pretty sure that its kana-kanji conversion lacks grammatical processing, but for average Japanese typing it still looks like a good system for the money. Of course, our JStar is a different class of machine. My old letter to Richard Willis explains the difference as well as I can, so I have enclosed a copy. Last week, our friends at FUJI XEROX installed a dedicated trans-Pacific phone line, and we can now transmit documents of the quality of this one across the Pacific in a few seconds. I know you can't afford that luxury now, but I do believe that our technology will "trickle down" quite fast, so I hope to be offering you a glimpse of what will be available to you in the future.

By the way, JStar is described in vast detail in the OA special edition of "Saiensu", No. 64 in their large-format paperback series. "Saiensu" of course is the affiliate of SCIENTIFIC AMERICAN, and I am told that SCIENTIFIC AMERICAN itself will have an article on "The Professional Workstation" next month, and I will have an article in it on "Multilingual Word Processing" by the end of the summer.

Here is the letter which Mr. Becker wrote to Richard Willis (dated January 6, 1984) in connection with Willis' article in the November, 1983 issue of BYTE magazine about the IBM 5550:

You were kind enough not to mention the bottom line on JStar vs. the IBM 5550: IBM will make a mint on the 5550, while XEROX has managed to lose a mint on Star! There are two reasons for this, which also happen to answer much of your letter:

(1) XEROX's top execs don't seem to have made up their minds yet whether they want to be in the computer business at all. So, they have not established the needed organization, market strategies, advertising, etc. (as you discovered). This policy doesn't make any business sense in view of XEROX's technical leadership in the development of personal computing systems — but who said that things make sense?

(2) There is a common fallacy about computer costs. In ancient times, like 15 years ago, a computer used to cost 1-10 million dollars. The cost has fallen so fast that people have a hard time telling where it landed. People who recently knew that computers cost millions of dollars now believe that they're nearly free. But this is simply false; the fact is that, as always, you still cannot buy something for nothing.

Since we cannot yet make a document-creation system that is both wonderful and dirt cheap, each company must make a choice of which type of system to develop and market: one that is lousy but cheap, or one that is wonderful but more expensive (although not millions of dollars). The former path leads to a fat bottom line, but it has one big flaw: It is easy to engineer down the price of a wonderful system, but it is impossible to build up a wonderful system from a bad foundation.

To my knowledge, XEROX is the only company that has taken the path of developing wonderfulness first, leaving cost-reduction to later. This fact leads to a fascinating result that you can easily observed: whenever any other companies wish to improve their systems, their only option is to start over with a completely new system based on technology from XEROX. This phenomenon affects the readers of BYTE.

Anyhow, in the 1983–4 time-frame it is impossible to create a really wonderful word/graphics processor in the alleged $5,000 price range, and so for a while the systems in that popular price range will continue to be crappy. In future years wonderful systems will eventually descend into that range, but thanks only to the fact that they were initially developed at a necessarily higher price. I'm glad to be one of the few who have the opportunity to be doing the job right.

The letter has a postscript about Fuji Xerox:

P.S.: Of course FUJI XEROX has no computer products of its own — they dealt solely in copiers until we dragged them kicking and screaming into the leadership of the "OA buumu". I think you understand the Japanese well enough to guess how pleased our FUJI XEROX associates were when they found out that the only good Japanese word-processing system was designed by Americans and made in USA!!

* * *


A laser printer would, naturally, be the ideal solution for our bilingual or multilingual printing needs. I have read that such printers have the capacity to print a maximum of 480 dots per inch, one of the highest resolutions commercially available, giving a printing quality on the same level as planography. The printers can be used to print graphics as well as text and are applicable to the multiple-copy requirements of publishing because they are based on copier/duplicator engines. However, such printers have been far too expensive for us lowly translators to afford. My understanding is that the cheapest laser printers used for word processing have cost about $20,000 or $30,000. Until now, that is.

An article by Richard A. Shaffer in the March 16, 1984 edition of the Wall Street Journal ("Laser Printers for Computers May Soon Dominate Industry") reports that within a few months several companies will begin "selling laser printers made by Canon USA Inc. [sic] at prices between $3,000 and $5,000." Somewhat later, the article adds, Ricoh of America Inc. and Xerox are also expected to begin supplying similar machines.

The Canon printer, the LBP-CX, is based on the company's PC-20 personal copier, which costs $1,000. Therefore, the Wall Street Journal reporter says, some industry followers are said to believe that its price could drop to as low as $2,000 in a year.

The article mentions some other types of inexpensive non-impact printers which are expected to appear shortly on the market. For example, Hewlett-Packard Co. this week introduced a $495 printer which operates by spraying ink on paper. Delphax and C. Itoh of Japan have a $5,000 machine which prints with a stream of ions, or charged particles. And Epson America has a printer using liquid crystals that should sell for a few thousand dollars.

The Nikkei sangyō shimbun (February 10, 1984) describes the Canon LBP-CX printer as an extremely compact semiconductor laser-beam printer. It is the smallest and lightest-weight printer of its type in the world, measuring 47.5 cm wide, 41.5 cm deep, and 25.4 cm high, and weighing only 26 kg. It is the first printer of this type adopting a cartridge (disposable) type construction in the electrophotographic circuitry such as the photosensitive drum and developer. The principle is the same patented principle as that adopted in Canon's PC series copiers. In other words, as Dataquest Trends puts it, "the components of the electrophotographic engine that normally require considerable maintenance are all contained in an operator-replaceable cartridge." (Autumn 1983, p. 11) If the need for routine maintenance is almost entirely eliminated thanks to the use of replaceable units in the electronic circuitry, this would be ideal for us translators, since the unavailability of maintenance has been the main obstacle preventing us from obtaining word processors capable of handling Japanese text.

Canon has already marketed a large-size laser beam printer, the LBP7121CS, for use with general-purpose computers. The new Canon LBP-CX will be usable with computers, word processors, facsimile devices, and CAD/CAM systems. Naturally, it can output both kanji and images. According to the article in the Nikkei sangyō shimbun the printing speed is "8 pages per minute." Canon has already begun delivering samples to a number of Japanese and overseas companies, and production and shipments will begin on a large scale as soon as OEM agreements have been signed. The article mentions a price of ¥500,000 per unit in Japan ($2,242 at a recent exchange rate) and $2,000 per unit overseas. According to the Dataquest publication quoted above, "projected end-user prices for the Canon printer are approximately $5,000."

Back now to that article in the Wall Street Journal. A Dataquest forecast predicts that sales of printers will peak during the next few years. 1988 sales will be about $2.5 billion, only slightly higher than in 1983. By contrast, sales of printers using alternate technologies are expected to rise to about $2.6 billion in 1988 from about $44 million in 1983. Laser printers will be the dominant non-impact technology, according to the article.

Apple Computer's Macintosh is one product in which laser printers may be vital to sales success. The printer sold currently with the Macintosh, and the only one with which it operates now, prints letters as patterns of dots and "can't space the dots closely enough to form the smooth edges that most business-quality machines product." An impact-type printer would solve this problem, but would not allow the Macintosh to print in a wide variety of styles and sizes of type, and to print images as well as text. Apple is therefore expected to sell the Canon LBP-CX laser printer, which can print with almost twice the detail of the printer now available for the Macintosh.

"Unless you are a graphic-arts professional, it's difficult to tell the difference between the output of this printer and pages that have been typeset," says Hideo Yamamoto, director of Canon's laser-beam printing division. "There is a catch, however," concludes the article. "The laser printer may cost almost twice as much as the Macintosh itself."

Readers of this newsletter will recall an announcement (see no. 12, p. 19) that a Japanese version of Macintosh with Japanese-language word-processing capability will become available in the fall of this year. Apple products are now being distributed in Japan by Canon Sales.

* * *


The Nihon keizai shimbun of March 24, 1984 has a dispatch from its correspondent Makiuchi in New York about attempts just beginning to market Japanese-language word processors in the U.S. It mentions that Fujitsu America will market its "My OASYS" word processor in the U.S. beginning in April (see our newsletter, no. 13, p. 14). Fujitsu intends to station sales and service personnel in New York, Los Angeles and San Francisco and will aim its sales at offices of Japanese business companies in the U.S. Fujitsu believes that there are about 4,000 such prospective users in the U.S.

Fujitsu recently held a get-together at the New York Hilton Hotel to announce its Japanese word processor. The demonstration was attended by nearly 100 Japanese businessmen.

The article continues that CPT, the Minneapolis-based word-processor manufacturer, also hopes to sell Japanese-language word processors to the same users. (See our issue no. 13, p. 16.) CPT has currently more than 60,000 systems installed in 65 different countries, and in Japan its products are used in about 800 companies, 200 of which also have offices in the U.S. CPT is planning to work together with its agent in Japan, a trading company called Kyowa Shokai, and will hold a word-processor seminar in New York in order to appeal to the Japanese businesses.

* * *


Fujitsu and the Korea Advanced Institute of Science and Technology (KAIST), which have been jointly developing a machine-translation system for translating Japanese into Korean (see our newsletter, no. 9, p. 3–4), recently unveiled in Seoul a simple demonstration system. A Japanese highschool reader was translated into hangul. The preliminary demonstration attracted attention because it is the first attempt at a machine-translation system for the Korean language.

In the demonstration, 600 words and 20 compound terms in Japanese and Korean were both entered into a Fujitsu computer and processed. The translation was successful with a 93% accuracy. Korean authorities were impressed, and high anticipations are entertained about the results which will be achieved in the future full-scale research.

Fujitsu and KAIST announced last October that they would engage in joint development of the machine-translation system, and during the next 1½ year they will develop a prototype system. The plan is to test and improve the prototype and to complete it during the first period of 1986.

In addition to machine translation from Japanese to Korean, the committee in charge of the project intends also to make systems for translating from Korean into Japanese and also from Korean into English. (Article in Nikkei sangyō shimbun, March 22, 1984)

* * *


Several translators I know have recently recommended a business dictionary entitled "Advanced Business English dictionary, English-Japanese, Japanese-English." The Japanese title is Eiwa waei Shin-bijinesu eigo daijiten. The editor's name is Hisao Kobayashi, and the dictionary is published by Pacific Management Consultants. Stuart C. Matthews, who has been a J-E technical translator for 8 years, has sent us this description of the dictionary:

"The title of this dictionary, originally published in 1977 by Pacific Management Consultants of 1-1-3 Shiba Daimon, Minato-ku, Tokyo 105, can be misleading especially as it sounds like one of those somewhat thick and not so professionally produced volumes of "Japlish" rushed straight off the press and into the welcome arms of your typical Japanese "kokusaika" businessman who will thumb through it diligently on the morning commuter train and reach the office not much the wiser.

"Actually, this is a combined English/Japanese and Japanese/English dictionary, and the 2,108 pages that make up this very solid and invaluable reference book containing about 80,000 terms, loan words, neologisms and abbreviations would naturally be quite out of place on a commuter train.

"The Advanced Business English Dictionary is a greatly expanded version of Pacific Management Consultants' "Gendai Business Eigo Daijiten" published in 1974 and what makes it a good buy for those of us translating from Japanese is the 1,347 pages (almost double the 715 pages in the E/J section) and the numerous examples of word usage in the Japanese/English section.

"For those of us faced from time to time with unfamiliar business terms that turn up in newspapers, specialist magazines and management reports, the dictionary definitely means "business" — a word defined by the publisher to encompass the legal, economic, trade, financial, accounting, political and tax fields. Terms from science, medicine and computers are also featured, incidentally.

"An added bonus is the appendix giving the English names of selected Japanese companies listed on the First Section of the Tokyo Stock Exchange, although this is superseded by the more extensive list in the Japan Times' Waei Honyaku Handbook which was revised in 1983.

"My experience with the dictionary has shown it to be the best available for the general business field and at 15,000 yen back in 1977 (the price may have gone up if an updated edition has since appeared) it is a most worthwhile investment. It has many terms and proper names which the big J/E Kenkyusha does not deign to list, and its style is more readable and up-to-date. All in all, it is a sounder and more reliable guide to the intricacies of Japanese business.

"Alongside the 1974 J/E Kenkyusha, Nelson's Character Dictionary, O'Neill's Japanese Names, the Japanese version of the McGraw-Hill Dictionary of Scientific and Technical Terms and the Japan Times' Waei Honyaku Handbook, the Advanced Business English Dictionary completes my short list of the J/E dictionaries most indispensable for general translation work."

* * *


An English-Chinese dictionary of engineering and technology, compiled by the Dictionary Editing Group, Zhong Wai Publishing Company, Hong Kong, (1036 pages, 1981) contains over 173,000 terms from fields such as mechanical engineering, optics, metallurgy, chemical technology, computer science, architecture and mathematics. It can be purchased for $49.50 by mail from John Wiley & Sons, Inc., One Wiley Dr., Somerset, N.J. 08873.

* * *


For your convenience, I am including in this issue a composite directory which embraces all the names submitted so far. Readers should let me know if you prefer to have the whole, composite directory reproduced in every issue, or whether it would be sufficient in each subsequent issue to have just a list of the additional names submitted in the meantime.

CHANDLER, Brian. 10, Simmondley Grove, Glossop, Derbyshire SK13 9NQ, England. Tel. Glossop (04574) 61906. Japanese to English Technical Translation. Specialties computers, electronics 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.

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).

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.

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)7626532. 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, 607 Franklin Street, Cambridge, Massachusetts 02139. Tel. (617)661-9784. Offers Japanese translation, typesetting, interpreting and research services.

MOFFETT, Linda, 749 Ash Street, Denver, Colorado 80220. Tel. (303) 3880284. BA, University of Colorado Japanese Studies, background in chemistry and pharmacy. Specializes in chemistry, batteries, textiles, medical.

MUNOZ, Roey. 7401 Cannon Mt. P1., Austin, TX 78749. Tel.: (512)288-1897. JapaneseEnglish and Spanish-English translation. ES 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.

PATNER, Richard. 4110 Derek Road, Madison WI, 53704. Tel. (609) 249-2506. Japanese to English translation of patents and technical documents with over ten years experience.

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.

SHKOLNIK, Alexander, 4434 Fulton St., Apt. 42, 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. 505 23rd St., Apt. 4, Oakland, CA 94612. Tel.: (415) 839-9527. Languages: Japanese. Subjects: Medical, biological, patents, chemistry, electronics, mechanical engineering, Oriental medicine, computers, legal, etc.

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.

* * *


It was an unexpected pleasure to meet Dr. James Unger, Associate Professor at the University of Hawaii. He was in San Francisco for a very brief visit, and we did this interview on March 31. I want to thank Dr. Unger for making time in his busy schedule to come by and be interviewed. I am sure the readers will find the contents of the interview very interesting.

Dr. Unger told me that he has found, in university departments where Japanese is taught, a considerable reluctance to offer courses in technical Japanese. The departments are usually oriented towards literature and do not consider it their duty to teach students how to read nonliterary texts.

Have any readers of the newsletter used the textbook Comprehending Technical Japanese by Daub, Bird, and Inoue? This seems to be the only textbook on technical Japanese. I understand it is for use by persons who know no Japanese — presumably scientists and engineers - and who want to learn just to read technical texts. I would like to ask someone to review it, and perhaps to evaluate its methodology in comparison with that used in the Sheffield University course.

Send in your tidbits, etc., and I will incorporate them in another issue. Those who send in glossaries (and problem lists) from now on should provide the kanji for the Japanese words. I will type them in Japanese if necessary. Don't forget to write if you want your name to be included in the Directory in the next issue.

April 10, 1984

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

* * *