Growing a Japanese Science City
The Nissan Institute/Routledge Japanese Studies series Editorial Board J.A.A.Stockwin...
136 downloads
1009 Views
2MB Size
Report
This content was uploaded by our users and we assume good faith they have the permission to share this book. If you own the copyright to this book and it is wrongfully on our website, we offer a simple DMCA procedure to remove your content from our site. Start by pressing the button below!
Report copyright / DMCA form
Growing a Japanese Science City
The Nissan Institute/Routledge Japanese Studies series Editorial Board J.A.A.Stockwin, Nissan Professor of Modern Japanese Studies, University of Oxford and Director, Nissan Institute of Japanese Studies Teigo Yoshida, formerly Professor of the University of Toyko, and now Professor, Obirin University, Tokyo Frank Langdon, Professor, Institute of International Relations, University of British Columbia, Canada Alan Rix, Professor of Japanese, The University of Tokyo Junji Banno, Institute of Social Science, The University of Tokyo Leonard Schoppa, University of Virginia Other titles in the series: The Myth of Japanese Uniqueness, Peter N.Dale The Emperor’s Adviser: Saionji Kinmochi and Pre-war Japanese Politics, Lesley Connors A History of Japanese Economic Thought, Tessa Morris-Suzuki The Establishment of the Japanese Constitutional System, Junji Banno, translated by J.A.A.Stockwin Industrial Relations in Japan: The Peripheral Workforce, Norma Chalmers Banking Policy in Japan: American Efforts at Reform During the Occupation, William M.Tsutsui Education Reform in Japan, Leonard Schoppa How the Japanese Learn to Work, Ronald P.Dore and Mari Sako Japanese Economic Development: Theory and Practice, Penelope Francks Japan and Protection: The Growth of Protectionist Sentiment and the Japanese Response, Syed Javed Maswood The Soil, by Nagatsuka Takashi: A Portrait of Rural Life in Meiji Japan, translated and with an introduction by Ann Waswo Biotechnology in Japan, Malcolm Brock Britain’s Educational Reform: A Comparison with Japan, Mike Howarth Language and the Modern State: The Reform of Written Japanese, Nanette Twine Industrial Harmony in Modern Japan: The Invention of a Tradition, W.Dean Kinzley The Myth of Japanese Uniqueness, Peter N.Dale Japanese Science Fiction: a View of a Changing Society, Robert Matthew The Japanese Numbers Game: The Use and Understanding of Numbers in Modern Japan, Thomas Crump Ideology and Practice in Modern Japan, Roger Goodman and Kirsten Refsing Technology and Industrial Development in Pre-War Japan, Yukiko Fukasaku Japan’s First Parliaments 1890–1905, Andrew Fraser, R.H.P.Mason and Philip Mitchell Emperor Hirohito and Showa Japan, Stephen S.Large Japan: Beyond the End of History, David Williams Understanding Japanese Society, Joy Hendry Ceremony and Ritual in Japan: Religious Practices in an Industrialized Society, Jan van Bremen and D.P.Martinez Understanding Japanese Society: Second Edition, Joy Hendry The Fantastic in Modern Japanese Literature: The Subversion of Modernity, Susan J.Napier Militarization and Demilitarization in Contemporary Japan, Glenn D.Hook
Growing a Japanese Science City
Communication in scientific research
James W.Dearing
London and New York
First published 1995 by Routledge 11 New Fetter Lane, London EC4P 4EE This edition published in the Taylor & Francis e-Library, 2002. Simultaneously published in the USA and Canada by Routledge 29 West 35th Street, New York, NY 10001 © 1995 James W.Dearing All rights reserved. No part of this book may be reprinted or reproduced or utilized in any form or by any electronic, mechanical, or other means, now known or hereafter invented, including photocopying and recording, or in any information storage or retrieval system, without permission in writing from the publishers. British Library Cataloguing in Publication Data A catalogue record for this book is available from the British Library Library of Congress Cataloguing in Publication Data A catalogue record for this book has been requested ISBN 0-203-21058-1 Master e-book ISBN
ISBN 0-203-02847-3 (Adobe eReader Format) ISBN 0-415-08134-3 (Print Edition)
Contents
List of boxes List of figures List of tables List of acronyms Series editor’s preface Preface
vii viii xi xiii xiv xvii
1 Introduction Meet Tsukuba Science City Three success stories in Tsukuba Science in Japan Large-scale science planning in Japan
1 4 17 26 32
2 Understanding a science city What is a science city? Communication, collaboration, and creativity Proximity The homophily-heterophily paradox Key questions about Tsukuba
34 35 36 39 41 44
3 History of the science city concept Problems in the capital The new city contest Settling on science Protest and political appeasement Tall boots, starry skies, mosquitoes, and wild dogs Tsukuba Expo ’85 Science as a theme in economic development
45 46 49 51 54 55 56 58
vi
Contents
4 Implementing the plan Changes to the master plans The loss of Ko¯no Ichiro¯ Project management Tsukuba’s role in Japan Summary of findings
63 63 68 73 79 81
5 Research communication in Tsukuba Establishing a science culture Informal study groups Evolution of five study groups Government exchange programs in Japan Government initiatives in Tsukuba Making sense of the role of communication initiatives
85 86 88 91 104 111 117
6 Collaboration networks Mapping communication structure Conceptualizing Tsukuba’s research collaboration network Structural views of the science city Number of articles published Number of authorships Are ministries a barrier to collaboration? Institutes and rate of publication First authorship of publications Summary
124 124
7 Lessons learned about growing science The development of Tsukuba Communication, collaboration, and creativity Access to knowledge in Tsukuba The privatization of Tsukuba Science City Comparison with Kansai Science City The future of Tsukuba Science City
157 157 163 170 172 173 177
Notes References Index
126 127 140 140 147 150 150 154
181 203 217
Boxes
1 2 3 4 5
International comparisons with Tsukuba Tsukuba in the mass media Inside a private lab in Tsukuba Communication at Tsukuba Research Consortium How not to communicate: Tsukuba Network
6 18 82 106 120
Figures
1 Tsukuba is near Tokyo, on the main island of Honshu¯, in Japan 2 Government laboratories are clustered together in the research and education district, the heart of Tsukuba Science City 3 The trends by year of the establishment of public versus private research laboratories are opposite, with the establishment of public labs having occurred in the late 1970s, and the establishment of private labs occurring in the late 1980s 4 Distribution of national newspaper articles about innovation and research collaboration in Tsukuba 5 National media coverage of Tsukuba Science City in four major economic and financial newspapers 6 Relational diagram of institutes, the number of researchers, and their parent ministries involved in superconductivity research in Tsukuba 7 Relational diagram among institutes in Tsukuba where LB film research is conducted 8 Rate of increase of residential and commercial land prices in Tsukuba Science City 9 Comparison of the Nouvelle Ville de Tsukuba (NVT) plan with the first, second, third, and fourth master plans 10 A conceptual relationship between the four Tsukuba master plans 11 Important organizations and key persons throughout the history of Tsukuba Science City
5
7
16 18 19
21 24 57
60 65 69
Figures
12 Degree of formality of Tsukuba Science City study groups 13 Multiple-answer responses by Tsukuba Science City study group members show what members expect from their participation 14 Institutes in Tsukuba and Tsuchiura and the number of their researchers who investigate heat transfer 15 Within-Tsukuba organizational links and their strengths for 1979 16 Within-Tsukuba organizational links and their strengths for 1982 17 Within-Tsukuba organizational links and their strengths for 1985 18 Within-Tsukuba organizational links and their strengths for 1988 19 While the percentage of all links within research institutes is decreasing, the percentage of links between institutes is increasing 20 For Japanese-language articles, the percentage of all links within research institutes is decreasing, while the percentage of links between institutes is increasing 21 For English-language articles, the percentage of all links within research institutes is decreasing, while the percentage of links between institutes is increasing 22 The percentage of links between Tokyo and other areas, and between Tsukuba and other areas, is increasing, while the percentages of Tsukuba-to-Tokyo and within-Tsukuba links are decreasing 23 Within Tsukuba Science City, the percentage of links between researchers at different national institutes is increasing, but decreasing between the University of Tsukuba and the national research institutes 24 The percentage of articles co-authored by researchers in different research institutes has increased, while the percentage of articles with one author and those with co-authors in the same research institute has decreased
ix
89
90 96 129 130 131 132
134
135
136
137
139
141
x
Figures
25 The percentage of Japanese-language articles co-authored by researchers in different research institutes has increased, while the percentage of articles with one author and those with co-authors in the same research institute has decreased 26 The percentage of English-language articles co-authored by researchers in different research institutes has increased, while the percentage of articles with one author and those with co-authors in the same institute has decreased 27 Trends in the percentages of authorships 28 Trends in the percentages of first-authorships of research journal articles by Tsukuba Science City authors and their co-authors
142
143 144
153
Tables
1 2
3 4
5 6
7
8
9
10
11
Number of researchers and support staff per government institute in Tsukuba Science City Number of members in the Tsukuba Applied Geoscience Society (TAGS) study group, by institutional affiliation Number of Tsukuba public researchers who study thermo-fluid dynamics, by institutional affiliation Cumulative number of attendees, by organization, at eight of the Genetics and Bioengineering study group meetings Topics of the first 35 weekly forums at the Tsukuba Research Consortium from July 1985 Number of articles, authorships, and links, by year and by language, published by researchers in Tsukuba Science City and their co-authors Results of four NEGOPY network analyses for research organizations and co-authors of research articles Tsukuba Science City research institutes with the most authorships of science and technology journal articles Tsukuba Science City research institutes with the most Japanese-language authorships of science and technology journal articles Tsukuba Science City research institutes with the most English-language authorships of science and technology journal articles The degree of interministry co-author relations is shown in matrix form, with the number of links
14
92 101
102 108
127
128
145
146
147
xii
12
Tables
between researchers of different ministries Productivity by Tsukuba Science City research institutes, as measured by dividing the total number of an institute’s researchers by the number of journal publications by those researchers in 1988
149
151
Acronyms
CRDC ERATO JETRO JICST JRDC KEK MITI NLA NVT RIKEN STA TAGS TRC
Capitol Region Development Commission Exploratory Research for Advanced Technology Organization Japan External Trade and Research Organization Japan Information Center for Science and Technology Japan Research Development Corporation National Laboratory for High-energy Physics Ministry of International Trade and Industry National Land Agency Nouvelle Ville de Tsukuba Institute of Physical and Chemical Research Science and Technology Agency Tsukuba Applied Geoscience Society Tsukuba Research Consortium
Series editor’s preface
It remains unfortunately true, halfway through the 1990s, that Japan is an underreported country. Despite significant increases in the amount of information available, it is still the case that few aspects of Japan and its people are discussed in comparable depth, or with similar assumptions about familiarity, to discussion of the United States, Britain or other major countries. Differences of language and culture of course constitute a barrier, though less so than in the past. As the patterns of our post-cold war world gradually consolidate, it is more than ever clear that the regional and global importance of Japan is increasing, often in ways more subtle than blatant. To borrow a phrase from Ronald Dore, we really should start “taking Japan seriously.” The Nissan Institute/Routledge Japanese Studies Series seeks to foster an informed and balanced, but not uncritical, understanding of Japan. One aim of the series is to show the depth and variety of Japanese institutions, practices and ideas. Another is, by using comparison, to see what lessons, positive and negative, can be drawn for other countries. The tendency in commentary on Japan to resort to outdated, ill-informed or sensational stereotypes still remains, and needs to be combated. For many years the relative lack of Nobel prizes awarded to Japanese scientists has been a reason for soul-searching in Japan. Indeed the “Nobel prize” problem became a symbol of the painful idea that Japanese were good at copying and improving on inventions made elsewhere, but were seriously deficient in respect of creativity. While recognizing that when industry was still at the stage of catching up with Western advanced technology it made sense to concentrate on “improvement engineering,” many
Series editor’s preface
xv
thoughtful Japanese already realized in the 1960s that this could not go on for ever. Eventually, Japan would reach a technological frontier and then, without a pool of creative scientists, national economic dynamism would be certain to decline. Some blamed the priority given to applied over pure science, others the concentration on fact cramming and rote learning in Japanese schools, still others the bureaucracy that seemed to pervade the scientific establishment. One reaction to these anxieties was the creation of Tsukuba Science City some way outside the Tokyo conurbation in the 1960s and 1970s. Its planners, in building a brand new city to specialize in science and education, were also motivated by the desire to stem the tide of left-wing student radicalism that tore universities apart throughout Japan in the late 1960s. In its early stages Tsukuba (like other purpose-built new towns started from scratch in the countryside, for instance Canberra) was a notoriously uncomfortable and uncivilized place to live. Eventually, however, it developed a life and an identity of its own. It also became home to many scientists and saw the creation of a great deal of new science. Dr Dearing examines the dynamics and achievements of Tsukuba Science City from a comparative perspective, and informs us about an important experiment in facilitating scientific progress. J.A.A.Stockwin
Preface
My interests in Japanese society date from 1983, when I began ten months of undergraduate study at Waseda University in Tokyo. Five years later, as a Ph.D. candidate at the University of Southern California, my familiarity with Japan and Japanese people made Japan a viable choice for dissertation field work. This book is a product of that year of field work (Dearing 1989a), four subsequent years of contemplation, writing, and presenting about the topic, and several later visits to Japan. I was initially aware of Tsukuba Science City because of publicity about the world exposition held there during 1985. From my studies at USC’s Annenberg School for Communication under the guidance of Everett M.Rogers, then Walter H.Annenberg Professor of Communications and now Professor and Chairperson of the Department of Communication and Journalism at the University of New Mexico, I became intrigued with better understanding the relationship between the diffusion of new ideas among scientists and engineers and the results of their communication. If communication is robust and occurs frequently among diverse specialists, is the likelihood of collaboration and creativity heightened? Must individual researchers initiate communication with each other (and thus perceive that they are autonomous and in control), or can government play a proactive and effective role in encouraging such relationships? And from a systems or community perspective, what happens when government intervenes in a massive way to “turbocharge” scientific proximity, communication, collaboration, and creativity? Tsukuba is a “living laboratory” in which to investigate answers to these questions. Moreover, this monumental social experiment in highly planned science had gone relatively
xviii
Preface
unstudied by both Japanese and foreign social scientists. For Japanese social and policy scientists, circumscribed access to information sources and a dearth of researchers interested in the sociology of science explain the lack of scholarly attention to Tsukuba Science City. For foreigners, the expense of international research, the Japanese language, and the personal contacts necessary for accessing sources and recruiting research assistants explain the absence of attention to Tsukuba. I had to overcome this latter set of obstacles. International research is expensive and requires funding beyond what a single academic department can normally provide. My project was no exception. The present research was supported in part by the United States National Science Foundation (Sociology Division, Doctoral Dissertation Improvement Grant #SES-8721459, and Supplemental Grant) and by the Annenberg School for Communication. The data-set of mass media articles discussed in Chapter 1 was donated by the Nikkei NEEDS Data-base in Tokyo. The mailed survey of the leaders of informal research groups discussed in Chapter 7 was underwritten by the Tsukuba Research Consortium. The data-set for the study of co-authorship reported in Chapter 8 was donated by the Japan Information Center for Science and Technology, a division of the Japanese Science and Technology Agency, in Tokyo. Follow-up data-gath-ering trips to Tsukuba were supported by the U.S. Association for Asian Studies, Michigan State University International Studies and Programs, Ibaraki Prefecture, and by the organizers of the International Science City Symposium in Kansai Science City, which is the Osaka-Kyoto-Nara area’s answer to Tsukuba. I have often heard it remarked by both foreign scholars in Japan and by Japanese scholars that the quality of access granted for interviewing individuals and inspecting institutes in Japan is greater for foreigners than for Japanese. My experience supports one half of this assertion. My sources were open and willing to discuss potentially sensitive information. The many Japanese researchers I came to know were particularly talkative, frank, and friendly. The chemists, physicists, engineers, and geologists were all interested in sociological interpretations of their science city, and willingly offered their own interpretations. Besides my interviewees and survey respondents, I am grateful to Professor William Dutton, University of Southern California and Brunel University; Dr. Kawahata Masahiro, Fujitsu Research
Preface
xix
Institute; Professor Steven Lamy, University of Southern California; Dr. Onda Masahiko, Mechanical Engineering Laboratory, MITI; and Professor Sato Hideo, University of Tsukuba. I also thank Professor Gerald R.Miller and Professor J.David Johnson, my chairpersons at Michigan State University, who encouraged me to finish this book despite knowing that it would land mostly outside of the academic field of communication. Gordon Smith and James Whiting of Routledge Books were very helpful in bringing the book along to its finish. Expert advice from Professor Arthur Stockwin, Oxford University, and from an anonymous reviewer, was most appreciated. Several people helped me in extraordinary ways. Professor Everett Rogers deftly guided me through a Ph.D. program and is a friend full of surprises. This book is dedicated to Ev and his unique and thoughtful style of scholarship. Kawamoto Tetsuzo¯, Managing Director of the Tsukuba Research Consortium, lent his credibility to my project and helped numerous times in opening doors which otherwise may have been closed to me. As an avid observer and facilitator of scientific communication in the science city, Mr. Kawamoto is the leading authority concerning the sociology of science in Tsukuba. Naito¯ Toshio, my primary research assistant, worked hard to organize a very competent research team, and facilitated some of the data analysis. Maenaka Hiromi translated some technical documents for me. Yoshida Hirotake and Yoshida Kanae graciously hosted me in their home, until a friend, Kazaoka Takako, found a house for rent. A number of Japanese whom I met were surprised to learn that a foreigner was studying Tsukuba Science City in a rigorous way. This sense of surprise extended to a Asahi Shinbun newspaper bureau chief, who wrote an article about my research, and to a television reporter with the Japan Broadcasting Corporation (NHK), whose camera crew accompanied me to several places in the taping of a national news feature. Their work exposed prelim-inary versions of my project to tens of millions of Japanese. Since that time, audiences in Japan, Britain, the United States, Egypt, and Canada have provided me with useful ideas as I have crystallized my thinking about Tsukuba in relation to other research communities throughout the world. These critics, respondents, and listeners represent many specialties, including academics from management, business, urban planning, sociology, political science, public policy, history, anthropology, engineering, communication,
xx
Preface
and scholars who specialize in studying Japan. University and science administrators, along with politicians and government planners, private consultants, and company represen-tatives have been members of these same audiences. This diversity of interested persons suggests to me that a phenomenon as complex as a city built for scientific work is truly curious to an attentive, interdisciplinary, and international audience. I hope that this book satisfies some of that curiosity. James Dearing East Lansing, Michigan
Chapter 1
Introduction
“I have found a very suitable place near Mount Tsukuba.” With those words, former Japanese Minister of Construction Kn¯o Ichiro¯, one of postwar Japan’s two most individualistic and persuasive political leaders, announced the highly competitive selection of a backward, undeveloped agricultural valley as the site for a novel planned city on 27 August 1963 (Asahi Shinbun 1963a). Two years later, in July 1965, Ko¯no died. The sudden death of the powerful politician shocked Japan. In the mid-1970s, the first researchers began working in the muddy new community. Not everyone agreed with the late Minister Ko¯no that Tsukuba was such a suitable place to live. For example, when urbanite, upper-class Tokyo wives of the first researchers arrived in Tsukuba, the housewives separated, bagged, and placed their garbage outside their new apartment buildings for pick-up twice weekly. They had done this same routine in the metropolitan capital prior to their move. They waited. And waited. After the garbage had accumulated into smelly, humid mounds, the women angrily marched to the local village office to demand that their garbage be collected. The rural office worker was perplexed. “We don’t have any garbage service,” he explained, “because there is no garbage here in Sakura Village.” Until that time, the only residents of Sakura, which would later be incorporated as a part of Tsukuba, were farmers. They processed their own garbage into mulch. To placate the angry urban housewives, a pit was dug at the site of a future elementary school. The women were told to throw their garbage into the pit. In the meantime, the village would learn how to collect and process garbage (Asakawa 1981). Today, there is much uncertainty and pessimism among Japanese about Tsukuba Science City (even though Tsukuba now
2
Introduction
has Japan’s most innovative garbage processing system). Foreign observers, too, are pessimistic about Tsukuba. Some critics of the national government refer to the science city as proof that the powerful national ministries and agencies can make monumental mistakes in wasting taxpayers’ money. Construction of the science city has cost the government about 1.5 trillion yen, or 11 billion U.S. dollars, which is seven and a half times more money than the Japanese government originally expected the project to cost (Jo¯yo¯ Shinbun 1963a). Science critics question whether the world’s best science will ever characterize the city (Hamilton 1992a). Political critics oppose the science city on ideological grounds. They believe that the city represents extraordinary government control of academic and university organization. Social critics consider Tsukuba to be a failed experiment in the social organization of science. To them, there is a lack of traditional Japanese urban culture in the city. To the present author, a native Californian, Tsukuba looks more or less like a familiar community. Indeed, the campus of the University of Tsukuba was based on the design of the University of California, San Diego, campus (Dohi 1989b). But to almost all Japanese, Tsukuba appears quite strange. The vast majority of Japanese would, if asked, probably respond with apathy about the science city, and consider it an uninteresting and inconvenient place to visit (as millions of Japanese seem to have perceived Tsukuba during a hot and humid worldwide science and technology exposition in 1985). Increasingly, however, Japanese who live in Tsukuba and even some people who have only heard about the science city think of the city as being quite beautiful. The lengthy period of Tsukuba being a difficult place for Japanese to live is ending. Why investigate 1 such a place? It is frankly too early to judge the science city on the extent to which it is achieving its mission of producing creative basic research. The city of Kyoto celebrated its 1,200th birthday in 1994. Tsukuba was initiated 31 years ago, and only for half of those years has Tsukuba been home to researchers. Perhaps no 15-year-old community, especially one with such a peculiar mission, can be expected to look or feel like a mature living and working environment. Yet the young science city does exhibit both problems and notable scientific successes, the latter most notably in superconductivity, solid state physics, extreme high-energy vacuums, genetic
Introduction
3
mapping, and electron microscopy. And there are other, less obvious indicators of progress in Tsukuba Science City which should not be ignored, for they contribute in important ways to a productive culture of scientific work. What is happening in Tsukuba should be understood by persons interested in cross-national learning about science and technology planning as well as urban policy-making (Masser 1990a). Tsukuba Science City is the world’s largest-scale coordinated attempt to “turbocharge” the rate and quality of scientific innovation (Rogers and Dearing 1990), as well as the world’s most planned research community. Although many governments throughout the world are actively planning or constructing research communities, an experiment focused on the sociology of science on the scale of Tsukuba may never happen again. Rarely are communities subject to this degree of planning and introspection. Between 1962 and 1983 the Japanese national government and Ibaraki Prefecture commissioned 359 different studies about how to construct aspects of the science city, totaling 37,485 pages (Wakabayashi 1985). According to a consultant who was hired by the Japan Housing and Urban Development Corporation to investigate and write a detailed history of the science city, by spring 1989, approximately 1,020 governmental and private planning studies had been conducted about the science city (Yajima 1989a). In 1992, 24 prefectural governments in Japan maintained offices in Tsukuba, for status or surveillance reasons (Kawamoto 1992a). In Japan, if not elsewhere, a long-range experiment on the scale of Tsukuba could only have been committed to in the early 1960s, during flush economic times when politicians and planners were seeking ways to spend money. As an indecisive U.S. Congress made clear in 1993 by stopping, restarting, and stopping again the construction of the superconducting supercollider in Texas (after U.S. $3 billion had been spent and 11 of 54 miles of underground tunnel had been dug), the inability to predict the outcomes of long-range, large-scale science projects makes these proposals too risky for most politicians to support. So we should pay close attention to Tsukuba Science City because of its distinctiveness (Merton 1968:329; Mitroff 1974), and for the lessons which decision-makers in other communities and in other countries may draw from the Tsukuba experience and adapt to their own planning initiatives. In terms of its organization of scientific and engineering research, Japan is extremely important for
4
Introduction
policy-makers in other countries to understand. Japan has moved from the periphery to the center on the world stage of research productivity (Narin and Frame 1989; Schott 1994). This book may also be of interest to people who are interested in the sociology of Japan, the ways in which the Japanese national government works, Japanese science, and the process of building personal relationships and gaining access to technical information in Japan. This book is about one research community, why it was created, how the intricate plans and objectives for the city were thwarted during the city’s lengthy implementation, and the resulting research relationships within the community. Concepts of communication, collaboration, and creativity are used to assess what has happened in Tsukuba Science City. Ko¯no Ichiro¯’s opening statement at the beginning of this chapter serves as a key point of departure for organizing the book into three parts: (1) the political and economic context during the early 1960s in Japan, when the decision to construct a new city was made; (2) the extraordinary planning and pivotal implementation of the science city concept; and (3) contemporary indicators of how Tsukuba Science City is functioning. Each of the three parts of the book corresponds to a distinct period in the history of Tsukuba Science City. Because different research questions were posed when I investigated each time period, the conclusions of each of the three parts of the book are quite distinct. My goal is that these conclusions complement each other in explaining the process of “growing” a science city.
MEET TSUKUBA SCIENCE CITY Many Japanese people say that Tsukuba Science City is “not really Japan.” I think that Tsukuba represents a distinctively Japanese experiment in the organization of basic scientific and engineering research. There is nothing like it anywhere in the world…. Americans would call it a “frontier town.” Sharon Traweek, Beamtimes and Lifetimes: The World of High Energy Physicists (1988:40) Tsukuba Science City is located 35 miles northeast of Tokyo, Japan’s capital city (Figure 1). From Tokyo Station one can arrive in Tsukuba in one hour if traffic is light via a nonstop bus which departs every 15 minutes during the daytime. Traveling by the same means from
Introduction
5
Figure 1 Tsukuba is near Tokyo, on the main island of Honshu¯, in Japan
Tsukuba to Tokyo often takes two and a half hours because of inbound traffic heading to the capital. By train the ride takes about one and a half hours either way and requires a change of trains, as well as a local bus or taxi ride at the Tsukuba end, after arriving at Arakawaoki Station, which is outside the city of Tsukuba. A major freeway, the Jo¯ban Expressway, passes close to the city, but automobile travel is expensive because of the high cost of gasoline and highway toll fees, and automobiles face the same traffic as do buses. Tsukuba is only 25 miles northwest of Tokyo Narita International Airport, but there is no direct train or freeway between Tsukuba and Narita. Because it does not have a rail station, Tsukuba is a city built for, and based around, automobile travel. Tsukuba was the first
6
Introduction
Box 1 International comparisons with Tsukuba Research communities may be distinguished and compared in a number of ways. Several variables which have been used in previous comparisons are the degree of city planning, levels of public and private investment, numbers of high-technology firms, quality of life indicators, availability of venture capital, university resources, the number of Ph.D. degree holders, and access to strategic, production, and marketing expertise (Saxenian 1988; Rogers and Dearing 1989; Rogers and Chen 1989; Rogers and Dearing 1990). Comparison of levels of investment and numbers of high-technology firms are attempts to measure, in a roundabout way, the productivity of a science city. Another type of data by which we can begin to understand the relative productivity of science cities is the number of science and technology journal and book article authors who work in each science city. This type of data is a more direct measure of productivity since it is an output of research activity, not an input. Although the intent of this book is not to compare different science cities,2 data catalogued in the Current Contents Address Directory (1987), Science and Technology Geographical Index (Institute for Scientific Information 1987), provide one measure of research activity in different areas of rapid innovation for the year 1986. Tsukuba Science City, with 3,454 first-article authors,3 ranked ahead of Akademgorodok Science City in the U.S.S.R. (which had 2,555 first-article authors), and ahead of Austin, Texas (which had 2,121 first-article authors).4 But Tsukuba was slightly behind Cambridge, England (which had 3,584 first-article authors), and well behind Research Triangle, North Carolina (which had 7,500 first-article authors).5
Japanese city designed for the benefit of drivers (Kato¯ 1992b). Ninety percent of Tsukuba residents own an automobile, a very high percentage of automobile owners for any Japanese city. Thirty-one miles of pedestrian pathways make bicycling a somewhat common alternative to automobile driving. Walking is not so common because of the city’s expansiveness. Tsukuba Science City, with just over 166,000 residents, has about half of the land area of Tokyo, which has a population of 12 million people. Tsukuba is unlike Japanese cities in other aspects, too. The total area of the science city, which incorporates five towns and one village, is 70,543 acres (almost identical in acres to North
Introduction
7
Figure 2 Government laboratories are clustered together in the research and education district, the heart of Tsukuba Science City
8
Introduction
Carolina’s Research Triangle Park). The center of the city is designated as a research and educational district, and measures 6,669 acres in an elongated eleven mile by four mile shape (Figure 2). Almost all of the 44 national research institutes and both universities are within this district. They are surrounded by over 200 private research facilities. The government institutes and universities are grouped by topic into a higher education and training zone, a construction research zone, a science and engineering zone, a biological and agricultural zone, and a common use zone. Because of an emphasis on internationalism, it is not unusual to see foreign exchange students and foreign visiting researchers in the science city. About 3,000 foreign students and researchers live in Tsukuba at any one time, from about 90 countries (Kato¯ 1992a). Tsukuba has the highest percentage of foreign residents of any Japanese city (Kawamoto 1992a). The city also has the highest education level per capita and the highest number of books checked out from libraries per capita, among cities in Japan (Kato¯ 1992a). Located in the heart of the city, the University of Tsukuba emphasizes science and engineering, and encourages close research relationships between university faculty, students, and researchers at many nearby national research institutes. The university faculty and student body are generally not on a par with elite public universities such as the University of Tokyo or the University of Kyoto, nor with elite private universities such as Waseda University or Keio University. For example, from 1981–91, University of Tokyo researchers published 10,982 articles in physical sciences journals, compared with 8,853 for the University of Kyoto, 7,549 for the University of Osaka, 6,037 for the Tokyo Institute of Technology, and 2,784 for the University of Tsukuba. As a measure of importance or impact, articles by University of Tokyo researchers were subsequently cited by others an average of 8.22 times, compared with 7.05 times for the University of Kyoto, 6.54 times for the University of Osaka, 6.49 times for the Tokyo Institute of Technology, and 6.09 times for the University of Tsukuba (Institute for Scientific Information 1992). Considering the newness of the University of Tsukuba, these ratings reflect well on its progress. The University of Tsukuba stands out in other ways. Educational innovations at the University of Tsukuba attract somewhat different types of professors and students. In 1987 the university itself had 26 problem-centered institutes. The
Introduction
9
university was the first in Japan to offer double-track two-year terminal master’s programs and five-year doctoral programs (Fujimoto 1987). The demand by private companies for University of Tsukuba graduates is greater than the university can supply, partly because the university’s faculty and students have a relatively higher degree of contact with basic researchers in the many national labs, compared to faculty and students of other universities (Tokumaru 1989). Thus, University of Tsukuba graduates in certain topics may have more, and better, hands-on experience with research. And some of the research at the university is acknowledged to be first rate: between 1990 and 1992, a research team at the University of Tsukuba produced a series of articles which were cited more often than the articles by any other Japanese researchers in the world (Institute for Scientific Information 1992). But in contradiction to the image of a growing research and academic city, the number of professors, associate professors and assistant professors at the University of Tsukuba decreased by 182 (about 12.5 percent) during the three years between 1984 and 1987 (University of Tsukuba 1984; University of Tsukuba 1987). Several factors may explain this decrease, such as attrition due to retirement of older faculty members from the former Tokyo Teachers’ College (this college was transformed into the new university, a transformation explained in Chapter 3), who moved to teach at the new university but soon thereafter retired. Discontent among faculty who found the rather isolated living conditions in Tsukuba less desirable than they had anticipated may have contributed to this decrease. Political pressure by other universities directed toward the Ministry of Education to correct what administrators at other universities perceived to be an unfairly favorable faculty/student ratio at the University of Tsukuba (this ratio was about 1:10 in 1984) may have led to fewer hirings at the new university. Lastly, competition between Japanese universities to hire foreign lecturers and professors, of which the University of Tsukuba had employed a disproportionately large percentage, may have contributed to a decrease in University of Tsukuba faculty. Initially, the Ministry may have over-hired faculty, including foreign faculty members, to impress people with the new university. One source suggests that the “brain drain” from the University of Tsukuba is more serious than the numbers alone suggest. Productive
10
Introduction
faculty members in Tsukuba are being recruited by universities in Tokyo, and their replacements are invariably younger, less proven professors (Kumata 1989).6 Tokumaru Katsumi, who in 1989 was Vice-President for Research at the University of Tsukuba, said that the number of positions allocated to the university by the Ministry of Education has increased slightly, and that the loss of professors may be especially high in the humanities and social sciences. There is no unusual loss of science and engineering faculty (Tokumaru 1989). In a bold move in 1992, the university hired Nobel Prizewinning émigré physicist Leo Esaki as its new President. Most science observers consider Esaki a dynamic personality who immediately raises the status of the university and, perhaps, the city as a whole. Esaki studied at the University of Tokyo, worked for Sony in Japan and IBM in New York. He stayed in the U.S. 30 years, which is where he won the Nobel Prize, for work which he had mostly done in Japan. President Esaki’s immediate goal is to rapidly increase university-industry-government lab research relations as one way of building the science culture in his new home. He has pushed this technology transfer agenda by creating in July, 1994, the high-profile Tsukuba Advanced Research Alliance (TARA), which is a consortium-like partnership between the University of Tsukuba, foreign researchers, Tsukuba’s national research labs, and corporate laboratories based in the science city. TARA research projects in the biological and material sciences are teamoriented and will further link together Tsukuba scientists across organizational and disciplinary boundaries. Esaki and TARA Director Murakami Kazuo hope that more and more corporate labs in Tsukuba will become financial supporters of TARA. Eventually, more than 200 professors will collaborate through TARA on temporary contracts (Nagakura and Kikumoto 1994). University of Tsukuba professors are also taking part in research teams located at the Ministry of International Trade and Industry’s new National Institute for Advanced Interdisciplinary Research, in Tsukuba. These MITI teams involve professors and graduate students from several universities, along with researchers from several of Tsukuba’s national labs and from over three dozen corporations. MITI sponsors the Joint Research Center for Atom Technology in Tsukuba, where several university professors lead
Introduction
11
large-scale multi-year research projects. These interdisciplinary and interorganizational initiatives in Tsukuba are at the forefront of a nationwide trend in Japan toward more interministry cooperation, although in other ways Japan’s ministries continue their longstanding isolation from each other. The Japan National Laboratory for High-energy Physics (KEK) is the most famous laboratory in the science city. KEK anchors the northern section of Tsukuba’s central research and education zone, just north of the University of Tsukuba. KEK, which like the university is under the jurisdiction of the Ministry of Education, Culture, and Science, was Japan’s prototype interuniversity research institute (Chapter 3 documents some early history of KEK when the laboratory was being planned and built). The laboratory was initially built with a proton synchrotron. In 1987, KEK debuted Tristan, which for two years was the world’s highest energy electronpositron colliding beam accelerator. Tristan focused the attention of physicists from around the world on Tsukuba Science City. A third accelerator is being built at KEK to study b-mesons (Taubes 1994). In addition to its large-scale and innovative initiatives in the organization of world-class science, the city of Tsukuba is unusual in other ways, too. For example, underground a 4.6mile network of utility tunnels spans the central city. Inside these tunnels are large pipes for collecting municipal waste (pushed along by powerful thrusts of air) and dust, and pipes for distributing steam, cold water, electricity, cable television signals, and telephone service (Housing and Urban Development Corporation 1984). No wires or utility poles clutter the above-ground scenery. In the middle of the central district is a city center composed of a bus and taxi terminal, city offices, a main post office, the prefecture’s tallest office building, a hotel, department stores with restaurants inside, movie theaters, supermarkets, a parking structure and parking lots, small retail businesses, Tsukuba Science City promotion and tourist offices, a 5,900-square meter art museum and library, and the world’s largest planetarium. Neatly manicured flowers spell out T-S-U-K-U-B-A in large letters to arriving visitors. The architecture of the central city is mostly similar. Postmodern buildings are drab, weather-stained shades of gray concrete and steel. Obvious by its absence is the rather haphazard mix of truck stops, gasoline stations, small
12
Introduction
restaurants, convenience stores, and homes which is so prevalent throughout Japan. In Tsukuba, many posted guide maps and signs along wide walkways inform visitors where they are in relation to the rest of the city. Streets were dug slightly lower than the clearly defined residential areas as a noise abatement measure. In 1990, there were 93 public parks, seven kindergartens, eight primary schools, six junior high schools, and three high schools in the science city. Rows of young trees line the wide boulevards. This progressive level of public investment is limited to the central research and education district. A stark disparity exists with the surrounding agricultural area, in which roads are narrow and services are few. One gains the general impression that Tsukuba Science City was very carefully planned. It was. Dozens of apartment buildings, containing over 9,000 apartments, surround this city center. These apartments are owned by the Japanese national government, and they are rented to government researchers and their families for extremely low fees. Further to the north, 5,500 apartments and dormitories house students and company employees. Walkways cutting through grass lawns and stands of red pine trees connect complexes with one another and with the city center. Young mothers pushing baby strollers and primary-school-age children pass by. When it is not raining, thousands of futons and clothes hang over the highrise apartment balconies, which makes the area look colorful and somewhat messy. The thousands of heavily subsidized apartments, coupled with a dearth of new houses in Tsukuba, are major reasons why most researchers have chosen not to buy a house in Tsukuba. Having been forcibly moved once from Tokyo, the older governmentemployed researchers are cautious about investing in property in case they are transplanted again. Subsidized apartments allow them to avoid buying. Moreover, later in their careers federal employees often seek employment in private companies, thus it is necessary for government researchers and administrators assigned to Tsukuba to maintain personal contact with potential private employers in Tokyo (Glasmeier 1987). In 1992, the total population of Tsukuba was just over 166,000. Of this total, about 47,000 lived in the central research and education district, and 119,000 lived in the surrounding suburban district. The population in the surrounding district has almost reached the target of 120,000, but the population in the central
Introduction
13
district is far short of the 20-year-old 100,000-person target (Housing and Urban Development Corporation 1989b). The number of new residents moving to Tsukuba each year increased steadily beginning in 1972, reaching a peak of 17,000 in 1979, the year when staffing at most of the national laboratories was complete. The influx of new residents has since dropped markedly. Each working day a number of commuters, most of whom live in Tokyo, swell the local population. 7 The frequent buses from Tokyo Station are filled with research administrators, scientists, and engineers clad in dark blue business suits, and a sprinkling of casually dressed university students. Of Tsukuba’s 166,000 people, about 100,000 work for small businesses or are farmers and their families; 15,000 are students at the University of Tsukuba; and about 50,000 are government employees and their families (University of Tsukuba 1987; Kawamoto 1988; Onda 1988). In 1990, total employment by public and private research facilities was 18,809. Of this total, 10,628 were publicly or privately employed researchers. Of the 7,400 government researchers, approximately 2,700 held a Ph.D. degree (Tsukuba Research Consortium 1988b; Tsukuba Research Consortium 1990; Kawamoto 1992a). In 1994, the number of publicly and privately employed researchers had increased to about 12,000. Since 1980, most of the increase in the number of Tsukuba researchers has occurred in the private laboratories. For example, the number of government researchers increased by only 42 persons from 1989 to 1990 (an increase of 0.6 percent), whereas in the private research facilities, the number of researchers increased by 474, from 2,749 to 3,223 (an increase of 14.7 percent). The government researchers are assisted by about half of the 2,500 graduate students at the University of Tsukuba (University of Tsukuba 1987), which helps to alleviate a personnel shortage in the national labs. Table 1 lists the 44 national institutes and two public universities in Tsukuba by ministry, listing numbers of researchers and research-support personnel. Just how important is Tsukuba to the Japanese government’s total basic research effort? Very important. In Tsukuba, 45 percent of Japan’s national researchers work in 33 percent of Japan’s national research facilities, supported by 48 percent of the nation’s research and development budget (Onda 1988; Kawamoto 1992a). The city’s research importance begets
14
Introduction
Table 1 Number of researchers and support staff per government institute in Tsukuba Science City
Introduction
Table 1 Continued
Source: Kawamoto (1990)
15
16
Introduction
Figure 3 The trends by year of the establishment of public versus private research laboratories are opposite, with the establishment of public labs having occurred in the late 1970s, and the establishment of private labs occurring in the late 1980s and 1990s
Introduction
17
prestige. Tsukuba’s mayor has received sister city proposals from Malmö, Sweden; Manchester, England; Braunschweig, Germany; and Austin, Texas and Irvine, California, in the United States. Tsukuba is also affiliated with Cambridge, Massachusetts, and Summerland, Canada (The Japan Times 1988). The relocation and growth of pharmaceutical, microelectronics, medical, chemical and other high-technology firms in the Tsukuba area have increased since 1980, and dramatically since the 1985 world exposition drew attention to the science city (see Figure 3). In 1990, there were 166 private research organizations (129 of which were laboratories) operating in one of Tsukuba’s 14 research parks and industrial districts. Planning and construction had begun on another 44 research organizations, 40 of which were designated to be laboratories (Tsukuba Research Consortium 1990). The 166 private organizations employed 5,372 people, 3,223 of whom were researchers. One year later, in 1991, 37 more private research organizations were in the planning or construction stages (Tsukuba Research Consortium 1991). The private labs in Tsukuba, like a number of the corporate labs in Japan, conduct basic research, investigations which are not driven by the goal or immediate desire to see a marketable product result (Kinoshita 1993). There is a shortage of women in the science city. For male researchers, who work long hours in relative isolation, this is very worrisome. In Japan, men are expected to get married in their late twenties or early thirties, and women in their mid to late twenties. It is not uncommon for male researchers in Tsukuba to be 35 or 40 years old and single, and under pressure from their parents to meet more women. When they marry, male researchers often marry administrative assistants and secretaries. Those are the women they tend to know. In 1988, 90 percent of the Ministry of International Trade and Industry’s (MITI) 2,670 Tsukuba employees were male (Asahi Evening News 1988). About 70 percent of the students at the University of Tsukuba are male. THREE SUCCESS STORIES IN TSUKUBA Scientific breakthroughs make research communities famous. From Oxford, England, to Austin, Texas, scientific breakthroughs are the outcomes that matter. Acknowledgement within the scientific and corporate communities that a locality possesses a critical
18
Introduction
Box 2 Tsukuba in the mass media Another indication of the importance of Tsukuba Science City comes from Japan’s national mass media. Most Japanese cities with populations equivalent to Tsukuba have a total of two or three full-time national newspaper reporters, and no shared press room. In Tsukuba, a central press room hosts 16 full-time reporters and about 14 part-time reporters, some of whom are from television networks. The largest mass media organizations also have their own news bureaux in Tsukuba. According to these reporters, Tsukuba is the third most important area of Japan for science and technology news, behind the Tokyo and Kyoto-Osaka metropolises (Tsukuba Press Room 1989). In terms of the importance of news sources within Tsukuba, these reporters ranked the National Laboratory for High-energy Physics first, the complex of nine Agency of Industrial Science and Technology laboratories second, and the University of Tsukuba third. What have these reporters written about Tsukuba? An analysis of the content of these articles reveals that the number of news ar ticles about research and science results and technological innovation in Tsukuba did not increase much in frequency between 1979 and 1989 (see Figure 4). The number of articles about research exchanges or research collaboration in Tsukuba only slightly increased.
Figure 4 Distribution of national newspaper articles about innond research collaboration in Tsukuba Source: Nikkei NEEDS Data-base
Introduction
Figure 5 National media coverage of Tsukuba Science City in four major economic and financial newspapers (N=1,290 news articles) Source: Nikkei NEEDS Data-base Prior to spring 1985, media coverage mostly concerned government institutes moving to Tsukuba, the construction of public institutes, and government decisions about the science city. News coverage peaked during the beginning of the exposition and focused on specific technologies on display and the crowds of people. Many news stories during and after the exposition focused on private companies moving to Tsukuba (Nihon Keizai Shinbun 1989). For example, the yearly totals of articles about the location, construction or development of private companies in Tsukuba was 2 in 1979, 3 in 1980, 0 in 1981, 3 in 1982, 6 in 1983, 13 in 1984, 35 in 1985, 44 in 1986, 63 in 1987, and 39 in 1988, as shown in Figure 5 (the 1987 peak of articles included news about private investment in Tsukuba, and the local incorporation of six municipalities into Tsukuba Science City). In summary, national news about Tsukuba from 1979 to 1985 focused on public investment. News after Tsukuba Expo ’85 was increasingly focused on private investment. As of yet there have been relatively few articles about scientific breakthroughs and technological innovation in Tsukuba. It can be concluded that during this decade, the science city made the national news in Japan mostly for reasons other than scientific results.
19
20
Introduction
mass of knowledge and innovation concerning certain topics leads to investment and job creation. Here, three scientific breakthroughs in Tsukuba Science City are reviewed: The determination of superconducting oxide crystal structures; a highly conductive optical film; and the construction of an extremely high-pressure vacuum chamber. What are the communication processes in which the central investigators took part during the course of the three breakthroughs? High tc superconducting oxides In late 1986 and early 1987, physicists, mineralogists, and crystalographers worldwide were paying close attention to rapid advances in superconductivity. In Tsukuba well over 100 researchers were actively working on aspects of superconductivity, particularly the identification and specification of the molecular structure of superconducting materials (Okamura et al. 1987). The activity and pace of new knowledge was brisk and exciting enough in Tsukuba that the science city was temporarily referred to by some Japanese as “Superconductor Valley,” and the period of time was referred to as “Superconductor Fever.” 8 World-class research on superconductivity continues in Tsukuba (Izumi et al. 1989). The researchers working on superconductivity in Tsukuba are a loose network of people. They have not met as a group, although some of them attend the same occasional symposiums and workshops in Tsukuba, and the same national conferences of the Applied Physics Society of Japan, and the Physical Society of Japan, which are held in the larger cities of Tokyo, Osaka, and Nagoya. Superconductivity researchers in Tsukuba belong to one of five national institutes (see Figure 6). A sixth institute, the National Laboratory for High-energy Physics (KEK), is a common-use facility where these researchers conduct experiments. These six institutes are under the jurisdiction of three ministries. Ministry of International Trade and Industry (MITI) researchers work with each other and with professors and students at the University of Tsukuba (which is under the jurisdiction of the Ministry of Education). Likewise, researchers in the two Science and Technology Agency (STA) institutes work with each other and with professors and students at the university. There is no research collaboration on superconductivity between MITI and STA researchers. So the
Introduction
21
Figure 6 Relational diagram of institutes, the number of researchers (in parentheses), and their parent ministries involved in superconductivity research in Tsukuba Source: Izumi 1989
Ministry of Education’s two Tsukuba institutes which are involved in superconductivity research (the university and KEK) serve as bridges between two rival and noncommunicative groups of researchers in MITI and STA. Graduating from the same university is another way that superconductivity researchers at different Tsukuba institutes know each other, but these networks tend not to actively transcend ministry rivalries (Okamura 1989a). Some highly productive superconductivity researchers in Tsukuba fit the Western stereotype of the isolated laboratory scientist in a white coat who works feverishly into the night. One such researcher 9 at the National Research Institute in Inorganic
22
Introduction
Materials, who specializes in computer analysis of neutron powder diffraction data obtained in experiments at KEK, communicates mostly with researchers in his own lab who prepare samples for him to analyze, and students from the University of Tsukuba who assist with his monthly experiments at KEK. Although his name is well known to superconductivity physicists throughout the world and to superconductivity researchers in Tsukuba, he knows few of them personally. Of the 110 Tsukuba researchers that he estimated work on superconductivity research related to his own research, he personally knows about 20. Most of these acquaintances he has met at national conferences in large cities. He had not met them in Tsukuba. This researcher learns of new knowledge claims by receiving preprints of journal articles. Thus he knows part of the contents of the relevant journals before they are published. He tends not to use the telephone because he thinks that talking to a person can take too long, and he does not use fax machines because they do not produce beautiful enough replicas of pre-prints. For each of his articles he mails out about 10 pre-prints to colleagues, who in turn circulate them to their colleagues. He attends about 20 to 30 lectures or meetings in Tsukuba each year, mostly because he is personally invited and feels obligated to attend. He knows of other meetings by scanning poster announcements. He does not take vacations (Izumi 1989). Other superconductivity researchers are more widely communicative, personally know virtually all of the superconductivity researchers in Tsukuba, and send out more copies of pre-prints to colleagues (Okamura 1989a). Now that the rate of knowledge claims has slowed in superconductivity research, initiatives to increase communication between superconductivity researchers in different Tsukuba institutes (and in Japan) are beginning (Okamura 1989b; Izumi 1989).
Optical film switching devices In April 1989, researchers at MITI’s National Chemical Laboratory for Industry announced the world’s first optical film which switches its electrical conductivity in response to exposure to changing light, similar to the visual neural system of animals. The switch is a Langmuir-Blodgett (LB) organic film made of 60 molecular layers
Introduction
23
which produce a conductivity six times greater than any previous thin film. A goal of the researchers is to create a superconducting thin film which will have applications in optical switches and optical memories. Probably fewer than 20 Tsukuba researchers work on LB films. About half of these researchers work at private laboratories; the rest are researchers at three of MITI’s Tsukuba institutes. Most of the privately employed researchers have occasional communication with small research teams in the National Chemical Laboratory for Industry and the Electrotechnical Laboratory (ETL), at which some of the privately employed researchers have spent from three to 18 months assisting senior investigators who research LB films. The LB film research, at least at some of the private R&D labs, benefits from state-of-the-art equipment, but the research is crippled by too few researchers, most of whom are quite young. During 1982–3, similar but separate LB film projects were started at the National Chemical Laboratory for Industry and at ETL. Two years later, a research leader at the Chemical Lab with training in applied organic chemistry and an interest in materials science telephoned who he perceived to be the top three LB film researchers in Japan. Two of these people were at universities outside of Tsukuba (the University of Saitama and the University of Kyoto). The third person, an electrical engineer, was a short walk away from the Chemical Lab at ETL. When the ETL researcher told his project director that he had an overture from a Chemical Lab researcher, the project director decided that ETL should not cooperate with Chemical Lab on LB film research. However, administrators at MITI’s central administrative office, near both laboratories, required that the researchers work together, along with researchers from a third MITI institute in the same complex, the Research Institute for Polymers and Textiles. Every day for six months the chemist walked over to the senior engineer’s office. The ETL engineer taught the chemist about LB film research methods (Mizukami, Sugiyama, and Ikematsu 1989). Two researchers from the Research Institute for Polymers and Textiles usually joined these discussions, but they were not as interested in developing highly conductive LB films. Based on what he learned from the electrical engineer, the chemist eventually isolated the chemical composition of the path-breaking highly conductive film, and produced it.
24
Introduction
Figure 7 Relational diagram among institutes in Tsukuba where LB film research is conducted Source: Kawabata 1989
This small network of researchers and their research assistants in each of the three MITI institutes was gradually augmented by researchers on paid loan from private chemical or electronics companies, mostly in the Tsukuba area10 (see Figure 7). A small amount of basic research into LB films goes on at each of these firms. The private researchers generally are not aware of their colleagues at the other private labs who are involved in LB film research, but they do know the researchers at the national institutes (Mizukami, Sugiyama, and Ikematsu 1989). The project leader at the National Chemical Laboratory for Industry knows the general
Introduction
25
approaches being taken at these private labs, but he does not know details about any of their research (Kawabata 1989). The chemist at the Chemical Lab played the central investigatory role in the LB film research (Yamagiwa 1989; Sakurai 1989) and in the LB film research communication network. He identified three ways of meeting researchers relevant to his work. He attends 60–70 percent of MITI’s monthly New Chemical Progress meetings in Tokyo, at which industrial researchers may meet national researchers. Private researchers read about his research in industry journals and telephone him to arrange private meetings in Tsukuba. Also, he attends large academic society conferences in different cities. The Tsukuba LB film researchers have not met as a group (Kawabata 1989). Extreme high-pressure vacuum chambers High-pressure vacuum chambers have numerous industrial applications, but a lack of knowledge about the molecular structure of metallic surfaces has stalled the development of the highest pressure vacuums. Many physicists thoughout the world gave up attempting to build extreme high-pressure vacuums because no known metal composition could withstand extreme high pressure. In 1980, a researcher at KEK who was still interested in developing high-pressure vacuums read a journal article about an aluminum surface. An accompanying photograph of the aluminum surface convinced him to contact the author, a specialist of material surfaces, who worked in Tsukuba at the nearby National Research Institute in Inorganic Materials. The physicist took a bottle of Suntory whiskey over to the house of the material surface scientist—a 10-second walk away—and proceeded to learn the molecular structure of the aluminum surface. Now the physicist directs a successful team of 14 researchers studying extreme highpressure vacuums (Ishimaru 1989a). The project has 50 corporate sponsors, five of which have labs in Tsukuba (Ishimaru 1989b). There are approximately 100 researchers in Tsukuba who investigate vacuums. Of these, about 60 now work on extreme highpressure vacuums, 10 of whom are senior researchers. The KEK physicist sees each of these senior researchers four to six times per year. He sends out about 50 pre-prints of his articles to peers, 50 percent of whom are in Tsukuba. Draft experimental data are exchanged by fax.
26
Introduction
The KEK physicist organizes one or two small meetings at KEK each year. These are presentations of work, followed by drinking parties. There is also one year-end party in Tokyo for his corporate sponsors which about 90 directors of small companies and research directors of large companies attend. These three stories of communication among successful researchers in the science city suggest several ideas about communication and scientific work: 1
2
3
4
Much more communication, representing many new relationships, could be expected to occur in Tsukuba than does occur. In some cases, scientists work across the street from each other, know that they investigate similar phenomena, but they choose not to meet. Some of this absence of communication can be explained by competition and rivalry. But not all of it. In some cases, the networks exist. Scientists know each other, they talk together, and they collaborate with important results for both science and industry. These networks are loosely bound, their strength deriving from interpersonal contact in informal situations. Top scientists have at least as much communication with researchers outside of Tsukuba as with researchers in the science city, and probably more. In Tsukuba, research administrators and research organizations may serve to impede as well as to facilitate communication.
SCIENCE IN JAPAN In order to understand Tsukuba, we must understand the nature of science in Japan. The word “science” as used in Japan (kagaku) has no English language equivalent. In Japan, the term broadly refers to the humanities, social sciences, jurisprudence and law, agriculture, engineering, medicine, and basic and applied natural sciences (Kondo¯ 1988). “Kagaku” refers more to the span of scientific fields than to the “scientific method” (Bartholomew 1989:4). In Japan, “Tsukuba Kenkyu¯ Gakuen Toshi” is translated into English as “Tsukuba Science City.” Yet more accurately, the Japanese term means “Tsukuba Research and Academic Town.”
Introduction
27
When Japan opened itself to the world in 1853, its people were impressed by the degree of industrial achievement in other countries. By sending thousands of Japanese students to study overseas and return home, and inviting eminent academics to teach in Japan, science entered the highly stratified and regimented feudalistic society. The development of science in Japan was most influenced by the organization of science in Germany (where 74 percent of Japan’s overseas students studied between 1900 and 1910, finding the collection of academic information particularly easy there) (Bartholomew 1989:71–2), and by the types of Japanese who were drawn into scientific careers, which was determined by precedent set during the Tokugawa period (1600– 1867). Young people from elite families involved in translation, medicine, astronomy (employed by the Shogunate), and merchants formed the class of people who were initially attracted to science (according to Bartholomew (1989:20), as many as 8,000 mathematicians may have been active during the Tokugawa period). No social strata, however, contributed as many people to the early scientific ranks as did the samurai. Later during the Meiji period (1868–1912), when Japan was actively seeking to gain scientific expertise from abroad, these same types of Japanese made up the ranks of students pursuing careers in science. So scientists came from the elite classes in Japanese society, as most of them do today. Unlike science as it evolved in Western countries, science in Japan always had a strong practical orientation (medicine dominated early Japanese science). This practicality, along with the regimented feudal social structure from the Tokugawa period, and the acculturation in the norms of Western science which Japanese students went through abroad, resulted in Japan’s approach to science today. For example, national politicians in the 1950s and 1960s believed in the ability of science to contribute to economic development, and government agencies and ministries such as the Science and Technology Agency and the Ministry of International Trade and Industry emphasized applied research (Kondo¯ 1992). By any measure, Japanese governmental and private investment in science and technology has increased dramatically since 1948, when the government dedicated itself to rebuilding the nation through science. The Science Council of Japan was created in 1949 to advise the prime minister about how to build its scientific
28
Introduction
infrastructure. The Council originally planned that basic research should be conducted in universities and applied research should be conducted in corporations, with the national laboratories playing an in-between role, especially for experiments and tests which required the use of very large or expensive research technology that neither universities nor companies could be expected to afford (Kondo¯ 1992). The actual conduct of research by these types of organizations has developed somewhat differently. Japanese universities have not been the source of all basic research in Japan. Much more basic research than applied research has been conducted in the national labs. And private industry has conducted much more basic research than expected. Japan’s share of combined research and development expenditures by the United States, the Soviet Union, West Germany, France, the United Kingdom, and Japan rose from 1 percent (last) in 1955 to 16 percent (second to the U.S.) in 1985. In 1955, 0.8 percent of Japan’s national income was spent on R&D; in 1987 R&D spending was 3.2 percent of national income (Science and Technology Agency 1988a). Japan spends a larger percentage of its gross national product on R&D than any other country (Anderson 1992:564). Although fewer researchers are employed in Japan than in the U.S., Japan ranks well ahead of other OECD member nations in number of researchers. For example, Japan had 25.3 percent of all OECD researchers, compared to 5.3 percent for France (Science and Technology Agency 1988d). Spending on research by private companies in Japan has increased rapidly since 1970, while spending by the national government for research at universities and national institutes has increased at a very modest rate. In 1986, private company spending was six times greater than either type of government spending, and this gulf widened in the early 1990s. By 1990, private companies’ share of basic research expenditures in Japan was 39 percent. Corporate R&D funding peaked in 1991, with private companies spending nearly 10,000 billion yen that year. Due to a weakened economy, Japanese corporate spending on research dropped 2 percent in 1992, and nearly 6 percent in 1993. Machinery, electronics and computers, and transportation industries reduced R&D budgets the most. Government funding of basic research at universities has increased somewhat, although its level is far lower than the amounts spent by industry. This latter development reverses a long-time real-dollar decrease in government funding for
Introduction
29
university research (Anderson 1992; Hirano 1992). Increased government support for university research is coupled with increasing corporate donations to universities, which grew from 15 billion yen in 1983 to 50 billion yen in 1992 (Normile 1994). The number of researchers employed at the national research institutes has remained constant for the past 17 years. In some cases the number of researchers has remained constant or increased slightly while research budgets have decreased. For example, at the National Institute for Environmental Studies in Tsukuba Science City, the budget for research decreased from the previous year’s funding level by 5 percent in 1983, 13 percent in 1984, 8 percent in 1985, 8 percent in 1986, and the level of funding in 1988 was 20 percent below the funding level in 1986. The level rose sharply in 1987 because the government capitulated to political pressure from the U.S. to buy foreign-made research equipment (Environmental Agency 1988a; Ueda 1989). This lack of growth in national institute employment has resulted in a manpower shortage, especially of young researchers. Although about half of Japanese researchers are under 35 years of age, these younger researchers are overwhelmingly joining industry (Science and Technology Agency 1988a). University graduates can make higher incomes by working as researchers at private companies. Status, however, remains with university and national laboratory positions. It is becoming more common for university graduates to start employment within industry and then seek a university or national lab position (Kinoshita 1993). The number of researchers employed at universities rose only modestly during the 1970s and 1980s, while the number of privately employed researchers mushroomed. In 1992, the national government committed itself to improving the dismal physical condition of Japan’s public universities after a lobbying effort by the President of the University of Tokyo, Arima Akito, led to extensive press coverage and public outcry about university decay (Anderson 1992). The successes of Japanese industry in the commercial development of technologies are well known, as are the nurturing and coordinating roles played by MITI. With the maturation of the Japanese economy has come a shared appreciation in Japan of the role of basic science as a generative source of new technologies. In 1986, the Japanese Cabinet adopted a “General Guideline for Science and Technology Policy,” which stressed the need for scientific and technological creativity, basic research, greater
30
Introduction
university-industry cooperation, and more basic research by private industry. These recommendations are being acted upon, evidenced by government programs which provide small interdisciplinary research teams of scientists and engineers with five-year grants (which may be spent in any way the principal researcher deems best) to pursue basic research (Brinkman et al. 1988), and by a new interest by Japanese corporations in basic research (Sun 1987). For example, the electronics giant Hitachi has operated its Advanced Research Laboratory since 1985. In 1992, it housed 114 researchers investigating electronic beam physics, computer software, and molecular biology, supported with an annual budget of U.S.$41 million (Hamilton 1992b). Thousands of private companies now make project-specific research grants to universities (Tokumaru 1989), and every year hundreds of Japanese firms cooperate with the Ministry of Education to allow industry and academic researchers to collaborate (Kinoshita 1993). Although some reports document impressive Japanese advances in basic research (Science 1986) and in the quality of patented ideas (Broad 1988), other analyses report that there is no evidence that the returns from basic research in Japan are relatively high (Mansfield 1988), and that the structure of funding basic research is flawed (Sun 1989). A study of citations to patents by country showed that Japanese technology is highly innovative, and that Japanese basic research is far less innovative but growing (Narin and Frame 1989). While expenditures for basic research rose at an annual rate of 8.9 percent between 1977 and 1986, the percentage of all research which is classified as basic research decreased from 16 percent in 1977 to 13 percent in 1986. The rate of applied research grew, despite the fact that private companies were conducting a larger percentage of all basic research every year. Dedication to conducting research will continue to grow in Japan. In response to a government survey, 62 percent of private company research department directors chose “insufficient in both quantity and quality” to describe their research personnel, while only 2 percent of the directors chose “sufficient in both quantity and quality” as descriptive of their researchers (Science and Technology Agency 1988d). One reason for progress in the development of science in Japan has been the relative consistency of Japanese science policy. The dominance of one political party for 38 years (until
Introduction
31
1993) and the weak presence of adversarial groups such as the military, organized labor, and environmental groups have enabled the steady accomplishment of scientific and technological policy goals (Lynn 1986). Moreover, each of the government ministeries and agencies concerned with science is staffed by career bureaucrats who are typically employed by the same ministry for the duration of their careers, even though they may be transferred among organizations. There are 21 Japanese national ministries, agencies, and organizations that spend money on science and technology. Overall science coordination between ministeries and agencies is the task of the Science and Technology Agency (STA), which reports directly to the Prime Minister’s Office and oversees a set of research organizations, such as the Research Development Corporation of Japan, which contracts with private firms in Japan for the development of certain technologies, and awards exploratory grants to senior investigators for highly promising basic research. STA is also directly responsible for massive laboratories such as the National Aeronautics and Space Development Association compound in Tsukuba, and the Institute of Physical and Chemical Research (RIKEN) in Tokyo. Approximately one-third of all government expenditures for research and development is under the jurisdiction of STA. The agency supports both basic and applied research. The Ministry of Education supports basic research at national and private universities and at university institutes, as well as at national interuniversity research institutes, one of which is the National Laboratory for High-energy Physics (KEK) in Tsukuba Science City. The Ministry did not allow national university faculty to consult or work outside of their university jobs for many years. This policy contributed greatly to the lack of technology transfer between universities and private industry in Japan, especially since the quality of research is generally higher at public universities (such as the University of Tokyo and the University of Kyoto) than at private universities (Marshall 1986). This prohibition resulted from fear by postwar academic planners that universities might again be subject to governmental-industrial control, as they were prior to World War II. This fear is still felt by academics, but university researchers more acutely feel the need for more funding for their research, by both industry and the federal government. Government funding through the Ministry of Education for university research is about
32
Introduction
one-third of the level of university research support in the U.S. (Kinoshita 1993:597). Japanese professors are currently prohibited from seeking funding from government ministries other than the Ministry of Education. The Ministry of International Trade and Industry supports basic and applied research, primarily through its Agency for Industrial Science and Technology (AIST), which operates a complex of nine research laboratories in Tsukuba. MITI’s percentage of the national science and technology budget is about half of what STA and the Ministry of Education are each alloted, yet its influence may be greater than this percentage indicates due to its close relationship with various industries. MITI, its reputation of adroitly managing Japan’s rapid industrial growth aside, has commited strategic errors in the past by investing in high-risk and unproven science (Anderson and Hamilton 1992). Many private companies sponsor projects by MITI researchers. Also, some MITI employees are transferred to STA high-level positions for a temporary duration, so MITI has some influence within STA (Lynn 1986:299). There is often intense competition between these three government organizations (STA, the Ministry of Education, and MITI) in the control of science and technology (Brinkman et al. 1988:18). Factionalism has long played a role in Japanese science (Bartholomew 1989:271–2). So the organization of science and engineering research in Japan is distinctive in some aspects, such as the large percentage of basic research being performed by private industry, the overall rapid increase in the amount of research being conducted when compared with other countries, and outstanding performance in applied research and development. For example, government agencies and professional associations in Japan play a much more proactive organizing role in transferring technologies than their counterpart organizations in the U.S. (Cutler 1989). Yet the history and evolution of Japanese science suggest that as a subsystem within the society, science in Japan plays roles that are quite similar to science in Western countries such as the United States (Westney 1987; Cummings 1990; Nakayama 1991; Schott 1994). LARGE-SCALE SCIENCE PLANNING IN JAPAN Some of the remarkable nature of Tsukuba Science City is due to the fact that it exists at all. Could a project of this magnitude only
Introduction
33
be accomplished in Japan? Is it the ability of Japanese ministry bureaucrats to coordinate large-scale projects that led to the establishment of the science city? Japanese government and industry planners have collaborated in the Fifth Generation Computer Project, the Very Large-scale Integrated Circuit Project (Sakakibara 1983; Sigurdson 1986), the nationwide Information Network System (Kitahara 1983; Ogasawara and Salvaggio 1983; Ito¯ 1986), the Technopolis Project (Tatsuno 1986; Masser 1990b), a general industrial policy (Johnson 1982; Frenkel 1984), and ongoing massive urban infrastructure developments in Osaka Bay and Tokyo Bay. Japanese bureaucrats have been acknowledged to have an unusual ability to forge shared visions and follow through on those visions. Why have Japanese government and industrial planners been effective with large-scale economic development projects? Such strategic economic development is poorly explained by the widely accepted model of Japan as a harmonious society (Benedict 1967; Noguchi 1979; O¯uchi 1984). Studies of Japanese organizational and political relations reveal intense, often ongoing, conflict (DeVos 1984; Krauss, Rohlen, and Steinhoff 1984; Saxonhouse 1986; Dore and Sako 1989)11. How does the Japanese government manage largescale economic development, and in particular science development, in the light of hypercompetition among private firms, vertically separate administrative structures at the national level which are manifest as animosity between bureaucrats employed by rival ministries,12 and condescension between industry and government in Japan? These questions are addressed in this book through tracing a developmental history of the science city.
Chapter 2
Understanding a science city
Despite the many purposes that science cities serve, such as job creation, regional development, heightening an area’s international status, even boosting the political or bureaucratic careers of key proponents, there is only one main rationale for creating any science city. Science cities exist first and foremost to increase the rate and diversify the pattern of research-based innovation.13 We know, fundamentally, what science cities will be like if and when they grow up and are “successful.” To be considered successful, a science city must exhibit higher rates and diversified patterns of collaborative research-based innovation than would have occurred had the city not been designated and planned as a science city. Moreover, the city’s rate and pattern of scientific innovation should favorably compare with similar data representing other science cities, and other research-intensive communities. Government-to-industry technology transfer, spin-off companies, technology commercialization, and regional economic development are important outcomes. But for a science city, the main rationale remains the same as that originally envisioned by Sir Francis Bacon (1624) in The New Atlantis. Scientific innovation is the key outcome measure of success or failure. Increasing the rate and diversifying the pattern of research-based innovation in a science city requires that researchers communicate with each other, collaborate on research projects together, and work creatively. These three processes lead directly to the generation of scientific innovations, which is the first of several stages in the diffusion of innovation process.
Understanding a science city
35
WHAT IS A SCIENCE CITY? A science city, if it is functioning well, is a “research community.” A research community is characterized by (1) communication among scientists and engineers, who represent an unusually large proportion of the population; (2) concentrations of scientific, technological and intellectual resources, such as laboratories, testing equipment, libraries, technicians, and university graduate students; and (3) a rapid rate of scientific innovation. Scientific innovations are based on scientific knowledge. Scientific knowledge exists in researchers, in their publications and tools, and in how they work. Characterizing research communities in this way means that to become productive, that is, to eventually generate knowledge in various forms, such communities must first and foremost function as social systems. A collection of researchers and resources may represent an area in which a lot of research work occurs, but it may not feel like a “community.” A community is a group of people who share the same social norms and characteristics, and perceive similar interests and goals. Spatial proximity is not necessary for a group of people to think of themselves as a community (Hyman 1968; Merton 1988), but it does make the establishment of a shared sense of community more likely (Hagstrom 1965; Mullins 1972). Spatial proximity helps to create a research community because it enables frequent and informal communication. If being located in the same place does lead to frequent face-to-face communication and makes it easier for researchers to engage in, research collaboration is likely to result even among diverse specialists (Kraut, Egido, and Galegher 1990). Defining science cities as research communities that are social systems means that science cities must be living environments. They must function as places where people want to live as well as to work. This indirect facilitation of research is very important. Being in touch with others and close to families, having social opportunities, feeling culturally comfortable as well as competitive are characteristics of human social systems that, when achieved, enable us to turn our attention to our work through prolonged concentration and deep thought. Of all the things that make Tsukuba Science City a distinctive place in the world, science is the most important. Tsukuba is a city which exists for science, for the accumulation of new knowledge,
36
Understanding a science city
for experimentation, and for the training, work, and results of researchers. The process of “doing science” is, for many people in Tsukuba, what life is all about. Yet even Tsukuba Science City has to function as a community through a shared appreciation of the same social norms, learned characteristics which become common to people who work there, and the perception by residents and those people who work in Tsukuba of similar interests and goals. Research communities differ on a number of attributes, such as the type of research conducted, the mix of public and private research, and the degree to which the research conducted is problem-directed (that is, applied research). For example, Cambridge, England, is world famous for publically funded basic research concerning genetics, physics, medicine, and chemistry (Segal Quince Wicksteed 1985; Saxenian 1988). The Palo Alto and San Jose area of California (known famously as “Silicon Valley”) is known for important contributions in microelectronics research, most of it conducted by small entrepreneurial private firms; this research has tended to be quite applied (Rogers and Larsen 1984). In Tsukuba, research is conducted on a broad range of topics, with a strong public orientation until the early 1990s. Most of the research does not have immediate commercial application. Applied research is the supposed function of Japanese cities which have been designated “technopolises” by the Ministry of International Trade and Industry. In Japan, there are more than 30 so-called technopolises, but until now, still only one functioning science city. COMMUNICATION, COLLABORATION, AND CREATIVITY A collection of researchers becomes a research community through communication, a process in which participants share meaning through the creation and exchange of information. Communication is not optional in the conduct of doing science. Just as modern industrial production is at its heart an information process of feedback and control (Clark and Fujimoto 1991), so too is science by necessity a communication process of sharing ideas, of analyzing others’ work, of mulling alternative hypotheses, of thinking and talking as well as doing (Pelz and Andrews 1966; Garvey 1979; Hoch 1987; Traweek 1988). As the noted observer of science, Michael Polanyi, wrote:
Understanding a science city
37
Consider…the effect which a complete isolation of scientists would have on the progress of science. Each scientist would go on for a while developing problems derived from the information initially available to all. But these problems would soon be exhausted, and in the absence of further information about the results achieved by others, new problems of any value would cease to arise and scientific progress would come to a standstill. (Polanyi 1962:54) Scientific work flourishes in those social systems in which communication is maximized; that is, social systems in which communication is uninhibited, loosely structured, interdisciplinary, and frequent (Griffith, Jahn, and Miller 1971; Griffith and Mullins 1972; Friedkin 1982; Pake 1985; Peck 1986). Interpersonally and in small groups, researchers think by talking. E.M.Forster’s aphorism, “How can I know what I think till I see what I say?” applies well to researchers who often need to talk about or in other ways “test” complex problems by communicating them to their colleagues. Then they themselves can understand the problem and suggest solutions (Amman and Knorr-Cetina 1988). By talking, writing, and visualizing, three primary means of communicating, researchers test and refine their ideas. The centrality of communication to scientific work is widely appreciated and encouraged in the design of physical work space. Industry has long recognized this truism. This is the main justification for corporate research laboratories being built with a university-like atmosphere. As Harvey Brooks (1988) has suggested, this occurs with basic research as well as with the development of still “generic” technology, in its early stages—prior to crystallization into specific product designs or the emergence of a dominant design. Communication is an iterative process of sharing meanings among participants. Communication is best studied by focusing on the relations among communicative participants over time. This perspective of communication as an ongoing relational process of creating and exchanging meaning implies that communication is the substance of social interaction, and ultimately forms the building blocks of abstract concepts such as community, social structure, and culture.
38
Understanding a science city
Collaboration is a work relationship characterized by cooperation, complementary skills, and common goals. Research collaboration is indicative of long-term relationships which are often social as well as work-related. When people communicate more than one type of information (social as well as work-related) with each other over long periods of time, strong networks and mutual influence result. Creativity is a process of making or recognizing innovations (Lasswell 1959:203). An innovation is an idea, practice, or object that is perceived as new by an individual or other unit of adoption (Rogers 1983:11). Creativity has often been conceptualized as a psychological process which is determined by factors such as intelligence or the propensity to take risks. Yet psychologists have long acknowledged a probable relationship between social relations and environmental conditions, and individual creativity (Stein 1963). At a minimum, social relations are necessary for recognizing the value of creativity (Csikszentmihalyi 1990) and of innovation. Even the most gifted individuals need stimulating social and work environments (Eyring 1959). As Michael Polanyi wrote: Scientific thought can flourish only in a scientific atmosphere. Scientists who try to settle in new centres of learning, in remote countries, know well the burden of isolation. There is no more back-breaking task than to uphold an interest in science in an environment which does not understand science and does not value it. (Polanyi 1945:322) Although the relative contribution toward creativity by psychological and social factors is not yet understood, the study of social factors is easily justified. “Social factors may be responsible for only a small part of the total variance in creative behavior, but they may account for the lion’s share of the variance that anyone can do anything about!” (Amabile 1990:72). By content-analyzing interviews with 120 scientists from over 20 corporations, Amabile and Gryskiewicz (1988) found environmental factors (which include relations among people) to be more important than personality factors in promoting and inhibiting creativity. In reviewing the results from this same research, Amabile (1988) concluded that whether considered from the point of view of individuals or of organizations, motivation is the single most
Understanding a science city
39
important factor in creativity. Resources have to be available and researchers must possess the skills to utilize those resources, but these two factors only determine what is possible. The degree to which a researcher is intrinsically motivated to pursue a task determines what will actually be accomplished (Amabile 1988:156). In other words, creativity occurs during those time periods when hard work is pursued with enthusiasm (Weber 1946) and is evidenced by self-indulgence (Csikszentmihalyi 1975). Environmental conditions which do not detrimentally impinge upon communication may motivate people (Dill 1986; Kanter 1988), but it is researchers who are encouraged to pursue their work for its own sake according to what interests them whose results are judged to be most creative (Amabile 1990). A better understanding of the sociology of research communities may assist in an eventual predictive understanding of “ecologies of innovation,” a term coined by Harold D.Lasswell (1959:217) to describe research environments which maximally facilitate creativity. Often, human beings realize their greatest creativity when they are in places that were not intended for such work. Writers, painters, and sculptors find inspiration and the ability to concentrate in beautiful natural settings, in bars and cafes, in old renovated buildings, or in the early morning hours when lying in bed. In short, creativity often comes to us when we are not looking for it and in places that were not created to stimulate it. Can a planned science city like Tsukuba support an “ecology of innovation?” PROXIMITY For decades urban planners and geographers have sought to explain the human phenomena of agglomeration, defined as the gathering of people and resources for the purposes of living, working, and recreating. Their research has addressed important questions such as: Why does agglomeration occur? What are the benefits for individuals and organizations of agglomeration? Are businesses which do not locate in a highly agglomerated place (such as a large city) at a strategic disadvantage? The impetus for much of this research was the French geographer Jean Gottmann (1961), whose pioneering work explored a chain of nearby cities as a single system. A key to the importance of Gottmann’s work was his emphasis on social networks as well as spatial structures in leading to a dynamic megalopolis.
40
Understanding a science city
What are “social networks” and “spatial structures”? A social network is the set of ongoing relations among the people in a social system. A social system may represent many types of agglomeration, including neighborhoods, towns, districts, and cities. Social networks both enable and constrain action by people. A spatial structure is the distribution of and distance among the units in a system. A unit may represent physical artifacts such as houses, offices, and roads. Successful agglomeration occurs because of the interaction of social networks and spatial structures. The interaction of social networks and spatial structures was a key theme of the sociologist Robert Erza Park (1926) at the University of Chicago in the 1920s, whose research in the slums of Chicago led to the research tradition of social geography. Park and his students mapped the networks of different types of people in the slums of Chicago. Park considered social networks to be the constantly evolving result of communication among people, and the meanings that they come to share and give to physical objects through their communication. 14 Park and his students found that people grow attached to and dependent upon both the social networks of which they are members, and to the spaces in which they live and work. Space is important to people because its structure provides an orientation or context for our behavior. People who have “no sense of place” are disoriented. To function productively as members of a social system, people need to be integrated into both social networks and spatial structures (Jackson and Smith 1984). Many economic problems throughout the world may be thought of as problems of agglomeration. Prior to the expansion and dense integration of markets and economies in the world, single organizations were considered to be the center or “locus” of production. Now, districts, cities, and regions function, through agglomeration and its interaction of spatial and social structures, as the most important loci of production. This change in central importance from the firm to the region can be illustrated in the recent history of the Palo Alto-San Jose area in California known as Silicon Valley. From 1970 to 1980, Silicon Valley was the world center for the microelectronics industry, not so much because of large, powerful firms in the Valley, but rather because of the frequency and quality of communication among people in different organizations, especially engineers employed by semiconductor and computer
Understanding a science city
41
design and manufacturing companies. By 1985, the tremendous growth of companies in Silicon Valley led to their need for professional managers. The introduction of professional management into Silicon Valley had two important results: (1) the greater control of information flows from firms, and (2) the greater provision of services in-house. As Silicon Valley companies founded offices and production facilities in other localities (many of them overseas), the external communication of Silicon Valley companies was increasingly with other branches of the same company. So external communication with other companies, especially concerning technical knowledge or informal “know-how trading,” a form of information bartering in which no money changes hands (von Hippel 1987; Carter 1989), was discouraged. The profuse personal networks among engineers which had created the dynamism of Silicon Valley dried up. Other geographic areas caught up with Silicon Valley as important centers for microelectronics research, development, and manufacturing. In the late 1980s, Silicon Valley regained its entrepreneurial spirit and high degree of interfirm information exchange. Why? Annalee Saxenian (1994) has documented how a new set of small start-up companies which produce small quantities of expensive “designer” microchips sprung up in the Valley, and dominated a new worldwide market niche. Small companies usually have few resources and are very dependent on other organizations in their business environment. So external communication is more vital for small companies than for large companies. For the region, however, external communication among people in diverse organizations is always necessary to enter a period of very rapid or synergistic innovation, such as happened twice in Silicon Valley. The expansion of communication among people from diverse organizations in Silicon Valley increased the rate and diversified the pattern of researchbased innovation there. THE HOMOPHILY-HETEROPHILY PARADOX One of the fascinating paradoxes of communication among people comes from combining research findings from studies about diffusion of innovations and social networking. Most people are only convinced of the personal value of an innovation when they learn of the innovation from a person similar to themselves, yet we stand to learn information which is of the greatest personal value
42
Understanding a science city
from people who are dissimilar to ourselves. This homophilyheterophily paradox is a key to understanding the relationship between communication, collaboration, and creativity in research communities. Homophily is the degree to which people are alike. People who are highly homophilous tend to trust each other and accord a high degree of relevance to each other’s opinions and ideas. For example, an analytical chemist who learns that another analytical chemist has begun using a new technique for determining the molecular components of fruit juice is likely to seriously consider adopting the technique herself. An innovation which is perceived as useful by one chemist will likely be perceived as useful by a second observing chemist. The degree of homophily between people is a persuasive factor which encourages the participants to adopt the same innovations. Communication among homophilous people is easier, more precise, more rapid, and has more readily understandable (short-range) rewards, than communication with heterophilous others. Thus, people primarily communicate with homophilous others, which means that we tend to know the same things and adopt the same innovations as do people with whom we frequently communicate. The perception of homophily with others is what binds together reference groups, which are real or imagined social collectivities in relation to which an individual regularly evaluates his or her own situation or status. Colleagues in the same laboratory, research institute, academic department, or geographically dispersed discipline 15 often serve as a reference group for each other through continual interpersonal communication and mentor relationships. Heterophily is the degree to which people are different. Two people who are very heterophilous tend not to readily understand the relevance of each other’s opinions or ideas to their own problems. An analytical chemist is likely to be quite hesitant to adopt an innovative technique for molecular identification which she observes a microbiologist using. Communication, let alone the decision to adopt a new idea, is difficult between heterophilous people. Heterophilous people know different things and different people. Heterophilous scientists have been trained differently, ask different questions about different phenomena for different reasons, and are slow to perceive the relevance of a heterophilous scientist’s ideas.
History of the science city concept
43
Scientists tend toward greater and greater specialization, learning more and more about less and less. Besides the disincentive of having to learn the technical jargon of another discipline in order to engage in heterophilous communication and possibly eventual collaboration, even greater obstacles to heterophilous communication exist, such as the lengthy time investment required, the risks of unappreciated results or disciplinary ostracism, and political and organizational barriers which prevent much heterophilous communication—something research administrators may perceive as unnecessary or superfluous. When applied to the study of research communities, the homophily-heterophily paradox suggests that the potential knowledge gained by a researcher from participation in interdisciplinary and interorganizational communication and collaboration is very high, but unlikely to be realized. Researchers tend to become routinized in patterns of within-discipline, and often within-organization, communication. Since homophilous researchers can often be identified by organizational as well as disciplinary boundaries, a general goal should be to establish patterns of interorganizational as well as interdisciplinary face-to-face communications among researchers. So the ideal communicative state in a research community, that of maximal innovation or synergistic creativity, is possible when there is frequent and regular interpersonal communication among many heterophilous researchers from different research organizations. This is a very difficult communicative state to achieve, but one which is most likely to be supportive of persons whom Derek de Solla Price (1961;1963) termed “research mavericks.” Mavericity is the property of making unusual associations in ideas, of doing the unexpected. Research mavericks are interdisciplinarians. They take unusual advantage of the potential innovativeness of heterophilous communication, and they exemplify the ideal of synergistic creativity, a process in which the creative whole of a group of researchers exceeds the sum of its members’ individual creative contributions. Synergistic creativity results when both the rates of innovation increase and the patterns of collaboration grow.
44
History of the science city concept
KEY QUESTIONS ABOUT TSUKUBA Three sets of questions about Tsukuba Science City organized this study. Each set corresponds to a distinct time period in the history of Tsukuba. First, why was this science city created? How did the political and economic context during the early 1960s in Japan influence the decision to construct a new city? How important was individual leadership? Was the decision to construct a science city a rational governmental process in which a time-ordered set of steps was taken in order to realize a specified goal? Second, did implementation proceed smoothly? To what degree were scientists involved in the building of the science city? How can government control of the development of this massive project best be characterized? Third, how successful is the science city? Which types of initiatives to get researchers to communicate, collaborate, and create new knowledge have been the most effective? To what degree is interdisciplinary, interorganizational communication in evidence? The following chapters explore answers to these three sets of questions.
Chapter 3
History of the science city concept
My only regret in planning Tsukuba is that I forgot to plan a cemetery. The cemeteries in the area belong to the old-time residents, and they don’t want newcomers buried in their cemeteries. In Japan we have a saying that you should be buried where you do your life’s work. So there is no place to bury the first-generation science city people. Takayama Eika (1989), President, Ko¯gakuin University In 1955, Mount Tsukuba overlooked a flat valley, the middle of which was dense with red pine trees.16 For centuries farmers did not cut down the forest because the soil was not rich enough to grow rice. For centuries, rice had been the unit of monetary exchange in Japan. Tsukuba farmers could only grow vegetables, which were not worth much. The farmers also kept the forest intact because they needed a renewable source of wood for heating their homes. The farmers could not have known in 1955 that three centuries of local poverty were coming to an end. Twenty thousand years ago, the dark blue waters of the Pacific Ocean reached all the way to the foot of Mount Tsukuba. What is now Lake Kasumigaura, Japan’s second-largest freshwater lake, was a bay. The headwaters of the Sakura River flowed from above the small town of Iwase, through the Tsukuba Valley, down to the town of Tsuchiura and into the bay. Irregular flooding of Edo (the former name of Tokyo) from the large Tone River, which flowed into Edo Bay, led the seventeenth-century Shogunate to rechannel the Tone River north of Edo Bay. Flooding in Edo was relieved, but silt from the Tone River then blocked off the saltwater inlet into Kasumigaura. The freshwater lake was born, and Mount Tsukuba became an inland mountain.
46
History of the science city concept
According to Saga (1987), during Japan’s Nara period (646– 794) and Heian period (794–1185), the Tsuchiura-Tsukuba region, complete with castle, was home to powerful rulers and fierce warriors. The area was (and is) home to two of Japan’s three highest ranking Shinto shrines, Kashima and Katori. 17 The Tsuchiura-Tsukuba region retained its national influence until the Tokugawa period began in 1600. The Tokugawa Shoguns forcibly relocated the strongest anti-Tokugawa clans as far away from Edo as possible. One of these was the Satake family, based in Tsuchiura. The Meiji period, beginning in 1868 with the overthrow of the Tokugawa Shogunate, further erased the historical prominence of the Tsuchiura-Tsukuba region. The castle fell into ruin. The area did gain prominence of a different kind, as home to Japan’s largest air base and training center for World War II’s kamikaze pilots. Girls from all over Japan came to Tsuchiura to work as prostitutes (Saga 1987). In the Tsukuba area, farmers were the object of much ridicule from residents in other valleys and from the cosmopolitan city dwellers of nearby Tokyo. The recent history of poverty and powerlessness, not its former centuries of glory, colored the perspectives of many Japanese living on the greater Kanto¯ Plain (along with howling winter winds and a large pig industry), when they thought of the Tsukuba area. This prejudice was the major reason why the Tsukuba area was the last remaining undeveloped area so near to the world’s largest city, Tokyo. And this is why the red pine forests still stood undisturbed in 1960.
PROBLEMS IN THE CAPITAL The concept of a science city arose as a convenient possible solution to six pressing problems. The three most important of these problems were unrelated to scientific or technological innovation. Science was not the reason why the Japanese government built a science city. Crowding in Tokyo, a bottleneck in the application and admission procedure to Japan’s elite universities, and a desire by conservative politicians in Tokyo to reduce the influence of liberal teachers and students, were the three main reasons why the national government decided to build a new city. Somewhat less important problems were former Prime Minister Ikeda Hayato’s
History of the science city concept
47
1960 plan to double per capita income without having any clear means of doing so, a Cabinet-level recommendation that scientific facilities in Tokyo were inadequate for enabling Japan to catch up with the West in scientific knowledge, and a shared belief by ministry planners that the era of heavy industries would soon be over. The proponents of each of these six problems were independently looking for solutions. Crowding in Tokyo had been recognized by national politicians as an important public issue for some time in Japan. People who lived in Tokyo complained of escalating land prices and property taxes, traffic jams, and air, water, and noise pollution. Constituents outside of the capital complained about the relative lack of industrial development and infrastructural improvements throughout the rest of the country. But even with the inconveniences of overcrowding, to live in Tokyo was to be part of modernity. In Tokyo’s bustling economy fortunes might be made. People from rural areas worked and saved their money to escape the poverty of the countryside; they were not about to give up their dreams of a better life (Kumata 1989). The national government repeatedly encouraged the public to stay and industry to diversify outside of Tokyo, but none of these attempts was successful. Rural development programs failed. A 1958 Tokyo Metropolitan Area Development Plan prepared by the Tokyo Regional Development Commission (adapted from the Greater London Area Plan) suggested a compromise in which areas surrounding Tokyo would be developed as satellite cities serving the metropolitan center (Onda 1988). In the meantime, Japanese continued flocking to the capital, making it the world’s most populous city in the 1960s. A government proposal released in August, 1960, suggested the wholesale move of all university faculty (about 30,000 individuals) and students (about 300,000 people) out of Tokyo to a selfcontained academic city (Wakabayashi 1985). Opponents of the proposal laughed at this idea. A subsequent plan reduced the number of universities to be moved to five. Proponents of these plans sought to lessen the additional governmental problems of hierarchical university admission procedures and aggressive political opposition. Japanese universities were numerically ranked in order by quality of education and graduate job placement. Consequently, a disproportionately high percentage of high school graduates applied
48
History of the science city concept
for entrance into the number one university, the University of Tokyo, to the detriment of other universities. Ministry officials suggested the creation of a university equal in status to the University of Tokyo by moving and combining five established Tokyo educational institutions: Hitotsubashi University, for its strength in the social sciences; Tokyo Institute of Technology, because of its excellence in science and engineering; Tokyo Ikashika University, for its medical school; Ochanomizu University, because of its female faculty and students; and Tokyo Education University, for its school of education (Kumata 1989). Each of the universities, except Tokyo Education University, rejected the idea.18 This latter school’s facilities were particularly poor, its educational charter was limited, and it was facing increasing competition from other teachers’ colleges in the Tokyo area (Ishikawa 1989). An ideological split quickly emerged on Tokyo campuses between faculty and students who opposed government intervention in academic matters (primarily those in the social sciences and humanities), and faculty and students who supported the idea of the government building an academic town (primarily those in the physical sciences and engineering). The former coalition pointed to the government domination of universities which accompanied Japan’s prewar fascism, while the latter coalition foresaw funding for new laboratories and testing equipment. Serious political protests occurred at both Tokyo Education University and at the University of Tokyo. Disruptions at the University of Tokyo were taken very seriously because of that university’s extremely high status; in some ways, its influence rivaled that of the national government. Rioting students locked professors out of some departments. In the University of Tokyo’s School of Urban Planning, which played a leadership role in planning Tsukuba Science City, some young faculty members resigned, torn between the concept of academic freedom and the lure of such a major design project (Takayama 1989). At Tokyo Education University, the faculty chose a prorelocation candidate as the university’s new president. The school administration agreed to move19 if it would be rechartered with a broader mandate as a general university. This offer suited the Ministry of Education, since Tokyo Education University was host to a particularly liberal and vocal faculty. The university was a training ground for new teachers. Conservative politicians believed that moving the university would dissolve some of the political
History of the science city concept
49
power of the National Teachers’ Union, many of whose members were also communist or socialist political party members. Rechartering the university in 1967 into the University of Tsukuba (which opened in 1973), weakened the liberal side of the faculty through the Ministry’s emphasis on changing the structure of the new university, which involved greatly reducing the power of senior professors and increasingly centralized administrative power, as well as moving the faculty members out of Tokyo (Kumata 1988). Many critics of Japanese higher education stress the rigidity of faculty advancement and insularity as major problems. So the new University of Tsukuba was portrayed by the Ministry of Education as a major experiment in lessening control at the micro level within a university, while also achieving the political goal of increasing control over higher education at the macro level of governmental education administration. A very different type of faculty, more conservative and dominated by the physical and life sciences, teaches and conducts research at the University of Tsukuba compared to many of their very liberal predecessors at Tokyo Education University.
THE NEW CITY CONTEST Faced with the political infeasibility of the massive university relocation idea, the government was forced to change tactics. On 1 September, 1961, the Prime Minister’s Cabinet directed the Administrative Agency to investigate the possibility of relocating government agencies outside of Tokyo. 20 Industry associations located in Tokyo protested against this idea, arguing that having to deal with government agencies located outside of Tokyo would be a burden. They proposed instead the reclamation of more land in Tokyo Bay. Something had to be done about the overcrowding in Tokyo. Politicians were worried that the increasing concentration of people and industry in Tokyo would stagnate the nation’s economic recovery, which would in turn jeopardize Prime Minister Ikeda Hayato’s widely publicized 1960 plan to double the national income within 10 years. This plan was the Japanese government’s most important attempt to direct the postwar economy (Pempel 1982:71). A site selection committee decided that the ideal location would be about 100 kilometers from Tokyo. This distance was far enough
50
History of the science city concept
away so that the relocation effort would be taken seriously, and it was a sufficient distance so as not to be subsequently engulfed by the advancing suburbanization of Tokyo. The selection committee identified three towns as candidates for the relocation of government agencies: Akagi, Nasu, and Fuji. In each of these three towns the national government owned large contiguous tracts of land which had been military bases in World War II. Thus, the government would not have to buy land for the new town development. Each site was supported by formidable political coalitions. Akagi had the support of Fukuda Takeo, Nakasone Yasuhiro, and Tanaka Kakuei, three Diet members who would each become prime minister of Japan. Lobbying for Nasu was Funada Yuzuru, a representative of Tochigi Prefecture in which Nasu is located. Nasu was also favored by then-Minister of Construction Ko¯no Ichiro¯.21 Fuji was a choice of several other less visible politicians, especially those from Southern Japan. Political lobbying was fierce between proponents of the three sites, and eventually became deadlocked. Finally, a compromise site, Tsukuba, was proposed. Why Tsukuba? Two future Ministers of Construction, Hashimoto Tomisaburo¯ and Akagi So¯toku, both of whom were from Ibaraki Prefecture, where Tsukuba is located, proposed the compromise site. Helicopter visits with planners and industrialists were made to each of the four contending sites (Jo¯yo¯ Shinbun 1962). Tsukuba gained rapid support as a compromise site from academics and researchers who owned homes in Tokyo because Tsukuba is about 40 kilometers closer to Tokyo than any of the other three towns, thus making for a shorter commute. 22 They did not like the idea of working in Tsukuba, but nevertheless, they considered Tsukuba to be the least of four evils (Takayama 1989). The decision that Tsukuba would be the site of a new city was made official in a 27 August 1963 Cabinet meeting. The decision was the lead story on the front page of the national Asahi Shinbun the next morning.23 Whereas the announcement was big national news, the decision did not arouse particular attention in the Tsukuba area. The local newspaper, the Jo¯yo¯ Shinbun, dutifully reported the choice of Tsukuba in a small box story one day later, 28 August.24 On the 29th, the same newspaper noted that the national government had changed the name of the project from a “government administrative city” to a “science and academic city” (Jo¯yo¯ Shinbun 1963b). Now Tsukuba would be a home to researchers, not bureaucrats.
History of the science city concept
51
Although the local media and farmers were somewhat apathetic, local politicians greeted the announcement with enthusiasm. “Our town will increase the number of employees to help with the land purchases,” said Mayor Yoshimura of Oho Village. “This is very good news for our prefecture, and thanks to all who helped us. I will work hard for the science city,” stated Governor Iwagami of Ibaraki Prefecture (Asahi Shinbun Regional Edition 1963). SETTLING ON SCIENCE The new city concept had very little to do with science. Once Tsukuba had been selected, it was designated to be a so-called “government and administrative city.” Why did the concept of a new city change to a science city, and how did this key conceptual transformation occur? In the early 1960s, the Science and Technology Agency (the STA) had begun to lobby Prime Minister Ikeda to convince him that heavy industry was a temporary means toward, and not an end state of, economic development. STA also argued that heavy industry, with its attendant pollution and massive consumption of natural resources of which the nation possessed very few, was not ideally suited to Japan, and that Japan needed a long-range plan to bolster its weak scientific research capacities vis-à-vis the U.S. and European countries. Moreover, the STA presented the government’s concentration of scientific laboratories in Tokyo as antiquated, crowded, and not conducive to world-class research (Tatsuno 1986). Because science and technology were becoming recognized as generative forces of economic development, these themes fit in nicely with the Prime Minister’s income-doubling plan. National competitiveness and pride played a role in the STA’s lobbying efforts. The early 1960s were a time of new technologyintensive manufacturing “cities,” such as Jurong in Singapore, Kaohsiung in Taiwan, and the Siberian science city of Akademgorodok, 15 miles from Novosibirsk, in the U.S.S.R., which had existed since the late 1950s (Kawamoto 1967; Birnbaum 1975; Kawamoto 1992b). The idea of creating an academic city began to change into a plan to create a science city. A political negotiation ensued between the Cabinet’s Administrative Agency and the various national ministries, with
52
History of the science city concept
the Administrative Agency asking each ministry which of its operations could be moved from the capital. No one wanted to move. Guidelines were issued which were supposed to help the ministries identify which of their operations should be moved. Officially, offices not required to be in Tokyo in order to function were targeted for relocation. These included clerical bureaux and regional offices which did not interact frequently with their parent ministry offices in Tokyo, and research, testing, and inspection units or institutes which functioned somewhat autonomously. In practice, the result was that the politically least efficacious institutes and bureaus were listed as candidates for relocation. It became clear to Cabinet members that research units of each ministry were being targeted for relocation. This focus on research institutes coincided nicely with the science suggestions by the STA. On 13 July 1962, the nation’s elite science policy body, the Council for Science and Technology, recommended the joint relocation of the national research institutes. There were about 90 national institutes in Tokyo at the time (Ishikawa 1989). A proposal was made that the land on which the old institutes stood in Tokyo be sold to private developers to pay for construction of the new institute facilities. Some individuals objected that selling the land to private developers who would inevitably build office buildings would only worsen the problems of overdevelopment in the capital, not help to alleviate it (Kawamoto 1989a). Nevertheless, revenues from the sale of land where the old institutes in Tokyo stood were used to buy land in Tsukuba (Ishikawa 1980:2). Thus, with great yet unheralded irony, the single largest rationale for creating a new city (overcrowding in Tokyo) was soon subverted by the government’s sale of much of the old institute land in Tokyo to private developers, who demolished the old laboratories and erected office buildings. In 1963, the Cabinet’s Administrative Agency asked the viceminister of each ministry to submit the names of its institutes for relocation, the amount of land they would require, and when they would move. Forming such a list was a politically difficult task, and the institutes which were listed for relocation changed a number of times, independent of any overarching theme of moving the organizations most appropriate for the establishment of a research and academic city (Kawamoto 1980). On 10 September 1963, the Cabinet gave its consent that Tsukuba be the site for the research
History of the science city concept
53
and academic city. The Capital Region Development Commission (or CRDC, which became the National Land Agency in 1974) was put in charge of urban planning for the new city, while the STA was given coordinating authority for science planning. Unlike the STA, which had the full executing power of a ministry, the CRDC had only advisory status, so the various ministries affected by the relocation plan were more eager to work with CRDC than with the STA25 (Ishikawa 1989). The city was designed to be twice the size of Akademgorodok in the former U.S.S.R. and, as the center of Japan’s future government research, was “devoted to study and research without commercial and industrial taint” (Birnbaum 1975:42). Between 1959 and 1966, the STA coordinated seven overseas study missions of areas of rapid science development in order to better plan the science component of the new research and academic city. During the 27-day 1966 trip, for example, nine study group members from the CRDC, the STA, Japan Housing Corporation, the University of Tokyo, and the Ministry of Education visited 14 planned science centers in England, France, Norway, West Germany, and the U.S. In the U.S. the group visited the National Bureau of Standards, the National Institutes of Health, the Stanford Research Institute, and Stanford University. The group’s objectives were to study the locations and conditions of research institutes, buildings and facilities, scientific equipment and laboratories, and joint research programs. The group paid close attention to how research topics were selected, how collaboration was arranged, and the disbursement of funding (Hashimoto 1967). Some of the participants of the seven science study missions helped to direct and design the new science city. The Cabinet wanted Tsukuba researchers to carry out long-range investigations of a basic research nature and not to be distracted by short-term product-oriented research. This exclusion of industry R&D may have been influenced by the shared belief that basic research and applied research were two quite dissimilar activities (Polanyi 1956). The planned new town was influenced by French suburban planning, and was named “Nouvelle Ville de Tsukuba” (NVT). The “garden city” plan featured expansive public parks and residential areas. This plan was soon revised because two ministries, MITI and the Ministry of Agriculture, had already bought plots of land which they deemed best for their facilities, and these plots conflicted with the NVT plan (Takayama 1989).
54
History of the science city concept
PROTEST AND POLITICAL APPEASEMENT Development of the science city progressed slowly, primarily because of complex negotiations between the national government and local farmers (landowners) about how much land, and which land, the government needed to purchase to construct the city. Between 1964 and 1966, protests from angry farmers (as well as from government employees who faced relocation) caused successive scaled-down revisions of the original Tsukuba Science City Master Plan by the Master Plan Core Working Group. The national government chose not to forcibly purchase the land on the basis of eminent domain in order not to aggravate local farmers. Instead, Japanese planners followed an acquisition policy of tochikukaku seiri, a land readjustment procedure for combining individual landholdings, adopted from Germany in the 1920s (Nishiyama 1986; Masser 1990a). Continued opposition from farmers might have frightened politicians of the leading Liberal Democratic Party (LDP) away from supporting the science city concept, since they depended heavily upon the voting support of farmers.26 Thus, the Japanese government had to buy plots of land on which farmers were least dependent. By the mid-1960s, fuel oil had replaced wood as the most popular source of home heating and cooking in the villages of Tsukuba, so the red pine forest became expendable. Thus, “the distribution pattern of these forests determined the basic shape of the [science] city” (Kawamoto 1980:9). Complicating the government’s attempt to purchase land in Tsukuba was student unrest and agricultural opposition to the development of the New Tokyo International Airport at Narita, which is 25 miles southeast of Tsukuba. The government’s airport plan, which also required the purchase of large contiguous tracts of agricultural land, served as a main rationale for violent student riots from 1967 to 1969. Land for the research and academic city at Tsukuba was first purchased on 9 December 1966. Despite a government objective to buy the land from farmers for the least expensive prices possible (Ishikawa 1989), during the first three months of 1967 alone the national government was able to buy 40 percent of the land it wanted. The government bought the land at a cost of 1,000–2,000 yen per tsubo (a tsubo is a unit of measure equal to 3.3 square meters). Then, when farmers in Tsukuba learned that the
History of the science city concept
55
government was placating Narita farmers by paying four to five times more for agricultural land in Narita than it was offering farmers in Tsukuba, the farmers in Tsukuba stopped selling their land. Eventually, the Ibaraki prefectural government helped finance a more lucrative deal for Tsukuba farmers by exchanging their agricultural land for farmland in Northern Ibaraki, providing irrigation, and paying relocation expenses, in addition to the sale price for their land in Tsukuba. By June, 1970, the government had title to 93 percent of the land in the future research and education district (Wakabayashi 1985:87–9). In all, the government was forced to buy 4,446 acres (or 60 percent) of the 6,669-acre central education and research district from 2,600 separate landowners.27 Construction began on the world’s largest earthquake simulator as part of the National Research Center for Disaster Prevention on 6 October 1968. Ground was broken for the construction of the University of Tsukuba on 27 March 1972, and three months later, 43 national institutes were committed to relocate from Tokyo. By 31 March 1980, each of the 43 national research institutes had been relocated and were operating in Tsukuba (Tsukuba Planning Department 1985; Bloom and Asano 1981).
TALL BOOTS, STARRY SKIES, MOSQUITOES, AND WILD DOGS Between 1970 and 1980, researchers and their families moved into newly completed apartments in Tsukuba. In the late 1970s, not everyone thought that Tsukuba was such a great site for a new city. For relocated housewives the reality of the science city was one of dirt roads and open fields. Early Tsukuba life was a shock for the migrants from Tokyo. Their memories are colored with images of tall boots, which they had to pull on to trudge through the mud, and starry skies, which they had been unaccustomed to seeing in the night glare and smog of Tokyo (Society for Recording Life Histories in Tsukuba Research and Academic City 1981). The conveniences of modern life did not exist in Tsukuba. A group of about 20 high-energy physicists were given the task of designing and overseeing the construction of a circular proton synchrotron at the future site of the National Laboratory for Highenergy Physics (KEK), beginning in July, 1971. The area
56
History of the science city concept
encompassed a one-hole golf course and a country club building.28 The rather plain country club became headquarters for living, sleeping, and planning the synchrotron. It was a common sight to see Japan’s most eminent physicists tromping through the mud carrying long sticks for protection against wild dogs, while swatting at mosquitoes. Just getting to the country club was an adventure. From Tokyo we had to take the train to Tsuchiura, and then drive on an old crooked road for a long ways. The road was so bad, with so many turns and jolts, that we called it Whiskey Road, because when you finally got to the country club you felt like you had drunk too much whiskey. (Nishikawa 1989) The boundaries of the new city of Tsukuba engulfed six existing towns and villages. This implied merger caused protracted disputes because each of these jurisdictions had its own history, identity, and local political leaders, and competed with each other for prefectural resources. They did not want to join with their rivals to create a new city. Leaders of the largest town, Yatabe, were particularly opposed to incorporation since they felt that Yatabe, as the largest town, had the most to lose. They agreed to incorporate only after the STA promised that the site for an upcoming world exposition would be in Yatabe, and that the site would become an industrial park after the exposition (Takayama 1989). TSUKUBA EXPO ’85 The 1985 International Science and Technology Exposition, entitled “Dwellings and Surroundings: Science and Technology for Man at Home,” was a landmark in the history of the science city. About 20 million Japanese and foreigners visited Tsukuba during the Expo, which lasted from 17 March to 16 September 1985. This event was originally the idea of a senior official of the Science and Technology Agency, who believed that the government should promote the positive aspects of science and technology by staging a large-scale exposition for the general public. A Tsukuba Expo steering committee was formed on 26 March 1979. The Cabinet allotted funds for planning the exposition in November, and a
History of the science city concept
57
Figure 8 Rate of increase of residential and commercial land prices in Tsukuba Science City Source: Kuramochi Real Estate, Tsukuba, 1989
government-industry management association for the exposition was formed, chaired by a respected industrialist, Doko¯ Toshiwo (Bloom and Asano 1981), who could command financial participation by leading Japanese corporations. The political and industrial thrust behind the Expo had important rhetorical goals of promoting favorable public opinion about the societal roles of science and technology in general, publicizing the emergence of the new science city, and fortifying the rationale for the massive amount of public funds which had already been spent to build Tsukuba Science City.29 Results of the Expo included an improvement of Tsukuba’s traditional image and an international recognition that Tsukuba was a place of science, helped along by visits from French President François Mitterrand and British Prime Minister Margaret Thatcher, skyrocketing land prices (see Figure 8), rapid local investment by private companies intent on building R&D labs in Tsukuba, and an increase in the pace of construction of the Jo¯ban Freeway, which links Tsukuba with Tokyo. Tsukuba Expo ’85 cost
58
History of the science city concept
about U.S. $1 billion dollars to stage, and by 1986 the Federal government had spent approximately U.S. $11 billion dollars (1.5 trillion yen) to build the science city (Tsukuba Research Consortium 1986; Kawamoto 1989c). The 1.5 trillion yen actually spent was seven and a half times more money than the Japanese government had said the project would cost in 1963 (Jo¯yo¯ Shinbun 1963a). Still, due to the tremendous land price escalation, the resulting value of the government land in Tsukuba is nearly as much as the 1.5 trillion yen that the government spent to purchase land and construct the science city. So Tsukuba Science City has been an excellent investment for the national government (Kato¯ 1992a).
SCIENCE AS A THEME IN ECONOMIC DEVELOPMENT The complete planning history of Tsukuba is complex because science was actually a post hoc rationale for the city’s development. The initial reason for building a new city was partially to solve several problems in Tokyo. Over a period of two or three years, these motivations coalesced into the idea of building an academic city. This idea, due to its political infeasibility, was replaced (some would say supplemented) 30 by an idea that survived to become a plan for a science city. Even after the idea of relocating national research institutes to Tsukuba was decided upon by the Cabinet, the plans only gradually reflected the key reason why a geographic clustering of scientific and academic resources and people could be advantageous: an increased rate of scientific knowledge creation and technological innovation might result. That science city planners in the mid-1960s came to appreciate the possible role of communication in facilitating this increased rate of knowledge creation and innovation is apparent in their study of research collaboration and joint research programs in other science cities (Hashimoto 1967), and in the geographic grouping of Tsukuba research institutes into five research zones: higher education and training, construction, science and engineering, biology and agriculture, and common use (Ishikawa 1980:6; Kawamoto 1990:3). The concentrated development of science and technology was used as a rhetorical theme for economic development by the national and prefectural governments. The prefectural government
History of the science city concept
59
lobbied hard not only for the Tsukuba project, but also for the new international airport project (which it did not get) on the grounds that the location of an international airport close to the science city would surely contribute to the growth of the science city (Jo¯yo¯ Shinbun 1963c). In this sense, Tsukuba Science City provides a ready example not only of how economics and politics can determine the development of science and technology, but also how the theme of science and technology can be used to achieve political and economic goals.31 Yet the planning and development of Tsukuba Science City also show that negative or irrational consequences may result when science and technology are used as rhetorical devices for the satisfaction of political and economic goals. Optimum conditions for research are compromised as the immediate pressure for the achievement of political and economic goals lessens. Examples of such compromises color the history of the master planning process between 1962 and 1969, and the subsequent alterations made to the fourth master plan after 1969. Revisions due to political appeasement and to practicality led to substantial changes in plans for the academic and research city. Planning centered around the seven years from 1962 to 1969, and was transformed by four iterations of creating basic plans and master plans (see Figure 9). Changes can be noticed not only in the amount of land area to be devoted to the science city (which decreased by 17 percent from the first master plan to the fourth master plan), but also in the functions which different parcels of land were to serve. For example, in the September 1963 NVT plan approximately one-third of the land was designated for green parks and another one-third was designated for residential housing. These proportions were dramatically scaled down in the subsequent plans of July 1965; February 1966; April 1967; and April 1969. In contrast, the land area devoted to research and development facilities remained relatively constant, with its proportion of the total land area increasing over time. Moreover, in the original plan of September 1962, about one-eleventh of the total land area was designated as an industrial and commercial district. This designation was excluded from subsequent plans (Tsukuba Planning Department 1985; Wakabayashi 1985:101, 108). This analysis of the history of the science city concept reveals aspects of Tsukuba that are largely unknown today. This is not the futuristic bellwether city that tourists see portrayed in glossy
60
History of the science city concept
Figure 9 (pp. 60–1) Comparison of the Nouvelle Ville de Tsukuba (NVT) plan with the first, second, third, and fourth master plans
History of the science city concept
61
62
History of the science city concept
brochures. What this analysis depicts is something much less certain, a controversial initiative which changed identities several times as people attempted to use it to solve diverse personal and public problems. Perhaps above all, the history of the science city concept is one of politics more than policy (Shima 1988). The history of Tsukuba demonstrates that the large-scale organization of science can be used for a variety of nonscientific problems. Science was a convenient and possible solution to vexing problems such as Japan’s long-term economy, overpopulation, and international prestige. The science city proved useful for weakening political opposition to the conservative ruling party. The scientific establishment in Japan was also able to use the science city concept for certain ends, such as the improvement of the scientific infrastructure in Japan. Personal careers were advanced (such as that of Ko¯no Ichiro¯). Moreover, important aspects of the city were not anticipated, such as the early exclusion of private business and changes in land values. The conceptual history of the science city is anything but a careful rational plan.
Chapter 4
Implementing the plan
Ko¯no’s idea for Tsukuba was grand, but not as large as what exists today. During those times, the ministries fought bitterly. His control was limited even though his influence was great. He couldn’t exactly direct them. Implementation was left to the planning board, not politicians. So much of what occurred was unexpected. Takayama Eika (1992), President, Ko¯gakuin University When they are conceptualized as parts of a time-ordered change process, implementation, the time during which an innovation is put to use, and routinization, the time during which an innovation is no longer perceived to be new, are the most important stages of a change process (Roberts-Gray and Gray 1983; Eveland 1987; Nord and Tucker 1987; Leonard-Barton 1988; Van de Ven and Rogers 1988; Dearing, Meyer, and Hermodson 1993). The best laid plans can, and often do, go awry. In large-scale attempts at planned change, which the science city initiative clearly was, it often occurs that individuals think that they are doing the right thing, when in aggregate their individual actions may delay, subvert, or defeat the change initiative for which they have been working. Behavior which seems rational to individuals is often irrational for their organizations, while rational actions for organizations may well be irrational in their consequences for workers, employees, or organizational members (Gerth and Mills 1946; Hilbert 1988). CHANGES TO THE MASTER PLANS The first master plan was the result of each ministry independently responding to the national government’s order that they designate
64
Implementing the plan
some of their staff for relocation from Tokyo to Tsukuba. The plan lacked overall coordination and specificity. It detailed only the facilities, area of land desired, and number of employees for each ministry. There was no attempt to integrate the functions of the ministries, nor was there reasoning behind the suggested proximity of ministry research institutes to one another. There were no precise details of common infrastructure and facilities, such as roads, landscaping, or community centers. According to Wakabayashi (1985:117–27), this first master plan was realistic in its objectives (the relocation of the units of each ministry) but overly technocratic in its means. The second master plan was very different from the first, and in many ways its antithesis. Whereas the first master plan for Tsukuba was realistic and technocratic, the second master plan was idealistic, thematic, and graphic (Wakabayashi 1985). A strong identity for the science and academic city was stressed. An integrated design emphasized the importance of interrelations between national universities, facilities within the central research and education district, and science and engineering institutes. Private universities were also to be located within the central research and education district. Just under 20 percent of the central district was set aside for private research institutes and hightechnology companies. The second master plan promoted a concept and image for a science city which would continue to orient all subsequent planning for Tsukuba. This dramatic reversal between plans 1 and 2 reflects the influence upon the second master plan of theoretically trained urban planners and architects at the University of Tokyo. The first master plan had been more the result of ministry bureaucrats. Japanese national ministries announce visions of change (Marshall 1986; Environmental Agency 1988b) which then proceed through processes of negotiation, conflict, and consensus during which these visions are reduced in their objectives and sometimes result in workable, manageable plans which are then implemented. The initial visions purposely contain little substance, so that affected “stakeholders” may then contribute to (as opposed to react to) the plan. The second Tsukuba master plan was in a very real sense the “first” science city master plan, since it followed the Japanese tradition of promoting a grand theme unfettered by technical details.
Implementing the plan
65
Figure 10 A conceptual relationship between the four Tsukuba master plans Source: Wakabayashi 1985
The third and fourth master plans were, according to Wakabayashi (1985), alterations of, and reactions to, the second master plan. The first master plan was largely discarded as a plan, although its technical details about institute facilities and their number of employees were incorporated into subsequent master plans. The second master plan’s concept of a science and academic city was retained. This conceptual relationship between the four master plans is shown in Figure 10. The idealism of a science city as expressed in the second master plan was severely compromised through the processes of negotiation, appeasement, and ministry politics throughout the development of the third and fourth master plans, and in further changes made after the fourth master plan. The prefectural government failed to buy all of the land which the master plan required (Tsuchida 1989). Government ministries rushed to claim certain parts of the new city for the location of their facilities. As a result, land originally designated for residential housing, private businesses, and public parks diminished, with 95 percent of the land in the fourth master plan designated for government research institutes. This one-sided land development plan greatly biased the population structure of the new city and caused the collapse of local
66
Implementing the plan
finance (Wakabayashi 1985). Some of the complaints by Japanese about Tsukuba in the 1980s can be traced to the lopsided land use designations in the third and fourth master plans: a lack of social opportunities, inadequate small business development, and an absence of traditional culture. These missing features were inserted in the second master plan, the framework for developing the science city, but they were subsequently pushed out of master plans three and four by the processes of institutional decision-making. The fact that the national government did not own all of the land for the project from the initiation of the science city project encouraged land price escalations and land buying competition between ministries. “This decision not to buy all of the land was one of Ko¯no’s biggest mistakes” (Takayama 1992). After Ko¯no died and late in the planning process, former Prime Minister Tanaka Kakuei who, like Ko¯no, was a very activist and detailoriented politician, sought to influence the implementation of the science city. Whereas Ko¯no’s ability to direct the ministries in the Tsukuba project was limited because the government had not bought all of the Tsukuba land, Tanaka’s ability to direct the ministries was limited because, by that time, the ministries did own Tsukuba land. Control of the land for the Tsukuba Science City project would have given either politician strong directive clout with the ministries (Takayama 1992). The idea of a science city in the second master plan exhibited a bicameral rationality. That plan emphasized a new community dedicated to outstanding achievement in science and education which would provide its residents with a higher quality of living than was available in Tokyo. The increasingly technocratic third and fourth master plans, by retaining the goal of establishing a science city but not all the means of doing so, crippled the new city’s chances of achieving social and cultural desirability. Incremental changes which seemed rational at the time to their initiators functioned irrationally, in the long term, to make the achievement of the larger goals set for the new city more difficult. For example, the virtual exclusion of all private research laboratories meant that Tsukuba would lack much of the innovative dynamism which characterizes industrial research in Japan, since 70 percent of research and development in Japan is carried out by private industry. Many of the latest trends in high-technology research, such as the formation of consortia and joint research associations, depend mainly upon private industry for their institutional members.
Implementing the plan
67
Between 1980 and 1987 in Japan, 45 joint research associations were formed. In a country in which job mobility has not been a primary means of facilitating scientific and technological communication, these associations serve to transfer technical information (Wakasugi 1988). Thus Tsukuba’s publicized image departed in important ways from its reality. Many people assigned to work in Tsukuba refused to live there. Many workers who chose to become residents complained about the socially sterile environment to which they had moved. The suicide rate of disgruntled researchers and lonely students was the highest in Japan. Work-related isolation, encouraged by traditions of interministry rivalries, continued in Tsukuba. Only since Expo ’85, with the rapid construction of many private labs and increasing numbers of small restaurants, stores, and service shops, has Tsukuba become more of a socially desirable and dynamic community. In the history of Tsukuba we see a pattern of large-scale economic development which is common in Japan. A prefectural government, with assistance from the national government, publicizes an ambitious idea to develop a region or locality. The plan is scaled down through the actual processes of building the infrastructural improvements. A massive event is held to promote the new area. Private industry, attracted by mass media attention to the event and by the general public’s awareness, perceives the area as a good investment opportunity. This sequence of economic planning and development dates at least from the promotion of Senri New Town for the 1970 Osaka Fair, and of Kobe Port Island in 1981. Since 1981, at least six expositions have been held each year in Japan. Noting the Osaka, Kobe, and Tsukuba successes in attracting industrial development, Kanagawa Prefecture planners together with planners from Yokohama City tried to reindustrialize (or “informationalize”) Yokohama by following the same sequence on an equally grand scale. A six-month-long “Yokohama Exotic Showcase”, which began in March 1989, had attracted over 16 million visitors by 1995, and featured the same types of science and technology pavilions, sponsored by the same corporations, which were at the 1985 Tsukuba Expo. Industrial parks for high-technology firms, a large new science park, and renovated city centers for Kanagawa’s two largest cities, Yokohama and Kawasaki, are already in place and nearly ready to greet an anticipated rush of private development, the potential fall-out from the
68
Implementing the plan
1989 showcase. If the number of visitors to the Yokohama fair equals the number of visitors expected, the indirect effect of the exposition may be 550 billion yen for the prefectural economy and 947 billion yen for the national economy (Foreign Press Center 1989). A total of fifteen expositions were held in 1989.32 ¯NO ICHIRO ¯ THE LOSS OF KO Individuals important in the history of Tsukuba Science City can be divided according to whether they played planning or development roles. A small group of these people exerted an influence on the city well beyond the scope of their organizational position (see Figure 11). They are viewed by others (and, in some cases, by themselves) as leaders who had strong personalities and were able to convince others that their vision of the science city, or an aspect of it, was right. The dominant figure in the history of Tsukuba Science City is known in political lore to many Japanese as an outspoken, domineering, individualistic leader. Japanese do not typically associate him with Tsukuba Science City. Ko¯no Ichiro¯ served as Minister of Construction 33 and functioned as Deputy Prime Minister34 of Japan from July 1962 to November 1964. Ko¯no died in July 1965, at the age of 67. Ko¯no’s career goal was to become Prime Minister of Japan (Ishikawa 1989; Takayama 1989).35 Ko¯no Ichiro¯ was born in 1898 in Kanagawa Prefecture. Kanagawa is spiritually somewhat the Texas of Japan (Kato¯ 1992b). Ko¯no was from a wealthy family, which sent the ambitious young man to an expensive private university, Waseda. National politicians typically graduate from the University of Tokyo or the University of Kyoto. But Ko¯no did not specialize in government. He graduated as a journalist and worked first for the Asahi Shinbun. 36 His beat was to cover the Ministry of Agriculture, in the process of which he befriended many agricultural bureaucrats, forming networks within the ministry and with bureaucrats in other ministries. Ko¯no was an excellent and sensitive listener to others. People enjoyed talking with him because he paid little attention to their social status but close attention to their ideas37 (Ishikawa 1992). The early friendships, especially with bureaucrats, would reap huge dividends later, when, as an elected official, he sought to influence those same men (Kato¯ 1992b). Elected for the first time to the Japanese Diet in 1932, Ko¯no became a protégé of Hatoyama Ichiro¯. Before the war, much of
Implementing the plan
Figure 11 Important organizations and key persons throughout the history of Tsukuba Science City
69
70
Implementing the plan
the Japanese national bureaucracy had been consolidated as the Ministry of Home Affairs. General Douglas MacArthur divided the Ministry of Home Affairs into eight ministries. In 1951, Ko¯no returned to politics with his mentor, Hatoyama. When two conservative political parties merged to form the Liberal Democratic Party, Ko¯no became part of then Prime Minister Hatoyama’s Cabinet as Minister of Agriculture (Johnson 1982). He quickly asserted himself as the most powerful person in the Hatoyama Cabinet (Ishikawa 1992), partly by raiding the ranks of other ministries to staff agriculture (he would later do the same thing in the Ministry of Construction). Whereas interministry job transfers had been common before the war, Ko¯no was the only man to do this in the immediate postwar era (Kato¯ 1992b). Then, under Prime Minister Kishi Nobusuke, he became Director General of the Economic Planning Agency. The famous 1956 Metropolitan Law, a long-range development plan for the Kanto¯ Plain on which Tokyo is located, was a product of Ko¯no’s work (Ishikawa 1992; Kato¯ 1992b), as was a 1955 plan to establish a superministerial budget bureau, and wrest control of the budget away from the powerful Ministry of Finance (Johnson 1982). The early 1960s were a time of mammoth construction projects in Japan. Tracks for the world’s fastest and quietest train were being laid. A massive highway system was being built around metropolitan Tokyo. Huge, new facilities were being built in Tokyo in preparation for the 1964 Olympics. For Ko¯no, who was in charge of the latter two projects, what was needed and wanted were more projects that would receive international attention and consequently bring him domestic acclaim—enough acclaim to be elected prime minister. International relations had always fascinated Ko¯no. The two initiatives in which he was centrally involved and of which he was proudest were negotiations to normalize Japan-U.S.S.R. relations (more than 50 percent of Japan’s Members of Parliament opposed Ko¯no’s plan, which he got passed), and Japan’s becoming a member of the United Nations. He would come to think of Tsukuba as a third most important accomplishment (Ishikawa 1992). In the early 1960s, money was not a problem. The Japanese economy was booming. If the science city project had arisen ten years later, with the oil shock and the sudden escalation of the yen,
Implementing the plan
71
the government could not have committed nearly so much money to scientific research. In the early 1960s, Ko¯no Ichiro¯ was the leader of the secondlargest faction in the Liberal Democratic Party, behind Prime Minister Ikeda Hayato’s faction. The power balance in the Diet was precarious, with neither faction able to secure the majority necessary to maintain Ikeda’s position or vault Ko¯no into the Prime Minister’s position. So Ikeda was forced to deal with Ko¯no to stay in power. Most such deals extract from a prime minister a single appointed position. Ko¯no, already Minister of Agriculture, demanded, and got, two positions: Minister of Construction, and Chairman of the Capital Region Development Commission. In addition, he demanded and received an official state residence and office in the Azabu district of Tokyo, a luxury normally reserved for the prime minister and the two leaders of the Upper and Lower Houses of the Diet. In this mansion, a spectacular example of modern architecture from the 1950s, 38 Ko¯no endeared himself to urban planners and architects by holding frequent late-night brainstorming sessions about construction and development. 39 He had a knack for identifying bright, young planners and turning them loose on grand ideas. Architects and urban planners were especially attracted to Ko¯no because he had the power to direct state funding of their plans. It was during one of his nocturnal “lab sessions” that the idea of an academic and science city was born. Ko¯no chaired the Capital Region Development Commission, a high-powered group in charge of relieving overpopulation in the capital. He formed a highly secret working group to investigate moving the capital from Tokyo to the city of Hamamatsu. Initially, this idea was not distinguished from the new town project, but he split the working group into two teams, one to investigate moving the capital, the other to investigate the establishment of a new academic city. He simultaneously had a third major project in the planning stages, popularly called the “1 million population cities project,” in which he targeted several cities in the Northern Kanto¯ Plain for industrial development. Ko¯no Ichiro¯ was the central person in coordinating interministry work on the academic and research city which eventually emerged in Tsukuba. He worked with key planners from the Ministry of Construction, the Japan Housing Corporation, the STA, and the University of Tokyo’s Department
72
Implementing the plan
of Urban Planning (many of whom were his appointees) to forge a consensus on the high value of the science city project. According to Takayama (1989), Ko¯no embodied the leadership of the project and had the power to make the various ministries, agencies, and organizations act to carry out his vision of a science city at Tsukuba. Ko¯no was one of only two postwar Japanese politicians (along with former Prime Minister Tanaka Kakuei) whom the Japanese mass media frequently described by the word “jitsuryokusha” or “man of clout” (Kato¯ 1992b). He was considered a dangerous man by bureaucrats because of his ability to remake national ministries through politicization and job transfers (Johnson 1982), his detailed knowledge of the ministries and their employees which he gained from his tenure as a journalist covering the government (Kato¯ 1992b), and his tendency to get involved in policy-making, which in Japan is typically left to ministry bureaucrats (Johnson 1982; Ishikawa 1992; Kato¯ 1992b). Ko¯no was establishing interpersonal networks of social and professional indebtedness within the ministries of agriculture and construction which would launch him into the prime ministerial position, just as Prime Minister Ikeda had successfully done within the Ministry of Finance, and Ko¯no’s chief political rival, Sato¯ Eisaku, was attempting to do as Minister of International Trade and Industry. The 1964 Olympics, staged in Tokyo and masterminded by Ko¯no, was for many foreigners their first look at the world’s new rising economic power. Ko¯no had new freeways (chiefly Ko¯soku Do¯ru) and waterways built so that the image of Japan would not be one of crowded streets and poor utility service. This publicity was essential for Ko¯no Ichiro¯. It made him a strong candidate for prime ministership when, shortly thereafter, Prime Minister Ikeda Hayato died. Ikeda’s death set off a battle between factions headed by Sato¯ Eisaku and Ko¯no Ichiro¯. Ko¯no, the constant entertainer who loved roses and art (Kato¯ 1992b) but never read books (Ishikawa 1992); Ko¯no the activist minister who, like former Prime Minister Tanaka Kakuei who had been MITI minister, liked to “dirty” his h a n d s i n p o l i cy - m a k i n g , d i d n ’ t c a r e m u c h f o r bu i l d i n g c o n s e n s u s . I n s h o r t , Ko¯n o h a d e n e m i e s . B u r e a u c r a t s a n d politicans were frightened of what he (and later Tanaka) might do as prime minister to upset the status quo of the Japanese government.
Implementing the plan
73
The Construction Minister’s tendency to act alone and his reputation for brutal political action brought with them an element of risk. Although Ko¯no’s many supporters were fervently dedicated to him (he tended to reward followers with power, not money, as Tanaka Kakuei would later do), a slight majority of the Liberal Democratic party supported Sato¯ in November, 1964. Sato¯ Eisaku would go on to dominate Japanese politics as prime minister, until 1972. So Ko¯no Ichiro¯’s greatest dream, of becoming prime minister of Japan, eluded him. In July, 1965, just two years after the choice of Tsukuba and months after his biggest political defeat, Ko¯no suddenly died.40 The Ko¯no faction of the LDP was carried on by his most identifiable student, Nakasone Yasuhiro, who would later become prime minister. But the science city project was left to a less certain fate. “If Ko¯no had not died, I have no doubt that he would have created a special Tsukuba Development Authority to coordinate the planning and development” (Takayama 1989). The various organizations still worked with the Capital Region Development Commission and the STA, which together oversaw the project, but its guiding light was gone. Able bureaucrats and planners inherited the project, but not the power to implement it.
PROJECT MANAGEMENT Certain individuals and the organizations for which they worked were particularly important in the implementation of the science city plan. The Dean of the Department of Urban Planning at the University of Tokyo, Takayama Eika, served as Vice-Chair of the Tsukuba Planning Committee, personally directed the master planning process for the city, and staffed the three-member core master planning working group with three of his students. Dr. Takayama wielded more influence over the planning of the science city than any other person. Takayama’s mentor had been Uchida Cho¯zo¯, who designed the master plan for the new University of Tokyo main campus after the old campus was devastated by the 1923 Kanto¯ Earthquake. Takayama has been a key person in linking the academic traditions of architecture, city planning, and master planning in Japan. In a country where academic movement among universities is the
74
Implementing the plan
exception, Takayama was not only the long-term Dean of the Department of Urban Planning at the University of Tokyo, but also concurrently appointed as the Dean of the rival Department of Architecture at the Tokyo Institute of Technology, until a suitable dean could be groomed for a new Department of Social Engineering. In 1980, eleven years after finishing the Tsukuba Master Plan, Takayama became president of an engineering university, Ko¯gakuin University, where he directed the master plan for a new campus. Students of students of Takayama now instruct their own graduate students on how to design master plans. For an architect and urban planner, Tsukuba represented a dream come true. The budget was large and the purpose was bold: create an entire science city from scratch. Takayama was one of the nine members of a study team which traveled to other science cities in 1966 in the U.S. and Europe to get ideas for Tsukuba. He also made repeated trips to planned communities such as Canberra, New Delhi, Brasilia, and Akademgorodok. Facilities were always first class, he noted, but none of these areas had (at that time) overcome the stigma of feeling like an artificial city. So Tsukuba was planned with spacious neighborhoods, many parks, and social and commercial districts (although these components were gradually deemphasized as planning proceeded toward development because of increasing ministerial influence). Takayama’s master plan core working group in turn coordinated with other working groups which were responsible for certain aspects of the city (traffic, agriculture, plants and parks, housing, and education), as well as with bureaucrats from the Japan Housing Corporation, which was responsible for purchasing the land in Tsukuba (Tsuchida 1989). A colleague of Takayama’s at the University of Tokyo was Yoshitake Taisui, now president of a technology and design university in Kobe. Yoshitake contributed greatly to the design of the new city, particularly to its accommodation for researchers and architectural design aspects of the University of Tsukuba. Before moving on to Kobe, Yoshitake became the Vice-President of Planning and Construction of the new University of Tsukuba in 1973 (Dohi 1989a). The Japan Housing Corporation, a public organization under the Ministry of Construction, contributed several key individuals to the Tsukuba project. Konno Hiroshi was, according to one source, “a quick thinker who made good but rather individual decisions” (Tsuchida 1989). He had two key employees under his direction,
Implementing the plan
75
Dohi Hiroshi and Wakabayashi Tokio, both of whom came to assume more responsibility for the project than their boss. Dohi became one of the two people most responsible for the design of the University of Tsukuba (the other was Yoshitake Taisui41), while Wakabayashi later documented the Tsukuba master planning process (Wakabayashi 1985; Kawamoto 1989a; Takayama 1989). Dohi and Wakabayashi worked closely with a key science planner from the Science and Technology Agency, Kawamoto Tetsuzo¯. Another key individual was Ishikawa Makoto, who began with the Ministry of Construction, moved to the Secretariat of the Capital Region Development Commission in 1967, and stayed with the Commission when it was transformed into the National Land Agency in 1974.42 Ishikawa was perhaps the key coordinating person of the Tsukuba project at the working level. Tsuchida (1989) and Yajima (1989a), members of the master plan core working group, both remember negotiating and arguing with Ishikawa, who would then go to each of the ministries which were bargaining for more (or more prestigious) land in the future science city and haggle with them. “The whole process was very patchwork and disorganized” (Ishikawa 1989). The planning of Tsukuba Science City almost certainly would not have deteriorated into such competition between ministries for land and resources if the idea’s initial leader, Ko¯no Ichiro¯, had lived longer. Minister Ko¯no had targeted both Ishikawa Makoto and Konno Hiroshi to become central members of his special Tsukuba Development Authority (Takayama 1989). As Minister of Construction, Ko¯no had direct organizational control over three of the five key organizations involved in planning Tsukuba: the Ministry of Construction, the Japan Housing Corporation, and the Capital Region Development Commission. A fourth key organization, the University of Tokyo’s Department of Urban Planning, was something of a feeder school for these three government organizations. The fifth organization, the Science and Technology Agency, maintained a close relationship with the Capital Region Development Commission, but the STA’s role was less pronounced in the planning of Tsukuba than in the development of the science city. So the planning of the science city was largely controlled by technocrats, specialists of construction, urban planning, architecture, engineering, and housing. With the beginning of actual development in 1968, the involvement of bureaucrats from the Ministry of
76
Implementing the plan
Finance and Ibaraki Prefecture also became important (Tsuchida 1989). The role of Ibaraki Prefecture in managing and determining future directions of Tsukuba has continued to grow. The prefecture has tried to expand the National Land Agency’s special emphasis on the science city to the prefecture as a whole. The regional government has had to reconcile its rhetoric about its concentration of basic research with the fact that the science city has so far created few jobs for the residents of the prefecture (Kawamoto 1992a). The responsibility for the science city has become regional, but the majority of its output benefits (students, new knowledge) accrue nationally and internationally, and not necessarily at the local or regional levels. The efficacy of the STA in coordinating development increased as more people began to realize that science and technology considerations might also be important in the building of a science city. The early involvement of the STA was due to interest in the idea of relocating research institutes by the STA Vice-Minister Umezawa Kuniomi. Although Umezawa was the top career bureaucrat in the STA at the time, he had no political power comparable with elected ministers. So Umezawa could not assume control over the research and academic city plan as could the Minister of Construction, Ko¯no Ichiro¯. Although the STA had been involved in the academic and science city idea since 1961, it was not until 1978 that the STA established a center which encouraged interpersonal communication between researchers, the Tsukuba Center for Institutes. The purpose of the Center, despite its name, was to bring individual researchers from different organizations together. The first director of the Science and Technology Agency’s Tsukuba Center for Institutes was Kawamoto Tetsuzo¯, a geographer employed by the STA who had increasingly been involved with the science city idea since 1963. Kawamoto had studied the Soviet science city of Akademgorodok in 1966 (Kawamoto 1967). In 1978 the STA named him the first Director of the Tsukuba Center for Institutes. More than anything else, Kawamoto was important in publicizing ideas and implementing policies concerning the desirability of communication between researchers at different institutes and in different fields of study. These ideas did not influence the early plans of the ministries and agencies that were required to relocate to Tsukuba. Facilities were planned without
Implementing the plan
77
many shared facilities or the sharing of expensive experimental and testing equipment. For instance, Bloom and Asano (1981:1245) report about 50 electron microscopes and 35 wind tunnels in Tsukuba, more than would be needed if resources were shared. Although the central research and academic district was planned in disciplinary zones, the dominance of planners concerned with the form of the science city precluded much involvement by planners concerned with the content of interaction in the science city. The lack of success of government-sponsored attempts at increasing scientific communication and research cooperation led Kawamoto to become the chief operating officer of a private research consortium in Tsukuba in 1983, where he has since continued many science communication programs for Tsukuba researchers. His initial appreciation of scientific communication and cooperation are now much more widely shared values among science city researchers and administrators (Ibaraki Prefecture et al. 1988). Two other individuals, both physicists, who were important in the development of Tsukuba Science City are Fukuda Nobuyuki and Nishikawa Tetsuji. Fukuda was Dean of Science at Tokyo Education University (which was rechartered into the University of Tsukuba) and a visible and vocal proponent of the need for an academic and science city. He is credited by many people with the rechartering and physical relocation of Tokyo Education University. Fukuda is perhaps the most colorful personality in the history of Tsukuba Science City. He was an able physicist with a stellar pedigree: he worked at the Institute for Physical and Chemical Research in the 1930s under the guidance of one of Japan’s most famous physicists, Nishina Yoshio. As a university student Fukuda lived in a dormitory with students who in the 1960s and 1970s became leading national politicians. At Tokyo Education University Fukuda worked with Tomonaga Shinichiro¯ (who also had trained under Nishina), who won the 1965 Nobel Prize in physics. Perhaps for these reasons, Fukuda was accepted as something of a public rhetorical spokesperson for Tsukuba Science City in spite of what many considered his unusual religious and political beliefs (Birnbaum 1975; Traweek 1988:135–6; Berton-Lewis 1989a; Ishikawa 1989; Kawamoto 1989a; Nishikawa 1989).43 Nishikawa Tetsuji directed the planning and building of the high-energy particle accelerators at the highly visible National
78
Implementing the plan
Laboratory for High-energy Physics (KEK), beginning in 1971. In 1977 he became Director-General of KEK. Perhaps because of the massive scale and commonly ascribed importance of KEK, Nishikawa was widely respected as a leader of the scientific community in Tsukuba. In 1977 he was elected by his fellow institute directors to be the first Chairman of the newly established Tsukuba Institute Directors’ Meetings, which were a first attempt to encourage cooperation and collaboration among administrators and researchers of the various Tsukuba national research institutes. He served as Chairman for five years, during which time, he says, bureaucratic wars between institutes under the jurisdiction of different ministries decreased (Nishikawa 1989). The individuals and their roles reviewed here made major contributions to either the planning or development of Tsukuba Science City, partly because they were working in uncharted territory. The familiar organizations for which they worked had yet to stake out clear spheres of organizational responsibility concerning the science city. Thus a handful of intelligent, highly motivated people were able to have major impacts on plans for the new city. Today, 20 to 30 years later, organizational influence through the implementation of bureaucratic routines is well entrenched in Tsukuba. Contemporary leaders in Tsukuba are less obvious than those of past decades, a difference that is not an artifact of historical clarity or simplification. The historical detail and dominant individuals in the planning and development of the science city are not known to young career bureaucrats assigned to direct the further development of science and scientific cooperation in the science city (Takagi 1989a). Ministry policies of rotating science administrative personnel every two years further obscure the historical individualism of the rise of the science city. Clearly, the early planning of the science city came from Tokyo. First-generation scientists, researchers, academics, and administrators assigned to Tsukuba, however, perceive that the development of the science city was largely a product of Tsukuba people and not of bureaucrats in Tokyo (Nishikawa 1989). This viewpoint is not shared by current national science administrators working in Tsukuba and Tokyo. In the words of one administrator assigned to Tsukuba: “The center of Tsukuba is Kasumigaseki [the governmental district of Tokyo]” (Takagi 1989a).
Implementing the plan
79
TSUKUBA’S ROLE IN JAPAN Although Tsukuba was the first designated “research and academic city” in Japan, it is no longer alone. Scientific and technological research resources have long been spread throughout Japan. Kansai Science City (Keihanna) has opened several complexes of buildings, and is under construction between Osaka, Kyoto, and Nara. Its planners are determined to avoid the mistakes made with Tsukuba (Anderson 1989). Included as Kansai Science City consultants are key administrators associated with Tsukuba Science City, such as Takayama Eika, Chiba Genya, and Kawamoto Tetsuzo¯. A major mistake by some Tsukuba planners was to distinguish too clearly between basic and applied research, and between governmental and industrial research. Tsukuba was to be a place of basic research as pursued by government scientists and university researchers. That research with relevance to industrial applications would be conducted in the science city was never in doubt with the relocation of nine of MITI’s Agency of Industrial Science and Technology laboratories, whose researchers maintain close contacts with industrial researchers. But the main emphasis was on a city where the general foundations of several sciences could be explored. Gradually, by the early 1980s, corporate R&D labs were sought. By the 1980s science city planners were beginning to realize that the emphasis on government-conducted basic research was doing little for economic development in the area, and that privately employed researchers working on problems of applied science and engineering had a great deal to offer to governmental and university researchers. The transfer of knowledge between basic and applied researchers is a two-way street; that is, researchers working at different levels of scientific exploration, whether the work is commercially driven or not, have valuable lessons to share. The prefectural government began to publicize Tsukuba as an advantageous location for hightechnology firms. At about the same time that Tsukuba was moving toward welcoming industrial research and small high-technology companies in the mid-1980s, a number of cities, 25 of which are designated as “technopolises” by MITI, began to emphasize to private companies their strengths as centers of scientific and educational resources. In other words, in Japan the heavily publicized concepts of “science
80
Implementing the plan
cities” and “technopolises” became difficult to distinguish. Government proponents of each concept were attempting to move onto the other’s turf. It is too early to tell whether or not this increasing rhetorical similarity will lead to a dynamic high-technology industry in Tsukuba or to improved science and university resources in other parts of Japan, or both. The number of high-technology companies in Tsukuba has indeed increased and will continue to do so. But many companies are dissuaded from considering Tsukuba as a location site because of its relative lack of infrastructure, such as labor, production facilities, subcontractors, and marketing expertise, as well as the perceived market inapplicability of much of the basic research being conducted there. As for the scientific and educational resources of the so-called technopolises, one can be even more pessimistic, and with more confidence. Begun in 1980, the technopolis program in Japan is conceptually similar to the 1962 “new industrial cities” regional development program. That plan failed for at least two reasons. Political pressure by the 561 municipalities in Japan forced the government to expand the initial number of new industrial cities from 15 to 96, thus greatly diluting the impact of the limited resources which the government was willing to contribute to the program. Also, private businesses refused to relocate out of the Tokyo-Osaka industrial corridor. As Masser (1990b) pointed out, the original technopolis concept was of two or three sites; intense lobbying forced MITI to accept 25 of the 40 prefectures which applied. Increasing the number or concentration of high-technology companies is easier than increasing scientific and university resources. Political support is usually rampant for industrial development, which creates tax revenues and employment. Political support is much more difficult to muster for scientific laboratories and universities. The pay-offs are longer range and indeterminate. Similar to the likelihood of political support, the supply of labor is a key variable in comparing the ability of Japanese cities to increase industrial, as opposed to scientific or university, resources. Japan produces about 40 percent more engineers than the U.S. The country does not have a shortage of trained engineers (Anderson 1984). Managers and engineers are the most crucial employees required by high-technology firms (Rogers and Larsen 1984:126,
Implementing the plan
81
140). Yet the supply of independently minded scientists in Japan could be larger (Gibson and Dearing 1988:232). It takes a long time to accumulate scientific and university resources (Botkin 1988). The dilution of government resources available to support economic development, the greater political feasibility of support for the development of industry rather than science or universities, and the relative lack of scientists and university faculty compared to the labor supply of engineers are all reasons why Tsukuba should be able to continue to attract high-technology firms, and why Japanese technopolises should have a relatively difficult time attracting scientific and university resources. Tsukuba will continue to become more like a technopolis, where applied research is stressed, but the other young technopolises will have difficulty expanding their resources outside of their industrial base. Thus, Tsukuba Science City will likely become and remain unique in Japan (and perhaps in the world) as a concentrated, generative source of basic knowledge, as well as fulfilling the applied research role of a technopolis. SUMMARY OF FINDINGS Here the main points of this chapter are reviewed. The number of high-technology firms in Tsukuba has increased rapidly since Tsukuba Expo drew attention to the science city in 1985. Japan’s largest corporations are now either investing in the science city or contemplating doing so, through the establishment of research surveillance offices. There are about six times more fulltime mass media correspondents in Tsukuba compared to other Japanese cities with equivalent populations. A content analysis of their stories about Tsukuba shows few stories about scientific breakthroughs or technological innovation. The large number of stories about public and private investment may mean that news stories about innovations may dominate future coverage of the science city. The impressive level of investment and future potential aside, the newness of the city, a lack of night-time meeting places, and a transportation system based on automobiles combine to make Tsukuba a sterile environment. When asked, researchers do not report meeting other researchers away from work in Tsukuba. The science city will mature with the coming of the New Jo¯ban rail line, but it is unclear whether maturity will mean becoming a
82
Implementing the plan
Box 3 Inside a private lab in Tsukuba An example of a private company which built an R&D lab in Tsukuba is Omron Tateishi Electronics. Omron is a large electronics company based in Kyoto which specializes in the development of automation technologies. 44 Omron employs researchers in over 20 separate divisions. The Tsukuba facility, which opened in 1987 and in 1989 employed 27 people (20 researchers, 3 administrators, and 4 secretaries) specializes in factory automation by designing industrial robots. Employees live in company-owned apartments and houses which are scattered throughout Tsukuba. The average age of Omron’s Tsukuba researchers is 27. The laboratory is located on a large plot in the 316-acre North Science Park, where expansive green fields and rows of young trees separate conspicuous, modern R&D labs from one another. The Omron lab is spacious, quiet, and new, and equipped with state-of-the-art electronics design and testing equipment. Strategic planners at Omron decided to build a Tsukuba laboratory because the science city is close to Tokyo and Omron is consciously trying to establish a corporate presence in the capital, and Omron identified the science city as an area of increasingly rapid technological development. Science administrators at the Omron Tsukuba lab maintain close ties to the Omron Central Lab in Kyoto by traveling to Kyoto by train about once a week. The researchers in Tsukuba may originate ideas for new research if the idea is agreed upon by administrators at both the Tsukuba lab and at its sister institute in Kyoto, the Product Development and Engineering Center. As a whole, the Omron Tsukuba lab has had little communication with its neighboring laboratories in the North Science Park. Directors of the various institutes met before each of the R&D labs began operation in order to discuss prefectural building regulations, and employees of the labs have had one joint softball game. As of 1989, the only joint-use facilities within the North Science Park were tennis courts, which employees use. Land is designated for more joint-use facilities, such as a post office and bank, but in 1989 there were not enough people in the park to justify the full-time operation of a post office or bank. Each lab has its own restaurant or eating area. Researchers at Omron’s Tsukuba lab remain rather insulated from other researchers in Tsukuba. The lab’s administrators send researchers to discussion forums at the Tsukuba Research Consortium if they perceive that the weekly lecture topic is relevant to Omron’s ongoing needs. At these forums Omron
Implementing the plan
83
researchers meet other researchers. Of course, researchers can meet each other In a multitude of other ways in Tsukuba, such as shopping, dining, or jogging, but they do not report doing so. The administrators at Omron in Tsukuba have more communication with Tsukuba institute employees than do Omron researchers. By virtue of the corporation’s membership in the Tsukuba Liaison Council (sponsored by the STA) the Omron administrators meet other administrators at the Tsukuba Center for Institutes every several months. Another way that Omron administrators meet administrators and some researchers is through attendance at monthly local business meetings (Tsukuba no yu¯be, sponsored by Ibaraki Prefecture). These meetings, which used to attract more government researchers but now attract more private research administrators and entrepreneurs, are held at the new Tsukuba Supporting Center (Kamada 1989; Onoda 1989).45
bedroom community for Tokyo commuters or becoming a more convenient and independent city of science. Landowners will be the big winners. The concept of putting many researchers together so that they might communicate more and be more innovative arose very gradually during the city planning process. The initial idea for a new city had little to do with science, and was certainly not begun on behalf of science. The idea gained support because politicians and bureaucrats saw the idea as a potential solution to the six problems of crowding in Tokyo, a bottleneck in admissions to elite universities, the power and influence of liberal teachers and students in Tokyo, low national per capita income, inadequate scientific facilities in Tokyo, and a perceived end to the era of heavy industry. The single largest rationale for creating a new city (overcrowding in Tokyo) was soon subverted by the government’s sale of public research institute land in Tokyo to private developers. So political and economic imperatives determined the development of science and technology. The vagaries of political commitment and a leadership void meant that optimum conditions for research were steadily compromised throughout the city’s planning stages. Complaints about Tsukuba in the 1980s can be traced to the rationalization of research plans during the 1960s. A
84
Implementing the plan
major mistake was the exclusion of private R&D labs from later plans. Only gradually did planners suggest that a geographic agglomeration of researchers might result in an increased rate of innovation via interpersonal communication. The dominance of planners concerned with the form of the science city precluded much involvement by planners concerned with the content of the science city. Individualistic leaders played vital roles concerned with idea formation, planning, and development of the city. The primary leader of the science city idea was a powerful national politician who died two short years after approval of the idea by the Diet. Able urban planners, engineers, and bureaucrats inherited the project but not the power to realize the vision. National ministries cooperated little in planning and developing the city. Now, the routinization of responsibility for the science city manifests itself in greater organizational control over the city, and lesser roles for individual influence. Lastly, compared to the many young technopolises in Japan, Tsukuba will continue to attract high-technology firms, becoming more like a technopolis where applied research is stressed, while maintaining its strong position as the center of governmental basic research. Japanese technopolises will not fare nearly so well in their attempts to develop strengths in scientific research. Technopolises will have difficulty expanding beyond their industrial bases unless they focus on attracting the basic R&D labs of Japanese corporations, which are growing rapidly in both their scope of research and their size of commitment.
Chapter 5
Research communication in Tsukuba
There are layers of relationships in Tsukuba. In the case of the University of Tsukuba, private companies now are allowed to donate money to specific university departments, for specific projects. That money is nontaxable. We have had about 600 such contributions. One level below those official relations are the many study groups. These groups function to develop personal relationships, but not really consulting or collaborative research. Really valuable information is shared one level deeper, between company people and professors or national researchers, in private. Tokumaru Katsumi (1989), Vice-president for Research Development, University of Tsukuba Ever since we were right and they were wrong in identifying the crystaline structure of superconducting oxides in March, 1987, researchers at the Electrotechnical Laboratory won’t talk to us. We are rivals. Izumi Fujio (1989), Research Scientist, National Institute for Research in Inorganic Materials If communication is the key to collaboration and creativity, to what degree is communication occurring in Tsukuba? Who talks with whom? What do they talk about? And what is the quality, or informational value, of their communication? For people who want to stimulate communication among researchers, these are key questions. To address these questions, the types of initiatives which have
86
Research communication in Tsukuba
been tried in Tsukuba for increasing communication among researchers were studied. Some of these initiatives were undertaken by the national government. Other initiatives sprang from private and nongovernmental organizations. Still other communication initiatives were the idea of individual researchers in the science city. Although the planning and implementation of the science city concept (as detailed in Chapters 3 and 4) were complex, the economic and political obstacles of establishing a city of science could nevertheless be overcome by government action. This is not so easily the case with the encouragement of communication among researchers. The social obstacles to creating a community in which researchers live and are creative are not readily overcome by legislation or money. A culture of frequent intellectual and social communication has to be cultivated. It cannot be bought.
ESTABLISHING A SCIENCE CULTURE During the mid-1970s, communication among researchers in Tsukuba was on a very small scale. Researchers met haphazardly when buying vegetables or when their children became friends at school and brought their new friends home (Nishikawa 1989). First attempts by researchers at re-establishing work networks consisted of telephoning previous co-workers who had also been relocated from Tokyo to Tsukuba (Kawamoto 1988). In 1979, thousands of researchers who had had little previous contact with each other moved from the anonymity of Tokyo and began to work in the quiet new science city. Most of these researchers were about 30 years of age (Kawamoto 1989b). They hoped that their new surroundings would free them from the restrictive research environments they had known in Tokyo, which were dominated by cumbersome administration and senior researchers. Relocation to Tsukuba Science City offered the opportunity to “prove themselves” to the scientific establishment in the capital, and, they imagined, to make a real contribution to the progress of Japanese science. The new Tsukuba researchers recognized that the organizational and administrative barriers which had segregated their research institutes in Tokyo had not yet developed in Tsukuba. Everything was new. The researchers sensed the opportunity to meet colleagues
Research communication in Tsukuba
87
informally in other Tsukuba research institutes in order to share ideas. Yet researchers dislike being told what to do, so when science administrators in Tsukuba tried to mandate communication among their researchers, participation and enthusiasm were low (Kawamoto 1989c). Scientists prize their personal autonomy from bureaucratic authority (Merton 1968; Rokeach 1979). Moreover, research institute administrators were more accustomed to competition than cooperation with each other, so administrative efforts at communication were halfhearted (Kawamoto 1987). Gradually, for reasons of loneliness and a feeling of rebellion against the scientific establishment in Tokyo, networks of researchers within the same or similar specialties formed. Researchers began to realize that, unlike in Tokyo, contacting researchers in different institutes (even those under the control of rival ministries) was possible in the new science city. What had been isolated groups of hardly penetrable national laboratories under the control of competitive ministries suddenly looked like new organizations whose administrations had not yet erected protective boundaries around their researchers. A few interdisciplinary meetings were formed, followed by a rash of intradisciplinary study groups.46 In just a few years, by 1982, about 80 informal groups of researchers were regularly meeting. Thirteen years later, about 150 groups existed. Informal groups are especially effective means of communicating scientific and technical information because members become familiar with each other, develop trust, and tend to perceive the relevance of other members’ work (Havelock and Elder 1987; Larsen 1988). Informal study groups have been vital to the establishment of a science culture in Tsukuba (Dearing 1989a; Dearing and Rogers 1990). Informal groups, particularly those groups which attract members from different organizations and different disciplines, lead to the creation of culture more rapidly than would otherwise be possible due to the far-reaching spread of information from discussions in the group back to the colleagues of group members, in their home organizations and disciplines. In this way, the existence of informal study groups eventually creates a science culture, which acts as the “turbocharger” of innovation and creativity discussed in Chapter 1. Other means of encouraging communication among researchers, particularly governmental or organizational initiatives, may lead to the creation of culture, but their impact is
88
Research communication in Tsukuba
usually less important (Dearing 1991). Intervention and control by science administrators of research work has been common in the history of science in Japan (Bartholomew 1989:267–9) as it has been in other countries.
INFORMAL STUDY GROUPS Informal study groups occur wherever scientists and engineers have a need for learning or contributing information which is relevant to their work, and which is either unavailable or costly to access through other means. Some of Tsukuba’s informal study groups (I surveyed them by distributing a questionnaire to the group organizers or leaders) have as few as seven members, while most study groups have about 30 members. 47 The number of active participants at any particular meeting is about half of the group’s total membership. Topics range widely, from catalysis to wind tunnels to viruses. Tsukuba Science City study group members are typically from 35 to 45 years of age. Very few researchers in their 20s or 50s belong to these groups. Most of the groups meet less frequently than once a month, and these meetings tend to be rather serious, centering around the presentation of research results or research in progress (see Figure 12). Many of the study groups have members employed at different institutes who are involved in collaborative research with each other. Of the study group leaders, 40 percent estimated that their members were or were soon to be collaborating in joint research projects. How effective are informal study groups in leading to collaborative research? Of the 30 out of 71 study groups in which collaborative projects characterized the relations among some members, 15 of the respondents thought that the projects had been inspired by first bringing the researchers together in the study group. About 80 percent of study group leaders wrote that their study groups had not changed, or had only changed a little, over time. In open-ended replies, some questionnaire respondents revealed that their study groups were changing for the worse, with comments such as “Everybody is getting too busy to meet,” or “There is a decreasing number of members,” or “It’s getting formal
Research communication in Tsukuba
89
Figure 12 Degree of formality of Tsukuba Science City study groups (n=60)
and ordinary, and discussion is less hot.” The key characteristic desired of new members is that they be energetic and interesting. Researchers mainly see the goals of study groups as generating new ideas, engaging in conversation with colleagues about research problems, and thus stimulating their own research (see Figure 13). Study group leaders overwhelmingly considered their study groups to be either fairly successful (79 percent) or very successful (17 percent) at achieving these goals. In 1979, the first study group was formed. By 1982, the number of study groups had already reached about 80. Why did the number of study groups grow so rapidly between 1979 and 1982 and then slow considerably after 1982? The answer is not clear.48 Perhaps this number more or less covers the intellectual breadth of research topics in the Tsukuba area. Another, more plausible, explanation is that at the time when most of these study groups were formed (1981–2), the majority of the research population had just moved to the new science city. By 1980, 36 national institutes were open and at least partially operative. The organizers of the various study groups were invariably young researchers in their 30s who wanted to give the
90
Research communication in Tsukuba
Figure 13 Multiple-answer responses by Tsukuba Science City study group members show what members expect from their participation (n=101)
new science city a distinct identity apart from Tokyo. For them, creating and leading a study group which purposely crossed organizational boundaries and invited participation from university professors and privately employed researchers was a new and somewhat radical concept. They were innovators in adopting a new type of scientific communication. The organizers (many groups had several organizers) closely identified with their groups, and they took pride in their accomplishments of communicating in a new way. Some of the groups met in the Tsukuba Center for Institutes, a neutral meeting place, since members were from different institutes and some were under the jurisdiction of different ministries. It is conceivable that a large degree of the satisfaction and rewards derived from participating in a research study group results from creating the group. To start a new research study group now in Tsukuba certainly would not be a radical concept. Researchers trying to do so might be socially rebuked by their senior researchers, who would prefer that the younger researchers joined the appropriate established study group. “Everybody likes to have their own club. I think that there are too many study groups,” said Fujita Takeshi, a study group member and the director of an
Research communication in Tsukuba
91
academic liaison program for Eizai Pharmaceutical Company in Tsukuba. EVOLUTION OF FIVE STUDY GROUPS Case studies of five informal groups of researchers who meet regularly in Tsukuba were conducted.49 These five groups were (1) the Tsukuba Applied Geoscience Society, (2) the Genetics and Bioengineering Study Group, (3) the Thermo-fluid Dynamics Seminar, (4) the Heat Transfer Colloquium, and (5) the Tsukuba Chemical Science Club. The goal of the case studies was to understand the roles of information, communication, and change in the groups, and to provide more detailed insight about when this means of bringing researchers together is most effective in leading to communication and collaboration, and when study groups are not so effective. The Tsukuba Applied Geoscience Society (TAGS) was the first study group organized by researchers in the science city. On 12 June 1979, the group held its first two-hour monthly meeting. In June 1989, the group held a 10-year anniversary meeting and party, highlighted by the publication of a book which includes the history of TAGS, field trip summaries and maps, and notes on lecture meetings. The group was initially formed by young researchers at the National Research Center for Disaster Prevention, the Geographical Survey Institute, the University of Tsukuba, the Public Works Research Institute, the National Research Institute of Agricultural Engineering, and the Forestry and Forest Product Research Institute. These researchers shared an interest in applied geoscience, which combines studies of geography, vegetation, surface soil science, geology, and hydrology in a focus on the prevention of natural disasters. A geographer from the National Research Center for Disaster Prevention who had become director of the Tsukuba Center for Institutes, Kawamoto Tetsuzo¯, initiated the idea and became manager of the fledgling group. Now the study group claims over 200 members at 12 institutes under the jurisdiction of six ministries. Table 2 lists the number of group members belonging to each institute. There is a wide discrepency between the total number of TAGS members and the number who actively attend the monthly meetings. Average meeting attendance is between 10 and 20 percent of the
92
Research communication in Tsukuba
Table 2 Number of members in the Tsukuba Applied Geoscience Society (TAGS) study group, by institutional affiliation
Note: Table is based on 1989 data
group’s membership, and the average age of active participants is over 40. A typical meeting has one or two speakers and a questionand-answer period. Communication through TAGS has resulted in one collaborative research project funded by the STA. This project concerned how to best predict slope failures due to heavy rainfall, and involved 20 researchers, some of whom the project director met through his involvement in TAGS. Active members of TAGS, nearly all of
Research communication in Tsukuba
93
whom have been participants for 8 to 10 years, cite several nagging problems with the study group. The group has lost its interdisciplinary nature, meeting attendance has decreased, there is a lack of interest in the group by young researchers, meetings have become repetitive and sometimes boring, and socializing after the monthly meetings has decreased, they say. The annual field trips to areas of geologic interest still attract many members, but the monthly meetings do not. Why is TAGS faced with these problems, how serious are they, and how might they be remedied (if indeed they should be)? TAGS initially attracted many monthly participants of diverse specialties partly because TAGS held the only geoscience meetings in Tsukuba. Gradually, however, competition arose in the form of other, more specific, geoscience study groups. For example, a group specializing in tectonics, the movement of the earth’s plates, began. Another group formed to discuss basic geology. Thus, members who had made TAGS such an interdisciplinary group began to attend other meetings which they perceived as being more relevant to their specialized research work. This winnowing of its original eclectic nature left TAGS increasingly narrow in topics of discussion, because only those researchers most interested in applied geoscience continued to regularly attend meetings. Thus, the group has lost membership at its disciplinary extremes, no doubt making the meetings increasingly predictable and similar, and hence less interesting to many. TAGS was originally formed by young researchers with a common interest in applied problems, partly because they perceived that older geological researchers in Tokyo did not concern themselves with the real world. Japan is a country in which geological activity, extreme weather, earthquakes and flooding are commonplace and difficult to ignore. But now, since the research institutes in Tsukuba emphasize basic research rather than applied research, most researchers perceive the applied focus of TAGS to be irrelevant to their work. These reasons help to explain why membership has decreased and become more homophilous, and partly, but not wholly, explain why young researchers in Tsukuba are not participating in TAGS. The founders of the group were in their early 30s in 1979 when the group was formed, and they remain the leaders of the group today. Younger researchers hired directly from graduate school to work in Tsukuba institutes since 1979 have not been incorporated into the
94
Research communication in Tsukuba
TAGS hierarchy. The satisfaction, responsibility, and control which the early founders enjoyed has not been offered to younger researchers. Younger members have only been encouraged to attend as participants, not leaders. Moreover, because there are no young leaders in TAGS, young attendees feel socially constrained against speaking critically. Negative comments might be perceived as disrespectful of well-known, senior researchers upon whom their own careers partially depend. Participants perceive the topics of TAGS meetings as repetitive or boring partly because the range of topics has narrowed but also because of the structure of the meetings. Lectures are given from one narrow end of a long, rectagular room, so some listeners are 10 or more chair rows away from the speaker. There is little time for questions and discussion afterward. There is time for socializing after TAGS meetings since they end at 9 p.m., but the familiarity among active members means that they already talk to one another often, so there is less need to socialize. Each of the problems associated with TAGS seems to stem from the similar root cause of an aging leadership which has failed to phase itself out by encouraging younger leaders. TAGS may become an old boys’ club. Young, active researchers who are well known in their specialty and not necessarily well known to active TAGS members would most likely attract, through their own personal networks, more young members, who in turn could restore the organization to its original interdisciplinary status. A few graduate students from the University of Tsukuba attend TAGS meetings, but their participation is of limited duration. Because the hiring of national researchers is controlled at the national level (theoretically, all applicants for jobs with the national government must take a standard set of written tests), the university does not serve as a feeder school for researchers at the Tsukuba national laboratories. Some graduate students are hired as part-time research assistants, but it is extremely rare for these students, even if they are exceptionally good researchers, to be hired full time by the institute. Thus, groups like TAGS are faced with a constant turnover of a small number of young people, with little chance that any of these students can continue in the group and eventually assume leadership positions. Both the Heat Transfer Colloquium and the Tsukuba Chemical Science Club are examples of thriving study groups. Each group
Research communication in Tsukuba
95
has strong ties to private researchers, and engineering and chemistry professors, as well as many members from the government research institutes in Tsukuba. Organizers of the groups consider their membership diversity to be a major reason why their members are enthusiastic about the study groups. Collegiate friendships often serve as the relational basis for study group organizers when they search for industry and academic researchers to join their group. In both the Heat Transfer Colloquium and the Tsukuba Chemical Science Club, the placement of graduate students with private firms is a main outcome of the interpersonal relationships between professors and industrial researchers. Privately employed researchers join the groups partly to meet professors and secure an inside track on the availability of the best graduates; professors join the groups partly to meet employers and to place their students with the best companies. Members of the Heat Transfer Colloquium are mostly engineers who study the diffusion and stability of heat to better understand how rapid-functioning devices such as supercomputers can be developed, or how heat fluctuations affect aerodynamics. Colloquium members include researchers who study fluid dynamics, heat transfer, and thermal dynamics, and come from a variety of Tsukuba institutes (see Figure 14). The colloquium was founded by, and is chaired by, Professor Nariya of the University of Tsukuba, but organized by Yabe Akira, a researcher at the Mechanical Engineering Laboratory. About 100 flyers are sent out for each meeting. An average of 40 researchers attend, 50 percent from national labs in Tsukuba, a further 25 percent from the University of Tsukuba, and about the same amount again from private companies. An advisory committee of 61 researchers from 31 research laboratories and universities oversees the operation of the Colloquium. The Colloquium, which is organizationally based at MITI’s Mechanical Engineering Laboratory and at the University of Tsukuba, enjoys a particularly close relationship with Hitachi’s Mechanical Engineering Research Laboratory, in Tsuchiura, a small city next to Tsukuba. In 1980, when the MITI lab moved from Tokyo to Tsukuba, the Hitachi lab moved to Tsuchiura. The link was strengthened when a well-known senior researcher at the MITI lab, Dr. Nakayama, was hired by Hitachi. The Colloquium’s organizer, Yabe, had been Nakayama’s student. This relationship and
96
Research communication in Tsukuba
Figure 14 Institutes in Tsukuba and Tsuchiura and the number of their researchers who investigate heat transfer (institutes in the box on the right play a central role in the study group)
others means that the Colloquium’s business connections are quite strong. The study group has used these connections to shift the locations of its meetings. Each meeting is held at a different lab, and includes the inspection of experiments and equipment following a research presentation and discussion. Because of this “hands-on” involvement of Colloquium members and access to their colleagues’ facilities, the close proximity of the meeting sites, and a lack of similar study
Research communication in Tsukuba
97
groups in the Tsukuba area, attendance remains high. Participants sometimes go to a bar to drink after the meetings. The average age of participants had increased from about 30 when the Colloquium was founded to about 40 in 1989, but young members are not so much sought as are engineers who are doing high-quality research. Graduate students are not actively recruited. Because of this emphasis on prominent, active researchers and their unusual concentration in Tsukuba, the Colloquium is able to partially set the research agenda in Japan concerning heat transfer studies. This ability is especially true in the field of mechanical heat transfer. In Tsukuba there are “two camps” of heat transfer researchers. Those heat transfer researchers trained at university science departments tend to study atmospheric heat transfer. Heat transfer researchers trained at university engineering departments tend to study mechanical heat transfer. The Colloquium concentrates more on mechanical heat transfer research. The Colloquium’s range of topics has expanded along with the growing field of heat transfer. Partly in recognition of the group’s agenda-setting role, the national Japan Society of Mechanical Engineers underwrites the Colloquium’s activities, in addition to funding from the Tsukuba Expo Memorial Foundation. In keeping with this leadership role, the Heat Transfer Colloquium occassionally offers two-day educational retooling programs to update researchers from other parts of Japan about heat transfer research. Several collaborative research projects have resulted from participation in the Colloquium, such as jointly determining how to measure the conductivity of extremely thin films at Hitachi and the National Research Laboratory of Metrology. More collaborative research might have resulted among heat transfer researchers at different Tsukuba laboratories if there were not such a high degree of duplication of experimental equipment. A number of the national institutes have the same type of equipment. More numerous are collaborative research relationships between Tsukuba heat transfer researchers and researchers outside of Tsukuba. For example, the Building Research Institute has ongoing joint research projects with the California Institute of Technology, the University of Hokkaido¯, and the NIST in the United States (formerly the National Bureau of Standards). More than any of the other informal study groups reviewed here, the Tsukuba Chemical Science Club has a strong industrial
98
Research communication in Tsukuba
membership. The Club was predated by another “Present and Future State of Chemistry” group begun by Professor Oai Shigeru of the University of Tsukuba in the mid-1970s, when Japan was in the midst of the oil shock that upset the national economy. Many chemical companies had moved from the Osaka area and established manufacturing facilities north of Tokyo, in Kashima, a heavy-industry zone not far from Tsukuba. Oai had previous relations with these chemical companies because he was from Osaka City University. A colleague of his at Osaka City University, Professor Ando¯ Wataru, also came to the Department of Chemistry at the University of Tsukuba. Together, Oai and Ando¯ persuaded several Osaka-based firms which had not yet established research labs in the Tokyo area to locate in Tsukuba. The firms came to Tsukuba because they wanted personal access to Kanto¯ area students, they wanted to establish working relations with senior and middle-level researchers in the national institutes, and land prices in Tsukuba were relatively inexpensive. In 1986, Ando¯ founded the Tsukuba Chemical Science Club, which closely followed the mandate of the previous group, which was to develop personal relationships between industry, government, and university chemical, biological, and materials science researchers. The Club has 173 individual members and 40 corporate members. Individuals pay an annual fee of about U.S. $24 to join the club, and companies pay $160. The Club also receives some money from the Tsukuba Expo Memorial Foundation. A board of advisors is made up of three professors from the University of Tsukuba, one national researcher each from the National Chemical Laboratory for Industry and the Fermentation Research Institute, and employees of five private companies in Tsukuba. Ando¯ (from the University of Tsukuba) chairs the meetings, assisted by two co-chairs, one from industry and one from the national institutes. The Club meets three times per year, in May, September, and December. Each meeting features two one-hour presentations, each of which is followed by a brief question-and-answer period. A dinner party follows the last speaker. Average attendance varies from 40 to 70, 30 members of which represent a core group of consistent attenders. Half of the participants are national researchers or university professors. The other half are company researchers. Professors from the University of Tsukuba, the University of Tokyo,
Research communication in Tsukuba
99
Saitama University, Ibaraki University, and Sophia University attend the meetings. No students attend the Club meetings. Attendance fluctuates in response to the advertised topics of discussion. Unlike most of the study groups that we investigated, the topics of the Chemical Club presentations tend to be diverse and are not research reports, in an attempt to interest more people. Examples of past lectures are the chemical theory of the origin of life, the Bhopal Dow Chemical tragedy, and Asian medicine. Meetings are always held on Fridays, and tours of the laboratory facilities of national research institutes are a regular feature. The Club strives to be as personal as possible, and does not encourage the establishment of official relations between organizations. “The chemical companies are too closed to release their priority information in an official way,” said the leader of the Tsukuba Chemical Science Club, Ando¯ Wataru. “With chemicals you have to be very careful about what you say, because the knowledge of a new product or formula is not very complex, as it is in electronics.” A problem with the membership of the Club is that privately employed researchers are typically assigned to a particular lab for only two to three years, and then are transferred to one of the firm’s other laboratories. These short assignments mean that there is a constant turnover in the corporate individuals who attend Club meetings. So publicity and recruiting for the Club has to continue at the private companies. Privately employed researchers will not come to the meetings unless they have the permission of their supervisors. A lengthy newsletter is published three times per year, to coincide with the meetings. The newsletter is important not only for introducing members to one another and giving them a common knowledge about which they can talk, but also as tangible evidence for the many corporate researchers that their time and membership fees have some return (the newsletter is important to persons in company “accounts payable” departments). A major goal of the Tsukuba Chemical Science Club has been to increase the communication of knowledge from industry members to researchers of the national laboratories. “Our public members do not know, as a rule, what is going on in the private labs,” said Fujita Takeshi, the director of an academic liaison program with Eisai Company. In general, private company researchers believe that they know what kinds of research are going on in the national laboratories
100
Research communication in Tsukuba
because of the open nature of the national institutes, the personnel exchange programs that they can take part in, and employment raiding by private companies of senior researchers at the national laboratories, which has decreased the quality of research at the national laboratories. On the other hand, national institute researchers feel that they know enough (which they admit is very little) about corporate research projects, because corporate researchers from the private labs have worked for them as research assistants, so the national researchers think that they know what the private researchers know and are interested in, and they are not convinced that private laboratory research is very relevant to their own work. Fujita, with the Eizai Company, has used his participation in the Tsukuba Chemical Science Club to meet organic chemists, biochemists, medical doctors and doctors of pharmacy, and life science researchers. These contacts have resulted in at least three collaborative research projects for Eizai: one with researchers at Hitachi Kasei Company, one with researchers at the National Chemical Laboratory for Industry, and a third joint research project with National Chemical Laboratory for Industry researchers and researchers at Japan Oil and Fat Company. Centered around the topics of combustion, turbulence, fluid mechanics and aerodynamics, the Thermo-fluid Dynamics Seminar was created for the purpose of exposing the work of young researchers to their peers and more senior researchers.50 The seminar has followed the career of its leader, Professor Tsuge Shunichi, who was trained and taught at the University of Tokyo where he cofounded the seminar. After 10 years as a researcher with the National Aeronautics and Space Administration (NASA) in the United States, Tsuge returned to the University of Tsukuba in 1979, from where he reestablished the Thermo-fluid Dynamics Seminar. Tsuge’s seminar is now operated by two of his former students, both of whom work outside of Tsukuba. The seminar meetings stress theory and methods, so the meetings have a very academic (as opposed to strictly problem-solving) atmosphere. The seminars are usually followed by either a cocktail hour or dinner. Announcements of meetings are sent to 90 people; about 15 to 30 attend. This number has remained constant over time. A quarter of the participants are graduate students from the University of Tsukuba, Ibaraki University, and Chiba Science University. Most attenders are national researchers in Tsukuba (see
Research communication in Tsukuba
101
Table 3 Number of Tsukuba public researchers who study thermofluid dynamics, by institutional affiliation
Note: Table based on 1989 data
Table 3). Less than 10 percent are privately employed researchers.51 In several cases presentations by students have led to public recognition of their work, and job placement. The seminar has not led to collaborative research. A senior researcher at the National Institute for Environmental Studies in Tsukuba who studies thermo-fluid dynamics went to several of the seminar’s meetings when it first began, but has not attended the meetings since. He cited three reasons why he does not attend the meetings: he no longer is on the mailing list for seminar date, time, and topic announcements; he is too busy; and most importantly, his specialty, while in general similar to the seminar’s range of topics, is in fact quite different in theory and methods. The intellectual connection is much stronger between this national institute and the University of Tokyo than with the University of Tsukuba, he said.52 The Genetics and Bioengineering Study Group has halted all its activities because the research environment in Tsukuba has changed, making the existence of the group less necessary.53 When it was active, the group attracted hundreds of attenders to its meetings (see Table 4). This group’s rise and fall seem to have paralleled the excitement of developments in the field of genetics in the early 1980s, as well as to have corresponded with the newness and subsequent adolescence of Tsukuba Science City. Clearly, there are times within disciplines and at points during research paradigm formation when informal study groups play important roles in diffusing scientific knowledge. But informal study groups can outlive the usefulness of their contributions to a set of researchers.
102
Research communication in Tsukuba
Table 4 Cumulative number of attendees, by organization, at eight of the Genetics and Bioengineering study group meetings
The need for the Genetics and Bioengineering Study Group, which was very strong at first, eroded when the Ministry of Agriculture, Forestry and Fisheries opened the National Institute for Agrobiological Resources in December 1983. This center brought together most of the researchers from various institutes who had been attending the Genetics and Bioengineering Study Group
Research communication in Tsukuba
103
monthly meetings. Also, private genetics research blossomed worldwide as the profit potential of biotechnology became apparent. Trade journals and academic journals proliferated, making knowledge about genetics easier to obtain. The demise of the Genetics and Bioengineering Study Group in no way reflects a lack of local research in plant molecular biology. Japan’s Rice Genome Research Program has its own five-storey building near the agricultural complex of buildings, in south Tsukuba. The Rice Genome Research Program has 160 corporate members who contribute funding to the program, in exchange for information about disease- and stress-resistant rice. The basic research goal of the program, however, is a complete sequencing of the rice genome (Stevens 1994). As more researchers moved to the science city, informal study groups tended to become less interdisciplinary and to become more intradisciplinary. It is easier to interest researchers in initiatives which are intradisciplinary. Researchers recognize the value of interdisciplinary initiatives, but they tend not to participate in them. Leaders were important in the establishment and operation of the study groups investigated here. Each group was formed and guided by a strong-willed, energetic researcher, several of whom are known as excellent researchers by their peers. The value of the knowledge communicated among group members differs according to whether or not the groups included researchers from private companies. In study groups without industrial members, or in which the topics of discussion are perceived to have little industrial application, the value of the knowledge discussed is relatively high. In study groups with industrial members, or in which the topics of discussion are perceived to have industrial application, the knowledge discussed is less valuable. For example, both the Heat Transfer Colloquium and the Tsukuba Chemical Science Club have many members who are industrial researchers. Because of the competing interests among company researchers, the degree of specificity of the engineering and, especially, chemical knowledge discussed in the study groups is slight. Membership in study groups bridges the gap between very formal communication, such as the written reports that academic and government researchers prepare based on company-funded research grants to Japanese universities and government research institutes, and very personal communication, such as paid, private consultations among professors and
104
Research communication in Tsukuba
government researchers on the one hand and company researchers on the other hand. To become an intimate member of the more valuable academicgovernment-industry personal networks requires a means of personal entry that the very formal institutional relationships do not provide. Research study groups, which bring people together, albeit in a somewhat formal atmosphere, provide the social opportunity for entry into the more valuable, very personal communication among researchers. GOVERNMENT EXCHANGE PROGRAMS IN JAPAN Several nationwide government initiatives to increase communication and innovation among researchers in Japan are sponsored by the Science and Technology Agency or coordinated by the Ministry of Education, Science, and Culture (Monbusho¯). Monbusho¯, for example, oversaw nearly 1,300 university-corporate collaborations in 1992. Most of these collaborations are modest, with an average budget of only U.S. $36,000 (Normile 1994). These programs impact on research communication in Tsukuba since many Tsukuba researchers take part in or lead these projects. The STA provides “special coordination funds” for interdisciplinary applied and basic research projects in which international joint research is strongly emphasized. Project proposals may be submitted by a team of researchers with different organizational affiliations, or projects may originate from the STA and be advertised to attract applications from interested researchers. These projects are funded for three- to five-year periods. The total amount of funding for these projects has doubled several times since the first of them began in 1981. In a brochure publicizing these projects, seven of the eight featured project directors worked in Tsukuba Science City (Science and Technology Agency 1988c). The STA also funds a select number of highly publicized research projects grouped under the Exploratory Research for Advanced Technology Organization (ERATO), which is administered through a public corporation, the Japan Research and Development Corporation (JRDC). ERATO began in 1981. Each project is proposed, organized, and administrated by a senior scientist who has autonomy in deciding how the funding (between U.S. $2 million
Research communication in Tsukuba
105
and $3 million per year for up to five years) is spent. Groups of 10 to 15 young researchers work on each project. Five of 15 research projects were based in Tsukuba in the late 1980s. The STA and Monbusho¯ operate government institute-industry exchanges and university-industry exchange programs. The number of private company researchers who have collaborated in year-long university-based research projects has greatly increased, from about 60 researchers in 1983 to 550 researchers in 1988, and into the thousands in the early 1990s. For both the researchers and the organizations involved there are distinct advantages in collaboration. Privately employed researchers tend to be young graduates without much experience or a strong theoretical background. They work with other young researchers, and their projects are limited by the projected commercial feasibility of the research. By working as assistants in university departments and national research institutes, private company researchers gain experience and guidance from senior researchers. Their firms benefit by tapping the knowledge of senior researchers, which may then be reflected in more innovative work by their employees. Both university and national research institute researchers tend to be less informed about the latest instrumentation and apparatus, which private company researchers take for granted. The turnover time for lab equipment is more rapid in private companies (Science and Technology Agency 1988a). University and national research institute researchers have less funding with which to conduct research. Their main dilemma, however, is a manpower shortage. Wages are comparatively low, so it is difficult to attract the brightest young researchers. Both university and national institute researchers use university graduate students as research assistants, but young privately employed researchers tend to be brighter (those selected to assist senior researchers in universities and national institutes are the cream of a high-grade crop), they have some experience, and they have their salaries paid by the participating companies. The universities and national institutes benefit by “retooling” the knowledge of their faculties and employees about state-of-the-art equipment and technologies, and they save money by not paying salaries. It is a highly symbiotic technology transfer system.
106
Research communication in Tsukuba
Box 4 Communication at Tsukuba Research Consortium The Tsukuba Research Consortium (TRC) is perhaps the most publicized nongovernmental organization in the science city. Much attention has been directed toward the Consor tium because its organizational structure is innovative, the Consortium leases laboratories to several high-profile research projects funded by the Science and Technology Agency, its research communication activities extend beyond the bounds of its member companies to include researchers from throughout the science city, and the involvement of two science administrators, Chiba Genya and Kawamoto Tetsuzo¯. The Consortium has yet to receive much attention for collaborative research results. The TRC opened in December 1982 as the realization of an initial vague notion by Hayashi Chikara, President of ULVAC Corporation, which specializes in vacuum technology. In 1979, Hayashi wanted to create a collegiate research atmosphere in which his researchers could communicate and work with, as well as learn from, researchers in other companies involved in high-risk research. Hayashi approached a well-known science administrator, Chiba Genya, who was at that time designing the STA’s Explorator y Research for Advanced Technology Organization (ERATO) research program (which Chiba now directs), to help them organize the new venture. Because of his role, with the STA, of providing money to small science-based companies, Chiba had many contacts with other small- to medium-sized companies. Chiba introduced Hayashi to Hiruma Teruo, President of Hamamatsu Photonics, which specializes in producing photo-detectors. In their spare time on weekends the three men discussed the idea. “I wanted to create a kind of kindergar ten or playground for young researchers. Of course Hayase and Hiruma wanted a place to free up the creativity of their researchers, where there would not be any company regulations, but they were also worried about where the future generation of leaders for their companies was going to come from,” said Chiba. The two company presidents were thinking of building a research center in London or New Jersey, partly because their firms were better known abroad than in Japan, and par tly to broaden the perspectives of their young researchers. With Chiba’s help on weekends, the presidents of six other small firms were convinced of the value of the idea each of their companies specialized in different but potentially complementary fields (Chiba 1989).54 Gradually they decided that they should instead try to improve the research environment in Japan by building a consortium at
Research communication in Tsukuba
107
home. Since 1976 bureaucrats from the National Land Agency had been asking Chiba how they could encourage more private companies to locate in Tsukuba.55 A professor, Tazaki Akira of the University of Tsukuba, convinced them that Tsukuba was the ideal location for such a novel research idea (Kawamoto 1989d). Hayashi, Hiruma, and the other company presidents decided that Dr. Tazaki should lead the still-vague project of a consortium, but as a public employee Tazaki was prohibited by the Ministry of Education from accepting a private job. So, on 7 July 1980, Tazaki’s wife became the first president of the Tsukuba Research Consortium.56 These eight, relatively small high-technology “core” companies jointly bought 8.5 acres of land in Tsukuba’s To¯ko¯dai Research Park,57 and divided the land into 10 lots, according to Chiba’s concept of a shared facility. Chiba thought that the facility should resemble a Japanese temple (a traditional place of learning), but the construction and design would have been prohibitively expensive. 58 Each of the eight companies was awarded one lot, on which laboratories have been built. A ninth lot was further subdivided into laboratories which could be leased to other companies wishing to be associated with the Consortium, and a parking lot. On the tenth, central, lot was built a common-use facility in which the headquarters office of the Consortium (TRC Ltd., technically a separate entity) was located.59 TRC is financed by each of the core companies, the lease money from the “satellite” company laboratories, and its own varied seminar and information activities. In 1987 the annual operating budget of TRC was 170 million yen (over U.S. $1.3 million). Carrying out the original idea and purpose of the Consor tium is essentially the charge of the common organization, TRC. TRC is overseen by the presidents of the seven remaining firms (one firm, Akashi Kaisha, went out of business), and each of them is responsible for its own laborator y and personnel within the consortium. Administration of the core and satellite company joint activities is the responsibility of TRC. The Chief Operating Officer of TRC is Kawamoto, former first Director of the STA Tsukuba Center for Institutes. Dr. Chiba and the company presidents had met Kawamoto in 1979 when they toured the Tsukuba area looking for land. Because of Kawamoto’s experience with the Tsukuba Center for Institutes, he began to advise Chiba and the company presidents about the organizational str ucture of the Consortium. “It seemed like I kept seeing him. I decided then that he would be a good chief executive officer for the Consortium, but it took three years to talk him into taking the job” (Chiba 1989).
108
Research communication in Tsukuba
Table 5 Topics of the first 35 weekly forums at the Tsukuba Research Consortium from July 1985
Research communication in Tsukuba
109
The headquarters building has a large courtyard, cafeteria, small meeting rooms, some laboratory space, tatami-style bedrooms for out-of-town guests, an elegant seminar room, and lounge, besides the headquarters office. The headquar ters office subscribes to magazines which some researchers check out to read, but the Consortium has no library. Researchers and staff of each of the core and satellite companies regularly eat lunch in the cafeteria, and a lesser number come to weekly forums and seminars (see Table 5). Some researchers come to see English-language feature films on video one night a week. The five STA ERATO projects housed in the Consortium in 1989 represented the largest concentration of ERATO projects anywhere in Japan. Projects on ultra-fine particles, “superbugs,” namo-mechanisms, solid surfaces, and molecular dynamic assembly were carried out in Tsukuba. If visitors to Tokyo want to see ERATO projects “in action” they are frequently advised to travel to Tsukuba and visit the Consortium. Nine subgroups of the five ERATO projects have been housed in the Consortium complex. ERATO projects are publicized as exemplars of Japan’s new thrust into basic, creative research. The concentration of ERATO projects is a product of Chiba Genya’s involvement in planning the Consortium. Attempts at increasing research communication and collaborative research by TRC extend far beyond the 8.5 acres of the Consortium’s land. This outreach is one of the most curious or innovative aspects of TRC’s activities. Even though TRC’s explicit purpose is to increase communication and innovation among researchers of its member companies (this would represent return on investment for each of the member companies), many other researchers in the Tsukuba area appear to be beneficiaries of TRC’s emphasis on research communication. This seemingly public nature of a private consortium is easier to understand if TRC—the coordinating office of the Consor tium—is considered as a somewhat independent, money-making information service company, and not just as the central office for the member companies of the Consortium. TRC charges attendance fees and membership fees for its colloquia and seminars. And of course Consortium researchers stand to benefit from communicating with researchers outside of the TRC who are employed at the national institutes, private companies, and in Tokyo and elsewhere. Nevertheless, the Consortium member companies appear to subsidize a large number of research communication activities which may benefit outside or competing interests as much as they benefit the member companies. This appearance makes TRC seem quasi-public in its effects on Tsukuba Science City; that is, TRC seems to operate partly for the larger
110
Research communication in Tsukuba
public good of a more communicative research community in Tsukuba by sponsoring or hosting some of the same types of activities that, for example, the public Tsukuba Center for Institutes sponsors or hosts.60 This blurring of the distinction at TRC between private interests and the public good can largely be explained by the chief operating officer, Kawamoto Tetsuzo¯. Kawamoto was the first Director of the public Tsukuba Center for Institutes. At that time there were no models in Japan upon which to base the Center’s activities, so through trial and error Kawamoto molded the way in which the Center for Institutes initiated communication among Tsukuba researchers. At TRC he continues to do many of the same things that he did at the Center for Institutes, only with better funding, a more efficient staff, and fewer organizational restrictions. His official authority may have lessened since he is no longer a public employee, but his reputation as a knowledgeable person who has a broad network of friends and acquaintances is well established in Tsukuba.61 In many ways TRC is the organizational embodiment of Kawamoto. In the words of one Consortium researcher, “It is difficult to separate TRC from him, when thinking about either one” (Namba 1989). Young researchers at the Consortium know between 40 and 60 other Consortium researchers (Lewis 1989; Namba 1989). Par ticipation in forums, seminars, barbeques, and excursions to other Tsukuba labs var ies greatly. Most Consortium researchers are not so active in joint activities organized by the TRC office. A smaller network of Consortium researchers, linked together by Kawamoto, appears to be highly active in TRC’s activities. The potential of communication and joint research among Consortium researchers is greater than what has been achieved to date (Lewis 1989). At least eight collaborative research projects had been started as of May 1988, involving Consortium member companies 12 times and six universities and national research institutes (Tsukuba Research Consortium 1988a). The lack of more collaboration can be par tly explained by the Consortium’s brief five-year history. And much of the research conducted there is exploratory. As of June 1989, a total of 257 individuals worked at the Tsukuba Research Consor tium, 185 of whom were researchers. One core company had 35 researchers at the Consor tium, while three core companies only had 5 researchers each assigned to the Consor tium (Tsukuba Research Consortium 1989). The quality of researchers sent by each corporate headquarters laboratory to its Tsukuba lab varies by company (Chiba 1989).
Research communication in Tsukuba
111
Two indications that the Consortium is perceived as being successful by its core company presidents were a unanimous vote of confidence for the Consortium by its board of directors in June 1989, and plans to construct additional laboratory facilities at the Consortium by four of the core companies (Chiba 1989).
GOVERNMENT INITIATIVES IN TSUKUBA Tsukuba Science City is most noted for its concentration of national research institutes (48 percent of the government’s research budget is spent in Tsukuba), so relations between researchers in these institutes, those researchers in private companies, and those at the University of Tsukuba, are very important. In a national survey of privately employed researchers, one question asked about access to national research institutes. Only 19.7 percent of respondents replied that money to pay for access or collaborative research was a problem, while 43.6 percent said that application procedures were too complicated, and 47.1 percent responded that access was too restricted (Science and Technology Agency 1988a). Thus, a major challenge for science administrators is to make the national research institutes in Tsukuba more accessible. One initiative, the Tsukuba Center for Institutes, has played an especially important role in providing an infrastructure and resources which researchers can use to communicate among themselves. 62 The Center has a large auditorium and serves as a distribution outlet and processing center for computerized information about science and technology articles and papers from throughout the world. The Center offers meeting rooms, publishes Science Communication, a widely read monthly notice in Tsukuba, offers translation services, and subsidizes, through the Expo ’85 Memorial Foundation, the activities of some of the locally organized study groups of researchers. The Center has a small restaurant and a comfortable lounge (where many people take short naps or escape the humidity or rain), which make the building a popular place throughout the day as well as a place where researchers sometimes congregate at night. The Tsukuba Press Room is also located in this building.63
112
Research communication in Tsukuba
The Science and Technology Agency has formal authority to oversee and coordinate science development in Tsukuba. The success of its Tsukuba Center for Institutes over the last 13 years, however, may actually be precluding better communication between researchers at institutes which are under the control of different ministries. The Center for Institutes is identified by many people, particularly research administrators in Tsukuba and in Tokyo, as an STA institute (which it is), and not as a neutral meeting place. This sense of ministerial ownership or rivalry has now led to the establishment of two other institutes dedicated to the exchange of research communication: Tsukuba Center Inc., by MITI, and the Agricultural Exchange Center, by the Ministry of Agriculture, Forestry, and Fisheries. The former project is located immediately across the street from MITI’s large complex of laboratories. The Agricultural Exchange Center is located near the center of this ministry’s vast Tsukuba complex, in between many of its research institutes, and includes a conference hall, restaurant, and meeting rooms. The Ministry of Construction was planning a similar place for its Tsukuba researchers to meet and exchange information. The establishment of these new research exchange centers may decrease the diversity of researchers who use the original Tsukuba Center for Institutes, thus reducing the interdisciplinary role of that facility. In a real sense, each of the ministries with a major research presence in Tsukuba is “building its own castle” for scientific information exchange. A large-scale initiative to increase communication among researchers is Tsukuba Center Inc., which opened 10 July 1989.64 This cooperative venture is a high-technology science park and incubator project, which features 30 on-site laboratories for lease, 20 rooms for incubator projects, training programs for new researchers and retraining for experienced researchers, office space for 40 firms, in-house technical, management, financial and marketing consultation, joint testing facilities, a restaurant, a bar, and banquet rooms. The largest stockholder in Tsukuba Center Inc. is Ibaraki Prefecture, which invested U.S. $3.8 million dollars (500 million yen). 65 The Japanese Ministry of Finance (through the Japan Development Bank) is the second-largest investor. The project was planned by Mitsui Corporation and the Ministry of International Trade and Industry’s Agency of Industrial Science and Technology, and houses the equipment and researchers for
Research communication in Tsukuba
113
MITI’s large-scale Super Heat Pump Project. MITI’s coordination garnered the financial involvement of 74 corporations, 70 of which are major firms on the Tokyo Stock Exchange. Each corporation has contributed between U.S. $77,000 and $769,000 (between 10 and 100 million yen), and, along with Ibaraki Prefecture and the Ministry of Finance, is a stockholder in the Center. The history of Tsukuba Center Inc. is very suggestive of how private business and national ministries sometimes interact in Japan. The original idea for a new center for small businesses came from participation by Mitsui employees in industrial working groups organized by MITI and the STA, and from within the Mitsui “family” of 24 companies. During the mid-1980s, several people within Mitsui wanted to link the conglomerate’s companies in a more productive way, as well as create a joint means of gathering information about new technologies. One of these employees, Okazaki Hideaki, had moved to Tsukuba in 1979. As he learned more about Tsukuba Science City, he began to suggest that Tsukuba would be an ideal location for an institute dedicated to technological information exchange. “Many corporations were then committing themselves to spending huge amounts of money for Tsukuba Expo. That struck us as a great way to waste a lot of money, since the pavilions are demolished afterwards” (Okazaki 1989). Okazaki’s idea was adopted by Mitsui’s top management, with the support of an elder Mitsui director, who was from Ibaraki Prefecture. The Mitsui overseas technical offices in New York, San Francisco, England, and Germany were enlisted to gather information on how other collaborative research institutes involving high-technology firms were organized. Mitsui was particularly interested in using the Center as a technology transfer linkage between industrial and academic researchers. About 20 outside companies were sought as partners (Okazaki 1989). Mitsui officials inspected several sites in Tsukuba and sent a report of the corporation’s intentions to the four ministries with the largest presence in the science city: MITI, the STA, the Ministry of Agriculture, and the National Land Agency. Mitsui was afraid that to identify the project too closely with any one ministry would jeopardize its relations with the other ministries—and it has to deal with each of them on a regular basis. MITI was the most enthusiastic about the project, and suggested that Mitsui take
114
Research communication in Tsukuba
advantage of a new law that MITI had recently sponsored in 1986 which allows for consortia-type research organizations. One year later, in 1987, Mitsui was still trying to negotiate an agreement with the 20 companies which were interested in financing the “Mitsui Technology Exchange Center,” as it was then called. Deciding on an appropriate organizational form was a problem because Mitsui had wanted the Center to be more than a real estate investment in which it would rent offices to other companies. At this same time, Ibi Toshio, the new director of MITI’s Agency of Industrial Science and Technology Planning Division in 1986, was trying to develop an idea for a “salon” where diverse researchers could not only communicate together but also where laboratories would be open to all members. “I couldn’t get enough public money for the salon idea, and then I heard about Mitsui’s plan,” said Ibi. “They wanted to rent office space and hold seminars, but I thought that was a very conventional idea. Mitsui did not like my idea of shared laboratories. Bad for trade secrets” (Ibi 1989). The final version of Tsukuba Center Inc. is a compromise between Mitsui’s goal of communication about research, and MITI’s goal of communication among researchers. Ibi had been partly responsible for planning the infrastructure of MITI’s large Tsukuba complex of research institutes, but that project was, for the most part, finished and operating smoothly. His salon idea, the planning for which in 1986 he had penciled into the 1987 MITI budget, seemed compatible with Mitsui’s goals. Ibi wrote a funding proposal for the Center to Research Core, a large public foundation endowed by the sale of Nippon Telephone & Telegraph stock, but in order to receive a grant there had to be many potential beneficiaries of the Center—not just Mitsui and its partner companies. Thus, Mitsui relinquished the role of the lead planning organization for the Center, perhaps begrudgingly, and became instead a major investor. Research Core provided a seed grant of approximately U.S. $7.6 million (1,000 million yen) for the Center. MITI’s rationale for playing a coordinating role was that a government partnership would give the new Center the distinctive organizational form that Mitsui was searching for, and increase the attraction for more industrial partners. “Mitsui was concerned that the Center should be profitable, and if MITI were involved it would look like a better investment to smaller companies” (Ibi 1989). Yet
Research communication in Tsukuba
115
over-identification with any one ministry might likewise doom the collaboration. The many former MITI officials now with Mitsui were influential in convincing the corporation to loosen its control over the project (Okazaki 1989). In an attempt to dampen the impression that the Center was becoming solely a MITI project, some of the directors of the Center were chosen from the STA, the Ministry of Agriculture, the National Land Agency, and the Ministry of Construction (Ibi 1989). Experts from different organizations were consulted about how to best prepare the Center. The Ministry of Finance, through the Japan Development Bank, invested in the Center. Next, a group of MITI officials went to visit the governor of Ibaraki Prefecture. The prefecture eventually became the largest stockholder in the Center.66 Okazaki had identified three potential sites in Tsukuba. One of these was immediately next to MITI’s complex of research institutes. Ibi liked this site, since it would encourage easy and frequent access to the new Center by MITI’s researchers. The land was owned by the Japan Housing Corporation, but it was designated for use by the STA, not MITI. So Ibi had to ask the STA for permission to buy the land. The STA refused, but then agreed to sell half of the site to a private entity (but not directly to MITI). The sale of public land to a private company, however, required the approval of all 16 ministry vice-ministers. Fortunately for MITI, in January 1987 the vice-ministers were already planning to meet to discuss the new Tsukuba College of Technology, so Ibi was able to get the land sale on the meeting agenda. The sale was approved. On 22 September 1987, officials from Ibaraki Prefecture, the Japan Development Bank, Mitsui and six other private corporations met at the Japan Federation of Economic Organizations Hall in Tokyo, to formally discuss the timeline for Tsukuba Center Inc. (Mitsui Trade News 1987). Completion of the Center was scheduled for January 1989. That same month, a MITI official joined the Mitsui project. This personnel assignment signaled that the Center was indeed a MITI project. Further identification with MITI came with the location of the MITI Super Heat Pump Project within one of the Center’s buildings. Tsukuba Center Inc. was established as a business in February 1988 (17 months before its opening). It has a staff of 25. The President, Hattori Tadashi, is an engineer by training and a longtime Mitsui employee and board member. The entrepreneurs who occupy the 20 incubator spaces have mostly signed three-year
116
Research communication in Tsukuba
contracts. “We do not want them to stay longer than five years. We want to keep the firms rotating in and out” (Hattori 1989). The Center derives income from rental fees, seminar fees, its restaurant and bar, and percentages of the profits of entrepreneurs who benefit from the incubation in the Center. An entrepreneur interviewed in the present study who is renting a space in Tsukuba Center Inc. uses it as a place to make contacts and evaluate information, not as a lab or technical product site. His actual research and production are conducted at other places in Tsukuba (Okazaki 1989). Likewise, some of the larger companies which are stockholders in the Center are mostly concerned with industrial and technological surveillance, or establishing personal contacts. An example of a Tsukuba Center Inc. stockholding company is Jo¯yo¯ Bank, one of the largest regional banks in Japan. According to Kobayashi Teruo, General Manager, Jo¯yo¯ Bank has three primary objectives for investing in the Center: Our main objective is to finance the construction of R&D facilities of those companies which have been laggards in coming to Tsukuba, but who will, through their experience with the center, decide to build here. Another objective is to learn about the technology supporting firms that are moving to the Tsukuba area, and establish business relations with them. A third goal is to establish personal relations with the employees of these companies, many of them researchers, who need to buy houses in Tsukuba. We can serve as a matchmaker with local realtors. And we can finance their home loans. (Kobayashi 1989) Jo¯yo¯ Bank is the only bank with an automatic teller machine at the Center, so the bank may benefit by gaining additional accounts. Jo¯yo¯ Bank has three full-time employees and one part-time employee stationed at Tsukuba Center Inc. The potential benefits to be gained by a multifaceted trading conglomerate such as Mitsui from its investment in Tsukuba Center Inc. are even greater. Besides learning of new entrepreneurial technological developments about which it would not otherwise have such direct information, Mitsui is in the position to offer an entrepreneur with a marketable product a
Research communication in Tsukuba
117
complete worldwide testing, manufacturing, marketing, and distribution system. In effect, for a conglomerate such as Mitsui, the Center is a breeding ground for new high-risk, high-profit firms that can be brought to market under the Mitsui umbrella. By using the information networks cultivated in the Center (and its many other evaluative experts) a conglomerate can minimize risk, while using its diversified resources to maximize profit. The longterm value of holding stock and maintaining a presence in a collaborative venture such as Tsukuba Center Inc. becomes very apparent. The University of Tsukuba, through its graduate programs, serves as a major government initiative to increase research communication in the area. Every year about 1,000 graduate students work as assistants at national research institutes, and in 1984 the university employed 44 lecturers from those institutes, while 62 university professors conducted research at 11 national institutes (Kawamoto 1984). Now, the university has established the equivalent of adjunct professorships for eminent researchers who normally work at the national labs or at local private laboratories.
MAKING SENSE OF THE ROLE OF COMMUNICATION INITIATIVES Is it possible to plan creativity, or is it only possible to plan the conditions under which creativity may thrive? This question addresses an apparent dichotomy between the degree of planning and the freedom of thought which is often considered a necessary condition for creativity. Although an answer to this question is elusive, the present results point rather clearly to the contribution which informal groups and individual-initiated communication can make toward fostering creativity within the constraints of a planned system of scientific work. Tsukuba Science City represents a very planned system of basic research. Yet plans for the city did not include the means of encouraging or rewarding interpersonal communication among researchers. The formal authority of science administrators was imposed on researchers in order to force communication among scientists of different research institutes, but a lack of administrative commitment combined with scientists’ disdain for top-down bureaucratic control led to failure.
118
Research communication in Tsukuba
Informal groups are most important in that they give birth to interpersonal networks through which relatively high-value knowledge may be exchanged. This is not the explicit purpose of informal groups. These groups were formed to solve the general need of scientists for more knowledge about hot research topics. The frankness of discussions in informal groups is limited, being a function of the degree of participation of private industry scientists employed by competing firms, and government researchers employed by rival ministries or laboratories. Thus the value of the knowledge discussed in informal group meetings is moderate. Common or complementary research interests, as well as social interests, may then lead to a hardening of friendships. These private, interpersonal relations serve as the means of transferring high-value knowledge among researchers. Such relations often link academic or government scientists with corporate scientists, government scientists with academics, and occasionally lead to collaboration between corporations. Informal groups in Tsukuba Science City are most effective if they include the participation of private industry researchers. Having participants who are interested in the same phenomena, represent diverse organizations, and have different uses for the phenomena under study, lessens competition among informal group members while increasing the potential for complementary and collaborative research projects. Study groups with strong representations of industrial researchers remain the most intellectually active groups over a period of years. Informal group participation by scientists leads to collaboration among those researchers. Over 60 percent of study group organizers replied that their members either had conducted collaborative research among themselves, probably had done so, or were planning to do so. Half of these survey respondents believe that collaborative research projects resulted from discussions in the informal group. Survey and case study results indicate that the function of informal groups in Tsukuba, particularly of intradisciplinary groups, has changed since their inception. Most of the groups, which can be categorized as minor rebellions against the established organization of research which prevailed in Tokyo during the late 1970s, have themselves become institutionalized. As a result of a changing research environment in Tsukuba and in scientific disciplines themselves, not all of the groups which were once effective in
Research communication in Tsukuba
119
technology transfer are still so effective. Groups organized to address interdisciplinary research topics tend to function rationally (that is, they serve their original purpose of transferring technology) but their life-span is short. As soon as worldwide knowledge about a new interdisciplinary field becomes somewhat well understood and traditional means of knowledge exchange (such as journals and conferences) appear, these groups disband. Sometimes the novel emphases of these groups become routinized in the establishment of formal research labs. Interdisciplinary groups tend not to outlive their function. Intradisciplinary groups, which bring together homophilous researchers, tend to address topics which are comparatively well understood. These informal groups last longer than interdisciplinary groups, but their function becomes irrational. Intradisciplinary groups in Tsukuba appear to exist primarily to confer status on their leaders. The institutionalization of intradisciplinary groups reflects the intellectual conservatism which characterizes aging scientists, which serves as a disincentive for young researchers to join the group. Government initiatives aimed at increasing communication among Tsukuba researchers are considered less successful by participants than are the informal study groups. An exception is the Tsukuba Center for Institutes, which has served as a common meeting place for diverse researchers. A private consortium appears successful thanks to aggressive and explicit communication strategies. As for MITI’s Tsukuba Center Inc., the original proponent (Mitsui) may profit greatly. Thus, communication initiatives may be profitgenerating schemes. Appreciation of the benefits of interministry research notwithstanding, ministry competition is alive and well in Tsukuba. The most striking evidence of competition is the concomitant establishment of research exchange centers, each of which is promoted as a neutral meeting place for researchers employed by different ministries. Each of the centers will appear very similar: conference halls, restaurants, meeting rooms, and some will have laboratories. The most successful of the ministry centers will be those with strong leaders who are effective science and technology networkers in the local community and in Tokyo. Collaboration among Tsukuba researchers may well be curtailed by this functional duplication (which also characterizes research and testing equipment in the city). Such a lack of
120
Research communication in Tsukuba
Box 5 How not to communicate: Tsukuba Network In 1987, the Science and Technology Agency began testing a small-scale combination electronic mail, bulletin board, science and technology data-base, and document retrieval system for researchers in Tsukuba Science City. On 28 March 1989, the expanded test system was named Tsukuba Network, and officially began as a free service for Tsukuba researchers who had access to a personal computer, a telecommunication modem, a telephone, and telephone lines. Tsukuba Network offers access to some archived information which would conceivably be of high value to researchers. For example, as one of its services, the Japan Information Center for Science and Technology distributes and collects an annual questionnaire to researchers at 650 universities, national research institutes, and prefectural and municipal research institutes throughout Japan concerning their ongoing research projects. By looking at this data-base, a researcher can find other researchers who are conducting complementary research. Tsukuba Network allows access to this service (Takagi 1989b). Tsukuba Network also links together both UNIX and MS-DOS computer operating systems, which is a convenience for researchers using traditionally noncompatible computers (Takahashi 1989). The idea for the electronic communication system originated with Professor Takahashi Hidechika of the Tsukuba College of Technology.67 Takahashi graduated in mathematics from Saitama University in 1960, and hand-built several computers with colleagues before joining the Institute for Nuclear Physics in Tokyo in 1965. Takahashi’s job was to design par ticle accelerators by computer, for the new National Laboratory for High-energy Physics in Tsukuba. He moved to Tsukuba in 1971 as KEK’s director of computer operations. At KEK, Takahashi learned much about the data-base needs of researchers. But each research institute in Tsukuba was only linked to its parent ministry’s computer system. At the University of Tsukuba, different departments have their own local area networks, which are not compatible (Matsuda 1989). There was no sharing of computer systems in the science city.68 In 1975, Takahashi became chairman of a small working committee which began, as one of its activities, to explore the possibility of a shared computer networ k. Committee members were mid-level engineers from the computer depar tments at various research institutes, so discussions centered around the technical requirements for such a system (Takahashi 1989).
Research communication in Tsukuba
121
In 1986 and 1987, the committee conducted two questionnaire surveys, one of Tsukuba researchers in the national research institutes, and one of researchers in private laboratories. The purpose of the questionnaires was to determine the information needs of Tsukuba researchers. The national institutes received a different, more extensive, questionnaire than did private companies. 69 Fifty-six public research institutes received copies of the questionnaire during August and September 1986. The committee received 168 completed questionnaires from people at 49 of these institutes. Questionnaires were sent to 70 private laboratories during December 1986 and January 1987, from which 46 completed questionnaires were returned. A majority of the responding national laboratory researchers answered that they were somewhat dissatisfied with current exchanges of research information in Tsukuba, thought the situation was worsening as no more information channels to Tokyo had been established, would use electronic mail and data retrieval systems, thought that Tsukuba and Tokyo were the most vital geographic areas to link together, and that the primary obstacle to establishing such a system would be administrative barriers to sharing information. Majorities of the responding private laboratory researchers answered that they were dissatisfied with current exchanges of research information in Tsukuba, would like information about research projects, information about the topics and meeting details for informal study groups, and a data retrieval service, and thought that cost was the primar y obstacle to the establishment of such a system. Additionally, 49 percent of private company respondents wrote that “information gathering” was the primary motivation for establishing a facility in Tsukuba, and 95 percent responded that their company did not have information or data that they would willingly contribute to the system (Tsukuba Institute Coordinating Committee 1987). Tsukuba Network has had a very slow start. Immediately after the opening on 28 March 1989 of Tsukuba Network, widely publicized in Tsukuba, problems began with the service. Many researchers who called the Tsukuba Center for Institutes, where the Network system is administered, were rejected for membership on the basis that they did not qualify for the free service. Others were told that they qualified but would have to wait an indefinite period of time to receive a user ID number (Takagi 1989b). Refusing and delaying potential adopters is not a very effective strategy for diffusing a new communication technology. Who qualifies for membership? Although Networ k is advertised as a system for all Tsukuba-based researchers, and its chief administrator wants to eventually encourage everyone
122
Research communication in Tsukuba
in Tsukuba, including nonresearchers, to use Network, user ID numbers have only been allotted to a select group of researchers (Takagi 1989a). Theoretically, any researcher employed by an institute which is a member of the Association of Research and Academic City Institutes is eligible for a user ID number. “ID numbers are only allotted to individuals, not to organizations, since a major purpose of Network is to break down institutional barriers” (Takahashi 1989). But individuals must be members of a certain organization. The number of researchers who are using Network is about the same number of people, 250, who were previously connected to the preceding test system. The capacity of the system is about 10,000 users. One service which the system’s founder, Takahashi, would like to see offered is a complete directory of researchers and staff who work at all of the national institutes in Tsukuba Science City. Such a directory is already compiled, printed, and widely used by Tsukuba research personnel, and a Network version would have the advantage of more frequent and easier updating of information. Having such a directory on Tsukuba Network would help diffuse use of the system, since the directory is an important information source with which Tsukuba researchers are already familiar, and one which they use regularly. Both hardware and software technical problems have slowed down the system since its March 1989 debut (Takagi 1989b). Also, Network is not yet connected to international data-bases or international communication systems, such as Bitnet. Another problem is inadequate funding for the project. In 1989, only one part-time technician was employed to operate, repair, and modify the Network system. Priority was being given to public researchers at the expense of private researchers. Of the researchers interviewed in the present study, all had heard of Tsukuba Network, some were curious about it but said they had little time to learn how to use such a system, and most had heard uncomplimentary comments from friends about the system. The largest problem for Tsukuba Network may be that the project was essentially the work and inspiration of one former director of the Tsukuba Center for Institutes, Takagi Joichi, who, like most public science administrators in Japan, changes jobs every two to three years.
Research communication in Tsukuba
123
interdisciplinary communication will partially thwart the purposes of the centers. The popular perception that communication is neither frequent nor close between Japanese universities and industry is false. Private consulting by professors is very common and allows the efficient communication of valuable information, while building interpersonal networks which are instrumental in finding jobs for university graduates. Even though approximately 1,000 graduate students from the University of Tsukuba work in the Tsukuba national laboratories each year, the university does not serve as a feeder school for the national labs. New government researchers are rarely assigned to their choice of laboratory or location. This assignment system is detrimental to the strengthening of communication networks within Tsukuba.
Chapter 6
Collaboration networks
In Chapters 3 and 4, the history of Tsukuba Science City was recounted and interpreted. Chapter 5 focused on communication among researchers in the science city, and the roles of both informal and formal communication initiatives meant to increase collaboration in a variety of fields. Are there identifiable patterns of research collaboration in Tsukuba? How has research collaboration in Tsukuba changed over time? To what extent do researchers at government research institutes, universities, and private R&D laboratories collaborate? Are researchers in different institutes which are under the control of the same ministry more or less likely to co-author articles than are researchers who are under the control of different ministries? Which research institutes are most productive? To what extent is Tsukuba an intellectually independent research community? This set of questions drove the data collection and analysis reported here. To gain a better understanding of the overall scientific social structure in Tsukuba, this chapter discusses patterns of research relationships, or collaboration, in the city. The data conceptualized as representing a scientific social structure comes from the coauthorship of research journal articles (Dearing 1989b).
MAPPING COMMUNICATION STRUCTURE Co-authorship is an effect of research collaboration, which has resulted from both social and research communication. Thus, coauthorship networks are descriptive of some of the communication patterns among researchers. Making inferences about communication among researchers based on their co-authorship is less problematic
Collaboration networks
125
than co-citation analysis, in which citations by different authors of the same prior articles are assumed to represent cognitive links between the citing authors. Co-citation analysis has been criticized because like references cannot automatically be construed to represent interpersonal communication (Mulkay 1974; Chubin 1976; Edge 1977; Chubin 1987). Co-authorship data have strong inferential advantages over co-citation data because co-authorship data represent a complex set of activities which requires direct, personal communication among participants, can be taken as evidence of similar or complementary cognitive orientations, whether they be theoretical, methodological, or practical, and is based on research results which are themselves the products of scientific research. Co-authorship is also a certain indication of a two-way communication relationship, though such relations are frequently not equal in the degree of contribution or in the status of the authors. Often, co-authorship may also be considered an indication of a social relationship between authors, an organizational relationship, in which the organizations to which authors belong also share identities in the research project, through funding or other provisions, and a long-lasting personal relationship. This latter possibility is important, since it means that co-authorship data represent enduring patterns of communication between researchers, and often not just short-term work relations. Co-authors may not communicate for years, but the network connection remains between them, and may be activated for various purposes throughout their careers. Over a long period of time, many co-authorship links may be weak in frequency of communication but strong in utility to the researchers (Granovetter 1973). Co-authorship data have been used previously to measure degrees of scientific integration (Price and Beaver 1966), and trends in international scientific collaboration (National Science Board 1987:95–100). Analyses of co-authorship data may give us an approximation of the extent of productive interorganizational communication and research in Tsukuba, and between Tsukuba and other geographic areas. And because our study is focused at the organizational level of analysis, we can test the extent to which interorganizational communication and work relations among researchers parallel or contradict the administrative jurisdictions of the Japanese government ministries which control the government research institutes in Tsukuba.
126
Collaboration networks
The present analysis uses co-authorship data to represent patterns of relations among researchers.70 This relational approach focuses on the communication among researchers as indicative of a more-or-less geographically bounded scientific social structure (Monge 1987:146–7). CONCEPTUALIZING TSUKUBA’S RESEARCH COLLABORATION NETWORK Data on co-authorship were supplied by the Japanese Science and Technology Agency’s Japan Information Center for Science and Technology (JICST) for the years 1979, 1982, 1985, and 1988. The present analysis centered around three variables: articles, authorships, and links. Articles refer to the data entries (computerized summaries) made by data entry clerks at JICST. Each entry represents one article published in a science or technology journal. Authorships refer to the authors and co-authors of the articles which are published in science and technology journals. Since many researchers authored and co-authored more than one article during our periods of study, and since we are interested in organizational rather than individual distinctions, we count authorships, not authors. A link is a co-author relationship between two co-authors, who are authors of the same article. There is one link between any two co-authors. But many articles are co-authored by more than two researchers. The number of links per article increases more rapidly than does the number of co-authors of an article.71 Of these three types of variable, links are perhaps the best measure of communication relationships. When a research team adds one new person, a whole set of new relationships may potentially develop. For example, when a five-person biomedical team of researchers adds a chemist (for a total of six people), five new relationships have potentially been created. Using links to represent relational networks, as opposed to relying on counts of the numbers of articles published or the number of co-authors involved, more accurately maps the complexity of interpersonal communication patterns. For the time-periods 1979, 1982, 1985, and 1988, the total number of articles was 11,460; the total number of authorships was 30,468; and the total number of links was 63,360. 72 The mean number of co-authors per article in our data-set was 2.7, while the
Collaboration networks
127
Table 6 Number of articles, authorships, and links, by year and by language, published by researchers in Tsukuba Science City and their co-authors
mean number of links per article was 5.5. Table 6 shows the breakdown by year and language of publication for these total numbers of articles, authorships, and links. STRUCTURAL VIEWS OF THE SCIENCE CITY What do structural perspectives of Tsukuba look like? First, we can distinguish between two types of link. One type of link represents relationships between authors who work in the same research institute. Another type of link (of particular concern to the present analysis) represents relationships between authors who work in different research institutes. Links between authors are evidence not only of communication among researchers, but also of collaborative and work relationships, and often of social relationships. Establishing collaborative, work, and social links among researchers in different institutes has been a primary objective of science administrators in establishing Tsukuba Science City. So the focus here is on those links which bridge organizations.73 The number of organizational links between Tsukuba institutes, as well as between Tsukuba institutes and research institutes located outside of Tsukuba, has continually increased. Yet the percentage of increase is lessening. In contrast to this slowed growth in the
128
Collaboration networks
Table 7 Results of four NEGOPY network analyses for research organizations and co-authors of research articles
number of organizational links between institutes, the strengths of those links, as measured by the numbers of co-author relationships between individuals in those organizations, is continuing to grow rapidly. Over time, this trend results in an increasingly strong network structure7475 (see Table 7). Whereas the percentage of increase in the number of organizational links between research institutes is lessening, indicating that the number of possible organizational linkages is being approached, the percentage of increase in the number of individual co-authorship links is continuing to grow rapidly. This latter fact suggests that the potential for co-author links is very great, and has not yet been approached within this research system. So even though progress has been made in Tsukuba to get researchers together, there is a long way to go, at least in terms of the potential for communication networking in the science city. Graphs of link data reinforce the magnitude of this change and its potential. Figure 15 shows the total number of organizational links within Tsukuba and their strengths for 1979. Figure 16 shows the total number of organizational links within Tsukuba and their strengths for 1982. The degree of integration between institutes, as well as the strengths of organizational links, for these time-periods is relatively low when compared with Figures 17 and 18, which show the total links between organizations in Tsukuba
Collaboration networks
129
Figure 15 Within-Tsukuba organizational links and their strengths (strength equals the number of co-author links between researchers) for 1979
Figure 16 WithinTsukuba organizational links and their strenths (strength equals the number of co-author links between researchers) for 1982
Figure 17 Within-Tsukuba organizational links and their strengths (strength equals the number of co-author links between researchers) for 1985
Figure 18 Within-Tsukuba organizational links and their strengths (strength equals the number of co-author links between researchers) for 1988
Collaboration networks
133
and their strengths for 1985 and 1988, respectively. Both the complexity and the strength of organizational research relations have increased considerably over time. In network terms, the University of Tsukuba is the most central research organization in the science city. There are more coauthorship and organizational links with the University of Tsukuba than with any other public or private institute in the science city. This finding is to be expected of a successful university, since a university encompasses a number of disciplines and a great variety of researchers. This increased complexity of the number of links between research institutes is also well illustrated at the more micro level of single research institutes. For example, in 1979, MITI’s Electrotechnical Laboratory (ETL) in Tsukuba had 135 links with nine institutes in Tsukuba and locations of research institutes outside of Tsukuba.76 In 1982, ETL had 298 links (a 121 percent increase) with 13 institutes in Tsukuba and locations of institutes outside of Tsukuba. In 1985, ETL had 490 links (a 65 percent increase over the previous time-period) with 19 institutes in Tsukuba and locations of institutes outside of Tsukuba. In 1988, the Electrotechnical Laboratory had 952 links (a 94 percent increase over the previous time-period) with 22 institutes in Tsukuba and locations of institutes outside Tsukuba. The strongest link during 1988 was among researchers at the University of Tsukuba and researchers at the National Laboratory for High-energy Physics. In 1988, 65 research relationships characterized this link. The second strongest link during 1988 was between the University of Tsukuba and the National Institute for Environmental Studies, with 57 co-author relationships. The present analysis indicates that most links are between researchers who work in the same research institutes. The absolute number of such intra-institute links is increasing. A lesser number of links are interinstitutional, and this number is also increasing. But the rates of increase are different for intra- and interinstitute links. While the percentage of all links which are intra-institutional is decreasing during our three time-periods of study, the percentage of all links which are interinstitutional is increasing (see Figure 19). This pattern of a decreasing percentage of links within research institutes and an increasing percentage of links between research institutes holds across languages. For both Japanese-(see Figure
134
Collaboration networks
Figure 19 While the percentage of all links within research institutes is decreasing, the percentage of links between institutes is increasing
Collaboration networks
135
Figure 20 For Japanese-language articles, the percentage of all links within research institutes is decreasing, while the percentage of links between institutes is increasing
136
Collaboration networks
Figure 21 For English-language articles, the percentage of all links within research institutes is decreasing, while the percentage of links between institutes is increasing
Collaboration networks
137
Figure 22 The percentage of links between Tokyo and other (not Tokyo and not Tsukuba) areas, and between Tsukuba and other areas, is increasing, while the percentages of Tsukuba-to-Tokyo and withinTsukuba links are decreasing
138
Collaboration networks
20) and English- (see Figure 21) language articles, the percentage of links within research institutes continually decreases, while the percentage of links between research institutes continually increases.77 These juxtaposed trends are particularly pronounced for English-language articles (see Figure 21).78 The numbers of links between Tsukuba researchers and those researchers in areas outside of Tsukuba as well as the number of “indirect” links between Tokyo co-authors and co-authors in other places (not Tsukuba and not Tokyo) are very large, compared to the number of Tsukuba-to-Tsukuba links. Besides increases in the absolute numbers of links by type, Figure 22 shows that the percentage of indirect links is increasing, the percentage of links between Tsukuba researchers and researchers outside of Tsukuba and Tokyo is increasing, the percentage of links between Tsukuba and Tokyo researchers is decreasing, and the percentage of Tsukubato-Tsukuba links is decreasing slightly. Of all co-authorship links between Tsukuba research institutes in 1988, only 11 percent involved privately employed researchers.79 Of the 775 co-authorship links between researchers in Tsukuba research institutes for 1988, 688 were between researchers in national institutes. When we analyze the links between researchers in different Tsukuba research institutes by type of institute, we can see that the absolute number of links between researchers at different national research institutes is increasing. Also, the absolute number of links between researchers at the national research institutes and private companies was about the same during 1985 and 1988, the absolute number of links between researchers at the University of Tsukuba and private companies is increasing, and the absolute number of links between researchers at the University of Tsukuba and researchers at Tsukuba national research institutes is increasing. The percentage of all links between researchers at the national research institutes increased rapidly from 1985 (when it was 34.6 percent) to 1988 (when it was 45.9 percent), while the percentage of links between researchers at the University of Tsukuba and researchers at the national research institutes continually decreased (see Figure 23).
Collaboration networks
139
Figure 23 Within Tsukuba Science City, the percentage of links between researchers at different national institutes is increasing, but decreasing between the University of Tsukuba and the national research institutes
140
Collaboration networks
NUMBER OF ARTICLES PUBLISHED Analysis of the number of articles shows a clear increase in those articles which had a single author, co-authors within the same research institute, and co-authors in different research i n s t i t u t e s . O n c e a g a i n , m a j o r d i ff e r e n c e s i n t h e d a t a a r e revealed by analyzing trends in the percentage of these types of articles. Whereas the percentages of articles with single authors and coauthors in the same research institute have continually decreased, the percentage of articles with co-authors in different institutes has continually increased (see Figure 24). These juxtaposed trends hold across language, whether Japanese (Figure 25) or English (Figure 26). In the case of English-language articles in 1988, the percentage and number are a majority, compared to those articles with a single author and those articles with co-authors in the same research institute.
NUMBER OF AUTHORSHIPS The absolute number of authorships has increased from 6,455 in 1982, to 8,816 in 1985, to 12,343 in 1988. Looking just at the institutional affiliation of the authors, the two most noticeable trends are a decrease in the percentage of authorships at Tsukuba national research institutes, and an increase in the percentage of authorships at research institutes which are not in Tsukuba or in Tokyo (see Figure 27). Within Tsukuba Science City, there is a large variance in the number of authorships by research institute. The research institutes with the most authorships have remained more or less consistent in ranking, relative to one another. In 1988, two research institutes had by far the highest numbers of authorships: the Electrotechnical Laboratory, with 1,346 authorships (from 559 researchers), and the University of Tsukuba, with 1,203 authorships (from 589 researchers, plus some graduate student authors). The research institute showing the largest over-time increase was the National Laboratory for High-energy Physics, which “ranked” 10th in 1982, 8th in 1985, and 3rd in 1988 among the 46 research institutes in Tsukuba (see Table 8). When the numbers of authorships per research institute are disaggregated by language, dramatic differences emerge. Authors
Collaboration networks
141
Figure 24 The percentage of articles co-authored by researchers in different research institutes has increased, while the percentage of articles with one author and those with co-authors in the same research institute has decreased
142
Collaboration networks
Figure 25 The percentage of Japanese-language articles co-authored by researchers in different research institutes has increased, while the percentage of articles with one author and those with co-authors in the same research institute has decreased
Collaboration networks
143
Figure 26 The percentage of English-language articles co-authored by researchers in different research institutes has increased, while the percentage of articles with one author and those with co-authors in the same research institute has decreased
144
Collaboration networks
Figure 27 Trends in the percentages of authorships
at some research institutes, such as the Public Works Research Institute and the University of Tsukuba, publish almost all their articles in Japanese. Authors at other research institutes, such as the National Institute for Inorganic Materials and the National Laboratory for High-energy Physics, generally publish in English. At some research institutes, such as the Electrotechnical Laboratory and the National Chemical Laboratory for Industry,
Collaboration networks
Table 8 Tsukuba Science City research institutes with the most authorships of science and technology journal articles
145
146
Collaboration networks
Table 9 Tsukuba Science City research institutes with the most Japanese-language authorships of science and technology journal articles
an approximately even number of articles are published in Japanese and in English (see Tables 9 and 10). The language of publication of a research article is determined by the discipline or field within which the researcher works. In some disciplines, such as high-energy physics, the salient audience or reference group of physicists is dispersed worldwide. The key journals in which high-energy physicists prefer to publish are all published in English (abstracts in some of the journals are duplicated in other languages). Thus, for a high-energy physicist to write an original article in any other language but English is somewhat counter-productive, and may signal that the article is not worthy of publication in a good journal. Even if Japanese physicists publish an article in a Japanese journal, the language of the original publication is likely to be English. Conversely, many Japanese researchers write in Japanese because their salient audience, or reference group, exists in Japan.
Collaboration networks
147
Table 10 Tsukuba Science City research institutes with the most English-language authorships of science and technology journal articles
Researchers at the Public Works Research Institute, for example, investigate very applied problems of highway engineering, erosion control, and waste water treatment. These topics are to a large extent specific to environmental conditions in Japan, and are read primarily by Japanese geologists and engineers. ARE MINISTRIES A BARRIER TO COLLABORATION? Are Tsukuba researchers in different institutes which are under the control of the same ministry more likely to co-author articles than are Tsukuba researchers who are under the control of different ministries? Given the history of animosity between several of Japan’s ministries, such a result would indeed be expected. In the case study research reported in Chapter 5, a number of researchers mentioned a lack of research collaboration between researchers in institutes which are under the control of rival ministries, such as the STA and MITI.
148
Collaboration networks
Moreover, the physical arrangement of research institutes in Tsukuba would seem to encourage intraministry collaboration rather than interministry collaboration. The ministries with many research institutes in the science city have designed the institutes in clusters. For example, the Ministry of Agriculture, Forestry, and Fisheries has all of its 13 institutes in southern Tsukuba, with eight of them within one large interconnected complex. MITI’s Agency of Industrial Science and Technology center, with 9 institutes, is even more tightly organized as a group of research institutes. Despite these expectations, structural results based on coauthorship do not support the hypothesis that researchers within different institutes under the same ministry collaborate more than researchers who are under the control of different ministries. In fact, results here of co-authors within the science city indicate that, by a ratio of 2 to 1, researchers cross ministry boundaries in their co-authorship of journal articles.80 Tsukuba co-authors at different national research labs were employees of the same ministry 32 percent of the time, whereas 67 percent of such co-authors were employees of different ministries. This is an important finding which suggests that policies such as exchange programs and the co-location of researchers pay off. On a ministerial level, differences emerge. For example, STA researchers in Tsukuba only co-author with other STA researchers in other institutes (there are five STA Tsukuba institutes) 11 percent of the time. In contrast, Ministry of Agriculture, Forestry, and Fisheries researchers co-author with researchers in other Ministry of Agriculture, Forestry, and Fisheries Tsukuba institutes (this ministry has 13 Tsukuba research labs) 81 percent of the time. Table 11 shows, in matrix form, the numbers and percentages of the co-authorship links between ministries when organizational links are aggregated by ministry, for 1988. The large number of co-authorship links with Ministry of Education researchers in Tsukuba results mostly from researchers at various national labs teaming up with science and engineering professors and graduate students at the University of Tsukuba. For example, of the 64 coauthorship links within Tsukuba involving the National Institute for Environmental Studies, 57 (89 percent) involved university researchers. In terms of co-authorship among researchers in Tsukuba, the University plays a central role. Whereas ministry
Key: STA=Science and Technology Agency; Environ=Environmental Agency; Edu=Ministry of Education; Health=Ministry of Health and Welfare; Agri=Ministry of Agriculture, Forestry, and Fisheries; MITI=Ministry of International Trade and Industry; Trans=Ministry of Transportation; Constr=Ministry of Construction
Note: * This value is not applicable since the Environmental Agency has only one research institute in Tsukuba Science City
Table 11 The degree of interministry co-author relations is shown in matrix form, with the number of links between researchers of different ministries (percentages of each ministry’s links with each other ministry are in parentheses)
Collaboration networks 149
150
Collaboration networks
affiliation may be a barrier to collaboration among researchers at some rival ministries (such as MITI and the STA), other ministries have low or no barriers to collaboration, especially the Ministry of Education.
INSTITUTES AND RATE OF PUBLICATION Because of the differences in the number of researchers employed at the many institutes in Tsukuba Science City, the number of authorships alone is not a good measure of productivity among research institutes. To better enable some comparison about research productivity among research institutes, a measure of productivity was derived by dividing the number of authorships per institute by the number of researchers employed in that institute.81 Table 12 ranks Tsukuba Science City research institutes by their degree of research productivity. 82 The National Institute for Environmental Studies, the only research institute of the Environmental Agency located in Tsukuba Science City, ranked first. During 1988, the 177 researchers there each published an average of 4.21 articles. Fewer than one article per researcher (.77) of these 4.21 articles were published in English. Table 12 also lists language of publication by research institute.
FIRST AUTHORSHIP OF PUBLICATIONS To what extent is Tsukuba an intellectually independent research community? One way of possibly answering this question is by analyzing where the first-authors of research articles work. One of the major trends stated previously was a rapid increase in the amount of co-authorship between Tsukuba researchers and researchers in areas other than Tsukuba or Tokyo. If the first listed author of a research article can be assumed to have initiatied the research project which is represented by an article, a question about these increasing co-authorship relations becomes: Is the percentage of Tsukuba first-authors increasing or decreasing? Information about the trends of article initiation may provide a measure of the directionality of the research relationship, which, when aggregated, may reflect on the degree of independence of the Tsukuba research community.
Collaboration networks
151
Table 12 Productivity by Tsukuba Science City research institutes, as measured by dividing the total number of an institute’s researchers by the number of journal publications by those researchers during 1988
152
Collaboration networks
Table 12 continued
Note: *An additional 330 researchers joined this research institute from Tokyo after this analysis was conducted.
Collaboration networks
153
Figure 28 Trends in the percentages of first-authorships of research journal articles by Tsukuba Science City authors and their co-authors
The concept of dependency implies an unequal relationship, in which one co-author may be dependent on another co-author for vital knowledge, insight, or skills. Obviously, there are many shades of dependency, including interdependency (or complementarity) in which two co-authors are mutually dependent on one another.83
154
Collaboration networks
Here, trends of first-authorship by location are compared. Findings suggest that while the number of first-authors located in Tsukuba is relatively large, there are increases in the percentage of articles that have first-authors either overseas, or in Japan but not in Tsukuba or Tokyo. Also, the percentage of first-authorships in Tsukuba national research institutes and at the University of Tsukuba decreased between 1985 and 1988 (see Figure 28). Two interpretations of a decreasing percentage of first-authorship by Tsukuba researchers are plausible. Does the decrease mean that Tsukuba researchers are increasingly dependent on outside researchers, or does the decrease mean that Tsukuba researchers are increasingly in demand as partners by outside researchers? Perhaps the second interpretation is more correct. In Japan and among academics, Tsukuba is a center of physical sciences research. The number of researchers throughout Japan who come to Tsukuba to conduct experiments is growing, as is the number of graduate and post-graduate students from throughout Japan who come to Tsukuba for short-term research visits. Previously we mentioned the rapid rise in the number of privately employed researchers who come to the science city to work for one to two years in the national labs. These trends imply that researchers from other areas in Japan come to Tsukuba because they know that good research is being done there. While “in residence” in the science city, they write, as the first-author, a research article, in which their Tsukuba collaborators will be included as co-authors. The same logic may apply to the increase in the number of overseas first-authorships. Researchers in other countries are aware of good research being done in the science city. Their inclusion of Tsukuba co-authors reflects access to Tsukuba facilities and researchers.
SUMMARY Because of a rapidly increasing number of co-authorship relations between researchers at different Tsukuba institutes, the structure of the science city research network is increasingly strong. The strength value of.33 is high relative to other measured communication networks. The graphic network of research links in Tsukuba reflects this increasing complexity. Importantly, while the percentage of all links which are intrainstitutional decreases during our three time-periods of study,
Collaboration networks
155
the percentage of all links which are interinstitutional increases. These trends of a decreasing percentage of links within research institutes and an increasing percentage of links between research institutes hold across languages. Although the number of co-authorship links between university and national laboratory researchers has increased during each timeperiod of study, the percentage of all university-national lab links is decreasing in the face of rapidly increasing numbers of interlaboratory research relationships between government researchers. The percentage of links between Tsukuba researchers and researchers outside of Tsukuba and Tokyo is increasing. Within the science city, the largest trend in number and percentage is an increase in the links between researchers at different Tsukuba institutes. The results of our analysis of authorships support our findings about links. The percentage of single-author articles, like the percentage of articles co-authored by researchers in the same institute, has steadily decreased, while the percentage of articles coauthored by researchers in different institutes has steadily increased. Reflecting the increasingly diverse locations of the co-authors who work with Tsukuba authors, the percentage of authorships at Tsukuba national research institutes has decreased. The Tsukuba research institutes with the most authorships have remained more or less consistent in ranking, relative to one another. There is a wide variance in the numbers of articles authored by research institute, as well as a dichotomy in the language of publication. At some institutes, Japanese is the predominant language of publication. At other institutes, publications are in English. Ministry affiliation appears to be a barrier to research article coauthorship in most cases. Yet our results indicate that Tsukuba researchers at different institutes under the control of different ministries co-author twice as many articles together as do Tsukuba researchers at different institutes within the same ministry. The Ministry of Education has the largest number of co-authorship relations with researchers from other ministries, because of the involvement of professors and graduate students at the University of Tsukuba. When compared by one measure of research productivity, the National Institute for Environmental Studies ranked first, with the
156
Collaboration networks
highest number of research articles published per researcher (4.21) in 1988. Of all Tsukuba institutes with one or more publishing researchers, the Tsukuba Space Center and the Geographical Survey Institute tied for last place, with the lowest number of research articles published per researcher (0.27) in 1988. In the analysis of first-authorship, the largest change between the 1983–5 and 1986–8 time-periods was an increase in the number and percentage of first-authors who work in Japan but are not located in Tsukuba or in Tokyo.
Chapter 7
Lessons learned about growing science
There is still the same tension between the needs of the group and the wishes of individuals that existed in the period discussed [1868–1920] and the same fixation with international standards of research creativity and whether, if at all, Japan’s system can meet them. James R.Bartholomew, The Formation of Science in Japan (1989:276) At a time in history when geographic areas of concentrated scientific and technological resources are increasingly of interest to local, regional, and national governments throughout the world, Tsukuba Science City stands as a rare example of a large-scale 30year experiment in the organization of basic scientific and engineering research. In this book, a number of questions have been posed about Tsukuba Science City and how the data collected about Tsukuba might help in addressing issues about science development and science policy. These issues extend far beyond this one science city. Here, results pertinent to questions posed throughout the book are grouped, summarized, and interpreted. THE DEVELOPMENT OF TSUKUBA How has Tsukuba Science City come to exist? Is this a case of a strong and unified central government decreeing that a science city shall be built? Can a Tsukuba only be built in Japan because of a cultural ability of the Japanese to work harmoniously together? Previous analysts have characterized the development of large-
158
Lessons learned about growing science
scale science-based economic development projects in terms of uncoordinated cooperation, in which public sector agencies play a reactive role and private sector actors play complementary but uncoordinated roles in development. In Silicon Valley, uncoordinated cooperation occurred in the form of fortuitous defense contracts, which spurred the location of eager supply firms and drew talented engineers to the Palo Alto area, and led to semiconductor development and refinement, the readiness and ability of the engineering departments at Stanford University and the University of California, Berkeley, to supply faculty and graduates, and rampant communication networks and job mobility among entrepreneurs and engineers (Rogers and Larsen 1984; Saxenian 1985; Saxenian 1989). Key engineers cooperated in early ventures (such as Fairchild Semiconductor, and Hewlett-Packard), and participate in very intricate communication networks in order to exchange information. Uncoordinated cooperation is achieved through the pursuit of individual goals for as long as the achievement of these goals does not become communally counterproductive.84 Uncoordinated cooperation also well describes high-technology economic development in Britain (Hall et al. 1987). Studies of hightechnology development in other areas also associate economic development with cooperation, either through the making of formal policies (Aydalot and Keeble 1988), the benefits of industryuniversity-government partnerships (Whittington 1985), the necessity of interfirm networks (Segal Quince Wicksteed 1985; Hakansson 1989), or informal project-oriented pacts between state and local level politicians and bureaucrats, businesspeople, and academics (Smilor, Kozmetsky, and Gibson 1988b). So the achievement and management of cooperative economic development has been used by analysts not only to explain large-scale economic development in Japan, but also to explain cases of economic development based on science and technology at multiple sites in the United States, and in Britain, France, Italy, and India (Hall 1986; Aydalot and Keeble 1988; Smilor, Kozmetsky, and Gibson 1988a; Williams and Gibson 1990). These analyses of high-technology development, like some studies of large-scale economic development in Japan, rely on a metaphor of cooperation or harmonious relations to explain successful high-technology development. These authors assume that complementary outcomes in the development of research
Lessons learned about growing science
159
communities are possible only through cooperative action in which stakeholders share a common developmental goal. In fact, instances where diverse stakeholders share the same developmental goal are worthy of study precisely because they are rare instances of economic development. The literature on science and technology-based economic development fails to make this point. Far more common in the real world are instances in which competing firms, bickering bureaucracies, public interest groups, and politicians identify their own goals for participation in economic development initiatives. Public and private argument ensues and the extent of disagreements becomes clear; alliances form. This conflictory means through which large-scale science-based economic development occurs has been relatively overlooked by academics. Yet conflict alone cannot explain the existence of large-scale projects like Tsukuba. Unless there is some other dynamic relational process at work, conflict typically results in an aborted project, such as the superconducting supercollider in Texas, or a greatly lessened project, the customary fate of hundreds of so-called science or research parks in the United Kingdom and the United States. The historical research carried out here suggests that Tsukuba Science City does not exist because of a strong and unified central government, nor because of a unique cultural ability of the Japanese to work in harmony. From the early 1960s until 1966, the plans for Tsukuba Science City progressed via strategic ambiguity, a process of encouraging diverse stakeholders to understand the developmental project as a solution to their own problems. 85 By encouraging collaborators to focus on their own developmental goals by strategically communicating with them, project coordinators avoided overly negotiated (that is, lessened) developmental outcomes. The developmental goal of creating Tsukuba Science City was presented to various project stakeholders in terms of solving their particular problems, without sharing with each stakeholder how the project was being portrayed to other important stakeholders. Other stakeholders were concerned with other problems (scientific research facilities, overcrowding in Tokyo, vocal political opponents, etc.). With a loss of leadership in the person of Kono¯ Ichiro¯, however, came a loss of ambiguity, and the project unraveled into a conflictory economic development process. Conflict simultaneously enabled the implementation of the science city plan, yet inherently limited its effectiveness. Beginning in 1966 and continuing through 1978, ongoing negotiation and construction took the evolving
160
Lessons learned about growing science
science city further and further away from the elaborate plans that Takayama Eika had overseen and that Kono¯ Ichiro¯ had been able to protect. The conduct of scientific and technological research today in Tsukuba Science City is partly the result of this conflictory means of large-scale economic development. Specific problems associated with the contemporary science city can be traced to these 12 years of conflictory, negotiated development. Political and bureaucratic opportunism, a lack of project coordination, and constant administrative rationalization of original strategic plans well characterize the developmental history of the science city. Each of these conditions continues to hamper the economic development potential of Tsukuba Science City and each has a negative impact on the rate of scientific discovery and technology innovation in the science city. What lessons for the planning and implementation of large-scale science-based economic development can be learned from the history of Tsukuba Science City? 1
The need to solve diverse economic, political, and social problems presents opportunities for leaders to coalesce solutions in terms of a single economic development plan. Combining solutions (or at least giving the appearance of doing so) is a requirement, since the competition for resource allocation typically necessitates a joint solution. In the case of Tsukuba Science City, the administrative cum education cum science city plan was interpreted as a solution to six national problems. A new city was not understood as a solution to each individual problem by experiencing it; rather, the solution offered by a new city was strategically ambiguous so that different people could interpret the city plan in different ways. For example, conservative politicians interpreted the plan as a means of ridding the conservative Liberal Democratic Party of the Tokyo Education University, which was a training school for educators, many of whom became outspoken socialists and communists. Science bureaucrats within the Science and Technology Agency, on the other hand, interpreted the plan as a means of providing scientists with new facilities, while economic planners within MITI interpreted the plan as a means of supporting and advancing Japanese industrial policy. The process of implementing the science city plan follows the Japanese government ministry tradition of promoting a grand
Lessons learned about growing science
2
3
161
theme unfettered by technical details, which is subsequently made more and more rational. This means of economic development planning and implementation gets results: Tsukuba Science City functions with some notable scientific success. A strong political champion can wield the power necessary to force compliance on the part of antagonistic government bureaucracies to a large-scale plan for economic development. Kono¯ Ichiro¯, driven by personal ambition, was absolutely necessary for managing and channeling conflict in productive ways during the city planning process. The loss of leadership during the beginning of the implementation process did not ruin the project, but it did severely compromise it, which was manifest in the loss of ability to control the bureaucracies. The project is only now recovering from his death as more control for daily operations of the science city has been transferred to Ibaraki Prefecture and as researchers and residents are beginning to perceive the science city as a desirable place to live. Conflict among government units during the implementation process of a large-scale economic development plan may lessen or delay the achievement of the plan’s goals. In the present case, competition among Japan’s national ministries and agencies largely determined how the innovation of a science city was implemented and how Tsukuba functions today. Conflictory tendencies among bureaucratic personnel were manifest in continual attempts to maximize each bureaucracy’s share of the new science city. This tendency was a rational course of action for each bureaucracy to pursue. The irrational result was akin to a “tragedy of the commons” in which each ministry sought to maximize its own share of the new city, to the detriment of the larger science city goal (of allowing government scientists a place to work with a comfortable environment where it is easy to communicate, collaborate, and be creative). Residential, social, and industrial aspects of the science city were sacrificed in order to placate the demands of the negotiators of each bureaucracy for ever larger shares of the city. Bureaucratic competition is the root cause of contemporary conditions such as the physical shape of the city, the lack of formal relations among many of its research institutes, a lack
162
4
5
6
Lessons learned about growing science
of shared scientific facilities, and a lack of social opportunities. To an extent, the people responsible for conceptualizing and planning the science city also are important in the implementation process. Although the plans for the science city were severely compromised during the years of implementation, the theme of a science city survived, and the city was built and is operating. Involving the people who will be responsible for the implementation of a large-scale innovation in the planning process engenders their commitment to seeing the project through. The meaning of large-scale economic development projects may change over time. The Japanese government today maintains that Tsukuba was conceptualized as a city where scientists could think and create, but the science city was not initially planned to be a research community. Science was a post hoc rationale. The science city (which is still referred to as an education city by some Japanese) was an academic town, then an administrative town, then a city of science. The conflictory developmental process described here is fraught with ill effects for the conduct of scientific research. The faults and negative characteristics of Tsukuba Science City result from alterations made to the plans to build the science city. Most of the contemporary problems of Tsukuba Science City had been anticipated and planned against by urban planners. Originators and planners of the science city concept foresaw the importance of creating an informal social environment in the city, for example, but their ideas were successively whittled away as planning proceeded into negotiation and development. The conflictory implementation process detailed here may have its most insidious effects on the contemporary research culture in Tsukuba Science City. The task of increasing communication among Tsukuba researchers is now a foremost, explicit concern among science planners assigned to the city, as well as among planners of other large-scale science-based projects, such as the new Kansai Science City. If other research communities are planned with a detailed knowledge of the influence which politics and economics had on Tsukuba’s development, it is possible that some of the contemporary problems in Tsukuba may be avoided elsewhere.
Lessons learned about growing science
163
Tsukuba itself is so young that current obstacles to better communication, collaboration, and creativity may well recede with appropriate local science policy.
COMMUNICATION, COLLABORATION, AND CREATIVITY If communication is robust and occurs frequently among diverse specialists, is the likelihood of collaboration and creativity heightened? Must individual researchers initiate communication with each other (and thus perceive that they are autonomous and in control), or can governments play a proactive and effective role in encouraging such relationships? What happens when a government intervenes in a massive way to “turbocharge” scientific proximity, communication, collaboration, and creativity? The personal interviews with researchers, the survey of the leaders of informal study groups, and the network analysis of coauthorship patterns provide the data from which the following lessons are derived. 7
In general, self-initiated, intrinsically motivated communication among researchers from different disciplinary backgrounds does in many cases lead to collaboration and creativity. When researchers organize themselves to communicate with colleagues, their initiatives serve multiple important purposes, such as the sharing of technical information, confirmation of research importance by professional colleagues, and social support and relaxation. The results of such informal group gatherings are impressive, and sometimes lead not only to small collaborations and publications, but also to the institutionalization of coordinated research programs involving over 50 researchers, and even to the establishment of national research laboratories, in the case of genetics research in Tsukuba. Over 60 percent of study group organizers replied that their members either had conducted collaborative research among themselves, probably had done so, or were planning to do so. Half of these survey respondents believed that collaborative research projects resulted from discussions in the informal group. Investigator-initiated informal groups can also become ineffective and wholly irrational (that is, they may
164
8
9
Lessons learned about growing science
persist and prosper without satisfying their original goals). Intrinsic motivation and individual control are both important for producing participation by researchers that is task-oriented and enthusiastic. Intrinsic motivation is positively related to creative output; extrinsic motivation is negatively related to creative output. Japan’s national ERATO program, which provides star researchers and their research teams with research grants for several years to study especially adventurous topics, is concentrated in Tsukuba and is an example of organizing basic research so that it is both intrinsically motivating to the people involved and controlled by the researchers, not research administrators. The results point clearly to the contribution which informal groups might make toward fostering creativity within the constraints of a very planned system. Yet most of the plans for the city did not include the means of encouraging or rewarding interpersonal communication among researchers. The strategy of agglomerating scientific resources and personnel is leading to more communication and collaboration among researchers in Tsukuba Science City. Results of the co-authorship analyses suggest that the Tsukuba Science City coauthorship network is growing in both the number of organizations involved and in the strength of bilateral research relationships. The present analysis did not test whether the rate of growth of scientific work in Tsukuba is greater or lesser than the rate of scientific work in any other research community or social system. Informal groups are most important in that they give birth to interpersonal networks through which relatively high-value knowledge may be exchanged. This is not the explicit purpose of informal groups. These groups were formed to solve the general problem of scientists not having enough knowledge about hot research topics, such as biotechnology. The frankness of discussions in informal groups is limited, being linked to the degree of participation of private industry scientists employed by competing firms, and government researchers employed by rival ministries or laboratories. Thus the value of the knowledge discussed in informal group meetings is moderate. However, common or complementary research interests, as well as social interests, may lead to a hardening of friendships. These private, interpersonal relations
Lessons learned about growing science
165
serve as the means of transferring high-value knowledge among researchers. Such relations often link academic or government scientists with corporate scientists, government scientists with academics, and occasionally lead to collaboration between corporations. The most successful of such groups provide resources without centralized control, and are the products of determined leaders and/or respected researchers. 10 While the percentage of all links which are intra-institutional has decreased, the percentage of all links which are interinstitutional has increased. Network and statistical analyses of data about journal articles authored by Tsukuba researchers and their colleagues indicate that the structure of the science city research network is increasingly complex and strong because of a rapidly increasing number of co-author links between researchers at different Tsukuba research institutes. These juxtaposed trends hold across languages and across the somewhat confounded variables of links, articles, and authorships. 11 By a ratio of 2 to 1, Tsukuba co-authors in different labs cross parent ministry boundaries in their co-authorship of journal articles. This finding suggests that the planned objective of breaking down institutional and parent ministry barriers to cooperative research is succeeding. 12 A research university, the University of Tsukuba, is central to the Tsukuba co-authorship network. With the high visibility and prestige of its new president, the emigré Nobel physicist Leo Esaki, the centrality of the university in the local coauthorship network can be expected to increase. Also, both the new professorial system (renkei daigakuin), which allows researchers at the national labs and at private labs an adjunct affiliation with the university, and the fact that the status of professors is higher than the status of researchers at either the national labs or in private industry will continue to drive collaboration. The University of Tsukuba would be expected to have a central position in a Tsukuba co-authorship network for two other reasons. Like universities in general, the University of Tsukuba is a multidisciplinary organization. Thus, it would be expected to form a bridge between many research institutes which have more narrowly defined research agendas. Second, approximately 1,000 graduate
166
Lessons learned about growing science
students of the university work as research assistants in Tsukuba national and private laboratories every year. Some of these students undoubtedly show up as research article coauthors. Relationships between University of Tsukuba faculty, notably in chemistry and engineering, and researchers at the nearby government research institutes and private research labs in Tsukuba and Tokyo are close. Paid consulting is common. Faculty consulting serves to transfer knowledge while maintaining interpersonal networks which are instrumental in finding jobs for university graduates. Faculty-industry networks are more efficacious than graduate student-government laboratory networks. Even though approximately 1,000 graduate students work in the Tsukuba national labs each year, the university does not serve as a feeder school for the local national labs because the government hires from a national pool of applicants, and new government researchers are not necessarily assigned to their choice of laboratory or location. 13 Informal groups in Tsukuba Science City are most effective if they include the participation of private industry researchers. Having participants who are interested in the same phenomena, represent diverse organizations, and have different uses for the phenomena of study lessens competition among informal group members while increasing the potential for complementary and collaborative research projects. Study groups with strong representations of industrial researchers remain the most intellectually active groups over a period of years. 14 Multidisciplinary communication, not just interdisciplinary communication, should be an aim of research community planners. Interdisciplinary research involves researchers from different disciplines. Multidisciplinary research brings together diverse specialists to focus on a problem of common concern. Multidisciplinary communication initiatives in Tsukuba have been more successful than interdisciplinary initiatives in leading to collaboration and creativity. Similarly interested people from different disciplines and from different organizations provide for the richest, most exciting informal research groups. 15 Means of considering pre-competitive corporate knowledge as public goods are essential for continuing to improve the
Lessons learned about growing science
167
communicative climate in the science city. Even though, in Tsukuba, there exist important rivalries between ministries and their researchers, a great deal of knowledge exchange does occur there in the realm of public goods, led by the national research laboratories. The increasing privatization of Tsukuba, however, necessitates a government-organized corporate effort to facilitate the regular and multiple direction flow of precompetitive technical information. Communication is a two-way process of sharing knowledge. Privately employed researchers can only expect to learn valuable information in discussions and group meetings if they also provide valuable information. If top management is not willing to allow its researchers to share knowledge, those researchers will not be welcomed for long in public discussions of research progress and results. 16 Geographic proximity is a necessary but not sufficient condition for many collaborations. Previous studies of proximity and communication show that co-location is positively related to communication. Collaboration, however, cannot be assumed to occur as a result of communication. Reliance on this unwarranted assumption characterizes many so-called research or science parks in the world: mere colocation is taken to represent a synergistic effect. Proximity is necessary for many eventual collaborations, but intrinsic interests of the researchers, manifest in multidisciplinary communication, is probably more important in leading to collaboration and creativity. 17 A lack of administrative commitment combined with scientist disdain for top-down bureaucratic control leads to the failure of several government communication initiatives. The formal authority of science administrators was imposed on researchers in order to force communication among scientists of different research institutes. Basic researchers employed by the Japanese government value autonomy very highly, as do scientists in Western countries. Clearly, however, there is an important role for institutions that exist to promote the free flow of knowledge among researchers. When locally created, institutions for furthering communication and collaboration are well received by researchers and supportive of direct initiatives by researchers. The important work in Tsukuba Science City by Kawamoto Tetsuzo¯ at two of these “bridging” organizations is an excellent example of how institutions can support interorganizational
168
Lessons learned about growing science
communication and collaboration at the individual and research group levels. 18 The function of informal groups in Tsukuba, particularly of intradisciplinary groups, has changed since their inception. Most of the groups, which can be categorized as minor rebellions against the established organization of research which prevailed in Tokyo during the late 1970s, have themselves become institutionalized. As a result of a changing research environment in Tsukuba and scientific disciplines, not all of the groups which were once effective in technology transfer are still so effective. Groups organized to address interdisciplinary research topics tend to function rationally (that is, they serve their original purpose of transferring technology) but their lifespan is short. As soon as worldwide knowledge about a new interdisciplinary field becomes somewhat well understood and traditional means of knowledge exchange appear (such as journals and conferences), these groups disband. Sometimes the novel emphases of these groups become routinized in the establishment of formal research labs. Interdisciplinary groups tend not to outlive their function. Intradisciplinary groups, which bring together homophilous researchers, address topics which are comparatively well understood. These informal groups last longer than interdisciplinary groups, but their function becomes irrational. Intradisciplinary groups in Tsukuba appear to exist primarily to confer status on their leaders. The institutionalization of intradisciplinary groups reflects the intellectual conservatism which characterizes aging scientists, which serves as a disincentive for young researchers to join the groups. 19 The groups of linked laboratories identified in the network graphs are centered around the two ministries with (1) the largest numbers of researchers in Tsukuba, and (2) the two most geographically proximate complexes of laboratories. The network graphs in Chapter 6 show clear groups of r e s e a r c h i n s t i t u t e s b y t wo b r o a d c o m b i n a t i o n s o f disciplinary fields. These are basically the MITI labs and private industrial labs in Tsukuba, and the agricultural labs u n d e r t h e c o n t r o l o f t h e M i n i s t r y o f Fo r e s t r y a n d Fisheries. For increasing co-authorship at least, a critical mass of researchers who are grouped close together produces results.
Lessons learned about growing science
169
20 The central position of the University of Tsukuba points to the quantity as opposed to quality bias of the co-authorship measure. The university, while it is strong in several departments, is not one of Japan’s top-level research universities.86 In contrast, the government researchers employed at the Tsukuba research laboratories are the cream of the crop of Japanese scientists. Work done at the Electrotechnical Laboratory, the National Institute for Inorganic Materials, the National Laboratory for High-energy Physics, and several other of the government research labs is acknowledged as worldclass basic research. Proportionately much less of the work performed at the University of Tsukuba achieves this level of excellence. If a Tsukuba research network based on quality of work could be graphed, the University of Tsukuba would appear peripheral, not central. 87 However, the recent establishment of adjunct professorships at the university may lessen this distinction between quality and quantity in publications, depending on how Japanese researchers choose to list their affiliations on journal articles. Because a university affiliation is very prestigious in Japan, many of Tsukuba’s best scientists working in the national labs may begin to list a University of Tsukuba affiliation in addition to their laboratory affiliation. Several findings from the network analysis may be explained by antecedent and current conditions in Tsukuba. Although the city was very comprehensively planned during the 1960s, implementation of the plan was severely compromised by warring bureaucrats of the eight national ministries targeted to move research institutes to the new science city. Housing and social opportunities were squeezed out, while the land and resources dedicated to research institutes increased. As a consequence, many researchers worked in Tsukuba in the late 1970s and early 1980s but refused to live there. This condition perhaps contributed to the low amount of research collaboration during 1979 and 1983 among researchers of different institutes, since they would have difficulty meeting. Similarly, a bureaucratic planning bias against involving private R&D labs in the design and initial construction of the city may have contributed to the pattern in our data of closely integrated government labs and peripheral corporate labs.
170
Lessons learned about growing science
The finding that the number of co-author links between Tsukuba research institutes and organizations in areas outside of Tsukuba, and in other places (not Tsukuba and not Tokyo) are very large compared to the number of Tsukuba-to-Tsukuba interorganizational links may be explained in part by a high demand from researchers throughout Japan to gain access to scientific expertise and equipment in the science city. Conceptualizing, measuring, and interpreting single coauthorship networks for geographically bound research communities provide useful insights about the strength of interorganizational relations, the extent of disciplinary groups of research institutes, and the overall degree of integration of a research community. Such knowledge is of practical use to science policy-makers. When network approaches such as the present investigation are combined with historical and verstehen approaches in the study of research communities, research communities may become recognized as a most fruitful conceptual level for understanding the sociology of science. One important issue with which science planners will have to grapple is how much communication should be encouraged between dissimilar researchers (Dearing 1993). A common assumption is that more communication is better. From a systems or community perspective, however, diminishing returns are theoretically possible from researchers becoming too similar over time. The present investigation only saw evidence of “diminishing returns” at the level of small groups, since some of the informal groups of researchers studied did indeed become boring to participants over time. It is suggested here that despite the possibility of diminishing returns as a result of researchers becoming somewhat similar in their perceptions of each other’s work, reducing perceptions of difference and inapplicability (that is, difference reduction) will continue to be a larger challenge for people who seek to increase communication among researchers. ACCESS TO KNOWLEDGE IN TSUKUBA By focusing on the different ways in which communication does and does not occur in Tsukuba, it is possible to suggest how people who do not live in the science city may best gain access to knowledge in the city. A large number of technology-intensive
Lessons learned about growing science
171
companies, for example, conduct regular surveillance of scientific advances in the city by stationing employees there. Simply living in Tsukuba, however, does not equate with being in communication with researchers, as the communication tendencies of the researchers themselves point out. The problem of access to scientific and technological information is a problem of gaining access to communication networks. Formal communication networks are good routes for being introduced into the scientific community of Tsukuba Science City. However, formal networks are poor sources of valuable science and technology information. If the Japanese government wishes to allow people more access to science and technology information in Tsukuba, its efforts should not stop with arranging participation (especially shortterm participation) in formal communication networks. In most cases, only limited access to information, or access to low-value information, will result. Arranging a “scientific mentor” who can introduce the visitor into informal communication networks in Tsukuba may be a worthwhile alternative or supplement to just arranging participation in formal networks (such as exchange programs). One means of indoctrination is participation in informal study groups. Informal groups, many of which meet monthly, discuss research, and socialize, are most important in that they give birth to interpersonal networks through which relatively high-value knowledge may be exchanged. This is not the explicit purpose of informal groups. These groups were formed to solve the general problem of scientists: not having enough knowledge about hot research topics. The frankness of discussions in informal groups is limited, being linked to the degree of participation of private industry scientists employed by competing firms, and government researchers employed by rival ministries or laboratories. Thus the value of the knowledge discussed in informal group meetings varies according to group membership. However, common or complementary research interests, as well as social interests, may then lead to a hardening of friendships. These private, interpersonal relations serve as the means of transferring high-value knowledge among researchers. Such relations often link academic or government scientists with corporate scientists, government scientists with academics, and sometimes lead to collaboration between corporations.
172
Lessons learned about growing science
THE PRIVATIZATION OF TSUKUBA SCIENCE CITY In the life-cycle of a city, Tsukuba is still an infant, and will change greatly. For example, whereas the budgets for Tsukuba national laboratories remained about constant from 1980 to 1990 (and have decreased for some of the national labs), hiring and funding levels are rising rapidly at the private laboratories (there is, however, a renewed commitment by the national government to fund basic research, at least at its national universities, starting with an overhaul of the infrastructure of the University of Tokyo). By the year 2000, research activity in Tsukuba Science City (in terms of number of labs, number of employees, and research budgets) will be dominated by private industry, which will tie Tsukuba more closely into worldwide corporate research and development networks. If corporate basic research does come to dominate Tsukuba, the city may become less distinct from other science cities, such as Kansai Science City which has at its core corporate basic research. A primary theme in this book is that a science city can be grown through increased communication, collaboration, and creativity among researchers. A research area that is dominated by corporate basic research will not likely be characterized by this developmental progression (this idea is discussed in the following section, too). A corporate-dominated research culture will develop differently due to corporate rules about proprietary rights and the need to withhold, not share, information. In some research communities which are dominated by private research firms, such as the Silicon Valley area of central California, a preponderance of private research and development firms does not for the most part inhibit the development of interpersonal communication networks among engineers and scientists. Yet Silicon Valley does not have corporate research at its core. Although many of the initial start-up high-technology firms which made the valley famous in the 1980s did mature into large corporations, the valley’s culture has since rebounded into what it once was: a network of entrepreneurs who are interdependent with other entrepreneurs to create, manufacture, distribute, and market high-technology products and services. Corporate dominance and professional management threatened the dynamism of Silicon Valley by closing off interorganizational communication of employees. Communication and formal as well
Lessons learned about growing science
173
as informal collaboration were vital to Silicon Valley. Entrepreneurs and their employees “trade know-how” with each other to survive and prosper. Corporations, on the other hand, control resources by prohibiting the giving of information by their employees. There is already evidence of a limiting of information among researchers in Tsukuba. Even though most of the research occurring in private labs in Tsukuba is of a precompetitive type, administrators of corporations recently locating to the science city worry about nearby competitors. They are secretive about what research projects are under way in their laboratories. Particularly secretive are lab employees of pharmaceutical and electronics corporations, industries in which strategic advantage can quickly reverse if trade secrets are learned by rival firms. At a time when many Japanese who only hear about Tsukuba or live in Tsukuba are becoming more favorable in their opinions of the science city, expert observers of informal communication trends in the city are becoming more pessimistic (Kawamoto 1992a). COMPARISON WITH KANSAI SCIENCE CITY Another example of large-scale science development is just taking shape in Japan. Kansai Science City (or Keihanna, in which “kei” comes from the kanji for the city of Kyoto, “han” is taken from the kanji for Osaka, and “na” stands for the smaller city of Nara) was planned as the Kansai region’s answer to Tsukuba Science City in the Kanto¯ region, and its proximity to Tokyo. The capital city, built on the Kanto¯ Plain, is a cultural, political, and economic rival of the cities in the Kansai region, of which both Nara and later Kyoto served as the capital of Japan.88 The total size of Kansai Science City is just over half the size of Tsukuba Science City (38,053 acres to 70,543 acres). Japan’s first graduate-level-only university, the Nara Institute of Science and Technology, and a new campus of Doshisha University are joined with small campuses of Doshisha Women’s College of Liberal Arts, Osaka International University, Kansai University of Foreign Studies, the Hotani campus of Kansai Gaidai College, and Osaka Electrocommunication University in the new science city. Kansai Science City’s central scientific and cultural facilities, Keihanna Plaza, opened in April 1993, and the formal opening of the science city occurred in October 1994. A consortium of 170 companies
174
Lessons learned about growing science
contributed 44 percent of the money for Keihanna Plaza. The Japanese government contributed the rest of the money (Holden 1993). A second shared-use facility, Takayama Science Plaza, is also open. Fifty billion yen had been invested in the new science city as of late 1994 (Shimoda 1994). Yet the economic recession in Japan has inhibited many possible investors from committing money to the project. Kansai Science City was largely the idea of former University of Kyoto Vice-Chancellor Okuda Azuma, in 1978. The national government assisted the original study group by conducting land use plans of the area (the National Land Agency and four other ministries have been involved in planning), and designating the city a national project, which gave high status to the initiative and made convincing corporate investors easier. Planning took 10 years, until 1988, when construction of shared-use facilities and laboratories began. The concept and its translation into a plan were and remain complicated because of the involvement of the three cities as well as 12 local districts. The land for the science city is still being purchased by Keihanna Incorporated, which is a problem due to wildly escalating land speculation (Takayama 1992). Whereas one tsubo (3.3 square meters) of land in Tsukuba was bought by the government 30 years ago at a price of 2,000 yen, one tsubo of land in Kansai Science City now commands more than 200,000 yen. 89 Kansai Science City has benefited from hindsight by Tsukuba’s earlier development. Nevertheless, not buying all of the land for the Kansai project early on in the planning process is a mistake which should not have been made, since the same mistake was made with the Tsukuba project 30 years earlier (Takayama 1992). Some of Kansai Science City’s shared-use facilities, such as the International Institute for Advanced Studies, stand nearly empty because of the joint negative effect of the very high land prices in the Keihanna area and the Japanese recession in the early and mid-1990s. The objective of Kansai Science City is cultural as well as scientific discovery. Whereas the arts, social sciences, and humanities are mostly ignored as topics of concentrated research in Tsukuba, they will be emphasized in Kansai Science City, as will the physical sciences and, especially, the life sciences. This weaving together of the sciences and the arts may be thought of as a unique Asian attempt to resolve the “third culture debate” in academia
Lessons learned about growing science
175
(Lakoff 1966). Shimoda Minoru of the ARPAK International Institute of Kyoto and designer of the science city is attempting to blend urban facilities, rural villages and agricultural land, and existing culture with the new influence of about 40 large corporation research laboratories, public research institutes, and universities and colleges, into one academic research community (Shimoda 1994). Kansai Science City is located in a beautiful area that is already inhabited by a number of suburban and rural communities. Some of Tsukuba’s most renowned problems, of social isolation and a culturally sterile environment, will not occur in Kansai Science City. It is a very desirable location in Japan. But distinct geography and purpose alone do not capture two essential differences between Kansai Science City and Tsukuba Science City. First, whereas Tsukuba was a top-down government project, Kansai is a bottom-up project that has involved local residents in decision-making. Early community involvement can be expected to lead to closer relations between indigenous residents and researchers and support staff. Second, whereas Tsukuba was planned and developed as a government basic research area into which industry was later invited (and only grudgingly accepted), the Keihanna project was initiated by academics and is being led by large corporate investors (Housing and Urban Development Corporation 1989; Kyoto Prefectural Government 1991; Chiba 1992). The 12 clusters of science and cultural development will be home to private laboratories first and government research labs second. Matsushita Electric (with about 150 researchers in 1994), the central research lab of Daiwa (with 150 researchers), Shimadzu (with about 60 researchers), Kyocera (with 50 researchers), Canon (with about 110 researchers), and many other companies have built or plan to build research facilities in the science city (Kansai Research Institute 1994a; Kansai Research Institute 1994b). Considering the primary role of Japanese corporations in the conduct of basic research, this is a wise planning decision. These corporations are joined by several public research labs, such as the Advanced Telecommunications Research Institute International (with about 250 researchers, the most yet in the science city), and the Ion Engineering Research Institute Corporation (with 10 researchers). In late 1994, about 1,500 researchers were working in Kansai Science City (Shimoda 1994). Increasing the rate and diversifying the pattern of research-
176
Lessons learned about growing science
based innovation collaboration has been increasingly difficult in Tsukuba as private corporations have built a larger presence there. It is likely that proprietary rights and secrecy will be a major obstacle to the development of Kansai Science City due to the central involvement of major corporations. Major corporations establishing laboratories in Kansai Science City have highly professional management, many forms of which tend to overcontrol employee communication, which will result in a stifling of interorganizational communication and know-how trading. An increase in managerial control over scientific and technological information hurt bottom-up informal communication among researchers in Silicon Valley and in Tsukuba. The problem of overcontrol by technology-intensive and competing corporations that are linked in a broader strategic purpose have been well documented by David Gibson and Everett Rogers (1994) in their study of the privately led U.S. response to the success of the Japanese microelectronics industry. Communication was curtailed in truly ingenious ways that are rational to management within any one firm, but not to policy-makers who work to realize the creative potential of a research community. So from a science communication perspective, Kansai Science City appears fundamentally flawed in its design. Attempts to plan against the privatization of knowledge in Kansai Science City resulted in the construction of shared-use facilities, but these are few and far between. Moreover, their presence does not address the critical question of research management and the control of knowhow trading by corporations. The flaws in the organization of science in Tsukuba derive not from how the city was planned, but from how those plans were curtailed as they were implemented over a 10-year period. What should be done in Kansai Science City? Intercorporate networks at the level of individual researchers and research teams should be strongly encouraged. Researchers resent managerial control over them and their resources. This usually represents the imposition of extrinsic motivation for the researcher. Management at different corporations must appreciate the synergistic advantages of horizontal linkages among organizations through oneto-one and team-to-team informal communication and competition. Perhaps most importantly, the leadership of public universities in creating the idea and overseeing the development of Kansai Science City should continue through the implementation and beginning
Lessons learned about growing science
177
operation of the science city. A difference between what was planned and what was implemented was the major flaw in the development of Tsukuba Science City. Outstanding and thoughtful plans for an ideal research community were severely and detrimentally altered during implementation. Competition and fighting among national ministries led to the exclusion, over time, of very important elements of the research community in Tsukuba’s master planning process. Small, incremental changes which seemed rational at the time to their proponents functioned irrationally in the long term, to make the achievement of Tsukuba’s larger goals more difficult. Kansai science policy planners must find ways of encouraging open communication and collaboration among researchers from different corporations at a pre-competitive stage in research. THE FUTURE OF TSUKUBA SCIENCE CITY Aside from the impressive level of public investment and the rapid rate at which private R&D labs are opening, the newness of Tsukuba Science City, a lack of night-time meeting places, and a transportation system based on automobiles combine to make Tsukuba a peaceful but still somewhat sterile place to live and work. This characterization will change. The future of Tsukuba Science City is being reshaped by the construction of a new 60-kilometer high-speed train line between Tokyo Station and the center of Tsukuba. This line will be operational by the year 2005. From Tokyo Station to the outskirts of Tokyo, the New Jo¯ban Line will be underground; it then becomes an above-ground train. The train must be underground in Tokyo because there is no more room for above-ground trains in the capital. Spectacularly high land prices in Tokyo mean that the very short underground segment from Tokyo Station to nearby Akihabara Station will account for about one-third of the total construction cost of the line. The Ministry of Transportation estimates that the line will cost 600 billion yen, but estimates commissioned by JR Higashi Nihon, the private rail company with rail construction and operating right-of-ways, are that the line will cost from 800 to 1,000 billion yen. This discrepancy has led JR Higashi Nihon to delay the purchase of land and construction in an attempt to extract public subsidies or a cost guarantee (Ministry of Transportation 1985).
178
Lessons learned about growing science
According to one transportation expert,90 there is no question that construction of the New Jo¯ban Line will be completed because passenger numbers are at dangerously high levels on the current Jo¯ban Line. Also, at least two other private rail companies want to build the new line if JR Higashi Nihon declines its construction and operation rights. Each of the three prefectures (Ibaraki, Saitama, and Chiba) through which the line passes is in favor of the project.91 The Ministry of Transportation and the prefectures have considered the creation of a special public/private company that would build and own the railway, and the hiring of JR Higashi Nihon to operate the service (Kurokawa 1989). Anticipation about the New Jo¯ban Line is encouraging land speculation in Tsukuba by Tokyo-based real estate investment firms. Land speculation is one reason why fields and relatively inexpensive buildings continue to exist in Tsukuba: owners are holding onto Tsukuba real estate until the train begins operating, anticipating that land prices will jump dramatically. It is expected that Tsukuba will then become an attractive “bedroom town” for commuters who work in Tokyo. The population of Tsukuba may rise suddenly to the levels foreseen by early science city planners once the New Jo¯ban Line opens. Residents and observers of Tsukuba have already seen massive land value appreciation. This scenario presents at least two possible futures for the science city. In the first scenario, Tsukuba will be swamped with residential housing and accompanying “normal” suburban development which will smother the unique character of the science city, to the detriment of scientific and technological innovation. The freedom and isolation which are important to the progress of scientific work (Holzner, Campbell, and Shahidullah 1985:310) will disappear. Tokyo will engulf Tsukuba. People who work in Tokyo will move to Tsukuba to live, since the one-hour commute via the new train will take less time than current commuting times from other suburban areas of Tokyo. This scenario represents a curious twist of what many Tsukuba planners and residents consider the major contemporary problems of the science city: its transportation link to Tokyo is inconvenient, and many government researchers continue to live in Tokyo and commute to Tsukuba, rather than living and working in the science city. Inconvenient transportation and Tokyo-to-Tsukuba commuters are the current problem; too convenient transportation and Tsukuba-to-Tokyo commuters will be the future problem.
Lessons learned about growing science
179
According to one former elected representative of the Tsukuba area who is most concerned with public welfare and education, “the New Jo¯ban Line will ruin the best things about Tsukuba” (Shima 1989). The second scenario is more optimistic. By the year 2005, science and technological collaboration will be frequent among Tsukuba researchers and will strongly bind the community. The city will be firmly established as a source of scientific and technological innovation. Therefore Tsukuba Science City will be able to withstand the influx of metropolitan nonresearchers, and the new train line will primarily serve to make access to Tokyo easier for Tsukuba residents (Dohi 1989b; Mikado 1989). Regardless of which scenario turns out to be closer to reality, many Tsukuba residents agree that the New Jo¯ban Line represents the fourth major phase in the science city’s history, from science city construction, to the relocation of the national research labs to Tsukuba, to the present-day rapid increase of private R&D labs, to the eventual opening of the New Jo¯ban Line and the maturing of the city. Grand government plans for the science city (National Land Agency 1989) and for the greater Southern Ibaraki Prefecture area in which Tsukuba is located (Tsukuba Mook 1989) center around the new rail line. The population of the science city is expected to double to 320,000, while Tsukuba and 15 surrounding cities, towns, and villages are expected to combine for a population of 1 million. The 30 years since the passing of the legislation to create the “government and administrative city” have unraveled in unforeseen ways for the people who live in Tsukuba. The farmers who did not sell their property to the government for the construction of the city and stayed on have changed from being disregarded by the knowledge elite who moved to the new city from Tokyo, to being envied as millionaires. The land that they own is now too costly for the construction of homes. Instead, they are courted for their land by major corporations which seek to build laboratories in Tsukuba. With the huge influx of private corporation money has come a number of new supportive businesses and services. This development makes the science city more desirable as a place to live and buy a home to most researchers. Yet they cannot pay the going rate for the available land that is still held by local farmers and real estate brokers. Researchers complain that the farmers are benefiting unduly
180
Lessons learned about growing science
from the researchers’ efforts, which have established Tsukuba as a place of science. The traditional and science cultures in Tsukuba have changed greatly since 1963, but they are perhaps no more integrated in 1995.
Notes
CHAPTER 1 INTRODUCTION 1
This investigation was conducted as an embedded case study, a research strategy in which one case study is investigated (Tsukuba as a whole), as well as specified cases within that case, which are interpersonal and interorganizational relationships (Yin 1984). Many investigators of “technology transfer,” for example, focus on the possible matrix of relationships among high-technology industries and a local university, and frame their analysis in terms of the city or community as a whole (Segal Quince Wicksteed 1985; Olofsson et al. 1987; McMullan et al. 1987; Smilor, Kozmetski, and Gibson 1988b). These studies have one unit of analysis, specific research relationships within the city or community. Some of these studies seek to characterize the larger city or community (a second unit of analysis) in terms of the specific technology transfer relationships that are investigated. This somewhat inductive approach to understanding technology transfer can be contrasted with other studies which take more of a deductive approach; they start with the city or community (Rogers and Larsen 1984) or country (Tatsuno 1986) as the explicit unit of analysis and then seek to characterize specific relationships within in terms of the larger social system. Each method of investigation yields findings which the other method does not necessarily reveal. For example, by targeting the entire Silicon Valley community, Rogers and Larsen (1984) were able to show such deleterious effects of rapid high-technology development as wage inequities, workplace health hazards, and increased traffic and pollution. On the other hand, by focusing specifically on department policies at the University of Calgary, McMullan et al. (1987) were able to show the importance of entrepreneurship courses to the growth of a local economy. The present investigation combines both of these approaches by conducting research at each of these two units of analysis: (1) specific interorganizational relationships, and (2) the city of Tsukuba as a whole. Case study strategies, unlike sample survey strategies or ethnographic strategies, encourage the use of a mixture of
182
2
3 4
5
Notes methodologies. Yin (1984) defines a case study as requiring multiple sources of evidence. Multiple sources of evidence, when each is applied to the same phenomena, result in a “triangulation” or web of evidence that may cross validate (or invalidate) findings (Lievrouw et al. 1987; Jick 1979). Triangulation is the combination of several types of data or methods of analysis in an investigation of the same phenomena. Either in validation or invalidation the researcher may be more confident of his or her conclusions. In the present analysis I triangulate (1) interviews, (2) archival records, (3) survey research, (4) aggregate indicators of innovation (mass media analysis, co-authorship analysis, and a comparison of science cities), and (5) participant observation. Preliminary findings were circulated back to expert informants for them to interpret. Such “member validation” is pursued in order to determine the diversity and depth of members’ views about a particular event, such as their participation in a communication initiative (Emerson 1981). Member validation is in itself a form of “endless triangulation” (Cicourel 1974). A comparative study of science cities should, among other things, attempt to standardize measures of a region’s research productivity by (1) dividing the number of research article authorships by the number of researchers working in that region, and (2) only comparing data which represent regions of similar geographic area. The Current Contents Address Directory lists only one article per published author. That is, if a researcher has four articles published in one year, he is listed only once. The number of first-article authors depends directly on which cities and towns are counted as being within the jurisdiction of a research community. The numbers of research article authors who list their institutional affiliations in Tsukuba, Austin (Texas), Research Triangle (North Carolina), Akademgorodok (Soviet Union), Cambridge (England), Tokyo, and Davis (California) were compared. Data were gathered for the year 1986 from the Current Contents Address Directory (1987), Science and Technology Geographical Index (Institute for Scientific Information 1987). Because of overlapping city and town jurisdictions, the institutes normally associated with these geographic areas may be listed under different geographic headings. Therefore, in the present analysis, Tsukuba Science City refers to authors belonging to all organizations listed in “Tsukuba Science City,” “Yatabe,” “Sakura,” “Toyosato,” “Kukiyaki,” as well as some of the institutes listed under “Ibaraki” (those included in the present analysis were identified as being within the new Tsukuba boundaries by a local research expert). Research Triangle refers to all authors belonging to all organizations listed under “Research Triangle,” “Durham,” “Raleigh,” and “Chapel Hill.” Akademgorodok refers to all authors belonging to all organizations listed under “Akademgorodok” and “Novosibirsk.” Note that geographic areas which have never been considered
Notes
6
7
8
9 10
11
183
“science cities” may also have large numbers of first-article authors. Cases in point are large metropolitan cities such as Tokyo (with 14,282), as well as small college towns such as Davis, California (with 2,602). For all but the most prestigious universities (of which the University of Tsukuba is not yet one) recruitment of senior professors is virtually impossible. Professors are promoted from assistant rank to associate rank to, finally, full rank, usually within the same department. Mobility is very low between universities. The research method of participant observation was used (along with other methods which are explained at other points in this book) to learn how individuals in the science city behave. In the present investigation, participant observation meant observing how people in Tsukuba live and work, and, more specifically, watching how researchers work and participate in communication initiatives, and how administrators organize initiatives. The present approach to participant observation was to understand how participants and administrators see the initiatives of which they are a part, in addition to trying to understand how participant perceptions reflect upon our research questions. Participant observation was used particularly during the last six months of field research by the present author to understand what life in Tsukuba is like. My observations were not hidden; it is more likely that I was somewhat conspicuous by my presence. Also, much of my participant observation took place while I was actively engaging researchers in conversation, whether on the job or late at night eating and drinking. Involving oneself with participants whose behavior is being observed is a much more natural means of collecting information than just watching participants. The information gathered is likely to be more accurate since participants will be more relaxed. “Superconductor Valley” was a reference to California’s Silicon Valley, while “Superconductor Fever” was a reference to a best-selling book in Japan about Silicon Valley, Silicon Valley Fever (Rogers and Larsen 1984). This researcher writes an average of one journal article per month. His articles are primarily published in Physica C, the Japanese Journal of Applied Physics, and Nature (Izumi 1989). It is not unusual for privately employed researchers in Tsukuba to spend long periods of time studying under the direction of senior researchers in the government labs. For example, researchers at Sanyo’s Tsukuba Research Lab said that 10 of their co-workers assisted senior researchers at MITI’s Electrotechnical Laboratory in 1987 alone. Similarly, prior to opening its lab in Tsukuba, Mitsubishi Seishi sent three researchers to national labs, each for six months. This type of personnel transfer is a very common means of technology transfer between the Tsukuba national institutes and private companies throughout Japan. Saxonhouse (1986) does not focus on conflict in Japan, but he does dismiss the idea of Japanese business and bureaucratic relations as harmonious.
184 12
Notes Although national-level administrative relations are best characterized as vertical, horizontal relations among administrators and policy-makers are common at the regional level (see Samuels 1983).
CHAPTER 2 UNDERSTANDING A SCIENCE CITY 13
14 15
The dominant concern in world affairs has changed from a bipolar security dilemma to a pluralistic economic dilemma. For many nations, partnership in a trading alliance, such as the European Economic Community (EEC), has eclipsed partnership in a military alliance, such as the North Atlantic Treaty Organization (NATO), as the most important national imperative. Yet whether priorities are militaristic or economic, a key determinant of national development remains constant: capability in science and technology. The question of how a nation can quicken the rate and improve the quality of its scientific and technological innovations is vital for retaining political independence or achieving autarky. Thus, in national science policy, science cities are means to a very important end. Park’s work formed the basis for the type of research known as “symbolic interactionism.” Reference groups are, for some individuals, invisible colleges that link together active researchers who are working on similar problems throughout the world.
CHAPTER 3 HISTORY OF THE SCIENCE CITY CONCEPT 16
Methodological decisions about investigating the history of the science city are explained here. Personal interviews were conducted with 52 individuals (57 interviews in total). Of this total, 24 personal interviews with 20 individuals, most of whom are considered authorities in Tokyo or Tsukuba, were held on the general topic of Tsukuba Science City. These interviews were a major source of information about the planning and development of Tsukuba Science City. “Authorities,” for our purposes, are taken to be persons who, in the process of notable career or social achievements, gain disproportionately large organizational, personal, and social influence. In the present case, authorities were identified through archival records and in conversations with people as being instrumental in the planning or development of Tsukuba Science City. Personal interviews were sought with those authorities who were identified by three or more other authorities as being instrumental in the planning or development of the city. The majority of authorities interviewed in the present study lived in Tokyo. One of those identified declined to be interviewed. The average age of those interviewed was about 60. Personal interviews averaged two hours each. Three most frequently mentioned authorities were asked to individually play a type of oversized “board game” in which they
Notes
185
identified (1) the duration for which different ministries, agencies, and departments were important to the Tsukuba Science City project, (2) key individuals at those organizations, and (3) the relative strengths of the relationships between the involved organizations. The combined results of these three historical interpretations are represented in Figure 11, in Chapter 4. Thirty-three of the personal interviews, with 32 scientists and administrators, were a major source of information about specific communication initiatives in Tsukuba. Scientists and administrators of specific communication initiatives in Tsukuba averaged about 40 to 45 years of age. They were identified by (1) authorities in Tsukuba, (2) a list compiled by the Tsukuba Center for Institutes, (3) each other, and (4) through publications. Personal interviews with these people averaged one and a half hours in length. In addition, approximately 50 less structured conversations without prepared questions, almost all with nonauthorities, were held concerning Tsukuba Science City. These conversations, several of which lasted longer than three hours but most of which lasted fewer than 45 minutes, served both to identify key interviewees and to discuss the information garnered from the personal interviews. Most of these sources were identified through intermediaries with whom the author was previously familiar. Certain respondents (from both categories) were especially informative, and were reinterviewed. Selection was not random, and therefore not representative of the Tsukuba Science City population. Because random samples may be ill-suited for cross-cultural case study research (Lonner and Berry 1986) and very specific information was often sought, our distribution of interviews shows a purposive bias toward government-employed researchers and science administrators. Measuring the success of the communication initiatives in question was necessarily done in part by the present investigator. Yet a major objective was to listen closely to how interviewees talked about and appraised the initiative in question. The largest determinant of to what degree initiatives were successful or not was the perception of the interviewees. The present investigator’s perception was influenced not only by interpreting what interviewees said, but also by participant observation of most of the initiatives. Individually tailored questionnaires were open-ended and many follow-up questions were asked (Converse and Presser 1986). Almost all of the interviews were conducted in Japanese and English, with the assistance of one of the author’s several interpreters. Each of the Japanese interpreters had lived for an extended period of time in a foreign country. The investigator had a limited command of spoken and written Japanese. Some interviewees (particularly scientists and younger government employees) had a good command of English because of graduate study in the United States. A number of interviewees who felt at ease speaking English did so, resorting occasionally to Japanese by consulting with the interpreter or with the investigator. For historical questions, cues taken from archival records about events were included to assist with interviewee recall. For interviewees who spoke good
186
Notes English, Japanese interpreters were often brought along as a courtesy, so that the interviewee could speak in either Japanese or English. To make sure that the content of questions was understood, questions were most often asked through interpreters, in Japanese. Most of the questions in the personally administered open-ended questionnaires were “back-translated” to check whether their meaning in Japanese was ambiguous (Brislin 1986). “Back-translation” is a process of translating from one language into another by one translator, and back again into the original language by a second translator. Back-translation is a safeguard usually taken for survey questionnaires which are not personally administered by the investigator. The process can be considered an additional validity precaution here, since both interpreter and investigator were present to clarify questionnaire ambiguity and respondent confusion (Ratcliffe 1983). Questions were designed to be (1) specific, (2) simple in their wording, and (3) commonly understood (Brislin 1986; Converse and Presser 1986), while at the same time using the language of the scientific subculture in Tsukuba, where appropriate (for example, scientific jargon would be worked into discussions with scientists and science administrators). Three objectives were kept in mind when drafting the present questions: (1) ask questions pertinent to the stated research questions, (2) make sure that question wording does not bias the results, and (3) allow respondents to teach the investigator about Tsukuba (Spradley 1979; Bradburn and Sudman 1988). An attempt was made to document how respondents think about Tsukuba, and particularly its scientific social structure, in an ethnographic sense. An ethnographic approach was somewhat difficult because the present investigation is a social scientific study in most regards, meaning that I have been trained to believe that I can understand the social situations and problems of others through my own analysis (Spradley 1979:30). Nevertheless, it was a conscious strategy not to impose preconceived notions of how people and organizations in Tsukuba function and communicate. Archival records used as references for the present research included dissertations, books, chapters, journal and magazine articles, newspaper articles, reference statistics, national and local government documents, lists compiled by government administrators, geographical maps and plans, university documents, unpublished manuscripts, and real estate land valuations. These data sources form a somewhat conflicting mesh of information about Tsukuba, but one which for the most part is conspicuously shallow in content. Much of what has been published about Tsukuba Science City is simply government rhetoric. Archival records about Tsukuba dating from 1960 to 1989 were collected by the author in the U.S., and throughout the one-year research period by the author in Tokyo and Tsukuba. An attempt was made to collect all archival records which mainly concerned Tsukuba, including samples of local newspaper articles and local government records. Almost all records were in Japanese. Translators were asked
Notes
187
to identify the content of each record, after which the present investigator chose certain topics for translation. For lengthy records, such as dissertations, translators were sometimes instructed to “talktranslate” by verbally interpreting the publication with the investigator present. Portions of interest to the investigator were then translated in full. 17 Japan’s third “grand shrine” is further south on the Ise Peninsula. 18 Tokyo Kasei College recently moved to a new campus in Tsukuba. Tokyo Rika University is considering a move to Tsukuba. Officials from the University of Texas at Austin discussed with Tsukuba officials the possibility of U.T.Austin opening a satellite campus in Tsukuba, but the idea did not develop into a proposal. 19 In a contemporary replay of the faculty battle played out at the old Tokyo Education University, physical science, life science, and engineering faculty of the University of Tokyo are seeking to move their departments outside of Tokyo to gain better facilities, while their colleagues in law, economic, and social science disciplines oppose such an idea (Takayama 1992). 20 Overpopulation in Tokyo remains a major problem today. Various Diet members and public policy consultants continue to suggest that some or all of the national government bureaucracy be relocated outside of the capital city. These proposals have continually failed to attract sustained and serious political attention. 21 According to one source, Ko¯no Ichiro¯’s selection of Nasu as the site for a new city was precluded by Ko¯no’s ownership of a ranch in Nasu (Kato¯ 1992b). 22 In 1960 the first shinkansen (“bullet train”) was under construction and modern freeways were still in the planning stages, so it took longer to travel 100 kilometers in Japan. Curiously, each of the three “losing sites” are now linked to Tokyo, and the rest of Japan, by bullet train. 23 It is possible that the choice of Tsukuba was determined before the public contest between the “initial” three sites of Akagi, Nasu, and Fuji took place. According to Yajima (1989b), a young urban planner first proposed the idea to the Chairman of the Capital Region Development Commission, Ko¯no Ichiro¯, that a science city be built in Ko¯no’s hometown of Hiratsuka, but Ko¯no rejected the idea as transparent porkbarrel politics. Tsukuba was proposed instead of Hiratsuka, and the competition between Akagi, Nasu, and Fuji was set up as a smoke screen so that real estate speculators would not drive up the cost of land in Tsukuba before the national government could buy land there. This possibility makes sense because the first rationale offered for the selection of Tsukuba when the decision was announced was less expensive land prices in Tsukuba (Asahi Shinbun 1963b), although in each of the other three sites the national government already owned large contiguous tracts of land which had been military bases in World War II. 24 According to articles appearing in 1962 and 1963 in issues of the Jo¯yo¯ Shinbun, national government officials, whose job it was to explain to local Tsukuba area farmers about the plan for a new city, apparently told
188
Notes
local residents that if built in the Tsukuba area the new city would be built on the southern slope of Mount Tsukuba, not in the valley where the science city is now located. Very few people lived on the slopes of Mount Tsukuba, so it may have seemed to residents that the plan for a new city would not affect them greatly. Indeed, the headline of the article in the late afternoon issue of the national Asahi Shinbun on 27 August 1963, reporting the same-day Cabinet choice of Tsukuba, announced Tsukuba mountain slope the best site for the construction of a new administrative city. 25 The government ministries with laboratories in Tsukuba have continued to resent the authority given to the Science and Technology Agency for science coordination. According to the president of the Science Council of Japan, Kondo¯ Jiro¯, this resentment has taken the form of bureaucrats from each ministry encouraging the ministry’s own set of favored companies to build laboratories in Tsukuba, to the exclusion of other companies (Kondo¯ 1992). 26 An illustration of this “politics of appeasement” was an irate farmer’s refusal to move his house which stood directly on the path of the planned Eastern Boulevard (Higashi O¯do¯ri), north of the Azuma neighborhood. The boulevard was built immediately around each side of his house. The house remained until the early 1970s. 27 The national government in Japan commonly acquires land for major public developments by a land division scheme (tochikukaku seiri), originally adopted from Germany, in which the government persuades each private landowner to give up approximately 30 percent of his or her land in exchange for a guarantee by the government that subsequent public investment in the acquired land will increase the value of the remaining privately owned land as well. Some Tsukuba land was purchased by this negotiation method (Dohi 1989b). 28 The one-hole golf course and poorly constructed country club were hastily built soon after the government announced its selection of Tsukuba as the science city site. The farmer who owned the land reasoned that the government was going to want to buy his land, and recreational land had a higher tax base and thus a higher legal value than farmland. Nine or 18 holes were not required to qualify the land as a “golf course,” so only one hole was built. The farmer’s scheme worked. 29 If measured by attendance, Tsukuba Expo ’85 may be considered a success. Yet its effects on participants are less clear. A majority of people questioned by the author mainly remembered (1) crowds of people, and (2) a long day of frustration with inadequate transportation. 30 Some officials of the Ministry of Education still consider Tsukuba primarily an academic, as opposed to a science, city. 31 The theme of constructing a science city was also used to accomplish very personal goals. A professor and noted architect at the University of Tokyo, Yoshitake Taisui, was very involved in drafting architectural plans for Tsukuba. Yoshitake complained bitterly to the Japan Housing Corporation when the first two apartment buildings for government researchers were built in Tsukuba. The buildings were
Notes
189
drab and the apartments very small. Yoshitake took his drawings to Prime Minister Tanaka Kakuei (who had previously been Minister of Construction and owns several construction firms) for approval. The drawings were of modern, relatively large apartments. Tanaka agreed that the nation’s scientists should have such nice accommodation, and then in abrupt fashion presented the architect with Tanaka’s own drawings. “You will build these,” he said. A complex of the apartments personally designed by Prime Minister Tanaka Kakuei was hurriedly built and ready for a tour by the Emperor when he visited the science city. Researchers still live in Tanaka’s apartments, unaware that the infamous prime minister designed their homes (Yajima 1989b).
CHAPTER 4 IMPLEMENTING THE PLAN 32 Most such expositions in Japan are unsuccessful in attracting economic investment to their locality due to their small scale, the lack of an adequate work-force and weak industrial infrastructure, and the repetitive similarity of the fairs. The 1988 Sapporo World Food Festival lost more than 8 billion yen (Foreign Press Center 1989). In spring 1989, one advertising agency, Dentsu Incorporated, signed contracts to stage about 70 such public fairs in Japan. Dentsu was hired for both Tsukuba Expo ’85 and the Yokohama Exotic Showcase. 33 In Japan career bureaucrats staff the national ministries, but the minister in charge of each is a political appointee. 34 There is no official office of “deputy prime minister” in Japan. However, the politician who leads the second most powerful faction in the ruling party functions as a deputy prime minister. 35 Ko¯no Ichiro¯ was a skilled politician adept at garnering large cash contributions from Japanese corporations. The intricate, intimate dependency of politicians on corporate money which was brought to light by the Recruit Cosmos stock scandal, which badly damaged public confidence in the Liberal Democratic Party during 1988–9 and led to many political resignations (including that of Prime Minister Takeshita Noboru, and of former Prime Minister Nakasone Yasuhiro from the LDP), indictments of top officials, and at least one suicide (by Takeshita’s private treasurer), can be traced back through LDP factions. Japan’s most infamous “money politician,” former Prime Minister Tanaka Kakuei (indicted for accepting bribes from Lockheed Corporation), was the direct factional predecessor of Prime Minister Takeshita. Prior to Tanaka’s rise to power in the early 1970s, Ko¯no was the leader of another faction which included Nakasone. A third faction controlled by Prime Minister Fukuda then was controlled by Abe Shintaro¯. Takeshita, Abe, and Nakasone were each centrally implicated in the 1989 Recruit Cosmos stock scandal. Their teachers of how to squeeze money out of influence-seeking corporations were Ko¯no Ichiro¯ and Tanaka Kakuei.
190 36
Notes
Waseda University, one of Japan’s two top-rated private universities, is famous for producing journalists. 37 For example, Ko¯no was renowned for asking the opinions of students and service workers. One large boulevard in Tokyo was enlarged from being a small street because a taxi driver convinced the politician of the inefficiency of the current road (Ishikawa 1992). 38 Ko¯no’s mansion was inherited by his son, Ko¯no Yo¯hei, who could not afford the annual taxes on the residence. The mansion was sold and became the Hungarian Embassy. The former residence-turned-embassy is now an exclusive corporate club. 39 The creativity of the late-night conversations was lubricated with whiskey and cigars (Yajima 1989b). Session host Ko¯no Ichiro¯, however, was a nondrinker. 40 Curiously, although Ko¯no Ichiro¯ never became prime minister, his son may. Ko¯no Yo¯hei, now in his 50s, was voted the new reformist leader of the Liberal Democratic Party in the wake of its loss of power to a coalition government in 1993, and in 1994 became Minister of Foreign Affairs. 41 Yoshitake Taisui was professor of design theory at the University of Tokyo, where he taught alongside Takayama Eika. Dohi Hiroshi was one of Yoshitake’s many students. When he left the University of Tokyo, Yoshitake became an urban planner for the Ministry of Education, and was named Vice-President of the fledgling University of Tsukuba. Yoshitake Taisui is now President of the Kobe Technology and Art University. 42 Ishikawa Makoto is the son of Ishikawa Eiyo, who was a prominent staff engineer with the Ministry of Construction following World War II. In 1952, the senior Ishikawa became the first chairman of the Capital Region Development Commission, which Ko¯no Ichiro¯ later chaired. 43 Fukuda is the only prominent historical figure in the development of Tsukuba about whom interviewees invariably prefaced their comments with a chuckle. 44 Omron Tateishi is well known in Japan as a progressive high-technology electronics company, but the firm is perhaps best known for its innovative personnel management policies. Omron requires its managers to take periodic month-long paid vacations from work. Vacation themes must be submitted to the company for approval, and those proposals which are work-related are turned down. The idea is to refresh and spiritually renew its employees. One manager’s vacation theme, made famous by national media coverage, was a bicycling trip to the more than 100 shrines on the island of Shikoku. A manager whom I interviewed had recently returned from four weeks in Hastings, England, where he studied English. 45 Not all entrepreneurs who attend the Tsukuba no yu¯be meetings are in high-tech businesses. When asked about the entrepreneurs that they had met at these meetings, the only person identified by administrators at Omron was the self-employed gardener who now takes care of the Omron grounds.
Notes
191
CHAPTER 5 RESEARCH COMMUNICATION IN TSUKUBA 46
A
In order to investigate informal groups in Tsukuba, and the extent to which these groups of researchers have been successful in transferring technology among their participants, we used both survey questionnaire and case study methods. three-page questionnaire was mailed to researchers who are listed as the organizers of research study groups in Tsukuba. A list compiled by the Tsukuba Center for Institutes in spring 1988, identified 76 contact persons. This list was created by assigning a secretary to telephone each of the national research institutes in Tsukuba and asking an administrative person in charge of research whether they were aware of any of their researchers meeting outside of work in research groups. Thus, this list of 76 was determined by the information (or lack thereof) which institute administrators had about their researchers’ activities. Three strategies for supplementing this list were followed. First, a Tsukuba resident experienced in academic research was hired to go to each of the institutes and collect information from the notices posted on bulletin boards about research group meetings. Second, three messages (two in Japanese and one in English) were placed on the “bulletin board” of a new electronic data-base linking researchers in Tsukuba. Members of research groups were encouraged to either telephone the author or send an electronic message to a certain user number. Third, back issues of Science Communication, a locally distributed notice printed by the Tsukuba Center for Institutes, were examined for possible contact names of study group organizers. Science Communication is a popular means by which study groups advertise their meetings. These supplementary strategies resulted in the identification of a possible additional 65 research groups. A questionnaire was drafted in English and translated into Japanese. Questions in Japanese were then tested for the original English meaning, and adjusted. One scientist and one science administrator in Tsukuba reviewed the Japaneselanguage questionnaire and made criticisms and suggestions for a revision. Several questions were added and coding categories were revised. The complete mailed packet consisted of (1) a Japaneselanguage cover letter written by Kawamoto Tetsuzo¯, a widely known and respected science administrator in Tsukuba Science City, (2) an English-language cover letter and explanation written by the present author, (3) a photocopy of a newspaper article about the present research investigation, which had been printed locally in Tsukuba two weeks before the packets were mailed, (4) the three-page questionnaire, and (5) a return-addressed stamped envelope. The packet envelopes and the return addressed envelopes both were printed with the name of the Tsukuba Research Consortium, a private organization in Tsukuba which is well known for science communication activities. One hundred and forty-three packets were mailed.
192
Notes
Within one week, 30 completed questionnaires had been received. After three weeks, 65 completed questionnaires had been received. Follow-up telephone calls brought the total number of completed questionnaires to 71. Judging from some mailed and telephoned responses, perhaps as many as 40 of the 143 total mailed packets went to the organizers of meetings which were not topic-specific informal study groups of researchers who regularly meet. The total number of such groups was estimated by local experts to be about 100 in 1989. So our survey response rate is about 70 percent. Respondents were asked to identify the institutional affiliations of the members of their research group, and the number of researchers from each institution who were group members, as well as to respond to several other closed and open-ended questions. Our case studies of five small groups of researchers were based on personal interviews with group organizers, leaders, participants, and nonparticipants. Researchers averaged about 40 to 45 years of age. They were identified by (1) authorities in Tsukuba, (2) a list compiled by the Tsukuba Center for Institutes, (3) each other, and (4) through publications. Personal interviews with these group leaders averaged one and a half hours. Selection was not random because very specific information was sought, from individuals with unique knowledge. Our distribution of interviews shows a bias toward government-employed researchers because the groups have more participants from the government research institutes than from universities or from private firms. Measuring the success of the communication initiatives of study was done by listening closely to how interviewees appraised each communication initiative. The main determinants of the degree to which initiatives were successful were the perception of the interviewees, and our observation of the groups and their members. Our individually tailored questionnaires were open-ended, and many follow-up questions were asked. Almost all of the interviews were conducted in Japanese and English, with the assistance of interpreters. Some interviewees (particularly younger government researchers) had a good command of English because of their graduate study in the United States. A number of interviewees who felt at ease speaking in English did so, resorting occasionally to Japanese by consulting with the interpreter or with the investigator. For interviewees who spoke adequate English, Japanese interpreters were often brought along as a courtesy, so that the interviewee could speak in either Japanese or English. To make sure that the content of our questions was understood, questions were most often asked in Japanese through interpreters. 47 The present data are based on written responses from the leaders of 71 study groups. The precise number of responses to each question varies since some respondents did not answer some questions, and not all questions were applicable to all respondents. 48 The consistency in the number of study groups identified by the Center for Institutes may be attributable to the method used to gather the data.
Notes 49
50
193
A primary issue in studying specific initiatives to increase communication among researchers in Tsukuba is which initiatives to study. How can the population of initiatives (1) be identified, and (2) be sampled? The present investigation uses local science and government experts in Tsukuba, as well as our definition of science communication initiatives, to identify the population of initiatives, and a “most different systems design” to select a subset of initiatives, each of which is the object of study in the present investigation. Przeworski and Teune (1970) identify a most different systems design as a strategy for selecting social systems. This strategy seeks to maximize the variance between the social systems selected, by choosing systems for comparison which are least alike in all respects except for the dependent variable of study. The strength of this strategy is in showing that phenomena are related in the same way under very different circumstances. The most different systems design begins at a lower than systemic level of analysis. The object is to rule out irrelevant system variables from explaining the dependent variable. “The most different systems designs eliminate factors differentiating social systems by formulating statements that are valid regardless of the systems within which observations are made” (Przeworski and Teune 1970:39). Both informal research group initiatives by researchers and formal governmental initiatives to increase communication among researchers were listed. Local experts were asked to distinguish more successful initiatives from less successful initiatives on the bases of (1) resulting research collaboration, and (2) longevity of the initiative. The most successful initiatives which were maximally different in other respects were chosen for case study. One informal research group was included which only became a “group” after an initial exciting discovery, and never advertised itself as a group. It was investigated as a case study because of the researchers’ spectacular results. The other researchergenerated communication initiatives were explicitly recognized as research groups, with regular meetings and various degrees of organization. The five cases of informal study groups are based on the following personal interviews and archival documents: Tsukuba Center for Institutes 1979; Kawamoto 1987; Japan External Trade and Research Organization 1988; Ando¯ 1989; Fujita 1989; Haseni 1989; Iseya 1989; Ishibashi and Furutani 1989; Kaneko 1989; Kuroda 1989; National Research Center for Disaster Prevention 1989; Oyagi 1989; Tanaka 1989; Tsuge 1989; Ueda 1989; Wakasa 1989; and Yabe 1989. A major applied and social problem in Japan has historically centered around gaseous combustion. Prior to 1960 when the use of gas for cooking and heating became popular, gas inhalation was a popular means of suicide because gas did not badly distort the color of a corpse (in contrast, drowning badly distorts the life-like look of a corpse), so people choosing to commit suicide preferred gas to preserve the way their bodies would look at funerals. As long as city houses were constructed of wood and paper, homes were well ventilated and would
194
51
52
53
54 55
56
57 58
59
Notes not explode when gas valves were left on. But as housing construction relied more on materials which resulted in tighter joints, windows, and fittings, gas-filled houses began exploding, leading to neighborhood fires. So combustion became an important problem for Japanese researchers (Kaneko 1989). One privately employed research manager suggested that the study group could attract more researchers from the private sector if more applied problems were addressed, but he felt that the study group was more valuable as an academic-type meeting, not a business trouble-shooting session (Kaneko 1989). The researcher in question, Ueda Hiromasa, was a graduate of the University of Kyoto, while Dr. Tsuge was a graduate of the University of Tokyo. Tsuge still maintains close relations with faculty at the University of Tokyo, and Ueda and Tsuge are good friends. So their lack of intellectual communication appears to result almost wholly from different intellectual orientations and research interests. Initial excitement about genetic engineering has matured into institutional form in Tsukuba. Following the establishment of a 50,000-seed-capacity automated gene bank in 1979, a Tsukuba Genetic Resources Storage Center was completed in 1988 with a capacity of storing up to 160,000 seeds and microbes for as long as 4,000 years (Japan External Trade and Research Organization 1988). Each of the core companies had about 1,000 employees in 1989 (Kawamoto 1989d). According to Chiba (1989), the NLA had sent a survey questionnaire about location decisions and the Tsukuba area to private companies in the late 1970s. Most of the company people indicated their negative impression of Tsukuba Science City by throwing away the questionnaires. The NLA bureaucrats were shocked. They had not realized that Tsukuba Science City was going to require such a “hard sell” to private companies. Once the company presidents had agreed that Tsukuba should be the site for the Consortium, they had to decide what a “consortium” meant. The original idea was simply entrusted to Professor Tazaki. Through his wife (the presidency was an unpaid position) the Consortium was capitalized with 1 million yen (about U.S. $7,500). The 8.5 acres had been earmarked for the construction of a United Nations University. The TRC building bears some traces of Chiba’s temple idea. In a courtyard area is a rock garden, and one can stand at the TRC building and see the sun set between two column-like buildings, which was a feature of traditional Japanese temples. According to Kawamoto (1989d), after a number of overtures he accepted a personal request from the STA Vice-Minister Umezawa on 1 October 1982 to become the operational leader of the Consortium. Kawamoto visited each of the eight company presidents to find out what they thought a consortium should be. Only the president of Akashi Seisakusho offered an idea of what the Consortium should be like. Tazaki and Chiba had an undeveloped idea that the Consortium should
Notes
60 61
62
63
64 65 66
67 68
69
195
consist of a laboratory and workshop to be shared by researchers from all eight firms, on a single small lot. Kawamoto dismissed this concept as unworkable. The concept, physical design, and communication activities of the Tsukuba Research Consortium have now been copied by other similar organizations from Hokkaido¯ to Kyu¯shu¯ (Chiba 1989). When researchers in Tsukuba are asked about Kawamoto they refer to him as a “big boss” or “famous person.” Professors and bureaucrats in Tokyo who have some knowledge or responsibility concerning Tsukuba, however, often do not know of him. In Japanese, the name for this center is Tsukuba Kenkyu¯ Ko¯ryu¯ Center, which literally means “Tsukuba Research Exchange Center.” When the Center was first built in 1978, it became the site for the Institute Directors’ Meetings, which were a series of meetings of the executive directors of each of the Tsukuba national research institutes. The first meeting in 1977 was held in the area of Tokyo known as Toranomon because there was no room in Tsukuba big enough or comfortable enough to seat the 50 or so directors together. With the completion of the Center one year later, the Institute Directors’ Meetings became a regular event at the Kenkyu¯ Ko¯ryu¯ Center, which probably led to the Center’s current English name. Since the Center is now more devoted to use by individuals, the literal translation of “Tsukuba Research Exchange Center” is more accurate than the name Tsukuba Center for Institutes. The Tsukuba Center for Institutes was the first national institute with the purpose of increasing interlaboratory and interministry communication. According to the Center’s first director, use of its facilities was highly circumscribed at first by the STA bureaucrats in Tokyo, and has since become less restrictive (Kawamoto 1987). In contrast to the other seven initiatives examined in this chapter, the present review of the Tsukuba Supporting Center was necessarily written before the initiative formally began. According to the president of Tsukuba Center, Inc., this contribution by Ibaraki Prefecture is the largest amount of money that the prefecture has ever invested in a commercial enterprise (Hattori 1989). According to several sources, MITI’s tactics may have been so successful because of a subtle threat that if the prefecture refused to invest in the new center, MITI might refuse small business aid to firms within the prefecture. The Tsukuba College of Technology is still in the planning stages. Its offices are located on the campus of the University of Tsukuba. In response to a question about whether their current data-bases were open to the public, 89 percent of responding researchers at the public research institutes replied “no” (Tsukuba Institute Coordinating Committee 1987:19). Differences in wording between the two questionnaires made it obvious that the electronic network was being planned as a public system primarily for researchers at the national research institutes, and that private researchers might be allowed access to the network. People receiving the questionnaire at private institutes might have sensed a
196
Notes depreciatory tone from the questionnaire, which could have affected their responses to the survey.
CHAPTER 6 COLLABORATION NETWORKS 70
Data based on journal research article authorship were collected in order to test (1) the degree of interorganizational integration in Tsukuba, (2) the degree to which Tsukuba is intellectually dependent upon Tokyo, and (3) the degree of collaboration between Tsukuba researchers and researchers in other geographic areas. Data for the co-authorship network analysis were obtained from the Japanese Science and Technology Agency’s Japan Information Center for Science and Technology (JICST) computerized data-base of published science or technology journal articles and scientific or technological papers presented at academic conferences throughout the world. The data consisted of a series of entries, each of which described a published research article. Each entry included information such as the title of the article, the journal in which it appeared, the author(s) and where they worked, and key words about the content of the article. The present investigation used only the information of where each author and coauthor worked (their institutional affiliation). The present data-set was derived in the following search sequence: (1) the JICST data-sets containing records for the years 1975–89 were specified (number of entries=6,246,083); (2) then, the number of published journal articles which were either authored or coauthored by an author whose institutional affiliation was listed as being in Tsukuba were specified; (3) then, all of the records for the four time-periods of July through December 1979, 1982, 1985, and 1988 were specified (number of entries=11,463). The 11,463 articlerecords, hereafter simply referred to as articles, contained three duplicates, perhaps because of original JICST data entry error, so the total number of articles in our data-set of four time-periods was 11,460. The number 11,460 does not represent articles published within the months and years of our four time-periods; rather, 11,460 represents the number of JICST data entries made during those specified time-periods. Because the receipt of journals and magazines sometimes is late, or because some issues of journals were occasionally lost in the mail or by JICST and had to be reordered, the entries made during any particular time-period include a small percentage of articles which had been published in previous time-periods. So, for example, the data in our time-period of June through December 1979 actually include the majority of articles published during that time-period, as well as a small percentage of articles which had been published in the first five months of 1979, and a smaller percentage of articles published during 1978, and decreasing percentages of articles (the absolute numbers are very small) for the years 1977, 1976, and 1975. See Dearing 1989a for the actual
Notes
197
distribution of the dates of publication of articles in the co-authorship data-set. Thus, our four time-periods of June through December for the years 1979, 1982, 1985, and 1988 actually contain very uneven distributions of articles published during the time-periods 1975–9, 1980–2, 1983–5, and 1986–8. Each of these time-periods is roughly equal to the one-year time-periods of 1979, 1982, 1985, and 1988. Therefore, results reported in Chapter 6 refer to the time-periods as 1979, 1982, 1985, and 1988. These data have several limitations. The number of journals which JICST subscribes to has increased during the time-periods analyzed. In 1979, JICST entered data about articles from 9,006 journals. The 1982 number of journals was 9,790, the 1985 number of journals was 10,741, and the 1988 number of journals was 13,024. Conceivably, this increase in the number of source documents might confound the analysis, particularly in the attempt to find over-time patterns in the distribution of the numbers of articles. For example, an increasing trend of the number of articles authored by Tsukuba researchers might be a result of increases in the number of journals subscribed to by JICST, and not due to increasing productivity of Tsukuba researchers. While this possibility cannot be dismissed, my repeated conversations with officials of JICST suggest that the increasing numbers of journals should not confound the results. When JICST was founded in 1957, the majority of research institutes which today are in Tsukuba then existed in Tokyo. To create its data-base, JICST officials asked each of the academic and professional societies in Japan and many such societies overseas to indicate which journals were important to members of their associations as publication channels. The JICST data-base was created by first including the most important journals, and, in subsequent years, supplementing this number of core journals with intellectually somewhat peripheral journals. Thus, additions to the data-base as late as 1979 (the first year of the present analysis) should not significantly affect our analysis. In general, the increase in the number of journals since 1979 represents businessoriented, high-technology journals. Although researchers may read some of these journals, they by and large do not publish in them (Igarashi 1989). Another limitation of our data-set is that we have no entries of nonJapanese language articles for the 1979 time-period. Therefore, some of our comparisons will be limited to the 1982, 1985, and 1988 periods. Also, the JICST data-base only covers science and technology, and our data do not include articles by medical researchers. Thus, our measurements of the University of Tsukuba, in particular, are quite conservative, because no measures of Departments of Social Science or Humanities, or of the School of Medicine, are included. The data-set also does not reveal the full extent of authorship and co-authorship by privately employed Tsukuba researchers. The names of public organizations, such as the University of Tsukuba, were specified by JICST staff when the computer search was done, so that all journal articles by University of Tsukuba researchers would be
198
Notes included for the time-periods for which I asked. Yet the names of most private companies which employ researchers in Tsukuba were not specified. Since all the public organizations were specified, our data-set includes privately employed authors if they co-authored an article with publicly employed researchers. Yet single authored articles by privately employed researchers, as well as articles coauthored by privately employed researchers, might not appear in our data-set. A Japanese coder was instructed by the present investigator and a Japanese research assistant how to distinguish between published articles by the organizational or geographical affiliation of the author(s) according to whether their affiliation was in Tsukuba or not. Coded data were double-checked for the accuracy of data entry. Authors and coauthors who identified themselves as being affiliated with an organization in Tsukuba were coded into the category for that organization. For example, an article solely authored by a researcher at MITI’s Electrotechnical Laboratory (ETL) would be coded in the category “ETL.” In order to measure the degree of dependence or integration of Tsukuba researchers with the large number of researchers in nearby Tokyo (which represents Japan’s largest concentration of researchers), Tokyo-based co-authors were coded into the category for that city. For example, an article co-authored by a professor at Keio University in Tokyo and a colleague at the Fruit Tree Research Station in Tsukuba would be coded twice, once as “Tokyo,” and once as “Fruit Tree Research Station.” To better understand the degree of dependence or integration of Tsukuba researchers with researchers in Japan who are not in Tsukuba or Tokyo, and with researchers based overseas, co-authors were coded into the categories “Japan other than Tsukuba or Tokyo,” and “outside of Japan.” For example, an article co-authored by a researcher at Lockheed Corporation in San Diego, a professor at Kyoto University, and a scientist at the National Space Development Agency (NASDA) in Tsukuba would be coded three times, once each into the categories “outside of Japan,” “Japan other than Tsukuba or Tokyo,” and “NASDA.” Articles without a Tsukuba co-author are not included in the present analysis. Articles solely authored by a Tsukuba researcher are included as a measure of research productivity in different Tsukuba Science City organizations. This specification that included articles must have a Tsukuba author or co-author overcomes one of the problems which confound many network analyses: that is, how to set clear network boundaries since real-world relational boundaries are rarely clear cut (Rogers 1987:297). The existence of co-authorship relations between researchers in different organizations was conceptualized as a ratio-level indicator of the strength of research relationships between the organizations involved. For example, if two researchers at the University of Library and
Notes
72
199
Information Science and three researchers at the Geological Survey of Japan co-authored an article, the measure of relational strength based on that article would be 6 (the 6 cells in a 2×3 matrix). Each of the two coauthors at the university would have a relationship with the three researchers at the Geological Survey of Japan. Or, each of the three researchers at the Geological Survey of Japan can be thought of as having a relationship with each of the two researchers at the university. An article with only two co-authors, each at a different organization, would have a relational strength value of 1. Because we conceptualize the strength of the co-authorship relation as a ratio-level value, an article representing a strength relation between organizations of 6 is six times stronger than an article representing a strength relation between organizations of 1. Because co-authors share (although certainly differentially) in the conceptualization, data-collection and analysis, and writing of a research article, we assume that all co-authorship relations are symmetrical and undirected. Relations are assumed to be reciprocated; that is, if we asked two coauthors who had co-authored an article together whether they had coauthored an article together, we expect that they would both answer affirmatively. Statistical distributions were computed for each of our four periods, and the data in each period were then analyzed with the NEGOPY 3.3 network analysis personal computer software program (Richards 1986; Richards 1989). NEGOPY was chosen rather than another network analysis program because my goal was to learn lessons about whether or not some organizations in Tsukuba are relationally closer than other organizations concerning research collaboration by their employees. According to Richards (1986:11), “the primary goal of NEGOPY is to define clusters of nodes [in our case, organizations or geographic areas] that have more contact with one another than with nodes in other clusters.” Useful overviews of NEGOPY are given in Rogers and Kincaid (1981) and Rice and Richards (1985). Conceptual overviews of network research are provided by Monge (1987) and Rogers (1987). 71 For every 1 addition in the number of co-authors (n+1), the number of links increases by n. Note that these are the total numbers of articles, authorships, and links for our data-set. These numbers are based on all published articles, within our four time-periods of study, which had one or more Tsukubabased author or co-author. Thus many of the authorships represent nonTsukuba authors, and many of the links represent relationships between Tsukuba researchers and researchers in other geographic areas (such as Tokyo), as well as some relationships between non-Tsukuba authors. Links between non-Tsukuba authors are included in the present analysis because these authors shared a common Tsukuba author. For example, an article with one co-author at the University of Tsukuba (in Tsukuba), one co-author at the University of California, San Diego (outside of Tsukuba), and one co-author at Waseda University (outside of Tsukuba)
200
Notes
has a link between the researchers at UC San Diego and Waseda University. The total numbers in the present analysis necessarily include such “indirect Tsukuba links.” 73 The presence of an organizational link between any two research institutes has a strength value equal to the number of individual links between researchers at the two institutes. 74 The “structure” value ranges from 0 to 1, where zero would describe a set of research institutes that is no more structurally differentiated than would be expected in a random set of research institutes; that is, the set would not constitute a group or network. The structure measure is based on the number of three-node “cycles.” A cycle is formed when two nodes have a relationship with a common third node. The number of observed cycles is divided into the number of expected cycles, resulting in the structure measure. Most communication networks have structure values from.15 to.35 (Richards 1986). 75 Because the data search conducted on our behalf by JICST may not have revealed the full extent of research article authorship and co-authorship at private R&D labs in Tsukuba, the present measure of network structure may be biased toward depicting a network which is more interconnected than is actually the case. 76 Our coding categories distinguish between research institutes within Tsukuba Science City, but only distinguish between locations of research institutes (for example, Tokyo) for those research institutes located outside of Tsukuba Science City. 77 Note that in Figure 20 we compare four time-periods. Figures 19 and 21 compare only three time-periods because JICST data-entry clerks were not able to provide us with any data about English-language articles for 1979. 78 Note that the y-axis scales change in these figures. Different scaling is used to show trends in the numbers and percentages of links most clearly. 79 Numbers and percentages of (1) privately employed co-authors, (2) links involving one or two privately employed co-authors, and (3) articles written by privately employed researchers are potentially conservative as a result of the computer search organizational keywords used by data entry clerks at the Japan Information Center for Science and Technology. 80 Only the institutes under the ministries with the five largest numbers of interministry links were included in this analysis. The included ministries are (1) the Ministry of Education, with 263 co-author links to other ministry research institutes in Tsukuba, (2) MITI, with 166 links, (3) the Ministry of Agriculture, Forestry, and Fisheries, with 113 links, (4) the STA, with 75 links, and (5) the Environmental Agency, with 64 links to other ministry research institutes in Tsukuba. The other ministries with such co-author links were the Ministry of Health and Welfare, with 2 links, the Ministry of Transportation, with 1 link, and the Ministry of Construction, with 4 links. To include these ministries in the present analysis would have
Notes
81
82
83
201
biased the results since we combined percentages, not numbers of links. The data we have on the number of researchers employed per research institute are annual for 1988. Since our data-set of numbers of articles from which the numbers of authorships are derived is only an approximation of annual data, the following formula was applied to our authorship data to get an estimated measure of annual productivity per research institute:
Note that this measure of productivity may be interpreted as counting quantity more than quality. One good article may be intellectually worth more than four bad articles, yet our measure obscures qualitative distinctions. On the other hand, publication itself is an indicator of quality (that is, bad papers are not supposed to be published). Our measure may also be biased toward counting shorter articles more than longer articles. For example, the majority of journals in one field might in general require shorter articles which only report experimental results, while the journals which predominate in another field might require longer articles. Such disciplinary differences might allow researchers in some fields to write more articles than researchers in other fields. Note that the interpretation of data about first authors rests crucially on the assumption that first authors are research initiators.
CHAPTER 7 LESSONS LEARNED ABOUT GROWING SCIENCE 84
85 86
87
A “tragedy of the commons” (Hardin 1968) certainly did result in Silicon Valley, as evidenced by familiar problems of unplanned overdevelopment such as traffic congestion, industrial pollution, exorbitant housing prices, and a cost of doing business which has led firms to move from the valley. Besides the goals of stakeholders being diverse, each stakeholder may have multiple goals (see Eisenberg 1984; and Eisenberg and Witten 1987). These remarks are not meant to imply that the University of Tsukuba is a typical Japanese university. In fact it is quite unusual because of (1) its newness, (2) its organization into a college system, (3) above-average facilities, (4) its very large campus, (5) an administrative structure which decentralizes the power of senior professors, and (6) its being the first university in Japan to offer double-track two-year terminal Master’s programs and five-year doctoral programs. The University of Tsukuba may be becoming more peripheral than
202
88
89
90
91
Notes central. The number of all professors at the University decreased by 182 (about 12.5 percent) during the three years between 1984 and 1987. Possible explanations are normal retirement attrition, discontented faculty who have found living in Tsukuba too isolated, a lack of research funding opportunities, competition from other Japanese universities for Tsukuba faculty, and complaints from other Japanese university administrators that the faculty\student ratio at Tsukuba was unfairly low. The perception by Kyoto and Osaka planners that Tsukuba Science City is a satellite of Tokyo and “belongs” to the capital is ironic. In the mid1980s, Japanese corporations were encouraged to move to or establish laboratories in Tsukuba by the national government. Many Tokyo-based corporate decision-makers declined this invitation since they had laboratories in the capital and they did not understand how having a laboratory in Tsukuba would be advantageous to their corporations. A number of Kyoto-based corporations, however, saw building a laboratory in Tsukuba as a way to establish a presence close to Tokyo at low cost. So many of the large Japanese corporations with labs in Tsukuba are Kansai-based firms, making for a rather intimate Kansai-Tsukuba link. So the perception that Tsukuba Science City is naturally aligned more closely with Kanto¯ region firms is not wholly accurate. Land prices in the Keihanna area are now more or less comparable to land prices in the Tsukuba area. For example, laboratory space at the Tsukuba Research Consortium in 1992 cost 250,000 yen per tsubo (Kawamoto 1992). To quote: “100 percent train ridership capacity means that all of the seats are full, and people are standing but with plenty of room to read full-size newspapers. At 150 percent, people standing can still read their newspapers, but they cannot open them out fully. At 200 percent they cannot read anything. That’s shoulder-to-shoulder. Between some stations on the old Jo¯ban Line density is 250 percent. People are in danger of being squeezed to death. It’s getting to the point where at a station like Kita Senju if another person falls off the platform from overcrowding and is killed, we say ‘So what’” (Kurokawa 1989). Although officials from each of the affected prefectures favor a new rail line, they disagree as to the exact proposed route. Saitama and Chiba prefectural officials suggest that the line should end in their prefectural territories, not at Tsukuba, in Ibaraki Prefecture.
References
Amabile, T. (1988) “A model of creativity and innovation in organizations”, Research in Organizational Behavior, 10:123–67. ——(1990) “Within you, without you: The social psychology of creativity, and beyond”, in M.Runco and R.Albert (eds) Theories of Creativity, Newbury Park, CA: Sage. Amabile, T., and Gryskiewicz, S. (1988) “Creative human resources in the R&D laboratory: How environment and personality impact innovation”, in R.Kuhn (ed.) Handbook for Creative and Innovative Managers, New York: McGraw-Hill. Amman, C., and Knorr-Cetina, K. (1988) “Thinking through talk: An ethnographic study of a molecular biology laboratory”, paper presented at the Symposium on Science Communication: Environmental and Health Research, Los Angeles, December 15–17. Anderson, A. (1984) Science and Technology in Japan, Essex, U.K.: Longman. ——(1989) “Kansai science city takes root”, Nature, 338:285. ——(1992) “Japanese academics bemoan the cost of years of neglect”, Science, 258:564–9. Anderson, A., and Hamilton, D.P. (1992) “Bright dreams despite a legacy of hits and misses”, Science, 258:577. Ando¯, W. (1989) Professor, University of Tsukuba, personal interview, Tsukuba, May 16. Asahi Evening News (1988) “No girls for the scientists”, A6, June 21. Asahi Shinbun (1963a) “Tsukuba mountain slope the best site for the construction of a new administrative city”, August 27 (in Japanese). ——(1963b) “Tsukuba mountain slope the best site for the construction of a new administrative city: Japan Housing Corporation to prepare the draft”, A1, August 28 (in Japanese). Asahi Shinbun Regional Edition (1963) “Tsukuba mountain slope a promising future: More great news for the prefecture: A new administrative city with 150,000 population”, 16, August 28 (in Japanese). Asakawa, M. (1981) “The problem of garbage collection suddenly arises”, in Society for Recording Life Histories in Tsukuba Research and Academic
204
References
City (eds) Tall Boots and Starry Skies, Tsukuba: Tsukuba Sho¯rin (in Japanese). Aydalot, P., and Keeble, D. (1988) (eds) High Technology Industry and Innovative Environments: The European Experience, London: Routledge. Bacon, F. (1624) The New Atlantis, in A.B.Gough (ed.) Oxford: Oxford University Press. Bartholomew, J.R. (1989) The Formation of Science in Japan: Building a Research Tradition, New Haven, CT: Yale University Press. Benedict, R. (1967) The Chrysanthemum and the Sword: Patterns of Japanese Culture, Cleveland, OH: Meridian Books. Berton-Lewis, H. (1989a) Lecturer, University of Tsukuba, personal interview, Tsukuba, March 28. Birnbaum, H. (1975) “Japan builds a science city”, Physics Today, February. Bloom, J.L., and Asano, S. (1981) “Tsukuba Science City: Japan tries planned innovation”, Science, 212, June 12. Botkin, J.W. (1988) “Route 128: Its history and destiny”, in R.W.Smilor, G.Kozmetsky, and D.V.Gibson (eds) Creating the Technopolis, Cambridge, MA: Ballinger. Bradburn, N.M., and Sudman, S. (1988) Polls and Surveys, San Francisco: Jossey-Bass. Brinkman, W., Oxender, D., Colwell, R., Demuth, J., Rowell, J., Skalak,R., and Wolf, E. (1988) JTECH Panel Report on the Japanese Exploratory Research for Advanced Technology (ERATO) Program, Washington, DC: National Science Foundation. Brislin, R.W. (1986) “The wording and translation of research instruments”, in W.J.Lonner, and J.W.Berry (eds) Field Methods in Cross-cultural Research, Newbury Park, CA: Sage. Broad, W.J. (1988) “Novel technique shows Japanese outpace Americans in innovation”, The New York Times, A1, March 7. Brooks, H. (1988) “The research university: Doing good, and doing it better”, Issues in Science and Technology, 4:2. Carter, A.P. (1989) “Knowhow trading as economic exchange”, Research Policy, 18:155–63. Chiba, G. (1992) Director, Research Administration, Exploratory Research for Advanced Technology, Japan Research and Development Corporation, Japanese Science and Technology Agency, personal interview, Tokyo, June 21. ——(1989) Personal interview, Tokyo, March 24. Chubin, D.E. (1976) “The conceptualization of scientific specialties”, Sociological Quarterly, 17:48–476. ——(1987) “Research evaluation and the generation of big science policy,” Knowledge: Creation, Diffusion, Utilization, 9(2):254–77. Cicourel, A.V. (1974) Cognitive Sociology: Language and Meaning in Social Interaction, New York: Free Press. Clark, B. (1987) The Academic Life, Princeton, NJ: Carnegie Foundation. Clark, K.B., and Fujimoto, T. (1991) Product Development Performance: Strategy, Organization, and Management in the World Auto Industry, Boston, MA: Harvard Business School Press.
References
205
Converse, J.M., and Presser, S. (1986) Survey Questions: Handcrafting the Standardized Questionnaire, Newbury Park, CA: Sage. Csikszentmihalyi, M. (1975) Beyond Boredom and Anxiety, San Francisco: Jossey-Bass. ——(1990) “The domain of creativity”, in M.Runco, and R.Albert (eds) Theories of Creativity, Newbury Park, CA: Sage. Cummings, W.K. (1990) “The Culture of Effective Science: Japan and the United States”, Minerva, 28:426–45. Cutler, R.S. (1989) “A Comparison of Japanese and U.S. High-technology Transfer Practices”, IEEE Transactions on Engineering Management, 36(1):17–26. Dearing, J.W. (1989a) Communication Among Researchers in a Science City: Tsukuba, Japan, Ph.D. dissertation, Los Angeles: University of Southern California. ——(1989b) “How Japanese scientists talk to each other: Communication networks in Tsukuba Science City”, MIT Japan Program Working Paper, Cambridge, MA: Massachusetts Institute of Technology. ——(1991) “Access to scientific information in Japan”, Preprints of the 3rd International Conference on Japanese Information in Science, Technology and Commerce, Institut de l’Information Scientifique et Technique, Vandoeuvre-les-Nancy, France. ——(1993) “Rethinking technology transfer”, International Journal of Technology Management, 8:1–8. Dearing, J.W., Meyer, G., and Hermodson, A. (1993) “A late look at diffusion”, unpublished paper, Department of Communication, Michigan State University, East Lansing, MI. Dearing, J.W., and Rogers, E.M. (1990) “Japan: Tsukuba Science City”, in F.Williams and D.Gibson (eds) Technology Transfer: A Communications Perspective, Newbury Park, CA: Sage, 211–25. DeVos, G. (1984) (ed.) Institutions for Change in Japanese Society, Berkeley: Institute of East Asian Studies, University of California. Dill, D. (1986) “Research as a scholarly activity: Context and culture”, in J.Creswell (ed.) Measuring Faculty Research Performance, New Directions for Institutional Research, No. 50, San Francisco: Jossey-Bass. Dohi, H. (1989a) Professor, University of Tsukuba, personal interview, Tsukuba, April 27. ——(1989b) Personal interview, Tsukuba, May 1. Dore, R.P., and Sako, M. (1989) How the Japanese Learn to Work, London: Routledge. Edge, D.O. (1977) “Why I am not a co-citationist”, Society for the Social Studies of Science Newsletter, 2:13–19. Eisenberg, E.M. (1984) “Ambiguity as strategy in organizational communication”, Communication Monographs, 51:227–42. Eisenberg, E.M., and Witten, M.G. (1987) “Reconsidering openness in organizational communication”, Academy of Management Review, 12(3):418–26. Emerson, R.M. (1981) “Observational field work”, Annual Review of Sociology, 7. Environmental Agency (1988a) The National Institute for Environmental Studies.
206
References
——(1988b) White Paper on the Environment, 1988, English translation, Tokyo: Foreign Press Center. Eveland, J.D. (1987) “Diffusion, technology transfer, and implementation”, Knowledge: Creation, Diffusion, Utilization, 8:2:303–22. Eyring, H. (1959) “Scientific creativity”, in H.Anderson (ed.) Creativity and Its Cultivation, New York: Harper & Brothers. Foreign Press Center (1989) PressGuide, XIV(4), April. Frenkel, O. (1984) “Flying high: A case study of Japanese industrial policy”, Journal of Policy Analysis and Management, 3(3):406–20. Friedkin, N. (1982) “Information flow through strong and weak ties in intraorganizational social networks”, Social Networks, 3:273–85. Fujimoto, K. (1987) “Government-university-industry cooperation in R and D activities in Tsukuba Science City”, Journal of Engineering Education in Southeast Asia, 17(1), March. Fujita, T. (1989) Director, Academic Liaison Program, R&D Division, Eisai Company, personal interview, Tsukuba, May 29. Garvey, W.D. (1979) Communication: The Essence of Science, New York: Pergamon Press. Gerth, H.H., and Mills, C.W. (1946) From Max Weber: Essays in Sociology, New York: Oxford University Press. Gibson, D.V., and Dearing, J.W. (1988) “Technopoleis: Themes and conclusions”, in R.W.Smilor, G.Kozmetsky, and D.V.Gibson (eds) Creating the Technopolis, Cambridge, MA: Ballinger. Gibson, D.V., and Rogers, E.M. (1994) R&D Collaboration on Trial: The Microelectronics and Computer Technology Corporation, Cambridge, MA: Harvard Business School Press. Glasmeier, A.K. (1988) “The Japanese technopolis program: High tech development strategy or industrial policy in disguise?”, International Journal of Urban and Regional Research, 12(2):268–84. Gottmann, J. (1961) Megalopolis: The Urbanized Northeastern Seaboard of the United States, New York: The Twentieth Century Fund. Granovetter, M.S. (1973) “The strength of weak ties”, American Journal of Sociology, 78:1360–80. Griffith, B.C., and Mullins, N.C. (1972) “Coherent social groups in scientific change”, Science, 177:959–64. Griffith, B.C., Jahn, M.J., and Miller, A.J. (1971) Science, 173:164–6. Hagstrom, W.O. (1965) The Scientific Community, Carbondale, IL: Southern Illinois University Press. Hakansson, H. (1989) Corporate Technological Behavior: Co-operation and Networks, London: Routledge. Hall, P. (1986) (ed.) Technology Innovation and Economic Policy, Oxford: Philip Allan. Hall, P., Breheny, M., McQuaid, R., and Hart, D. (1987) Western Sunrise: The Genesis and Growth of Britain’s Major High Tech Corridor, London: Allen & Unwin, 154–5. Hamilton, D.P. (1992a) “Has Tsukuba put its worst days behind it?”, Science, 258:572.
References
207
——(1992b) “Industry steps in to fill the gap in basic research”, Science, 258:570–1. Hardin, G. (1968) “The tragedy of the commons”, Science, 162:1243–8. Haseni, Y. (1989) Research scientist, Head, Fire Safety Division, Building Research Institute, personal interview, Tokyo, June 22. Hashimoto, S. (1967) “The research environment at research institutes and universities in Europe and in the United States”, in Science and Technology Agency Planning Division (eds) Special Issue: Research and Academic City, Tokyo: Science and Technology Agency (in Japanese). Hattori, T. (1989) President, Tsukuba Supporting Center, personal interview, Tsukuba, May 22. Havelock, R.G., and Elder, V. (1987) “Technology transfer in Japan: An exploratory review”, unpublished paper, Center for the Productive Use of Technology, George Mason University. Hilbert, R.A. (1988) “Bureaucracy as belief, rationalization as repair: Max Weber in a post-functionalist age”, Sociological Theory, 5:70–86. Hirano, Y. (1992) “Public and private support of basic research in Japan”, Science, 258:582–3. Hoch, P.K. (1987) “Migration and the generation of new scientific ideas”, Minerva, 25(3):209–37. Holden, C. (1993) “Japan science city shapes up”, Science, 260:757. Holzner, B., Campbell, D.T., and Shahidullah, M. (1985) “Introduction: The Comparative Study of Science and the Sociology of Scientific Validity”, Knowledge: Creation, Diffusion, Utilization, 6(4): 307–28. Housing and Urban Development Corporation (1984) Tsukuba Science City, Tsukuba: Tsukuba Development Bureau. ——(1989a) Kansai Science City, Kyoto: Kansai Science City Development Bureau. ——(1989b) Tsukuba, Tsukuba: Tsukuba Development Bureau. Hyman, H. (1968) “Reference groups”, International Encyclopedia of the Social Sciences, 13:353–66. Ibaraki Prefecture, Housing and Urban Development Corporation, and Urban Planning Association (1988) (eds) Tsukuba: A Challenge for the 21st Century, Tsukuba: Tsukuba STEP (in Japanese). Ibi, T. (1989) Director, General R&D Division, Industrial Technology Department, New Energy and Industrial Technology Development Organization, personal interview, Tokyo, June 7. Igarashi, R. (1989) General Manager, Data Processing Division, E.D.P Department, Japan Information Center of Science and Technology, personal interview, Tokyo, June 8. Institute for Scientific Information (1987) Current Contents Address Directory, Science and Technology Geographical Index, Philadelphia: Institute for Scientific Information. ——(1992) Science Indicators Data Base, Philadelphia: Institute for Scientific Information. Iseya, F. (1989) Researcher, Environmental Research Center and Institute of Geoscience, University of Tsukuba, personal interview, Tsukuba, May 9. Ishibashi, K., and Furutani, H. (1989) Ph.D. candidates, Institute of Engineering Mechanics, University of Tsukuba, personal interviews, Tsukuba, May 31.
208
References
Ishikawa, M. (1980) “Studies of Moving from Large Cities and of the Characteristics of a New City Built Around Research and Educational Institutions”, Ph.D. dissertation, English summary, University of Tokyo. ——(1989) Vice-President, Nagaoka Technology University, personal interview, Tokyo, April 8. Ishikawa, T. (1992) Personal Secretary and Policy Advisor to Kouno Yohei; Personal Secretary to Ko¯no Ichiro¯, personal interview, Tokyo, March 28. Ishimaru, H. (1989a) “Ultimate pressure of the order of 10 -13 Torr in an aluminum alloy vacuum chamber”, Journal of Vacuum Science and Technology, May-June. ——(1989b) Research Scientist, National Laboratory for High-energy Physics, personal interview, Tsukuba, May 17. Ito¯, Y. (1986) “Telecommunications and industrial policies in Japan: Recent developments”, in M.S.Snow (ed.) Marketplace for Telecommunications, New York: Longman. Izumi, F. (1989) Research Scientist, National Research Institute in Inorganic Materials, personal interview, Tsukuba, May 1. Izumi, F., Takayama-Muromachi, E., Nakai, Y., and Asano, H. (1989) “Structure refinement of La Ca Cu O with neutron powder diffraction data”, Physica C, 157:89–92.1.9 1.1 2 6 Jackson, P., and Smith, S.J. (1984) Exploring Social Geography, London: George Allen & Unwin. Japan External Trade and Research Organization (1988) “New gene bank completed”, New Technology Japan, 16(6):6–7. The Japan Times (1988) “Tsukuba in demand as a sister city”, 3, October 27. Jick, T.D. (1979) “Mixing qualitative and quantitative methods: Triangulation in action”, Administrative Science Quarterly, 24, December. Johnson, C. (1982) MITI and the Japanese Miracle: The Growth of Industrial Policy, 1925–1975, Stanford, CA: Stanford University Press. Jo¯yo¯ Shinbun (1962) “Ko¯no pays particular interest to water availability, climate, and the ease of land acquisition”, December 15 (in Japanese). ——(1963a) “Mount Tsukuba chosen as site for government administrative city”, August 28 (in Japanese). ——(1963b) “Let’s try to avoid the notorious real estate brokers”, August 29 (in Japanese). ——(1963c) “Local residents against development send smoke signal to Ibaraki officials”, November 5 (in Japanese). Kamada, K. (1989) Manager, Omron Tateishi Electronics Company, Tsukuba R&D Laboratory, personal interview, Tsukuba, May 22. Kaneko, H. (1989) Manager, Total Energy System Development and Service Section, Tokyo Gas Company, personal interview, Tokyo, June 2. Kansai Research Institute (1994a) Kansai Science City Volume 1: Guide to Kansai Science City, Kansai Science City: Kansai Research Institute. ——(1994b) Kansai Science City Volume 2: Guide to Facilities in KSC, Kansai Science City: Kansai Research Institute. Kanter, R. (1988) “When a thousand flowers bloom: Structural, collective, and social conditions for innovation in organizations”, Research in Organizational Behavior, 10:169–211. Kato¯, E. (1992a) “Conditions for the information producing city”, Quest for the
References
209
New Age in Cities: An Interim Report of the History and the Prospect of Cities, Tokyo: Nippon Toshi Center, 47–63. ——(1992b) Professor, University of Tsukuba, personal interview, Tsukuba, March 23. Kawabata, Y. (1989) Research Scientist, National Chemical Laboratory for Industry, personal interview, Tsukuba, May 16. Kawamoto, T. (1967) “Novosibirsk Science City—A prototype of a planned academic city”, in Science and Technology Agency Planning Division (eds) Special Issue: Research and Academic City, Tokyo: Science and Technology Agency (in Japanese). ——(1980) “Life and research activities in Tsukuba Science City”, Toki no Ho¯rei, 1084, September 3 (in Japanese). ——(1984) “The status of research collaboration in Tsukuba Science City”, unpublished paper, September. ——(1987) “Frankly speaking, on research communication in science and technology”, Nikkei-Sangyo¯ Shinbun, April 14, 15, 16, 21, 22, 23 (in Japanese). ——(1988) “Lessons learned about scientific communication in Tsukuba Science City: Looking at the grassroots”, paper presented at the Symposium on Science Communication, Annenberg School of Communications, University of Southern California, Los Angeles, December 15–17. ——(1989a) Chief Officer, Tsukuba Research Consortium, personal interview, Tsukuba, March 28. ——(1989b) Personal interview, Tsukuba, March 31. ——(1989c) Personal interview, Tsukuba, April 20. ——(1989d) Personal interview, Tsukuba, June 6. ——(1990) “Tsukuba Science City”, Science and the Future, Encyclopaedia Britannica. ——(1992a) Personal interview, Tsukuba, March 27. ——(1992b) “An Assessment of Science Cities: Lessons from the experience of Tsukuba Science City”, paper presented at the International Symposium on Development Strategies for Science Towns, Taejon, Korea, November 19–20. Kinoshita, J. (1993) “Is Japan a boon or a burden to U.S. industry’s leadership?”, Science, 259:596–8. Kitahara, Y. (1983) Information Network System: Telecommunication in the Twenty-First Century, London: Heinemann Educational Books. Kobayashi, T. (1989) General manager, Jo¯yo¯ Bank, Tsukuba Science City Branch, personal interview, Tsukuba, June 14. Kondo¯, J. (1988) “Science in Japan”, unpublished paper. ——(1992) President, Science Council of Japan, personal interview, Tokyo, March 26. Krauss, E.S., Rohlen, T.P., and Steinhoff, P.G. (1984) (eds) Conflict in Japan, Honolulu: University of Hawaii Press. Kraut, R.E., Egido, C., and Galegher, J. (1990) “Patterns of contact and communication in scientific research collaborations”, in J.Galegher, R.E.Kraut, and C.Egido (eds) Intellectual Teamwork: Social and Technological Foundations of Cooperative Work, Hillsdale, NJ: Lawrence Erlbaum Associates.
210
References
Kumata, Y. (1988) Personal conversation, Tokyo, September 5. ——(1989) Professor, Department of Social Engineering, Tokyo Institute of Technology, personal interview, Tokyo, March 26. Kuroda, K. (1989) Research scientist and Chief, Hydrology Section, Geological Survey of Japan, personal interview, Tsukuba, May 9. Kurokawa, T. (1989) Professor, Department of Socio-economic Planning, University of Tsukuba, personal interview, Tsukuba, May 8. Kyoto Prefectural Government (1991) Kansai Science City, Kyoto: Office of Science City Public Works and Construction Department. Lakoff, S.A. (1966) “The third culture: Science in social thought”, in S.A. Lakoff (ed.) Knowledge and Power, New York: Free Press. Larsen, J.K. (1988) “Lessons learned about technology transfer”, paper presented at the Symposium on Science Communication: Health and Environmental Research, Los Angeles, December 15–17. Lasswell, H. (1959) “The social setting of creativity”, in H.Anderson (ed.) Creativity and its Cultivation, New York: Harper & Brothers. Leonard-Barton, D. (1988) “Implementation characteristics of organizational innovations”, Communication Research, 15(5):603–31. Lewis, R. (1989) Research Scientist, Tsukuba Research Consortium, personal interview, Tsukuba, May 8. Lievrouw, L.A., Rogers, E.M., Lowe, C.U., and Nadel, E. (1987) “Triangulation as a research strategy for identifying invisible colleges among biomedical scientists”, Social Networks 9:211–48. Lonner W.J., and Berry, J.W. (1986) “Sampling and surveying”, in W.J. Lonner, and J.W.Berry (eds) Field Methods in Cross-cultural Research, Newbury Park, CA: Sage. Lynn, L. (1986) “Japanese research and technology policy”, Science, 233: 297, July 18. McMullan, W.E., Long, W.A., Ray, D.M., and Vesper, K.H. (1987) “New venture development: The Calgary experience”, unpublished paper, School of Management, University of Calgary, Alberta. Mansfield, E. (1988) “Industrial innovation in Japan and the UnitedStates”, Science, 241:1769–74, September 30. Marshall, E. (1986) “School reformers aim at creativity”, Science, 233: 267–70, July 18. Masser, I. (1990a) “Cross-national learning for urban policy making”, Cities, February:25–30. ——(1990b) “Technology and regional development policy: A review of Japan’s Technopolis Programme”, Regional Studies, 24(1):41–53. Matsuda, N. (1989) Professor, University of Tsukuba, personal interview, Tsukuba, May 8. Merton, R.K. (1968) Social Theory and Social Structure, New York: Free Press. ——(1988) “Reference groups, invisible colleges, and deviant behavior in science”, in H.J.O’Gorman (ed.) Surveying Social Life: Papers in Honor of Herbert H.Hyman, Middletown, CT: Wesleyan University Press. Mikado, E. (1989) Chief, Tsukuba Bureau, Asahi Shinbun, personal interview, April 18.
References
211
Ministry of Transportation (1985) Plan for a High-speed Transportation Network in and around Tokyo, Tokyo: Research Center for Transportation Economics, Regional Transportation Office (in Japanese). Mitroff, I.I. (1974) “Norms and counter-norms in a select group of theApollo moon scientists: A case study of the ambivalence of scientists”, American Sociological Review, 39:579–95. Mitsui Trade News (1987) “New center to link government, academic, and business sectors”, 24(6):7–8, November/December. Mizukami, A., Sugiyama Y., and Ikematsu, M. (1989) Research Scientists, Tsukuba Research Center, Sanyo Electric Company Research Laboratory, group interview, Tsukuba, June 9. Monge, P.R. (1987) “The network level of analysis”, in C.R.Berger, and S.H.Chaffee (eds) Handbook of Communication Science, Newbury Park, CA: Sage, 239–70. Mulkay, M.J. (1974) “Methodology in the sociology of science: Some reflections on the study of radio astronomy”, Social Science Information, 13:109–19. Mullins, N.C. (1972) “The development of a scientific specialty: The phage group and the origins of molecular biology”, Minerva, 10(1): 51–82. Nagakura, S. and Kikumoto, H. (1994) “New developments in the science policy of Japan”, Science, 266:1189–90. Nakayama, S. (1991) Science, Technology and Society in Postwar Japan, London: Kegan Paul. Namba, K. (1989) Research Scientist, ERATO, personal interview, Tsukuba, May 8. Narin, F. and Frame, J.D. (1989) “The growth of Japanese science and technology”, Science, 245:600–5, August 11. National Land Agency (Metropolitan Areas Development Bureau) (1989) “New Tsukuba project: Aiming at an international science and technology city”, pamphlet (in Japanese). National Research Center for Disaster Prevention (1989) “Study for the prediction system of slope failure by heavy rain”, Research Notes of the National Research Center for Disaster Prevention, 82, March. National Science Board (1987) Science and Engineering Indicators—1987, Washington, DC: National Science Board. Nihon Keizai Shinbun (1989) “Industries still interested in opening a branch in Tsukuba”, May 3 (in Japanese). Nishikawa, T. (1989) Former Director General, National Laboratory for Highenergy Physics, personal interview, Tokyo, April 21. Nishiyama, Y. (1986) “Western influence upon planning administration in Japan: Focus on land management”, in H.Nagamine (ed.) Urban Development Policies and Programmes: Focus on Land Management, UNCRD, Nagoya, 315–55. Noguchi, T. (1979) “The formation of the Japanese zaibatsu—the political merchant in the original accumulation of capital”, Keio Business Review, 16:169–87. Nord, W.R., and Tucker, S. (1987) Implementing routine and radicalinnovations, Lexington, MA: Lexington Books.
212
References
Normile, D. (1994) “Universities and companies learn benefits of teamwork”, Science, 266:1174–5. Ogasawara, R., and Salvaggio, J.L. (1983) “Public policy for the information society: The Japanese model”, in J.L.Salvaggio (ed.) Telecommunications: Issues and Choices for Society, New York: Longman. Okamura, F. (1989a) Research Scientist, National Institute for Research in Inorganic Materials, personal interview, Tsukuba, May 12. ——(1989b) Personal interview, Tsukuba, May 29. Okamura, F.P., Sueno, S., Nakai, I, and Ono, A. (1987) “Crystal structure of Ba YCu O y, a superconductive compound determined by x-ray single 2 3-x 7crystal diffraction method”, Materials Research Bulletin, 22:1081–5. Okazaki, H. (1989) President, Techno Culture Tsukuba, personal interview, Tsukuba, May 31. Olofsson, C., Reitberger, G., Tovman, P., and Wahlbin, C. (1987) “Technologybased new ventures from Swedish universities: A survey”, in N.C.Churchill, J.A.Hornaday, B.A.Kirchhoff, O.J.Krasner, and K.H.Vesper (eds) Frontiers of Entrepreneurship Research, Wellesley, MA: Babson College. O¯nda, M. (1988) “Tsukuba Science City complex and the Japanese technopolis strategy”, in R.W.Smilor, G.Kozmetsky, and D.V.Gibson (eds) Creating the Technopolis, Cambridge, MA: Ballinger. Onoda, H. (1989) Director, Omron Tateishi Electronics Company Research Laboratory, personal interview, Tsukuba, May 22. O¯uchi, W.G. (1984) The M-Form Society: How American Teamwork Can Recapture the Competitive Edge, Reading, MA: Addison-Wesley. Oyagi, N. (1989) Research Scientist, National Research Center for Disaster Prevention, personal interview, Tsukuba, May 9. Pake, G.E. (1985) “Research at Xerox PARC: A founder’s assessment”, IEEE Spectrum, 54–61, October. Park, R.E. (1926) “The urban community as a spatial pattern and a moral order”, reprinted in C.Peach (1975) (ed.) Urban Social Segregation, London: Longman. Peck, M.J. (1986) “Joint R&D: The case of Microelectronics and Computer Technology Corporation”, Research Policy, 15:219–31. Pelz, D.C., and Andrews, F.M. (1966) Scientists in Organizations: Productive Climates for Research and Development, New York: Wiley. Pempel, T.J. (1982) Policy and Politics in Japan, Philadelphia: Temple University Press. Polanyi, M. (1945) “The planning of science”, Political Quarterly, 16, October, 316–28. ——(1962) “The republic of science”, Minerva, I(1). Price, D.de Solla (1961) Science Since Babylon, New Haven, CT: Yale University Press. ——(1963) Little Science, Big Science, New York: Columbia University Press. Price, D., and Beaver, D. (1966) “Collaboration in an invisible college”, American Psychologist, 21(6):1011–18, November. Przeworski, A., and Teune, H. (1970) The Logic of Comparative Social Inquiry, New York: John Wiley. Ratcliffe, J.W. (1983) “Notions of validity in qualitative research methodology”, Knowledge: Creation, Diffusion, Utilization, 5(2), December.
References
213
Rice, R.E., and Richards, W.D. (1985) “An overview of network analysismethods and programs”, in B.Dervin, and M.J.Voigt (eds) Progress in Communication Sciences, Norwood, NJ: Ablex. Richards, W.D. (1986) The NEGOPY Network Analysis Program, Burnaby, Canada: Department of Communication, Simon Fraser University. ——(1989) NEGOPY 3.3 personal computer software program, April. Roberts-Gray, C., and Gray, T. (1983) “Implementing innovations”, Knowledge: Creation, Diffusion, Utilization, 5(2), 213–32. Rogers, E.M. (1983) Diffusion of Innovations, Third Edition, New York: Free Press. ——(1987) “Progress, problems and prospects for network research: Investigating relationships in the age of electronic communication technologies”, Social Networks, 9:285–310. Rogers, E.M., and Chen, Y.A. (1989) “Technology transfer and the technopolis”, in S.A.Mohrman, and M.A.Von Glinow (eds) High Technology Organizations. Rogers, E.M., and Dearing, J.W. (1989) “University-industry technology transfer in information societies”, in B.D.Ruben, and L.A. Lievrouw (eds) Information and Behavior, 5, New Brunswick, NJ: Transaction. ——(1990) “Turbocharging innovation in three cities”, IEEE Engineering Management Review, Spring: 2–10. Rogers, E.M., and Kincaid, D.L. (1981) Communication Networks: Toward a New Paradigm for Research, New York: Free Press. Rogers, E.M., and Larsen, J.K. (1984) Silicon Valley Fever, New York: Basic. Rokeach, M. (1979) “From individual to institutional values: With special reference to the values of science”, in M.Rokeach (ed.) Understanding Human Values, New York: Free Press. Saga, J. (1987) Memories of Silk and Straw, Tokyo: Kodansha International. Sakakibara, K. (1983) “From imitation to innovation: The Very Large Scale Integrated (VLSI) Semiconductor Project in Japan”, Cambridge, MA: Alfred P.Sloan School of Management Working Paper, Massachusetts Institute of Technology. Sakurai, M. (1989) “MITI lab nurtures ‘neural eye’”, Electronic Engineering Times, 535:39, April 24. Samuels, R.J. (1983) The Politics of Regional Policy in Japan, Princeton, NJ: Princeton University Press. Saxenian, A. (1985) “The genesis of Silicon Valley”, in P.Hall, and A.Markusen (eds) Silicon Landscapes, Boston, MA: Allen & Unwin. ——(1988) “The Cheshire cat’s grin”, Technology Review, 91(2): 67–75. ——(1989) “Regional networks and the resurgence of Silicon Valley”, paper presented to the Association of Collegiate Schools of Planning, Portland, OR, October 4–7. ——(1994) Regional Advantage, Cambridge, MA: Harvard University Press. Saxonhouse, G.R. (1986) “Why Japan is Winning”, Issues in Science and Technology, Spring: 72–80. Schott, T. (1994) “Emerging and Declining Centers of Engineering Science: Japan and the United States”, Knowledge: Creation, Diffusion, Utilization, 15(4):417–56. Science (1986), 233, July 18.
214
References
Science and Technology Agency (1986) Research Institute Index for Tsukuba Science City, Tokyo: Science and Technology Agency. ——(1988a) White Paper on Science and Technology 1987: Toward the Internationalization of Japan’s Science and Technology (English summary), Tokyo: Foreign Press Center. ——(1988b) “Execution plan for research and experiments”, Tokyo: Science and Technology Agency (in Japanese). ——(1988c) “And the next”, pamphlet, Tokyo: Science and Technology Agency (in Japanese). ——(1988d) White Paper on Science and Technology 1988: Toward the Establishment of a New Creative Research Environment (English summary), Tokyo: Foreign Press Center. Segal Quince Wicksteed (1985) The Cambridge Phenomenon: The Growth of High Technology Industry in a University Town, Cambridge: Segal Quince Wicksteed. Shima, M. (1988) An Apron Representative and the Tsukuba Renaissance, Tsukuba: Tsukuba Sho¯rin (in Japanese). ——(1989) Former Representative, Sakura Village, Ibaraki Prefecture, personal interview, Tsukuba, May 16. Shimoda, M. (1994) “Kansai Science City and regional development policy”, paper presented at the International Science City Symposium, Kansai Science City, October 17–18. Sigurdson, J. (1986) “Industry and state partnership in Japan”, Discussion Paper 168, Lund, Sweden: Research Policy Institute. Smilor, R.W., Kozmetsky, G., and Gibson, D.V. (1988a) (eds) Creating the Technopolis, Cambridge, MA: Ballinger. ——(1988b) “The Austin/San Antonio corridor: The dynamics of a developing technopolis”, in R.W.Smilor, G.Kozmetsky, and D.V.Gibson (eds) Creating the Technopolis, Cambridge, MA: Ballinger. Society for Recording Life Histories in Tsukuba Research and Academic City (1981) (eds) Tall Boots and Starry Skies, Tsukuba: Tsukuba Sho¯rin (in Japanese). Spradley, J.P. (1979) The Ethnographic Interview, New York: Holt, Rinehart and Winston. Stein, M. (1963) “A transactional approach to creativity”, in C.Taylor, and F.Barron (eds) Scientific Creativity: Its Recognition and Development, New York: John Wiley & Sons. Stevens, J.E. (1994) “Japan picks a winner in the Rice Genome Project”, Science, 266:1186–7. Sun, M. (1987) “Japan’s inscrutable research budget”, Science, 238:22, October 2. ——(1989a) “Japan faces big task in improving basic science”, Science, 243:1285–7, March 10. Takagi, J. (1989a) Director, Tsukuba Center for Institutes, personal interview, Tsukuba, April 20. ——(1989b) Personal interview, Tsukuba, June 14. Takahashi, H. (1989) Professor, University of Tsukuba, personal interview, Tsukuba, April 27.
References
215
Takayama, E. (1989) President, Ko¯gakuin University and Professor Emeritus, University of Tokyo, personal interview, Tokyo, April 12. ——(1992) Personal interview, Tokyo, March 28. Tanaka, K. (1989) Research Geologist, National Research Center for Disaster Prevention, personal interview, Tsukuba, April 20. Tatsuno, S. (1986) The Technopolis Strategy: Japan, High Technology, and the Control of the Twenty-first Century, New York: Prentice Hall. Taubes, G. (1994) “Japan learns to accommodate its global research partners”, Science, 266:1181–3. Tokumaru, K. (1989) Vice-President for Research Development and Professor, University of Tsukuba, personal interview, Tsukuba, June 14. Traweek, S. (1988) Beamtimes and Lifetimes: The World of High Energy Physicists, Cambridge, MA: Harvard University Press. Tsuchida, A. (1989) Partner, Urban Environment Research Consultants, personal interview, Tokyo, April 6. Tsuge, S. (1989) Professor, University of Tsukuba, personal interview, Tsukuba, May 17. Tsukuba Center for Institutes (1979) Science Communication, June 1. Tsukuba Institute Coordinating Committee (1987) The Use of a Communication Network: Final Report, Tsukuba: Special Committeefor Computer Usage. Tsukuba Mook (1989) “The future of Southwest Ibaraki Prefecture”, May 18 (in Japanese). Tsukuba Planning Department (1985) “Looking toward tomorrow’s international brain city”, Tsukuba: Tsukuba Planning Department (in Japanese). Tsukuba Press Room (1989) Newspaper correspondents, personal conversations, March 24. Tsukuba Research Consortium (1986) “Tsukuba Research Consortium”, pamphlet. ——(1988a) “Main theme of joint research”, May. ——(1988b) “Research organizations list”. ——(1989) “Employee list”, June 12. ——(1990) “Summary of research organizations in Tsukuba”. ——(1991) “List of located companies in Tsukuba”, June. Ueda, H. (1989) Research Scientist, National Institute for Environmental Studies, personal interview, Tsukuba, June 9. University of Tsukuba (1984) “Outline of the University of Tsukuba”, pamphlet. ——(1987) “Outline of the University of Tsukuba”, pamphlet. Van de Ven, A.H., and Rogers, E.M. (1988) “Innovations and organizations: Critical perspectives”, Communication Research, 15(5):632–51. von Hippel, E. (1987) “Cooperation between rivals: Informal knowhow trading”, Research Policy, 16:291–302. Wakabayashi, T. (1985) “A study of the role and function of the Tsukuba Research and Academic City master planning process”, Ph.D. dissertation, Tokyo: University of Tokyo (in Japanese). Wakasa, K. (1989) Research Scientist, National Institute of Agrobiological Resources, personal interview, Tsukuba, May 12.
216
References
Wakasugi, R. (1988) “Technological innovation and diffusion in Japan and Asian NIES”, paper presented at the Third International Technical Innovation and Entrepreneurship Symposium, Gold Coast, Australia, September 5–8. Weber, M. (1946) “Science as a Vocation”, a speech to students of Munich University (1918), in H.H.Gerth and C.W.Mills (eds) From Max Weber: Essays in Sociology, New York: Oxford University Press. Westney, D.E. (1987) Imitation and Innovation: The Transfer of Western Organizational Patterns to Meiji Japan, Cambridge, MA: Harvard University Press. Whittington, D. (1985) (ed.) High Hopes for High Tech: Microelectronics Policy in North Carolina, Chapel Hill, NC: University of North Carolina Press. Williams, F., and Gibson, D.V. (1990) (eds) Technology Transfer: A Communication Perspective, Newbury Park, CA: Sage. Yabe, A. (1989) Research Scientist, Mechanical Engineering Laboratory, personal interview, Tsukuba, May 17. Yajima, K. (1989a) Partner, Urban Environment Research Consultants, personal interview, Tokyo, April 6. ——(1989b) Personal interview, Tsukuba, May 1. Yamagiwa, K. (1989) “New LB film shows promise of future super-conductor behavior”, Nikkei High Tech Report, 4(11):7–8, April 10. Yin, R.K. (1984) Case Study Research: Design and Methods, Newbury Park, CA: Sage.
Index
Abe, Shintaro¯ 189 Advanced Telecommunications Research Institute International 175 Aerological Observatory 130–2 Agency of Industrial Science and Technology (MITI) 79, 112, 114, 148 agglomerations 39–40; and communication, benefits from 164 Agriculture, Ministry of 68, 70, 92; in collaboration networks 148–9; initiatives of 112–13, 115; laboratories, geographically proximate complexes 168–9 Akademgorodok science city (Russia) 51, 53, 74, 76 Akagi, as science city site 50 Akashi, S. 194 Akito, Arima 29 Amabile, T. 38–9 Amman, C. 37 Anderson, A. 28, 29, 32, 79–80 Ando¯, Wataru 98–9, 193 Andrews, F.M. 36 ARPAK International Institute of Kyoto 175 articles published, in networks 140– 2; by language 127 Asahi Evening News 17 Asahi Shinbun 187, 188 Asahi Shinbun Regional Edition 1
Asakawa, M. 1 Asano, S. 55, 57, 77 authorship of articles 127, 140–7; and co-authorship see coauthorship; English 136, 138, 142, 143, 147; by institute 145–6; Japanese 133, 135, 140, 141, 146; single authors 140, 141, 142, 143; trends in 144 Automobile Research Institute 129, 132, 151 automobiles, travel by 5–6 Aydalot, P. 158 Bacon, Francis 34 Bartholomew, J.R. 26, 27, 32, 88, 157 Beaver, D. 125 Benedict, R. 33 Berry, J.W. 185 BertonLewis, H. 77 Birnbaum, H. 51, 53, 77 Bloom, J.L. 55, 57, 77 Botkin, J.W. 81 Bradburn, N.M. 186 Brinkman, W. 30, 32 Brislin, R.W. 186 Broad, W.J. 30 Brooks, Harvey 37 Building Research Institute 96; coauthorship, network analysis of 129–32, 146; publications and productivity 150
218
Index
Cambridge (Great Britain) 36 Campbell, D.T 178 Canon Corporation 175 Capital City Development Commission (Japan) 53 Capital Region Development Commission (Japan) 69, 71, 75 Carter, A.P. 41 Chen, Y.A. 6 Chiba, Genya 79, 106–7, 109, 110–11, 175, 193, 195 Chubin, D.E. 125 Cicourel, A.V. 182 Clark, K.B. 36 co-authorship of articles: English 136, 138, 142, 143, 147; Japanese 133, 135, 141, 143, 146; and links, analyzed 126, 127, 134, 165, 199; and ministries 147–52, 165; organizational links within Tsukuba 129, 130, 131, 132; Tokyo, links with 137–8, 144; university-institutional 138, 139, 140, 144–6, 165 collaboration networks 124–57; and agglomerations, benefits from 164; articles published 127, 140–3; authorship of articles 127, 140–2, 143–7; and communication, benefits from 163–4; conceptualizing 126–7; failure of 167–8; first authorship 152–4; and informal study groups 164–5; institutes and publication rates 152, 201; investigation methods 196–9; lessons learned 163–70; links within Tsukuba 139; mapping of 124–6; and ministries 147–52; network analysis of 128; and proximity 167; in science cities 37–9; structural view of 127–39, 200 communication: and access to knowledge 171; and agglomerations, benefits from 164; government initiatives at 111–17; lessons learned 163–70;
motivated, benefits 163–4; multidisciplinary, as goal 166; ofpre-competitive corporate knowledge 166–7; problems, at Kansai 176; in science cities 37–9, 158; and Tsukuba Center Inc. 112; at Tsukuba Research Consortium 106–10 computer systems, Tsukuba Network on 120, 122 concept of science cities: contest for Tsukuba site 49–51; and crowding in Tokyo 46–9; domestic irritations 55–6; history of 45–62; investigation methods 184–7; and protest over land values 54–5, 188 conflict: among ministries 161–2; impact on research 162; in implementation of Tsukuba 159–60, 161–2 Construction, Ministry of 68, 69, 71, 73–5, 92; in collaboration networks 148–9; initiatives of 115 Converse, J.M. 185, 186 creativity: and communication, benefits from 163–4; lessons learned 163–70; in science cities 37–9; synergistic, in science 43 Csikszentmihalyi, M. 38, 39 Cummings, W.K. 32 Cutler, R.S. 32 Dill, D. 39 Dohi, Hiroshi 2, 69, 75, 179, 188 Dore, R.P. 33 Doshishi University 173 economic development: changes in meaning of 162; conflictual, impact on research 162; science in 58–62 Edge, D.O. 125 Education, Ministry of 21, 48–9, 69; in collaboration networks 148–9; exchange programs 104–5, 107; and TAGS 92 Egido, C. 35
Index Eisenberg, E.M. 201 Elder, V. 87 Electrotechnical Laboratory (MITI) 23, 24, 96; co-authorship, network analysis of 129–32, 143, 145–7; publications and productivity 150 Emerson, R.M. 182 employment 13–15 English co-authorship of articles 136, 138, 142, 143, 147 Environmental Agency (Japan) 29, 64, 92; in collaboration networks 148–9 Esaki, Leo 10 Eveland, J.D. 63 Exploratory Research for Advanced Technology Organization (ERATO) 104, 106, 108 Eyring, H. 38 Fujimoto, T 36 Foreign Press Center (Japan) 68, 189 Forestry and Forestry Products Research Institute 131–2 Forster, E.M. 37 Foundation for the Advancement of International Science 131–2 Frame, J.D. 4, 30 Frenkel, O. 33 Friedkin, N. 37 Fruit Tree Research Station 129, 132, 151 Fuji, as science city site 50 Fujimoto, K. 9 Fujita, Takeshi 90, 99–100, 193 Fukuda, Nobuyuki 69, 77, 189, 190 Furutani, H. 193 Galegher, J. 35 Garvey, W.D. 36 Genetic and Bioengineering Study Group 91, 101–3, 194; attendees at meetings 102 Geographical Survey Institute 130–1, 151
219
Geological Survey of Japan 129–32, 145–6, 151 Gerth, H.H. 63 Gibson, D.V. 81, 158, 176, 181 Glasmeier, A.K. 12 Gottman, Jean 39–40 government: exchange programs in communications 104–11; failure of collaboration networks 167–8; initiatives in communications 111–17; ministries of see ministries Granovetter, M.S. 125 Gray, T. 63 Griffith, B.C. 37 Gryskiewicz, S. 38 Hagstrom, W.O. 35 Hakansson, H. 158 Hall, P. 158 Hamilton, D.P. 2, 30, 32 Hardin, G. 201 Haseni, Y. 193 Hashimoto, S. 53, 58, 69 Hashimoto Tomisaburo¯ 50 Hatoyama, Ichiro¯ 70 Hattori, Tadashi 115–16, 195 Havelock, R.G. 87 Hayashi, Chikara 106, 107 Health, Ministry of 149 Heat Transfer Colloquium 91, 94–7; collaborative research in 97; institutions involved 96 Hermodson, A. 63 heterophily: in paradox 41–3; in study groups 103 high-pressure vacuum chambers 25–6 Hilbert, R.A. 63 Hirano, Y. 29 Hiruma, Teruo 106, 107 Hitachi, in Tsukuba Science City 95–6, 100 Hoch, P.K. 36 Holden, C. 174 Holzner, B. 178 homophily: and heterophily paradox 41–3; in study groups 93
220
Index
Housing and Urban Development Corporation (Tsukuba) 11, 13, 175 Hyman, H. 35 Ibaraki Prefecture (Japan) 69, 76–7, 112 Ibi, Toshio 114–15 Igarashi, R. 197 Ikeda, Hayato 46, 49, 51, 71, 72 Ikematsu, M. 23, 24 implementation of Tsukuba 63–84; conflict in 159–60, 161–2; and Kono Ichiro 68–73;lessons learned 160–9; master plans, changes to 63–8; and planning 162; political champion, needed 161; project management 73–8; as solution to problems 160–1; Tsukuba’s role in Japan 78–81; uncoordinated cooperation in 158 Industrial Products Research Institute 96; co-authorship, network analysis of 130–2; publications and productivity 150 informal study groups 88–91;and access to knowledge 171; changes in 168; evolution of 91–104; expectations of 89–90; formal nature of 88–9; growth of 87; and interpersonal networks 164–5, 166; investigation methods 191–2, 193; in joint research 88–9; role of 117–19, 123; see also Genetic and Bioengineering Study Group; Heat Transfer Colloquium; TAGS; Thermofluid Dynamics Seminar; Tsukuba Chemical Science Club 91–104 innovation, in science cities 34, 38–9, 42 Institute for Scientific Information 182 International Science and Technology Exposition (Tsukuba, 1985) 56–8, 188 Iseya, F. 193
Ishibashi, K. 193 Ishikawa, M. 48, 52, 53, 54, 58, 68, 69, 75, 77, 190 Ishikawa, T. 68, 70, 72, 190 Ishimaru, H. 25 Izumi, F. 20, 22, 85, 183 Jackson, P. 40 Jahn, M.J. 37 Japan: first authorship, analysis of 153; national problems, and implementation of Tsukuba Science City 46–9, 160–1; science in 26–32; science planning 32–3, 183–4 Japan External Trade and Research Organization 193 Japan Housing Corporation 69, 74–5, 115 Japan Information Center for Science and Technology (JICST) 120, 126, 196–7 Japan Research and Development Corporation (JRDC) 104 Japanese co-authorship of articles 133, 135, 141, 143, 146 Jick, T.D. 182 Johnson, C. 33, 70, 72 Jo¯yo¯ Shinbun 2, 50–1, 58, 59, 187 Kamada, K. 82–3 Kaneko, H. 193, 194 Kansai Science City 79; comparisons with Tsukuba 173–7, 202; objectives 174–5 Kansai University of Foreign Studies 173 Kanter, R. 39 Kato¯, E. 6, 8, 58, 70, 72, 187 Kawabata, Y. 24–5 Kawamoto, T. 3, 8, 13, 173, 202; on communications at Tsukuba 86–7, 195; on development of Tsukuba 51, 52, 54, 58; implementation of plan 69, 75–6, 77, 79; on study groups 91, 107, 110, 117, 191, 193, 194
Index Kazuo, Murakami 10 Keeble, D. 158 Keio University 8 KEK (Japan National Laboratory for High-energy Physics) 11, 20, 21, 31, 78; co-authorship, network analysis of 130–2, 143, 145–7; and high-pressure vacuum chambers 25–6; planning of 55–6; publications and productivity 150; and study groups 96; and Tsukuba Network 120 Kikumoto, H. 10 Kincaid, D.L. 199 Kinoshita, J. 17, 29, 30, 32 Kitahara, Y. 33 Knorr-Cetina, K. 37 Kobayashi, Teruo 116 Ko¯gakuin University 74 Kondo¯, Jiro¯ 26, 27–8, 188 Konno, Hiroshi 69, 74, 75 Ko¯no, Ichiro 62, 159; choice of site 1, 4, 50, 187; loss of 68–73, 189–90; as political champion 160, 161 Kozmetsky, G. 158, 181 Krauss, E.S. 33 Kraut, R.E. 35 Kumata, Y. 10, 47, 48, 49 Kuroda, K. 193 Kurokawa, T. 178, 202 Kusunose, Sho¯taro¯ 69 Kyocera Corporation 175 Kyoto, University of 8 Kyoto Prefecture (Japan) 175 labor supply at Tsukuba 80–1 Lakoff, S.A. 174 land use and purchase at Tsukuba 65 land values at Tsukuba: increase 57; and master plans 66;protest over 54–5, 188 Larsen, J.K. 36, 80, 87, 158, 181, 183 Lasswell, H. 38, 39 LB (Langmuir-Blodgett) thin film 22–5
221
Leonard-Barton, D. 63 Lewis, R. 110 Lievrouw, L.A. 182 Lonner, W.J. 185 Lynn, L. 31, 32 MacArthur, Gen. Douglas 70 McMullan, W.E. 181 Manabe, Mitsuhiko 69 Mansfield, E. 30 maps of Tsukuba Science City 5, 7, 60–1 Marshall, E. 31, 64 mass media, on Tsukuba Science City 18–19 Masser, I. 3, 33, 54, 80 master plans for Tsukuba 63–8; and land use 65; and private research 66; and relocation 63–4; of science and academic city 64–5; and social life 66–7; working group on 69, 74 Matsuda, N. 120 Matsushita Electric 175 mavericity in science 43 Mechanical Engineering Laboratory 101, 129, 132, 145–6, 150 Merton, R.K. 3, 35, 87 Meteorological Research Institute 131, 145, 151 Meyer, G. 63 Mikado, E. 179 Miller, A.J. 37 Mills, C.W. 63 ministries: collaboration networks in 147–52, 165, 200–1; conflict among 161–2, 188; laboratories, geographically proximate complexes 168–9 MITI (Ministry of International Trade and Industry) 10, 17, 20, 21, 27; in collaboration networks 148–9, 152; initiatives of 112, 113, 114, 115; laboratories, geographically proximate complexes 168–9; and optical film switching devices 22–
222
Index
5;research supported 23, 32, 79; and study groups 92, 95; Tsukuba as solution to problems 160 Mitroff, I.I. 3 Mitsui Corporation 112–13, 115, 119 Mitterrand, François 57 Mizukami, A. 23, 24 Monge, P.R. 126, 199 Mulkay, M.J. 125 Mullins, N.C. 35, 37 multidisciplinary communication, as goal 166 Nagakura, S. 10 Nakasone, Yasuhiro 50, 73, 189 Nakayama, S. 32, 95 Namba, K. 110 Nara Institute of Science and Technology 173 Narin, F. 4, 30 Narita International Airport 54–5 Nasu, as science city site 50 National Agriculture Research Center 131–2 National Chemical Laboratory for Industry 23, 24, 96, 100; coauthorship, network analysis of 130–2, 145–7; publications and productivity 150 National Food Research Institute 129, 132, 150 National Institute for Environmental Studies 96, 101; co-authorship, network analysis of 130–2, 145–7; publications and productivity 148, 150 National Institute of Agrobiological Resources 131–2 National Institute of Agroenvironmental Sciences 129, 132, 151 National Institute of Animal Health 130–2, 151 National Institute of Animal Industry 130–2, 150
National Institute of Inorganic Materials 130–2, 145, 150 National Institute of Sericulture and Entomological Sciences 129, 132, 150 National Land Agency (Japan) 53, 69, 75, 174, 179; initiatives of 113 national problems, and implementation of Tsukuba Science City 46–9, 160–1 National Research Center for Disaster Prevention 129–32, 150, 193 National Research Institute for Pollution and Resources 96, 101, 131; co-authorship, network analysis of 129–32, 145–6; publications and productivity 150 National Research Institute of Agricultural Engineering 131–2 National Research Laboratory for Metrology 96, 101; coauthorship, network analysisof 129–32; publications and productivity 150 National Science Board (Japan) 125 NEGOPY network analysis of coauthorship 128, 199 Nishikawa, Tetsui 56, 69, 77–8, 86 Nishina, Yoshio 77 Nishiyama, Y. 54 Nobusuke, K. 70 Noguchi, T. 33 Nord, W.R. 63 Normile, D. 29, 104 Oai, Shigeru 98 Ogasawara, R. 33 Okamura, F. 20, 21, 22 Okazaki, Hideaki 113, 115, 116 Okuda, Azuma 174 Olofsson, C. 181 Omron Tateishi Electronics 82–3, 190;co-authorship, network analysis of 130–1 Onda, M. 13, 47 Onoda, H. 82–3
Index optical film switching devices 22–5 Osaka, University of 8 Osaka International University 173 O¯uchi, W.G. 33 Oyagi, N. 193 Pake, G.E. 37 Park, R.E. 40 Peck, M.J. 37 Pelz, D.C. 36 Pempel, T.J. 49 planning: and implementation of Tsukuba 162; revisions in 59–62, 188–9; and science in Japan 32– 3; sequence, in Japan 67; of Tsukuba Science City 12–13, 53 Polanyi, Michael 36–7, 38, 53 Presser, S. 185, 186 Price, D.de Solla 43, 125 Primate Center for MedicalScience 130–2, 152 private research: electronics, laboratory for 82–3, 190; growth of 172–3; and informal study groups 166; in master plans for Tsukuba 66; in Tsukuba 17 project management at Tsukuba 73–8 protests over Tsukuba Science City 54–5, 188 proximity: and collaboration 167; in science cities 39–41 Przeworski, A. 193 Public Works Research Institute 130–2, 145–6, 150 railway access to Tsukuba 5, 177–9 Ratcliffe, J.W. 186 reference groups in social networks 42, 184 research and development: conflict, impact on 162; distinction between basic and applied, at Tsukuba 79; growth of laboratories 16; importanceof 13– 15; institutions 14–15, 31; private
223
see private research; in science cities 35; spending on 28–30 research communications 85–123; attitudes to at Tsukuba 86–7; government exchange programs 104–11; government initiatives 111–17; informal study groups 88–91; investigation methods 191–2, 193; limiting of 172–3; role of 117–23; science culture, establishing 86–8; study groups, evolution of 91–104 research communities see science cities Research Institute for Polymers and Textiles 24, 130–2, 145–6, 150 Rice, R.E. 199 Richards, W.D. 199, 200 Roberts-Gray, C. 63 Rogers, E.M. 3, 6, 87, 158, 176, 181, 199; on implementation of Tsukuba 63, 80; on science cities 36, 38, 183 Rohlen, T.P. 33 Rokeach, M. 87 Saga, J. 46 Sakakibara, K. 33 Sako, M. 33 Salvaggio, J.L. 33 Samuels, R.J. 184 Sato¯, Eisaku 72–3 Saxenian, A. 6, 36, 41, 158 Saxonhouse, G.R. 33, 183 Schott, T. 4, 32 science: in economic development 58–62; in Japan 26–32 Science and Technology Agency (Japan) 20, 28, 29, 30; andchoice of site 51, 52, 53; co-authorship data 126; in collaboration networks 148–9, 152; exchange programs 104–5, 106, 108; as government initiative 111–13, 115; and planning of Tsukuba 69, 75–6; and study groups 92;Tsukuba as solution to problems 160; and Tsukuba Expo
224
Index
’85 56–8, 188; and Tsukuba Network 120–2, 195–6 science cities 34–44; communication, collaboration and creativity in 37–9; concept of, history of 45–62; contest for site of 49–51; definition 35–6; homophily-heterophily paradox 41–3; international growth of 51; investigation methods 184–7; key questions 44; new, in Japan 79–80; proximity in 39–41 science culture, establishing at Tsukuba 86–8; investigation methods 191–2, 193 scientific publications 6, 8 Segal Quince Wicksteed 36, 158, 181 Shahidullah, M. 178 Shima, M. 62, 179 Shimadzu Corporation 175 Shimoda, Minoru 175 Sigurdson, J. 33 Silicon Valley (California) 36, 40–1, 158, 183; private research in 172–3 Smilor, R.W. 158, 181 Smith, S.J. 40 social networks: reference groups in 42; and spatial structure 40 Society for Recording Life Histories in Tsukuba Research and Academic City 55 So¯toku, Akagi 50 spatial structure and social networks 40 Spradley, J.P. 186 Stein, M. 38 Steinhoff, P.G. 33 Stevens, J.E. 103 strategic ambiguity, and implementation of Tsukuba 159, 201 Sudman, S. 186 Sugiyama, Y. 23, 24 Sun, M. 30 superconductivity 20–2, 183;and thin films 23
synergistic creativity in science 43 TAGS (Tsukuba Applied Geoscience Society) 91–4; collaborative research in 92; decline of 93–4; membership 92 Takagi, J. 78, 120, 121–2 Takahashi, Hidechika 120, 122 Takayama, E. 45, 79, 174, 187; choice of Tsukuba 50; planning of Tsukuba 48, 63, 66, 68–9, 72–5, 160; on science cities 54, 56 Takayama Science Plaza (Kansai) 174 Takeo, Fukuda 50 Takeshita, Noboru 189 Tanaka, Kakuei 50, 66, 72–3, 189, 193 TARA (Tsukuba Advanced Research Alliance) 10 Tatsuno, S. 33, 51, 181 Taubes, G. 11 Tazaki, Akira 107 technology transfer 181–2 Teune, H. 193 Thatcher, Margaret 57 Thermo-fluid Dynamics Seminar 91, 100–1, 193–4; membership 101 Tokumaru, K. 9, 10, 30, 85 Tokyo: co-authorship of articles 137–8, 144; crowding in 46–9, 187; first authorship, analysis of 153 Tokyo, University of 8, 48, 69; and planning of Tsukuba 73–4 Tokyo Education University 48–9, 160, 187 Tokyo Institute of Technology 8, 74 Tokyo Kasei University 187 Tokyo Rika University 187 Tomonaga, Shinichiro¯ 77 Toshiwo, Doko¯ 57 Totumaru, K. 10 Transport, Ministry of 92, 177; in collaboration networks 149 Traweek, Sharon 4, 36, 77 Tropical Agriculture Research Centre 131–2
Index Tsuchida, A. 65, 69, 74–6 Tsuchiura 96 Tsuge, Shunichi 100, 193 Tsukuba, University of 8–10, 9, 13; central role of 169, 201–2; coauthorship, network analysis of 129–32, 143, 145–7, 165, 169; first authorship, analysis of 153; as government initiative 117; private research at 85; publications and productivity 150; and study groups 94, 101; and Tsukuba Network 120 Tsukuba Center for Institutes 69, 76, 193; and communication in research 107, 110, 119, 195; as government initiative 111–12; and Tsukuba Network 121 Tsukuba Center Inc. 112–17, 119, 195 Tsukuba Chemical Science Club 91, 94–5, 97–100; collaborative research in 98; membership 98 Tsukuba Institute Coordinating Committee 121, 195 Tsukuba Life Science Center 130–2, 150 Tsukuba Network 120–2, 195–6 Tsukuba Planning Department 55, 59 Tsukuba Research Consortium 13, 17, 58, 82, 110; communication at 106–11, 194–5; forum topics 108 Tsukuba Science City: access to knowledge in 170–1; choice of site 1, 4, 50–3, 187; comparisons with Kansai Science City 173–7; description of 4–17; development of 157–63; Expo (1985) 56–8, 188; failures of 2; future of 177– 80; international comparisons 6, 182–3; key questions 44; location and layout of 5, 7, 8;
225
organization and key personnel 69; privatization of 172–3; role of in Japan 78–81; transport access to 5–6, 177–9, 202; see also collaboration networks; concept of science cities; implementation; lessons learned; planning; research communications Tsukuba Space Center 130–1, 151 Tucker, S. 63 Uchida, Cho¯zo¯ 73 Ueda, H. 29, 193, 194 Umezawa, Kuniomi 76 uncoordinated cooperation, in science city implementation 158, 201 universities: in Kansai 173; in Tokyo 47–8; Tsukuba see Tsukuba, University of University of Library and Information Science 130–2, 151 Van de Ven, A.H. 63 von Hippel, E. 41 Wakabayashi, T. 3, 47, 55, 59, 64–5, 69 Wakasa, K. 193 Wakasugi, R. 67 Waseda University 8, 190 Weber, M. 39 Westney, D.E. 32 Whittington, D. 158 Williams, F. 158 Witten, M.G. 201 Yabe, Akira 95, 193 Yajima, K. 3, 69, 75, 187, 189, 190 Yamagiwa, K. 25 Yin, R.K. 181, 192 Yoshitake, T. 69, 74, 75, 188, 190