<論説>技術と戦争 : 第二次世界大戦から冷戦期までのアメリカにおけるコンピュータ技術を例に (特集 : 戦争)

Abstract

本稿では、まず、第一次世界大戦以降の戦争と技術の関係について概観し、戦争によって発展する技術とその平時利用の問題と、科学者・技術者の戦争協力体制の問題を論じる視点についての見取り図を示す。それを礎に、今日の情報通信技術の発展の基礎となった、デジタルコンピュータとそのネットワークを基礎づける通信に関する技術の研究開発について、第二次世界大戦から冷戦期にかけてのアメリカの軍事研究費との関係を論じる。デジタルコンピュータに関連する技術は、第二次世界大戦中に各国で戦略的に取り組まれて、大いに発展した。そして第二次世界大戦後には、特に冷戦期のアメリカにおいて、レーダーおよび通信技術と結びついた大規模な研究開発と、宇宙開発に資する研究開発が行われ続けた。こうした軍事的あるいは政治的な技術の発展の加速が、やがて平時の経済発展へとつながっていった過程は、「国力としての技術」の発露であったことを検証する。

This paper provides an overview of the relationship between computer technology and war during World War II and the early Cold War era. In doing so, it develops three themes, namely the development of civilian technologies as a result of military tensions, conflict, and strategy; the articulation of technical goals and objectives resulting from the mobilization and integration of scientific and engineering Knowledge systems; and the broad relationship between federal research expenditures and technological innovation. On this basis, the paper examines the development of digital computers, primarily in the United States. The first case is about the Electronic Numerical Integrator and Computer, ENIAC. As Mitchell Marcus and Atsushi Akera pointed out in their paper "Exploring the Architecture of an Early Machine: The Historical Relevance of the ENIAC Machine Architecture, " ENIAC was not developed as a realization of theory, but a compound of practical choices of relevant technologies optimized for the military need of the time. The second case, the British Bombe (along with various other wartime digital and analog computers), utilized analog devices to implement fast computational procedures also to meet narrow military needs. Chikara Hoshino pointed out in his book, Yomigaeru Turing (Recalling Turing), that the decision to use analog devices was essential to the overall conceptualization of the apparatus. Through these instances, it is clear that wartime efforts to develop computers were not derived from any formal theory of computers or computation, but from the practical need for doing computation in military fields such as exterior ballistics, code-breaking, and hydrodynamic simulations. Nor did this wartime work bring most practitioners to theorize about "digital computers." A significant exception was John von Neumann and his famous description of the fundamental architecture for digital computers. As a mathematician, von Neumann pursued various applied mathematics projects for the military, but began to develop a profound interest in theorizing computing machines through his wartime activities. Indeed, von Neumann was a symbolic figure who transgressed the boundary between pure and applied mathematics, and through no coincidence, the practice and theory of computational machines. It was during the way that the general belief in the distinction between science and technology as discrete spheres of activity became blurred as a whole through the fact that many scientists were mobilized for applied, military projects. This blurring continued into the postwar period as the technological fruits of war were declassified and aggressively promoted for peaceful purposes. This technology transfer of what we would now call "dual use technologies" was an explicit policy of the Office of Scientific Research and Development (OSRD), the science mobilization agency directed by Vannevar Bush. In his Science: The Endless Frontier, published in July 1945, Bush clearly foresaw the need for the broad circulation of knowledge across the military and civilian spheres amidst the postwar expansion in peacetime federal research expenditures. For Bush, "Scientific Manpower" meant both scientists and engineers. His usage of "science" in his visionary report as OSRD Director called for fundamental research in both pure and applied science (e.g. engineering) as a means of systematically augmenting the U.S. national potential for radical technological advances, both military and civilian. After the war, in the 1950s, the knowledge and community of practitioners surrounding digital computers were gradually institutionalized into existing academic fields. During this embryonic phase of computing as a new interdisciplinary field, those involved with digital computers interacted with those from adjoining fields such as communications, control theory, and telecommunications engineering. A prime example of the fruits of this interdisciplinary exchange, and one that again spanned the military and civilian sectors, was the SAGE (Semi-Automatic Ground Environment) air defense system developed through heavy funding by the Air Force. Despite its clearly defined military purpose, the civilian spinoffs of this technology were astounding. In addition to opening up the area of computer networking that provided a foundation for today's Internet, work on the SAGE air defense system contributed extensively to the development of new manufacturing methods and knowledge, which contributed substantially to IBM's subsequent commercial success. Then in the wake of Sputnik, the Defense Department created the Advanced Research Projects Agency (ARPA). While originally conceived of as a way to centralize all space systems R&D efforts, when this function was absorbed by the National Aeronautics and Space Agency (NASA), ARPA redefined its mission around the notion of military command and control. Digital computers emerged as an important cornerstone for this work. ARPA's expenditures in computer research (both hardware and software) represented a significant portion of its overall budget, and this funding in turn transformed computing through the development of time-sharing systems, computer graphics, artificial intelligence, networking, and the development of academic computer science itself. The other federal agency whose work spanned military and civilian interests, and whose expenditures fueled other fundamental advances in digital computer technology, was NASA. In the wake of Sputnik, NASA's investments as well as mission specific requirements led to the miniaturization of computer systems and the development of more specialized digital (and analog) electronic systems. Especially during the early 1970s, the research and procurement bankrolled by NASA contributed significantly to the development of the U.S. semiconductor industry. Protests against the "military-industrial-academic complex" became intense during the same period, leading the military to again redefine its R&D mission around more narrowly defined military interests. Ironically, this curtailed the military's ability to contribute to civilian technologies. But by then, military research had laid a firm foundation for the U.S. digital computers industry and for academic computer science research. In this way, this paper demonstrates how the development of computer technology, born out of the military necessities of the "endless frontier, " produced both the technological foundations and the industrial infrastructure necessary for the continued and phenomenal expansion of computing within the civilian sector.