Spatial diffusion of electric power looms in the Fukui silk Weaving industry, modern Japan

Abstract

Over thirty years have passed since the time when innovation diffusion was first recognized as an interdisciplinary research theme. Geography has exclusively focused on the spatial dimension of diffusion while most other disciplines have paid attention to its time dimension. The research interests of Japanese economic historians (Kiyokawa, 1995) and agricultural economists (Sakiura, 1984) seem, however, to overlap those of geographers. In the field of economic history, in particular, studies on the adoption of the power loom in various textile industries have attempted to explore spatial factors influencing intra- and inter-regional diffusion. Specifically, the focus has been on the reasons for the adoption of power looms in each textile producing area. Case studies concerning one such area, and focusing on the situation prior to the introduction of power looms, show that a strong desire to reduce costs formed the background to their introduction. On the other hand, comparative studies of a number of textile producing areas point out other factors causing regional differences in the introduction and adoption of the power loom. Specifically these include: factory system diffusion; the availability of motive power; union activities incorporating members of the same profession; and the kind of textile product.Such studies, however, make use of data aggregated at the level of the whole production area, so that the reasons for adoption of the power loom by individual enterprises remain to be understood. Taking Fukui City, a representative location for habutae-silk production in Japan, as the study area, this paper investigates the spatial diffusion of electric power looms in silk weaving factories in the late 1900s to early 1910s. For spatial diffusion research in general, such a study on innovation diffusion at the intra-urban level will contribute substatially to examinations of the neighbourhood effect. The first half of this paper describes the circumstances just before electric power loooms began to be introduced into the weaving areas of Fukui City. By this description it is hoped that a deeper insight into the spatial diffusion process will be gained.The latter half of this paper, using logit type trend surface analysis, elucidates the spatial diffusion process of electric power looms among silk weaving factories with ten workers or more. The determinants of this process are examined using the logit model. The results are summarized as follows:The silk weaving industry in Japan, especially habutae-silk, was dependent on overseas markets in the early modern period. Although a major earner of foreign currency, the industry remained highly sensitive to worldwide business fluctuations. The background to the rapid introduction of power looms is that, despite recession following the Russio-Japanese War of 1904, the Japanese Government was afraid that the overseas market share would be lost due to mass production of inferior quality habutae-silk. The Government therefore ordered textile manufacturers to improve product quality, and ensured that all local producers were made aware of the situation. The power loom proved superior to the hand loom in production efficiency, and so was also an effective means to hold back the trend for rising wages during the depression.Power looms were first adopted by influential textile manufacturers at a time when there was panic that Japan would lose its overseas market share and the industry would decline. These people, after all, had first hand knowledge of the overseas markets through visits to their trading partners. After initial adoption there was rapid spread throughout the weaving community, despite the high expense of the machines. In fact conditions were favourable for power loom adoption in the sense that habutae weaving is a relatively straight forward manufacturing process suited to mechanization.