Chapter Seven
Conclusion
Traditional Industry and Indigenous Capitalism in Nineteenth-Century Japan

"Herring. . . are my life," explained Saga Kichijiro , a seventy-nine-year-old fisher from Rumoi, when asked by a reporter in January 1988 why he persisted in setting his gill nets every spring, despite the fact that there had not been a substantial herring run in the area since 1954.^[1] Saga's patience apparently paid off, as he harvested small but growing catches of herring in 1985, 1986, and 1987. Nevertheless, there is no doubt that his dreams of seeing the good old days of the fishery return will never be fulfilled. For one thing, ecological damage, once done, is hard to repair.^[2] The herring catch in Rumoi in 1987 was an "excellent" seven hundred tons, but that is minuscule when compared with even the worst years of the late nineteenth and early twentieth centuries.^[3] Nor would there be much point in processing hundreds of thousands of tons of herring meal anyway, now that chemical fertilizers are so readily available. Even in the unlikely event the herring did return and there was a market for them, the labor-intensive production of the old days would no longer be practical, necessary, or desirable.

Still, Saga is not the only one nostalgic for the heyday of the fishery. Every spring, newspapers in Hokkaido receive letters from old-timers misty-eyed about the vibrant and prosperous past. A favorite image is the tantalizing aroma of fresh-caught herring cooking on an open fire—a far cry from the putrid stench that hit Thomas Blakiston's nostrils when he arrived in Hakodate in 1861.^[4] (In fact, most fishers had more exotic tastes: the robust flavor of makuri —herring that had been held in a storage shed until it had begun to ferment—was always a crowd pleaser.)^[5] The demise

of the herring fishery was a tragedy for fishers, environmentalists, and seafood lovers alike. Analyzing its demise does, however, allow us to draw conclusions about what occurred in the past without having to risk speculation on the future.

My goal in this study has been to clarify the origins and nature of the capitalist transformation of the Hokkaido fishery. I have been concerned in particular with the indigenous roots of capitalism in the fishery and with the immiserating impact of capitalist social and economic relations on small fishers. In both cases the connection between social relations within the fishery and the state structure surrounding it has been critical to my argument. I will conclude by briefly placing the Hokkaido fishery within the broader context of early modern and modern Japanese history, with particular reference to these issues.

Between about 1830 and 1860 the herring fishery underwent the beginnings of a transformation to capitalism. The impetus for structural change came from within the fishery itself, but all developments were shaped by the political and institutional environment of the state. Matsumae, while hardly a typical domain, was nonetheless integrated both politically and ideologically into the Tokugawa state. Its institutions, having originated in the Ainu trade, had built into them a tolerance—even a need—for economic growth. Indeed, the unique features of Matsumae's origins and later development make it an extreme example of a domain dependent upon commerce for its prosperity and stability. In this respect it was rivaled only by Tsushima, which functioned as an intermediary in trade and diplomatic relations between Japan and Korea.^[6]

Matsumae was an integral part of the Tokugawa economy. Its development is best seen within the broader context of Tokugawa economic history, for without the growth of commercial agriculture to feed demand for fertilizer, the Hokkaido fishery would never have been a viable enterprise in the first place. Commercial agriculture was itself a product of the integrating forces of the bakuhan state, for it was in response to state needs that the transportation routes and markets necessary for cash cropping developed. The same was true for the fishery: the shipping routes that carried herring meal and other commodities between Hokkaido and Osaka (and many other ports) were charted in the 1670s by Kawamura Zuiken (1618-99), who had been ordered by the bakufu to find a way to transport tax grain from Sakata to Edo.^[7]

Matsumae's dependence on commerce differed only in degree—not substance—from that of the bakufu and other domains. The Tokugawa state, despite its ostensible relegation of merchants to a lowly position,

relied on them to get goods to market in order to generate cash for samurai stipends, expenses arising from alternate attendance duties, and other feudal obligations. The bakufu and all domains thus had to reach some sort of accommodation with the commercial class. The adaptation of domain institutions in Matsumae to commercial growth—the way, for example, trading posts evolved into contract fisheries—was consistent with practice elsewhere. Indeed, domains throughout the country made extensive use of merchant contractors (ukeoinin ) to regulate the production and distribution of essential commodities.^[8] In the bakufu, Tanuma Okitsugu was aggressive in his attempts to tailor policies to economic realities rather than simply to force the economy to conform to preconceived ideal forms.^[9] Unlike commerce in other domains, commerce in Matsumae was not linked to an agricultural economy, so the effects of commercialization on domain institutions were not obscured by the fiction that subsistence agriculture was the only legitimate economic activity. This difference made Matsumae an ideal environment for proto-industrial development. The distinctiveness of Matsumae, then, lay not in the simple fact of commercialization or the presence of merchant capital but rather in the way commercial and political institutions interacted.

The emphasis on rice cultivation in other jurisdictions inhibited the sort of broad regional specialization found in Europe and thus precluded the demographic transformation necessary to fuel full factory industrialization. Under the kokudaka system a daimyo's place in the institutional hierarchy of the Tokugawa state was measured in terms of the putative agricultural productivity of his domain. Although it very quickly ceased to reflect actual economic conditions, the kokudaka system retained its institutional importance throughout the Tokugawa period. Even as many domains came to tolerate and even actively foster a wide variety of economic activities, peasants were expected to grow grain—preferably rice—unless there was some compelling reason for them not to do so. When officials pressed peasants to produce as much rice as possible—even where climatic or technological conditions made rice cultivation impractical—they were responding to the position of rice as a measure not only of wealth but also of status in the feudal polity.

The kokudaka system thus was an institutional hurdle to economic development, although not an insurmountable one. For daimyo and their officials the critical distinction was not between subsistence and commercial agriculture, or even between agriculture and industry, but rather between rice and nonrice production.^[10] Whether proto-industrialization occurred—or, more precisely, whether other commodities could supple-

ment or even replace rice in the domain economy—became largely a question of a domain's attitude toward nonrice production.^[11]

In Hokkaido, rice cultivation was impossible, so the herring fishery became a proxy for agriculture. In a sense, herring was "rice" in Matsumae: its economy revolved around the fishery in a way that satisfied the requirements of the feudal polity, while opening the door for considerable proto-industrial and eventually even capitalist development. In places where rice cultivation was more feasible, the authorities could adopt a narrower definition of "rice," with the result that support for industry—and even commercial agriculture—was not readily forthcoming. For instance, the southern Kanto was the locus of a number of industries, most notably soy-sauce brewing and sardine-fertilizer processing, that drew labor from the peasant population. However, local authorities were either unwilling or unable to appreciate the value of these industries. Instead of seeing their potential benefits, all they saw was the decline of agriculture. Rather than profiting from industrial growth, they vainly tried to get villagers to go back to the fields.^[12]

The inability of the Tokugawa state to take full advantage of proto-industrial development reflects the rigidity of a feudal polity. Economic institutions may come under pressure and begin to change, but they cannot complete the process of transformation so long as political impediments remain.^[13] In Japan proto-industrialization had its ultimate origins in the political integration of the early Tokugawa period and the subsequent development of transportation routes and markets to deliver tax grain and other commodities to a burgeoning urban population. Later, however, the sort of structural change prompted—indeed, required—by proto-industrialization was impeded by the institutional structure of the Tokugawa polity. Once the feudal polity was eliminated by the reforms following the Meiji Restoration, economic change proceeded at a rapid pace, so that Japan was a genuinely capitalist economy by the beginning of the twentieth century. One casualty of this transformation was rice, which lost its ideological and institutional place of honor in the Japanese political economy with the enactment of the land-tax reform of 1873.

In Matsumae the accommodation of the polity to commerce opened the door to proto-industrial development and eventually to capitalism. In contrast, the Nanbu domain—the archetypical "backward" region^[14] —offers an example of proto-industrialization without capitalism. Nanbu can be seen as a microcosm of the entire country, with the central Kitakami river valley representing the "advanced" agricultural regions and the mountains and coast the "backward" centers of proto-industrial

development. Agricultural, but not industrial, production in the domain was centered on the Kitakami valley, which included the castle town of Morioka. Proto-industry—most notably large-scale commercial fishing and fish processing (not herring, alas!) and ironworking—was found along the Pacific coast and in the mountains separating the coast from the Kitakami valley.

Interestingly, the most highly industrialized parts of the domain saw the most serious occurrences of unrest among Nanbu's notoriously contentious peasants; the best known instances of conflict (the Sanhei rebellions of 1847 and 1853) involved disputes over domain commercial and industrial policies, and were directed in large part against merchants who had purchased samurai status from the domain. The disorder in Nanbu may be attributed to the fact that the impetus for economic growth came from the proto-industrial hinterland, but the feudal institutional structure—located as it was both physically and ideologically in the agricultural core—could not adapt.^[15]

In Nanbu the regional diversity of economic activity encouraged the development of proto-industry, but manufacturing remained under the domination of merchant capital, backed by the sanction of domain monopolies and monopsonies. Incidents like the Sanhei rebellions, in which peasants reacted against tight domain (and hence merchant-capital) control over the economy, may have represented failed attempts to open the door to future capitalist development.

The Nanbu example is presented as a counterpoint to the Hokkaido fishery to accentuate the fact that institutional structure, rather than the activities of merchant capital, is ultimately the key to understanding the causes and outcomes of economic change. It was not the prominence or quiescence of merchant capital that determined the course of proto-industrial or capitalist development in Hokkaido because merchant capital is inherently conservative, choosing domination at the point of exchange rather than at the point of production whenever possible.^[16] This is not to say that the state alone caused capitalism in Hokkaido, merely that the state is a key variable in understanding how changes in social and economic relations evolved. Given the fact that Matsumae did not develop in political or economic isolation from the rest of Tokugawa Japan, nor did it deviate in function from the basic institutional norms of the bakuhan state, we can conclude that the potential for capitalist development was present in the Tokugawa economy as a whole.

Recognizing the potential for capitalism in the Tokugawa economy is not the same, however, as positing a single, inevitable course of develop-

ment. Paradoxically, the same innovations that spurred growth often inhibited more profound change. Edward Pratt argues that the institutional adaptation of the Tokugawa polity to commercial growth after the middle of the eighteenth century fostered the rise of a class of rural elites, the gono (usually rendered "wealthy peasants"). These rural elites built commercial and proto-industrial ventures around marketing networks that had been established under official auspices to serve urban consumer demand.^[17] The actual course of proto-industrialization in a given area was always subject to a variety of constraints, both political and environmental. Gono in areas with high agricultural productivity retained a presence in farming longer than others, for the simple reason that they lacked a strong incentive to gamble on industry or commerce. Although a commitment to agriculture on the part of rural elites did not prevent proto-industrialization, it may have softened the disruptive impact of economic change on social relations within the village community. Conversely, in localities with low levels of food-crop production—typically mountainous regions involved in sericulture—poor peasants often lost their land and became wholly dependent upon industrial employment. Whether productivity was high or low, a farmer's initial decision to enter industry or to remain solely in farming was subject to other factors, such as tax levels (and whether taxes were paid in cash or in kind, and whether they were calculated as a percentage of each year's crop or set at a fixed rate). These same conditions continued to affect the gono's behavior once he had become involved in proto-industrial endeavors.^[18]

A Tokugawa peasant's decision to participate in a proto-industrial enterprise, whether as entrepreneur or as worker, can be understood only in the context of the institutional structure of the bakuhan state. But every time a peasant switched from growing rice to spinning thread, and every time a well-to-do villager invested money made in sericulture in new land, a nail was hammered into the coffin of the system that had given rise to these activities in the first place. By the time all the nails had been hammered in, the participants in the economic system found that their positions had been altered irrevocably.

Still, not every instance of proto-industrialization developed into factory industrialization. Indeed, at times the very vigor of proto-industrial enterprise hindered further development. For instance, Pratt finds that after the Meiji Restoration silk-reeling regions with a history of proto-industrialization were often slow to adopt modern technology, while other areas "enjoyed far greater maneuverability." This difference occurred partly because "the shift to mechanized production did not simply entail

the application of modern technology to traditional crafts." Native technology "maintained a powerful hold over production not only because of its profitability, but because it meshed well with the social structure and needs of the populace at the time."^[19] Merchants in established proto-industrial regions had little immediate motivation to abandon tried-and-true methods, while competitors new to the silk business had every incentive to start at as advanced a level of technology as possible.

Modern technology may have prevailed in the long run, but, as Kären Wigen points out, early-twentieth-century Japanese silk production essentially retained its labor-intensive, handicraft character.^[20] The persistence of proto-industrial production in textile manufacturing, which nevertheless was Japan's first highly mechanized industry, indicates that the economics of structural transformation was not simply a matter of applying the most advanced technology available to a chosen sector. Studies of the Meiji econonay often assume that because Western technology was inherently superior, traditional industries were doomed to disappear quickly. In fact, as Pratt and Wigen demonstrate, traditional methods were often the more "rational" choice in the silk-thread industry. Even after the opening in 1872 of the Tomioka filature—big, modern, expensive, and not at all profitable—native manufacturing techniques endured and were in fact the driving force behind economic development in the area for decades.^[21] The parallel to this development in Hokkaido was the continued expansion of gill-net fishing throughout the nineteenth century, despite the appearance of the pound trap.

Henry Rosovsky once asked, "Is it really necessary to move back farther and farther in time in order to appreciate the significant dimensions of a recent and entirely different past?"^[22] He said no; but now we know that the Tokugawa past was not nearly so "entirely different" and unconnected as it once seemed. Japan was not magically transformed within just a few decades by the introduction of Western technology and institutions after 1868; but neither did every instance of proto-industrialization develop smoothly into full-blown industrial capitalism. Whether there was a direct causal link between given instances of proto-industrial development and capitalism, however, the evolution of capitalist social and economic relations was profoundly affected by earlier conditions.

The connection between proto-industrialization and capitalism is particularly clear in places like Hokkaido, where both the "before" and "after" pictures depict a changing fertilizer-processing industry. A more discerning eye is needed in other regions, where the pressures of proto-industrial development at the local level and institutional change at the

national level sometimes combined to alter regional economies beyond recognition. Wigen's careful study of the Shimoina region of southern Shinano reveals the subtle process of long-term change. During the Tokugawa period Shimoina's paper and transport industries were the center of a thriving proto-industrial complex; Wigen presents evidence that suggests the evolution of capitalist production in at least some sectors of the paper industry. After the Meiji Restoration silk and forestry took over, and Shimoina was integrated into not only the national economy but the world economy as well. The cost, however, was that the region's geopolitical position—even its integrity as a geographical unit—was compromised in the process, and a once vibrant area was reduced to a backwater.^[23] An analysis that had begun or ended in 1868 would have missed the important connections between Shimoina's proto-industrial development and the impact of capitalist development directed by the Meiji state.

After the Meiji Restoration the state implemented policies that affirmed the predominance of the capitalist fishery in Hokkaido. Indeed, the transformation was not really complete until state institutions had fully adjusted to the advent of capitalist production through the formalization of property and fishing rights at the beginning of the twentieth century. It was during this second stage of the emergence of capitalism that small fishers experienced relative immiseration.

Relative immiseration in the context of the fishery meant that small fishers gradually lost their ability to function as independent proprietors. Some experienced a decline in their material living standards as a result of this process, and others did not. But, as I have emphasized throughout this study, the question is not of primary importance here because aggregate measures of economic growth tell us little about fishers' perceptions of their relative position in society and nothing about the way production was organized.

The same is true when we look beyond the fishery to the way the changes of the Meiji period affected ordinary people throughout Japan. The flip side of the systematic recognition of property rights by the state was the dispossession of those whose rights were not recognized. After the land-tax reform such people became tenants whose independence was curtailed by the demands of landlords. Thus, just as the proletarianization of small fishers was a construct of the state's policies toward the fishery, the differentiation of the Honshu peasantry into landlords and tenants was a function of the formalization of property rights in land. In both cases the impetus for structural change came from within the econ-

omy, but the change was not completed until the state had provided an institutional context for the new relations among producers.

When faced with the reality of the capitalist mode of production, people had to either accommodate themselves to it or succumb to proletarianization. Most people who became wage laborers did not, of course, have any real choice at all: they were proletarianized because they no longer had the resources to maintain independent production. Many Japanese, particularly in urban areas, compensated for this loss of independence by embracing the ideology of "imperial democracy," which affirmed the right of citizens to participate in the emergent new order.^[24] Of course, those who had the wherewithal to do so protected themselves in other ways too. Thus fishers moved to places like Karafuto in an attempt to establish themselves as petty capitalists before it Was too late.

The irony is that the state preferred to see the Japanese people resign themselves to capitalism rather than accommodate themselves to it. Thus conflicts like the one over fishing rights in Karafuto were really about the fishers' right to participate in a capitalist economy that did not need any new capitalists. The desire of small fishers in Karafuto to preserve their independence contrasts with the attitude of officials in Toyama prefecture, who in 1891 studied ways to send "surplus" residents of several coastal districts to Hokkaido, where they could be put to work as wage laborers in the herring fishery.^[25]

Today, the functional equivalents of the Hokkaido fishery workers are the rural men who spend each winter far from their homes in northern Japan loading the barrels of pork that ballast postwar Japan's ship of state—working, that is, on dams, bridges, and other big-ticket public works projects. Whatever the discomforts of their lives, they are without question better off than their forbears on the herring-fishing crews. But their prosperity—if indeed they see themselves as prosperous—is an incidental by-product of state-driven economic development rather than a central goal of government policy. In that sense their situation is not at all different from that of the Hokkaido fishers whose entrepreneurial successes and immiserated failures alike were of little immediate consequence to a Meiji state eager for rapid growth whatever the proximate costs.