The pattern of social relations in the Tokugawa period was of great significance in shaping the form in which Japan, from the last third of the nineteenth century, was to industrialise. The form of the Tokugawa ruling class, and its relations with the rest of Japanese society, affected the way in which the political revolution of 1868, the necessary prelude to industrialisation, would proceed. Japan’s revolution of 1868 was – as is well discussed in Smith (1960) – an “aristocratic revolution”.
Among the countries of Asia, Japan alone was successful in the nineteenth century in beginning the processes summed up in the term “modernisation”. It is reasonable to try to explain this in terms of the character of the preceding society. Many scholars – among them E.H. Norman, Samir Amin, Barrington Moore Jr., Jon Halliday, Perry Anderson – suggest that Japan’s success is due to the fact that the society of the Tokugawa period was “feudal”; it thus shared important characteristics with the society of medieval Europe, the birthplace of capitalism.
The argument for the “feudal” character of Japan rests on the existence of military warfare among competing warlords before the 17th century, and seemingly on the perceived diffusion of political sovereignty from around 1600 until the 1860s.
Other authors are less disposed to apply the term “feudalism” to Japan – at least after about 1600.
Some issues are agreed. Japan in the Tokugawa period was “pre-capitalist”. In pre-capitalist class societies, some form of “extra-economic coercion” is normally required as a means of appropriating surpluses from the immediate producers, the peasants. The form that this “extra-economic coercion” takes is thus of great significance in determining the character of the exploitation process. As Perry Anderson suggested:
The superstructure of kinship, religion, law, or the state necessarily enters into the constitutive structure of the mode of production in pre-capitalist social formations. They intervene directly in the “internal” nexus of surplus extraction. (Anderson, Lineages, p.403)
Thus pre-capitalist modes of production may be distinguished from each other both by the organisation of the direct producers and their relation above all to the land, and by the social and political structures through which surpluses are extracted from them. All of this is directly in line with Karl Marx’s argument of in Capital Volume III; there – in the course of a discussion of “Labour Rent” – he breaks off to make a general methodological statement on the analysis of modes of production:
The specific economic form, in which unpaid surplus labour is pumped out of direct producers, determines the relationship of rulers and ruled, as it grows directly out of production itself and, in turn, reacts upon it as a determining element. Upon this, however, is founded the entire formation of the economic community which grows up out of the production relations themselves, thereby simultaneously its political form. It is always the direct relationship of the owners of the conditions of production to the direct producers – a relation always naturally corresponding to a definite stage in the development of the methods of labour and thereby its social productivity – which reveals the innermost secret, the hidden basis of the entire social structure, and with it the political form of the relation of sovereignty and dependence, in short, the corresponding specific form of the state. (Karl Marx, Capital III, ch.xlvii, Moscow, 1962 ed., p.772)
The key focus here is not simply on the organisation of the ruling class, but on its relations with the subordinate class of direct producers.
Using criteria of the kind, the application of the term “feudalism” to Tokugawa Japan seems inappropriate. The similarities between Japan and Europe are much less interesting or significant than the differences.
Max Weber also – although his criteria were different from those of Marx – excluded Tokugawa Japan from the “feudal” category, preferring to treat it as a case of “patrimonial bureaucracy” (Weber, Economy and Society, chs.7 & 8). Land in Tokugawa Japan, he pointed out, was not the hereditary and autonomous property of the lord, but was rather a “benefice” held at the will of the ruler. Feudalism in Europe permitted the development, on the basis of the lords’ private property in land, of “Estates” capable of bargaining with the royal power, an outcome notably lacking from Japanese development. There the daimyo, in Weber’s terms, held “prebends” rather than “fiefs”, while the samurai received for their services not land but stipends.
Trimberger (1977) suggests that Japan is best understood, in the Tokugawa era, in terms of a “tributary mode of production” (a term she draws from the work of Samir Amin, who does not himself apply it to Japan). The “tributary mode of production” has some similarities with the “feudal” mode. In both, the peasantry have some control over the means of immediate production, and produce agricultural surpluses. The non-producing ruling class uses its control over the means of judicial and military power to appropriate a large percentage of these surpluses. Appropriation takes place, not through economic exchange, but as an outcome of the political, legal and cultural authority of the rulers. As in the feudal system, the peasantry and the lordly class struggle over the distribution of the surplus.
But, these similarities apart, it is the differences between the “feudal” and the “tributary” modes of production which are more significant.
In feudalism, appropriation of the surplus is undertaken by a lord who has direct and personal control over the serf-tenants who work on his manor or estate. He collects the surpluses in the form of labour-services or “rent”. In the tributary mode, by contrast, a state-class uses its control of the centralised political apparatus to “tax” the peasantry, who work the land in autonomous villager communities. Individual members of the ruling class, in the latter mode, do not participate directly in the process of production as directing agents.
In feudalism, the ruling class lives in the rural areas, on its manors and estates, and exerts individual authority over the peasants. The landlord takes on, in his person, the functions of the state, acting as judicial authority. On the other hand, in the tributary mode of production, the state directs and regulates agricultural production, but only through impersonal and indirect controls over the peasant communities. The rulers live in the towns – in Japan, anyway – and the state takes on, through its regulations, the functions of the landlord.
The difference between Japan and Europe lies, crucially, in the simultaneous conversion of the samurai into state bureaucrats and the peasants into members of autonomous village communities.
A feudal mode of production exists when agricultural surplus – in the form of labour or rent – is personally extracted from peasant producers by landlords who exercise political authority in the absence of a centralised state apparatus. A tributary mode of production exists when agricultural surplus – in the form of labour or tax – is impersonally extracted from peasant producers by state officials in the absence of landlords with personal control over the producers. (Trimberger, 1977, p.88)
Two differences may be further noted between the European feudal system and the Japanese tributary system, both differences being of major significance for the later development of Japanese society. The first concerns the relation between the noble class and the land: the lords are not tied to the land by relations of direct property-interest. Japan thus lacked a conservative landowning class of the type of the Prussian Junkers, who might later oppose moves towards industrialisation. The second concerns the relations between the aristocratic ruling class and the commercial sector. The European states had far less control over the activities of the trading sectors, who developed considerable legal and political autonomy from the nobility. In the Japanese case, however, the state-class maintained far more despotic control over the activities of trade, handicraft production and merchanting than they did over the activities of the peasants. In this sense, Europe and Japan may be contrasted in terms of the degree of despotic control over the peasantry on one hand and the urban classes on the other.
Trimberger (1978) in fact suggests that a more fruitful set of comparisons can be made between Japan and Turkey than between Japan and Western Europe. These comparisons concern not only the forms of society in the pre-capitalist era, but also the forms of their “modernising” political revolutions and their subsequent capitalist development. She writes:
Tokugawa Japan and Ottoman Turkey as patrimonial states were thus structurally distinct from the feudal states of Western Europe. The power of patrimonial aristocracies in Japan and Turkey was based on office, not on land. Even during the decline of feudalism in Europe, the landed nobility “continued to own the bulk of the fundamental means of production in the economy, and to occupy the great majority of positions within the total apparatus of political power.” Because of this congruence of economic and political power, the absolute monarchies in Europe acted to protect and stabilise the social and economic power of the landed nobility. In Europe, “the increase in the political sway of the royal state was accompanied, not by a decrease in the economic security of noble landownership, but by a corresponding increase in the general right of private property. The age in which “absolutist” public authority was imposed was also simultaneously the age in which absolute private property was progressively consolidated. It was this momentous social difference which separated the Bourbon, Hapsburg, Tudor, or Vasa monarchies from any sultanate, empire or shogunate outside Europe. (Trimberger 1978 pp 60-1. The quotations are from Perry Anderson, Lineages, p 429)
Tokugawa Japan was anything but economically stagnant. The introduction of the “tributary” mode of production both directly shaped aspects of social and economic life, and opened space for new economic forces to develop.
One immediate result of the Baku-han system was the speeding up of urban development. The noble caste was shifted from the countryside into castle towns, within whose protection there grew up artisan and merchant quarters to supply and coordinate the needs of the ruling class. By the end of the Tokugawa period, in the mid-nineteenth century, some ten per cent of the population were living in cities and towns with a population of over 10,000. The enforcement of the alternative residence system on the daimyo helped Edo (modern Tokyo) to grow enormously: in the eighteenth century its population may well have been as high as a million, and it was certainly a much more populous city than any in Western Europe in the same period.
The internal peace which the Tokugawa shogunate succeeded in imposing on Japan for a period of two and a half centuries provided a framework within which – inside the “tributary” organisation of the economy – quite rapid processes of commercialisation of economic activities developed. This commercialisation of economic relations began to compete with, and to undermine, the Tokugawa system of stratification and organisation. For through it, sections of the lower classes rose to greater wealth and economic power, while the ruling samurai class found its fortunes undermined.
The processes of commercialisation proceeded apace, affecting both rural and urban relations.
The most complete treatment of the question of agricultural change in this period is that provided by Smith (1959), who summarises the whole process of agricultural change as follows:
At the beginning of the period, farming was generally carried on through the cooperation of families organised into actual or putative kinship groups, who to some extent shared land, labour, animals, tools and even food or housing. By the end of the period, however, such cooperation had largely disappeared. Although it lingered on for a generation or more in isolated places, in the end the individual family nearly everywhere clearly emerged as the centre of production organisation and economic interest. (Smith, 1959, p.ix)
Several factors encouraged the development of smaller, nuclear-family-based farming units. The first was a series of technical changes in farming (for details, see Smith, 1959, chapter 7). New implements, fertilisers and seed strains became available, partly in response to and partly as a result of commercial and urban development. The growth of urban areas promoted demand for new, marketable crops, including silk, cotton, paper, wax, rape seed and indigo. From about 1700, partly through the influence of the small Dutch trading station at Nagasaki, scientific treatises on agriculture began to appear, and these became available to the more active farmers. Techniques of irrigation were improved, the paddy was better used, and the rice crop improved. Commercial fertilisers improved both in availability and usage, as did new threshing devices.
The effect of this kind of technological development was not – as is sometimes the result of technological development – to displace and reduce the demand for labour in the countryside, but the opposite: Japanese agriculture suffered from a labour shortage. Not only that, but the quality of labour needed was rising:
Far from simplifying and making more uniform the multitude of tasks that confronted the labor force (as mechanical innovations presumably would have done), innovations actually increased the demands made on every farm worker. They demanded of him more specialised knowledge and skill, more attention to detail, the exercise of more initiative and judgment. Weeding, seed selection, planting in rows, the use of strong and costly fertilisers, the levelling of fields, the use of water as protection against frost – these and many other operations depended for their effectiveness on the alertness, effort and skill of individual workers. To speak metaphorically, rather than impelling farming forward to a manufacturing stage of production these operations served to strengthen its handicraft character. The increasing emphasis in farming on just such operations put the larger labor force at an ever greater disadvantage in competing with the small one. The large labor force with its hereditary servants and nago, its part-time workers and degrees of family membership, was a loosely organised and relatively heterogeneous group. By contrast the small labor force, which in most cases coincided with the nuclear family, was tight, disciplined, and socially homogeneous. It consequently not only could supervise its members more successfully, but could rely on them to a far greater degree for spontaneous effort since it gave them stronger and more immediate incentives. Under the circumstances, technical innovations brought the opposite of the economies of scale we tend mistakenly to associate with all technological advance; that is, beyond a certain small size, the larger the farming unit the more inefficient it was likely to be. (Smith, 1959, p.105)
As labour became more difficult to obtain, the cost of obtaining it rose. Wages available to those working outside their immediate families rose, and this exercised a pull on the large, heterogeneous labour force of the early period, tending to disintegrate it. The result was a gradual dissolution of the extended family group into its various nuclear family components.
... the extended family was a product of the older mode of cultivation. Just as that mode had required a large family organisation, so the shift on large holdings to tenant cultivation now required a small labor force. With the trend to tenant cultivation in full tide, the raison d’être of the extended family disappeared. (Smith, 1959, p.106)
Smith also notes how this pattern of agricultural development was different from a typical Western pattern:
The long-term significance of the trend to smaller units of farming will be more clearly understood, perhaps, if seen in a larger context. We must remember that it is by no means the only historical pattern of agrarian development. In some Western countries, indeed, the overall trend during the past three or four centuries has been towards larger farming units. This trend has encouraged mechanisation in farming; has increasingly divided the farm population into a few capitalist farmers on the one hand and many wage labourers on the other; has weakened the solidarity of the village community by increasingly differentiating its members as to status, role, and wealth; has largely deprived the family of significance in the organisation of production; and has drastically altered the ratio of urban and rural population by increasing food output while simultaneously reducing the labor requirements of agriculture.
Inadequate as this series of statements patently is to describe the jigsaw of agrarian development in England and some other places, it at least suggests how remote that development was from the experience of Japan, despite the fact that the starting points in the two cases were similar in important respects. For in Japan the trend towards smaller units of farming made mechanisation virtually impossible; it kept the agricultural population a relatively homogeneous class of small peasant farmers despite the presence of landlords and obvious differences of wealth; it preserved the organic unity of the village community despite the growth of a nonfarming population within it; it enhanced rather than diminished the role of the family in farming; and it maintained the farming population at a constant level, and so at a very high ratio to urban population despite industrialisation. (Smith, 1959, pp.106-7)
A third factor, additionally, played a major part in the victory of small farming over large: the nuclear family labour force was uniquely capable of combining its agricultural work with other occupations. The early Tokugawa period was characterised by “a prodigious waste of labour” in farming. “Only for a few brief periods during the year could the labour force be fully employed; the rest of the time it was unemployed in varying degrees” (Smith, 1959, p.129). With the growth of commercialisation, and the demand from urban areas for commodities, this “spare labour” could now be put to profitable use, in cottage industries. And the most efficient way of organising such labour was in nuclear family units.
But, if individual farming offered economic advantages, it did not require large-scale ownership of the land. The unequal distribution of land as between families which characterised the villages of the early Tokugawa period was not overcome, but the pattern of relations obtaining between richer and poorer peasants did change quite markedly. The better-off peasants reduced the amount of land they worked themselves, but expanded the size of their holdings, relying increasingly on tenant-farmers, paying rents, to work their lands. The old kinship system of agricultural organisation was thus gradually displaced by landlord-tenant relations among the peasantry themselves. The nobility, who had withdrawn from direct management of agriculture at the onset of the period, took no part in this. Indeed, since the exchange of land-holdings was strictly illegal, much of this real process of land-concentration in the hands of peasant-landlords took place under legal fictions such as “permanent mortgages”.
The old pattern, under which labour-services had been given by clients and dependants to the wealthier village families, was gradually displaced by one where families paid money-rents to landlords. The process of transformation was slow and uneven, and included such transitional forms as the commutation of part of the money rent by labour services. Smith comments:
When this occurred we can no longer properly speak of labor services. For labor services had now lost their social meaning and become mere substitutes for payments in money or kind. This of course was much more than a change in the character of labor services: it bespoke as well a transformation of the relations of the persons who received and gave them. No longer were these persons bound to one another by powerful mutual obligations rooted in cooperative farming. Cooperation on anything like the old scale had disappeared, and the two parties now stood in a relationship impersonal to such a degree that one would no longer give the other so much as a day’s labor without specific compensation. (Smith, 1959, p.139)
The form in which rents were paid to landlords varied, though Smith suggests that we can see these different forms as a kind of developmental sequence:
A massive governmental survey of tenancy practices originally published in 1921 revealed five distinct types of rent extant at that time. Buried in this mass of data concerning diverse practices relating to tenancy, these variations were looked upon as representing little more than intriguing local idiosyncrasies until Professor Ariga, who on other occasions has also turned his ethnological interests to the aims of history, brilliantly showed that at least the first three types represented successive evolutionary stages. Although instances of the last two were rare at the time of the survey, which perhaps caused Professor Ariga to give them but little attention, they clearly represent further stages just then emerging.
The five types of rent were: (1) kariwak, or sharecropping, under which rent was a percentage of the tenant’s crop; (2) kemmi, under which rent was paid in kind, but instead of being a fixed percentage of the crop, the percentage was fixed annually after an estimate of the coming harvest was made; (3) jomen, which was a stipulated payment in kind rather than a percentage of the harvest and was fixed for a period of about five years at a time instead of annually; (4) daikin, a pseudo-money rent fixed in kind (for a number of years, like jomen), but actually paid in money; (5) a real money rent that was fixed as well as paid in money. Let us now trace the evolution of landlord-tenant relations through these five types of rent, in order to glimpse the shift to individualistic values within a cooperative relationship.
Kariwake, or sharecropping, was the most primitive type of rent; it occurred mainly in relatively backward areas, where crop yields were low and farm income could not be significantly supplemented by earnings from other employments. Under this form of rent the landlord shared the risks of farming with the tenant inasmuch as his share of the crop declined in proportion to any drop in crop yields; but the average rent for this reason was 20 or 30 percent higher (in relation to crop yields) than other types of rent. Kariwake thus not only was a product of impoverished and backward farming but tended to keep the tenant and his farming poor.
Since the landlord shared with the tenant the risks of farming to a greater extent than under other types of rent, he naturally exercised a firmer and more detailed supervision over the tenant’s farming decisions and even his daily routine of work, advising him what crops to sow, what fertilisers to use, when to plant and harvest. He also contributed capital from time to time to improve the tenant’s yields – after all, yields were shared – and he might even contribute labor if that was necessary to avoid a drop in yields. These functions obviously required that the landlord himself be an active farmer. Otherwise he would not have equipment to loan, knowledge to supervise the tenant intelligently, or the skill or labor to cut, transport, and thresh his own share of the crop after it was divided, according to common kariwake practice, as it stood ripening in the fields; and he would not be able to make use of the straw and chaff that were the chief reasons for dividing the crop at this time rather than after the harvest. Virtually every distinctive feature of kariwake – poverty, technical backwardness, dearth of accompanying by-employments, and the active role of the landlord – suggests therefore that it represents the first stage of the evolution of tenancy from cultivation by the holder (tezukuri).
Kemmi marked a considerable advance over kariwake, and its adoption was both a stimulus to better farming and a sign of it. Because it meant lower rents in relation to yields, kemmi strengthened the tenant’s incentive to invest and left him more to invest with. For yet another reason it signified a more advanced agriculture than kariwake. Although admitting of a reduction of rent in especially bad growing years, the reduction was not automatic as in the case of kariwake, nor was it necessarily in proportion to yields; hence kemmi could be practiced only where yields were relatively high and stable, or earnings from by-employments unusually reliable. Since kemmi rents were usually not lowered at all until yields fell more than 10 percent below normal, the landlord shared considerably less in the risks of farming than under kariwake. As a result he had less to say about how the tenant farmed, and he contributed capital less frequently and labor almost never. Since in addition the tenant paid the rent in threshed grain rather than in crops as they stood in the field, it was unnecessary that the landlord be a farmer himself. It is possible that the spread of kemmi to some extent reflected the withdrawal of large holders from cultivation, tending of course to reduce the supervisory role of the landlord. Nevertheless, the landlord’s paternal role toward the tenant did not entirely disappear. Rent was fixed annually, rather than for longer periods, explicitly to permit a closer adjustment to fluctuations in crop yields – for the tenant’s sake as much as for the landlord’s; kemmi in this respect as in others seems to have stood midway between kariwake and jomen, incorporating features of both.
Jomen fixed rent for a period of years; it permitted adjustments in the meantime but only when yields fell drastically – by custom 20 to 30 percent below normal. The entire loss from all but the most severe annual fluctuations therefore was placed on the tenant, so that it was only in areas where agriculture was especially productive that jomen was feasible. At the same time, by guaranteeing to the tenant the whole of any increase in productivity during a period for which the rent was set, jomen provided a powerful incentive to progressive farming. Like kemmi, jomen represented a restriction of the role of the landlord in farming since he had less reason than ever to contribute to the tenant’s efforts and less grounds for interfering with them. Indeed, jomen was compatible with a complete withdrawal of the landlord from farming, even eliminating the need present under kemmi of inspecting the tenant’s crop annually in order to set rent.
Some responsibility remained to the landlord for the tenant’s welfare, however. The landlord was still expected, when a harvest was exceptionally bad, to make an adjustment in rent – to revert temporarily, as it were, to the practice of kemmi. Admittedly his concern for the tenant was self-interested: in the long run his losses would be smaller if he made concessions than if he forced his tenant into failure by stubbornly demanding rent in full. This is another way of saying that the landlord was not purely a rentier, but was in a position that obliged him to contribute, though somewhat passively, to the tenant’s farming success. It is significant, however, that the size of his contribution, in the form of reductions in rent, could not always be based exclusively on crop yields. The tenant’s family and farm economies were so closely meshed that, in deciding whether and how much to reduce rent, the landlord was obliged also to consider the state of the tenant’s health and capital equipment, the condition of his finances, and numerous other factors that lay at the very heart of his family life. Adjustments in rent by the landlord therefore often had something – or appeared to have something – personal about them.
From jomen it was but a short step to the pseudo-money rent, since it was natural under certain circumstances to commute rent levied in kind into a money payment. This helped both landlord and tenant. The landlord need not market the commodities after receiving them and the tenant need not transport them to him, tasks that were especially onerous when the landlord lived in town or in another village. Commutation was possible, however, only if the tenant could market a very much larger share of the harvest than he had been accustomed to, and if the landlord were not interested in rent in its original form of commodities. These two conditions could not be satisfied unless there were an extensively developed market and both parties were deeply involved in it. When one adds to these conditions that payment in money was especially advantageous when the landlord had withdrawn entirely from farming, particularly if he no longer even lived on the land, it is not surprising that with this type of rent landlord-tenant relations became quite businesslike. This was evinced above all in the fact that landlord-tenant relations, hitherto based on oral agreements, tended to be reduced to detailed, written contracts. These documents less left to chance or goodwill, attempting to stipulate rights and obligations as fully and precisely as possible. Despite all this the sense of mutuality did not utterly disappear. So long as rent was levied in kind, the question of its amount could not be completely separated from the question of crop yields, so that in exceptionally bad years the landlord was still bound to make some concession and the tenant was beholden for it.
The true money rent appeared when rents were not only paid but also fixed in money. Because this severed any direct link between rent and crop yields money rents tended to be more rigid than other types. This meant of course that the tenant accepted more, if still not quite all, of the risks of farming. An even more significant change was that the tenant was now fully exposed for the first time to the vicissitudes of the market. So long as rent was levied in kind, the landlord shared the risk of loss from unfavourable price movements by accepting rent in kind (or its monetary equivalent at current prices). But with the true money rent the entire risk of adverse prices was assumed by the tenant, who undertook to pay the landlord a certain sum of money whatever the state of the market. Added to the internal danger to the tenant of crop failure, there were the hazards (and opportunities) of the businessman betting on price. The tenant could venture such grave risks only when an expanding economy gave him reasonable hope of gain and some margin for loss, and further when he had learned to think of himself as an autonomous economic unit whose fate was divorced from others’. It would be foolish to say that there were then no traces left in tenancy of cooperative relations, but whatever traces were left were mere vestiges and clearly did not characterise the institution as a whole.
A word of explanation is perhaps required at this point lest the impression be given that this stage of development had been widely attained even in 1921 when the government survey of tenancy was made. Actually it was to be found only here and there as a faint beginning. It must be remembered that the various types of tenancy, or more strictly speaking rent, did not follow one another in simple sequence anywhere. All five types existed in 1921 and no doubt they had much earlier, too. Nor were the diverse types regionally distinct; on the contrary they were intricately mingled within regions and even to some extent within villages and on individual estates. Evolution therefore consisted of no simultaneous movement over the country from one type of tenancy to another, but of the continuous development, at different rates, of particular instances of tenancy from more personal to less personal forms. This had the effect of continuously altering the proportions of the various types of tenancy, but so slowly and in such complex ways that the changes cannot be followed. All that can be discerned is the general drift. (Smith, 1959, pp.152-6)
The major means by which landowners increased their holdings was by foreclosing on loans for which land had been pledged as security. Peasants, as everywhere, were unwilling to sell land – but “if the peasant were loath to sell land he would borrow money on it to avoid selling, and in the end he often lost his land anyway.” (Smith, 1959, p.158) Thus, as in Western Europe, the local moneylender came to increase his landholding and his power in the village. Not for nothing did the Russian peasants call the similar figure in their villages kulaks (“fists”).
This developing pattern of agriculture produced larger surpluses than the earlier form. But these surpluses did not necessarily flow into the hands of the noble class, the daimyo or samurai. The samurai might hold, in the words of a rhymed couplet, that “Peasants are like sesame seeds; the more you squeeze, the more comes out” (cit Moore); but the problem was the squeezing. Taxes were raised. But in the new structure of the village, where taxes were effectively levied on the landowners, higher taxes were passed on in the form of higher rents. This tended to contribute to the breaking down of solidaristic relations within the village community, sharpening class antagonisms within the ranks of the peasantry, and partially weakening the village as an agency of control of the ruling noble class. There was an increasing wave of peasant revolts and disturbances in the late eighteenth and the nineteenth centuries, aimed at both grasping landlords and tax-collectors. The state was sufficiently alarmed at rural unrest to retreat from its attempts at tax-hikes, with the result that the shogunate and daimyo governments were unable to get hold of a good part of the new surpluses flowing from the individuation and commercialisation of agriculture.
It was not only peasant-landlords whose wealth increased. The merchant class also prospered in the commercialising environment of peaceful Tokugawa Japan. The daimyo collected taxes from the peasants in the form of rice, most of which they could not consume directly, even when it was paid out as stipends to their samurai (who likewise could not consume it directly). What the daimyo and the samurai wanted was cash for spending on all the expenses of civilised and city life. The rice therefore had to be marketed, and the noble class itself was precluded for status reasons from participating in trade. The great han, therefore, contracted with merchants to establish warehouses for their rice and other han products in Osaka and Edo, and to act as their commercial agents. Other merchants took their profit from the purchase and resale of the daimyo rice taxes and the samurai rice stipends. Needing cash for their enforced urban life, the warrior class was forced to develop ties with the merchants, much as they despised them.
Traditionally, the feudal aristocracy had considered money matters unclean and beneath the samurai’s dignity. As in medieval Europe, usurious profit was held in disesteem. The chonin (merchant) style, which included profit taking and the accumulation of capital, was consequently misunderstood and suspect. Thus the Tokugawa merchant remained far more vulnerable to the arbitrary actions of government than his European counterpart. But on the other hand his profits were never so systematically taxed.
While the samurai scorned the merchant’s way of life, in actual fact he became deeply dependent on his services. Restricted to an “inn-like existence” in the castle towns, the samurai was forced to rely upon the chonin to bridge the gap between town and country ...
Increasingly, the samurai class became dependent on financial agents: the shogunate on its currency monopolists and its large financial agents, the hatamoto (Tokugawa vassals below daimyo rank) on money changers who converted their rice stipends to cash, the daimyo on their Osaka or Edo warehouse agents. Conversely, as the commercial houses became involved in nearly every aspect of the fiscal transactions of the administrative class, they were bound to become a major creditor group. (John Whitney Hall, 1970, pp.206, 208)
To the considerable resentment of the samurai, the merchants waxed very fat on the proceeds of their activities.
The growth of commercial capital is revealed in the estimate that by 1761 there were in Japan over two hundred commercial houses each valued at over 200,000 gold ryo. (The ryo was roughly equivalent to a koku of rice.) Thus in total capital worth the great merchants had become the equivalent of many daimyo. (Hall, 1970, p.209)
The samurai class’s rice and other commodities were never sufficient to provide the cash they wanted, for in Edo and the other cities the world of pleasures was always expanding, expensively. The daimyo and the samurai fell, more and more, into debt with the merchant class, whose wealth provided them with the means for the development of a vigorous and sophisticated urban cultural life. At least one of the great capitalist concerns of modern Japan – Mitsui – began its life as a merchant family engaged in draperies and then banking for the great noble houses of the Tokugawa period.
Urban life, and the growth of a commercial network linking the various parts of Japan together – though it was still inhibited by internal tolls and controls erected by the various han authorities as an additional source of taxation – also promoted the growth of handicraft industries. These often took the form of cottage industry in the villages, as we noted earlier. Here and there, factories developed, through they were all quite small and relatively unimportant. In several of the textile trades, workplaces employing up to a hundred people are recorded, and small factories also existed in sake brewing and in wax manufacture. In gold, silver and copper mining, and in the iron industry, even larger enterprises grew up. The expansion of this commodity production provided the basis for a steady accumulation of capital for the traders and money-lenders, and also for the rural landlords emerging from among the peasantry who were diversifying their activities from agriculture into handicraft production.
Thus, already in the later Tokugawa period, labour was taking the partial form of wage-labour within handicrafts industries. The situation in this respect was not unlike that in sixteenth century England.
But the possibilities of capitalist growth were limited. Commercialisation took the form of petty commodity production rather than of full-scale capitalism. Commercialisation, in both urban trade and in rural handicrafts and agriculture, was sufficiently powerful a tendency to provoke a growing crisis in the tributary economy of the nobility, but it did not lift up a new class of urban bourgeois capable of challenging the political status quo. Japan, given its different structure, lacked the crucial dynamic element that appeared in Western European medieval society: the “free town”.
It was not among the chonin (merchants) that the leadership needed to usher in the new world was mainly found. Some of them, notably Mitsui, helped to finance the revolt against the Tokugawa. But, as a class, the chonin owed their wealth to their position as financial and commercial agents of the daimyo and Shogun; the interests of most of them were bound up with the old regime, even though they might sometimes suffer from its exactions. Insulated by the Seclusion Edict from the vitalising impact of foreign trade, they had become as a class conservative in temper and restrictionist in commercial policy. With a few notable exceptions, the great merchants failed to seize, either before or after the Restoration, the new opportunities offered to them by the opening of the country to the commerce of the world and by the development of new industries. (G.C. Allen, 1962, pp.28-9)
While the merchant class saw their wealth increase, they remained in a pattern of symbiotic dependence on the aristocracy. They never developed a general critique of the social relations of their period; their writers and actors satirised existing society and its values, but the merchant class never developed anything approaching a systematic challenge for power. Like the rural landlords, the merchants remained excluded from political power, and seemingly accepting of their exclusion. If in Europe the bourgeoisie played a “revolutionary part” (Marx), their role in Japan was one of passive bystander.
Commercialisation in Japan never led to capitalism. The profits of trade were never accumulated by merchants for investment in production, which remained primitive in technique and scale, and mostly artisanal in form. Japanese capital, restricted in its development by the shogunate’s ban on foreign trade, remained at the level of usury or merchant capital. If anything, the peasant-landlords were a little more dynamic: they were the ones who developed the rural handicrafts and manufactures. But their activities were limited by the commercial blocks erected between the han and by Tokugawa edicts restricting their movements.
Nor did the commercialisation of agriculture lead towards capitalism. Land was rented to tenants, not by improving capitalist landlords as in England, but by peasant landowners who could not obtain the needed labour to work the land themselves. The only surplus population in existence was that section of the samurai class (perhaps half the samurai, or three per cent of the total population) who had no form of activity, and they were banned by status reasons from wage work, from commerce and the like. Given the existence of land, and of rural handicrafts, few peasants were willing to become full wage workers. The late Tokugawa period experienced a severe labour shortage, not a surplus population searching for employment; that condition, and the limit placed on increasing exploitation by peasant revolt, massively inhibited the development of a proletariat.
Thus, the effect of commercialisation in Japan was very different from that in England. There commercialisation, with other developments, broke down the traditional structures of agricultural production much more severely. With the ending of serfdom in England, the peasants gradually lost in large part the protection of their use of and access to the land. In becoming free, they became free to lose the land (see e.g. Brenner, or Saville). But, in Japan, the forces of commercialisation did not break down – even if they did weaken – the pattern of paternalism and communal solidarity in the villages. Japan experienced no “functional equivalent” of the enclosures in England, no mechanism by which the population might be forced off the land to create a labour reservoir and the basis for a free labour market. (Trimberger, 1977)
Commercial capital in Japan was thus “constantly reined in and re-routed towards parasitic dependence on the feudal nobility and its political system” (Anderson, Lineages, p.454 – though he is probably wrong about the term “feudal”).
If commercialisation ran increasingly into contradiction with the tributary system, and no capitalist development of an indigenously inspired variety seemed likely, the only other possible outcome was a slowly maturing crisis in the whole system of social relations. Had Tokugawa Japan remained isolated, it seems quite possible that the whole system constructed in the late sixteenth and early seventeenth centuries would finally have fallen apart. How this might have happened it is obviously difficult to say, but one possible outcome might have been a “re-feudalisation” of the social system, with local warlords – not necessarily the daimyo themselves, but perhaps members of their bureaucratic staffs – attacking the shogunate directly and breaking away from it. In a brief but suggestive discussion based on Weber’s account of “patrimonial bureaucracy”, David Lockwood has suggested that this form of resolution of the contradictions of such social forms may be a likely outcome (David Lockwood, 1964)
Certainly the signs of incipient crisis within Japan were plain to see.
Some of the literature appears to suggest that Japan was running up against some kind of Malthusian growth limit. The evidence takes two forms: the rising level of peasant unrest in the eighteenth and nineteenth centuries, and recurrent famines. The authorities were also campaigning constantly against the practice of infanticide both in the villages and among the lower samurai. Akamatsu (p.63) gives cautious credence to the idea that Tokugawa Japan had a “population ceiling level” of around 27 millions. In both 1782 and 1832, the population was approaching or had surpassed this limit when famine broke out, forcing the population down. Hall (1970, p.203) is more cautious, and seems to point to the spread of landlordism and commercial activity as the cause of the difficulties.
The ruling class’s finances, in any case, were becoming steadily more critical:
... the total seignorial value of the (shogunal) domain around 1800 was slightly below 4,500,000 koku and the Bakufu’s receipts, in principal dues, were slightly above 1,500,000 koku, or almost thirty four per cent. This proportion gradually fell to an average of thirty per cent in the six years between 1836 and 1841. Apart from this, the total seignorial value of the domain was then barely 4,200,000 koku: average annual receipts were around 1,320,000 koku. Moreover, the shogun’s personal gold-stocks, which amounted to 1,717,000 ryo in 1770, fell to 417,000 ryo after the famines and then to 377,500 ryo in 1816.
As for the daimyos’ administrations, they were all suffering financial difficulties and many of them – and not the least important, like Hagi or Kagoshima – made good their deficits by credits from merchants at Osaka or Edo, which they were incapable of paying back. In these circumstances, the seignorial economy was stagnant, if not declining, in the first half of the nineteenth century. (Akamatsu, p.38)
If the financial situation of the Shogun and the various daimyo was becoming more difficult, that of their samurai was much worse. Their stipends were paid at the will of their lord – and according to his capacity to pay. To the degree that the Shogun and the daimyo were suffering restrictions, they passed them on in more than full measure to their servants, the samurai officials, who suffered cuts in their stipends. Many lived in penury, even being forced to abandon the pride of their warrior status to take up commercial or agricultural pursuits. Some entered into (strictly illegal) marriage alliances with merchant families, or “adopted” sons from merchant families in return for financial assistance.
The growing indebtedness of the whole class – which, precisely because it was a patrimonial bureaucracy, needed a steady flow of cash income for the payment of officials’ salaries – constituted a major and persistent problem for the Tokugawa regime. Various not very effective measures were adopted at different times to “reform” the system. These included repeated devaluations of the coinage, the extraction of forced loans from the politically powerless merchants, the raising of land taxes, the imposition of new taxes on commodities, attempts to stem the flow of the population to the cities, placing legal limits on interest charged on loans to members of the samurai class, cancellation of samurai debts by fiat, enforced price-cuts on commodities, and injunctions against the luxurious life. None really worked for long: the symptoms of crisis would rapidly reappear.
One might perhaps summarise the problem by saying that the cash spending of the samurai class was rising faster than their tax revenues from the peasantry. In other words, there was a a crisis of “luxury”, promoted by the very commercialisation of economic relations. “Consumption requirements expanded beyond the inelastic limits of a backward and exploitative agrarian society, as a result of population growth and the insatiable demands of a parasitic ruling class” (W.W. Lockwood, 1968 p.5). But this is rather too static a view. The heart of the problem appears to be the one already suggested: the contradiction between (limited) commercialisation and the tributary mode of production. The form in which the crisis was manifesting itself was a “fiscal crisis”.
The system, based on the absence of lordly private property in land, lacked the degree of flexibility provided by the Chinese:
The Chinese gentry could be financially secure even though his dynasty was bankrupt, because his family usually owned land, and independently of his authorised salary he was able to squeeze a good living out of his local gentry functions. The Japanese warrior of the Tokugawa period, however, had no such freedom, for his entire income, regardless of its form, was usually derived directly from his lord’s tax sources. The lord’s poverty was equally the vassal’s. (Conrad Totman, Politics in the Tokugawa Bakufu, cit. Trimberger 1977)
In the short run, one way out of the crisis might have been simply to reduce the size of the samurai class, by pushing large numbers of the lower samurai into lower strata – a development which was anyway occurring here and there, in a piecemeal fashion, through the operation of economic forces. Yet the tensions created by such a “Thatcherite” solution – the trimming down of the bureaucracy in response to financial crisis – might well, in view of the armed character of the Japanese “civil servants”, have provoked a wave of civil wars and rebellions – leading, as suggested above, to the “re-feudalisation” of Japan.
Already, the signs of danger in that respect were apparent. Numbers of samurai, discontented and burning with many partial grievances, had – either deliberately, or as a result of failure to find a salaried office – become ronin, masterless warriors, bands of aristocratic rowdies involved in various semi-criminal conspiracies in the cities. Such men were to provoke a series of crises in the 1850s and 1860s.
Certainly many of the samurai – the only class permitted to engage in any form of political activity – were involved in numbers of political and cultural movements provoked by the crisis in Japanese society. The growth of market relations and commerce, the emergence and growing wealth of the rich peasant-landlords and merchants, were partially dissolving the inherited rigidities of the status-class structure. Samurai writers bitterly criticised the new developments:
Fear for the future was no doubt one reason that Tokugawa writers, most of them warriors, evinced the keen interest they did in “gono”, as wealthy peasants were called. Fascinated by this new social phenomenon, they rightly regarded it as a threat to their way of life. If the mere sight of great wealth in the hands of peasants did not fill them with misgivings, bitter experience with the power of such wealth drove the lesson home. Large numbers of the military class of every rank depended on loans or contributions (goyokin) from wealthy peasants to remain solvent. What better proof to them of institutional disintegration could there possibly be, unless it was the fawning respect for wealth that they perceived in every class and that they thought was destroying all sense of rank, and belittling the traditional values of frugality, industry, and modesty. “High and low compete in ostentation while government becomes more and more lax!” Fujita Yukoku exclaimed, and then he subsided with a sigh: “This is an age when money buys anything.”
This was a favourite theme of political and economic writers. Love of money was corrupting all classes, none more dangerously than the peasants, who after all were the prime producers of the economy and the base of the social order. But the gono were the worst offenders of all; they lived (it was said) in the style of the city rich, foreclosed the land of their neighbours, welcomed crop failures for the opportunity of buying up land at distress prices, kept concubines, corrupted officials with gifts and bribes, were pretentious of culture, and much more. Buyo Inishi’s indictment of them is more or less typical:
“Now the most lamentable abuse of the present day among the peasants is that those who have become wealthy forget their status and live luxuriously like city aristocrats. Their homes are as different from those of the common folk as day from night or clouds from mud. They build them with the most handsome and wonderful gates, porches, beams, alcoves, ornamental shelves and libraries. Some give money to the Shogun and receive the right to swords and surnames in return.... Others lend money to daimyo and local officials and ... exercise influence in their localities and abuse the common peasants. Still others despise the minor officials and win favour with imperial princes, with members of the royal family who have taken Buddhist orders, and with people versed in court affairs.... Moreover, village officials and others of wealth entrust cultivation to servants; they themselves wear fine clothes and imitate the ceremonial style of warriors on all such occasions as weddings, celebrations., and masses for the dead.”
There was more than a grain of truth in such descriptions; in numerous ways wealthy peasants were taking on the social characteristics of the warrior class. The most important distinction between the warrior and the peasant was that only the warrior had the right to bear a surname and to wear a sword; swords particularly were forbidden to peasants on pain of severest punishment. Exceptions to the general ban on weapons were occasionally made, but until the latter half of the Tokugawa period they were confined to a relatively few village officials whose descent from warriors was both relatively recent and generally acknowledged. By the early nineteenth century, however, both the Shogunate and baronial governments as a financial measure were resorting to the sale of the right to both arms and names – the sale above all to wealthy peasants. Not even a pretense was made the purchasers had aristocratic origins; a man had only to make a sufficient contribution to his lord’s perennially straightened finances to qualify; who he was made little difference.
Nor was this all: warrior offices which carried the reality as well as the appearance of power were coming into the hands of peasants. This seems to have occurred mostly frequently in the Kinai, where petty lords, especially Tokugawa vassals called hatemoto, commonly lived in Edo and entrusted the collection of revenues to local agents called daikan. Partly for reasons of economy, but chiefly to facilitate the collection of revenue and the raising of local loans and contributions, absentee lords often appointed wealthy peasants to this enormously powerful political and traditionally warrior office.
In still other ways wealthy peasants were taking on warrior characteristics. From at least the early Tokugawa period the upper stratum of peasants had been literate, but by the last century of the period the literacy of wealthy peasants in many cases went far beyond its former utilitarian limits. Peasants began to cultivate the fine arts and invade the field of scholarship and speculative thought, all previously the special province of warriors and the city rich. (Smith, 1959, pp.176-7)
Not only did the samurai complain about the “jumped up” peasants and merchants, but they also directed their anger at the their own lords. Smith, in a discussion of “merit” as an ideology among the samurai, notes that, for the samurai, access to office was crucial to their economic and personal security. In practice, all manner of considerations other than “merit” determined possession of office, promotion and the like: kinship links, anxiety among the daimyo to preserve rank-order among the various strata of the samurai, wealth, connections, bribery and others. If it was little honoured in practice, the principle of “merit” as a basis for appointment to office was “warmly regarded” by the samurai at large:
Partly this was because few samurai held office and all wanted it. Office gave power and privilege and was nearly the only means of advancement in income and rank. In peacetime it was also the only way of serving the lord, and the urge to serve was exceedingly strong. Besides, given the taboo on business and the special temperament required for the arts and scholarship, it was nearly the only escape from inactivity and boredom. For a large number of samurai, therefore, the merit principle held out of the only hope of office and office the only hope of self-fulfilment....
Still another reason for the strong appeal of the merit principle was that, after all, things were going badly in the country. After the late seventeenth century, the bakufu and the han were in serious financial difficulties, which was perhaps the least of their troubles. Peasants were abandoning farming for trade, merchants growing so powerful that they could bring all but the greatest lord to heel by withholding credit, samurai losing their martial qualities, and the lower ranks being ground down by frightful poverty. To all these classic symptoms of dynastic decay was added, toward the end of the eighteenth century, the threat of barbarian domination and debauchment. If things were to be put right and the country saved, clearly the bunglers (proved so by the troubled times) must be got rid of and abler men given power. (Smith 1967, p.?)
Throughout the Tokugawa period, there was a steady groundswell of complaint about the barriers to talent, which mixed in the late eighteenth and nineteenth centuries with new issues – especially those concerning Japan’s relations with the West, and the relationship between the Emperor and the shogunate – to form a truly explosive mixture. In crucial senses, the top layers of the ruling class were losing their legitimacy in the eyes of their own cadres. The existing rulers were judged, by their immediate subordinates – samurai who were members of the same class – to be feckless and incompetent, resistant to a renewal of the regime through the opening up of careers to talent.
On no broad political principle, not even perhaps on the Emperor, were the anti-bakufu agitators so nearly united [after 1853, CB]. And for the majority of the samurai, one must remember that the opening of careers to talent was no abstract issue but a matter of the greatest personal urgency – offering hope of escape from poverty, boredom and helplessness. (Smith, 1967, p.90)
Thus the Tokugawa system was being mined from within by many of its own personnel.
Those writers therefore seem correct who emphasise, beside the impact of the West on mid-nineteenth century Japan, the accumulation of tensions and contradictions within Japanese domestic society. Large sections of Japanese society – even if they had no clear vision of the method – were already anxious and prepared for some kind of thorough-going social change.
The country was full of restless spirits, dissatisfied with their conditions and thirsting for activity. There were nobles who wanted independence and foreign trade, to develop the resources of their domains; samurai who wanted opportunities to use their talents, whether as soldiers or as officials; merchants who wanted to break the monopolies of the guilds; scholars who wanted to draw knowledge from new springs; humble peasants and townsmen who wanted just a little freedom from tax and tyranny. Every force but conservatism was pressing from within at the closed doors: so that when a summons came from without they were flung wide open, and all those imprisoned energies were released. (G.B. Sansom, Japan: A Short Cultural History, p.524, cited by W.W. Lockwood 1969, p.9)
Last updated on 28.2.2002