Kazuo Nimura 'From Trade Unions to the "NAKAMA" in Feudal Society: Organizational Characteristics of Occupational Groups in Japan'

In this study I would like to investigate the characteristics of Japanese occupational groups seen in an international context. I shall take as my starting point the post-war trade union and work back from there to the ironworkers' union [鉄工組合Tekkō Kumiai ] that were the first Japanese labor unions in the modern period, and then finally consider the pre-modern NAKAMA [仲間=brotherhoods or associations]. Historical studies normally proceed in a chronological fashion, but in this case I have taken the liberty of heading in the opposite direction, because just as it is often said that the key to understanding the present lies in the past, so the clues to comprehending the past can also be said to lie in the present. Palaeontologists seek to reconstruct a picture of a palaeontological whole from fragments of rocks. In doing so, as they compare similar rocks, they are also able to study modern organisms which have the same mineral structure. The typological configuration of the organism can be reconstructed from the rock alone, but its surface texture, colours, eating habits and vocal characteristics cannot be understood solely from examination of the rock. This would require reference to an organism A of similar ecotype. In a similar manner, the attempt will be made here to reconstruct occupational social groups of the past. The material evidence for the existence of prehistoric organisms mostly consists of physical remains such as bones and teeth that can be subjected to techniques of chemical analysis, but in the case of historical documentary evidence, the further back one goes, the scantier such evidence becomes. Certainly, vast amounts of documentary data are available for the pre-modern period in Japan, but documents that record the customary practices of occupational groups or the mentality of the members of such groups are very limited in number. History is an empirical science, so it is indispensable to have knowledge of the background to historical documents of the same period, but when issues are being studied, for which the likelihood of available documentation is in any case low, then there are limits to the extent that an understanding can be based on the documents alone. The problem surely calls for an approach that refers to the practices and organisation of the pre-modern and modern occupational groups that understand the actual situation, and to studies of occupational group practices in other countries.

I do not actually intend to make pre-modern occupational group research my main objective; I would rather refer to specialists in pre-modern Japanese history for that. My main interest lies in elucidating the characteristics of modern Japanese labor relations. However, in doing so, it will be necessary to understand the characteristics of occupational groups in the Tokugawa, or Edo Period (1603-1867). The foundations of the western trade union movement were formed by the craft unions, but they cannot be understood without consideration of the traditions of the guilds that existed in mediaeval European towns and cities . In the same way, to get to know the nature of Japanese labor unions, one has to understand the nature of occupational groups in pre-modern Japan. I have occasionally expressed this view before, but it has not received much attention from most scholars of pre-modern Japanese history. On this occasion, therefore, when so many specialists in pre-modern historical studies are present, I have it in mind to take advantage of this rare opportunity and to step boldly outside my own specialist field in what I have to say.

Ⅰ. The Characteristics of Japan's Postwar Labor Unions

1. Why enterprise unions?

One of my principal research themes over the last ten years and more has been to examine the nature of Japan's labor unions and labor relations in a comparative international perspective. Beginning with an historical study of the debate about enterprise unions, I went on to write a number of related papers on the condition of the labor movement in the immediate post-war period and then a comparison of Japanese and Korean labor relations . The first half of my presentation today will cover the same ground, but first I would like to introduce the main points of my argument and I shall end with how I see the special features of pre-modern Japanese occupational groups in historical and international perspective.

I started out in this research by considering why labor unions in Japan took the form of enterprise unions. Western labor unions had mainly been organised on occupational (craft union), industrial sector (industrial union) or general union lines, all of which went beyond the bounds of the individual enterprise yet the great majority of Japanese labor unions are enterprise-based unions [企業別労働組合 kigyōbetsu rōdō kumiai]. I wanted to clarify why Japan was different. As you will know, this was a central theme in Japanese labor history in the 1950s and 1960s. The academic consensus then was that Japan's labor market was divided along enterprise lines and that consequently, labor unions that developed in that context did so as enterprise unions. The leading representative of this view was Ōkōchi Kazuo . There are a number of variations to Ōkōchi's thesis and over time, different elements were emphasised so that the thesis came to present various problems. On this occasion, however, I shall not be dealing with them. For the details, I refer you to my "The Historical Background of Enterprise Unions". I was always doubtful about the 'labor market causation theory', which was almost the conventional wisdom at the time. I had a number of reasons for my doubts but I was sure that the theory was wrong when I realized that it could not explain the actual situation in post-war unions, in which blue collar and white collar workers belonged to the same union. In Ōkōchi Kazuo's extremely concise view, "labor unions are organisations of members of the labor force who sell their labor, so those who have common interests in the same labor market will naturally organise themselves in the same union". However, I consider that to regard all the workers who belong to one enterprise as belonging to one and the same labor market is itself a mistake. Blue collar and white collar workers may indeed work for the same company but their labor markets are clearly different. Moreover, white collar workers themselves belong to different labor markets according to whether they are university or high school graduates. Even if Ōkōchi's view that "those who belong to the same labor market will naturally organise themselves in the same union' were correct, Japanese labor unions ought not only to be enterprise-based but also educational attainment-based. And even if they share the same level of educational achievement, male and female workers face completely different prospects when it comes to promotion and salary increases, so we also ought naturally to have gender-based unions. This of course is not how things are.

At this point I will of course be asked what I think are the origins of the growth of enterprise unions. In fact, my answer is extremely simple. When workers are faced with a problem that needs solving together, it is only natural that they should get together with the workmates they see and work alongside with every day and whom they know well. That is my view. Western unions too originally began from local organisations of workers who knew each other from meeting in the same bars.

What is more, I also think it necessary to ask not only "why did Japanese labor unions not become occupationally-based or industrial sector-based organisations?" but also "why did western labor unions start out as occupationally-based groups that went beyond the single company?" Labor unions as associations that people form based on the unit of those they know well at their place of work or at their company seem very natural, whereas occupationally-based unions seem rather peculiar. Why did westerners not join with their comrades at the same workplace whose names and faces they knew, rather than with those with whom they merely shared the same type of work and whom they had never met? Furthermore, there must have been particular reasons why there was sometimes even strife between those at the same workplace.

2. Why did 'blue- and white-collar mixed unions' develop?

I will come back to this problem again later. Next, I would like to look at another special feature of post-war labor unions, namely, the question of why blue collar and white collar workers belonged to the same union, a rather unusual development in the international labor movement.

The labor movement was, of course, originally a movement of blue collar workers. Both in the West and in Japan until before the Second World War, 'labor movement' meant 'blue collar' workers. The influence of that history is still strong in the West; white collar workers have tended to form their own organisations, and there are very few examples of them belonging to the same unions as blue collar workers. Yet in Japan, blue collar and white collar workers belonging to the same union - the so-called 'blue- and white-collar mixed union' [工職混合組合 kōshokukongō kumiai] - is the norm. This does not mean a combination in which a blue collar group is 'allied' to a white collar group; there are many examples of unions in which the fundamental unit is the enterprise itself, and people with different occupations within the enterprise all belong to such unions. It is an important point, in my view, not to overlook this special characteristic of post-war Japanese labor unions - that blue collar and white collar workers formed a single organisation. Why then were so many post-war Japanese unions of the mixed occupation type? To answer this, I believe the following three questions need to be addressed:
1. why did staff workers who had had no previous links with the labor movement join unions?
2. why did staff workers not choose to form their own union and instead opt to join a mixed occupational union?
3. why did blue collar workers did not oppose the membership of staff workers?

As I answered these questions in detail in some detail in my The Labor Union Movement in the Early Post-War Society, here I shall just restate the concluding points.

First, staff workers joined the labor movement because they too, like the blue collar workers, were suffering as a result of the post-war inflation, food shortages, and the damage caused by the wartime bombing. In such a context, the Occupation authorities' policy of the protection and promotion of labor unions facilitated staff workers' participation in the movement.

The second question - why staff workers did not form their own unions but joined together with shop floor workers - has not been given much consideration up to now, but it is actually a key point when thinking about post-war labor unions. This is because "it is very rare to find really successful attempts at forming a union that includes both manual laborers and white collar workers" in any country. In Britain and elsewhere, the main impulse that led white collar workers to join the labor movement was the wish to prevent the loss of their status relative to that of blue collar workers, so naturally, they formed their own unions. However, in post-war Japan it was mainly staff workers who opted for the 'mixed occupational union'. Having said that, their main motive was not at all the same; broadly speaking, there were two groups of people with completely contrary standpoints. One group were positive participants in the labor movement, people who wanted to see social reforms as a result of Marxist influence or their wartime experiences. They opted for the mixed occupational union because they saw the labor union as an organisation that functioned as one wing of a movement for social reform and revolution and regarded themselves as members of the working class. On the other hand, there were those staff workers who were wary of forming labor unions; for them, the emergence of the communist movement represented a danger. They felt that labor unions must be encouraged to take a moderate position. They too chose the all-in-one union inclusive of all employees.

Thirdly, Japanese blue collar workers accepted the idea of forming unions together with white collar staff because their own tradition of 'laborism' was weakening. Originally, in the background to the development of the labor movement in the West was the fact that the English working class typically had a 'them and us' mentality. Office workers, and even the shop floor boss, were naturally regarded along with the management as 'them', and for shop floor workers, belonging to the same organisation as 'them' could hardly be imagined. Such an attitude was far from the minds of Japanese workers. I shall return to this point later.

3. The Removal of Status Differences on the Shop Floor

One of the important causes advanced by post-war labor unions was the movement for 'managerial democratization' [経営民主化 keiei minshuka]. What was this 'managerial democracy' movement aiming at and what did it achieve? Understandably, different emphases were placed on objectives according to who was representing the movement. For blue collar workers and lower-ranking staff workers, 'managerial democratization' meant, above all, the removal of status differences within the company. On the other hand, for managerial staff or for university-graduates, who had been assured of an eventual managerial position, 'managerial democratization' meant employees' participation in management. Behind the drive to remove status differences was the fact that office workers were organized into clearly defined levels based on educational achievement and the fact that there were large differences in the way workers at those levels were treated. The situation was approximately as follows: employees at the top level of educational achievement were seishain [正社員'regular' employees], all of them university, high school, or technical college graduates, and they were employed at head office or sent to company offices throughout the country. They were soon promoted to middle management and were assured of eventual senior management positions. Middle high school graduates were called junshain[準社員] or junin[准員](junior or associate employees) and worked in clerical or technical departments; most were only employed in factories, and their promotion prospects were basically restricted to middle management positions. Below them were those who had only completed compulsory education - office boys and the like, who were termed yōin[傭員] and were treated like blue collar workers. On the other hand, workers employed for manual labor [職工 shokkō] were limited to those whose educational attainment was below the level of compulsory education. There were certainly examples of junshain and seishain being drawn from the ranks of shop floor workers, but in general, there was a gap between blue and white collar workers that was by no means easy to cross.

Between the treatment of regular employees and others, or between office staff and shop floor workers, there were large differences in pay, bonuses, promotion, and access to company accommodation. Especially great differences were evident in pay scales. Regular employees were paid annually and monthly; however hard they worked, they received no allowance for extra hours put in, but if they failed to turn up for work, their pay was not immediately cut. Their bonuses depended on the company's results. If business was good, they received an annual bonus several times the amount of their monthly salary. There are still remarkable stories told of people who built themselves a house with such bonuses. Junior employees were paid daily or monthly. Absences for them meant less pay, and Sundays and public holidays were also unpaid. Shop floor workers were either paid piece rates, or else by the hour, or a daily wage. Of course, absenteeism and lateness among workers had to be controlled, but beyond that, they were also required to enter and leave the workplace separately from other workers and were marked out by the colour of their hats. At many sites, workers leaving work for home were subject to body searches at the gate to check if they were carrying materials or company products. It was often said that such differences in the treatment of shop floor and office workers were rooted in the pre-modern nature of Japanese society, but the actual situation is not so simple. I still lack sufficient research data here, but it seems that along with the factory system, technical workers brought back a number of habits from the West. In fact, in Britain, the country many regarded as the model of a modern society, there was a similar pattern of discrimination. It was evident in various ways: hourly and monthly pay differentials between blue and white collar workers, differing lengths of the working day, different times of clocking on and off and the number of days off, different pension arrangements and allowances, and different lengths of notice on severance of employment e.g. a week or a month. Having to use different entry and exit points, car parks, canteens and even toilets - all these forms of discrimination existed until about 1965. British blue collar workers accepted such treatment as a matter of course. Post-war Japanese blue collar workers, however, regarded such divergent treatment as "feudal discrimination" and called for its removal. Many demands were made: for the removal of discriminatory nomenclatures such as rōmusha[労務者 laborer] and shokkō[職工 worker]; for the equalization of standards of pay, bonuses, allowances, working hours, holidays, days off, retirement; for the right to use the same gates for entrance and exit; for the ending of body checks, and for the extension to shop floor workers of access to various facilities used by office staff. Behind this movement for the removal of discrimination at work was a long history of resentment on the part of Japanese workers. The demand for the removal of such discrimination had erupted in various ways in pre-war labor disputes. This may seem at first glance unrelated to the issue of occupational groups, but the fact that post-war Japanese labor unions took on the internationally rare form of the 'blue- and white-collar mixed union' is actually profoundly linked to this very topic, so I would like to go into it to a certain extent.

Ⅱ. The Japanese Labor Movement before World War Ⅱ

1. Characteristics of Labor Disputes

A feature of Japanese labor disputes that has been noted since before the war has been their highly emotional nature; anger over prejudice at work has been a particularly prevalent element. There have been many cases, especially in major or long-running disputes, of blue-collar workers' anger erupting in a very violent form. Much of this anger has been caused by the perception of discrimination at work. Problems included engineers and foremen treating workers with arrogance and contempt, or acting illegally by demanding bribes as of right when fixing wages, and owners and managers behaving in a manner lacking in good faith or human decency.
Japanese workers were naturally not uninterested in the issue of working conditions, but they felt that wage levels did not only have an economic significance but also reflected 'status' within the enterprise. On the issue of absolute wage levels, they were very sensitive about their wage levels relative to other workers. This was shown very plainly in the Japan Railways engine drivers' dispute of 1898. In that dispute, in addition to demanding higher wages, they called for changes in job titles : from kikankata[機関方] to kikanshi[機関士] (engine driver) [translater's note: whereas the character '方=kata' simply means 'a person in charge', that of '士=shi' originally meant 'samurai' ], from kafu[火夫] (fireman or stoker) to kikanjoshu[機関助手] (lit. engine assistant), and from sōjifu[掃除夫] (lit. 'cleaning person') to kuriinaa (cleaner). kikankata[機関方] was mostly likely associated with 'wagon driver' or 'pack-horse driver' 'umakata[馬方]' [and thus regarded by engine drivers as an inferior designation for their job - trans.]. The engine drivers also demanded that they be given status equivalent to that of office staff. They regarded themselves not as mere workers but as engineers operating expensive machinery, who therefore deserved to be treated like members of staff. They also resented the fact that station staff could be promoted to stationmasters and be in a position to give orders to train drivers.

In the movement that culminated in the Ashio riot of 1907, among the various demands of the Shiseikai labor confederation, the miners made much of the demand to "sell Japanese rice to workers!" At the time, the Ashio mine shops sold all kinds of items to staff and workers, but home-grown, short grain Japanese rice was known as 'staff rice' [役員米 yakuinmai] and could only be sold to staff members, whereas ordinary workers had to buy long grain, imported rice. If they wanted to eat white rice, they were forced to buy it at higher prices from rice merchants in town. Needless to say, rice was not the only problem of discrimination between staff workers and miners. There were issues about all aspects of daily necessities in food, clothing, and shelter, including the location of company accommodation, the number of rooms, and access to shared or private toilet facilities. Miners were especially aggrieved by the unfairness of wage levels that were dependent on whether or not bribes were given to members of staff. The demand that white rice should also be sold to workers was characteristic of the anger felt over discrimination in the treatment of blue collar vis-a-vis white collar workers. Such anger was a widespread undercurrent throughout the Japanese labor movement.

Workers demanded an end to discrimination not only at work but also in society at large. In the readers letters' columns of pre-war labor union organs, phrases such as "workers and female workers are human beings too!" can often be found, as well as the repeated claim that "the goal of the labor movement is not higher wages; it is human freedom". The labor movement after World War I presented such workers' feelings as the ' demand for recognition of a person's humanity'. Not a few researchers have based their usage of the key concept jinkakushugi ('moral personality ism') on this historical background. Nevertheless, however often labor activists of the period may have used this term, I feel somewhat uneasy about the use of the western concept jinkaku, which signifies "a subject who acts autonomously as a reasoning being", because I feel that the use of this as a key concept leads to the emotional element within Japanese workers' demand for equality being lost sight of. It is not particularly well-formed as a Japanese expression, but for the time being, I would prefer to say "the demand that workers, managers, and office staff be commonly recognised as human beings".

Many Japanese managers and even some researchers, with an eye to making an impact abroad, have argued that "'respect for human rights' is itself characteristic of 'Japanese-style management'". The word 'Japanese-style' here implies that management in Japan has always had this respect for the human being, and in fact, a number of researchers have claimed precisely this. However, this view is flatly contradicted by the facts. It was only well after World War II that 'respect for the human being' began to be extolled. Post-war labor unions demanded 'democratization of management' and 'removal of status discrimination'; it ought not to be forgotten that their achievements in this direction were considerable.

2. Background to Anger = The Pretence and Reality of the Removal of the Status System

Blue collar workers' great resentment against discrimination has, as might be expected, to do with the particular nature of the reforms of the Meiji Restoration (post-1868). The so-called 'equality of the four classes' [四民平等 shimin byōdō] was no mere pretence; the freedom to choose one's work and the freedom to move where one wished did become realities. For lower class samurai and commoners who, under the former feudal clan system, had been restricted by the status into which they were born, this reform was a very positive one. However, in post-Restoration society there was no equality in the literal sense of the word; occupationally-based differences in social status continued to exist. People chose to work in factories and mines because they were forced to do so by the difficult circumstances of their lives, and these jobs were especially looked down on by society as a whole. Factory workers, disliking the term shokkō (factory hand, operative), frequently referred to themselves as shokunin (artisan), reflecting the fact that factory work was held to be lower in status than the work of the traditional artisan.

Factory workers were not only discriminated against in society at large but also within management structures. The organisation of production in an enterprise itself requires a division of labor and a hierarchy of duties. People who had lived within a status-based social order (e.g. pre-Meiji Restoration) naturally tended to regard this hierarchy of duties as being related to status. Factory workers of course resented the fact that they were seen as members of 'the lower classes' and were ranked lowest within the enterprise. Most Japanese workers did not think like English workers, who considered it natural that a worker's son should himself be a worker and who were proud to be members of the working class. To put it plainly, Japanese workers were people who wanted, if possible, to stop being workers. If this was not possible for oneself, then at least one could get an education for one's child so that he could stop being a worker. One episode makes this crystal clear. It comes from the account of Masumoto Uhei, chief engineer at the Toba shipbuilding yard, which was involved in a fierce labor union campaign. He was elected Japan's first labor representative to the ILO. I would to quote something from his account that is based on his own experience. "No-one would wish for the life of a factory hand. Society does not think of factory hands as being the same as other human beings; rather, they are thought of as being like oxen or horses. Although that cannot be helped, until quite recently, the feeling was common amongst Japanese workers that even if the parents fall to begging, they do not want their children to become factory hands.

When I called for the education of factory hands at the school associated with the Mitsubishi shipyard in Nagasaki, the feeling at the shipyard was 'educating factory hands? What's this? One generation of work at a factory is enough! There's surely no more stupid notion about than sending your children to school to learn how to be factory hands.' This was before the Russo-Japanese War [1904-05], but the feeling then in the shipyard about a school for educating factory hands was actually just like that" .

Japanese workers did not necessarily oppose discrimination in itself. They felt that it could not be denied when those of superior ability were given higher status, but there was great frustration when those of inferior ability were given the same higher status. To put it another way, they felt that those with higher status must be people of superior personal quality and abilty. It was not that everyone really believed that seriously, but such was the outward claim. In a backhanded application of this claim, inept superiors were frequently criticized. Such an attitude had already existed in the Tokugawa era (1603-1867), but after the Meiji Restoration of 1868, meritocratic values spread rapidly along with the diffusion of the school system, and this led to outbursts of indignation against unjust discrimination. British workers' 'pride in belonging to the working class' probably also included an element of refracted anger against discrimination, but Japanese blue collar workers would have found such an attitude difficult to understand.

The main standard in the determination of status in Japanese companies was academic achievement. Elementary school graduates became blue collar workers, middle school graduates became lower-ranking staff members, and university graduates became regular employees. Academic achievement of course was not unrelated to ability, but the problem was that it tended to be influenced more by parental income than by individual talent. There were exceptions such as the Gakushuin school (attended by members of the imperial family and the aristocracy), but compulsory education in Japan drew in students irrespective of their parents' social status. When the children of landowners and the children of laborers studied side by side, what counted was results and ability. Yet once the children were through elementary school, differences began to tell in terms of their parents' economic standing and the number of their siblings. If the family were poor, the children would soon have to go out to work, and those with only an elementary school education would only end up as blue collar workers, no matter how able they were. There are many examples in the autobiographies of labor leaders of bitter memories of not being able to go on to higher education despite having had an excellent record at elementary school. People who had experienced such blocks in their education tended to resent discrimination in the workplace all the more. There were many such people among labor union activists in the post-war period until the first half of the 1960s. Not unrelated to this is the fact that in the Sōhyō labor movement, conspicuous among the leaders of the public sector workers were those who had been first class pupils up to middle school level. Furthermore, a similar logic can be observed in the fact that throughout the pre-war, wartime and post-war periods, most of those who had leading roles within the labor movement were not from the lowest ranks but rather, were skilled workers who were relatively highly paid. There is no stronger motivating impulse for a social movement than a resentment of discrimination shared by a number of social groups. We should not lose sight of the fact that the labor movement is also a social movement. Naturally, this fact does not apply only to Japan.

3. Iron Workers' Union and Craft Unions

Thus far I have focused on the characteristics of labor disputes, and next, I would like to turn to the pre-war labor union. There are considerable differences between pre-war unions in the early period of the labor movement and those of the World War I years, but here I shall focus on the early period. Japan's first modern labor union - the Iron Workers' Union (tekkō kumiai) - began, as you know, in 1897. This union, known as the tekkō, was for a time thought of as an occupational union (shokunōbetsu rōdō kumiai), or craft union, because it was a horizontal organization that included workers from a number of different companies. For example, both Ōkōchi Kazuo and Sumiya Mikio both argued that at that time there was still a horizontal labor market in Japan, and that consequently, occupational unions that transcended the company unit were able to develop . However, later research by Hyōdō Tsutomu and Ikeda Makoto clearly showed that the internal arrangements of the Iron Workers' Union did not justify it being called a craft union, and today there are unlikely to be any researchers who would designate the Iron Workers' Union as an occupational union. Nevertheless, Hyōdō and Ikeda differed in their interpretation of the reasons why the Iron Workers' Union was unable to become an occupational union. I shall discuss this later; first, I would like to look at the ways in which the Iron Workers' Union differed from the English craft union.

The main difference between the two was the fact that the qualification for becoming a member of the steelworker's union was extremely loose. Article 6 of the Iron Workers' Union rules stated that:

"In accordance with the form stipulated in other paragraphs, machinists, blacksmiths, canners, foundry workers, model makers, coppersmiths, steel shipbuilders, electrical workers, engine drivers and stokers at steel works who have applied for membership to the main office or to a branch office and have been accepted by the executive council shall be recognized as members of this union."

In other words, this union did not target workers in a particular type of job, but sought to organise all workers who had jobs in steelyards. The union members worked in key industrial organisations - factories, railways and shipyards such as Japan Railways (railyards in various regions), Shimbashi Railway, the Ōmiya factory, the Shibaura factory, and the Ishikawajima shipbuilding yard, as well as the Tokyo arsenal. Those who joined the union were machinists, metalworkers, canners, foundry workers - all so-called 'steelworkers', but at the same time, there were also members whose jobs clearly had nothing to do with making steel, such as carpenters, dyers, engine drivers, and stokers. There were provisions in the union rules that recognised engine drivers and stokers, whereas for carpenters and dyers, there was 'a loophole rule' that allowed those to become members who 'belonged to various trades engaged in work at steelyards such as engine drivers and stokers etc.', so with such an interpretation, they cannot be seen as infringing the rules. However, later on, irrespective of such rules, whoever applied to join was allowed in, no matter what their place or type of work. For example, what was originally a completely separate union, the Yokohama Western Furniture-makers' and Cabinetmakers' Association, was taken in its entirety and recognised as branch number 41 . And as if that were not enough, even 'factory bosses and engineers' were allowed to join.

When these points are compared with practices in craft unions in Britain, the mother country of trade unionism, the difference is clear. The basic organizational principle in the craft union is that of a severely restricted membership qualification. Those able to join a union had completed a fixed period of training as an apprentice under a regular member of the union. The required number of years' training differed according to the period and the occupation, but normally, it was between 5 and 7 years and the training had to be completed before the age of 21. Attention needs to be drawn here to the fact that 'skilled' in the term 'skilled worker' in craft unions referred not to 'occupational ability' but, more than anything, to 'qualification' . This weight given to qualification is evident not only in Britain but in skilled and specialist occupations throughout the West. In fact, as this is the central point in my talk today, I would like to spend a bit more time on it.

It is natural that the period of apprenticeship should vary according to the occupation, but in the mid-19th century, it appears that in many jobs, the long 5-7 year period of apprenticeship stipulated by union rules no longer existed. There were not a few jobs in which the possibility of training was comparatively easy if there was experience in a related field of work. For example, at least a few carpenters would have had the ability to apply their skills immediately to make the wooden moulds for castings and some furniture makers would have been able to work as carpenters. Alternatively, quarrymen would certainly have been able in a short time to familiarize themselves with the skills necessary to do a stonemason's job. However, in a craft union, even though applicants for membership may have had the technical skills for the job that were in no way inferior to those of union members, if they did not have the 'qualification' of having served an apprenticeship for the prescribed period of time until the prescribed age, then their admission into the union was not recognised. And not only that - as a so-called 'illegal man', such a worker was prevented from doing that job. Even if he was a bona fide full member of another union, it made no difference to the fact that he was an 'unqualified' man. If it happened that an employer employed an unqualified man, union members would refuse to work with such a man and would quit the workplace. This was a stubbornly maintained habit among skilled workers and was prohibited by many provisions in union rules. Craft unions strictly controlled the qualification for membership in order to prevent the lowering of working conditions that would invariably result from an excess in the labor supply in any one occupation. Consequently, the unions regulated not only the length of apprentices' periods of training, but also the number of apprentices relative to the number of union members. It followed that the boundaries of 'the trade' were historically well-defined.

Naturally, the content of jobs altered with changes in technology, and as a result, labor disputes broke out over the definitions of the boundaries of 'the trade'. For example, when, in the shipbuilding industry, wooden ships gave way to iron and steel construction, fierce disputes broke out between ship's carpenters' unions, which had been the base of union organization in the shipbuilding industry, and the new metal machinists' unions such as the boilermakers. In a sense, the definition of trades was determined historically by just such boundary disputes. However, in Japan, such boundary disputes have hardly ever occurred, because trades were never clearly defined. Also, there had hardly ever been controls on the number of apprentices. For example, in one engineering works, there were 50 or 60 apprentices working, who had all been brought in by one boss (oyakata) . In other words, workers in Japan's engineering works, which provided the organizational base for the Iron Workers' Union, did not share the craft union mentality, which saw as basic the need to control entry into the trade in order to ensure that the power relationships within the labor market would be to one's own advantage.

4. Did single trade bodies with the 'closed shop' and internal regulatory powers also exist in Japan?

At this point I come to a topic I would have been keen to take up at this conference, namely, Tōjō Yukihiko's paper entitled Meiji 20~30 nendai no 'rōdōryoku' no seikaku ni kansuru shiron (An Essay on the Nature of the Labor Force 1887-1907) and the argument developed in his book Seishi dōmei no jokō tōroku seido (The System of Registering Female Workers in the Silkworkers' League), which includes his essay as a supplementary paper. In passing, I note that despite the many achievements of Japanese labor research, Tōjō's paper is the only one by a researcher into the pre-modern period that has drawn any attention. This is another reason why this issue deserves to be addressed.

As am I, Tōjō is strongly conscious of the particular characteristics of Japanese society; he argues against using the West as a standard and is in favor of the necessity of a comparative perspective. Thus far, we share the same view, but the argument he develops is actually quite far removed from my own way of thinking. Or rather, I ought to say that in terms of its content, it is completely the opposite of my approach. In his book, as those of you who have read it will know, Tōjō freely applies his own definitions and develops his own thesis about pre-modern Japanese society. Not so much a work of historical research, rather, it is the fruit of laborious philosophical contemplation that makes use of labor history materials, which those unfamiliar with ' Tōjō's language' will find extremely difficult to comprehend in the limited time we have available. For a critic of his to give merely a short introduction to his ideas would be to invite the charge of unfairness, so rather than using my own words, I think it better if you come to a judgment on the basis of your own reading of Tōjō's works themselves.

I would like to look a little more closely at his argument by citing his own words. For example, he argues as follows:

"Although the labor force in heavy industry at that time had a 'traditional' pattern, it had a closed shop and internal regulatory system that bears comparison with classical western occupational unions" (Tōjō op.cit. p.424, the following citations in parentheses are from the same book).

"They formed 'single trade associations' that had social status and quite firm boundaries (I shall use the term 'single trade associations' to generalize the groups studied in this book which operated as intermediaries in relation to workers' personal interactions and combinations. It was not a question of whether 'single trade associations' had organizational, legal, or systematic regulations and mechanisms. What was of importance was a) that they recognised each other as individual members and the fact that the groups had self-evident and clear boundaries b) that workers realised that being a member meant being aware that in some sense there was something to be shared through the medium of the group - among workers in heavy industry this meant 'skills' that represented social norms c) that regulatory conditions of a kind were laid down that included a customary (traditional) pattern and workers' frequent changes of job basically went on within these groups." (p.426)

"Based on that autonomous solidarity, skilled workers formed 'single trade associations', and the workplace was regulated by these. What went on within them was something of a 'black box' for capitalist owners." (p.3)

The problem is that Tōjō does not actually prove his argument here but simply affirms his view in the course of developing his particular thesis that 'pre-modern Japanese society' = multiple layered civic society'. He argues that
the labor force in heavy industry "had a closed shop and internal regulatory system that bears comparison with classical western occupational unions", but that 'closed shop and internal regulatory system' actually was not necessarily clear. On that point, he says the following:

But did workers in Japanese heavy industry actually have a 'family business-based' consciousness? A case I have studied in depth is that of mineworkers before the Meiji Restoration, in the Ani copper mine before and after the 1840s, where even furnace workers (seirenshokunin), whose families had been in the trade for generations, showed no signs whatever of having had a 'family business' mentality and who, if they had the chance, would cheat mine officials in order to make some profit for themselves on the side.

Still less can one imagine the workers at the Tokyo arsenal, the main complement of the Iron Workers' Union, as regulating themselves in accordance with a 'family business' mentality. Of course, it can readily be granted that "status relationships...and...personal connections" existed between skilled workers and apprentices who were linked by their technical skills and customs, but it is hard to imagine that, as the term 'personal connections' implies, there was a direct personal relationship between the oyakata and the totei and that this relationship was the basis of the power that regulated the whole 'single trade association'. If one argues that 'single trade associations' were closed shops, whatever the trade, one needs to show concretely how admission to the association was regulated. Tōjō, however, comments that in Japan the definition of 'occupation' itself is vague, so he acknowledges that there are difficulties with his own argument.

By acknowledging in this way that a crucial point in his discussion of 'single trade associations', namely, the definition of 'occupational types', is unclear, Tōjō loses the ground under his argument. I am completely at a loss to comprehend how he reconciles his assertion that "they recognised each other as individual members and the fact that the groups had self-evident and clear boundaries" with the statements I have just quoted.

With regard to the entry requirements, Tōjō acknowledges research on the early period of the Nagasaki shipyards by Nakanishi Yō, who writes that "in Japan's guilds and classical artisans' groups clear-cut 'controls' had always been weak, and [that] at the end of the Tokugawa period and the beginning of the Meiji Restoration (1860s) there were almost no systematic controls on the length of apprenticeships or the number of apprentices taken on" (p.424).
In a society where there were almost no systematic controls regarding apprentices' numbers and length of training, how could "firm controls and autonomous single trade associations" have existed? Also, even assuming that they had, why did they decline within a few decades, leaving almost no trace?
The British craft unions, with which Tōjō claims Japan's 'single trade associations' can "bear comparison" by the same standard, were able to sustain and develop their organization into the new century without any loss of strength in their regulations, despite suffering repeated major defeats in labor disputes where the issue was the 'craft-based rules' that were the basis of admission to the union. Ignoring such differences, what meaning can it have to claim that Japan's heavy industrial workers "had a closed shop and internal regulatory system that bears comparison with classical western occupational unions"ō

Furthermore, Tōjō argues that "based on that autonomous solidarity, skilled workers formed 'single trade associations', and the workplace was regulated by these. What went on within them was something of a 'black box' for capitalist owners." However, this claim differs widely from the situation as described by Japanese labor historians. If one is going to argue against this conventional understanding, then it is essential that one should put forward an argument backed up by solid evidence. If it is argued that in 19th century Japan, there were companies where workers were allowed autonomy in the workplace, and that those companies were consequently 'in the dark' as to what went on in the union, then one would like to know the names of those companies and to have some proof of this claim. I myself did some research into the employment and training of miners at the Ashio copper mine from the 1880s into the early years of the 20th century and found that, far from the workplace being a 'black box' for the owners, the management of production was from first to last almost entirely in the hands of educated technicians . If indeed it were the case that the Japanese workplace had been a black box for its owners, then surely Japanese capitalism would not have been able to develop so rapidly. In this connection I would like to refer to the work of Sugayama Shinji, who subjects Tōjō's argument to detailed and careful scrutiny in a paper entitled 'An Aspect of Skills Training in the Japanese Industrialization Process' . In his study of the careers of older workers and work group leaders at state-owned steelyards, Sugayama makes clear that Japanese skilled workers were not a closed group and that it was certainly not unusual for them to mix personally with people in other levels of society. He concludes: "In state-owned steelyards, from the time of their founding onwards, moving workers around, including moving them to 'faraway workplaces', was frequently used as a method of managing workforce requirements, and this plainly shows up the weakness of the autonomy of workplace associations organised by senior workers (oyakata).

I too certainly do not intend to deny that there were many single trade associations in Japan and that these groups enforced their own style of 'autonomous' regulations. It is only natural among workers belonging to the same trade that necessary regulations should be applied in the group when the duties of the job are contravened. For example, in the case of longshoremen who are often cited as a typical example of an unskilled job, to carry heavy loads while walking along boards from swaying barges to ships was a job that certainly required its own ability and experience. If workers without such ability or experience were included, then they could endanger the work of others and sometimes even cause disasters. In such a case, a longshoremen's association would enforce discipline and keep out unsuitable workers, which at first glance would seem like a case of exclusivity. This could also be called the 'autonomy' and the 'exclusivity' of a single trade association. However, it is not a case of control called for by the technical requirements of the job itself and it is something different from the rules 'that bear comparison with the classical occupational unions of the West' which were based on socially recognised qualifications; it clearly cannot bear such comparison.

There is a saying that has been handed down amongst Japanese workers : "as long as you have a skill, you have respect" (ude sae areba ichininmae). This clearly shows that Japan's occupational groups were above all concerned with a person's 'ability on the job'. By contrast, European guilds and graft unions were concerned with 'qualifications'. If this significant difference is not kept in mind, various issues cannot be understood, for example, why in Japan one did not 'look for a job' but 'look for a company', or why specialist occupations did not thrive. In Japanese society, although the important thing was the possession of ude (skill), what kind of formal process (shikaku - qualification) a worker went through in becoming familiar with that skill was not considered significant. What changed that situation somewhat was the introduction of western-style systems for specialist occupations such as lawyers, doctors, and nurses that required state-certified qualifications. Yet even in that case, a number of differences continued between practices in Japan and in the West. I consider that if this difference between the emphasis on 'skill' and the emphasis on 'qualification' is overlooked, then Japanese labor relations cannot be understood, nor indeed can labor relations in the West.

5. The reasons why the Iron Workers' Union could not become a craft union

I would now like to return to the question of why the Iron Workers' Union could not become a craft union. The issue was first addressed by Hyōdō Tsutomu and Ikeda Makoto, but the two men answered the question very differently.
Hyōdō sees the leaders of the Iron Workers' Union as aiming at an occupational union (shokugyōbetsu kumiai) but the combination of pressure on the labor movement from management and government together with the workers' own psychological immaturity meant that "in heavy industry after the Sino-Japanese War [1894-5] there was not much leeway that permitted activity characteristic of an occupational union that could maintain working conditions autonomously in accordance with its own union rules". In other words, "it was extremely hard for an employment system based on union rules to gain ground in a situation where a craft-based apprentice system in which senior workers (oyakata rōdōsha) trained apprentices and had firm control over them had more or less collapsed, where employers were gaining control of the process of training technical skills by means of learning-on-the-job systems, and where the more such systems took hold, the more workers trained under it were able to demand high wages and could move around freely."

Against this view, Ikeda maintains that labor leaders such as Takano Fusatarō and Katayama Sen "were not aiming at a classical occupational union which sought to maintain high wages artificially by means of controlling the labor supply and the dole" and develops a very different interpretation of the aims the Iron Workers' Union from that of Hyōdō. He points out that "Japan was trying to transplant technical skills which in advanced western countries belonged to trades (shokugyō) that were already being divided into 'shops' (shokumu)" but in Japan, "workers with training or skill in various crafts did not establish themselves as a stable class" and "there was little continuity between the artisans of the old iron industry and the workers in modern heavy industry, either in technical skills or personnel." In addition, with the rapid increase in demand for labor in heavy industry, "the proportion of workers who had started as artisans working with iron diminished and that of farmers and city-dwellers with miscellaneous skills increased considerably. The dramatic increase in demand for labor, combined with the fragmentation of skills, made the regulation of labor by unions very difficult."

When one considers that Hyōdō says that "a craft-based apprentice system in which senior workers (oyakata rōdōsha) trained apprentices and had firm control over them had more or less collapsed", and Ikeda says that "there was little continuity between the artisans of the old iron industry and the workers in modern heavy industry, either in technical skills or personnel", both men conclude that in Japan, among the former artisan class there existed a solid apprentice system and that controls on admission to the trade were possible. However, in fact, the collapse of the apprentice system in the former artisan class was widely recognised by contemporaries. For example, Yokoyama Gennosuke in 1896 records 65 different occupations in the Tokyo metropolitan area as having their own artisans' unions, including: plasterers, saw-workers, cellar carpenters, tile-makers, shipbuilders, tatami (straw) mat makers. He goes on to say:

Takano Fusatarō and others who were trying to promote modern trade unionism in Japan were aware of the important role played by artisans in western labor movements and directed their energies towards artisans too. Japan's first labor movement propaganda pamphlet, Takano Fusatarō's "To All You Workers" gives concrete examples of organization, and the first one presented is that of the carpenters' union. He also attended meetings of the Tokyo ship's carpenters' union and the doll-makers' union and worked towards their participation in the trade union confederation. However, the only traditional craft artisans that joined the movement as a result were the Yokohama ship's carpenters. In the later labor movement, even after the founding of the Friendly Society (Yūaikai), very few traditional craft workers joined the movement. It was during the First World War that Japanese artisans began to take part in the labor movement and apart from a very few areas, they did not leave much trace in it. This was a great difference from the central role played by skilled artisans in the early period of the labor movement in the West.

As the western model of labor organization (e.g. in the Iron Workers' Union) was not the only form of workers' movement in Japan, if it was the case that the old artisans' unions had been active in seeking to protect the interests of their crafts, then they too ought to be studied as one form of the Japanese labor movement. However, here too we find that the workers' movement in traditional Japanese occupations was not very vigorous. An exception to this trend was the tomoko dōmei (mining brotherhoods) - autonomous organizations of miners which maintained their organization and were relatively active for a long period. These mining brotherhoods were single trade associations that existed throughout the pre-modern and into the modern periods. On the basis of oyabun-kobun [senior-junior] relations that linked older and younger men through the acquisition of skills, the loose-knit, widespread brotherhoods functioned to provide mutual assistance among mineworkers throughout the country, and were found from Hokkaidō [in the north] to part of Kyūshū [in the south]. To become a full member of a mining brotherhood, one had to go through a fixed period of training under an oyakata (boss or senior worker), and there were restriction on how many apprentices a boss could have; among Japan's single trade associations, the mining brotherhoods were the closest to a craft guild. However, even the mining brotherhoods showed no sign of deliberately trying to control the labor market for miners. If they had, then they could have dominated the Japanese miners' labor market, even if only temporarily. Until the early years of the 20th century, the mechanization of mining presented various technical difficulties, so along with the growth of the mining industry, the number of miners rose dramatically. During this period, the brotherhoods played a significant role in familiarizing miners with skills. In the end, however, they did not function as controllers of the labor supply. This is clear from the fact that one could work as a miner without belonging to a brotherhood, and also from the fact that mineworkers who did not join a brotherhood were called murakata kōfu (village miners). Also, although apprenticeships were supposed to last 3 years 3 months and 10 days, this was not strictly adhered to, and once a certain level of skill had been acquired, a miner might move to another mine and pass himself off there as a fully-fledged miner. It has been noted that this looseness with regard to the period of apprenticeship was not just to be found among miners but also steelworkers and indeed every type of artisan.

In addition, if Japanese artisans' groups had had the same character as European craft guilds, one might expect to find more records of their activity at the end of the Tokugawa period (1850s and early-mid 1860s) and the early Meiji period (after 1867), but in fact there are very few such records. As you will know, in 1868 the Meji Restoration government issued the Commercial Code that liberalized the situation for merchants' associations (kabunakama) - a company could now have as many NAKAMA as it wished - but prohibited workers' groups (NAKAMA) from regulating wages and prices. If such NAKAMA resembled European guilds in their organization, this law would certainly have brought about dramatic changes in daily working practices. Yet on the contrary, there hardly seems to have been any opposition from the artisans' or merchants' groups (NAKAMA).

Alternatively, it could well be imagined that with all the reforms following the Meiji Restoration, artisans who had supplied samurai would have suffered a great deal. There were many artisans and suppliers who catered for the needs of the samurai: those who made military equipment such as swords and guns; makers of items of special clothing for samurai such as formal dress; artisans who supplied samurai patrons with all kinds of luxury items, as well as the merchants who were the intermediaries between the artisan and the samurai. Other craftsmen and artisans affected would have been those whose occupations were damaged or destroyed by the arrival of products and skills imported from the West after the opening up of the country. Despite all this, here again one sees hardly any evidence of an organized movement among artisans.

These then are the points on which I disagree with the interpretations of Hyōdō and Ikeda. The rapid development of Japanese capitalism after the Meiji Restoration destroyed the traditional apprentice system, and the reason why the Iron Workers' Union did become a craft union was not because of the weakness of the continuity of skills and personnel between the former artisan class and heavy industrial workers, but because Japanese society was originally not one that was accustomed to craft union-style controls. In other words, I am of the view that society in general did not regard as justified either the monopolization of certain trades and the sale of certain products by particular groups, or agreements by those groups on minimum wages, however the parties involved may have regarded such things. An assessment as to the correctness of this conjecture will require an examination of the form and characteristics of single trade associations in the Tokugawa period.

Ⅲ Occupational Groups in Pre-Modern Japan

I would now like to come on to my main topic - a comparison of the characteristics of Japan's pre-modern occupational groups and European guilds. As both of these areas of study are outside my own field I shall only be offering conjecture based on the results of research in modern labor history, so naturally enough, what I have to say by way of interpretation will only be hypothetical. Whether scholars of pre-modern history will see it as containing some points that can be helpful for their studies or whether they feel it is merely the conceit of an amateur I must leave up to them.

1. Free Cities and Castle Towns

Both Japan's 'NAKAMA' and European guilds had in common the fact that they were both organizations in which workers belonging to a particular line of work sought to defend the interests of their craft, support each other and cultivate friendship. Yet the two were decisively different in their relations with the authorities. Whereas in Europe, the guilds themselves were actually involved in the management of urban communities, Japanese NAKAMA were under the control of the shogunal authorities. They were also concentrated by the authorities in castle towns where an eye could be kept on them, and the organizations themselves were formed with the publicly declared, or else unspoken, recognition of the shogunal or local fief authorities. Recent research has shown that mediaeval European 'free cities' were not all as uniformly 'free' as had previously been thought, whereas the municipal government of Japanese towns (chō) has drawn more attention, so this kind of argument may be a bit out of date. However, first I think I need to reconsider the difference between the character of Japanese towns and European towns.

It goes without saying that European towns were established with the permission and under the protection of royal authority, but essentially, they were free cities based on citizens' self-government. On the other hand, while there were self-governing cities in mediaeval Japan such as Sakai, Hakata and Kuwana, they could not be not free cities. With regard to the difference between 'self-governing city' and 'free city', I would like to quote something from Nihon kinsei toshishi no kenkyū ("Historical Studies of Pre-Modern Japanese Cities") by Wakita Osamu who has studied the Japanese self-governing city.

Apart from the last sentence - "they therefore left the original urban arrangements and modes of economic activity as they were and left these to 'self-government' by the townspeople" - I find all the other points I have quoted persuasive.

2. Guilds and NAKAMA - Social Concurrence with Limits on Competition

A major difference between the guild and the NAKAMA that has drawn attention is the question of whether or not society in general concurred with their methods of "protecting the interests of the trade". In Europe, guilds fulfilled their objective of "protecting the interests of the trade" by monopolizing a particular craft or trade within a certain geographical area. Many townspeople belonged to some kind of guild, and so the policy of restricting competition between guilds was widely accepted as something natural. In practical terms, this meant controlling the number of apprentices so that the number of competitors would be limited, putting effort into maintaining the quality of products, determining criteria for the quality and quantity of raw materials and finished goods, carrying out autonomous inspections, and punishing those who infringed the rules. The quantity of products was also regulated, thus maintaining fair prices. To this end, customs such as the prohibition of night-time work and the regulation of working hours were also widely observed. To become a full member of a guild, one had to observe the formalities laid down by that autonomous guild; as an apprentice one had to train for a fixed number of years and then obtain the acknowledgment of all the guild members for one's admission. Because of these regulations, in craft guilds the number of trained men increased but the right to become a master (oyakata) was limited, so more and more men spent their lives without being able to rise to the status of a master, and the system of 3 levels emerged: master, artisan, apprentice. In this situation artisans formed their own artisans' organizations similar to those of the masters and developed their own movement that combined mutual assistance with demands to the bosses for higher wages and reductions in working hours.

It is known that in order to defend the interests of the trade, Japan's NAKAMA also restricted the number of members and came to arrangements on how to raise prices and wages. However, the central (shogunal) or local feudal authorities frequently refused to allow these 'arrangements'. As you know, the official proclamation after the great fire of Edo in 1657, referred to carpenters, sawyers, roofers, stonecutters, plasterers, straw mat makers setting up their own groups and arranging wage rises and prohibited this activity. The official proclamation of 1710 censured the thatchers for excluding a man who had broken their arrangement and ordered him to be reinstated in the NAKAMA. These ordinances show that artisans' groups did try to protect the interests of their trade 'autonomously' but at the same time, it is clear that the shogunate did not recognize the legitimacy of the groups' 'arrangements'. Just because such official prohibitions were issued does not mean that artisans always abided by them, and it can easily be imagined that they secretly attempted to maintain their 'arrangements'.

Of course, the shogunate recognized the groups' right to monopolize the craft and control the number of members, but its reason for doing so was not because it considered that the monopoly of an artisans' group over a particular trade was justified in itself, as with the European guilds. Rather, it posited other 'official' reasons that necessitated this, such as the need to maintain a stable supply of socially required goods. The shogunal authorities did not consider as justified that NAKAMA should be able to monopolize a particular craft, limit the number of members and fix the price of products. Within the NAKAMA it was seen as only natural that such means should be used to defend members' interests, but outside it was not a justification that could be avowed openly.

Another point that is often referred to is that Japanese NAKAMA did not necessarily enforce strict limits on membership numbers, and there are cases when even merchants' associations (kabunakama) actually deliberately pushed membership up. For example, the medicine wholesalers' association of Doshōchō in Osaka is thought to have been one of the few examples of Japanese NAKAMA that strictly held to the agreed number of shares. However, the management of the kabunakama was not in the hands of a single person but was directed by a 'household' (ie) that included non-consanguinary servants, and because the NAKAMA did not regulate the number of servants, it is supposed that the actual number of members of the NAKAMA fluctuated considerably. In the same way, even though membership numbers were controlled, the unit of membership was not the individual but the 'household' (ie), and in my view, more attention should be paid to this point as a characteristic of Japanese merchants' NAKAMA. However, in this medicine wholesalers' association, clerks with long service, even if they had no regular shares, could take part in buying and selling as a member of the NAKAMA by means of the loophole of creating sub-groups (kumishita). In this case, the rules were simply a formality, and the operative value system was one that emphasised the individual's particular personal situation. It was a society that regarded rigid adherence to rules as pedantry and which valued a wise and understanding compassion in human affairs [cf. the judgments of Solomon - trans.] That tradition still lives on today, and because people interpret things on the basis of the accumulation of established facts and show a cool disregard for theories and principles, even the clear clauses of the modern Japanese constitution have become a dead letter.

Of the pre-modern NAKAMA, merchants' associations that operated the shimekabu system were very few in number. Also, many merchants were involved in the transport of materials and products, and whatever the geographical region of the artisans' NAKAMA, a monopolistic business was recognized only for a limited number of trades such as the hairdressers who doubled as watchmen at gates and bridges. It is well known that even those hairdressers did not join NAKAMA, but there were a lot of them. You will also be familiar with the fact that, historically speaking, there was a period in which monopolies in business and limits on NAKAMA membership were completely forbidden.

At any rate, Japanese artisans' NAKAMA were not allowed, either by their employers or, of course, by the shogunate, to limit apprentice numbers or try to control the labor market, as European guilds did. Towns in Japan had not been surrounded by walls, and in a city like Edo [Tokyo], where there was constant traffic of great numbers of people in and out from all parts of the country, there were practical difficulties for artisans' NAKAMA in trying to control membership numbers and the labor market. In especially busy periods such as after a great fire disaster, particularly in trades such as carpentry, where there could be alternatively hectic and then slack periods, large numbers of migrant labor would have to be brought in from other provinces. For work such as repairs to the feudal lord's mansion, craftsmen would be called in from all over the native province. It can well be imagined that the artisans of the Edo NAKAMA were in no position to refuse such work under the aegis of the lord (daimyō) of the local domain. If building workers, who constituted the largest proportion of artisans, were incapable of effecting a monopoly in their trade, it would have been hard for other artisans (NAKAMA) to enforce regulations based on the kind of trade monopoly that was common in European guilds. The licences issued to the carpenters led by the Nakai household in the Kansai region are well-known, but there were still many 'novice carpenters' who worked without licences. In this case, licensed carpenters did not demand the exclusion of unlicensed men but called for them to be brought into the NAKAMA in order to reduce problems for each man's job.

In such a society, it was undoubtedly extremely difficult to enforce length of service for apprentices with any rigor. One reason why the European apprentice system continued to exist with such stubbornness for so long was the persistence of the social custom which held that as long as a man did not have the qualification obtainable only after completing the requisite period of apprenticeship, he had no chance of getting into that trade. By contrast, in Japan, where there was no social consensus about a single trade association's right to monopolize the labor market, the apprentice system merely meant a system of acquiring skills. Of course, the apprentice system was defended as one which provided a substitute family relationship and helped a person's character; the transmission of skills alone did not bind the master and apprentice. However, for the apprentice who had learned a respectable amount of skill, moving somewhere he could get higher wages was a temptation that was hard to resist. The master-apprentice relationship could not but be adversely affected when apprentices were severely reprimanded by their masters or when they were tempted by other jobs offering high wages. Of course, being removed from the register and becoming masterless was certainly a disadvantage to some extent. However, in view of the fact that many servants or apprentices did run away or go missing, and that employers in the same trade would hire such leave-takers and absconders, it is clear that there were limits to the effectiveness of NAKAMA rules. In his book Nihon no kasō shakai ("Lower Class Society in Japan" see pp.84-5), Yokoyama Gennosuke regrets the fact that the period of apprenticeship, "compared to what it used to be", became shorter and that even this regime could not be maintained and many apprentices still ran away. But one can see similar laments in documents from the end of the 16th century. What accounts for such a situation is not only the development of the urban economy in the Edo period (1603-1867) but also the critical fact that the Japanese apprentice system was not determined by the obtaining of 'qualifications'.

3. Self-help, Self-defence and Dependence on Authority

A characteristic of the European guild and craft union was that craftsmen's interests and benefits at work were defended by the guild members themselves. This was not a matter of having to protect themselves by their own efforts because they could not obtain protection by the state. Rather, it was because they affirmed the value of not depending on the state and helping each other; this was collective self-help. In England in the 19th century, alongside the craft unions, mutual assistance was also at the core of the many friendly societies and cooperative societies, and of the host of voluntary associations that emerged. Many of the early trade unions were actually friendly societies. What cultivated this spirit of voluntarism that eventually passed on to the trade union movement was the guild basis of urban society.

By contrast, in Japanese occupational groups, as we have seen, there was no absence of autonomous efforts to protect the interests of their members, but when problems arose, there was a strong tendency to turn to the town elders, or to solve problems by appealing to the magistrate's office. Examples are evident of problems such as matters relating to national officials or disputes that were difficult to solve within the NAKAMA such as issues over stockholding rights, but also, there were even supplications to urban authorities to rule on cases of infringement of arrangements made within the NAKAMA. Naturally, such attitudes can be ascribed to the authorities' constant refusal to acknowledge the existence of independent organizations with any real autonomy. However, the question of the relationship between the NAKAMA and the authorities should certainly not only be seen in terms of the one-sided application of power by the authorities. Under the supervision of leadership by the authorities the relationship was one of interdependence, a 'leaning on' each other. In a previous paper I actually emphasised the aspect of domination by decree and wrote that "artisans' organizations in the Tokugawa period were a means by which the authorities could levy forced labor on men belonging to particular crafts". But this was after all too one-sided. I learned from the concrete evidence in Inui Hiromi's book Naniwa Ōsaka Kikuyamachi (Naniwa Ōsaka Kikuyamachi) that in dealing with various issues relating to the NAKAMA or town elders, the magistrate's office was accustomed to make enquiries of the parties concerned before further proceedings.

4. The Problem of 'Organizational Culture'

Another problem which I feel requires examination, but which is not directly derived from the themes discussed here, is that of the characteristics of Japan's organizational culture. This is not just a matter of occupational groups - it is a broad theme that relates to all kinds of social and political organizations - and it is also important in considering occupational groups. I actually became aware of the importance of this theme when doing comparative research into Korean and Japanese trade unions. The organizational pattern of labor unions in Korea is overwhelmingly the enterprise union model; there is a very high proportion of mixed occupation unions, so the pattern closely resembles that in Japan. But when I went to Korea and studied how labor unions there were actually run, I discovered that there was large difference between the two countries, namely, in the method of electing officials and in the manner of running the organization. The Korean method of electing union officials and the organizational management based on it is actually 'presidential' in style. The candidates do not put themselves up for posts individually but form a team of 'running mates' to contest 4 posts: chairman, deputy chairman, secretary, and head of the policy department. The team that gains a majority of the votes cast by all union members then has the authority to appoint the members of the executive committee. Naturally enough, decision-making and management of the organization is carried on in a completely top-down manner. This applies not just to labor unions; it is normal throughout the Korean peninsula, north and south, in the political system and in companies.

When the organizational management of pre-modern Japanese occupational groups is looked at from this perspective, a number of discoveries can be made. First, there were many cases in use of a system of rotating officials. This is evident from the repeated appearance of official posts with names such as tōban (current duty), tsukiban (monthly duty) tsukigyōji (monthly referee, nengyōji (annual referee).

Another striking feature is that several people carried out the same post at the same time. As can typically be seen from the fact that there were two city magistrates in Edo who rotated on a monthly basis, there were many points in common with the political system of the Tokugawa shogunate, but I shall not go into this now.

What has attracted interest in the method of decision-making is of course the custom of unanimous decisions. In various organizations the custom was to require the agreement of all concerned. For example, a number of cases are known in which if a single member of a NAKAMA did not give his assent to a candidate's admission, then it did not succeed. It is conceivable that this custom was indirectly related to the fact that when problems occurred, all members of the NAKAMA were held to be collectively responsible.

Such organizational practices can be found in almost the same form in the tomokodōmei (mining brotherhoods) of the early Meiji period [post-1868], as is shown concretely in the tomokodōmei wage claim campaign that preceded the Ashio mine riot [1907]. These practices have not, of course, been passed on in the same form to today's labor unions. Officials are not rotated; they are chosen by election. Decision-making by majority vote is widespread. However, when we look at the state of organizational management, we see that those earlier practices can still exercise an influence. For example, however excellent a leader may be considered to be by his peers, in Japan it is not thought suitable for the same person to stay in the same executive post for a long period. By contrast, Samuel Gompers, who founded the American federation of Labor in 1886, remained as president for 36 years until his death in 1924 and lost only one an election during that whole period. Such a thing would be unthinkable in Japan, where the behaviour thought desirable for high officials is to complete one term of office and then make way for the younger generation while retaining influence oneself. This does not only apply to labor union officials; the same tendency can also be seen in company management. While fastidiousness in sticking to the principle of unanimity in decision-making is gradually fading, in many organizations, before important decisions are made, one sees that the custom of seeking the views of all concerned still stubbornly survives. Naturally, there are examples of company founders remaining in the same post for long periods, ultimately, majority voting is seen as the natural rule; it goes without saying that not everything can be explained by the customs of the past.

Conclusion - Issues for the Future

In my presentation today I have emphasised that in the main, the key for understanding the past is in the present. How effective this method is I must leave to you to judge. With regard to the last problem of organizational culture, I indicated the way in which the past controls the present. It's only natural that an examination of the ways in which the past does control the present should be at the centre of historical research, but I feel that rather than a 'one-way street' approach from the past to the present, a two-way research method deserves more study. For this to happen, it would be good if there were an increasingly lively exchange between scholars of pre-modern and modern history.

Finally, as there is a limit to what I myself can do with regard to future research in this area, I would like to conclude my rambling remarks with the self-indulgent request that someone take up the torch of this particular research.
1) It is surely important to bring an international comparative perspective into studies of pre-modern urban life and of the NAKAMA. That does not mean only Europe; it is important to include other geographical regions. What is needed here is not just theoretical analysis but discoveries of individual themes that can be verified on the basis of concrete data. It is on that kind of level that comparative research into guilds and NAKAMA can be conducted. I may not be fully up to speed here but I wonder if you share my own feeling that despite the great progress made in recent years in historical studies of the pre-modern urban environment, compared to previous years, there has actually been a falling back in international comparative studies in this area.
2) When considering the question of the continuity or lack thereof between occupational groups and especially labor unions after the Meiji Restoration, there seem to be key regions and types of work. For example, in the post-Meiji era labor movement, the main centers of organization were Edo and Ōsaka, and a study of the NAKAMA in those areas is certainly key. Especially important in connection with the Iron Workers' Union will be research into the NAKAMA of the Edo blacksmiths, the jewelers, and the shipwrights. Studies of the pre-modern NAKAMA have already produced some very good results, but I would like to see extensions of the period researched beyond the Meiji Restoration to investigate subsequent changes and to elucidate the actual extent of continuity.
3) As for single trade associations in the building trade, the carpenters had special weight as a trade association that united the many craftsmen working in the building trade and as one that passed on technical skills from the Tokugawa period to the Meiji period and beyond, and they would certainly be an important subject for research from a comparative international perspective. In this connection, a very interesting research theme would be the differences between the Tokyo stonemasons' union, which started a number of labor disputes in the Meiji period, and the carpenters' union, which had almost disappeared by the late 1880s.
4) Undoubtedly related to the question of frequent leadership changes is the fact that Japanese society lacked a system for producing specialist workers. Despite importing so many systems from China, why did Japan only try out the civil service examination system for a short period? When compared with a society like that of Germany, which puts such emphasis on qualifications, Japan can be said to be a society in which traditionally, qualifications have been valued less than an individual's ability to perform. One would certainly like to know what accounts for this difference. In this Japanese society there is a group that is an exception to the norm - a group which has created a western-style specialist organization: the bar association. If I may make so bold, I would say that that the bar association is the organization that exhibits the most craft guild-like, the most craft union-like functions in Japan today. In this sense, I think that an historical study of doctors' and attorneys' specialist professional bodies would be a most interesting theme for research.

A translation of "Nihon ni okeru Shokugyo-shudan no Hikakushiteki Tokushitu:Sengo rodokumiai kara jikan wo gyakkoshi kinsei no NAKAMA ni tsuite kangaeru" , in 'Keizaigaku Zasshi'(Osaka City University) Vol.102 No.2.
Translated by Terry Boardman

Writings of Kazuo Nimura

From Trade Unions to the 'NAKAMA' in Feudal Society
Organizational Characteristics of Occupational Groups in Japan

Kazuo Nimura

Ⅰ. The Characteristics of Japan's Postwar Labor Unions

1. Why enterprise unions?

2. Why did 'blue- and white-collar mixed unions' develop?

3. The Removal of Status Differences on the Shop Floor

Ⅱ. The Japanese Labor Movement before World War Ⅱ

1. Characteristics of Labor Disputes

2. Background to Anger = The Pretence and Reality of the Removal of the Status System

3. Iron Workers' Union and Craft Unions

4. Did single trade bodies with the 'closed shop' and internal regulatory powers also exist in Japan?

5. The reasons why the Iron Workers' Union could not become a craft union

Ⅲ Occupational Groups in Pre-Modern Japan

1. Free Cities and Castle Towns

2. Guilds and NAKAMA - Social Concurrence with Limits on Competition

3. Self-help, Self-defence and Dependence on Authority

4. The Problem of 'Organizational Culture'

Conclusion - Issues for the Future

Writings of Kazuo Nimura

From Trade Unions to the 'NAKAMA' in Feudal Society Organizational Characteristics of Occupational Groups in Japan Kazuo Nimura

Ⅰ. The Characteristics of Japan's Postwar Labor Unions

1. Why enterprise unions?

2. Why did 'blue- and white-collar mixed unions' develop?

3. The Removal of Status Differences on the Shop Floor

Ⅱ. The Japanese Labor Movement before World War Ⅱ

1. Characteristics of Labor Disputes

2. Background to Anger = The Pretence and Reality of the Removal of the Status System

3. Iron Workers' Union and Craft Unions

4. Did single trade bodies with the 'closed shop' and internal regulatory powers also exist in Japan?

5. The reasons why the Iron Workers' Union could not become a craft union

Ⅲ Occupational Groups in Pre-Modern Japan

1. Free Cities and Castle Towns

2. Guilds and NAKAMA - Social Concurrence with Limits on Competition

3. Self-help, Self-defence and Dependence on Authority

4. The Problem of 'Organizational Culture'

Conclusion - Issues for the Future

From Trade Unions to the 'NAKAMA' in Feudal Society
Organizational Characteristics of Occupational Groups in Japan

Kazuo Nimura