Writings of Kazuo Nimura

The Labor Union Movement at the Beginnings of Post-WWII Society in Japan

NIMURA, Kazuo

Introduction

The topic of this study is "the upsurge in labor unionism in Japan in the early post-WWII era considered in relation to 'continuation' and 'discontinuation' in the later enterprise union". There has long been international interest in the question as to how modern Japanese unionism, known for its particular characteristics, actually developed. So about 10 years ago I published a paper on the post-WWII union movement entitled 'Enterprise Unionism - the Historical Background' and followed this a few years later with another paper, 'The Historical Characteristics of Labor Relations in Japan'. *1  

In both papers I argued the fact that there was no tradition of autonomous craft guilds in Japan; that this was the main reason why occupationally-based unions did not develop in Japan; and furthermore, that this fact has had no small influence on modern labor relations. I also emphasized that a key feature of post-WWII Japanese unions was not just that they were enterprise unions but that they were so-called 'mixed occupation unions' [kôshoku kongô kumiai] which included both blue-collar and white-collar workers. There had been many enterprise-based or workplace-based unions in pre-war Japan, and in that sense, there was no big difference between the pre-war and post-war situation. However, pre-war Japanese unions were blue-collar-only organizations, and moreover, there have been a few examples of 'mixed occupation unions' outside Japan, so in this sense, I emphasized that it is precisely this combination of the 'mixed occupation union' with the 'enterprise union' that ought to be seen as the real characteristic of post-WWII Japanese unions.

These two papers, however, merely provided a broad overview of the post-war period. The present paper will limit itself to a more in-depth analysis of just the early years of the post-war labor union movement, which so transformed labor relations in Japan. There have been many studies of this subject, and it might seem as if there is little left in it to discover, but in fact, a surprising number of points remain to be substantiated by solid empirical evidence. In Japan, there is the debate about how to move on from a 'company-centered society', while abroad, there has been much discussion of 'Japanese-style management'. Today, when there is even debate about the 'Japanization' of labor-management relations, a detailed examination of Japanese labor unions from an historical and international perspective is surely timely.

1.    The Development of the Post-War Labor Movement - Quantitative Change and Characteristics

(1)  Changes in the number of unions and of union members

I shall start with a statistically-based confirmation of the situation during the period of the formation of labor unions immediately after the war. This may seem a rudimentary approach, but in fact it is one that has hardly ever been taken previously, and various statistical inaccuracies are still circulating even now. The statistics generally referred to with regard to labor unions are the Labor Ministry's "Basic Survey of Labor Unions[Rôdô kumiai kihon chôsa]. *2

However, this national survey began in December 1946, and figures for 1945 and up to June 1946 are those of the 'Statistics of the Establishment and Dissolution of Labor Unions', which were based on figures submitted by the unions themselves, and there are a number of problems with these 'Establishment and Dissolution' statistics. For example, at the point of survey, the values differed widely. The statistics were published promptly in the Central Labor Review[Chuô rôdô jihô] , but the values were constantly changing. As became clear later on, there were a number of cases in which figures were not submitted despite the formation of a union. The first set of near-complete figures worth looking at is that appended to Vol.1 of the"Documental History of the Japanese Labor Movement"[Shiryô rôdôundô-shi]. *3 (Table 1)

However, there are also doubts about these figures. Unions were first obliged to report their establishment in accordance with the Trade Union Law enacted in March 1946. It is unlikely that many unions formed before that date would have voluntarily reported their establishment.
   To look into this, I checked the order, by month and day, in which unions were clearly established in 1945 by examining histories of of labor unions and of regional labor movements and then the Directory of Japanese Labor Unions [Nihon rodokumiai meikan],*4 which was based on data reported to the Labor Ministry before the end of September 1947. I do not have space here to go into all the details, so I shall restrict myself to noting the results below.

Activity preparatory to union organization got underway throughout the country already in the month of the surrender, August 1945, but no unions were actually formed that month. From snippets of evidence it can be deduced that in September 1945 five unions were established, but there is nothing clear on the situation of any of them. The earliest new post-war union for which we have clear details of its formative process and subsequent activity was the All-Japan Seamen's Union [Zen Nippon Kaiin Kumiai], which held its inaugural meeting in Kobe on Oct. 5th 1945. Forty-one other unions can be confirmed as having formed in the same month. Other results for the rest of 1945 showed that 173 unions were established in November, 499 in December, and precise dates for a further 16 remained unclear. A total of 734 individual unions were therefore founded in 1945, which far exceeds the 509 recorded in the 'Establishment and Dissolution' statistics. However, the data I used often lacked figures for union membership, and the materials I had to hand happened to be secondary sources only, so they were not comprehensive. I was able to track down one other source that afforded more reliable statistics. This was 'Labor Union Numbers by the Year of Formation' included in the 1947 "Labor Union Survey Report" [Rôdôkumiai chôsa hôkoku] (Table 2)

Of the unions in existence at the end of 1947, 855 had been founded in 1945 when there were 602,706 members, figures far in excess of the figure of 509 unions and their membership of 380,000 which are normally used. As the number of union members may include members who joined after the founding of the unions, the numbers may be slightly smaller. On the other hand, it is certain the number of unions was greater than 855, because in the case of large companies, there are examples of workplace unions being set up before unions that covered the whole enterprises. For example, at the Ashio copper mine owned by Furukawa Mining Co., 12 workplace unions were formed in 1945 at the Tsûdô mine, the Honzan mine, and the Kotaki mine by workers in the mining, refining, water filtration, assaying, maintenance and construction, procurement, and general affairs sections, and another separate one by office workers.*5  Similar examples can be found at the Takahagi coal mine. However, many of these 'workplace unions' were soon amalgated into factory or mine-based unions, as happened at Ashio in March 1947.

At any rate, the 'Establishment and Dissolution statistics' for 1945 clearly did not represent the actual situation. The development of the post-WWII labor union movement evidently occurred much faster than has been thought; by mid-December 1945 the number of union members had certainly passed the pre-war peak of 420,000. I ought to point out here that errors in the 'Establishment and Dissolution statistics' were partly due to differences in officials' jurisdictional responsibility. Seamen, who come under the jurisdiction of the Ministry of Transport, were omitted, so the All Japan Seamen's Union,*6  the largest union at the time, was not included in the statistics.

What was the situation the following year, 1946?  Here too the 'Establishment and Dissolution statistics' for the end of June were generally used. However, the publication of the "Documental History of the Japanese Labor Movement"made it clear that just one union had disbanded before June 1946, and "30 Years of Labor Union Basic Surveys" was revised to record 12,006 unions and 3,679,971 members in those unions. On the other hand, Table 2 shows that at the end of 1947, the number of unions then in existence that were founded before June 1946 was 10,361, with 3,404,353 members, both of which figures are less than those shown in Table 1. There are unlikely to have been any large omissions in the 'Establishment and Dissolution of Labor Unions statistics'  for June 1946.

(2)   Estimates of the Number of White-Collar Union Members

What characteristics did these Japanese unions have that were formed in the immediate aftermath of the war?  For an answer to this question, we will need to consult "An Investigation of Post-War Labor Unions" [Sengo rôdôkumiai no jittai], *7 published by the Institute of Social Science Research, Tokyo University. Research based on a survey done in August 1947 discovered a number of important characteristics in post-war Japanese labor unions, beginning with their patterns of organization and types of executives, which became a cornerstone of subsequent research. Most noticeable in the survey results was the 'enterprise union', which was not very evident in other countries, and, alongside this, the fact that white-collar workers and even management staff  belonged to the same unions as blue-collar workers.

How those characteristics emerged and the elucidation of their origins became themes for subsequent studies in the field of labor research and will continue to be so. Here, I would like to focus on another characteristic - the large-scale participation in the post-war labor movement of white-collar workers and the leading roles such workers often played in the movement. Scholars have been aware of this before now, but there has been no statistical basis for the scale of the phenomenon,*8  and neither has its significance been adequately addressed.

I shall begin with a quantitative confirmation of the proportion of white-collar workers in the post-war labor union movement. Table 3 shows the numbers of labor union members at the end of June 1948 in three categories: white-collar worker-only unions, blue-collar workers' unions, and mixed occupation unions. There were 9196 unions with a total membership of 1,430,000.

The problem here is the number of white-collar workers in mixed occupation unions. The exact number is unknown, but an estimate can be made. First, we can deduce the numbers of white-collar and blue-collar workers from the National Census of 1950, using the numbers, 'according to industrial sector categories',  of  'those in employment'  aged 14 and above, and omitting owners and family businesses (Table 4).

There are no problems associated with white-collar workers in specialist technical occupations and in clerical posts. Normally, class composition theory considers managerial work to belong to the capitalist class, but in view of the fact that post-war labor unions sought to recruit managerial staff, they ought to be included too (white-collar A). blue-collar workers are considered to be those working in mining, transport, and manufacturing processes. It is hard to classify retail and service workers into the two groups, but as their numbers in mixed occupation unions are low, they will be ignored here.
   The proportions of white-collar and blue-collar workers are 4.3 and 5.7 respectively. Based on the figures in the table, we arrive at a figure of 1,760,000 white-collar workers in mixed occupation unions. When this figure is added to that of white-collar workers in white-collar unions, which was 3,190,000, the result amounts to nearly half the total number of unionized workers. Managerial staff, however, here also includes company executives, who were not the object of union recruitment.*9

If we then consider the category white-collar B, which does not include managerial staff, the proportion is 4:6. The number of white-collar workers in mixed occupation unions that results from a calculation on this basis is 1,637,000. This would seem to exceed the actual situation, because most specialist technical white-collar staff worked in the education, public service and financial sectors, where white-collar-only unions were numerous and mixed occupation unions few. In sectors such as shipbuilding and transport, where there were many more mixed occupation unions and many union members in absolute numbers, the number of white-collar workers was very low. If we now assume a compositional proportion of 3:7 of white-collar to blue-collar in mixed occupation unions across the board, we get a number based on this of 1,230,000 white-collar workers in mixed occupation unions and an overall total of 2,650,000 white-collar workers in white-collar-only unions. If we reduce the proportion still further to 2:8, the respective figures will then be 818,000 and 2,245,000.

According to the Table of Indices of Industrial Production [Kôgyô tôkeihyô] which considered only manufacturing industries, the percentage of white-collar workers was calculated to be 17.7% in 1947 and 17.5% in 1949, so a comparative ratio of 2:8 for the proportion of white-collar to blue-collar workers in all mixed occupation unions would be rather too low an estimate of the number of white-collar workers,*10  because there were other sectors which consisted almost only of white-collar workers, such as finance, education and the civil service.

However, my object here is to show the high degree of unionization of white-collar workers, so I shall therefore employ the final, more modest estimate of 2,245,000. This amounts to more than one third of the total number of unionized workers; 34.4% of Japanese labor union members at this time were white-collar workers. For a still more accurate estimate, one would have to add analyses of narrower occupational and industrial categories, and other statistics and surveys of smaller private enterprises, but even if the total were to be higher than 34.4%, it is unlikely to be lower.

We now need to work out the level of white-collar unionization. The number of all workers in employment in June 1948 which was used to calculate the level of unionization was 12,590,000. When this figure is multiplied by 34.9%, the percentage of white-collar workers, including those in managerial employment, according to the 1950 national census, is 4,400,000. But as the total number of unionized workers calculated includes those in the retail and service sectors, and those too are then considered to be white-collar workers, then the proportion of white-collar workers in the overall total of all workers in employment can be conjectured at 40%, or 5,040,000. With the unionized white-collar element as the denominator, the first figure amounts to 51.0%, while for the second figure it is 44.5%; with a numerator lower, and a denominator higher than the actual numbers, these are the figures that result. A comparison with other countries soon makes clear how high this figure is. For example, in Britain, arguably the most advanced country in terms of the labor movement, the number of labor union members in 1948 totalled 9,360,000, which was 3,000,000 more than in Japan, but of that total, only 21% (1,964,000) were white-collar workers. 28.8% of white-collar workers were unionized as against 53.1% of blue-collar workers workers; 45.1% of all British workers belonged to a union.*11   In the USA in the same year, 34.5% of the workforce were unionized: 44.1% of blue-collar and 16.2% of white-collar workers.*12

(3)   White-collar workers' leading role in the labor union movement

White-collar workers did not only make up a large proportion of union members; they actually led the whole labor union movement in the early post-war period. Our attention is once again drawn to this fact by the testimony of contemporary observers. "Witness of the Labor Movement" was a small booklet co-authored by Yakabe Katsumi, Murakami Kanji and others, published in September 1947. Its first article, 'Disputes Start from the Press' states:*13

"The first Yomiuri Newpaper dispute, which played such a big role in Japan's post-war labor movement, began on October 24th, only ten weeks after the defeat.......The 50 day-long dispute, which was not resolved until December 12th, drew great attention throughout society, as it employed strategies to control production, and because it was a newspaper. The dispute had a tremendous influence on the later development of the labor union movement. .......It was a long-running dispute that became a focus of national attention, and sympathizers visited the campaign headquarters at Yomiuri for days on end not only from the capital but also from as far away as Kyûshû and Tohoku with the intention of learning about the union and starting unions in their own factories. Thus, for a while Yomiuri became a kind of advice center for the union movement... the Yomiuri experience spurred on the formation of unions in Tokyo naturally enough but also throughout the country, and the manipulation of production became a new tactic employed by the union movement everywhere."

Thus those responsible for the first use of the "production control tactic" (seisan kanri senjutsu) - one of the characteristics of the post-war labor movement strategy - and of spreading it throughout the country were the 'university-educated journalists' elite of columnists, office chiefs and their deputies' *14 who led the Yomiuri Newspaper dispute. But at this time, when national union organization was not yet underway, the Yomiuri union did indeed function like 'an advice bureau for the labor union movement'. The Hokkaido Shimbun newspaper and other regional newspapers played a similar role.

The national union organization that led the union movement in the years immediately after the war was, indisputably, the Congress of Industrial Labor Unions of Japan (Sanbetsu Kaigi). An examination of the process that led to the formation of the Congress and of the composition of its leadership clearly reveals the prominent role played by white-collar workers in the post-war labor movement. The first to call for the formation of the Congress was the Japan Newspaper and Communications Union - a white-collar organization - the so-called 'Press Union'. The first Chairman of the Congress was Asahi Newspaper columnist and Press Union leader Kikunami Katsumi (educated at Kwansei Gakuin University) and the deputy chairman was the Meiji University-educated Dobashi Kazuyoshi, Chairman of the National Postal Workers' Union. Other university graduates among the Congress executives were Tsuzura Wataru (Tokyo Commercial University), Matsumoto Shin'ichi (Tokyo University) and Nakahara Junkichi (Kyoto University). The few blue-collar workers were men such as Deputy Chairman Sakaguchi Yasuo (National Railways), the experienced labor activist Hosoya Matsuta, Office Deputy Head and executive secretary, and Yamazaki Ryôichi. Moreover, the secretariat, which in later years maintained such a grip on the authority of the Congress that it was criticized for its 'secretariat politics', consisted overwhelmingly of university graduates and students.

By comparison with the Congress of Industrial Labor Unions of Japan, most of the leaders of the Japan Federation of Labor (Sôdômei), which had organized itself through personal connections between pre-war union activists, were from a blue-collar background, but there too, the executive secretariat at head office tended to be university and high school graduates. Hirasawa Eiichi, who was appointed JFL Founding Conference Secretary in May 1946, later reflected that the post-war generation men at JFL head office at that time were all intellectuals and said "I was about the only one who wasn't an intellectual".*15

If one had to choose a labor dispute that was representative of the post-WWII era, it would likely be the General Strike of February 1st 1947, and at the center of it was the National Public Workers' Joint Struggle Committee, composed of white-collar unions. Furthermore, there is no shortage of materials attesting to the major role played by white-collar workers in the many famous disputes of the early post-war period, such as the electric power industry (Densan) dispute, the postal workers' dispute, and the Tôhô film studio dispute.

White-collar workers were also to the fore in the establishment of unions in individual factories and workplaces. The Institute of Social Science, Tokyo University, carried out a detailed survey and analysis of the academic record and occupational status of leading figures involved in the establishment of new unions and of union executives, according to 'academic pedigree', industry and scale.*16  Apart from the mining industry, in which blue-collar unions predominated, the central figures in the formation of unions were more or less evenly divided between blue- and white-collar workers, and the latter were especially numerous in the industry-wide category. The history of individual, non-industry-wide unions also reveals that white-collar workers played a significant role in initiating and leading the process that led to the formation of unions.One such man was Masuda Tetsuo, known as the charismatic leader of the Union of Nissan Moter Co.

Statistically, the highest percentage of union participation among white-collar workers may well have been among those in personnel departments, who were involved with labor-management issues, often in cases of unions set up to serve employers' interests.*17  There were even examples of personnel department heads doubling as executive secretaries of white-collar unions.*18
   Here, I should add that I am not claiming that white-collar workers were at the center of the post-WWII labor union movement. More than half of union members were blue-collar workers, who actively campaigned for wide-scale wage hikes and the abolition of status differentials. However, if the Japanese labor movement in the early post-war period is seen in the context of the history of the international labor movement, it was one in which the proportion of white-collar workers was conspicuously high in comparison to other advanced countries. The leadership of the movement especially is noticeable for the large role played by white-collar workers, or rather, well-educated intellectual types. The focus of this paper is on the labor union movement, particularly at the company level, but if the labor movement is thought of in a broad sense, the characteristic outlined above becomes particularly distinct.

2.   The Origins of Enterprise Unions and Mixed Occupation Unions

(1)   Workplace Unions

Why did post-WWII labor unions in Japan become enterprise-based organizations rather than occupation-based or industry-based-organizations as is normal in the West?  A substantial response to this question referred to the fragmentation of the labor market along company lines. The representative proponent of this view was Ôkôchi Kazuo, who first put forward his 'migrant labor thesis' to account for the fragmentation of the labor market and later developed the following argument:

Until the 1910s in Japan, workers did not stick with one company but moved around in search of higher wages. From the 1920s, however, large companies employed means to retain their workers, and as a result, the pattern of 'long-term employment' was established whereby a worker would join a company soon after leaving school, be trained up in the company's training centers, receive regular wage increases in accordance with age and seniority and remain with the same company throughout his working life. A labor union is an organization for selling labor, and its solidarity is based on the fact that those in the same labor market share the same interests. Consequently, in a labor market that is enclosed within or restricted to a certain company, a company-based union will be established.*19

I put forward a detailed analysis and critique of this argument in the paper mentioned at the beginning of this essay. No counterarguments to this appeared, and I had come to think of the issue as settled, but recently, I became aware that there are still some who believe that the pattern of labor union organization is determined by the structure of the labor market. For example, Tabata Hirokuni writes that : "In the inter-war period the labor movement was focused on horizontally-structured unions but was forced into stasis by the closed structure of the labor market. As is already known, the formation of internal labor markets in large companies in this period was the origin of post-war company-based unions."*20

"The formation of internal labor markets....was the origin of post-war company-based unions" is rather unclear; presumably, it means that company-based unions emerged after the war because of the formation of internal labor markets in large companies during the war. If so, it differs not at all from Ôkôchi's argument. Although I shall be repeating myself on a number of points, I would like to re-examine the issue.

First, it is hardly tenable to maintain that Japanese labor markets were fragmented and enclosed within companies when unions were forming during the early post-WWII period. Certainly, in the second half of the 1920s, more employees in large companies did tend to remain with their companies, and labor markets did show a trend towards fragmentation along company lines. However, during the war, with labor shortages due to conscription and the rapid increase in production for the military, there was a large-scale flow of labor in and out of companies, and with the surrender, many factories either ceased, or scaled back, production, which led to massive unemployment. Many people just quit work of their own accord and left the cities, as they had either lost their homes through bomb damage or else had too little to eat. Company and union histories attest to the large numbers of those who gave up work or else were laid off; many workers were looking for work long after having left school. In other words, labor markets in the 1940s were not 'closed' but rather, extremely fluid.

Secondly, there is a problem with the meaning of the phrase 'pattern of lifetime employment', namely, that the 'lifetime employment pattern' that we speak of today which is deemed 'almost a workers' right' began after the war and during the period of high economic growth after the anti-rationalization disputes of the 1950s. During the inter-war years, those entitled to a 'lifetime employment pattern' were office staff and key skilled blue-collar workers only, and levels of employment insurance were also low. As for the 'seniority wage system', the harsh post-WWII inflation and consequent wide differentials in company wage levels weakened companies' abilities to hold on to their workers, which made the rules of the 'internal labor market' very different from those that obtain today.

Thirdly, the argument that labor markets determined the pattern of union organization cannot explain the fact that post-WWII unions became predominantly those of the mixed occupation type, the membership of which included both blue- and white-collar workers. Japanese labor markets have been almost consistently divided on the basis of educational achievement and gender. Even within the same company, the interests of those from high school or university backgrounds, or of male or female staff frequently differ. If it is argued that labor unions are organizations of those who share interests as a result of belonging to similar labor markets, then there ought to exist Japanese labor unions on the basis of academic achievement and gender, but in fact, there are no such unions.

Why then did post-war Japanese unions become workplace- or company-based organizations?  In my view, solidarity with one's daily workmates is the most basic and natural form of workers' combination. In any country and in any period - unless special conditions prevail - forming a combination with one's workmates will be seen as a natural course of action.  Especially, as in post-war Japan, at times when workers are laid off due to factory closures or when all employees in a workplace face common threats to their livelihood, such as inflation, it is only natural that workers combine with those who work alongside them every day and make demands of their common employer; this hardly requires any particular explanation.

On this point, there are some special conditions that certainly do call for explanation, namely, craft unions, which ought to be regarded as the original model for the western labor union movement. Why did they transcend the boundaries of the workplace and unite only those within the same occupation?  Why, despite the fact that workers saw each other and worked alongside each other every day, did they combine in craft unions only with those of the same occupation and join in solidarity with other workers of the same occupation whom they had never seen before? There must be particular reasons to account for this. And not only did the workers opt for this kind of organizational pattern, employers also accepted it.

I went into this question in some detail in another paper but suffice it to say here that my view is that the behaviour of craft guilds in the free cities of mediaeval Europe was continued by the labor union movement in the modern period. Common attributes of craft guilds are workers in the same occupation banding together, limiting the number of apprentices, and maintaining their working conditions by regulating working hours and amouts of production. The feeling of solidarity between those in the same occupation was created over the course of centuries. By contrast, in Tokugawa era Japan (1603-1867) there were no free cities; townspeople all lived under the control of the samurai.

There were of course tradesmen's organizations, but there was none of the autonomy seen in the guilds, and conditions of admission were lax. Whereas in the West, most of the early trade unionists were artisans, Japanese artisans' organizations and campaigning activities were very weak. The reason why Japanese workers did not opt for occupation- and industry-based unions*21 was because there had been no historical formation of a 'working class society' based on the firm association of workers in the same trade, which provided the solidarity that could go beyond the individual enterprise.

What gave Japanese workers their uncomplicated and non-intellectual feeling of comradeship was the workmates with whom they worked in the same workplace or enterprise. Feeling solidarity, as members of the same social class, with employees in rival companies did not come naturally. Conscious acquisition of such a 'class consciousness' was only possible through the teachings of Marxism or the example of the history of the labor movement. In cases where there was fierce competition between companies, it was not surprising that the employees in other companies all became the 'enemy', and one's 'friends' were those in one's own company, including the management.

Acknowledgment of the lack of a tradition of craft unionism not only makes it easy to understand why trade unions that transcended individual companies and were able to regulate working practices did not develop in Japan, but also makes other factors comprehensible, such as the lack of resistance to the introduction of new technology as long as jobs were not threatened, the easy acceptance of job transfers, mixed occupation unions that include both blue- and white-collar workers - indeed, all the characteristic features of Japanese labor-management relations. Enterprise unions are not only found in Japan. They are the main form of union organization*22 in many countries in Latin America, and there are also a number of countries in Asia where they exist.*23 In pre-revolutionary Russia too, workers are known to have had a stronger consciousness of affiliation to their company than to their occupation.*24  In all these countries, the craft union tradition seems to have been weak.

(2)   Mixed Occupation Unions

Why then did blue- and white-collar workers join together in the same unions?  Koike Kazuo explains this as follows:

On the basis of the idea that 'internal advancement' was taken to be something progressive, the other special features of Japan's labor-management relations can easily be explained. For example, why Japanese labor unions adopted a form of organization which includes white-collar workers is easily explainable. .... Because Japan's labor unions are of the mixed occupation type, they have succeeded in organizing white-collar workers. The most important reason for this is that organized labor in Japan came to share many characteristics with white-collar workers. In large companies, systems of internal promotion were extended widely to key blue-collar workers, who, as a result, came to have basically the same character as white-collar workers. This group of key skilled workers in large companies is the central constituent of Japanese labor unions. Consequently, although miners may work for large companies, because they do not have internal promotion systems, even today blue- and white-collar workers in the mining industry belong to their own unions.*25

However, it would appear that the cause and effect relationship has reversed here. It is not that 'mixed occupation unions with blue- and white-collar workers' developed because "key blue-collar workers in large companies came to have the same character as white-collar workers", but rather, that post-WWII Japanese labor unions became 'mixed occupation unions with blue- and white-collar workers', and that as a result of campaigning for the abolition of status differences between blue- and white-collar workers, Japanese blue-collar workers "came to have the same character as white-collar workers". Furthermore, to explain the internal promotion system in terms of the formation of the mixed occupation unions makes little sense from the outset, because there were many examples of companies in which there was little difference in their internal promotions systems and which were of a similar size in the same industry, where one company opted for a mixed occupation union for blue- and white-collar workers while the other one set up unions divided according to occupation. For example, there were mixed occupation unions in mining companies (see Table 3); municipal transport unions were mixed in the Kantô area (Tôkyo) but unmixed in the Kansai region (Ôsaka). There are even examples of varying patterns of union organization among the different factories of the same company. A union's organizational form is not determined only by the structure of the labor market; it ought to be taken into account that there is also room for the subjective choice of the organization involved. Nevertheless, I am not arguing that the nature of a union is completely unaffected by the characteristics of the labor market. It is hard to deny that once a labor market is established, its particular nature imposes limits on unions and union members.

Why then, did so many post-war Japanese unions become mixed occupation unions, with both blue- and white-collar members?  To answer this question, the following three issues need to be addressed. First, why did white-collar workers, who had previously had nothing to do with the labor movement, now join unions?  Second, why did white-collar workers not choose to set up their own unions rather than join mixed occupation unions?  Third, why did blue-collar workers not oppose white-collar workers' admission to the union?

The first question presents little difficulty. The inflation that set in after the end of the war and the shortages of foodstuffs, as well as the damage caused by the aerial bombing exposed blue- and white-collar workers to the same privations. Lay-offs caused by factory closures and scaled-down plant affected both shop floor workers and office staff. In a situation where there seemed few bright prospects for their company, even those who managed to avoid being laid off had to experience great anxiety. Against such a background, when the Occupation authorities declared their intention to safeguard labor unions, many white-collar workers felt relieved and joined the labor movement.

The second question - why did white-collar workers not set up their own unions instead of seeking to create unions with blue-collar workers - has not previously been seen as much of an issue, but it is actually an important topic when thinking about post-war labor unions. This is because in labor movements around the world, there are "very few cases of successful attempts to organize manual workers and white-collar workers in the same union".*26  Again, in England and elsewhere, the main reason why white collar workers join the labor movement is to preserve the relative superiority of their social status and working conditions vis-a-vis those of blue-collar workers;*27 consequently, the type of union organization chosen in such cases is the white-collar-only union. Even in unions that include blue-collar workers, the white-collar group is limited to a clearly demarcated organization. In Japan, on the other hand, those who opted for mixed occupation unions were actually the white-collar workers.

Why was this? The reasons are not uniform. Two parties whose standpoints were completely contradictory ended up making the same choice. One was younger and middle-aged office staff, educated mainly at university, high school, or technical school, who joined the labor movement with a positive attitude. This was the generation educated at high school and university in the 1920s and 1930s, when the influence of Marxism was prevalent in educational establishments, and even if those workers were opposed to Marxism, they could not be indifferent to it. This generation had experienced the horrors of war firsthand, which was one reason why it produced so many people who desired fundamental change in Japanese society. From this point of view, unions were an organization that functioned as one wing of forces for social reform and revolution, so this generation naturally looked to the labor union as the one organization that transcended class interests.

On the other hand, owners, managers and white-collar staff in general realized that it would be difficult to avoid the formation of unions, and there were many who, fearing the rise of the communist movement, thought that unions would have to be induced to adopt a pliable and moderate stance. To this end, white-collar workers would have to take a leading role, and they too moved in the direction of setting up organizations that would include all the employees in an enterprise. Among the white-collar staff, there were some who were wary of collaborating with shop floor workers, but few of them took a leading role in forming unions, and most followed the general trend.

As we have already seen, some white-collar workers did create their own unions, but most of them were in sectors such as public service, in finance and education, where the proportion of 'blue-collar workers' was low. There were examples of white-collar employees forming their own unions in industries with a majority of blue-collar workers, but this was not so much a case of a strong feeling that they must form their own union in order to defend their own interests; rather, it was a passive acceptance of the fact that where blue-collar unions had already been formed, they too would just have to create their own, white-collar union.

The third question - why did blue-collar workers accept white-collar staff as fellow union members? - can likely be ascribed to the weakness of the tradition of 'laborism' among Japanese blue-collar workers. At the root of the labor movement as developed by blue-collar workers was the consciousness of 'them and us', classically represented in the British working class. Office workers, bosses and managers were 'them', and it was impossible to imagine forming any kind of joint organization with 'them'. Such a mentality was very foreign to Japanese workers, the overwhelming majority of whom wanted to cease being blue-collar workers as soon as possible. The feeling of pride in belonging to the working class was largely absent from Japanese blue-collar workers apart from a few activists in revolutionary movements, and deep down, even they wanted to send their children to a good university. If they had been told that "it's only natural that the children of workers should be workers themselves", they would have regarded such a sentiment as discriminatory.

However, this mentality of Japanese workers does not of itself completely suffice to explain why blue-collar workers accepted white-collar workers into unions, because there is a surprisingly large number of cases in which blue-collar workers opposed mixed unions and sought to form their own exclusively blue-collar unions. Many such cases of blue- and white-collar workers forming their own unions are to be found in the mining industry, and there were also many blue-collar-only unions in the shipbuilding industry and in municipal transport. Finally, even where mixed occupation unions were created, there were a number of cases in which, during the process of forming the union, there was strong opposition among the blue-collar workers to a joint body with office staff. For example, the Tokyo Public Transport Workers' Union finally opted to form a mixed occupation union, but the process of doing so was marked by fierce debate.*28  Why was there opposition to mixed occupation unions in these areas of industry? 

In coal mining and metal mining there were few opportunities for blue- and white-collar workers to work alongside each other; miners were paid according to a piece rate system while office staff and engineers were paid a monthly salary. Furthermore, while almost all employees lived in company accommodation, blue-collar workers tended to have to live in miners' barracks (nagaya), while office and technical staff had detached houses. Company accommodation facilities differed in terms of the number of rooms, external water supplies and toilets for common use or internal water supplies and toilets for private use. Accommodation for miners would be on the side of the valley with little sunlight, while that for white-collar staff would be in pleasant locations with good views. The two groups, while living on the same site, did not mix in social life; they lived in virtually different societies. When we look at the reverse side of the coin - situations where blue- and white-collar workers worked together at the same site, but in their daily lives either did not live near each other or else related to each other very little, as for example, in small or medium-sized companies more than in large ones, or in big cities rather than company towns - then, mixed occupation unions were more accepted.

Meanwhile, what public transport and the shipbuilding industry had in common was that during the inter-war period and even in the early years of the war, unions maintained a particular form of organization and remained under the leadership of labor activists who played a key role in the reconstruction of the union after the war. It goes without saying that in these industries white-collar staff had been 'the bosses' men'. Union activists who had had the experience of being under pressure by white-collar staff felt strongly that unions should only be for blue-collar workers. It is often assumed here that the fact that the number of people who took part in the labor movement before the war was small in absolute numbers was one reason why blue-collar workers accepted mixed occupation unions. However, the All-Japan Seamen's Union was an exception to this. From pre-war times, the seamen, having established a solid union organization, had been pre-eminent in terms of the numbers of their members who had experience of the labor movement, but after the war, senior crewmen and ordinary seamen joined the same organization. The background to this was that many executives of the Japan Seamen's Union - which had organized ordinary seamen - as staff of the Japan Merchant Marine Patriotic Group during the war and regular full-time members of the Shipping Transport Council, were in effect 'white-collarized'. It is said that of the 100 Council members of the reconstructed union (corresponding to Central Committee members in other unions), "the majority had never set foot on a ship since the war years".*29

3.    'Company Culture' in the Immediate Post-WWII Years and the Movement for the Democratization of Management

(1)   The Food Crisis and Employees' Dependence on the Company

I shall omit the subject of wage rises, which were the main focus of labor disputes after WWII, as they are well-known, and instead I shall concentrate on those points which have tended to be overlooked or ignored. One was the fact that during and after the war, employees, especially those of large companies, became to an unprecedented degree ever more dependent on their companies. When commodities were scarce through the war years and afterwards, the company became not only a place of work and production, but also a route for the supply of goods, an indispensable source of support for employees' livelihoods. That tendency was reinforced drastically by the severe food shortages of the immediate post-war era. In the company towns in the coal and metal mining industries where employees had always been dependent on company shops, food shortages were not seen as a problem of the responsibility of the State or of the individual but as a failure of the company to fulfil its duties to its employees. In the big cities, people managed to survive by relying on all kinds of connections and relationships. Their relatives in the countryside were a real source of support at such a time, and those who remained in the cities were people without such rural roots to return to. The company became for them their last recourse. Not only wages rises but the company's raw materials and products, machinery and land and how to use their various connections obtained through the company in order to guarantee their supply of food - all these became topics of urgent concern for them. The official history of Nippon Kokan (NKK = iron and steel tube manufacturer) Kawasaki Works Union describes the situation as follows:

At the Union Conference of May 17 (1946), a Food Crisis Alleviation Committee was set up that would supervise the mobilization of all the company's departments and functions in order to guarantee that employees would be supplied with food. ... The goal of the committee was to put all its energies into organizing employees' efforts to combat famine, and the method employed was for the committee, with the cooperation of the company Welfare Department, to manage the supply and distribution of foodstuffs, and take charge of all materials supplied to the Head of the Welfare Department; all foodstuffs and materials would be apportioned in accordance with the decisions of the Committee. ... First, permission to make salt had to be sought from the Manager of the Works, and each department began to make its own salt. After the end of the war, company production ceased, and people formed groups, collected sea water, and using the company's fuel, made salt, which they exchanged for food. To encourage the will to recommence production, those departments which produced more received more food in accordance with the amount of production. Next, when employees needed to take steps to guarantee their food supply, paid leave was granted. The Personnel Department Chief at the time, Mr. Horikiri, made the strong announcement that "Faced with this food crisis, although production is falling, we must get through the crisis, so first we must take time off and attend to our food supplies. For the purpose of barter, coke and other company materials are provided to each department. We must cultivate company land and get ourselves vegetables and sweet potatoes for food". Mr Horikiri's words gave everyone courage.*30

Here a spirit of mutual assistance gradually developed and spread; it was literally the appearance of a 'company-centered culture'.

(2)   The Demand for the Abolition of Status Differences at Work

The campaign for the 'democratization of management' was a significant part of the post-war labor union movement in Japan and it too was to play a part in the development of mixed occupation unions. According to the survey "Post-War Labor Unions" carried out by the Institute of Social Science, Tokyo University, "The concept of mixed occupation unions, practical aims such as the abolition of status differences, the democratization of management, the introduction of monthly salary systems, and participation in management are all associated with practical aims, and the mixed occupation union is regarded as the ideal campaigning organization to engage with these aims in a positive way and to realize them".*31

What then was the goal of the campaign for the democratization of management with which the post-war labor union movement was so pre-occupied, and what did the campaign achieve?  What should be noted here is that emphases in the goals of the campaign differed according to who was prominent in carrying it. For blue-collar workers, the democratization of management meant above all the abolition of status differences between themselves and white-collar workers. The background to this was the fact that the pecking order in companies was clearly determined according to one's educational attainment, and the way workers were treated depended on the large differences in status between levels in that pecking order. The situation varied between companies and changed with the passing of time, but in general, it was as follows:*32

At the top of the hierarchy were the 'senior employees' (sei-shain), all university or high-school graduates, who were mainly employed at head office, from where they were sent to companies' offices and factories throughout the country. Middle school graduates were known as 'junior employees' (jun-shain) and worked as clerical staff and in technical posts; many were restricted to factory work. Below them were office boys (kyûji) and casual laborers (yômuin) who had only completed elementary school, who were treated like blue-collar workers. Some of the junior employees were promoted to become senior employees, but promotion came only after a long wait, and there was a limit as to how high one could go. Employment as manual workers on the shop floor was restricted to those who finished or received elementary school education. There were a few cases of promotion from the shop floor to junior or senior employee status, but between shop floor and office staff there was normally a gulf that was not easily crossed. There were substantial differences in wages, bonuses, and promotion prospects, as well as differences in the provision of company accommodation between senior employees and those below them, and between white-collar staff and blue-collar workers. Especially important was the basic difference in the wages system. Many other forms of discrimination had their origin in their relation to this. Senior employees received an annual or monthly salary; however hard they worked, they could not earn overtime allowances, but on the other hand, their wages were not cut as soon as they failed to turn up for work. Junior employees, by contrast, were paid monthly salary on a daily basis, their wages were cut if they failed to report for work, and Sundays and holidays were not paid. Shop floor workers were on a piece rate system or daily or hourly wage system. Of course, those paid daily or by the hour had to have their attendance and hours closely checked; they had to use different entrances and exits from staff workers, and had to wear differently colored hats. At many factories they had to submit to body checks on leaving work at the end of the day to see if they had any company property on them.*33

However, these differences in status between blue- and white-collar workers existed not because Japan, as has often been claimed, was a pre-modern society; rather, they were behavioural elements which were transplanted to Japan along with the factory system. In other words, white-collar workers were paid annual or monthly salaries because it was hard to control their work on an hourly basis. By contrast, the most effective way of controlling shop floor workers was with a piece rate system or else according to their working hours. A similar form of discrimination existed in Britain, which was regarded in Japan as the model of a modern society. Hourly rates vs. monthly salaries, differing lengths of work periods, different times for coming to and leaving work, different numbers of holidays, different pension and bonus schemes, a week's notice for blue-collar workers and a month's notice for white-collar staff; it even went as far as both groups using different company gates, different car parks, canteens and even different toilets. This was not just the situation in the past during the period of the Industrial Revolution; such things were still common in Britain in 1965.*34

However, such differences in treatment were regarded by Japanese blue-collar workers as 'feudal, status-based discrimination', and they demanded their abolition. These demands differed according to the company in question, but many called for the revision of discriminatory job titles such as rômusha (laborer) and shokkô (factory hand); the same standards of salary, bonus, and allowance systems, working hours, holidays, days off, and retirement ages as white-collar workers; the same company gates for all workers, the abolition of body checks, and the use of common facilities for all workers.

The demand for the abolition of status differences was not something that suddenly emerged after WWII. It stemmed from Japanese workers' long-held and deep anger about such discriminatory practices. The campaign for the abolition of these practices did not begin after the war; it had frequently erupted in various forms in pre-war labor disputes.*35

However, although Japanese workers had a long history of resentment of discrimination, just because they had long demanded its abolition does not mean that they were egalitarians. Their demands were not based on western egalitarian values such as "all human beings are born with equal rights". In pre-war labor disputes workers had frequently called for the 'acknowledgement of humanness (jinkaku)', and in the post-war campaign for the democratization of management they had extolled 'fundamental human rights' and 'the equality of all human beings'. But these sentiments were only employed to justify their demands; the real demand of Japanese workers was 'treat [us] like human beings', in other words, the demand to be recognized as a responsible member of the company and to have their ability and effort rightly appraised.

Among the examples which provide clear evidence of this 'egalitarian sentiment' of Japanese workers*36 was the annual conference of the Fuji Electric Workers' Union in July 1948, which passed a resolution calling for 'the abolition of discrimination' in connection with a demand for higher wages. The resolution began with the words "as we are all human beings" and thus appeared to be a demand based on fundamental human rights. But what followed this was: "we should therefore be given an equal opportunity to participate in the operation of the company". It was not a demand based on fundamental human rights but a demand to be treated as a responsible person and a proper member of the group to which one belonged, in this case Fuji Electric Company. The resolution continued: "Discrimination should be strictly limited to character (jinkaku) and ability (nôryoku)."*37 The demand which began with "as we are all human beings" thus became a demand for 'strictly limited' discrimination. In the resolution there is no indication how 'character (jinkaku) ' was to be assessed nor even any doubt that 'character' can be assessed. What is clear, however, is that there existed a meritocratic idea - in the broadest sense - that wage discrimination on the basis of  'a strict assessment of character and ability' was natural.  This reflected the view that workers who could be respected as people of character and ability ought to receive the appropriate compensation.

This would prove to be the beginning of the later systematization of meritocratic competition between workers. 'Meritocracy in the broadest sense' means that the way the Japanese felt about 'egalitarianism' was that, while talent was emphasized, an 'egalitarianism' that did not include individual factors such as effort, cooperativeness, age, and family situation was held to be the 'wrong egalitarianism' or 'bad egalitarianism'. Consequently, when all employees were badly off in the immediate post-war years, in the name of 'democratization', an egalitarian spirit emerged which reflected social need in that ordinary employees who had larger families were given larger company living quarters than management staff. This was not a concept of egalitarianism based on human rights, but rather, treatment that took into account the individual circumstances of those who were members of the group.

This 'sense of egalitarianism' was a reason why labor unions were not very enthusiastic about the removal of gender-based discrimination. Male union members, and most female union members too, felt that a woman's real place was in the home, that work was just a temporary occupation until marriage and that therefore women were not regarded as full company employees. Unions' campaign goals for women focused on demands for monthly leave for women for the sake of support for mothers; they said nothing about systems for married leave or early retirement. Talk of male-female parity in wage levels was just talk; all-female unions practiced wage discrimination against their own clerical staff, claiming that "it is normal in Japanese society for women's wages to be lower than men's".*38  Teachers' unions, however, kept up a stubborn campaign for the abolition of gender-based wage differentials, and the fact they forced through social acceptance of this issue is of great significance.*39  That unions as a whole did not actively press for abolition of discrimination against women also reflects the fact that there were few female union executives. In 1949 there were 5,130,000 male, and 1,520,000 female union members, but whereas there were 10,019 male union executives, female executives numbered only 416.*40

(3)   Employees' Demands for Participation in Management

Whereas for blue-collar workers, the demand for democratization meant the abolition of status differences, white-collar workers showed a keen interest in participation in management. Where employees' participation in management took on real form was in the "production control struggle" [seusan kanri tôsô].*41 However, in the early post-war period one reason why this struggle spread was the often-cited 'sabotage of production' by management. In other words, when prices were rising steeply, rather than increasing expenditure to produce more goods, employers would 'sabotage' the restarting of production by diverting available resources into blackmarket channels in order to increase profits. In such situations strike action was ineffective; the tactic employed by the unions was to seek 'production control'. Certainly, in October 1945, a communist leader, Tokuda Kyûichi, who had just been released from prison, called for 'workers' control of production' to prevent 'capitalists' sabotage'.*42  However, was 'the sabotaging of production' really the main cause of the 'production control disputes'?

There is no evidence of any 'sabotage of production' at well-known companies such as the Yomiuri Shimbun, Keisei Electric Railway, Takahagi Coalmines and Tôhô Movies. In the newspaper, railway, coalmining and film industries 'sabotaging production' would have made little sense. Documental History of Japanese Labor Movement [Shiryô Rôdôundô-shi] Vol.1 includes 10 examples of 'production control' campaigns, but the only case relating to 'production sabotage' is that of the Shôda Works, where a factory closure was planned on account of an alleged shortage of capital. A clear conclusion is not yet possible as all the examples have yet to be examined, but at least in the early period of production control disputes, the main cause seems not to have been the 'sabotaging of production'.

Rather, the common thread in companies where disputes over the production control occurred in this early period was employees' lack of trust in, and anger towards management. Naturally, many employers at this time, particularly those at large enterprises, felt prospects were bleak and saw little way forward. The numerous problems subsequent to the shock of defeat - the loss of military supply commissions, the damage caused by bombing and the degradation of plant and equipment during the war, anxieties that equipment would be seized in reparations, the breakup of conglomerates in accordance with trust-busting legislation (Law 207 - the Elimination of the Excessive Concentration of Power), fears of purges of managerial posts, shortages of raw materials and capital - all made it well-nigh impossible for management to develop any clear goals. For their part, employees, both blue- and white-collar, who had entrusted their livelihoods to the company, were afraid for the company's future, and their lack of faith in their employers' feebleness only increased. The Yomiuri dispute showed such workers a way to address the problems that faced them. The dispute, which began with criticisms of employers' irresponsibility and developed into a takeover of production at the enterprise by the employees' union, taught many people that the way forward was to form a union and reconstruct companies themselves. Anger at a management that showed neither 'sincerity' nor any understanding of employees' demands*43 turned the dispute into a ferocious struggle that proved to be highly significant.

Another reason for the spread of "production control tactics" was that, unlike strikes, they were a form of struggle that had no adverse effects on company results. They thus easily gained the support of all employees, including that of managers who were concerned for the company's future, as well as the understanding of society at large. It is more likely that the reason why there were so many 'production control' disputes at this time lay in this direction rather than 'production sabotage'.

It is noteworthy that even in disputes that led to workers' takeover of production, workers' anger and distrust was directed at individual managers rather than at the capitalist system itself. Among the leaders of the campaign there were those who envisaged a 'workers' management' that would exclude the bosses, but this formed no part of the actual goals or prosecution of the campaign. On the contrary, although there were demands for new management, there was no desire for 'employees' associations' that excluded managers. What they were calling for was a company in which employees would be able to speak on an equal basis with managers. Ideally, this was felt to be a situation in which a company would be managed in such a way that all those working in it would be able to express themselves about any problems with the management of the company and contribute to the solving of those problems.*44  Management were forced to accept the principle of workers' participation in running companies because it was hard for them to object to the fact that democracy was the fundamental principle of post-WWII Japanese society.*45

Of course, while there was talk of all workers having a say in all the company's problems, it was clearly impossible for all employees to participate directly in the management of the company, so there was widespread recourse to management councils composed of representatives of labor and management.*46  It was very much the case that these councils differed greatly according to the prevailing power relationships in each company. In companies where the workers were strong, or in other words, where management were losing their ability to manage, the unions expressed their views on all kinds of management problems and took part in decision-making. In particular, they demanded, and usually obtained, prior consultation on all matters pertaining to personnel matters such as entering and leaving employment. It was only natural that unions should call for employment insurance, as large-scale lay-offs were occurring one after the other. But the unions did not stop at demands for employment insurance; they also demanded the election of executives, heads of department and section chiefs, union agreement to promotions and transfers, and purges of: anti-democratic groups, company executives involved in war crimes, and those who were engaged in the unfair distribution of supplies.*47 A reason for the demand for the union's agreement to transfers was that personnel were being moved around unfairly.*48

However, the reason why Japanese unions laid emphasis on personnel issues and made various declarations with regard to them was not simply because they were aiming to defend the union's right to a say in the matter, but because Japanese workers had always been keenly concerned about issues of promotion and status. For example, among the 22 'provisional demands' set out at the establishment of the preparatory committee of the National Railways Workers' Union were 1) the removal of incompetent or undemocratic managers, 2) opposition to monopolistic control of promotions by educational elite factions (notably, faction of the Law Department of Tokyo University), 3) the extension of training centers and the establishment of publicly funded educational institutes from middle school to university level (so that managers could be promoted from the shop floor).*49 The demand that directly affected their own interests - 'promote managers from the workplace' - was clear evidence of employees' strong concerns about the personnel of top management, as reflected in the call for 'regulation of the faction of the Law Department of Tokyo University'.

The white-collar group who maintained a grip on the leadership of the labor unions were concerned about the danger of company collapse in the wake of the surrender in 1945; they wanted people of real ability in management and managers who had risen from the workplace. In post-war unions, especially those in which white-collar workers were central, there were examples of members who had got into management. Characteristic of the opposition to participation in management of people from outside the company and of the demand for promotion from within were the conditions applied to the appointment of company executives by the Chiyoda Insurance Employees' Union in 1946, stipulating that 1) executives from outside the company were not to be allowed; 2) appointees should be people of distinguished character who understood the labor union movement; 3) they should be people with a positive attitude towards management, irrespective of academic attainment and age.*50

Also, at its annual conference held from the end of 1946 into the New Year, the Japan Fire and Marine Insurance Company Employees' Union "held discussions on the new management group with the aim of the dissolution of the wartime management structure and the democratization of the management, and then realized these aims".*51

Personnel issues were not of course the only matters addressed by labor-management councils. Rather, in many companies such councils gradually became arenas in which working conditions could be negotiated. As in America, at places where industry-wide unions were the focus of collective bargaining, working conditions were negotiated at the industry level, and problems at individual companies were normally discussed at factory councils. However, in Japan working conditions and other issues are dealt with by the parties involved themselves, and because working conditions were intimately related with the other issues, American-style demarcation between collective bargaining bodies and labor-management councils made little sense.

Under the slogan of 'democratization of management', Japanese unions' increasing participation in all kinds of issues relating to management meant that unions became indispensable to companies, and this naturally gave rise to the idea that companies should subsidize unions. They began to provide rooms for union offices free of charge and paid the lighting and heating costs, or else they would pay the wages of employees who did no work at all as company employees but who, as union executives, devoted themselves to union activities; they also began to pay traveling costs and allowances to union activists who went on trips on union business. Such provision was not only demanded by the unions but seen as natural by management. In addition, the practice of union meetings during working hours became widely established, and companies would deduct union dues from employees' salaries. Of course, workers received more such benefits in companies where unions were relatively strong. The stronger the union, the more numerous the company benefits, which tended to increase the union's dependence on the company, as is clearly shown in Table 5 below.

Table 5  Numbers and Proportion of Senior Union Officials June 1948

	Full-time Executives			Full-time Secretaries
	Total Nos. of officials	Nos. of officials paid by their companies	Proportion (%)	Total Nos. of officials	Nos. of officials paid by their companies	Proportion (%)
Japan Federation of Labor	948	722	76.2	1,136	657	57.8
Congress of Industrial Labor Unions of Japan	2,784	2,563	92.1	3,209	2,847	88.7
Other national organizations	3,014	2,417	80.2	4,640	2,847	61.4
Others	641	583	91.0	885	365	41.2
Independents	314	224	71.3	428	216	50.5
Total	7,701	6,509	84.5	10,298	6,932	67.3

The Congress of Industrial Labor Unions of Japan (Sanbetsu Kaigi), which was known for its militancy, was the organization the highest proportion of whose officials were paid by the companies for whom they worked. Post-War labor unions were not just company unions in terms of their internal structure; they were also company unions in the sense that they were financially dependent on the company. This was the case with many other 'militant' unions besides the Congress of Industrial Labor Unions of Japan. It was the Japan Federation of Labor (Sôdômei) rather, that kept something of a distance between itself and the company.

4.    The Consequences of Post-War Labor Activism - Conclusion

(1)   The Turning Point in the Post-War Labor Movement

Aided and encouraged by the policies of the Occupation authorities and helped by the weakness and confusion of employers, the Japanese labor union movement achieved rapid growth in the early post-war years. However, the 'honeymoon period' for this oddball 'marriage of convenience' between the State, the employers and the unions - something hardly seen before in the history of trade unionism - soon came to an end. There has been much debate as to exactly when this occurred. I shall try to clarify it with reference to Changes in Union Membership Numbers (Table 6), a basic indicator of union strength (Labor Ministry Secretariat Survey Section "40 years of Labor Statistics" p.532).

According to this, there is a large gap between 1949 and 1950. In June 1949 there was a substantial drop compared with the previous year; the consistent rise seen since the end of the war stopped. In the following year 1950, the number of unions fell by 5,544 and the number of union members declined by more than 880,000. This contraction was especially dramatic in the case of the Congress of Industrial Labor Unions of Japan, which had led the labor movement since the war. The Congress, which was formed with 1,630,000 members in 21 publicly recognized industries, and which even a Labor Ministry survey of June 1948 recorded had 4644 unions as members with a total union membership of 1,211,423, had fallen to 1,020,190 by June 1949, and in 1950 there was an even more drastic fall to 290,087 members.

There was no single cause of this contraction; it was the result of a number of factors that occurred simultaneously. Within the union movement itself, there was a deepening rift during this period. In the Congress, with its Communist Party union leadership, the campaign for the democratization of union management had spread from its head office secretariat to many industrial federations and local unions, and this led to splits. The Communist Party itself split in early 1950 over criticism of the Cominform and was badly hit by the Red Purge, which rapidly weakened its influence over the labor movement. On the other hand, unions were also hard hit by external factors affecting the labor movement. In line with the Nine Principles for Economic Stabilization, (late 1948) and the Dodge Line which followed them in March the following year, subsidies to companies were cut, and this opportunity was taken to effect large-scale lay-offs, which was termed 'company reorganization'. In the public sector, there was the removal of public workers' right to strike, in accordance with General MacArthur's Letter, and the reductions in administrative personnel enforced by the 'Law on the Total Number of Positions in Administrative Organs' of May 1949. In organizations and companies such as the National Railways, the Post Office, and Toshiba, which had provided the main body of the post-war labor movement, large-scale lay-offs were announced, and just when the unions were about to begin their campaigns against these, the gruesome Shimoyama, Mitaka and Matsukawa incidents occurred one after the other; the movement was forced onto the defensive. The unions suffered further major setback from the revision of the Trade Union Law. The policies of the Occupation authorities (SCAP) were decisive in all these developments. From the MacArthur Letter to the Nine Principles for Economic Stabilization, the Dodge Line, the revision of the Trade Union Law, and the Red Purge - all the main policy directives issued from SCAP.

Of these, the one that had the greatest direct effect on labor-management relations at the company level was the revision of the Trade Union Law. Restrictions were placed on the level to which companies could subsidize unions. The previous, rather loosely phrased legal stipulation to the effect that unions were not "principally bodies entitled to look to employers for support" was revised to read that unions were not "bodies entitled to support from employers for the provision of the running costs of the organization"; this clearly prohibited company payment for union executives and secretaries. The earlier law had excluded from membership of a union "those who represent the interests of employers", but this was now revised more tightly to exclude specifically management staff and some personnel department staff.

Encouraged by SCAP policies, employers adopted a tougher, more confrontational stance towards unions and sought to recover lost ground. There were three points at issue: 1) preservation of management rights, especially the right to reject or block a union's right to express itself on personnel matters  2) the annulment of management staff membership of unions  3) the cessation of financial support for unions. Before the passing of the Trade Union Law, SCAP and the Japanese government had given 'individual guidance' with regard to points two and three direct to employers and unions,*52  and the employers had been able to get their way with relative ease. At the end of 1949, half a year after the enactment of the revision of the Trade Union Law, according to the Ministry of Labor's "Survey into the Current State of Compliance with the Revised Trade Union Law", the number of unions that were 'excluding those representing [company] interests' and which had 'already stopped receiving subsidies from employers' exceeded 20,000, 91.0% and 93.5% of all unions respectively.*53

As for point one above - the maintenance of employers' rights - many fierce struggles continued between labor and management, but here too, employers were very much helped by the revised Trade Union Law, which now enabled one of the parties to a labor contract to terminate the agreement unilaterally once it had run its course. If the union side did not agree to proposals for a contract from the management side, negotiations could now be terminated by management simply declaring that 'no agreement' had been reached. This would change the character of post-war labor unions, but a detailed examination of that process lies beyond the scope of this paper.

(2)    The Outcome of the Campaign for the Democratization of Management

Companies responded in varying ways to the demand for the abolition of discriminatory barriers between blue-collar and white-collar workers, but those responses nevertheless showed a certain common tendency. First to be abolished were the more blatant forms of discrimination, for example, the removal of separate entrance and exit gates and body checks on leaving work. Discriminatory job titles such as shokkô (factory hand) and rômusha (laborer) were dropped and replaced, in communication with the outside world, by jugyôin and shain (employee). However, the difference between manual work and desk work could not in itself be changed, and de facto demarcation continued within the company in the use of job titles such as ginôshoku (technician), jimushoku (clerical worker), gijutsushoku (technical worker). Nor was there any great change in the practices of employing university graduates at head office, and those with a high school education or less at factories, or of basing the speed or upper limit of advancement in the company on academic achievement.

Industries and companies reacted differently to the main demand in the campaign for removal of discriminatory treatment, namely, an end to discrimination in the wages system. Reductions were made comparatively quickly in the gaps that had existed between levels of family allowance, travel expenses, accommodation allowance, or in the criteria for assessing bonus levels. However, it was not so easy to remove systematic differences between blue- and white-collar wage levels, because the wages system was intimately bound up with the efficiency of production. Nevertheless, the unification of blue- and white-collar income structures continued for years to be one of the unions' main demands. In time, some companies did effect changes, and even the slowest had by the end of the 1960s replaced 'monthly salary by daily basis' systems with monthly salary schemes and regular fixed period increments for all regular, full-time employees.

It is well-known that Japan differs from other countries in that the curve of blue-collar wage increases is similar to that of white-collar workers.*54  It would seem that this difference dates back to the early post-war period. With the adoption of the tactic of demanding the same fixed amount of pay rise for all plus additional extras for some sections or age groups (to be dependent on company circumstances) in the annual spring wage hike campaigns ( shuntô), the difference between blue- and white-collar wage levels reduced still further,*55  continuing the 'white-collarization' of blue-collar workers. The culture of modern Japanese blue-collar workers does indeed have its own characteristics compared to that of blue-collar workers in other countries, but rather than these characteristics being rooted in traditional Japanese culture as is often claimed, they are more directly the 'results' of campaigns by mixed occupation unions.

These changes were of course not only due to the achievements of the labor movement; they also owed something to employers' requirements in the management of labor. It was not so much a matter of compromise between the two sides as of collaboration. That monthly salary systems became the norm for blue-collar workers in the 1960s was also related to the fact that the number of youngsters going on to high school also rose at this time, and companies had to employ high school graduates as blue-collar workers as it was no longer possible for them to satisfy their need for labor from middle school graduates only. In other words, high school graduates were now to be found among both white and blue-collar employees, which meant that demarcation between them inevitably came to be seen as discrimination. As a result, during this period, along with the implementation of monthly salary systems went the call for 'transparency in personnel management', and an increasing number of companies harmonized blue- and white-collar promotion paths and abolished upper limits on advancement.*56

Although it would be going too far to say that the anger and resentment of blue-collar workers, which had for many years been the driving force behind the Japanese labor movement, completely abated after this series of changes, the latent energy of the movement was considerably weakened. At the same time, the realization of 'transparency in personnel management' greatly spurred competition between employees for better pay and status. As we have already noted, Japanese workers' 'sense of fairness' was of the kind that, while opposed to 'unjust discrimination' in matters of blue-collar status, considered an egalitarianism that did not take account of individuals' abilities and efforts to be 'wrongful egalitarianism'. When all employees were suffering in the period immediately after the war, their 'feeling of fairness' was satisfied by the formula 'living wage = seniority pay + family allowance', which put the needs of their livelihoods first. However, in the period of high economic growth, when everyone's minimum living standards had been adequately met, dissatisfaction with a system in which seniority pay and promotion based only on length of service began to emerge among employees, and there were even examples of unions demanding 'wages based on type of occupation and ability'.*57 Employers, who had originally been averse to a wages system based on seniority pay that did not take into account workers' individual efficiency, warmly welcomed these new demands that were coming from the employees themselves. Efficiency pay systems spread from this new collaboration between labor and management.

(3)    'Continuation' and 'Discontinuation' in the Post-WWII  Labor Union Movement

The task assigned to me by the editor of the book was to consider the period of the flowering of the labor union movement in the immediate post-war years in relation to 'continuation' and 'discontinuation' in later enterprise unionism. This paper has limited itself to consideration of the first period of the post-war labor union movement from 1945 until 1949. Obviously, a study such as this is insufficient to confirm the 'continuation' and 'discontinuation' between post-war unions and modern Japanese enterprise unionism. Half a century has already gone by, and if I were to list up all the events that have happened since - among them the anti-rationalization struggles of the 1950s, the campaign of the General Council of Trades Unions of Japan (Sôhyô) to defeat enterprise unionism by means of campaigns in individual companies, the high growth and spring labor offensives (shuntô) which changed Japanese society so much, the oil shock and the labor movement during the subsequent period of low growth - there would be many topics requiring investigation. Moreover, without a detailed examination of various counter-labor movement tactics and of labor management, it will be impossible to answer this question. Consequently, what follows can only be of a hypothetical and tentative nature. First, in very general terms, it is clear that the legal and structural framework of the post-war labor movement was incomparably more stable than that of the movement before the war or in the immediate post-war years. There was no later repetition of the severe 'discontinuation' or rupture that occurred after the war. Despite the major setback to the labor movement in the crucial year 1949, there were none of the disastrous situations that frequently occurred before the war. The main change was the generational shift in the leadership of the movement at the national level; there was no parallel significant change in unions at the company level. Enterprise union organization continued as usual, and there was no change either in the mixed occupation unions that combined both blue- and white-collar workers in the same organization. The movement's aims remained fixed on the situation within individual companies, and the operation of unions continued to be dependent on companies, even if that dependence became relatively slight. To that extent, it can be said that the main feature was 'continuation'; there was no decisive change that could be termed 'discontinuation'.

However, there were certainly some decisive changes in this period, not only in the organizational pattern of unions but in labor-management relations within the company, that is, in the reciprocal relations between unions and management. After the end of the war, with the organization of almost all company employees, unions had the power to put pressure on companies and in some cases even controlled them. Later, however, employers were able to regain complete authority for management. This was a major change that can indeed be termed 'discontinuation'. With regard to how the change occurred, this paper has not been able to do more than indicate the various factors relating to the initial turning point. I would like to research this subject further in the future.

The second turning point that followed this, from the 1950s to the first half of the 1960s, was the period of major disputes in leading companies in every sector of industry around the issue of rationalization. In this series of disputes employers put their efforts into excluding union leaders and activists who were deemed uncooperative and created alternative unions within their companies. The first to quit the original unions were white-collar workers and then foremen.*58  This process showed only too plainly the weak point of those 'fully unionized' workplaces i.e. those with mixed occupation unions, which had been so effective in the production contrl disputes. From this time on, the overwhelming majority of white-collar workers left the original unions, which had been confrontational in their stance towards management, and went for career options that would lead to management positions; some university graduates were even selected for direct promotion to management posts.*59  Those who persisted in an oppositional stance were either forced out of the company or isolated.

The changes at company level were not of course restricted to issues of management's authority. There were a number of changes in the labor union movement itself which cannot be overlooked. One was the formation of the alternative unions, but another important change that must be taken into account is that the workplace-based unions of the immediate post-war years transformed into the enterprise unions of the 1960s.*60
   There were also changes in the leadership profile of the union movement. Whereas the movement as a whole had been led in the 1940s by white-collar workers, and notably by university graduates, in the 1950s and subsequently, when the movement experienced a revival of activity, leadership at the company level in white-collar unions was increasingly taken over by ordinary clerical workers who were middle school graduates, and in mixed occupation unions it passed to blue-collar workers, especially foremen. In the 1960s, with the implementation of 'transparency in personnel management', some ordinary clerical workers and foremen were selected for management posts and became non-union members. The fact that there were always some key figures in the union movement who would be promoted to management, combined with the selection for promotion by management of some university-educated white-collar workers who were union members early in their career, had the effect of blurring the boundaries between management and union. A situation in which leading union members aimed at becoming non-union members and competed against their former union colleagues inevitably led to a hollowing out of the movement.*61

The background to the emergence of company-oriented labor unions was the fact that the post-WWII labor union movement had 'achieved' its goals. The main motivation for social movements - of which the labor movement was of course one - was resentment against discrimination. In that sense, the abatement of this long-held grudge and the fact that now, all workers had become 'white-collarized' in a union that consisted only of full-time regular employees made it difficult to escape from a 'company-centered' approach to unionism. From this time onwards, if there is any revival of union activity, in the present company-centered climate, it will most likely be a union movement in which those workers who feel that they are not being given their due recognition as 'proper members of the company' demand that they be treated 'like everyone else'.

However, the role that current unions can play is certainly not a minor one. For the 'ideal image of the company' that was agreed by both sides in the labor-management confrontations of the early post-war years still survives, that is, the idea that the company does not exist only for the benefit of capitalists (the stockholders) but rather, is composed of all those who work in it and should function for the benefit of all. Of course, this is only one side of the coin; on the other, there is always the notion that employees can enjoy a prosperous life precisely because the company does well, and that therefore it is only natural that each employee should make sacrifices for the sake of the company as a whole. Especially in times of management crisis, the coin is flipped over, and then the message tends no longer to be that the company is there for all the employees but that the employees are there for the company. To give substance, as far as is possible, to the ideal image of the company, labor unions are indispensable. The traditional struggle of the people of Japan has for centuries always been to turn their rulers' fine statements (tatemae) against them and force them to live up to the logic of what they profess.

Epilog

In the theme given to me by the editor, besides the topics I mentioned at the beginning, there was also the secondary topic of 'the relation between the wartime industrial patriotic (sanpô) movement and post-war labor unionism'. I made a start on preparing materials that would make possible an answer to this question, but space considerations have led me to omit it here.

Another theme that I was unable to take up in this paper was the social dimension to the activity of the labor movement. Obviously, in the immediate post-WWII period and then again from the Korean War until the Vietnam War in the 1970s, one face of the Japanese labor movement was that of a peace movement. That the General Council of Trade Unions of Japan (Sôhyô), despite having been trained by SCAP, developed from 'an ugly duckling into a swan' with its proclamation of the Four Peace Principles, despite having been trained by SCAP, is due to the fact that from its executive right through its membership, the organization was permeated by a strong desire for peace that was rooted in its experience of war. The participation of post-war unions in the peace movement was one factor that showed that while they may have taken on a clear 'enterprise unionist' character, they were no mere tools of the Establishment.

However, from the 1980s onwards the majority of members have belonged to a generation with no experience of war, and the social dimension of labor union activity has become thin. This is certainly a signal example of 'discontinuation' in the post-war union movement. I was unable to touch on this problem either, but I would like to do so in the future.

(1) 'Kigyôbetsu kumiai no rekishiteki haikei' (Enterprise Unionism - the Historical Background) (Ohara Institute for Social Research, Hôsei University, "Kenkyû Shiryô Geppô" (Monthly Research Bulletin) No. 305, March 1984. 'The Historical Characteristics of Labor Relations in Japan' (Shakai Seisaku Gakkai Nenpô (Annual Bulletin of the Society for the Study of Social Policy) Vol. 31 "The Characteristics of Labor Relations in Japan" Ochanomizu Shobô, 1987.

(2) 1947 only 'Rôdôkumiai Chôsa' (Survey on Trade Unions); from 1948 'Rôdôkumiai Kihon Chôsa' (Fundamental Survey on Trade Unions), from 1983 'Rôdôkumiai Kiso Chôsa' (Basic Survey of Trade Unions). When the actual state of trade unions was displayed, the number of 'individual labor union (tan'i rôdôkumiai)' is employed, but this is not appropriate. For data from 1953 the index number of 'unitary trade unions (tan'itsu rôdôkumiai)' should be used, because the number of 'individual labor union' includes branches and chapters as independent units in the case of large trade unions. In passing, the number of 'unitary trade unions' in 1953 was 18,228 and 'individual unions' was 30,129. Further, the number of union members in the case of individual labor unions was lower than the actual situation as the figures excluded those working at union head offices.

(3) Ministry of Labor, "Shiryoô Rôdôundô-shi" ("Documental History of Labor Movement) 1945-46", Rômu Gyôsei Kenkyûjo , 1951, p.999.

(4) Ed. Kanô Michiji, supervised by Ministry of Labor , "Nihon Rôdôkumiai Meikan (Directory of Japanese Trade Unions)", Kokusai Rôdô Hosei Kenkyûjo, 1948.

(5) Ed. Ashio Copper Mine Labor Union, "Ashio Dôzan Rôdô Undôshi" (The History of the Labor Movement at Ashio Copper Mine), 1958, p.196.

(6) As of June 1946, the All-Japan Seamen's Union had 26 branches and 61,419 members (Ministry of Labor, "Shiryô Rôdôundô-shi 1945-46" Rômu Gyôsei Kenkyûjo, 1951, P.406.

(7) Eds. Institute of Social Science, Tokyo University, "Sengo Rôdô Kumiai no Jittai" (An Investigation of Post-War Labor Unions), Nihon Hyôronsha, 1950. The main body of this report was later republished without the appended tables and with a new title as Ôkôchi Kazuo ed., "Rôdô Kumiai no Seisei to Soshiki" (The Formation and Organization of Trade Unions), Tokyo University Press, 1956.

(8) Estimated figures exist for a slightly later period. For example, with reference to 1956, Izumiya Hajime puts 'the number of organized salarymen' at 2,590,000 and the rate of union membership at 36% (Matsunari Yoshie, "Nihon no sarariiman" [Japan's Salarymen], Aoki Shoten, 1957), Solomon Levine estimates that of the total number of union members in 1960, 36% were white-collar workers (Solomon Levine, "Unionization of White-collar Employees in Japan", in Adolf Sturmthal (ed.), White-collar Trade Unions, Urbana & London, University of Illinois Press, 1966, pp.223-225.)

(9) Care needs to be taken in making comparisons between Japanese and foreign rates of union density. In many other countries the total of possible union members is normally used as the denominator, whereas Japanese statistics use the number of 'employees' that includes company executives and those such as Self Defense Forces personnel and police officers, whose right to organize is not recognized. Consequently, the actual rate of union density in Japan is always from 3% to 5% lower than the proper figures. For details, see Nimura Kazuo, 'Rôdôkumiai Soshikiritsu no Saikentô' (A Re-examination of Union Density Statistics) in "Journal of the Ohara Institute for Social Research", No. 330, March 1986.

(10) Koike Kazuo, "Chingin - Sono Riron to Genjô Bunseki" (Wages: An Analysis of Theory and Current Practice) Daiamondosha, 1966, table 30

(11) George S. Bain, The Growth of White-Collar Unionism, Oxford University Press, 1970, pp.22-27

(12) C.Wright Mills, trans. Suzuki Masataka, "White Collar", Sôgensha, 1957, p.280.

(13) Labor and Agriculture Correspondents' Group, "Rôdôundô Mita mama" (Witness to the Labor Movement) Vol. 1, Jijitsûshinsha, 1947. pp.15-18.

(14) Yamamoto Kiyoshi, "Yomiuri Shimbun Sôgi (1945-46)" (The Yomiuri Newspaper Dispute) Ochanomizu Shobô, 1978, p.23.

(15) Hirasawa Eiichi, "Sôgiya" (Strike Organizer), Ronsôsha, 1982, pp.22-23.

(16) Ôkôchi Kazuo ed., "Rôdô Kumiai no Seisei to Soshiki" (The Formation and Organization of Trade Unions), Tokyo University Press, 1956, ch.5 'The Character of Union Executives'.

(17) At Hokkaido Coal and Steamship Company the employers got heavily involved with the process of the formation of the union; in October 1945 Maeda Hajime, the Personnel Dept. manager got all the personnel dept. section chiefs in Hokkaido together to decide on a set of 'Guiding Principles for the Establishment of Labor Unions' and directed the formation of unions in the region (Hokutan rômubu hen, "Hokkaidô tankô kisen kabushikigaisha rô dô undô shi shikô , Dai isshuu"[Hokkaido Coal and Steamship Company Personnel Dept. ed., "The MSS of the history of the labor union activities within the Hokkaido Coal and Steamship Company" Vol. 1]. The management at the Ministry of Post and Communications too worked toward the formation of a single union on the basis of guiding principles. ("Zentei rôdô undôshi, Dai 1 kkan" [The History of the Post and Communications Union Movement" Vol. 1], pp.105-113). Many other union histories also record examples of management staff playing a key role in the formation of unions.

(18) Sumitomo Coal Company Office Workers' Federation ed., "Jûnen shi" (A Ten year History) 1957, p.6. However, local labor committees did not accept such cases, and ordered reforms; such cases did not therefore continue for very long. (Bank of Kobe Employees' Union, "Jûnen no Ayumi" (Ten Years On), 1958, p.73).

(19) Ôkôchi Kazuo developed this idea in numerous papers on various occasions. To give one example: 'Wagakuni ni okeru rôshi kankei no tokushitsu' (The Characteristics of Labor Relations in Japan) in "Rôshikankeiron no Shiteki Hatten" (The Historical Development of Labor Relations Theory"), Yûhikaku, 1972.

(20) The Institute of Social Science, Tokyo University ed., "Gendai Nihon Shakai 5 Kôzô" (Modern Japanese Society 5 - Structure) University of Tokyo Press, 1991, p.222.

(21) The style of union organization that post-war Japanese workers wanted was the 'company-based employees' union'. This fact was grasped early on in surveys conducted by researchers of the Institute of Social Science, Tokyo University. The statement of aims adopted by the Japan Federation of Labor Preparation Committee that was inaugurated in January 1946 reads as follows:

"While it is to be deplored that most employees want enterprise-based union organization, these are nevertheless the circumstances with which we are faced. It goes without saying that this illusion will have to be shattered."

Ministry of Labor "Documental History of Japanese Labor Movement 1945-46", Labor Administration Research Center, 1951, p.447). The leaders of the Japan Federation of Labor tried to organize on an industry-wide basis, but this was not wanted by the workers.

(22) Efren Cordova, Industrial Relations in Latin America, Praeger Publishers, New York, 1984, pp.30-31.

(23) Ross M. Martin, Trade Unionism:Purposes and Forms, Oxford, Clarendon Press, 1989, p.174.

(24) Steve Smith "Craft Consciousness, Class Consciousness: Petrograd 1917", History Workshop Journal No.11, Spring 1981.

(25) Koike Kazuo, "Shokuba no rôdô kumiai to sanka" (Unions and Participation at the Workplace), Tôyô Keizai Shinpôsha, 1977, p.240.

(26) R. Branpan ed., Hanami Tadashi trans. "Rôdômondai no Kokusai-hikaku" (Labor Issues in International Perspective), Nihon Rôdô Kyôkai, 1983, p.238.

(27) Richard Hyman & Robert Price (eds.), The New Working Class? White Collar Workers & Their Organizations, London, The Macmillan Press, 1983, p.163.

(28) All-Japan Municipal Transport Workers' Union ed. "Toshikô Sanjûnenshi" (30 Years of the Municipal Transport Workers' Union) Rôdô Junpôsha, 1978, pp.158-160.

(29) Sasaki Hiroshi, 'Sengo ni okeru nihon no kaijô rôdô undô' (The Post-War Labor Movement in Japan's Merchant Marine) "Shôsen daigaku kenkyû hôkoku" (Research Bulletin of the Merchant Marine University, No. 3B, Feb. 1953).

(30) Nippon Kôkan Kawasaki Works Union, "Jûnen no ayumi" (Ten Years On), pp.17-18.
The role of union president sometimes even included the provision of food and clothing (Taiheiyô Coalmines Union, "Rôdô Kumiaishi" (The History of the Union") 1955, p.23).

(31) Ôkôchi Kazuo ed., "Rôdôkumiai no Seisei to Soshiki" (The Formation and Organization of Trade Unions), Tokyo University Press, 1956, p.94.

(32) Ujihara Shôjirô, ' Sengo rôdô-shijô no henbô' (The Transformation of the Post-War Labor Market' in "Nihon no rôshi kankei" (Labor Relations in Japan) Tokyo University Press, 1961.

(33) For the pre-war state of demarcation between blue- and white-collar workers, see Sugayama Shinji, 'Senkanki koyô kankei no rôshoku hikaku - 'shûshin koyô no jittai'' (A Comparative Study of Blue- and White-Collar Employment Between the Wars - The State of 'Lifelong Employment') ("Shakai keizaishigaku" (Socio-economic History) Vol. 55 No. 4, Oct. 1989).

(34) George S. Bain, The Growth of White-Collar Unionism, Oxford University Press, 1970, pp.48-65.

(35) See n.1 above and Nimura Kazuo, "Ashio Bôdô no Shiteki Bunseki - Kôzan rôdôsha no shakaishi" (The Ashio Riot of 1907 - A Social History of Mining in Japan).

(36) In a noteworthy study 'Chingin taikei to rôshikankei - nihon no jôken' (Wage Structures and Labor Relations - The Case in Japan) in his book "Chingin no shakaikagaku - Nihon to igirisu" (The Social Science of Wages - Japan and Great Britain) Chuô Keizaisha, 1990, Ishida Mitsuo analyses Japanese workers' 'meritocratic' motivations using the words 'fairness' and 'equitableness'. But to use the words 'fairness' and 'equitableness' to express Japanese workers' attitudes is, in my view, rather too conceptual. Here I shall use the phrase 'sense of justice'.

(37) Fuji Electric Workers' Union, "Kumiai undôshi" (History of the Union) Vol. 1, p.98.

(38) Shiozawa Miyoko, "Hitamuki ni ikite" (A Single-minded Life) , Sôgensha, 1980, p.166.

(39) Japan Teachers' Union ed., "Nikkyôso Jûnenshi" (Ten Years of the Japan Teachers' Union), 1958, p.230. In deciding for a contractual stipulation that women teachers be appointed central committee members and in campaigning for the abolition of gender discrimination, the Japan Teachers' Union played a prominent, vanguard role.

(40) Statistical Survey Section, Ministry of Labor,"1949-nen rokugatsumatsu kihon chôsa rôdô kumiai chôsa hôkoku" (Fundamental Survey on Labor Unions Report, June 1949), 1950.

(41) For production control disputes, see Yamamoto Kiyoshi, "Sengo Kiki ni okeru Rôdô" (The Labor Movement in the Years of Post-War Crisis), Ochanomizu Shobô, 1977.

(42) Yamamoto Kiyoshi, op. cit. pp. 111-112.

(43) In labor negotiations in Japan, few words are used as frequently as the word 'sincerity'. See n.1 above, Nimura Kazuo, "'Kigyôbetsu kumiai no rekishiteki haikei" (The Historical Background to Enterprise Unionism).

(44) Kurita Ken ed., "Gendai Nihon no Rôshikankei - Kôritsusei no Baransu-shiito" (Labor Relations in Modern Japan - The Efficiency Balance Sheet) Rôdô kagaku kenkyûjo shuppanbu (The Institute of Labor Sciences Research), 1992, p. 29, and Kurita Ken, 'Sengo minshushugi to nihon rôshi kankei' (Post-War Democracy and Labor Relations in Japan) in Nagasu Kazuji ed., "Gendai shihonshugi to tagenshakai" (Modern Capitalism and Pluralistic Society) Nihon Hyôronsha, 1979.

(45) Among employers in their organization Keizai Dôyûkai (Japan Association of Corporate Executives), some acquiesced in production controls and urged that "all those whose work related to production be involved in the management of production".
   Yamamoto Kiyoshi, ''Sangyô saiken' to sho-seijishutai' (''Industrial Reconstruction' and Political Actors' in Institute of Social Science, Tokyo University, "Sengo Kaikaku 5 Rôdô Kaikaku" (Post-War Reforms 5 Labor Reforms) Tokyo University Press, 1974.
   Shinobu Seizaburô, "Sengo Nihon Seijishi" (The Political History of Post-War Japan)" Keisô Shobô, 1966, pp. 412-415.

(46) Various studies have been made of labor-management councils (keiei kyôgikai); see Kurita Ken's paper (n.44 above),
   Nakajima Masamichi, 'Sengo gekidôki no 'shita kara no keieikyôgikai' shisô' ('The Philosophy of Bottom-up Labor-Management Councils' in the Post-War Period of Rapid Change) in Shimizu Shinzô ed., "Sengo rôdô kumiai undôshiron" (Essays on The Post-War Labor Union Movement) Nihon Hyôronsha, 1982;
   Endô Kôshi, "Nihon senryô to rôshi kankei seisaku no seiritsu" (The Occupation of Japan and the Formation of Labor Relations Policy) Tokyo University Press, 1989; Nishinarita Yutaka, 'Senryôki nihon no rôshi kankei' (Labor Relations in Occupied Japan) in Nakamura Masanori ed., "Nihon no kindai to shihonshugi - kokusaika to chiiki" (Capitalism and Modern Japan - Internationalization and Local Regions) Tokyo University Press, 1992.

(47) Ministry of Labor, "Shiryô rôdôundô-shi Shôwa 20-21-nen" (Documental History of Japanese Labor Movement 1945-46) : records of demands made in labor disputes.

(48) For example, Seika (Sumitomo) Mining Company tried to send Fujita Wakao, Chairman of Seika Kôgyô Karatsu Mine Office Workers' Union to their mine in northern Japan, which led to a dispute (Seika Kôgyô Federation of Office Workers' Unions eds., "Rengôkaishi" (History of the Federation) 1951. Fujita Wakao, "Sarariiman no shisô to seikatsu" (The Salaryman's Philosophy and his Life) (Tôyô keizai shinpôsha, 1959) is a most interesting account of a single labor issue by a labor researcher who was himself a participant in it.

(49) Ministry of Labor, "Shiryô Rôdôundô-shi Shôwa 20-21-nen" (Documental History of Japanese Labor Movement 1945-46) p. 410.
   For the role of the National Railways Union in personnel management and regulations issues, see Aoki Masahisa, 'Kokutetsu rôdô undô (1945-49) - Jinji kanri no 'minshuka' o chûshin ni' (The National Railway Labor Movement (1945-49) - The 'Democratization' of Personnel Management) in Rôdô sôgishi kenkyûkai (Labor Disputes Historical Research Group) eds., "Nihon no rôdô sôgi (1945-80)" (Labor Disputes in Japan 1945-80) Tokyo University Press, 1991.

(50) Chiyoda seimei gaimu jugyôin kumiai (Chiyoda Life Insurance Sales Representatives' Union), "Jûnenshi" (The First Ten Years) 1962, p.13.

(51) All-Japan Songai Hoken Union, "Zen Sonpo no Ayumi" (The Story of the All-Japan Indemnity Insurance Union) 1962, p.69

(52) In December 1948, preceding the revision of the Trade Union Law, the government issued a Labor Vice-ministerial Ordinance, according to which every prefectural labor section was to receive assistance from the local military administration, and before the revision of the law, indicated that in accordance with the main thrust of the revision, union regulations and labor agreements would be amended. Specifically, companies' financial support for unions was to be terminated, union members' qualifications were to be clearly regulated, and management staff, personnel staff concerned with allowances and bonuses, and security staff were to be non-unionized. ("Shiryô rôdôundô-shi Shôwa 23-nen"(Documental History of Japanese Labor Movement 1948), Rômugyosei Kenkûjo, 1952, pp.1117-1122).

(53) Ministry of Labor, "Rôdô-gyôsei-shi:Sengo no Rôdô-gyôsei"(The History of Labor Administration: Post-War Labor Administration), Rôdô Hôrei Kyôkai, 1969, pp.469-471.

(54) Koike Kazuo, "Nihon no Jukuren"(Skills Training in Japan), Yûhikaku, 1981.

(55) "The characteristic of wages in Japan compared to those in the West is that wage differentials are relatively slight" (Koike Kazuo, op. cit. p.71)

(56) Orii Hyûga, "Rômu kanri nijûnen" (20 Years as a Personnel Manager) Tôyô keizai shinpôsha, 1973. Kumazawa Makoto, 'Shokuba shakai no sengoshi' (The Workplace in the Post-War Period) in "Shin-hen Nihon no rôdôsha zô" (New Edition The Image of the Japanese Worker) Chikuma Gakugei Bunko, 1993). This historical background also accounts for the fact that many Japanese companies that have located in the West have promoted employment insurance together with the single status system as the selling point of Japanese-style management.

(57) Hashimoto Shûichi, 'Nôryokushugi to chingin taikei' (Meritocracy and the Wage System) in Kurita Ken ed., "Gendai Nihon no Rôshi-kankei : Kôritsusei no Baransu-shiito" (Labor Relations in Modern Japan - The Efficiency Balance Sheet) Rôdô kagaku kenkyû shuppan-bu, 1992).

(58) Takeda Makoto, "Ôji Seishi Sôgi (1957-60) - Nihonteki rôshi kankei kakuritsu o meguru rôshi kôsô" (The Ôji Paper Mill Dispute :1957-60 - A Key Labor Dispute in the Formation of Japanese-style Labor Relations), Taga Shuppan, 1992, Rôdô Sôgishi Kenkyûkai (Labor Disputes Research Group), "Nihon no rôdô sôgi" (Labor Disputes in Japan) Tokyo University Press, 1991; see papers by Matsuzaki Tadashi and Hirai Yôichi.

(59) According to a 1981 survey of the 313 member companies of Keidanren (then the Japan Federation of Economic Organizations; today [2007] the Japan Business Federation), as many as 232 companies (74.1%) had executives who had formerly served as members of union executive committees.

(60) Takagi Ikurô, 'Nihon no kigyôbetsu kumiai to rôdô seisaku' (Japanese Enterprise Unions and Labor Policy) in "Konnichi no shihonshugi 7 nihon shihonshugi to rôdô kaikyû" (Capitalism Today 7 Japanese Capitalism and The Working Class"), Otsuki Shoten, 1982.

(61) Among today's trade unions, the fact that a philosophy of 'company-ism' is weaker in unions in the public sector, where the right to strike is not recognized, than in the private sector, is not unrelated to the fact that there is no competition between public sector companies, and there is no 'career cadre group' members within public sector unions.