Lives of Labor

Work in a Maturing Industrial Society

PETER N. STEARNS

For Patricia, who helped the whole thing

©1975

ISBN 0-8419-0192-9

CONTENTS

Tables                                                                                             vii

Introduction                                                                                      1

Part I: The Framework for Labor

1.  The Evolution of the Labor Force                                   19

2.  Job Choice in a Maturing Industrial Economy                  45

3.  Unemployment                                                              85

Conclusion                                                                       114

Part II: Forces for Change

4.  Technology                                                                  121

5.  Relationships with Employers                                       148

6.  A Frenzied Pace                                                            193

Conclusion                                                                       229

Part III: Reactions and Compensations

Introduction                                                                    239

7.  Reducing Work Time                                                    241

8.  Off the Job: Family and Consumption Patterns              269

9.  Protest                                                                         300

Part IV: Towards the Future of the Working Class

10. Conclusion             335

ACKNOWLEDGEMENTS

I would like to thank the American Philosophical Society and the Social Science Research Council for grants that supported much of the research for this book. The Research Councils of the University of Chicago and Rutgers University provided valuable assistance as well, as did the Computer Laboratory of Rutgers University. Research on workers' aging was part of a larger project supported by the John Simon Guggenheim Memorial Foundation.

A number of people helped in the chore of pulling together statistical data: Lawrence Gurman, Robyn Levine, Roseann Erck, and Howard Zehr. For some calculations I was fortunate in having a domestic putting-out system, and my thanks go to Deborah Stearns, Nancy Uttrachi, and Duncan Stearns. My typist, Margaret Harmon, was consistently helpful. I am, finally, grateful to Patricia Branca Uttrachi for her careful reading of the manuscript and her many suggestions.

TABLES

Chapter 1 :

I Evaluation of the Industrial Labor Force

(Manufacturing and Transport Occupations)                         24

II Patterns of Industrial Growth: Summary and

Comparative Tables                                                             26

III    Women in Industry: Summary and Comparative Tables 30

— IV Employment of Children                                                     34

V    Domestic Manufacturing Workers and Employers                  37

Chapter 3:

I Annual Unemployment Rates                                              91

II Trends in Unemployment                                                  102

Chapter 5:

I Differences in Company Size                                              155

II Company Size: Comparative Suggestions                            156

HI Evolution of Company Size                                                157

IV    Evolution of Company Size: Summary                               161

V    Strikes for Dismissals (or More General Personal Issues) 184

Chapter 9:

I Demand Rate                                                                    305

II Industrial Variations in Leading Demands, 1899—1914 306

III    Evolution of Strike Rate                                                    314

IV    Industrial Strike Rates, 1899-1914                                   316

Part IV:

I Characteristics of Industrial Categories                                341

Appendix: Chapter 1

I Distortions in Census Data: Employed Personnel

Alone Compared to General Totals                                     358

II Patterns of Industrial Growth                                             360

HI Women in Industry                                                            365

IV Married Women in Industry                                                369

Appendix: Chapter 9

I Britain: Evolution of Strikes, 1899--1914                          372

II Britain: Rate of Demands and Results in Key Industries,

1899-1914                                                                      374

III    Britain: Rate of Demands and Results in Key Industries,

over Time                                                                         375

IV    France: Evolution of Strike Rate, Demands, Results           377

V France: Industrial Incidence of Strikes                              379

VI France: Evolution of the Rate of Strike Failures                 380

VII Germany: Evolution of Strike Rates and Results                381

VIII Germany: Incidence of Demands and Results, by

Industry, 1901-1913                                                        383

IX Germany: Losses per Demand                                            385

X Germany: Lockouts, 1901-1913                                       385

XI Germany: Trade Union Statistics                                       386

XII Germany: Trade Unions, Demands and Results                  388

XIII    Germany: Trade Unions, Patterns in Selected Industries, 1901-1913                                                                      389

XIV    Belgium: Incidence of Strikes and Strike Demands             390

XV Belgium: Patterns in Selected Industries, 1899-1913         392

XVI Belgium: Demands and Results, 1899-1913: Strike

Losses                                                                              394

INTRODUCTION

This is a study of work, and workers' reaction to work, from the 1890s until World War I, in four countries: Britain, Belgium, Germany and France. The topic is obviously not unprecedented. Earlier studies of industrialization commonly discussed the outlines of the work situation, stressing such elements as hours of work, safety conditions, and employment of women and children.¹ More recently, scholars such as E. P. Thompson have attempted to get at the nature of the industrial work experience, the psychic as well as material aspects that would strike a worker entering factory employment for the first time.²

Studies of this sort are immensely valuable and form an important precedent for the present effort. But with rare exceptions they have focused on the first stage of industrialization, when it seems reasonable to assume that the encounter between traditional concepts of work and the pressure of the factories was most abrasive. Relatedly, the emphasis has been on difficulties, of adaptation, on the compulsion under which workers were placed to convert them to effective instruments of industrial labor.³

Far less attention has been given to the more mature period of industrialization that opened up after the 1870s. Some outlines are familiar, of course. The growing size of the unit of employment presumably heightened impersonality at work and encouraged a more rigorous supervision of the labor force. New technology intruded even into the artisanal trades, converting some, such as shoe manufacture, into factory operations and reducing or altering skill levels in others. In all branches of manufacturing, this was a period of speed-u p, in which human arrangements were changed, beyond the requirements of machines themselves, in an effort to increase productivity.⁴ At the same time, some of the problems that have preoccupied labor historians during the earlier period now had a reduced impact; notably, the standard of living rose, though its extent can be debated, particularly for the years of inflation that opened up after 1900. Yet with rare exceptions the details of this framework have not been filled in. Economic historians have dealt only fleetingly with the workers, while labor historians have been consumed by a desire to trace the intricacies of rising socialist parties and trade unions.

It seems increasingly obvious that social historians must shift some of their attention from the early industrial period to more recent times. Fascination with the first shock of change has obscured the treatment of the more immediate background to our own age. At the time when concern for the quality of work is spreading once more, it is desirable to know something of the evolution of labor as industry matured. This study is not just a somewhat more proximate slice of a familiar subject. The evolution of industrial work was not steady during the nineteenth century. Many workers were brought into an essentially industrial work system, not in the first phase of industrialization but only at the century's end. Others had been forced into new kinds of labor much earlier, particularly in the factories, but had developed systems of adaptation, in the work place itself, that were challenged severely after the 1880s. Both aspects of this new stage of adaptation form vital parts of this study.

Among the workers newly encountering an industrial work system, there were three major types: new migrants, traditional unskilled groups, and artisans. Newcomers were constantly pouring into the factories. Particularly in Germany and Trance, this frequently reproduced the interaction between traditional work patterns and a new and intensiye work system; the contrast may indeed have been more vivid than ever before, given the sophistication of the technology and management that could now be applied even to new workers.⁵ Ironically, this was not the group of workers that felt the greatest strain at work. New, unskilled workers from an essentially traditional background, like Silesian miners employed in the Ruhr, could manifest far less anxiety about their industrial setting than their colleagues who had greater experience on the job. Among other things, they were accustomed to long hours on the job. Agricultural workers in East Prussia, for example, still worked from five in the morning until sundown, with only two hours' respite.⁶ Many immigrants felt tension in the new setting, but their pattern of life frequently allowed immediate redress, for, sometimes at their own choice, sometimes by compulsion, they alternated periods of work and leisure. In essence, by changing jobs and moving they restored some of the aspects of seasonality to which they had been accustomed. Thus large groups of workers, even in advanced industries such as metallurgy, fought with some success to preserve a traditional concept of work.

At least as interesting to our purposes are workers touched by new expectations or prodded into them by radical changes in the work setting. These could include immigrants, like the Dorset agricultural

laborers who poured into the mines of Wales, attracted by clear notions of rising pay rather than being forced out of the countryside by sheer population pressure.⁷ They embrace many unskilled workers who become part of the working class toward the end of the nineteenth century precisely because their traditional systems of labor are challenged. Dockers in Hamburg found their work becoming newly-intense during the 1890s. They previously had maintained solid ties with surrounding villages, but now they had either to abandon the countryside or suffer absences from their families for days at a time. For the docks were now mammoth in their extent. And the pace of work picked up, with new loading machines and more rigorous supervision.⁸ Their reaction was to mount an unprecedented protest effort, asking for regular and reduced hours. But this too posed problems, for it violated traditional work patterns that called for periodic, rather than regular, physical output.⁹ The next decades, and not only in Germany, would see recurrent tension between pressures from the outside, including a new technology, a traditional willingness to accept long but irregular periods of labor, and a new system of regular, limited, but intense work which industrialization and the labor movement both had to encourage. Similar tensions crop up in other areas of transportation work, including even the railroads, and among unskilled construction and factory workers. We are aware of them through movements for reduced hours, such as the London dock strike of 1889 or Scottish railway agitation during the 1890s. We can, by studying work itself, go beneath these important movements to ask why they proliferate in this period and not before — for the number of workers who suddenly declare that twelve- or fourteen-hou r work days have become inappropriate, despite their long standing in the trade, is astonishing — and what they mean in terms of the Larger work experience.

Journeymen, such as bakers or skilled construction workers, had clearly undergone intense stress in the first phase of industrialization. The work systems we find them defending around 1900 were not purely traditional. But part of the furor that has been eloquently described by labor historians of the earlier period involved trades that were dying; those that survived maintained strong traces of traditional protective devices. Hence English craft unions — in construction, furniture, hat manufacture, brass, — had long regulated hours and amount of production, the number of apprentices per worker, and so on.¹⁰ Artisans maintained traditional residential patterns, either clustering in the center city, as Charles Booth found to be the case

among London bookbinders and shoemakers,¹¹ or, as often in Germany and Belgium, commuting from surrounding villages. In either case, work had been their life; center-city artisans lived, drank, and played around their work experience, intermarrying with their own.¹² Artisans from the villages alternated rural and urban work, with similarly little time for activities unrelated to the work experience. All of this was now challenged, most obviously by intrusions from the outside such as new, mechanized equipment that spread for the first time to virtually every artisanal trade, but also by new expectations on the part of artisans themselves, as they sought more time for family life and new kinds of residential patterns.¹³

In terms of exposure to machines and to more regular, routinized work methods the working class was becoming more united by 1900 in all the advanced industrial nations. The reactions of key groups were not the same, for the ability and desire to develop compensations off the job varied as did the extent of pressure on traditional patterns. But there was something of a common framework, which carried over into protest efforts. For, unlike the first period in which artisans struggled to end the industrial revolution, reaction to the new work patterns now took shape within the industrial system. Unskilled laborers, a bit newer to the game, occasionally ventured an anachronistic protest against the whole system — as in sporadic acts of Luddism — but they too largely worked within the order. Protest in a broad sense encouraged further accommodation to industrial work, and while this was not always to the workers' disadvantage it did leave many with a sense of dismay.

Factory workers had undergone a steadier pace of change in technology and work organization since the early industrial period, but for those with lengthy factory experience the nature of work altered more dramatically during the decades around the turn of the century than at any point since the installation of factories themselves. Skilled workers in the factories were particularly challenged, for their ties with an artisanal mode of production had by no means been shattered. This applied to new workers in German machine-building plants from a small-town craft background or to shoe mounters everywhere who tried to defend traditional skills in a new factory setting, as well as to British turners or fitters who faced new challenges from automatic machinery and the semi-skilled who could operate it or to the iron moulders who still insisted on a seven-year apprenticeship. The erosion of proudly-won skills was a common theme throughout the advanced industrial countries. English engineers resisted the imposition of American production systems, claiming that English machine skills were the best

in the world.¹⁴ British shipbuilders maintained that certain kinds of work were 'theirs by custom.'¹⁵ Workers in textiles and mining, where technological innovations were less marked, nevertheless faced steady pressure to increase production, ranging from persistent efforts to increase the number of looms per worker, in weaving, to even more general efforts to use a piece rate to heighten work intensity.¹⁶ Traditional respites from work, already challenged by early industrialization, were now brought under rigorous scrutiny. Welsh mine-owners tried to abolish 'Mabon's Day,' an annual holiday, while British mine companies generally pressed workers to end their tradition of 'playing' on Saturday.¹⁷ In Belgian as in British mines, customary meal breaks were reduced, while in the factories foremen's supervision became more intensive. Even British textile workers, for all their factory experience, encountered a new level of work; previously able to nap periodically on the job, they now found themselves driven to a new level of constant labor which could, as with miners, eat into traditional respites.¹⁸

There is no need to argue that changes in work were more disruptive in this period than in the early industrial revolution, though in certain industries this was clearly the case. What we are talking about is the creation of the modern work system and, to an important extent, the modern worker himself. The early industrial period was marked by the passage of traditional types of workers and the reaction against the whole advent of modernity. It was full of drama, but it constituted more the death of an old working class than the making of a new one. The cluster of changes in work systems that developed toward the end of the nineteenth century in the most advanced industrial countries was accompanied by other developments that set the workers apart from their predecessors, without breaking with traditions entirely. This was the period when education became general (though it lagged in Belgium, with important effects on the reaction to work). Everywhere workers' birth rates began a rapid decline. Protest movements of course developed a real mass base for the first time. All of this forms something of a package, with the work experience a possible key.

The 1890s admittedly constitute only a convenient point at which to embark on this study. The breakdown of artisanal work patterns can be traced in London to the late 1870s, while for English textile workers the pressure to increase the pace of work was more intense in the 1880s than in the 1890s. But by the 1890s the conversion to new work systems was fairly general, across all four countries considered. Discussion of the next twenty years covers a manageable cross-section

of the reaction to the new forms of work, as well as the steady proliferation of new pressures. The introduction, at the end of the period, of American-devised efficiency systems, particularly in metal-working and machine-building, reveals the connections between this period and contemporary manufacturing work. The World War extended the changes that had already been outlined, but there is no claim that it marked a decisive break. The War disrupted the data base for a study of this sort, eliminating some useful sources such as normal strike records and many trade union publications; but again this should not be taken as a judgement that the mature of work dramatically changed. For the whole point in examination of the history of labor in the two decades before World War I is to assess the origins of a new but durable work experience.

In the early 1900s a survey by Adolf Levenstein revealed the following reactions to work: 60 percent of the miners reported no joy in work, while 15 percent found pleasure in their jobs, and 18 percent were indifferent; among textile workers the rates were 75, 7, and 14 percent, while among metal workers they were 57, 17, and 17 percent. Alfred Williams, a worker in a locomotive plant in Swindon, discussed the impact of new machines and the piece rate only a few years later. He found his colleagues exhausted, with no energy to enjoy life off the job; but they felt they had to go on working, for there was no alternative. One worker, whose wife locked him in his room to enforce a rest, escaped through the window to get "back to his job.¹⁹ Vignettes abound: Flemish miners refusing to join a strike because they found it normal to continue working.²⁰ The weaver who reported to Levenstein, against the majority, that, 'My daily work really gives me pleasure because I don't do it slavishly, but like all my jobs, with love and awareness.'²¹

This study deals both with adaptation and with inability to adapt. Its essence is simple, however complex and ambiguous the conclusions may be. With a generation or more of industrial work behind them, did workers find the changing work experience m ore or less onerous than before? This is not, despite Levenstein's statistics and Williams' impression, merely a tale of woe. Tempting as it is to seize on the frequent expressions of torment, it would distort the capabilities of workers themselves if we ignored the means by which they could come to terms with hardship. The important groups of workers who did maintain means of enjoying their jobs should not be ignored. Why, even in Levenstein's figures, the differences in work enjoyment among the three industries considered? For the workers who found decreasing

pleasure in work, there were some means of compensation, notably in new enjoyments off the job and the development of new protest outlets.

Both of these latter factors will demand attention after the work experience itself is assessed. In one of the only efforts to date to discuss the nature of work in a mature phase of industrialization, Eric Hobsbawm posits the increasing conversion to a market mentality toward work, which accepts changing technology and organization — a non-traditional work experience — in exchange for increasing rewards.²² Several studies in the mid-twentieth century confirm this notion of development, where factory workers view their jobs mainly as instruments and live for the time they have away from work.²³ Yet we cannot move automatically from the conversion Hobsbawm describes to the contemporary materials, and in some countries such as the United States many workers continue to express pleasure in work itself.²⁴ We do not know in what order various categories of workers effected the instrumental conversion and why. In a previous essay I have tried to portray important groups of workers who were subject to non-traditional work experiences and yet incapable of a full transition to an instrumental view.²⁵ At the other extreme, as Hobsba wm himself notes, we must not neglect the workers who found interest in the new technology and pace of work. Above all, as we realize how much of the pre-industrial work mentality had survived, even in Britain, into the early twentieth century, a careful examination of the ambiguities of the transformation to a new view of work may suggest complexities even in the approach of contemporary workers toward their jobs.

That protest outlets increased during the turn-of-the-century period is obvious. Much of the source material on which this study relies comes from protest situations. And conclusions about the nature of work in a maturing industrial society undoubtedly illuminate various conventional topics in labor history. Some insight into the reasons for different kinds of socialism, for example, emerges from examination of various work settings. Union policies and strike patterns directly relate to problems at work. We shall return to an assessment of protest in the concluding section. But it must be stressed that the study is not designed solely as an explanation of protest. We are fully as interested in, if often less knowledgeable about, workers who did not protest as in those who did. Unrest is treated as it flowed from the work situation, relieving some of its tensions and exacerbating others, and not, save in the concluding remarks, for its own sake.

And this is not merely a history with the politics left out. The study

has not been based extensively on the kind of cultural materials — poems, songs, and so on — that have gained such favor after the work of E. P. Thompson. This partly reflects lack of time on the author's part, given the extensive geographical area investigated, and to this extent it represents a weakness in the data base. But too much reliance on such evidence is risky in any event, because of the familiar problem of representativeness. The most soulful German miners, the young intellectuals in the Welsh mines who went to Ruskin College, the French worker poets may have articulated more general concerns, but they were unusual types. Some were demonstrably more upset about their work than their fellows were. Max Lotz, who described his horror at work in the mines so vividly, talked to be sure of the 'secret rage' that animated his colleagues, but he also described their phlegmatic character and their inability to feel, much less to articulate, his kind of torment: 'The miners are commonly made of thick, grainy wood.'²⁶ If for the early period it is chancy to qualify the evidence articulate workers produced, given the lack of abundant alternative materials, for this period it is essential. The dramatic phrase can easily obscure the common experience. This is not, as will be seen, a declaration that most workers were contented clods. It amounted to five percent of the total in some instances (as suggested by Florence Bell's study of the number of workers in Middlesbrough who read widely), but often to less than this; in Germany, where most workers did not even use union libraries, less than ten percent of all books borrowed from their libraries was other than escapist fiction.²⁷ The perceptions of the worker intellectuals were distinctive at least in degree, and the culture they produced has therefore not been given special place.

Hence there is nothing on workers' "theater and little on the labor press (though the latter has been extensively consulted). When we turn to the question of recreational outlets, as they developed as a complement to or compensation for the work experience, this may be seen as an unfortunate omission. There is a bit more on general popular culture, particularly sports, for a discussion of changing work patterns necessitates conclusions on the kind of leisure ethic workers could develop. But there is no pretense at a full survey of the relevant materials. Treatment of family developments falls into much the same category, for while important conclusion s can be suggested not enough is yet known to claim definitive coverage. This was a crucial period in working-class family life, as the reduction of the birth rate suggests, and the interrelationship of family patterns with work is obvious.

Twentieth-century studies that claim that family problems are more likely to cause work difficulties than the converse might suggest the family as the principal entrée into working-class culture, rather than the job itself. This study will postulate that it was in the period around 1900 that the balance was swinging toward family concerns, but that the disruptive effects of work could still predominate. But again, while going beyond the work experience into its links with a more general culture, we cannot pretend to offer a thorough assessment of this culture.

We can legitimately excuse some of this incompleteness with the standard invocation of having to start somewhere, given the absence of much historical attention to working-class life in this later industrial period. But there is no need to take too apologetic or tentative a tone. Despite the precedents cited earlier, work has received little historical attention, and if its study illuminates only a part of working-class life, it is a vital part. We have spent far more time on developments off the job: the standard of living and the labor movement being the most obvious cases in point. As we grope even in the present day to understand what a worker sees in his work, with immensely confusing results,²⁸ a historical focus on this phenomenon alone seems overdue. We know a bit about the traditional expectations that early workers brought to their jobs, their willingness to work long hours at a slow pace, and their lack of a modern sense of time. We know something more about the kind of work ethic that management hoped to foist upon them.²⁹ We know something about individual disputes that developed over work patterns. But the overall reality of wo rk and the varieties of response it elicited remain elusive.

The question of varieties of reaction will preoccupy us consistently. What might seem the most obvious analytical approach to th is question will not be used extensively. This is a cross-national study, rather than a rigorously comparative one. We will not, save in some final conclusions, try to isolate distinctive national experiences. I have dealt with the inadequacy of the most common national characterizations for working-class history in this period in another essay.³⁰ We will note subsequently, for example, that German workers, far from being unusually assiduous, were less work-oriented in the modern sense than British, whose productivity was higher in almost all the modern industries. German workers hung around the job more, but it is much more sensible to explain this by their more recent exposure to industrial work than by some peculiar German Arbeitsfreude. Hence the most fruitful comparisons concern recency of industrialization. We

will be paying particular attention to Germany, as an area of rapid changes in work systems for a population still highly traditionalist, in contrast to England, where work changes involved a population with much greater industrial experience. But the purpose is more to explain stages of development than particular national patterns, for given industrial categories could depart from any general norm. British construction workers, for example, were less open to change at work than French or German, as they talked of 'the likes of us' and found it difficult to contemplate progress.³¹ It is within industrial categories, not across whole nations, that comparison is most revealing.

Hence our efforts to encompass variety will stress key groups of workers, who may then be compared regionally and nationally to determine secondary characteristics. Most obviously in the case of journeymen but also among textile workers, metallurgists, railway workers, and even dockers, a certain selectivity operated in determining entrance to a type of work, and this was becoming more pronounced with the maturing of industry. With varying combinations of voluntary choice and compulsion, certain mentalities and even physical types tended to cluster in the main working-class occupations. Along with differences in the organization and technology applied to work, these differences produced varying reactions to work changes and efforts to develop compensatory enjoyments. Related to these categories are distinctions of age and sex, which will also be given prominent place. The work experience of women and the attitudes they developed to the job were different than those of men, while older workers took a different stance from the young.

Nevertheless, while variety must be treated in any analysis that is more than a collection of laments, we can begin with something of a common framework. All four nations considered had advanced industrial economies. Changes in technology and business organization were not substantively different in any of the four. France, long seen as an industrial laggard, had in fact an experienced working-class nucleus and a comparable pace of productivity increase during the nineteenth century as a whole.³² Its inclusion, at first sight perhaps unusual because of the common confusion over per capita and total growth rates, raises no real analytical problems. Even France's distinctive working-class protest tradition, manifest in this period in revolutionary syndicalism, neither reflected nor provided a really distinctive pattern of work.³³ Oddly enough, apart from differences due to length of exposure to an industrial economy, it is Belgium that produced the most distinctive worker experience, for despite long industrial exposure

much of the Belgian manufacturing labor force was not urbanized by 1900. Nevertheless, the framework established by advanced industrial capitalism covers all four countries considered and provides our starting point. Further comparison, particularly with the United States, where so many changes in work experience were now originating, is obviously desirable, but we can begin with this grouping.

More tentatively, we can suggest a few common ingredients in the culture that workers themselves brought to their jobs and indeed their whole lives. Despite the tremendous variety in situations, to which we return in the next chapter, workers tended to share a considerable fear Of change. They were easily convinced, more often than not correctly, that change would cheat them. They had seen previous changes redound to the benefit of their employer. They remembered, from their own experience or that of their fathers, periods of massive unemployment that seemed the result of change; recollections of the crisis Of the 1870s were commonly echoed in the 1890s decade, which then produced its own proof that change could be dangerous. And even sophisticated workers, capable for example of advanced forms of protest, had not shaken off the sense that what was traditional was good.

Hence a judgement of work in terms of past standards made almost universal good sense. In 1902 workers at Krupp complained about new arrangements at work, saying that they 'expected justice' from the company; and the management, recognizing what was meant, urged that changes in old work customs be introduced more subtly.³⁴ South Wales miners in 1907 claimed in court that they deserved regular overtime, and the extra pay resulting, again as a matter of tradition.^{3 s} Tradition at work commonly dictated a certain slowness of pace in most factories as well as artisan shops; this was true even in British metallurgy, where it was attributed, well after 1900, to the persistence of craft customs.³⁶ It dictated a sense that a certain amount of work deserved a set price. Hence workers in a Halifax metal plant found it perfectly natural to strike on behalf of a colleague who had refused a wage because he found it unfair 'for the amount of labour in it.'³⁷ German artisans expected to follow the customs of the place (ortsublich was the adjective commonly used) in the amount of work they did.³⁸ And the approach was shared by a host of other kinds of workers. Even relatively recent developments could be assimilated into the cake of custom and vigorously defended, for the cast of mind here inhibited a sense of constant change. Hence miners would agitate when work arrangements such as the time of rising from the pits after

completion of a piece of work, introduced a few years before, were altered, claiming great antiquity for the 'customs' they sought to preserve.³⁹ Factory workers could join their employers in concealing violations of safety laws from government inspectors, again from a desire to hold on to what they had against outsiders.⁴⁰ Even workers normally designated unskilled had their sense of tradition on the job. German dockers had a strict demarcation of tasks, differentiating types of goods loaded, work on the ship as opposed to that on the quay, carting, and so on, and they were as capable as any other group of pointing to the special customs of their job.⁴¹

And the traditional bent extended beyond work itself. Workers' wives could protest a new system of pay because it called for new thinking. Belgian miners' wives thus resisted the replacement of weekly with biweekly paydays; the employers claimed this reflected simple lack of budget knowledge, but the mentality involved went deeper.⁴² Within the home, male workers not uncommonly resisted any alteration in decor or furnishings. Traditionalism could extend to diets. A London wife reported: 'My husband, he'll eat nothing but boiled bacon, or a chop, or a bit of fat beef over pertaters; if it won't run to that, "Gimme a bit o' braun' cheese and I'll know weer I am," he'll say.'⁴³ And certainly» the traditional attitude covered the common approach to medicine. Workers typically remained Loath to call in a doctor, even when their wages might have sufficed.⁴⁴

At an extreme, traditionalism led to apathy, a sense that life was hopeless and beyond control. Many German workers dreaded the future or claimed that God or fate would rule their lot; such was the outlook in the Daimler automobile plant, near Stuttgart, where not a few workers said they 'hoped they would die first' when asked about the possibility of changes in the future.⁴⁵ At least as typically workers were unable clearly to answer questions about the future, even about how they intended to provide for their old age; there was simply no basis for thinking about such things. Where religion remained, as with many women workers, ignorance was not without comfort: T put my fate in God's hands ... then everything will be all right.'⁴⁶ More commonly one cannot escape the conclusion that workers could not bear the thought of looking far ahead, because no one would provide for them.

But traditionalism had another side to it. Workers had a pronounced sense of their own dignity, closely associated with the traditions of their jobs. During a Belgian miners' strike in 1901, the rumor spread that a hated employer had called his workers 'vagabonds'; the miners,

deeply offended, responded by trying to stone the man.⁴⁷ Workers in Belgium and elsewhere were revolted at the notion of classifying themselves as unemployed, some maintaining that to do so would lose them all claims to respectability. Unskilled workers, whatever their feelings of worthlessness in the past, produced frequent statements of their sense of dignity. French carters insisted that they be treated as something more than 'beasts of burden,' while sailors announced that 'they are not pariahs, having the same right as any man to aspire to improve their lot.'⁴⁸ The language of British workers bristled with references to their manhood, which they often associated with defense of their rights against their employers.⁴⁹

Thus in stressing the deep-seated traditionalism of the workers we are far from claiming that workers were passive creatures, incapable of understanding the changes that were occurring about them. And of course traditionalism as we suggest it here is a very vague term, for it is obvious that some workers were able to adapt to new situations, developing new aspirations of their own, far more rapidly than others. It is in many ways far-fetched to embrace, in a single generalization, uneducated Flemish artisans who returned to their native village after every work-day and manifested few interests beyond those of a customary peasantry, with, say, London or Parisian printers who were adopting many of the trappings of the lower middle class. Yet the theme of traditionalism must recur frequently in a discussion of work, und it does cut across national and professional lines. Workers measured their present by their past, and this gave them deeply-felt standards of what was appropriate. It left many of them more than a bit handicapped in developing new aspirations for the future. Even when they had hopes for a better life, they often realized that the hopes were futile or left the hopes deliberately vague, as with the many German workers who, bothered by persistent sociological pollster s — another sign of modern times — simply talked of wanting 'something better in the present life.'⁵⁰

The turn of the century period gains much of its drama from the confrontation between workers' values, still largely turned toward the past, and a new tide of change. A culture that had been painfully established or re-established after the first shock of industrialization was now challenged. What work was, what wives were for, what children were for all had to be rethought. Out of the variety of responses to these challenges came a recognizably modern, though highly differentiated, working-class life style.

Notes

1.    Jurgen Kuczynski, Darstellung der Lage der Arbeiter in Deutsch-land von 1871 bis 1918, 2v. (Berlin, 1962-67); on women's work, Margaret Hewitt, Wives and Mathers in Victorian Industry, (London, 1958).

2.   E. P. Thompson, 'Time, Work Discipline, and Industrial Capitalism,' Past and Present (1967>, 56-97.

3.    Herbert Gutman extends this theme into the later nineteenth century for the United States by stressing the adaptation required of each successive wave of immigrants: 'Work, Culture and Society in Industrializing America, 1815—1919,' American Historical Review, (1973), 531-88.

4.   David Landes, The Unbound Prc*metheus (Cambridge, 1969), 301-23.

5.   Peter N. Stearns, 'Adaptation to Industrialization: German Workers as a Test Case,' Central European History (1970), 303 ff.

6.   Arno Hoffmeister, Die wirtschaftliche Lage der Landarbeiter im Ostpreussen (Halle-Saalen, 1908).

7.    Brimley Thomas, 'The Migration of Labour into the Glamorganshire Coalfield (1861-1911), Economica (1930), 298 ff.

8.   Ferdinand Tônnies, 'Hafenarbeiter und Seeleute in Hamburg vor dem Strike 1896—97,' Archiv fur Gesetzgebung und Politik (1897), 673 ff.

9.   E. Prancke, 'Die Arbeitsvexhaltnis se im Hafen zu Hamburg,' Jahrbuch fur Gesetzgebung, Verwaltung, und Volkswirtschaft im Deutschen Reich (1898), 943-50.

10.    See, for example, Journeymen Felt Hatters, Annual Report, 1891.

11.    Over half of whom lived in the center city. Charles Booth, Life and Labour of the People in London (London, 1897), X, 25.

12.   Gareth Stedman Jones, 'Working-Class Culture and Working-Class Politics in London, 1870—1900; Notes on the Remaking of a Working Class,' Journal of Social History (1974).

13.   Thomas Bell, Pioneering Days (London, 1941), 61; J. Zitzlaff, Arbeitsgleiderung in Maschinen-Unternehmungen (Jena, 1913).

14.    Amalgamated Society of Engineers, The Lockout in the Engineering Trade (London, 1898), 3.

15.   Royal Commission on Labour, Digest of the Evidence, Group 'C (Textile, Clothing, Chemical, Building, and Miscellaneous Trades (London, 1893, C 6894), II, 30.

16.   At an extreme, the piece rate could multiply worker output by 100 per cent in five years, as was the case in Belgian glass manufacture between 1903 and 1909: Asa Briggs- and John Saville, eds., Essays in Labour History (London, 1 960), 1 38.

17.    H. Stanley Jevons, The British Coal Trade (London, 1915), 835; Miners' Federation of Great Britain, Executive Committee Meetings, 1909.

18.   Arthur Shadwell, Industrial Efficiency (London, 1906), I, 114 ff.

19.    Alfred Williams, Life in a Railway Factory (London, 1915), 183 ff. ; Adolf Levenstein, Aus der Tiefe, Arbeiterbriefe (Berlin, 1908), 54 ff.

20.    Archives de la Sûreté publique de la province de Liège (Archives de l'Etat, Liège), Carton XV, report of 1913.

21.    Levenstein, Tiefe, 90.

22.    E. J. Hobsbawm, 'Custom, Wages and Work Load, in Labouring Men (New York, 1964).

23.    John H. Goldthorpe et al., The Affluent Worker in the Class Structure (Cambridge, 1969); Robert Blauner, Alienation and Freedom; The Factory Worker and His Industry (Chicago, 1964).

24.    Work in America (Cambridge, Mass., 1972), passim.

25.    Stearns, 'Adaptation,' passim.

26.    Levenstein, Tiefe, 60 and passim.

27.    Mrs. Hugh Bell, At the Works (London, 1907); Adolf Weber, Die Lohnbewegungen der Gewerkschaftsdemokatie (Bonn, 1914). For more extensive comments on the sources used, see the bibliographical essay.

28.    It is plausible to argue both increasing pleasure and increasing distress in factory work since World War II: see Goldthorpe, Affluent Worker; Ferdynand Zweig, The Worker in an Affluent Society (New York, 1962); Sar A. Levitan, éd., Blue-Collar Workers (New York, 1971), 30 ff., 160 ff. See Conclusion.

29.    Sidney Pollard, The Genesis of Modern Management (Cambridge, 1964).

30.    Peter N. Stearns, 'National Character and European Labor History,' Journal of Social History (1970), 95-124.

31.    Raymond Postgate, The Builders History (London, n.d.), 38.

32.    Francois Crouzet, 'Essai de construction d'un indice annuel de la production industrielle française au XIXe siècle,' Annales: Economies Sociétés, Civilisations (1970), 56—101.

33.    Peter N. Stearns, Revolutionary Syndicalism and French Labor (New Brunswick, N.J., 1971).

34.    Gerhard Adelmann, éd., Quellensammlung zur Geschichte der sozialen Betriebsverfassung: Ruhrindustrie unter besonderer Berùcksichtigung des Industrie- und Handelskammerbezirks Essen (Bonn, 1960), I, 306.

35.    Jevons, Coal, 528 ff.

36.    D. L. Burn, The Economic History of Steelmaking, 1867-1939 (Cambridge, 1940), 147.

37.    General Union of Braziers and Sheet Metal Workers, Annual Report, 1913.

38.    Zentralverband christlicher Holzarbeiter Deutschlands in den Jahren 1908/09 (Cologne, 1910), 65.

39.    N. Dethier, Centrale syndicale des travailleurs des mines de Belgique (Brussels, 1950), 27.

40.    M. Loane, The Queen's Poor (London, 1905), 88.

41.    Paul de Rousiers, Hambourg et l'Allemagne contemporaine (Paris,

1902), 249 ff.

42.    Association houillère du Couchant de Mons, Historique de la grève du Borinage (Mons, 1911), 7.

43.    M. Loane, The Next Street but One (London, 1907), 28.

44.    Shadwell, Efficiency, I, passim.

45.    Fritz Schumann, Auslese und Anpasrung der Arbeiterschaft in der Automobilindustrie (Leipzig, 1911), 18.

46.    Ross Kempl, Das Leben der jungen Fabrikmàdchen in Mùnchen (Leipzig, 1911); see also Heinrich Herkner, Problème der Arbeiter-psychologie (Leipzig, 1912), passim.

47.    Fr. Vercruysse, La Contribution belge à la statistique internationale de chômage (Liège, n.<L), 37; S. H. Scholl, De Geschiedenis van de Arb eidersbe-weging in West-Vlaanderen (Brussels, 1953), 4.

48.    Adrien Veber, 'Mouvement social,' ±a Revue socialiste (1901), 361, 362, statements by individual workers in the 1901 port strikes, in Le Havre and Marseilles.

49.    Peter N. Stearns, 'Working-Class in Britain, 1890-1914,' in M. Vicinus, éd., Suffer and Be Still (Bloornington, 1972), 113.

50.    Max Morgenstern, Auslese und Anpassung der industrieller Arbeiterschaft betrachtet bei den Offenbacher Lederarbeitern (Breiburgi. Br , 1911), 110.

PART I THE FRAMEWORK FOR LABOR

CHAPTER 1 THE EVOLUTION OF THE LABOR FORCE

Establishing the cast of characters for a study of the working class is no easy task, yet it must be attempted for it provides both the framework for further analysis and the outlines of the job structure within which workers operated. Beginning with the premise that manual, non-agricultural labor demarcated both a class and a distinctive work experience, even a brief study of occupational structure introduces the theme of diversity. While modern industries commanded an increasing share of the working class, older sectors, including a large domestic manufacturing segment, did not yield quickly. Rapid growth itself produced diversity. Expanding industries like metallurgy and mining drew in people who abandoned their ties to the countryside reluctantly if at all. Railroads and street car lines allowed many new workers to maintain rural residence; around Hannover some iron workers came in from 30 to 40 kilometers away. Similarly, in one iron factory near Charleroi, over half the workers lived in a variety of towns distant from the plant. Even in England, where the industrial labor force was undeniably more fully formed, residential distinctions persisted; a railroad factory near Swindon found only the skilled workers living in town, while the laborers usually lived in the villages and came in on the train.¹ Clearly, commuting was less of a jolt than moving, but it therefore perpetuated important divisions within the working class. The stabilization of the industrial economy, after the first decades of frenzied growth, often made it more difficult to recruit new workers, for the worst rural misery had been skimmed off; hence an increasing resort to foreigners, who brought their own rural attachments with them. Polish workers in the Ruhr, for example, made a concerted effort to keep their traditional costumes and festivals.² Diversity of customs and residence affected the crafts as well. Thoroughly urbanized Parisian construction workers every summer faced colleagues who wandered in from small towns. Ten to twenty percent of the workers in craft centres such as Antwerp and Liège lived outside town, while in Frankfurt a full 72 percent of all masons lived in villages around the city.³

Thus diversity and the necessity of dealing with strangers and strange ways on the job were important elements of the work situation still. They loom larger in Germany, with its fast-growing economy, than in

England, but they were not negligible anywhere. And if the number of newcomers to industry was less overwhelming than it had been before, some of them seemed stranger, if only because of the distances they could now travel, and the chance to complain was greater. Attacks on foreigners, on country people, on domestic manufacturers, and on women at work abound, and though the y are often out of proportion to the real number of strangers on the job they reflect serious dissonances within a working class that was still incompletely formed. Dockers in Barry and elsewhere called for a national strike against Chinese labor, while sailors in Marseilles struck several times against foreign, particularly Arab, workers.⁴ German miners complained that their foremen were treating them more harshly because of their impatience with foreign workers, who understood little German; it was easy to carry over a cursing intolerance to the native labor force.⁵ British union organizers, though rarely faxed with real foreigners, could still point up the gap between urban and rural manufacturing workers. When a 1909 law required the establishment of conciliation boards in putting-out trades such as chain-making, machine lace manufacture, and wholesale tailoring, the worker representatives were virtually inarticulate, incapable of understanding the notion of dealing directly with employers.⁶ German and French bakers and tailors split between a modern group that fought customary ties with the craft master, and a large traditionalist segment that wanted nothing better than to live in and court the boss's daughter. Divisions in background, residence, and outlook abound, and no structural outline can capture all of them. We can, however, hope to convey some of the constant jostling with newcomers - newcomers, if not to industry itself, at least to the type of industry they were now involved in — that remained an important fact of life in most branches of manufacturing. Here is part of the background of the organized labor movement itself, which could provide some protection against strangers; a letter from a London worker caught the mood:

We must keep the question of who we shall work with to decide among ourselves, and not allow the employers to say that we shall work with non-unionists, blacklegs, or any other undesirable persons.⁷

Strangers became more menacing because of another, newer trend within the working class as a whole: the growth of the class was slowing up. Fear of dwindling opportunities was of course not new for artisans,

who had often anticipated the problem before it hit home. Important branches of industry were still growing lustily. Again, diversity complicates any generalization. But the slowing up of growth in the factory population was new, and it touched huge segments of the working class. Despite industrial growth only 12.1 per cent of the male manufacturing labor force in France was under 24 in 1911 compared to 19.1 per cent in 1896. If, then, a general point can be ventured beyond the invocation of continued, often abrasive contact with newcomers, it is that the working class rather generally became conscious of limitations to opportunity — opportunity not in the middle-class sense of chances for advancement, but in the more genuinely working-class sense of being able to continue doing what one had started out doing. Some workers were directly displaced from their jobs. Far more had to help their sons or daughters decide to enter a different trade. Reshuffling, more than new growth, was coming to characterize industrial work.

The outline of occupational structure and its evolution is of necessity purely descriptive, heavily numerical, and complicated. What follows is based on national census materials from the four countries under study.⁸ The material raises a host of obvious problems. It is not congruent chronologically; here the twelve-year gap between German censuses and the absence of material after 1907 are particularly unfortunate, among other things exaggerating comparative German growth because the 1895 census picked up the effects of the mid-'nineties crisis while that in 1907 predated another serious recession by a full year. Occupational categories varied from one country to the next and changed within the same country, complicating comparisons across time as well as across national boundaries. The French census of 1911 is indeed almost unusable because of its erratic job classifications and the absence of breakdowns by sex.⁹ More normally, as in Britain, census categories steadily improved in precision, but it remains impossible fully to rectify earlier fuzziness.

Furthermore most of the general tables, designed for economic more than social analysis, do not distinguish among workers, clerical help, and employers, which muddies any study of the working class per se.¹ ° Here, some controls are possible concerning the labor force as a whole (though not for more specific categories such as women, where figures, invariably include employers as well as workers).¹¹ Clerical employees cannot be sifted out. Presumably their numbers advanced; by 1906 in France they ranged from 1.4 percent of the total work-force in artisanal industries like wood, to 12.2 percent in highly-organized chemicals, but given the absence of figures from other censuses trends cannot be

clearly established. Employed personnel can, however, be sorted from employers in several cases, and with interesting results. Until about 1900 the presence of employers in the general census listings tends to reduce growth rates, for the work-force was increasing more rapidly. The one exception concerns stagnant or depressed industries — particularly crafts (French leather, French and English construction between 1901 and 1906) but also English textiles in the late 1890s. Workers normally escaped a sinking ship more rapidly than employers, so that general census data either minimizes the decline in numbers or conceals a loss of workers outright. Workers had greater need to abandon a deteriorating industry, but in some cases, perhaps, more choice in the matter than a small owner. After 1900 the general census data, where it distorts at all, exaggerates overall growth. Despite important tendencies toward rationalization, the number of employers was growing more rapidly than that of workers. This was true even in highly-organized sectors such as French, and Belgian mining or English shipbuilding where, despite consolidation or perhaps because of it, managers outstripped workers; it was certainly true in the crafts, as the number of small owners grew (as in French and British metals) or held firm as when the labor force shrank. But here the distortions caused by the presence of employers in calculations simply reduce the visibility of major trends; they do not conceal the trends or seriously alter the comparative judgements from one country to the next. And if they do tend to inflate the size of the crafts compared to the factory sectors, this is partially offset by the larger clerical contingent in the latter.

In general it should be clear that judgements on trends are fairly solid, even where absolute numbers leave serious doubts. Comparisons across time are therefore far firmer than those from one country to the next, where a whole variety of differences in the bases of calculation may intervene. Comparisons among industries are at an intermediate level of probability; they are not exact, if only because of the inclusion of non-worker elements, but they correctly indicate general orders of magnitude. It should also be noted that detailed studies of the breakdowns in the printed censuses could greatly improve the precision of our picture of occupational structure while, in my opinion, rarely changing the picture itself.

The number of people engaged in industry or transportation grew steadily between 1890 and 1914. Germany had the highest growth rate, followed by Belgium, France, and England. The French growth rate was bouyed by an unusually rapid gain in transportation; industry alone was more sluggish, for here as in England and to a lesser extent Belgium,

reshuffling was more significant than absolute growth. Except in Britain, where the industrial-transport sectors already commanded 90 percent of the population, the importance of industry in the occupational structure steadily advanced, which meant a steady if modest pull of newcomers, mainly from agriculture, into the working class. Here France held its own, suggesting that its slow overall growth was in large part a function of the general demographic lag. Finally, in all countries offering more than two censuses in the pre-war decades, the: decline in rate of growth after 1900 is marked. (Table I.)

Breakdowns for the major industries help explain both some of the peculiarities of the national patterns and the general evolution before and after 1900.¹² (Table II.) The industries themselves behaved rather similarly. Belgian miners faced declining opportunities, compared to the national average, whereas mining was a big gainer elsewhere. German transport behaved oddly. Efforts to improve efficiency, as on railroads, may have limited growth, and the German ports did not draw the hordes of new dockers characteristic of France and Britain; but the statistical comparison is almost certainly the result of census error. Everywhere else transportation was a burgeoning sector. Metals and machine building consist, quite simply, of far too many diverse segments; here the German distinctions between the two industries is useful. Machinery gained ground everywhere, led by electrical equipment and vehicles.¹³ Traditional metal work, in contrast, lagged; hence in France, between 1901 and 1906, lock manufacturers gained but .8 percent (and the number of workers alone probably shrank) while automobile manufacture increased by over 180 percent. Among the generally low growth industries, France stands out in its rapid curtailment of artisanal sectors (wood, leather, food) which still enjoyed modest gains elsewhere.

In terms of the large categories, then, construction, transport, and mining gained ground rapidly. So in fact did machine construction, but it was held back by metals which was by far the larger employer (its labor force more than twice as big, in Germany, in 1895). Printing and paper, a smaller employer, enjoyed more modest growth. Chemicals and metallurgy were spectacular gainers, but averaged a relatively small percentage of the total industrial labor force.

Generally below the national average in growth were artisanal, small-shop sectors two of which — wood and food — were substantial employers of labor. Clothing, in 1890 the province of craftsmen and women working at home, grew little and textiles, in most cases even a bigger employer, grew scarcely at all - yet these were still among the

Table I Evaluation of the Industrial Labor Force

(Manufacturing and Transport Occupations)

three or four largest users of workers in 1911. In other words, the modest but steady average growth rate of the industrial countries resulted from dramatic differences among major industries — no great surprise, perhaps, but an important fact for workers themselves.

Specific national growth patterns resulted in part from peculiarities in the industrial balance. It should be noted that Table II B, taken from the end of the census periods, tends to homogenize the national figures, for two decades of differential growth had already reduced certain peculiarities. Important distinctions remained, however. Because of relatively meager resources, France had only half the percentage of miners of the other industrial countries; this would seriously affect the 'average' characteristics of the French labor force, as in the question of overall growth where France was deprived of the full impact of a rapidly-expanding sector. Germany's high growth resulted in considerable measure from its existing specialization in metals and machines (which, of course, growth only heightened); Belgium similarly benefited in metallurgy. But Belgium also had a large textile sector, which Germany lacked. French specialization in crafts and clothing (though with food processing excepted) obviously hampered overall labor growth at this point but provided a pool from which other industries might gain. Britain's apparent bulge in food processing results in part from a more inclusive census category. Germany's concentration in construction probably has greater base in reality, for this was a less efficient industry in Germany, while a harsher climate required a larger labor pool since winter work was more limited. Differences in printing follow interestingly from national literacy rates, but they were not, of course, particularly significant for the general labor force. Overall, differences in the rank order of industries point up the peculiar situation of Germany, where four of the top seven industries were rapid gainers of labor. Elsewhere considerable concentration in artisanal sectors (France, Belgium) or in stagnant factory industries (Britain, Belgium) made the reshuffling of workers more important than the recruitment of newcomers to industry. Except in France workers were not being forced out of major categories, but growth rates were so slow that many of their children would have to decide to enter other lines of work. In fact, of course, the distinctiveness of Germany was less marked than the figures show. German sons and daughters were also being pushed out of important traditional sections of metals and textiles, for the craft/rural tone of German manufacturing was at least as pronounced as in France still in 1900. And where reshuffling seems called for on paper it did not necessarily occur in practice. French

Table II Patterns of Industrial Growth: Summary and Comparative Tables

A. Annual growth rates, major census categories

B. Percentage of total industrial labor force in major industries, last census date

metallurgy, for example, found it difficult to recruit craftsmen from other industries, particularly because the metal trades were still vigorous enough to provide a haven. So French heavy industry, like German, had to recruit unskilled rural labor extensively.

One further remark about the concentrations of labor can apply to all the mature industrial countries. All the biggest industrial categories had a strong traditional component, even in 1911, whether they continued to gain ground or not. Construction, one of the leaders, increased its labor force in part because technical innovation was modest. Employment in metals and certainly in textiles and clothing lagged somewhat but it was still substantial; yet much of this work occurred in small shops or at home. Mining had its own traditions as did major branches of transportation (in British transport, road, employment was almost double that of railroads and water combined, yet it had undergone relatively few changes in method).¹⁴ And the outright artisanal sector, as in wood-working, remained considerable. The most obviously modern industrial sectors, such as chemicals, were still small, even though their growth rate propelled them forward rapidly. A long background of artisanal production plus the earlier stages of indus-trialization left a massive population in industries in which change occurred rather modestly (but could be all the more shocking for its contrast with a traditional framework) or in which stagnation or decline were painfully forcing workers to choose new careers or at least to fear for the future.

The theme of painful readjustment applies particularly to women workers, for the changes in the labor force affected their traditional concentrations in manufacturing with unusual intensity. In the 1890s there were marked similarities in the situation of women in manufacturing labor. France employed the highest percentage by far, for 32 percent of its manufacturing force was female. But this resulted almost entirely from the unusual preponderance of women in the manufacture of clothing and the large size of this industry in France. In other words, French women worked disproportionately, but mainly in the home.¹⁵ In shop and factory industries such as printing and metalwork they did not loom as large as their counterparts in Belgium and Britain. Their substantial total employment in France is significant, a function in part of the need for labor, given the demographic sluggishness, that forced employers to seek women in their homes whenever the nature of the industry allowed; and it may have had a significant effect on the women involved. But it was relatively traditional in its reliance on family environment. Germany in the 1890s suggests an even more

traditionalist framework. The low percentage of women in the labor force reflects above all the slight importance of textiles and clothing in the German economy, but percentages in other industries were low as well. Belgium and Britain were quite similar. Belgium stands out as the only country still using significant numbers of women in the mines, but its employment of women was at a slightly higher rate than Britain's mainly because of the greater importance of the clothing industry. Employment of women in chemicals and printing/paper was advanced in both countries (as in France) in contrast to Germany; in Britain (as in France) women had begun to make some inroads in the wood industry. But everywhere the overwhelming fact was the concentration of women in only three industries: clothing, textiles, and to a lesser extent food processing. And only in textiles were they extensively employed in factories.

A variety of forces converged to alter this situation — if slightly — by World War I. The stagnation of textiles and clothing meant that the employment rate, if not the absolute number, of women in manufacturing would automatically decline if they did not find alternate jobs. In fact women's employment in these industries, as generally in food processing, increased. Men were better able to flee the declining jobs and, as the main wage-earners in most families, had more need to do so. In France the employment rate of women held up mainly because they were able to switch from textiles to dressmaking; no great initiative was necessarily involved though the rise of clothing sweatshops meant that many women were changing locales.¹⁶ Only in France would this alternative work, and even here a rapid rise in the percentage of young manufacturing workers (48 percent under 24 in 1911 compared to 33 percent in 1895) shows that older women now chose, or were forced to drop out. Elsewhere the increase of women in textiles and dressmaking was not sufficient to keep women abreast of the general increase in the labor force and in the female population.

In addition to considerable economic compulsion, women's employment was facilitated by two other factors which will receive fuller consideration in subsequent chapters. Technological changes in industries such as engineering, printing, and furniture manufacture (wood) reduced skill requirements and opened the way for new recruitment of women. Many employers were eager to take advantage of this, given the lower wages women commanded and, in some cases such as Germany, the difficulty of finding unskilled male workers in sufficient number. Finally, many women became more eager to work outside the home, partly of course in r esponse to new opportunities.

Decline or stabilization of the servant population in France and Britain suggests the interest of unmarried girls in alternate employment.¹⁷ The rapid rise of married women in the factory labor force in Germany and probably France denotes another kind of change in attitude, even more fundamental in terms of working-class culture.¹⁸

Against a rapid readjustment of women to the new occupational structure stood their ties to the home, which could restrict their ability to respond to new jobs or to leave their accustomed areas, as well as deter both themselves and their husbands from contemplating their entry into a factory. And on the job, vigorous opposition from male workers, and often organized labor, could be expected. Belgium provides a case study of what happened when the opposition was strong and, in all probability, the change in women's consciousness insufficient to break through it. The percentage of women employed in manufacturing actually declined in Belgium, as the textile and clothing industries faded. Women responded mainly by taking an increasing role in these stagnant sectors. They were actually pushed out of printing and chemicals, in terms of their relative standing. This was common in chemicals after the 1890s, as males gained a greater role. In printing it reflected the Belgian lag in adopting new composing machines, which opened the door to semi-skilled labor, and the related strength of craft unionism in the industry. Finally, the Belgian working class, unusually concentrated in a rural residential environment, was hardly conducive to new occupational strivings either by wives or by daughters. So, though there was some absolute growth, Belgian women did not keep pace.

France produced a rather similar pattern, all the more surprising given the shortage of labor generally. Here there was some ground gained in paper/printing, but women fell back, relatively, in metals and chemicals. An increase in leather could reflect technological change, but this was a sick industry and more probably women were brought in, often as domestic or small-shop workers, because of their low wages alone. French women mainly opted for traditional jobs, so that their distribution in the labor force changed very little. The maintenance of a high level of female labor thus did not indicate significant new developments in the setting for work. One must assume that the relative vitality of the French clothing industry, in terms at least of employment maintenance, plus the absence of other motivations to seek new kinds of work, discouraged venturesomeness, except among new numbers of teenage girls.

In contrast, Britain and Germany showed some modest break-

Table III Women in Industry: Summary and Comparative Tables A. Annual Growth Rates (in brackets, general growth rate):

B. % Totals in Major Industries at Last Census Date:

D: Changes in % Totals in Major Industries:

C: % of All Women in Industry at Last Census Date:

E: Change in % of All Women in Industry:

throughs. In both, the employment of women increased as a percentage of the total and in both the balance of women in industry changed noticeably. Textiles, clothing and food processing remained the leaders, of course, but printing, metals, engineering, and wood picked up significant numbers. In Germany this often brought relative totals only up near the levels achieved already in France; change, rather than absolute development, is the notable feature. Britain went further, if only because the decline in textiles hit harder — causing a drop in the relative levels of female employment during the 1890s. Levels of employment in printing, wood-working, and to a lesser extent the metal industries were unusually high.¹⁹

The entrance of increasing numbers of women into the industrial labor force, despite the doldrums of their traditional industries, was in many ways a remarkable development. Across the board in metals and machine building, in printing everywhere save in Belgium, and widely in wood-working, leather, and food processing the role of women, and certainly their absolute numbers, increased. Women pressed forward also, though in small numbers, in construction and, in Britain, general unskilled labor, a clear sign of the hunger for new jobs once textile employment stopped keeping pace with population growth. The industrial map was not remade, but there was enough change to affect the lives of many women workers and certainly to provoke serious reaction from male workers and from the labor movement. Here was an obvious case in which the reshuffling compelled by changes in the balance among key industries caused serious anxiety as well as, possibly, some important new opportunities.

Children, on the other hand, were increasingly withdrawn from industry. In 1901, in Britain, they were a significant factor only in textiles, and noticeable also in mining, printing, and clothing. By 1911 a general and rapid decline in all industries save textiles, where the traditional reliance on young labor and children's earnings still played an important role, made child labor a negligible factor. (Table IV.) Children remained more significant in Continental industry. In Belgium, 5.6 percent of all industrial workers in 1910 were under fifteen. Available figures for Germany are more difficult to interpret, for they list children under sixteen, well after most would plan to begin an industrial job even in England. Here too, however, while there was no absolute decline, the rate of increase failed to match the overall industrial average. The massive rise of young people in a few industries - metallurgy, chemicals, engineering most notably -reflected less an increase in child labor than the means by which the

more general increase in these industries was affected. New entrants to the work-force were consciously choosing jobs with greater opportunities. Even in 1896, in France, industries that seemed desirable to enter, such as printing, metals, and even textiles, had above average concentrations of people under 24. The delay in the average age of entry to industrial work, plus the changing balance of economic advantage in the industrial spectrum, both prompted the sense of choice. Here, as with women, new concentrations may have promoted a new sense of competition; in a few cases, such as German printing, trade union efforts to restrict entry were reflected in a lower rate of growth than the general average, despite the relative vitality of the industry. But in most of these cases opportunity outweighed the tensions created by new patterns, and the big howls over the entry of young blood came in the declining trades. In these trades, efforts at exclusion, the advice of disappointed parents in the trade, and the young person's own advantage created a slow growth rate, or even, as in leather, a marked drop.

Correspondingly, in terms of the labor force involved, industries aged and, sometimes, feminized, more than they died. Beneath the large categories and without pretending even to mention all the relevant job classifications, some further sense of the contrasting fortunes of various branches of industry must be created. We can begin with the surprisingly important sector of domestic manufacturing, as a case study in decline. We must realize, again, the incomplete extent to which manufacturing was synonymous either with urban or with factory life even in the 1890s. Britain had relatively little domestic production, though we have firm figures only for 1901; by then only a few industries had more than 5 percent of its labor force outside shops or factories. By 1907 only 2.9 percent of all manufacturing labor consisted of outworkers.²⁰ But the 1890s had seen a major battle over outwork, in industries ranging from shoes to needles; in the latter, 60 percent of all workers had been villagers in 1891, after which they rapidly gave way to new factories.²¹ By 1901 the decline of outwork in Britain was relatively gradual; only in printing, small metal -work, and textiles was it abrupt enough to reflect significant displacement of workers rather than natural attrition.

Germany's battle against home work is easier to trace.²² In 1895 7.7 percent of all manufacturing workers were in the home; by 1907 this was down to 4.7 percent, and the numerical decline was 10.4 percent. This was a rate that involved more than attrition. Textiles, for example, dropped from 27.8 percent to 16.1 percent in the twelve-year census interval. Metal work fell with similar precipitation, to 1.9 percent. Only

Table IV Employment of Children

A. England, Wales, Scotland, under 14:

Table IV Continued B. Germany, below 16:

in clothing - the woman's industry - did home work really survive, with 19.3 percent of the labor force. Otherwise the decline left most home industries filled with older workers — some aged in the industry and incapable of change, others rejects from other jobs. Some workers must have been forced out explicitly. Others had to advise their sons to seek another kind of career. The crisis in home industry was not new, but here, as in Britain a few years earlier, was its final agony. Because it had involved such a significant percentage of workers as late as the mid-'nineties, the death pangs of the system had major ramifications. Some displaced workers could shrug off their loss, when they had worked only to supplement the family income.²³ In other cases there was dark despair. Female British outworkers were held to suffer unusually frequent emotional depressions, while many explicitly noted the lingering downfall of their trade.²⁴ And apart from the sentiments of workers directly involved, the decline of home work seriously affected the outlook as well as the numbers of new workers available for factory labor.

Home work retained greater vitality in Belgium and France.²⁵ Fourteen percent of all manufacturing workers were in the home in Belgium in 1896;²⁶ 57 percent of all clothing workers were so located. By 1910 the absolute number of domestic workers had risen by 27 percent — an annual increase almost as great as that of industrial labor generally; almost all of them (83 percent) were in textiles and clothing, most of them women. In France the percentage of workers in domestic industry fell only slightly between 1896 and 1906; here, wood, metals, and leather joined clothing and textiles in extensive involvement in this system.²⁷ Thereafter the percentage of home workers declined in metals and wood, though not so fast as to force many workers out directly; it fell precipitously in textiles, clearly compelling many workers to drop out of industry, enter factory employment, or most commonly, switch to clothing. For the system held up in clothing and also in leather, keeping the national rate relatively steady. Its vitality was aided by a desire to avoid social legislation and the greater manipulability, in terms of pay and conditions, of a labor force isolated in the home and, sometimes at least, self-satisfied that it had avoided the perils of urban life.²⁸ Maintenance, even revival, of home work could be found elsewhere, as in the German tobacco industry during the 1890s; here too, light, simple equipment and the need to find cheap labor that was easy to dismiss during seasonal slumps spurred the system along.²⁹

The point is not so much to stress the immense variety concealed

Table V Domestic Manufacturing Workers and Employers

Kngland:

within the general occupational structure as to emphasize the odd combination of persistence and rapid upheaval that characterized much of the older sector of industry as the economy matured. Almost any work system that had existed during the past two centuries could still be found, sometimes, in little pockets, enjoying a certain vitality. An untold number of workers were directly forced from their initial jobs, as in the cases of precipitous decline in the domestic manufacturing system. Far more commonly workers aged with their trade, often increasingly frustrated as the system declined, and their children were pushed or drawn into a new work setting — either relieved to escape their parents' dilemma or bitter that they could not continue the old ways.³⁰

Almost all the slow-growth industries, and a few catchalls such as metal work, embraced sectors of outright decline or aging. Most included a large number of shops as well, only a step removed from the same process of decline and only a bit more advanced in terms of industrial organization. In 1907 33 percent of all shops in German furniture-making, and 96 percent in basket-making, had no machinery, only hand tools. Quarries in Baden had their own traditional skill categories, and the workers lived a semi-rural existence little changed from a century before. Except for a few Italians recently brought in, most were of local agricultural origin, and they kept gardens and farm animals; their companies were small and their employers often workers risen from the ranks (while some workers had been employers in their own right before). And their numbers increased significantly — by 60 percent between 1895 and 1910 — falling back only in the years immediately preceding the war.³¹ Of course industries of this sort were fated for decline, though there were always islands of prosperity for the small concern — like the precious metals firms in Britain — to slow the overall process.³² In Britain, between 1897 and 1907, the number of workers in shops fell from 14.8 percent of the total in manufacturing to 12.5 percent — a decline that involved relatively litle direct displacement but a good bit of switching out of the fathers' industry.³³ Furniture suffered radically, falling from 44 percent to 29 percent in shops; here the shop workers declined by 29 percent, more than natural attrition could account for. But shop-employed tailors dropped only 7 percent in the decade; again, there was considerable variety. The fate of the small shop was clear — in Britain the number of such units stopped expanding after 1912, promising further reductions in the labor force involved — but its hand moved slowly. Industries such as furniture-making and clothing were long polarized between shops and machine

factories, quite apart from the lingering outworker element.

Yet decline there was in traditional settings and, even where the numbers of workers were still more heavily involved in the older industries than the new, one's eye is inevitably drawn to the high growth, mechanized, factory settings. The overall evolution, however qualified by invocations of diversity and gradualism, raises a final structural question, which at this point can be more clearly stated than answered: What was happening to skill levels in the working class? We know that a new type of worker, the semi-skilled, was rising in engineering, printing, and some other areas; while census categories prevent its being statistically traced, the growth of such industries, and of women workers in them, provide some outline. But as for the more traditional, if gross, distinction between skilled and unskilled the situation is less clear.³⁴ In wood, leather, and some branches of food processing, slow or negative growth, feminization, and the few instances of an increase of domestic manufacturing denote a decline in skills. Again, skilled workers were not necessarily directly replaced, but entry was curtailed and the skilled group aged. This is the pattern suggested, for example, by the wide gap between the growth of tailors and of clothing manufacturers generally in Britain.³⁵ On the other hand, several of the large, slow-growth industries, such as textiles, traditionally offered only modest skill levels, so their relative decline could have been part of an upgrading. And all of the high-growth industries involved a significant skilled component, even if aone was devoid of ambiguity. The increase of transport workers, for example, included a large leaven of dockers and road-workers whose skill levels could be quite low; but evidence from Britain, at least, suggests that the railroad labor force, which did involve significant skills, grew more rapidly than the other components in transportation. Mining and quarrying involved skill, but often of a rough sort; certainly the extractive industries recruited widely among unskilled elements and there was some sense, particularly in the quarries, that the unskilled were taking over instead of receiving proper training in a job hierarchy. Hence Welsh quarrymen tried to insist that 'the skilled quarryman is the best man to employ on every kind of rock.'³⁶ Construction work provided an even more obvious mixture of skilled and unskilled elements. Both gained in number during the period, but both were susceptible to unusual fluctuations; everywhere the construction labor force declined significantly between 1901 and 1906, and most of the evidence suggests that, as expected, an industry subject to such fluctuations took a particular toll among the unskilled. In France the

unskilled (terrassier) segment did lose slightly less ground than the skilled, probably because of the unusual activity still connected with the construction of Paris subways. Even so there was significant loss, and the concomitant rapid rise of the transportation labor force suggests the possibility of a massive influx into dock work and hauling. When prosperity returned, however, the skilled increased far more rapidly. In Britain the unskilled comprised a smaller percentage in the first place (46 percent among bricklayers, but only 24 percent among masons) and when the slump hit they suffered far more substantially, though the rise of a new category of builders' laborers helped compensate somewhat. More consistently, over the period as a whole, the rapid decline, feminization and probable aging of the 'general labor' pool in Britain suggests the channeling of unskilled males into more specific job categories, some of which — like important aspects of dock work — were not really unskilled.³⁷

In sum, the number of unskilled in industry and transport increased, though their vulnerability to economic fluctuations was still great. They recruited from agriculture, particularly on the Continent, and also probably from a general labor pool in the cities. New builders' laborers in Britain, for example, were a mixture of rural and urban, with the urban predominating; some had tried other jobs, others had never received training, and all tended to be older than the new entrants to the skilled ranks.³⁸ It is improbable that the increase in the unskilled outstripped that of the skilled or that large numbers of skilled before their later years, were forced into unskilled ranks — hence the need for a distinctive recruitment base of many unskilled among the rural, the foreign, and some women. Furthermore, it is probable that the nature of unskilled work changed, becoming a bit more job-specific and involving a bit more training than had characterized early industrial society.³⁹

But this made it all the more possible that skilled workers, even when increasing in numbers, felt a certain threat as they saw not only a rapid influx of newcomers but a diminishing gap between their attainments and those of the unskilled. This is not an issue, unfortunately, that census data can resolve, for we do not know enough about the balance in key industries such as mining or engineering. And the real historical problem involves perception as much as raw data; it must be part of the detailed examination of work, for which occupational structure provides only a framework.

The maturation of industry confronted relatively few workers with direct displacement, except in periods of recession. Where census data reveal significant numerical decline, they apply mainly to rural areas.

Far more common was attrition through aging, that could leave important pockets of older workers cut off from the bulk of the working class. In some industries, such as woodwork, many threatened workers became small proprietors instead of leaving the industry, which helps account for the rise of employers relative to workers; their existence may have profited little from this switch — which continued even after World War I in Germany - but the sense of ownership may have provided compensation not available to ordinary workers. Dis-placement was thus masked as well as limited.

For many workers the theme of expanding opportunity must have been predominant, as the rapid growth of a number of solid industries, offering high skill levels and good pay, was evident to many new entrants to the labor force. Young males benefited particularly from the resultant freedom to choose, but some young women and unskilled workers may have found new opportunities as well.

Yet the slowing of growth for the whole working class and the major reshuffling that was occurring within the industrial labor force left their mark. Generalization is risky of course, given the great diversity of situations. But large numbers of workers were exposed to strangers, some of them with a rural outlook but others with a background in mother traditional industry that was failing to keep pace. Large numbers of workers had to choose — or wanted to choose — occupations that differed from those of their parents. Parents had to worry about placing their offspring, and it is surely no accident that this period of a slowing of growth saw new working-class efforts at birth control, beginning particularly in the artisanal trades and textiles which were the first to see the threat materialize. Change there was, then, and it could be exciting and challenging; but it was not open-ended.

Notes

1.    Walter Timmermann, Die Entlohnungsmethoden in der Hannoverschen Eisenindustrie (Berlin, 1906); Alfred Williams, Life in a Railway Factory (London, 1915), 11 ff.; B. Seebohm Rowntree, Land and Labour: Lessons from Belgium (London, 1911), 289.

2.    Wilhelm Brepohl, Industrievolk im Wandel von der agraren zur industriellen Daseiform dargestellt am Ruhrgebiet (Tubingen, 1957).

3.    Rowntree, Land, 290; Zentralverband der Maurer Deutschlands, Der Kampfum die Arbeitsbedingungen (Hamburg, 1909).

4.   Docker's Record, December 1913.

5.   Archives départementales des Bouches-du-Rhône M6—2050, reports of 1909—10; Max Fleischer, Zwei Krisenjahre im rheinischer Bergbau (Frankfurt, 1905).

6.   Mary A. Hamilton, Mary MacArthur (London, 1925), 80 ff.

7.    Operative Masons' Journal, 1 April 1914.

8.    All tables in this chapter are based on the four national census series: Ministère de l'industrie et du travail (Belgium), Recensement general de la population, 1890, 1900, 1910 (Brussels, 1892-1912); Census of England and Wales, 1891, 1901, 1911 (London, 1893 — 1916); Tenth through Twelfth Decennial Census of Scotland (Glasgow, 1893-1903; London, 1913); Ministère du commerce (later, Ministère du travail), Résultats statistiques du recensement général de la population, 1896—1911; Statistik des deutschen Reichs, Neue Folge 102-19 and 202-22.

9.    J. C. Toutain, La Population de la France de 1700 à 1959 (Paris, 1962).

10.   A few obvious categories of dealers (in dress), innkeepers (in food) have been eliminated in all calculations on the British materials. But in other groups — e.g. metals — not only employers but also many tradespeople are included. This inflates British figures over those on the continent, where commercial categories were more fully developed.

11.    See Appendix, Table I.

12.    See Appendix, Table II, for more detail.

13.    See Appendix, Table IA and Table IID.

14.    See Appendix, Table IIC.

15.    See Appendix, Table III.

16.   The generally low level of growth in France between 1901 and 1906 makes the absence of sex-specific data for 1911 particularly unfortunate. By 1911 French developments were probably more comparable with those in Britain and Germany. But see note 18.

17.    Theresa M. McBride, Rural Tradition and the Process of Modernization (Unpublished Ph.D. Dissertation, Rutgers University, 1973), 143 ff., discusses the economic advantages involved in servant work that persisted until the 1870s in France.

18.    See Appendix, Table IV. In Britain, in contrast, the employment of married women in industry barely kept pace with population growth. Peter N. Stearns, 'Working-Class Women in Britain, 1890-1914,' in M. Vicinus, éd., Suffer and Be Still (Bloomington, Ind., 1972), 100 ff.; On Germany, Peter N. Stearns, 'Adaptation to Industrialization: German Workers as a Test Case,' Central European History (1970), 303 ff.; Ernest Dubois, Le Travail des femmes mariées dans la grande industrie allemande (Paris, 1902). French data are far less firm; in 1895 22 and 33 percent of the women in clothing shops and textile plants were married, and a whopping 45 percent of all women in out work, but this (as the declining average age of workers in. manufacturing suggests) could

not hold steady. It is known that the contribution of women to family budgets increased at least until 1906, when it averaged 12 percent of total family earnings in the working class; but by 1914 it was down to 5.4 percent, which may reflect the immense difficulties of the traditional women's industries in France. Maurice Halbwachs, 'Revenus et dépenses des travailleurs. Une enquête officielle d'avant-guerre,' Revue d'économie politique (1921), 52. 19. Stearns, 'Women,' passim.

20.    Aaron L. Levine, Industrial Retardation in Britain (London, 1967), 27 ff.

21.    G. L. Allen, The Industrial Development of Birmingham and the Black Country (London, 1929), 228 ff.

22.    Heinrich Koch, Die deutsche Hausindustrie (Monchen-Gla.dbach, 1913), 173 and passim.

23.    Ibid., passim.

24.    Select Committee on Home Work, Report and Minutes of Evidence (London, 1908 HC 246), 11.

25.    Rowntree, Land 85 ff.; Fernand Baudhouin, Belgique 1900 -I960 (Louvain, 1961).

26.    This percentage is of manufacturing alone — not mining, construction, or transport. The 1910 figure, for example, is but 0.9 percent of the entire manufacturing-transport labor force, which is in fact probably too low but which cannot be corrected from the census data. What needs stress is the persistence of home work, given the clothing industry above all, not the comparative rate, for it is possible but not proved that Belgium had a higher rate overall.

27.    The French percentages include independent artisans; this accounts, among other things, for the French rate in construction, where self-employed, rather than domestic, workers were really involved. In contrast British figures cited here and in the text do not include self-employed, and the German material is unclear. So it must be emphasized that France did not necessarily have a much higher rate of real domestic work than Germany and not so much more than Britain as the figures here might suggest. Despite stereotypes we do not know that the French were especially addicted to domestic manufacturing outside of clothing; It is the persistence that commands attention.

28.    Pierre Clerget, Les Industries de la soie en France (Paris, 19 25), 97 ff.

29.    Walther Frisch, Die Organisationsbestrebungen der Arbeiter in der deutschen Tabakindustrie (Leipzig, 1905).

30.    Sidney Pollard, History of Labour in Sheffield (Liverpool, 1959).

31.    Fôhlich, Die Steinindustrie in Grossherzogtum Baden (Karlsruhe, 1913), passim; Bernard Schildbach, 'Arbeitstarifvertrage in der Holzindustrie,' Jahrbùcher fur Nationalôkonomie (1910), 807 ff.

32.    Allen, Development, 248.

33.    Levine, Retardation, 22 ff.; Pollard, Sheffield, passim.

34.    Application, certainly, in the mid-nineteenth century; William Sewell, 'La classe ouvrière de Marseilles sous la seconde république,' Mouvement social (1971), 27—63.

35.    See Appendix, Table IIC.

36.    W. J. Parry, Penrhyn Lockout, 1900-1901 (London, 1901), 113.

37.    See Appendix, Table IIC and D; Norman B. Dearie, Problems of Unemployment in the London Building Trades (London, 1908).

38.    Dearie, Problems, passim.

39.    In Western Europe today it takes at least five years to 'train' an unskilled worker coming in from a non-industrialized area. The category unskilled is far from being a constant, and changes in the background required almost surely developed in this period. Stephen Castle and Godula Kosack, Immigration and Class Structure in Western Europe (London, 1973).

CHAPTER 2 JOB CHOICE IN A MATURING INDUSTRIAL ECONOMY

The entrance of workers into the labor force has often been discussed in general terms, particularly during the formative stages of in-dustrialization. Attention has focused on push-pull factors: whether workers were impelled to journey to the cities because of population pressure and, sometimes, additional restrictions on work in agriculture; or whether they were drawn into industrial work because of higher potential earnings, the ability to escape parental restraints at a particularly vulnerable stage of life, the desire to abandon the monotony of village routine, and so on. Undoubtedly pressure on the land set the stage for the decision to enter industry, but for a teenager, traditionally compelled to remain within the parental orbit while waiting to inherit land, the 'pull' hypothesis has merit as well. By the late nineteenth century the terms of analysis can be refined somewhat. 'Push' still operated. Continued population growth, even though now reduced in rate, combined with the agricultural crisis, improved rural technology, and the displacement of small owners in many rural areas as a minority of peasants increased the size of their landed holdings. Traditional rural alternatives to peasant effort, notably domestic manufacturing and rural craft labor, continued to stagnate or even decline outright. As this work increasingly became the exclusive province of older workers, it offered no opportunities for new entrants to the labor force.¹ Paid agricultural labor itself became slightly more attractive, in terms of wages offered, but this merely reflected the fact that increasing numbers of people were seeking other alternatives, thus shrinking the agricultural labor pool. And, vital in terms of the present analysis, the work aspects of paid agricultural service did not improve. Hours from sunup to sunrise, during the growing season, and frequent periods of unemployment combined with strict direction from employers.² Indeed, in this latter regard, there were growing complaints that many peasants and estate managers were becoming more peremptory³ — doubtless an effect of the need to improve competitive position in a tight market; so some of the possible paternal cushions of traditional agriculture were being eliminated.

But there was less sheer compulsion to abandon agriculture than had existed earlier, at least in the sense of inability to survive in some form

of rural labor. This was reflected in the diminished rate of migration to the cities in the advanced industrial countries. In turn, this suggests that the rural people who did move were making a positive decision to seek something different, while manufacturing jobs were more and more commonly filled by people born in cities in the first place. Yet when we ask more detailed questions about why people chose the jobs they did and indeed whether they had any sense of choice at all, the element of compulsion still looms large. Growing numbers of people were seeking higher earnings and more regular work; yet many of them had little sense of positive selection in the jobs they took to attain these new goals. Here in fact, even before a job was begun, was a basis for the separation between the fruits of work — wages — and work itself. If there was little reason for the job, as opposed to other jobs, there was scant basis for enjoying the job itself. Key groups of workers had distinctive reasons for not discerning a positive choice in the jobs they took, and in combination the sense of randomness, even purpose-lessness, was widespread.

Other workers did assign reasons to their commitment to one job instead of another. They tended to cluster in the more prestigeous occupations, but they can be found in other ranks as well. Certain industries tended to attract workers with particular characteristics and goals. Some of these were frustrated, having had definite hopes that they were unable to fulfill, but others really had some sense of wanting to be in the occupation they had chosen. What we are trying to get at is the diversity of motivations that brought workers to their work, a diversity that would to a significant degree predict the kinds of satisfactions that could be derived from the work.

Obviously, materials on job choices are impressionistic, but collectively they prove fairly convincing — even though they are certainly inapplicable to every individual case. Thanks to the fervor of German sociologists, we have most detailed information on the dilemmas of German workers, a large number of them fresh from the countryside. This can be contrasted particularly with the more settled situation in Britain, where industrial experience was of far longer standing. British information is by no means comparable in detail, but it conveys a probable and somewhat surprising transition from the more fluid patterns of a rapidly-growing industrial economy.

In focusing on job choice, rather than the more general fact of being in the industrial labor force, we are exploring a rather new topic with a high element of subjectivity. The whole notion of choice may seem inappropriate to those who prefer to judge workers in terms of their

powerlessness in the overall economic structure. But one can admit the compulsions which capitalism imposed upon labor and still attempt to examine the real belief that workers — some ardent socialists among them — could have in explaining why they, personally, were doing one job and not another. Within this framework, the initial question to be asked is in essence simple: What kinds of workers felt they could account for why they had their jobs, as opposed to those who could assign no real reason? It would be tidy if we could link this question to a more general process, such as modernization, and assert that those workers who could discuss reasons for their holding a particular job had advanced farther toward a modern level of decision-making than their less articulate fellows. But the question of choice is more complex than this. A worker might have a quite clear notion of why he had a job and yet be bounded in fact by very limited regional horizons. Parental guidance was one of the most common factors cited by workers who could explain their situation, and this was indeed an important element in the lives of many workers. Yet was a worker thus guided more independent than one who could give no reason at all for his job? He might after all be a more rigid traditionalist. The definition of choice in a broad pattern of industrial change is thus difficult. And those workers who lacked a sense of choice were in no sense necessarily behaving irrationally or without cause; they may, as we will see, have made some extremely important decisions to enter the industrial labor force in the first place. But here we are back to the push-pull framework, in which the worker's own thinking about choosing his particular job is either unobtainable or never really existed at all. When we deal with workers who did think they made choices we can talk about the various kinds of factors that influenced them. In so doing we can begin to explain reactions to the nature of work itself, for anticipation would play a major role in ongoing judgements. We can further advance our understanding of the complexity of the industrial labor force. For a study of job choice, by the time that industry was reaching a certain maturity, produces a number of important conclusions. While some workers seem almost random elements in the industrial process, others played an active role in finding their occupational slot — ev en if they found themselves disappointed in it (for choice is not equivalent to satisfaction). Nor were workers, even apart from regional restrictions on the labor market, tools of simple market forces alone. Their own criteria might play down wage advantages in favor of more intangible factors, even to the point of affecting the rate and nature of industrialization itself. Nor, finally, were workers who sensed a choice

united in what to make of it. If we begin with the simplest distinctions, between workers following traditional criteria of choice and those bereft of choice, we must end with a complicated list of factors to explain why workers thought they had the jobs they had.

It is easy to identify the workers who could most readily account for choosing their jobs. The traditional crafts long stood at the top of the list of occupations to be sought. Not only sons of urban artisans but also newcomers from the countryside with some manufacturing background assumed that craft labor was to be preferred over all other manufacturing work — even at the expense of maximizing earnings. A Berlin shoemaker in the late 1880s urged his profession on his son: 'There is no need to be pressed into factories or to work outside your home and according to the steam whistle.' In shoe-making 'something of the old craft Gemutlichkeif remained.⁴ In France many shoemakers refused to leave their homes to enter the factory, even after machine competition severely reduced wages, for they valued their freedom, seeking not only to work at their own pace but also such simple pleasures as the freedom to smoke on the job.⁵ And if this choice still made sense in a declining craft, the attractiveness of high-paying artisanal jobs would be all the more obvious. In Bremen in the 1890s jobs were chosen in a rather traditional community setting. The Protestant pastor helped advise children on their profession, and of course paternal guidance played a great role. Competition for apprenticeship jobs was intense, and a boy would be very proud if he was able to win training in printing or a comparable craft; he could even hope to have a two-years' Wanderszeit (tramping period) once he completed the apprenticeship.⁶

Obviously the craft framework was strongest in countries like Germany where industrialization was still relatively new, but the prestige of traditional skills continued to have great appeal even in England, particularly in the villages and smaller towns. Robert Tressell described the tendency for sons to follow fathers into the construction trades in Hastings, adding that in part this derived from the fathers' desire to confirm their own worth by making sure that their sons did not rise above them.⁷ Parisian masons, who still clustered around City Hall and the He Saint-Louis, preserved even in a great city the kind of community structure in which entry into the crafts would be a logical job choice, urged by the fathers and accepted without question by most sons.⁸

Yet this pattern of selection was increasingly threatened by the end of the nineteenth century, since opportunities in the crafts did not

expand as rapidly as claimants upon them. General population growth played a role here, though urban artisans were quick to reduce their birth rate which compensated somewhat. In many trades stepped-up pressure from employers plus the introduction of new machinery reduced the quality of apprenticeship and gave a variety of newcomers access to previously skilled jobs. This was a constant threat in printing once composing machines began to spread in the 1890s. British printers' unions, having long tried to limit the number of workers who could operate the machines, had to yield to employer demands in 1895 and allow workers to learn machine skills on their own.⁹ Here as elsewhere there were recurring complaints that apprenticeship was decaying because the novice was no longer taught the whole skill.¹⁰ Machines in Belgian cabinet-making even more seriously cut into apprenticeship, particularly in Brussels; here, only the smaller companies continued any formal training, and these taught only portions of the traditional skill, and in bad conditions.¹¹ The rest of the trade was being taken over by factories.

Where the factory system was not triumphant, however, dilution of apprenticeship, though often discussed, was not too serious. Printers, for example, were widely successful in maintaining a thorough apprenticeship even for workers destined for the machines. Lyons dyers, fighting the intrusion of unskilled labor, insisted on a fixed period of training and a set proportion of apprentices.¹² Pressure on apprenticeship set part of the framework for the crisis of the artisanry, but other factors, including key decisions by artisans themselves, played a more direct role.

Very simply, the increasing invalidity of rural and village crafts, against factory competition, forced more and more trained journeymen into the cities. This, along with the pressures described above, prompted urban craftsmen to develop new devices for limiting entry to their ranks. The result, inevitably, was widespread frustration and confusion in job choice.

Rural journeymen had long faced declining opportunities. In one home industry after another fathers at some point had to stop urging their sons into their own profession. By the turn of the century this was occurring in shoe-making and clothing manufacture particularly, but also in metal work and in pockets of rural weaving.¹³ Some sons gladly escaped into the factories, where they could earn good money.¹⁴ But far more were prompted by their own background and their fathers' advice to try to get a job in the urban crafts. For many, the job choice was thought to have been made before migration to the cities occurred.

Villages and small towns continued to maintain a very high ratio of apprentices to journeymen. In France, printers in towns of less than 5000 population had one apprentice per three workers, compared to one in five in the larger cities and one in seven in Paris.¹⁵ The same pattern prevailed in Belgian cabinet-making, British construction work, German locksmithing, and many other trades. When home-town opportunities dried up it was logical to try to move one's profession to a city. A large number of skilled construction workers in London, for example, had been trained in the provinces.¹⁶ But the chances of entry into the urban crafts grew more and more remote.

Large numbers of traditionally transient workers were settling down by the 1890s. There were still the East German village masons who went every summer to the larger cities to earn big money, but generally this pattern was declining. Parisian masons still came disproportionately from the Marche or the Limoges area, but many of them lived permanently in Paris and scorned newcomers from the countryside for their willingness to accept poor wages and working conditions.¹⁷ They, and even more the members of trades such as printing, worked on devices to keep newcomers out. Craft exclusivism is of course no new theme, but it is vital to realize how effective it was still around the turn of the century. Everywhere construction workers sought to define their skills with increasing precision. French construction workers turned against formal apprenticeship altogether, limiting entry mainly to their own sons whom they trained themselves. Glovemakers in Limoges won an agreement in 1904 limiting the training of apprentices to urban workers alone and forbidding the assignment of any work to new rural journeymen; for three years there was to be no new apprenticeship at all, save for the sons and brothers of existing urban workers. The number of apprentices in printing declined dramatically. Once 55 percent of the total in Germany, they formed only 26.5 percent of all printers by 1906, and less than that in the largest cities; the union-enforced scale of apprentices thus had a dramatic effect. And in Germany legislation supporting formal apprenticeship carried this process still further; the 1908 'proof of capability' law allowed artisan masters to test and judge apprentices, and only when these latter had reached 24 years of age, in any local profession in which a compulsory Innung had been formed.¹⁸

Artisans did not adopt limitation measures without some regret, for they could threaten other aspects of the defense of the profession. While most French journeymen resisted apprentices because of the competition they provided, a few expressed fears that their craft would

die out for want of proper training. The sense that good workers were declining because of changes in apprenticeship was widespread, and it could add to the anxieties older artisans felt about their whole way of work.¹⁹ But there is no doubt about what was actually happening. Entry to the most dynamic crafts was increasingly confined to the urban-born, leaving out the masses of rural aspirants. In Berlin printing, for example, 50 percent of all workers were born in the city itself, far above the overall population ratio; while in all of Germany only 29 percent of all printers had been born outside large cities, again a proportion well below the ratios even within the big-city population. In the early 1890s 81 percent of London bookbinders and 66 percent of all London printers had been born in the city, in contrast to 53 percent of factory laborers and 22 percent of all railway laborers.² ° And within the urban group, an increasing number of journeymen simply followed their fathers into the profession. Mobility was not cut off, but it was definitely restricted. Again in German printing, a union survey revealed that only 18 percent of the grandfathers of existing printers had been printers, whereas 50 percent of their fathers had been in this trade.²¹ There is little doubt that, were similar data on other crafts available, they would suggest the same pattern of increasing concentration within existing craft families.

Yet even this does not exhaust the complicated behavior of the urban artisanry. From a number of crafts came reports of sons' increasing reluctance to follow their fathers' career choice. In Grenoble glove manufacturers had to give special subsidies to induce entry into apprenticeship, for new, higher-paying industries lured youths away and made their parents loath to pay out the 200-300 francs that traditional apprenticeship would cost.²² In Dijon and elsewhere, children of furniture-makers were drawn into the iron industry. Some were encouraged by their parents, who were disheartened by the prospects of their own trade and convinced that apprenticeship had deteriorated anyway; they wanted their children to better themselves by earning more money. But others went against the wishes of their fathers, who thought that their ability to obtain jobs for their sons and provide them with tools at low cost should override any strivings for independence.²³ Here, and even in better-paying trades such as printing, the sense that sons were impatient with a training period and eager to earn right away was widespread. It is not far-fetched to suggest a considerable erosion of parental authority in these cases, a desire to choose work not because of tradition or the intrinsic nature of the work itself but because of its instrumental value in providing higher pay. The confusion

caused among older journeymen by the many changes in the techniques and organization of the crafts could only have furthered this partial rebellion. Even in Germany where, as we will see, the instrumental view of work was less fully developed than in France and Britain, there were indications that a growing number of families in the most prestigeous crafts sought to send their children into higher professions; one survey in Berlin and Leipzig found over 40 percent of the sons of all printers who were of working age serving as booksellers, technicians, teachers, lawyers, and bureaucrats. Less definitively, the same survey reported the impression that this kind of mobility had become newly-important.²⁴

We can, then, suggest a twofold movement in the urban crafts. Fathers and sons, or sons alone, were in many cases concluding that craft work had lost its validity. This impelled a career choice based on earning power and/or status in the community at large. The impression was — though it can only be an impression — that this kind of choice was spreading. It was naturally more widespread in countries in which artisanal traditions had been more thoroughly eroded, as in France in contrast to Germany. But everywhere a still larger number of fathers and sons worked with great success to preserve a career choice in which a young man would count himself lucky to follow his father's footsteps, for a greater percentage of the ranks of urban craftsmen were filled through this kind of choice than at any point since industrialization had begun.

And this in turn meant that more aspirants for craft jobs were disappointed than ever before. Even many of the young men who opted for the high wages of factory work probably were influenced by their fear that the crafts would be closed to them. Still more ended up in the factories simply because the jobs they really wanted were unavailable. Carl Severing described a distressingly common pattern. His father, a cigar maker, hoped that his son would become a preacher, while Severing himself wanted to be a musician. Relevant training was too expensive, to the family's great distress. So the father decided on a career as a tailor, then as printer; but printing proved impossible because of the number of candidates. So Severing became a locksmith, and after apprenticeship in a small shop (where he also learned his socialism) he ultimately found work in a sewing machine factory.²⁵ This kind of frustration was recounted by many German workers, who were most vulnerable because of the rapidity of industrialization and the sudden decline of the small town crafts, but it operated in all the industrial countries to some extent. In a period of increasing change,

the most logical job choice involved following one's father, where the paternal profession offered any kind of security and status. Where this choice was impossible, as in the many stagnating crafts, it was then most logical to seek a job in another, but economically more viable, craft. Yet this was more difficult than ever before. So many young men took work which was at best their third choice, and found it difficult if not impossible to shake off the frustration, the sense of declining status, which this kind of imposed choice placed upon them.

We find the frustrated journeymen in the factories above all, though not a few, hampered by some family disaster or their own ill health, tumbled into unskilled casual labor; the ranks of London dockers, for example, were swelled by refugees from decaying crafts in the East End.²⁶ In the factories the craftsmen tried to recreate much of the ambiance of the crafts themselves. German locksmiths prided themselves as being from the best families, using the familiar form of address only among friends, rather than with their workmates generally.²⁷ Limoges porcelain painters similarly insisted on being called 'Monsieur' by the ovenmen.²⁸ Certainly there was a widespread effort to limit access to the factory skill. British iron founders pressed for limitation of apprenticeship agreements, succeeding in Sheffield (where they cut the number of apprentices to one per three workers). British engineers In one dispute declared that 'we were determined that we would not in any way assist, instruct, or help those whom [employers] selected other than skilled turners.'²⁹ Much of the familiar history of internecine rivalries within a factory labor force stems from the kind of compensations workers sought for a career that was not their first choice. Some of course found genuine interest in their work, like the most highly skilled workers in precision manufacturing in Berlin; many of these further hoped to earn or learn enough to rise to independent status or into the company bureaucracy. But others, trained in a more traditional craft, could only feel that their profession had deteriorated as they found themselves locked into factory work.³⁰ The metal trades, particularly in Germany, provide a host of examples of this kind of anxiety. The workers characteristically had fathers who were journeymen and who had warned them against factory labor. They had been taught that factory workers were idle people, 'Bummler' or louts, who took no joy in their work, and as children some of them had seen the workers dragging home, dirty and tired, after a day in the factory.³¹ Then, for want of craft jobs, they found themselves in the factories. High wages might compensate to some extent, but these were not workers who found high wages appropriate as the principal basis for a

job choice; the instrumental view of work was forced upon them, not freely adopted, and the work itself was all the harder to value as it had been reviled in their own upbringing.

And everywhere the effort to maintain craft traditions in the factories came under increasing attack at the turn of the century. German furniture-makers, citing their long apprenticeship and their purchase of their own tools, resented other workers in the new factories who rivaled their earnings even though 'their technical ability was less.' In contrast to the crafts, few factory groups were able to limit training; the British Amalgamated Society of Engineers lost a bitter battle on this subject in 1898. Here is where -formal apprenticeship really declined, for where factories needed traditional skills they had only to recruit from the small town craftsmen.³² This was the approach taken by the rising furniture factories in Brussels and in German metal production, where former locksmiths were used so extensively.³³ Erosion of skills and job controls meant erosion of personal status within the factory. The skilled worker, often frustrated in his actual job choice, could easily sense a deterioration in the evolution of factory labor after about 1890, because it departed still further from the criteria he sought in a job in the first place. An exhaustive study of German textile and metal workers produced overwhelming evidence that the vast majority of these workers had had higher goals in their youth than they were able to realize in their actual jobs.³⁴

We can suggest, then, that the workers who felt most completely masters of their destiny were still to be found in the urban crafts, save where these were under direct attack by factory competition. These workers chose their jobs with a sense of their fathers' example before them, according to values promoted by an artisanal community that remained quite real. They had to fight to maintain their sense of control, particularly by limiting entry to their ranks, and they were by no means immune from changes in the nature of their work. But they were generally successful in preserving their status and their ability to base family continuity on continued access to the craft. Though a minority from this group would opt for more substantial social mobility or the lure of higher wages, most chose a job which would provide a combination of relatively high status, relatively high earnings, and considerable intrinsic satisfaction. This was the only large group of workers capable of this kind of choice, but far more wanted nothing better than to emulate it. Hence the theme of frustration echoed so often in individual statements by workers and in general, typically abortive movements to reproduce the same combination of values in

other settings.

Most groups of factory and transport workers had a considerable element drawn from the craft orbit, often (particularly in Germany) with direct craft training and even more widely with important elements of the craft mentality. Most German studies of worker origins revealed a far higher percentage of artisanal (including, almost certainly, domestic manufacturing) than of peasant ancestry, for factory labor; in a leather plant 40 percent of workers reported grandfathers who had been Handwerker compared to only a third from peasant stock, while in a textile factory in Eastern Germany workers with artisan fathers outnumbered peasants about three to one.³⁵ The fact that many of these workers clustered at the top of the factory labor force in terms of skill and earnings makes their potential disappointment all the more significant, for they were often in a position to translate it into active protest.

Against the artisanal background we can measure the other large group of relatively new entrants to factory life, clustered mainly in the lower skill levels. These were people from a traditionalist environment also, but a much more static one. Where would-be artisans customarily made a conscious decision to enter a trade, in close consultation with fathers and even community leaders, many of the unskilled had no pattern of choice before them. They might dream of an artisanal life, but their dreams would be quickly dashed. Were they better off for their relative freedom from expectations that they could control their lot? Was purposelessness — from the standpoint of occupational choice — better than frustration? The goals of many new workers were not really wrapped up in the job at all, for they expected no joy from work. Ironically, many of the unskilled were closer to an instrumental view of the job than were the workers with artisanal values. But we must look at their apparent motivations before we can even conjecture about the relative satisfactions they might derive, in comparison with the artisans, from a factory setting.

A study of textile labor in a Mônchen-Gladbach factory gives a precise, if statistically limited, framework for the type of workers who seemed, by their own admission, to drift through their lives of work. Only 38 percent of his workers could give a precise reason for their choice of industry. Almost 13 percent assigned need, rather vague in itself. Another 13 percent said they were following their fathers' profession, though in fact in a full 31 percent of cases the fathers had been textile workers, suggesting that following the parental example was less compelling in some factories than in the crafts. Only 12

percent said they chose textiles because it offered a good job. Thirty-five percent specifically said they could not account for why they took the job they did. This uncertainty was confirmed by the fact that 59 percent of all workers under 21 had already had two jobs and 41 percent more than two, while 23 percent of all workers between 22 and 30 had had five jobs.³⁶ Less precise data suggest that many other workers would have had similar difficulties explaining why they did what they did. Many German miners, brought in by company recruiters, were so disappointed at what they found that they embarked on a life of constant job-changing. One young worker at 17 years of age left a Ruhr company to go to Holland; he evinced no sign of making a conscious choice of profession, simply hoping to see the ocean. He never made it, for he was arrested by the police as a tramp before he crossed the border; ultimately he ended up back in the mines.³⁷ In Ruhr mining generally, a full 40 percent of the sons of miners left the mines; this does not mean they necessarily lacked a job choice but it suggests at the least the weakness of traditional paternal example.³⁸ And if the evidence on aimlessness is most substantial in Germany, again because industry was newest there, the phenomenon was widespread. General unions in Britain were based in part on the fact that semi-skilled engineering workers, for example, had no sense of choosing a job and thus felt kinship with building laborers; there was in fact some drifting from one job to another, while in both cases laborers could rarely hope to rise into skilled categories.³⁹ The growth of semi-and unskilled factory labor everywhere, associated with the more sophisticated new machines, denotes the rise of a group at least unaccustomed to the notion of selecting a job. Without pretending to assign a percentage to the group overall, we can agree with contemporary social scientists who noted that many workers seemed to do whatever came along.⁴⁰

Three types of workers were most likely to be unable to assign a reason for their holding the job they did; for most members of each category the whole notion of explaining job choice was really foreign. The unskilled form the largest group. Their lack of choice was not new, but evidence available at the end of the nineteenth century allows some new insights into motivation; and in certain respects their situation had changed dramatically from that of the early industrial revolution. Within the unskilled category large numbers of older workers were more directly deprived of choice. And the growing number of female manufacturing workers probably fell within the 'choiceless' group, though for rather special reasons. Each of these categories deserves

some exploration before we return to the question of job choice more generally.

Let us begin again with an extreme, albeit a statistically significant one. Into France, Germany, and Belgium poured a host of foreign workers who were bent on making as little choice as possible.⁴¹ This is ironic, given the fact that in retrospect the decision to move to the city and to industry seems so monumental. And unskilled migrants now moved farther in search of work than skilled workers did, a railway-based reversal of early industrial trends.⁴² In the Daimler plant in Stuttgart, for example, a third of all day workers came from some distance, compared to only a fifth of the locksmiths.⁴³ The movement of foreign workers simply heightened this general trend. We cannot of course pretend to fathom what prompted particular migrants to move, as compared to their contemporaries who remained in the village. Once arrived in the industrial labor force, however, what stands out is the sense that extensive geographic movement was cushioned by the lack of emotional abandonment of the past: work in industry was meant to fulfill traditional goals. Thus many unskilled workers, like the Polish miners employed in the Ruhr, were bent on earning a nest egg to buy land back home. Of course they had some desire to earn more money — hence a decision to go to mines in the Ruhr rather than stay in Silesia where wages were lower — but their acquisitive instinct was limited. They had no intention of remaining workers permanently, even though many did precisely that. They did not move into industry to better themselves — a number of observers noted that Flemish miners were devoid of ambition, in what was really an extreme of a typical pattern — but to restore something that had been lost or was imagined to have been lost. The purpose was to go back, to endure the present in order to restore a position back home. A few extra enjoyments resulting from relatively high monetary pay, such as the ability to buy cigarettes, did not lead to quick conversion to an acquisitive standard of living or to a really instrumental view of work.

The need to avoid choice and the desire to preserve a traditional pace of work was illustrated by frequent returns to the home setting. Among 4105 Italian workers in the Longwy steel industry in 1903, for example, there were 1618 departures for home; the next year, 1589 departures from among 4148 workers. Most of these workers deliberately left their families at home, because of the firm intention to return. Their intermittent work was the product of earnings higher than traditional expectations and their revulsion at factory labor — both instances of the fact that no choice had been made, in terms of

commitment to a new job, but that poverty alone had pushed these workers to a new setting, supplemented in some cases by a recruitment agency and rather formless expectations of easy earnings.⁴⁴ And they were able regularly to send significant savings home, affirming their moderate personal expectations and their ties to the past; 500,000 francs were annually sent to Italy from the Villerupt post office alone.⁴⁵

The situation of foreigners simply exaggerated the characteristics of native unskilled migrants. German shipbuilding, traditionally the province of local craftsmen, was supplemented during the 1890s by masses of unstable rural workers, many of whom regularly left to do agricultural work. The fact that this migration heightened, if massively, a traditional seasonal influx of workers from the Weser confirmed the minimal amount of choice involved in the process.⁴⁶ Breton workers, sometimes recruited as strikebreakers, entered the Parisian transport and construction fields and the mines of the Nord. Some remained, though keeping contacts with the home village from which they continued to receive food (thus enabling them to accept low wages and avoid any sense of class consciousness). But where the nature of the work was really novel, as in the mines of the Nord as opposed to the casual labor of construction or hauling, upwards of three-quarters of the recruits left within four months.⁴⁷

Of course there were casual laborers whose situation was less clear-cut. The German merchant marine was flooded by immigrants from many areas, and we cannot presume to determine the motivations involved at this point. Big cities, particularly in Britain, spawned a variety of unskilled workers from products of broken homes, the ill, and the retarded, and while these people had little choice of job they lacked the continuity with tradition that many unskilled migrants still tried to maintain.⁴⁸ And other migrants, even though without skills, had newer motives in mind; British agricultural laborers, attuned to market incentives, moved to the docks and the mines quite frankly because of a desire for steadier or higher earnings and showed no disposition to leave.⁴⁹ But, particularly in the factories where the need for conscious choice of job would seem most obvious, little career decision was really involved. There was strong motive to leave the village, little knowledge of what work was in store, and a great desire to use the work to restore a pre-crisis village life. Because the work was so strange few migrants could persist to achieve their original goal, but their durable adaptation to industrial labor — if they stayed in manufacturing at all - was a protracted process that had little to do

with original motives.

As we have seen, the massive influx of women into manufacturing was a phenomenon common to almost all the industrial countries. In Britain between 1891 and 1911 the number of women in manufacturing and transport work rose by 40 percent, while the increase of women in the overall population was barely half that rate (at 24 percent). Between 1895 and 1907 the increase of women in German manufacturing was 51 percent, with the bulk of the gain coming after 1900. By 1910 there were 33 percent more women workers than in 1905, and their expansion in manufacturing jobs was double the rate for men.⁵⁰ This was a development of tremendous importance in working-class life, for it involved an unprecedented commitment to women's working outside the home. It is increasingly recognized that despite the horrors of female labor in the early factories - indeed perhaps because of them — most working-class women did not work outside the home during the initial stages of industrialization. Their earning activities were much more commonly devoted to servanthood, domestic manufacturing, or caring for lodgers. The breakthrough to factory and office work that occurred toward the turn of the century was thus genuinely novel; for the first time the majority of working-class women spent at least a portion of their lives working outside a domestic environment.⁵¹ Here, even more than with the long distance migration of many unskilled workers, we must ask about the choice involved.

Some women in the textile industry could undoubtedly have explained why they held the jobs they did. They could look back to a long tradition of work and they feared the boredom they would face if they were confined to the home; as a Lancashire spinner said, when asked why she did not devote herself to household and children: 'It 'ud give me the bloomin' 'ump.'⁵² Some women who had worked as servants had a definite sense of alternatives. Workers who came to Bristol confectionery from domestic service found 'more life' in the factory. German women who made a similar switch noted that, 'the foremen were not nearly so coarse as the gracious ladies.' Other German women who claimed that industrial jobs conveyed higher status than servanthood, in part because of some opportunity to rise to supervisor level, were making an even clearer career choice in the factories.⁵³ But none of these constituted a normal pattern. The women in textiles who kept on after marriage were a rarity among working women; overall, only 10 percent of all married women worked in Britain, and many of these did so only because their husbands were dead or ailing. The

percentage was rising slowly, but in all the industrial countries it remained true that the typical working-class wife did not hold a job outside the home and that the typical female factory worker was an unmarried young woman.⁵⁴ Nor did many women come over to industry from servanthood. The number of female servants declined (in France after 1895), stagnated (in Britain), or grew less rapidly than before (in Germany), but this resulted mainly from women not entering the occupation in the first place. It is safe to assume that they had some awareness of the alternatives before them, and chose factory work because they anticipated greater freedom and higher pay.⁵⁵ But they were not, in most instances, actively choosing jobs.

For women still worked to supplement family income and usually on a transitory basis. Not a few came from the ranks of artisans and even professional people; in several German factories the social origins of women workers were higher than those of men, though in sweated industries the daughters of dockers and construction laborers predominated.⁵⁶ Certainly in the former case, parents did not expect massive earnings from their daughters, allowing them to avoid the most arduous jobs; many young women thus preferred weaving to spinning, despite lower pay, because the work was less demanding.⁵⁷ Even in the sweated trades of London most girls were working for pocket money and, perhaps, to escape their parents' domination during the day.⁵⁸ None of this minimizes the significance of this change in the lives of young working-class women, or the fact that they made an important decision in going to work. And of course male opinion was normally hostile to women in industrial jobs. Ardent socialists, despite a theoretical commitment to women's rights (or perhaps in part because of it) could urge women workers to return to the home and take care of their children, thus fulfilling their proper role and allowing men's wages to go up.⁵⁹ A French typographer expressed the same view in talking of his female colleagues: 'They're too impulsive. One day they're enthusiastic for our work, the next day they've changed their minds. The only way to resolve this problem would perhaps be by absorption. Each male typographer would marry a female typographer. That would be a solution.'⁶⁰ This climate, combined with the newness of the industrial work experience and the tradition of temporary jobs before marriage explains the lack of job choice, but the fact remains that few women had much sense of selecting a career in industry.

Thus the departure from industry into marriage was a logical, completely expected pattern, for the job had been endurable in large measure because it was known to be temporary. The same mentality

prompted British girls, when asked about problems on their jobs, to talk more about tensions at home, with their parents.⁶¹ In Britain and elsewhere women changed jobs more frequently than men, and as with the male migrants this was both symptom and cause of a lack of commitment to a clear job choice. Even low levels of unionization among women, although owing much to male resistance, were related to this absence of professional commitment on the part of this rapidly growing segment of the labor force.

If, then, we focus on an industry with large numbers of women (aside from some traditional female industrial occupations such as cotton textiles — but even then perhaps only in Lancashire) or large numbers of unskilled migrants (like German mining, or construction labor generally) we have gone some considerable way to explaining, in admittedly gross terms, the large minority of workers who could not express a sense of choice in discussing their jobs. Both groups had made important decisions: to move a long distance, or to work outside the home, but where they found themselves in industry was almost irrelevant to this decision. In fact what they experienced on the job differed greatly from their expectations precisely because they had lacked the basis for a definite career choice. So many of them drifted. Or, unable to move readily - one wonders about the single women who were unable to marry — they lived what must have been confused, unhappy lives at work. But there is a final category of choiceless workers, less readily identified by contemporaries or in retrospect, that lacked even the luxury of making an initial decision to quit a traditional environment. Large numbers of older workers could have explained their position in the work-force only by the necessity of survival; their range of alternatives had narrowed to virtually nothing.

For many workers, old age, from an occupational standpoint, began in the mid-'thirties, certainly by forty. British railway companies, in a fairly typical example, rarely hired anyone over thirty. Unskilled workers could get jobs after they reached forty only with difficulty though they had to keep trying.⁶² British metal workers, claiming that work speed-ups caused premature aging, lamented that too many of their employers judged them 'too old at forty.'⁶³ By no means all older workers found themselves in serious trouble. Some died before reaching even a young old age — deaths helped account for the low percentage of older workers in mining — and as usual in history it is difficult to count the dead against the living. Others flourished reasonably well in industry, a point to which we will return. In terms of our present concern, essentially choiceless older workers can be found in several situations. Many

were buried in the countryside, clinging to domestic manufacturing jobs or even entering such jobs afresh because they could find no other work. Ignored by all but some meticulous government investigators, they were assumed to be contented because they were incapable of protest and rural, but in fact they were often barely surviving and quite aware of their choiceless plight. We have noted already that departures of young workers plus accretions of the old left sectors like rural clothing manufacture, in Belgium, or ribbon manufacture, around Saint-Etienne, in the hands of workers over forty, who were normally underemployed. ⁶⁴ This recourse had little value for workers in the big cities or for British workers generally; and the number of older workers increased dramatically with improvements in longevity and the increase in population growth through mid-century. Hence a variety of unskilled occupations developed or expanded to provide some casual employment to older urban workers, many displaced from decaying crafts. Taxi-driving was one, and the field rapidly became overpopulated. Parisian drivers included former valets, artisans, and mechanics, while in both Paris and London there were about 50 percent more drivers than cars by World War I.⁶⁵ Displaced older workers also headed for the docks. Here too their lack of real choice in a final job was matched and fed by the casual, irregular nature of the work.⁶⁶

Within the factories older workers were threatened by two related forces: the possible irrelevance of their initial skill and the pressure of work speed-up (both enhanced by diminished physical vigor). The latter provoked the loudest complaints around the turn of the century, particularly in Britain. Alfred Williams, in a locomotive factory, wrote vaguely of the employers' tendency to fire older workers. British railroad workers complained that senior workers were passed over in a desire to promote young hotshots, and special anger was directed against eye tests that could result in dismissal of older engineers. The cotton spinners' union criticized a tendency to fire older workers to make room for the promotion of piecers — the product of increased longevity as well as the speed-up drive.⁶⁷ Piecework hit many older workers hard. In Parisian masonry it increased production enough that some older workers could be fired, while the stronger, young workers responsible for the increase simply enjoyed their higher earnings. The older workers who kept their jobs found their pay reduced, as employers insisted on tying wages to product. Similarly, older German workers complained that their earnings were below those of a rank beginner, again the fruit of the confrontation between piecework and diminished strength and skill.⁶⁸ And there were more general

complaints that large companies were less loyal to older workers than small, constituting another ominous trend.⁶⁹

These pressures were exceedingly important, and while not entirely new they were undoubtedly on the rise. Many older factory workers suffered declining earnings (though some family expenses might be down as well) and severe blows to their self-esteem. And there was some outright unemployment even among healthy workers. Yet in terms of the question of job choice it would be inaccurate to generalize too glibly about older factory workers. They were not dismissed en masse. Only 25 percent of the unemployed in a York study were over forty — a percentage lower than the size of this age group in the working population. Among British iron workers, unemployment did rise for men after their thirty-eighth year, but it was highest for those in their early twenties.⁷⁰ In fact, as we shall see, British labor had developed protection for older workers that speed-up could dent but not overturn; the vigor of the complaints reflected expectations that age should be defended in factory jobs. Older workers were undoubtedly conscious of the need to watch their step, because of the difficulty of finding new work, but they were not under constant threat.

The situation was rather different among older Belgian, French and German factory workers. There is some evidence that older workers formed a larger percentage of the unemployed than in Britain; in France in 1906 they were twice as likely to be without a job.⁷¹ Certainly those who kept jobs tried to work steadily; it was noted that their production stayed even or rose a bit in the spring, whereas younger workers, distracted by the season, let it drop notably.⁷² They were less likely to change jobs: in one textile factory, where 23 percent of all workers between 22 and 30 held over five jobs, only 16 percent of the workers over 40 claimed the same.⁷³ Above all, large numbers of older factory workers found themselves demoted to the ranks of the unskilled. This was of course endemic in mining, where workers were brought to surface jobs at age 40. But Belgian and German mine-owners exacerbated the problem by using older pit workers for hauling, whenever they could hire strong young replacements.⁷⁴ Even more significantly, report after report on German metal and engineering factories suggested that the bulk of the unskilled labor force was composed of workers over thirty-six, who had lost their original skill rating and had given up all hope of improvement, and who formed an unusually stable element because their occupational survival was so tenuous. In France in 1911 29 percent of all male workers over 65

were listed as general laborers.⁷⁵

There can be no pretense at measuring the overall size of this group of older workers or at comparing it, save impressionistically, with its British counterpart. But what was going on was fairly clear. Against the general backdrop of speed-up, workers who lacked strong union protection and who came into factories with artisanal skills, often learned in the village and initially useful, found themselves displaced by younger workers with more up-to-date techniques. Lower nutritional levels played a role as well, sapping the strength of many German, French or Belgian workers earlier than their British analogues.⁷⁶

In sum, older workers were forced to change jobs without real choice, or more rarely to suffer outright unemployment, in two general settings. Declining crafts released a number of workers to the ranks of casual labor. Older factory workers might be pressed - and in some industries and areas regularly were — into the unskilled category, where, unlike many other unskilled workers, they could not even risk the luxury of switching jobs or taking time off. As with women and migrants, industries with unusually large numbers of older workers contained an unusually large number of workers incapable of explaining any choice, of job. Such would undoubtedly have been the case on the docks, had any pollster ventured there. Such was definitely the case in textiles in comparison with metallurgy, for a considerably larger proportion of the textile labor force was composed of older workers,⁷⁷ some displaced elsewhere and drawn by the lighter demands of the industry and others demoted within the industry itself.

Describing the situation of the choiceless workers gets us only part of the way to an understanding of their relationship to their job, but because most of these workers were truly inarticulate it is difficult to penetrate much further. They rarely protested because they lacked the job expectations to form the appropriate basis; and it is no easy matter to judge whether lack of protest reflects satisfaction or the most profound alienation. The older workers just described, locked into mediocre jobs in German factories, would not have claimed choice in their work, but this does not necessarily indicate intense dissatisfaction. For they might have expected old age to be a time of deterioration - even today workers rate the early twenties as the peak period of life — and have been grateful for surviving as they did. Locksmiths in the Daimler plant commonly talked of staying at their job 'as long as I can' — a fairly clear reference to anticipated problems in the later years of work.⁷⁸ Immigrant workers could derive pleasure from helping their families at home and relieve tension by changing jobs

and areas. Women, who were given some chance to speak about their jobs, manifested relatively little discontent, and in this case often talked of positive enjoyment at the social contacts possible in factory work; but this reflects the fact that, of all the 'choiceless' groups assessed, women had the greatest sense of alternatives in entering factory work and were more probably pulled rather than pushed into their jobs. But none of the groups expected much durable, intrinsic satisfaction in the job itself. If they were contented - and it will remain difficult to determine this basic point — it was because the job was n ot troublesome, in terms of expectations of workload, not because it brought pleasure and even more because of compensations peripheral to the work itself — social contacts, earnings, or sheer survival.

It would be possible, combining the two large groups thus far described, to paint an exceedingly bleak picture of workers' ability to choose a job, outside of the increasingly restrictive crafts. Frustrated at one end because of artisanal expectations, choiceless at the other, could factory workers feel they had a role in selecting their job? In fact, while both mentalities are exceedingly important, they do not describe the bulk of the factory population. A final venture into contemporary polling can introduce us to the mainstream of attitudes toward job selection.

The Daimler automobile plant near Stuttgart drew most of its workers from nearby towns and villages (again, only among the unskilled did significant numbers travel long distances); 44 percent of its workers were from villages under 2000, though among the top skilled workers (the locksmiths) the urban element was much larger. The average worker was 32 years of age and had been with the company four years; the skilled element was noticeably younger than the average, at twenty-nine years, but it had the same longevity. Artisanal backgrounds loomed large. This was, in sum, not a completely atypical factory setting, in terms of the elements we have already described, and though it had a greater home town atmosphere than the big factory cities of the north this might merely have heightened the frustration at factory life. Indeed some skilled workers did express a desire for the kind of training that would allow them to enter an independent trade, while admitting that they lacked the money they would need for this. But the vast majority of all workers — and particularly of the skilled group — said that they chose their profession according to their own free wish or the recommendations of their parents; the largest single response was that the profession was enjoyable. In contrast to the textile factory cited earlier, most of these

workers had always been in their current profession and most said they intended to stay in it.⁷⁹

Even among relatively new industries in Germany, there was thus considerable diversity in the willingness of workers to express job choice. If we cast our net wider, admitting the lack of precise polling data, to more settled industrial countries the sense of diversity can only increase. Two problems, then: to account for the reasons workers might have for believing they had chosen their jobs, despite the considerable potential for randomness or frustration, and to provide greater precision on the differences in choice within the working class.

To begin with, it is easy to exaggerate the frustration element. There is no doubt that far more German workers wanted to become artisans than were able to do so and could thus tell an investigator that their factory job was not their first choice. But three qualifications to this pattern, apart from the obvious fact that the dilemma was greater in Germany, as a newly-industrial country with a rapidly growing labor force, than in Britain or probably even France. First, we must not exaggerate the universality of acceptance of artisanal life. Apprenticeship could be exploitative at worst, dull and confined at best. Many were glad to escape it. One German metal worker was trained by his father, a master baker, from the age of ten; he hated the life. At 16 he began to wander, taking various unskilled jobs and often begging; finally he gained a permanent factory job at 28.⁸⁰ Second, workers who expressed a sense of disappointment at not having followed their first choice may have been developing a kind of ambivalent culture that has taken firm root in the working class. Recent surveys⁸¹ of the English working class suggest the ability to express career aspirations for selves or children that are almost explicitly admitted as unrealistic. German workers at the turn of the century could similarly talk of their desire to be a bureaucrat or an independent master and then note that they had no real control over their lives, that they would go 'where the wind blows us.'⁸² The larger aspirations were not insignificant, but they did not preclude a sense of positive choice, even of quiet satisfaction, in the job that had been acquired. And finally, in many factory industries the artisan mentality, as we can understand its survival, could find considerable expression. There was the need to grit one's teeth and accept a large workplace and considerable direction from above; some former artisans and peasants would always find this distasteful. But if this could be done, the criteria for job choice did not have to undergo total change.

There was, first of all, a widespread program of training. Formal

apprenticeship was under increasing attack and promotions of the unskilled posed a growing problem in many industries, particularly engineering and shipbuilding, and this trend must not be neglected.⁸³ But even in 1914 a considerable majority of all manufacturing workers were rated, and doubtless would have rated themselves, as skilled. The sense of choice already discussed in the Daimler plant was obviously related to the fact that skilled workers outnumbered unskilled and semi-skilled by almost two to one. British engineering, hard pressed by new methods, had a labor force that was 60 percent skilled by 1914 (the remainder evenly divided between unskilled and semi-skilled). And in many areas threat to skills in one company might be compensated by the development of new demands elsewhere; French Lorraine, for example, cried out for skilled metal workers throughout the pre-war years.⁸⁴ The threat to skills was real. It could be projected for the future even when not yet operative. Yet it was still possible for many workers to choose a factory job confident that it would involv e training and skill. And we should not minimize the inventiveness of individual workers in protecting this sense of training, even when union efforts at defending apprenticeship programs failed. Many German machine workers, for example, changed jobs explicitly to learn more; turners wanted experience in building steam engines, saying that this would not only increase their earning power but improve their satisfaction with their careers.⁸⁵

After training came promotion possibilities. Considerable attention has been given to the traditional mobility within the artisanry, from apprentice to journeyman to master, and how this was increasingly blocked in the early stages of industrialization. American investigators have devoted great effort to the study of more modern kinds of mobility, defined occupationally in terms of movement from one major job category to another, as from unskilled to skilled.⁸⁶ But what has been missed is a process somewhat in-between, that could be a normal life experience for a given factory worker, sometimes susceptible of being passed on to children, and quite possibly a very sensible pattern for workers who remembered the artisanal process which was now closed to them. What follows does not, of course, involve the nominal tracing of individuals that is essential to establish mobility definitely, and we will return to the problem of mobility more generally in a later section. For now, what can be established is that many workers could choose a factory job because it provided a fairly well-organized ladder of promotion covering a considerable portion of one's working life. Workers at Krupp, predominantly from artisanal backgrounds in the

1860s, saw a clear ladder of advancement established from one skill level to the next. And their descendants could rise higher still; only about a third of the children of second-generation unskilled were in the same category, whereas a full half were skilled and the remainder were in the company bureaucracy. And while this does not account for the workers who left the company, job stability was sufficient to give the sample significance. Metal-working and metallurgy everywhere involved advancement after the training period, even if, as with traditional artisans, this sometimes required explicit defense. A similar pattern was found at Daimler, where all the children of skilled workers old enough to hold jobs had risen higher than their fathers, either in the company or outside.⁸⁷ Carmaux metallurgists struck to protect the advancement of rollers by seniority. British smelters complained — successfully — that the customary promotion of the oldest changewheelers to furnace work had been ignored by one company.⁸⁸ This kind of training and promotion system in metal work could produce tensions quite similar to those discernible among artisans. Assistants in British iron manufacture (and in cotton) were reluctant to agitate lest this damage their chances for promotion by seniority; their concern was directed not only at employers but also at the skilled workers who supervised them, many of them their own fathers, and who therefore were free to dominate the union. Sons of Belgian metallurgists were reported avoiding their fathers' profession in favor of taking training as turners, where they could earn money faster, a job-choice dilemma that again tends to confirm the assimilability of factory and artisanal skills.⁸⁹

It is obviously impossible to suggest how many workers chose an industrial job with promotion in mind, or how many benefited from it, aside from citing the general figures on skill levels. And it must be remembered that only certain workers could hope to get on the ladder at all. Just as construction laborers were doomed to unskilled work, so railway maintenance men were apart from the careful seniority system used for the operating crews.⁹⁰ And not all of the major industries had clear systems at all, to compare with those in many kinds of metal work and on the railroads. But British and French mining had a definite ladder, from assistant or boy through hauler to miner, with the latter graded at five levels or more. The German system here was more confused, because of massive new entries of workers, except for the possibility of rising to master miner at the top; no matter at what age they were hired most workers moved after ten years as a hauler to an apprentice miner, and then on to miner when they were capable of

digging coal on their own — a ladder that may not have given much sense of movement, but which avoided complete stagnation.⁹¹ Textiles allowed movement from piecer to spinner or apprentice to weaver, and again there is evidence of a firm sense of movement within a given factory; workers in a rug factory in Neuville, in the Nord, struck against the hiring of an adult rug weaver who was not in the weaving 'corporation,' noting that this would prevent apprentices from rising. They ultimately beat up the worker involved, which seemed to settle the matter. Individual workers might choose jobs explicitly for the opportunity to rise. Contrasting his trade with coal mining, a German metallurgist proclaimed: 'If I work hard and remain with the company I can be foreman and perhaps even later even supervisor.'⁹² Many kinds of industrial jobs, then, could provide not only training but a sense of subsequent advancement, which in certain instances — railroads being the clearest case — gave seniority advantages and possibilities for supervision of assistants even late in the working career.

Finally, as has been obvious from the discussion of promotion itself, factories would quickly develop systems of favoring the children of existing workers, providing another kind of advantage that artisans sought. This is one of the reasons that a substantial minority of factory workers would explain their jobs in terms of parental choice; and there is no question that parents actively sought the possibility of assuring work for their children. In many Belgian metal factories, the children of existing workers were almost the only new workers hired. Not only metal workers but also tram and railroad workers often agitated for confirmation of the rights of children over outsiders, and a similar system was clearly at work in many textile centers.^{9 3} In Arrnentières linen weaving 78 percent of all workers were related to another textile worker in the city in 1901, 42 percent in the same factory; few children (3 percent) went into other industries. Where fathers had a son at work, 70 percent of the sons were weavers, most of them having started as their fathers' assistants. Parental guidance had its ambiguities for both parties. In an isolated textile area, and even more in the mines, parental example might operate only because there was no other work ready to hand, and we cannot be sure that it provided many workers with an active sense of job choice. In the Rhondda valley, 95 percent of all workers depended on the mines. Some children, hearing their parents complain about their labor, wanted an alternative but could find none. Others, who also had no sense of choosing among jobs, were nevertheless drawn to the mines by the possibility of earning relatively good wages early and by the mystique of work, the challenge to

manhood, that they developed as boys.⁹⁴ Lacking direct inquiries into miners' choice, it is impossible to sort those who felt a positive commitment from those who could have explained their job only by the absence of anything else. But the probability is that the group who could explain why they were in the mines increased with time. This is surely one reason that a growing percentage of sons followed their fathers into the Ruhr mines after 1900 despite an earlier tradition of seeking other work and the availability of factory jobs in the general area.⁹⁵ After the World War, it would also be a factor in the reluctance of miners' families to turn to other work even as the mines declined.

Overall, the ability to choose a job or at least explain why one had a particular job increased with the maturation of industry. The sense of frustrated choice did not disappear, but it is no coincidence that around 1900 it was far stronger in Germany than in Britain. Many workers, particularly with craft backgrounds, came to the cities with a desire to reproduce the best features of artisanal life, and without question not all of these could be found in the factories. Independence, in the artisanal sense, was impossible; it could only be ironically echoed by frequent job-changing. Other elements of artisanal job choice could be reproduced, but it took some time either to realize or to develop this. Promotion systems, particularly ones that would protect older workers, were even by 1914 less clearly established in Germany than in Britain. But as workers settled down they could develop a system of job choice which included parental guidance and example, a sense of training and some hope for a lifetime career.⁹⁶ Not all workers settled in this way, of course; and those who did undoubtedly narrowed their horizons somewhat. But within this framework many industrial workers could claim some understanding of why they were what they were.

Because many workers did seek to choose a job, but with somewhat varied criteria and important differences in their ability to make a selection, certain industries tended to attract distinctive types of workers. This process would be most apparent in a country like Germany, which was rapidly expanding its manufacturing labor force, but its results were still apparent in England, because an important minority of workers did move from one industry to another and because parental example could perpetuate certain attitudes and physical characteristics. Information is inadequate to cover the whole industrial roster, but a few key examples can illustrate how the process of job choice, blending with the technical requirements of various industries, produced a variegated working class.

Railroad service was consciously selected by a large number of

workers — as with many artisan jobs, its popularity greatly exceeded the number of jobs it could provide. In France there were 30,000 applicants for approximately 1300 vacancies each year, and these were carefully screened to select the most 'stable' types. As noted, pension systems were general as well, attracting workers with foresight for their old age. Many families sent several members into railroad work, which encouraged pride in the calling. It is not far-fetched to suggest that many applicants were rather stolid types in the first place, for the big attraction of railroad work was the steadiness and security of work, along perhaps with fringe benefits such as uniforms and, in Belgium and Germany, the prestige of being in state service. Promotions played a role, as we have noted, attracting workers with foresight for their old age ; pension systems were general as well. High pay might come at the end of a cureer, once one became an engineer, but it was not an obvious draw. Railroad work, and to a lesser extent employment on tram lines, where job tenure was also a realistic goal, tended thus to recruit a cautious, security-minded work-force, and in so far as the work was passed from father to son some of the same qualities were transmitted as well.⁹⁷

Metal work also drew workers who felt they had a job choice, but their interests were somewhat different. Early high pay was more important, concern for the long run rather less so; hence the average age of entry was low, compared to metallurgy or even mining, and young workers changed jobs so often. Cliques around particular skills divided the category still further, for if metal workers did not have the obvious trappings of prestige, recognizable to the larger society, that railroad workers might boast, they had a vigorous sense of place within their own community. They distinguished themselves by residence — the better-paid often shunning company housing to avoid bad associations — and by conversational groups in the factory. But status concerns aside, this was a more flamboyant work-force than the railroads employed, proud of their skills and their physical strength, touchier in their relations on the job.⁹⁸

Textiles provide a different picture still. There were good reasons to choose textile work, especially in places like Lancashire where parental example was obvious. But the percentage of textile workers who sensed a job choice was lower than in metallurgy. There were more women, more elderly, and skill levels were not as high even for young males. Textiles thus drew a labor force that was less literate than that in the crafts or metallurgy, and above all one that was physically weaker. In Germany only 29 percent of the textile workers did military service, most of the remainder excused because of physical ailments. Textile

workers were thus more likely to be disappointed by a failure to succeed in initial job choices or incapable of articulating a conscious choice at all. Even in Lancashire the physical health of the cotton workers was inferior to the working-class norm, if only through inherited characteristics exacerbated by relatively poor care for children and a tradition of relatively early child labor."

Obviously, distinctions of this sort are extremely general; they do. not predict the characteristics of any given worker. But the tendency for certain kinds of workers to gravitate to certain kinds of work deserves serious attention, for it helped shape the nature of each particular work-force. The characteristics involved expectations and the very ability to account for one's presence on a job, but also, in an age of uncertain nutrition and medical care, important physical traits as well. Differences might be enhanced by the actual work experience; for example, the reasons the artisans and metal workers chose their jobs were quite similar, but the subsequent behavior of the two groups divided by the organization of work imposed on the metal workers. But the initial job choice would continue to color the work experience itself, affecting job stability, protest, even family patterns to a degree. By the same token differences in job choice, among factory industries as well as between artisans and the unskilled, constituted serious divisions within the working class.

For the simple fact was that workers could readily perceive those of their workmates who were on the job for different reasons than they were. In the mines workers who were drawn to the peculiar excitement of mine work or following their parents' lead encountered unskilled newcomers whose presence had nothing to do with the work itself. Ethnic hostilities owed much to this kind of distinction; the Walloon miners who railed against 'Flemish cowherds' perceived the newcomers' strange motivations as well as their language and religiosity.¹⁰⁰ Tension was equally obvious in the Ruhr mines. Silesian newcomers thought that more experienced local workers were odd, if not reprehensible, because they seemed to be seeking something more than a daily grind, while the locals ridiculed the Silesians for their willingness to drag through every day so long as they were able to earn little treats like cigarettes.¹⁰¹ Neither judgement was fully accurate; Silesian workers, for example, might drag through many days but they were more likely to quit their jobs frequently. And judgements of this sort reflected more than differences in job choice. But what was sought in a job would affect ongoing experiences, so differences had lasting impact.

In stressing the range of job choice available we are not trying to

suggest some illusory degree of freedom. Direct evidence o>f opinions comes from only a few cases in Germany, itself not the most typical mature industrial setting by the turn of the century. Opinion and reality might differ; what workers told a bourgeois investigator might, whether deliberately or not, be intended to appeal to his sense of appropriate rationality, not to what really happened; though even the calculation of what was an acceptable reason for holding a job has considerable interest. After grappling with the workers who admitted they could not articulate on this subject, one must be wary of accepting the more communicative workers at face value. And many of these workers did not explain their job in terms of a choice among alternatives. Some who did weigh alternatives were doomed to disappointment, for the risk was great for the worker who thought beyond his parents' station. And even those who did think in terms of alternatives — the boys who were placed as printers rather than carpenters or the young men who became turners, in an engineering plant, instead of ovenmen in a smelting factory — operated under severe constrictions. Apart from the limitations on any human choice, workers had to select a career with obvious disadvantages in placement contacts, interregional mobility, funds available for training, and in many instances physical stamina.

Yet the sense of choice was by no means idle. Even a decision to enter industry could contrast favorably with traditional pursuits, as in the case of many women; or it could be taken amid the compulsions that induced a lonely migration. Choice of specific job could frustrate. It could fulfill a parent's dream. It could lead to an anticipation of satisfaction, the result depending on individual personality but also some broad categories of background and industrial setting. The reasons for taking a job were strong enough to distort a labor market, for within a range of economic compulsion — the need to survive — workers were not simple economic men. The desire to remain artisans retarded the formation of a large, skilled factory labor force in Germany or in the newer industrial areas of France. It kept printing skills alive that were no longer technically necessary. Related motivations made railroad recruitment so easy that wages stayed low despite a growing work load.

For on balance the criteria of workers who could explain their choices were rather conservative. We must not take the purely artisanal yearnings too seriously. Many workers had to choose the factories and could do so with a sense of positive commitment. But whether from artisanal background, as was so often the case in Germany, or the

product of several generations of factory experience in which independent criteria could be established, workers were rarely in a position to strike out too far. The increased ability to explain why one had a particular job, which we have asserted for the more mature industrial period, was not the fruit of an obvious modernization of working-class outlook. It was not the expression of great individual venturesomeness or the desire to maximize individual opportunity.¹⁰² Rather it reflected a desire to regain some job values that had been threatened during the earlier stages of industrialization.

This appears most obviously in the role taken by parents. Just as artisans continued to guide their sons, so factory workers could begin to think in terms of placing their boys. Correspondingly the workers most likely to lack specific job commitment were those in which parental referents were missing: the migrants, the older workers displaced from their chosen job, and women, for whom parents had not yet developed occupational (as opposed to marital) criteria. The working-class family was an economic unit still, and the activity of parents not only in advising but in placing their children followed from this.¹⁰³

Closely related to parental guidance were other criteria expressed in job choice: social status, at least within a working-class community, and a hierarchy through which some advancement was conceivable, even normal. Regularity of work was an important factor, though some kinds of workers would stress it more than others; artisans, for example, were less concerned than railroad workers. Provision for employment in old age might have been a logical consideration for security-conscious workers, but traditional guidelines were shakier here and the expectation of survival did not keep pace with the fact. Workers developed a concern for what would happen in their later years after some time on the job, more than in their initial decisions.

But this in a way merely confirms the desire to seek in a job many of the satisfactions traditionally associated with successful work: status, parental approval, and — as the Daimler workers noted so often — some intrinsic pleasure in the work itself. This was not, in so far as we can determine, an instrumental view of work. The view that a job should be taken not for its own sake but for extrinsic benefits was most common, in however inchoate a form, among new, unskilled workers, who saw their work as a means of returning to the countryside or escaping parents/finding a mate. Of course wages were part of the satisfactions that articulate workers said they sought in taking a job. And the absence of explicit mention of wage levels as a reason for job choice in

the German polling materials - it is virtually complete, despite an appropriate spot in the questionnaire — may relate to the recency of German industrialization, which tied the workers to an unusually traditional set of justifications. We have seen important signs of a tension between the parentally-approved reasons for taking a job and a youthful desire to earn money quickly. This was surely one reason that certain skilled trades, such as construction work in London, had constantly to be replenished by small-town trainees. Complaints by French artisans and metallurgical workers that the younger generation pursued the quick franc more directly express the potential clash among job criteria, as does the significant variation in the percentage of young people in the prosperous trades and industries. But the wage-based choice was nowhere yet predominant. Even where the absence of job alternatives was deplored, as among some Welsh miners, few young workers were sufficiently venturesome to seek their fortunes elsewhere. This was still the product of unusual necessity, as with the foreign migrants. Correspondingly the pecking order of occupations, for big-city workers who had the clearest range of alternatives, was not completely correlated with earnings. Jobs that offered other satisfactions, such as many crafts and railroad service, could rate above those with much higher average wages. High wages might in fact have to be offered precisely because the job was unpopular according to the general criteria workers used, and even then they might not suffice to eliminate a serious labour shortage.

Hence this stage of industrialization can be associated most clearly with a widening of a sense of job choice according to sensible but rather conservative criteria. This pattern was threatened less by youthful rebellions — which were by no means necessarily new -— than by disruptions of skill levels within the factories. The rapid influx of semi-skilled workers into engineering, metal work, and even crafts such as printing raised a host of problems about the future course of job choice. Could the semi-skilled themselves develop a sense of selecting one job over another, particularly given the hostility of those more experienced, skilled workers who might have served as logical models? Would the threat to established skills be sufficient to force even the most settled sector of the labor force to reconsider its reasons for selecting a manufacturing job? These problems were not yet predominant by 1914, but they were clearly visible and their significance was all the greater for the challenge they posed to the carefully-reconstructed framework within which most manufacturing workers had learned to operate in selecting a job.

And this raises the final question: if most workers by 1914 had not yet explicitly switched to a new system of job choice or to a belief that there was no reason for one job as opposed to another, how would their criteria coincide with their actual work experience? The relationship was frankly complicated. There was almost surely à correlation between those workers who could express reasons for job choice and those who felt satisfaction with their job. German metalworkers, for example, were much happier with their work than their colleagues in textiles, just as they more commonly believed they had chosen their jobs.¹⁰⁴ The same would surely have applied to artisans as opposed to the unskilled. But there was a massive gap between the figures for job choice and those for job satisfaction, in both metals and textiles. It would be ridiculous to play with precise statistics, and the biases of the pollsters — conservative/social-scientific in the one case, socialist in the other — explain part of the gap. But the gap had reality in the workers' experience as well.

For the very fact that many workers had developed or retained criteria for choosing a job gave them bases for protest when the job did not turn out as expected. Their protest could relate to the job itself, in an effort tp convert it into a more satisfactory experience, or it could assume more instrumental lines, when workers essentially gave up on their initial expectations and decided that their only hope was to pry more earnings out of an unpleasant situation. On the whole, therefore, a relationship developed between workers capable of job choice, and the industries in which they were most concentrated, and rates of protest.¹⁰⁵ In this sense, as in its relation to the work experience itself, the growing ability to decide on a job was a vital, if complicated, part of the maturation of the working class.

Notes

1. In Belgium it became virtually impossible to find apprentices in domestic industries such as clothing, for workers rarely urged their children to continue in the profession. Ministère de l'industrie et du travail, Les Industries à domicile (Brussels, 1908), IX, 89. With an aging labor force, natural attrition tended to reduce the number of workers. Friedrich Behr, Die Wirkung der fortschreitenden Technik auf die Schuhindustrie (Leipzig, 1909).

2. On the contrast between the irregularity of agricultural work and the relative stability of jobs in industry, Fritz Darmstàdter, Die Lage der Arbeiter im Kalisatzbergbau (Munich, 1911), passim.

3.  In Germany former agricultural workers complained that farmer-employers insisted on being called Herr, abandoning formerly friendly behavior at joint meals. C. Moszeck, éd., Aus der Gedankenwelt einer Arbeitsfrau (Berlin, 19Q9), 22.

4.  Bruno H. Bûrgel, Vont Arbeiter zum Astronomer (Beilin, 1927).

5.  Ministère du travail et de la prévoyance sociale: office du travail, Enquête sur le travails à domicile dans l'industrie de la chaussure (Paris, 1914).

6.  Bernhard Adeling, Sein und Werden (Offenbach, 1952).

7.  Robert Tressell, The Ragged-Trousered Philanthropists (London, 1955), 471.

8.  Le'on de Seilhac, Syndicats ouvriers, fédérations, Bourses du travail (Paris, 1902), 57.

9.  A. E. Musson, The Typographical Association (London, 1954), 231.

10.  Ibid., 253.

11.  Commission syndicale du parti ouvrier et des syndicats indépendants, Rapport présenté au XlVe Congrès syndical extraordinaire du 28 Avril 1912 (Brussels, 1912), 83 ff.

12.  Musson, Typographical, 258; Archives départmentales du Rhône, Series M, General strike by Lyons dyers, 1903.

13.  Ministère de l'industrie et du travail, Industries, IX, 89.

14.  B. Seebohm Rowntree, Land and Labour: Lessons from Belgium (London, 1911), 85 ff.

15.  Ministère du commerce; office du travail, Rapport sur l'apprentissage dans l'imprimerie (Paris, 1902), lxviii.

16.  Charles Booth, Life and Labour of the People in London (London, 1895), VI, passim.

17.  Amedée Dunois, 'Le Lockout de la maçonnerie,' Pages libres (1908), 52.

18.  A. Sartorius von Walternhausen, Deutsche Wirtschaftjgeschichte 1815-1914 (Jena 1923); Raymond Joran, L'Organisation syndicale dans l'industrie des bâtiments (Paris, 1914), 94 ff.; Max Morgenstern, Auslese und Anpassung der Arbeiterschaft (Leipzig, 1912), passim; P. Cousteix, 'Le Mouvement ouvrier limousin de 1870 à 1939,' L'Actualité de l'histoire (1957), 74 ff.; Fanny Imle, Die Tarifvertràge zwischen Arbeitgebern und Arbeitnehmern in Deutschland (Jena, 1907), passim.

19.  Ministère du travail, Chaussure, 105, 307.

20.  Walter Abelsdorff, Beitrage zur Sozialstatistik der deutschen Buchdrucker (Tubingen, 1900). In 1907 40% of the population of Germany's big cities had been born there. Wolfgang Kôllman, 'Industrialisierung, Binnenwanderung, und "Soziale Frage" ', Vier-teljahrschrift fur Sozial-und Wirtschaftsgeschichte (1959), 53; Booth, Life, X, passim.

21.  Hans Hinke, Auslese und Anpassung der Arbeiter im Btichdrucke-rgewerbe (Berlin, 1910), 35 ff.

22.  Grenoble et sa région, 1900-1925 (Grenoble, 1925), 464 ff.

23.  Ministère du Commerce, office du travail, Rapport sur l'apprentissage dans les industries de l'ammeublement (Paris, 1905), 525,531 ff., 634.

24.  Abelsdorff, Beitrdge.

25.  Carl Severing, Mein Lebensweg (Berlin, 1950), I, 13 ff. Even unskilled workers could briefly harbor craft aspirations. An East German laborer described his desire to be a locksmith, though this was completely beyond any realistic attainment. After a stint as an agricultural worker he ended in construction labor, hating his job at every stage. Luth Franz, Aus der Jugendzeit eines Tageslohners (Berlin, n.d.), 19.

26.  Paul de Rousiers, The Labour Question in Britain (London, 1898), 336 ff. In Germany many unskilled workers were former shoemakers, bakers and the like who felt their status loss heavily. Heinrich Reichelt, Die Arbeitsverhàltnisse in einem Berliner Grossbetrieb der Maschinen-industrie (Berlin, 1906). For an individual case in Germany see Wendel Holek, Lebensgang eines deutsch-tschechischen Handarbeiters (Jena, 1909).

27. Paul Gohre, Drei Monate Fabrikarbeiter (Leipzig, 1913), passim.

28.  Maurice Kahn, 'Les Evénements de Limoges,' Pages libres (1905), 9-10.

29.  H. A. Clegg et al., A History of British Trade Unionism since 1889 (Oxford, 1964), I, 430; Sidney Pollard, A History of Labour in Sheffield (Liverpool, 1959), 239 ff.; A. L. Levine, Industrial Retardation in Britain (London, 1967), 477.

30.  Clemens Heiss, Auslese und Anpassung der Arbeiter in der Berliner Feinmechanik (Leipzig, 1910), 111.

31.  R. J., 'Erlebnisse eines Metalldrehers,' Thunen-Archiv (1909), 218 ff.

32.  Pollard, Sheffield, 211 ; Darmstâdter, Lage, passim.

33.  Archives de la ville de Bruxelles 197—2, report on cabinet-makers' strike, 1912.

34.  Marie Bernays, 'Berufswahl und Berufsschicksal des modernen Industriearbeiters,' Archiv fur Sozialwissenschaft und Sozialpolitik (1912), 123 ff.

35.  Max Morgenstern, Auslese und Anpassung der industrieller Arbeiterschaft betrachtet bei den Offenbacher Lederarbeitern (Freiburg i. Br., 1911); Elise Hermann, Auslese und Anpassung der Arbeiterschaft in der Wollhutsindustrie (Munich, 1912).

36.  Marie Bernays, Auslese und Anpassung der Arbeiterschaft der geschlosene Grossindustrie (Leipzig, 1910). German textile employers correspondingly could not count on children 'naturally' entering the industry: Schulz-Gaevernitz, The Cotton Trade in England and on the Continent (London, 1895).

37.  Adolf Levenstein, Aus der Tiefe, Arbeiterbriefe (Berlin, 1905).

38.  Max Metzner, Die soziale Fursorge im Bergbau (Jena, 1911). Part of this stemmed also from restrictions on child labor in the mines; some of the 40% may have returned later.

39.  E. J. Hobsbawm, 'General Labour Unions in Britain, 1889-1914,' Economic History Review (1949), 138. A building laborer by age 24 could expect to remain a laborer all his life. Booth, Life, VI.

40.  Pollard, Sheffield, 230 ff.; Heinrich Herkner, Problème der Arbeiterpsychologie (Leipzig, 1912).

41.  In France in 1911 14% of all transport laborers, 13% of all construction laborers, 9% of miners, and 5% of metal and woodworkers were foreign. Georges Mauco, Les Etrangers en France (Paris, 1932), 49.

42.  On early industrial patterns, Arthur Redford, Labour Migration in England (London, 1926); John Modell, 'The Peopling of a Working-Class Ward: Reading, Pennsylvania, 1850', Journal of Social History (1971), 71-96.

43.  Fritz Schuman, Auslese und Anpassung der Arbeiterschaft in der Automobilindustrie (Leipzig, 1911). There were, to be sure, some isolated factories in which the unskilled were recruited locally, living at home and trying to preserve as much as possible of their traditional peasant life; but this was now uncommon in manufacturing. The main exceptions occurred in Belgium, where the extensive commuting network allowed many industrial workers to spend evening or weekends by their customary patch of earth. H. Demain, Migrations ouvrières à travers la Belgique (Louvain, 1919), 70 and passim. For more limited German examples: in one Magdeburg factory in 1898, 97% of all workers lived in their native villages, up to 23 kilometers away from the job. Hermann Beck, Lohn- und Arbeitersverhàltnisse in der deutschen Maschinenindustrie (Dresden, 1902). In Offenbach 78% of the leather workers were from the area, almost 50% of them rural and continuing to live in the countryside: Morgenstern, Lederarbeiter, 24 ff.

45.  Léon de Seilhac, 'Les Grèves du bassin de Longwy,' Musée social (1906), 390 ff.

46.  Josef Neumann, Die deutsche Schiffbauer (Leipzig, 191 0).

47.  de Peyerimhoff, 'Les Charbonnages français,' Musée social (1913), 148; H. Leduc, 'A Villeneuve-Saint-Georges,' Voix ouvrière (1910), 658 ff.;AftMéejocw/(1908), 291.

48.  E. Fitzer, Die wirtschaftliche und technische Entwzcklung der Seeschiffsfahrt (Leipzig, 1902). A study of the unemployed in York noted that over a tenth were locals under 19 years of age, who had entered a blind alley job, serving for example as delivery boy, and lacked the strength or wit to seek alternatives once their exploitable boyhood was over. B. Seebohm Rowntree and Bruce Lasker, Unemployment; A Social Study (London, 1911) 49.

49.  Royal Commission on L&boxxx,Minutes of the Evidence, Group 'B' (Transport by Water and Transport by Land (London, 1892, C 6708), I, 392.

50. Abstract of Labour Statistics (London, 1915, Cd. 7733), 293—319; Franz Hitze, Geburtenruckgang und Sozialreform

(Mônchen-Gladbach, 1917); Arthur Feiler, Die Konjunktur-

Periode 1907-1913 in Deutschland (Jena, 1914). 51. 77% of all unmarried women over 15 were employed in Britain

by 1911: Peter N. Stearns, 'Working-Class Women in Britain

1890-1914,' in M. Vicinus, éd., Suffer and Be Still (Bloomington

"1972), 110. 52 Mrs. J.R. MacDonald et al., Wage Earning Mothers (London, n.d.),

14.

53.  Royal Commission on Labour, The Employment of Women (London, 1893, C. 6894), 36; Heinrich Herkner, Problème der Arbeiterpsychologie (Leipzieg, 1912); Bernays, Auslese, 171.

54.  Steams, 'Women;' Hitze, Geburtenruckgang, passim. In Germany the massive increase in women working in industry was accompanied by an^- actual decline in the percentage of married women in the manufacturing labor force, from 6% to 5%.

55. Theresa McBride, Rural Tradition and the Process of Modernization: Domestic Servants in Nineteenth-Century France (Unpublished doctoral dissertation, Rutgers University, 1973) demonstrates that a strict economic calculation before the late nineteenth century would have sent women into servanthood, where earning possibilities were greater than in the factories; but after the 1870s the balance shifted. Former German servants reported a similar finding by 1900.

56. Booth, Life, passim; Hermann, Auslese.

57.  F. Hecht et al., Die Stôrungen in deutschen Wirtschaftsleben (Leipzig, 1903), II, 77 ff. and passim.

58.  Stearns, 'Women,' passim.

59.  Léon de Seilhac, Les Mariniers du Nord et leur dernière grève (Paris, 1905), 151.

60.  Archives départmentales du Nord M621/31; see also A. and Z., 'Pour la Réduction des heures du travail,' La Revue socialiste (1906), 131 ; Fédération française des travailleurs du livre, Dixième Congrès national (Paris, 1910), 53.

61.  Royal Commission on Labour, Women, passim.

62. W. T. Stephenson, 'The Railway Conciliation Scheme, 1907,' The Economie Journal (1911), 505; Rowntree and Lasker, Unemployment, passim.

63.  Friendly Society of Iron Moulders, Monthly Journal, 1907.

64.  Ministère de l'industrie et du travail, Industries, IX, passim; Pierre Clerget, Les Industries de la Soie en France (Paris, 1925).

65.  Public Record Office HO 55943, report of 1890; L. Bonneff and M. Bonneff, 'La Grève des taxi-autos,' La Grande Revue (1912), 353 ff.

66.  Edward G. Howarth and Mona Wilson, West Ham (London, 1907).

67. J. E. Williams. The Derbyshire Miner (London, 1962), 304; Amalgamated Society of Railway Servants, The Railwaymen's Charter (London, 1907), 8—17; General Railway Union, Executive Council Meetings (1905), 111; S. J. Chapman, 'Some Policies of

the Cotton Spinners' Trade Unions,' The Economic Journal (1900), 467.

68.  J. Lapierre, 'Les Grèves des platrières du bassin de Paris,' La Vie ouvrière (1912), 129; Cari Fischer, Denkwùrdigkeiten und Errinerungen eines Arbeiters (Leipzig, 1904); A. Picart, 'La Grève de la maçonnerie parisienne,' Musée social (1909), 381.

69.  Dora Lande, 'Arbeits-und Lohnverhàltnisse in der Berliner Maschinenindustrie zu Beginn des 20. Jahrhunderts (Leipzig, 1910).

70.  Rowntree and Lasker, Unemployment; Friendly Society of Iron-founders, Investigation into the Mortality, Sickness and Unemployment Experienced among Members (London, 1913).

71.  Louis Variez, Recensement des chômeurs gantois (Glient, n.d.).

72.  Ernst Bernhard, 'Auslese und Anpassung der Arbeiterschaft,' Jahrbuch fur Gesetzgebung, Verwaltung, und Volkswirtschaft im Deutschen Reich (1911), 1426. It may be more than a random contrast to note that many older skilled British workers took a day or two off during the week to rest: Arthur L. Bowley and A.R. Burnett-Hurst, Livelihood and Poverty (London, 1915), 111.

73.  Bernays, Auslese.

1 A. L'Ouvrier mineur (1907), 17; N. Dethier, Centrale syndicale des travailleurs des mines de la Belgique (Brussels, 1950), 127.

75.  Bernhard, 'Auslese,' 1405.

76.  On nutrition levels, a detailed and reasonably exact comparison is provided by the several studies of the British Board of Trade: Report on an Enquiry into Working Class Rents, Housing and Retail Prices . . . in the Principal Industrial Towns of Belgium (London, 1910, Cd.5065), of France (London, 1909}, Cd.4512); of Germany (London, 1908, Cd.4032. Whether so many workers were demoted within factories in France as in Germany is not clear; as many workers crop up, after age 65; in metals, textiles, and construction as the overall manufacturing average; Slightly lower percentages in metallurgy as mining relate to formal pension plans as well as wearing work. But from most industries some older workers flowed into the casual labor force. See Chapter VIII; estimates based on the 1901, 1906, and 1911 censuses.

77.  8.8% over 55 years of age in French textiles, compared to 6.2% in metallurgy and 6.1% in mining in 1901; Charles Benoist, 'Le Travail dans la grande industrie,' Revue des Deux Mondes (1902), 853.

78.  Fritz Schumann, Auslese und Anpassung der Arbeiterschaft in der Automobilindustrie (Leipzig, 1911), 111.

79.  Ibid., Ill and passim.

80.  Richard Ehrenberg, 'Schwàche und Starkung neuzeitlicher Arbeitsgemeinschaften,' Thunen-Archiv (1911), 415 ff.

81.  John H. Goldthorpe et al., The Affluent Worker in the Class Structure (Cambridge, 1969).

82.  Peter N. Stearns, 'Adaptation to Industrialization: German

Workers as a Test Case,' Central European History (1970), 303 if.

83.  Lande, Arbeitsverhaltnisse; A. L. Levine, Industrial Retardation in Britain (London, 1967); E. Backert, Geschichte der Brauarbeiterbewegung (Berlin, 1916).

84.  Levine, Retardation; Schumann, Auslese; Lucien Brocard, 'La Grosse métallurgie française et le mouvement des prix de 1890 à 1913,'.Revue d'histoire économique et sociale (1922), 394.

85.  Ehrenberg, 'Schwâche,' 517 ff.

86. Theodore Hamerow, Restoration, Revolution, Reaction (Princeton, 1956); Stephan Thernstrom, Poverty and Progress: Social Mobility in a Nineteenth Century City (Cambridge, Mass., 1964); Clyde Griff en, 'Occupational Mobility in Nineteenth-Century America,' Journal of Social History (1972), 310-30.

87.  Richard Ehrenberg and Hugo Racine, Krupp'sche Arbeiter-Familien (Jena, 1912); Schumann, Auslese.

88.  Adrien Veber, 'Le Mouvement social,' La Revue socialiste (1900), 367; British Steel Smelters, Mill, Iron and Tinplate Workers Association, Monthly Reports, 1913.

89.  H. A. Clegg, General Unions in a Changing Society (Oxford, 1964), 3—6; Jules Lekeu, A Travers le Centre; Croquis et moeurs; enquête ouvrière et industrielle (Brussels, 1907), 32. J. R. Clynes described how often agitation among cotton piecers was dampened by the criticism of the spinners, who were typically their fathers or uncles: Memoirs (London, 1937), I, 57.

90. Amalgamated Society of Railway Servants, General Secretary's Report (1905), 11. On the British railways one normally moved from a training period to fireman, where one served for eight years or so before expecting to rise to engineer; cleaners could ultimately hope to rise to foreman's positions, while brakemen and signalmen had their own hierarchies.

91.  H. Stanley Jevons, The British Coal Trade (London, 1915), 335; Adolf Levenstein, Aus der Tiefe, Arbeiterbriefe (Berlin, 1908), 21—2; David Crew, 'Definitions of Modernity: Social Mobility in a German Town, 1880-1901,' Journal of Social History (1973), 51-74.

92.  Adolf Zimmermann, Von Haspe biz Duisburg. Industrielle Reisebriefe (Berlin 1912); see also Archives départementales du Nord M625/111, report of September 1904; John Jewkes and E. M. Gray, Wages and Labour in the Lancashire Cotton Spinning Industry (Manchester, 1935), 93.

93.  Lekeu, Centre, 32 ff.; Edward J. O'B. Croker, Retrospective Lessons on Railway Strikes (London, 1 898), 11.

94. Wil Jon Edwards, From the Valley I Came (London, 1958), 142 ff.; Jevons, Coal, 596 ff.; William Reddy, 'Family and Factory: French Linen Weavers in the Belle Epoque,' Journal of Social History (197'4).

95. Die Bergarbeiterzeitung, 1899 ff.; Heinrich Mvinz, Die Lage der

Bergarbeiter im Ruhrrevier (Essen, 1909).

96.  Joan Scott has described the settling process for Carmaux glass workers, where the percentage of workers born locally rose dramatically in the 1880s and 1890s. 'The Glassworkers of Germany 1850—1900,' in Stephan Thernstrom and Richard Sennett, Nineteenth Century Cities (New Haven 1969), 3-42. The contrast between British and German mining, in teims of the stabilization process, is equally striking, though German miners were picking up the same trend. Up to 70% of all new Scottish and up to 85% of all new Welsh miners were sons of miners, and a large portion of the older workers had been many years with the same company; only 40% of the Ruhr miners were sons of miners even by 1914, and in representative companies only 19% of the miners had 10 years or more experience. B. McCormick and J. W. Williams, 'The Miners and the Eight-Hour Day,' The Economic History Review (1959), 239 ff.; Gerhard Adelmann, Quellensammlung zur Geschichte der sozialen Betriebsverfassung: Ruhrindustrie unter besonderer Berùcksichtigung des Industrie-und Handelskammerbezirks Essen (Bonn, 1965), II, 103.

97.  R. Godfernaux, Aperçu de l'évolution des chemins de fer français de 1878 à 1928 (Paris, 1928), 32: Charlotte Leubu-scher, Der Arbeitskampf der englischen Eisenbahner im Jahre 1911 (Munich, 1913), 21 ff.; Archives nationales (France) F 713826.

98.  Paul Gôhre, Drei Monate Fabrikarbeiter (Leipzig,- 1913); Otto Jeidels, Die Methoden der Arbeiterentlôhnung in der rheinisch-westfàlischen Eisenindustrie (Berlin, 1907).

99.  Max Morgenstern et al., Auslese und Anpassung der Arheiterschaft (Leipzig, 191-2); The illiteracy rate in French textiles in 1901 was 15%; in metallurgy 6%; in printing, 2%: Ministère du commerce, Résultats statistique du recensement général de la population (Paris, 1904); Jewkes and Gray, Wages.

100. Eugène Rousseau, La Grève générale dans le Centre en 1913 (Ghent, 1913), 6.

.01. Stearns, 'Adaptation,' passim.

.02. This point needs some possible qualification for workers who strove to escape manual labor altogether, a topic tak&n up in the Conclusion; but their number was not vast.

.03. On the economic roles of the stabilized working-class family, see Michael Anderson, Family Structure in Nineteenth-Century Lancashire (Cambridge, 1971).

.04. Levenstein, Tiefe, passim.

05. The one possible exception, and a massive one, is mining, in which direct-action protest occurred at the highest rate of all. There i_s too little information available about miners' job choice to determine if it had anything at all to do with protest. German miners undoubtedly chose their jobs less often than British, and their protest rate was much lower — though still high compared to protest rates of other German industrial categories. British miners

may have chosen their jobs, if only through parental example, and then rebelled because they failed to find what they were seeking, or they may have entered without alternatives and protested steadily as a result. There are, of course, many other factors involved in the massive strike rate of miners in all the industrial countries. Clark Kerr and Abraham Siegel, 'Interindustry Propensity to Strike — An International Comparison,' in A. Kornhaus, R. Dubin, and A. M. Ross, eds., Patterns of Industrial Conflict (New York, 1954).

CHAPTER 3 UNEMPLOYMENT

Unemployment, like the choice of a job, touched almost every aspect of the work experience. The vast majority of manufacturing and transport workers — except those with jobs in the mines and metallurgy — could expect to undergo some unemployment during their work career, and many would be frequently unemployed and for long periods of time. This stark fact requires a careful assessment of the nature of unemployment, despite its relationship to topics that will be discussed subsequently such as technological change. The subject is exceedingly complex, which is doubtless why it has been avoided by most students of the working class heretofore. No simple set of factors explains unemployment during this period; direct displacement by machines, for example, played only a small role, despite many workers' impressions that they were the principal villains. Complexity is compounded when we treat differences from one industrial category to the next; we must again deal with artisans in quite different terms, dividing them between small-town and urban, various groups of factory workers and the unskilled: the rates of unemployment varied widely. Within this framework we must try to determine what unemployment meant. For a few workers it was deliberate, for most a great burden. But some of those most severely burdened in a material sense were not surprised by unemployment, while others learned to fight it vigorously, some with considerable success. And, as with the choice of jobs, there was a certain evolution at work as industry itself matured, in which many types of workers learned to bring unemployment under some controls save in years of outright crisis. Fatalism most assuredly declined, although everywhere the fact or fear of unemployment cut severely into the stability of working-class life.

It is distressingly difficult to establish the incidence of unemployment with any precision. Given the difficulty with present-day unemployment figures we could expect many problems in the interpretation of unemployment statistics, in terms of seasonal, sex, and age adjustments. In fact, for this period, the figures themselves are often close to useless. Their existence is tantalizing, and we will utilize the outline they provide for several purposes. But we must turn quickly to more limited examples and impressions that provide no more general

statistical precision than the materials on job choice. A word, first, about the sources themselves.

All the industrial countries provided annual and monthly unemployment figures by 1900. The series sometimes changed categories, but at least numbers are available. They derive, unhappily, from trade union records; only in France, during this period, was unemployment checked as part of the census process. But the trade union figures covered only a minority of the working class. They were heavily biased toward the skilled. Unskilled unions were small and normally composed of unusually stable workers. This could lead some government statisticians, as in Germany, to claim that skilled workers were unemployed more often than unskilled, which overall was nonsense.¹ Even among the skilled there are serious questions about the representativeness of the membership sample. Did unemployables flock to unions to receive benefits? Or as was more probable, were union members among the healthier, diligent workers and therefore less likely to be unemployed than the average in their trade? Union figures undoubtedly minimized unemployment overall and possibly even in highly unionized trades. Furthermore, apart from the limitations of their staff work, unions had different motives in reporting unemployment figures. German unions tended to minimize rates because of the responsibilities they had in the social security system. French unions may, for doctrinal reasons, have exaggerated their rates. All told, the British unemployment figures are most reliable because of the size and experience of the unions and the fact that they had little to gain from exaggeration in either direction. Belgian and German figures are ludicrously low and deserve attention only for their indications of orders of magnitude over time and among industries; they cannot be seriously compared with the British. French figures seem more reasonable, despite the small size of unions, and the census check supports their utilization.

One final problem, with any of the sources on unemployment, relates to the reaction to unemployment itself. Many workers were desperately eager to avoid being called unemployed. The anxiety comes out strongly among Belgian workers, who were in so many cases close to the countryside and conscious of their neighbors' scorn for failure. A study of the unemployed in Ghent noted how few of the jobless registered at the Fonds du Chômage, because they felt they had 'dropped out of humanity' if they admitted their situation.² The workers who were still rural residents could often conceal unemployment by working in the fields. Some were claimed to welcome

the chance to catch up on home chores. But unfortunately for those observers who noted the dignity of workers who stayed home instead of pacing the streets, peak unemployment rarely coincided with the chance even to putter in one's garden; the observer who in December 1910 explained away a 7 percent unemployment rate in two of the backward provinces (Louvain and Nivelle) that supplied commuting workers for heavy industry, although far more accurate statistically than the government sources, simply ignored the seasonal factor. He was misled not only by the myth that things can't be too b ad if they're rural but by the workers' own sense of dignity.³ A similar desire to refuse to admit unemployment, plus the sheer necessities of survival, prompted many unskilled or even skilled workers to change jobs when unemployment struck. Hence in East Germany many textile workers did building trades work during the summer, at considerable cost to their professional skill; their own work was slack in this pe riod, and by leaving machines unmanned they believed they reduced the chance of overproduction later. These workers, too, would rarely report unemployment, for alternatives in rough work were preferable to a confession of failure.⁴

One can only suggest the kind of culture that produced this reaction, to which we must return in discussing the fears of even more sophisticated workers. A long tradition of underemployment in the countryside combined with village scorn for idleness and th difficulty of appealing to communal resources produced the culture, which may then have been enhanced by bourgeois promptings about the virtues of self-reliance. There is no question that it distorted the description of effective unemployment in many areas and industries. At the same time, the culture was declining around the turn of the century. Many artisans had long been loath to take work outside their skill, preferring unemployment instead; hence many engineering worker in Berlin factories cited the lack of 'suitable' work, not of work itself, in explaining their frequent unemployment.⁵ This mood spread in the factories. Alfred Williams described the unwillingness of fitters and boilermakers in his locomotive plant to take work outside their trade ; they preferred wandering, even begging.⁶ More and more workers came to see unemployment not as a personal failure but as a fault in the system, and their anger mounted as a result. Demonstrations by building trades workers in London, between 1901 and 1906, occurred amid unemployment that was undeniably uncomfortable, but relatively mild by traditional standards; what was new was the reaction.⁷

All of this means that unemployment figures were most accurate in

countries like Britain, and industries like machine building or printing, where workers defined their dignity in terms of their skill, and not the mere fact of holding a job. It means that the accuracy of the figures probably improved with time, not only because union membership became more widespread, which increased the reporting sample, but because more and more workers stopped seeing unemployment as a matter of shame and began to press for solutions. But the new mood could tend less to correct than to reverse the statistical problem: from underreporting workers might turn to exaggeration. Certainly the fears of unemployment stressed by the most articulate workers did not correspond to the discernible statistical trends; they mounted much more quickly. The repeated laments of skilled British workers - much more widespread than those in Germany — did not reflect a comparative difference in actual unemployment of comparable magnitude. It stemmed instead from a consciousness of unemployment as a social problem that was only beginning to spread in Germany or Belgium. This, too, complicates the task of the historian forced to go beyond the general statistics to more impressionistic sources.

National unemployment figures produce two unsurprising, if significant, conclusions, and one of high probability which is less expected. Unemployment oscillated greatly from year to year, by as much as 300 percent. It varied far more from one industry to the next, and this gives even yearly national figures quite limited utility in measuring the human fact of unemployment. But if we can attribute any value to the national figures they suggest that unemployment was undergoing an interesting decline during the years before World War I, most marked in France and Belgium while somewhat debatable in Britain where stability rather than decline is a more probable conclusion. This differential trend coincides with the thesis that can be offered concerning the overall relationship between maturing industrialization and unemployment: that the advance of industry could be compatible with apparently low rates of unemployment, where the economy of the countryside was not massively disturbed — the possible case in Germany; that more normally it ate into the rural bases of employment, swelling joblessness among the unskilled particularly — the Belgian and even more the French case early in our period; but that further advance would reduce unemployment levels, not only because of abstract processes of the economy but also because of actions by workers themselves, including new kinds of job choices, while leaving workers still highly vulnerable to economic crises — the British case and increasingly the French and Belgian.

Before dealing with differentials and trends, however, it is vital to establish the range of unemployment during the period as a whole, and this is precisely where the national figures are least helpful. Between 1891 and 1913, official statistics⁸ revealed an average unemployment rate of 4.3 percent in Britain (excluding the sick, and superannuated, and strikers), 2.2 percent in Germany, 2.9 percent in Belgium, and 7.0 percent in France (none of these last figures distinguishing so carefully among types of joblessness). The role of economic crises is rather clearly delineated by the national figures. Unemployment rose massively (by 80 percent) between 1891 and 1892 in Britain.⁹ The effects of the 1902—3 slump, so widely discussed at the time, were more muted; unemployment rose 21 percent in Britain, 23 percent in France, and scarcely at all in Germany in the first year of the crisis. The unskilled were hit harder than this, of course, and even unionized workers were much more seriously affected in many centers; for many, the employment effects of the crisis were felt more seriously later, in 1904-5— this was particularly true in the building trades, which influenced the national figures substantially and help account for their otherwise surprising pattern. The recession of 1908—9 was more sharply felt, partly because union reporting had improved but also because unionized workers were more vulnerable than before. The increase between 1907 and 1908 was only 25 percent in France, but 81 percent in Germany, 111 percent in Britain, and 156 percent in Belgium. In all these cases — and particularly in the slumps of the mid-nineties and 1902—3 — official figures were only the tip of an immeasurable iceberg. In Germany in 1901, for example, metal workers in Berlin reported up to 25 percent unemployment, plus vast amounts of short time, while the official figures showed the most modest national rate. In 1908, similarly, German unions on their own spoke of jobless rates of over 15 percent; and here as in the other countries the unskilled were scarcely counted.¹⁰ Thus, at a conservative estimate, between one in ten and one in five workers was out of work at any point of a crisis year, and between 1891 and 1914 there were at least eight crisis years.

Within a single year of normal prosperity, seasonal fluctuations were severe. They were endemic in the building trades. While about 5 percent of French construction workers lacked a job during the summers, the rate rose to ten percent or above during the winters. Seasonal variation was a bit less in Britain, at 6.3 percent between January and May. But fluctuations in Germany were still more radical, in part because of more severe winters but in part because of the poor organization of

small-town construction work. Hence in 1909 — admittedly a recession year — unemployment of unionized workers was 2.9 percent in small towns and 5.6 percent in large during the summer, but rose to 38 percent and 22 percent respectively by December.¹¹ Winter unemployment was also high in quarry work.¹² Obviously winter unemployment could be palliated by taking other jobs; many German masons learned to serve as brewery workers during the winter. But increasing rationalization made this particular interchange less feasible, and more and more masons were reduced to small repair jobs periodically during the winter.¹³ Of more lasting utility, in some areas, was the fact that dock employment peaked in the winter, but this could draw workers to these jobs who would find themselves on the streets again come spring. Seasonal factors in other crafts were different, but only a bit less severe. Employment among furniture workers varied 169 percent during an average year in Britain. Printers encountered greatest joblessness in the summer; their employment rates varied as much as 90 percent during a normal year (the range in Britain was from 3.3 to 6.3 percent during the years from 1900 through 1914).¹⁴ Jewelry and precious metal workers were victims of seasonal irregularities in the popularity of their products, as were the semi-skilled workers in the sweated clothing trades.

Seasonal factors were much less significant in the factories than in the crafts or among unskilled workers, but they were far from negligible. Shipbuilders reported long waits between jobs after a ship was completed, though overall metal and engineering work was relatively stable, ranging in Britain by 24 percent from a low in May to a peak in December. Workers in the light industries, such as textiles and shoe-making, suffered more seriously from seasonal variations in demand. These were reflected more commonly in short time than in outright unemployment, but the impact could be serious even so. The production of bicycles in Saint-Etienne dropped radically every autumn. A Belgian weaving factory reported a gap of five to six weeks between pieces on each loom, so that workers in theory responsible for four looms actually worked only three, with a corresponding loss of income, except during the autumn clothing boom. The percentage of workers on full time could fall to a low of 57 percent in British shoe-making during a slack season.¹⁵ Oscillations in heavy industry were far less significant, and again took the form of short time rather than unemployment; between 1900 and 1914 the average monthly variation of total shifts worked per coal miner in Britain was 10 percent (the peak occurring during the winter months), while that of iron and

Table I Annual Unemployment Rates

steel workers was only 2 percent.¹⁶ But these important exceptions only place the instability of employment for most workers into sharper relief.

In virtually all the unstable trades, furthermore, a minority of workers were favored with unusually regular employment, as employers sought to content a nucleus of particularly adept or docile workers. A few privileged workers regularly worked four looms in the Belgian textile factory described above. An elite of leather workers in Offenbach was exempted from seasonal short time.¹⁷ Even miners consistently complained of favoritism in the allocation of above-average numbers of shifts. And on the docks and in the building trades, inequality in the access to work was endemic. Variations among workers who had daily contact with each other could produce

self-satisfaction among the fortunate and apathetic despair among those condemned to instability. Increasingly, it caused resentment.

Unpredictability, within a year as well as from year to year, obviously colored workers' reactions to their jobs, particularly when it was concluded that part of the instability was due to human agency. And the massive oscillations in employment levels, along with great variations among industries and among groups of workers within a single trade, limit the desirability of positing an 'average' unemployment experience. Yet the effort to establish a composite picture has some merit in getting at the human experience which unemployment involved.

We can assume that at any point during a given year (lumping crisis and prosperity together) an average of at least 4 percent of all workers were unemployed in Britain and France. In fact this is surely a minimum figure, typical in fact of prosperous years, because of the underreporting of the unskilled, as the French statistics, which took more account of dockers and builders' laborers, already suggest. Actual rates in Germany, as we shall see from local reports, were probably comparable if not a bit higher. What did this mean, in terms of a worker's experience? Simple mathematics reveals that an annual unemployment rate of 4 percent, if spread equally over the entire labor force during 313 working days each year, would produce the loss of twelve and a half days for each worker. A series of local studies suggests that this calculation accurately conveys the days lost to unemployment among the kinds of workers included in trade union figures, but that a minority of workers bore the brunt in any single year. Available British statistics confirm the pattern closely. London compositors, for example, reporting an average of 5 percent unemployment, saw this spread among 20 percent of all workers during the year — suggesting that each unemployed worker was out nearly three months. More typically Scottish blacksmiths, with an average rate of 13 percent unemployment during 1894, spread it among 45 percent of all workers at some point, for a typical unemployment period of about a month.¹⁸ In 1892 60 percent of all members of the Amalgamated Society of Engineers were unemployed less than three days; 17 percent three days to four weeks; 7 percent four to eight weeks; 4.5 percent eight to twelve weeks; and 11 percent twelve weeks or more. In a better year, in 1895, a larger percentage were unemployed a few days or not at all (76 percent). The amount of moderate-term unemployment, up to four weeks, dropped 7.2 percent, but the other figures varied more modestly: 5 percent were unemployed four to eight weeks, 3 percent

eight to twelve weeks, and over 8 percent more than twelve weeks.¹⁹ The engineers' and compositors' figures reveal that in industries relatively stable in terms of seasonal factors, unemployment rates of 2 up to 5 percent annually involved over a fifth of the entire labor force at some point, for an average term of several weeks.

A study of the 1278 unemployed workers in York in June 1910 suggests a similar pattern in which most British unemployment was relatively long-term; the study occurred when union unemployment in the city was listed at 3.7 percent, but did not depend on union documentation. Of the unemployed only 1.1 percent had been jobless less than a week. Figures for other intervals reflected a large hard-core group; thus whereas 7.8 percent had been jobless between two and four weeks and 14 percent between eight and twelve, 11.5 percent had been out of work between seven and nine months and 5.6 percent over two years.²⁰

German materials display a different and in some ways more distressing pattern, for in addition to a hard-core group — visible in part because the German union statistics did not distinguish clearly between those lacking work and those incapacitated by illness, accident or old age²¹ — there was also a sizeable minority affected by several weeks of joblessness each year. The result was more general if possibly less intense suffering than in England. Among unskilled Berlin machine workers in 1906, for example, 41 percent suffered prolonged. unemployment during a good year because of illness, lack of work, or old age. The average skilled worker was less vulnerable, though more so than official union figures reflected, but still over a third suffered unemployment of two weeks to a month each year.²² In the crafts the situation was at least as bad. In 1907 17 percent of all unionized printers were unemployed at some point each quarter, over 40 percent at some point each year. Partly because of the poor economic situation 25 percent of all woodworkers lost time to illness alone in 1902; but even in 1897, a good year, 23 percent had done so. In 1902 unemployment and illness combined caused 52 percent of all skilled woodworkers to lose time from their jobs. German building trades workers suffered at least as severely. In 1900 32 percent of all roofers were unemployed at some point, while 27 percent of all masons were out of work in December alone.²³ And the length of unemployment among German union members was extraordinary. Of the masons identified in December, 1900, as unemployed, 22 percent had lacked work for at least thirteen weeks. The woodworkers in 1902 averaged, if unemployed, 34 days each; if sick or injured, 29 days. Unemployed

Berlin printers in 1910 were out 18.4 days, on the average, and in 1912, 23 days: some were unemployed three or four different times during the year to reach this figure.²⁴ In 1907 the average unemployed miner was out 27 days; the printer, 23 days; the woodworker, 15 days; the metal worker, 14 days; the baker, 11 days; the leather worker, 17 days; and the ship's carpenter 61 days. Interestingly in this case heightened unemployment in 1908 did not increase the average duration of unemployment overall; considerable duration seemed endemic to German unemployment. This same pattern prevailed in France. In 1896 only 12 percent of all unemployed males had been out only a week or less; 14 percent had been jobless for 26 weeks or more. Only a few trades such as printing had commonly brief stints, and by 1906, when only five percent of all unemployed suffered for a week, and a full 35 percent for over 26 weeks, even these privileged pockets had disappeared.²⁵

None of this adds up to a complete statistical picture of unemployment, and to some the figures may already have become excessive. Yet a few general conclusions are obvious. Most important, a listed rate of four or five per cent unemployment was no idle figure to large segments

of the labor force; it cannot be dismissed as the sad lot of a small minority. In Britain it was compatible, in prosperous years, with long-term unemployment of up to a fifth of a skilled labor force,²⁶ while in crisis years another group of up to 20 percent might be hit with up to four weeks of unemployment. In Germany prosperous years saw a higher percentage of workers affected by unemployment than in Britain, with up to a third of all skilled workers involved, for often substantial periods of time - a pattern not too dissimilar from British crisis years save that the hard-core group must have been smaller, for average duration was less. And during a year of recession, over half the labor force might be touched with unemployment at some point. In a situation such as this few workers would be immune from a fear of losing work and most would endure the experience several times during their working life.²⁷

Who were the unemployed? Several approaches might be revealing here. Women were less likely to be unemployed than men in a given industry - 50 percent less likely in Saxony in 1910—11.²⁸ This reflected some employer preference to keep cheaper workers on, perhaps, but more the withdrawal of unemployed women from the labor force. More significant is the apparent urban-rural pattern. The very biggest cities drew in so many unskilled workers, plus excessive skilled workers in trades like printing, that their rates would be higher

than those of other cities. In 1901, for example, the Parisian unemployment rate was more than twice as high as that in the industrial Nord.²⁹ Berlin and to a lesser extent London were also pockets of massive unemployment; so was Brussels, where unemployment could hit 19 percent while in other Belgian cities it ranged between 4 percent and 10 percent.³⁰ But city unemployment generally was lower than that in small towns, except the outright factory center. In 1910 a small-town construction worker in Germany was 60 percent more likely to be unemployed than his big-city counterpart. Similar patterns prevailed in printing and other crafts.³¹ These urban patterns relate in turn to the distribution of unemployment among key industries, which is the most important single way to break the phenomenon down.

Relatively few groups of workers enjoyed unemployment rates markedly below the average, but they included a large minority of the labor force. We have seen that railroad workers (not including maintenance men) chose their profession in part because of the regularity of employment. Miners and metallurgical workers were the oilier large categories among whom unemployment was normally not a problem.³² Scattered returns from Germany suggest a normal unemployment rate among miners of about .2 percent; only in the crisis year of 1901, when the rate was 4.1 percent, were over one percent of all miners unemployed. French miners suffered briefly in 1905-6, when rates in some months soared past 5 percent, but normally they too were under one percent, even in the recession of 1908—9. The French censuses of 1901 and 1905 recorded .05 and .5 percent unemployment in the mines, respectively. Unemployment among British miners may have run slightly higher, and between 1907 and 1909 it topped 5 percent; but again the normal rate was so low that it undoubtedly included only voluntary job-changes plus some incapacitated workers. Thus short time was a greater problem in the mines than unemployment, but even here workers were rarely deprived of at least live days of work per week. Metallurgical workers were even better off. The French census caught them with .03 and .17 percent unemployment in 1901 and 1906; German rates were a bit higher, averaging something over 1 percent between 1901 and 1905 and thea pulling up to 2.1 percent and 3.9 percent in 1912 and 1913. No rates are available for Britain, but short time was less significant than in the mines. Given the normal unemployment experience, the distinctive situation in heavy Industry was a tremendous boon for the workers involved. Metallurgical workers, who often lived near other categories that suffered much more

seriously, could particularly appreciate their advantages; one may hear echoes of their foresight in their efforts to keep their sons out of the more remunerative but chancier jobs in metal work. Miners, with no comparative base immediately visible, were less likely to reflect a sense of satisfaction with their job security.

Most factory workers endured unemployment rates close to the national averages, though there are some areas of uncertainty. Reported rates in the textile factories were usually rather low, though we have noted considerable seasonal underemployment. German textile unions reported an average of 1.4 percent unemployment; the French censuses recorded .04 and 1.7 percent unemployment in the textile factories. British rates were usually between 2 and 3 percent. Unemployment in Ghent textiles was a bit over 3.5 percent in 1910. But French unions recorded an average of 6.9 percent unemployment between 1901 and 1914, and there were complaints about massive joblessness in Roubaix and Rheims at various points.³³ So textiles present some uncertainties, beneath the official rates. Shoe manufacture was distinctly chancier. Germany reported rates of over 6 percent on several occasions, while in France the unions claimed an average of almost 8 percent; there were unusually wide fluctuations in the industry. Metals and engineering provide more ample information. Generally unemployment in these industries was close to the national average or a bit higher, placing them at the risky edge of the factory group. Thus the French censuses, listing 2.6 and 3.7 percent rates in 1901 and 1906, reported no higher rates in a predominantly factory industry. German recorded rates ranged between 1 and 3 percent, but we have seen that these should probably be doubled. French unions averaged 7.0 percent between 1901 and 1914, British unions 5.9 percent between 1891 and 1914 (a rate pulled up by shipbuilding, whose average was 8.9 percent); engineering specifically was 4.3 percent. Thus 5 percent unemployment was a very common pattern in the industry. These rates, plus the skill involved in much machine work, understandably produced important efforts at job defense. Hence a high rate of demarcation disputes, particularly in Britain; ship's carpenters, for example, seemed almost permanently at war with the riveters. Other workers tried to refuse to train new workers, to protect their jobs. And throughout the industry there was a widespread sense of threat, for machines and unskilled labor seemed to combine to menace the whole future of the industry.³⁴

But in general factory workers were less afflicted by unemployment and the fears it produced than other workers who laid claim to special skills. The urban crafts were significantly less stable, a distinction that

was not new but could be seriously troubling. Printers approached a 5 percent unemployment level: British typographers averaged 4.6 percent between 1891 and 1913, London compositors 5.3 percent. The French census found printers hovering under a 4 percent rate (3.8 and 3.6 percent in 1901 and 1906 respectively) — reflecting an interestingly low level in France which union figures confirm — while unionized printers in Germany averaged 4.6 percent. Printers, then, suffered about the same unemployment rates as metal workers. Both groups were highly skilled and divided between large shops and small. Woodworkers, also mixed in factories and shops, had comparable unemployment rates: 4.0 percent in Britain, but higher in France (at 10 percent according to union figures) and probably in Germany (at 2.5 percent even "by official statistics) where factories were less widely developed. The more artisanal the situation in this industry, the higher the unemployment rates: hence furniture makers in most countries (except Britain, where factories were well established) had extremely high unemployment rates. In France furniture workers often refused to train apprentices because of the threat to jobs, while Belgian furniture workers often suffered over 10 percent unemployment even in prosperous years.³⁵ Artisans in all these trades, caught between the traditional attractions of their craft and new machine competition, could easily echo the fears of their colleagues in the engineering plants: a French printer lamented, 'Unemployment, one of the greatest wounds of present-day society, has installed itself as master in all the shops, where previously there was security for the morrow.'³⁶

But the crafts more remote from the factory orbit were worse off still. Artisans in jewelry and precious metals suffered from unemployment averages bordering on 10 percent. Bakers fared just as badly. German bakers averaged almost 12 percent unemployment, never falling below 6 percent. In contrast to factory textiles, workers manufacturing hats and gloves suffered massive unemployment, again ranging over 10 percent in theoretically prosperous years. As could be predicted from their severe seasonal problems, skilled construction workers were often without work. British carpenters and joiners averaged 4.6 percent unemployment, all English builders 4.5 percent (5.3 percent in London); their Continental counterparts were far worse off. As we have seen, German unemployment rates surpassed 12 percent when averaged through the year. Here too complaints multiplied. British workers claimed, in the winter of 1905—6, that they could 'never recollect being so badly off since we have been in the trade.³⁷ French workers sought an explanation in new machinery: 'A

building which, fifty years ago, took two years to build, is now done in six months.'³⁸ And at one point Parisian builders resisted the imposition of a weekly day off, saying that they were already forced to miss more work than they wanted.³⁹

The professions that contributed to the reports on unemployment thus displayed a range of almost 1000 percent, with articulated fears varying accordingly. Three groups of workers, reflected rarely in the official evidence, operated in quite a different climate. Workers in domestic industry, workers in the sweated trades, and male laborers were often employed at best half the year. Many of them could thus not think of work as a normal experience, and their whole culture was profoundly conditioned by uncertainty.

We have touched on domestic industry already. High unemployment rates in shoe manufacture or woodworking reflected a division between factory or urban shop workers and manufacturers in a putting-out system. The latter rarely appeared in trade union reports, and their unemployment was much more common than that of the unionized workers, but they put pressure on the whole employment situation. In French shoe manufacture only 36 percent of all domestic workers reported little unemployment; the remainder were out two to six months a year - a full third missing half a year's work.⁴ ° Many saw-makers working at home in Sheffield were almost permanently unemployed.⁴¹

Workers in the sweated trades showed up in some general rates, such as those for food processing and clothing. German clothing workers in 1912—13 suffered almost 29 percent average unemployment, those in Brussels often recorded at over 26 percent unemployed. Leather workers in France - here divided between largely unmechanized factories and outwork — were unemployed at a 20 percent rate. Thirteen per cent of French food workers were unemployed during an average year, again according to trade union figures. The French censuses, taken in the spring during a peak period of employment, nonetheless found clothing and leather workers second only to ditch-diggers in their unemployment rate (4-4.5 percent), with over 3 percent of all food processers out of work.

Finally there were the unskilled males, dependent on their ability to lift and haul. Here we return mainly to local evidence, but it is compelling. France provides the only general rates that can be taken seriously: in the summer of 1901, when their jobs were usually at a peak, 6 percent of all ditch-diggers were unemployed; five years later the rate was 7.1 percent. Unions of building trades workers (here

mixing skilled and unskilled) reported over 15 percent unemployment between 1901 and 1914, while unionized dockers averaged 26.9 percent annually from 1904 onward. The notion of average unemployment ranging above 20 percent per year is confirmed in other cases. A British Select Committee in 1896 described the hordes of workers seeking low-paying navvies' jobs in Wales. Over a third of the York unemployed were casual workers, serving as builders' laborers, porters, cab drivers, even agricultural laborers; while in London a mass of irregularly employed men shifted among small jobs in construction, printing, and engineering, as well as going to the docks.⁴² Thus big cities produced a host of sickly, slow, or misguided young people condemned to a life of casual labor; and their numbers were swelled by older workers displaced from other industries and by newcomers from the countryside, to whom the wages of a laborer's job could initially seem quite attractive.⁴³ The result was chronically high unemployment.

Dock workers provided the best documentation of the appalling pressure because, far more than other laborers, they could hark back to more stable conditions in the past.⁴⁴ A few ports provided havens of regularity even around the turn of the century. Belgian dock workers were organized in nations, each consisting of up to sixty chiefs or bazen, who served essentially as subcontractors, and about three hundred workers. The nations imposed stiff rules on absenteeism, drinking, and the like and in return could assure quite regular jobs in small ports like Bruges.⁴⁵ Vestiges of similar arrangements existed in small ports in Germany, such as Flensburg; and new trade unions here could replace the older corporations in assuring stable jobs. In no case, given the haphazard arrival of ships, did such arrangements prevent significant unemployment, but they could keep it within certain limits and provide a fair distribution of the work that was available. But in ail the larger ports such arrangements were a thing of the past. Nations existed at Antwerp, for example, but they were powerless to prevent massive unemployment; the pattern of thirty-six hours of steady work followed by days without a job, so common to dock workers everywhere, held full sway.⁴⁶ Rates, again, are hard to determine, but 50 percent unemployment was common in Liverpool and Bristol during the 1890s. In London in 1901 there were 7000 more dockexs than were used at maximum, and 9000 more than were used on the average — which assured that on a typical day 36 percent of all dockers would lack work.⁴⁷ Rising unemployment was also reported in Marseilles and to a lesser extent Le Havre.⁴⁸ In Hamburg in 1895, only

9 percent of all dockers worked over 210 days a year; 7 percent worked 106 to 210 days, while 83 percent had between a single day and 106 days. Obviously here and elsewhere figures could be swollen by occasional job-seekers from other fields. In Hamburg 59 percent of all dockers were in this category. Even so, only 22 percent of the permanent dockers approached regular employment (over 210 days), while 46 percent worked less than 106 days.⁴⁹ And for their part many dockers did not move easily to other work, like the unskilled construction workers. Some observers claimed that this reflected a lack of knowledge and experience; it may also have mirrored an attachment to a profession that, in its odd way, had something of a craft background without the attendant skills.⁵⁰ In any event, the lot of the docker would be truly pitiable. Statistically he could count himself lucky to work half the time, which meant that survival was always precarious. A London docker described a normal situation: 'I had three days' work last week, but three days doesn't go far with many mouths to feed at home.'⁵¹

Work among dockers, and to a slightly lesser degree among ordinary merchant seamen, was a constant battle. 'I daresay none of you know what it is to be out of work,' one man noted to a British Royal Commission. 'I wish that some of you had that experience. There are always crowds at the dock gates and wharf gates, even at the best of times, who are fighting and struggling for employments.'⁵² Periods of intense work alternated with anxious, usually fruitless waiting, in makeshift sheds and according to rather random periods of call.⁵³ Unemployment rates among the unskilled encouraged favoritism and toadying beyond anything known in the factories or mines. Most employers tried to keep a cadre of steady workers, while periodically distributing work among enough others to maintain a labor force adequate to any sudden boom. A minority of dockers thus had relatively steady jobs. In Hamburg workers were divided into three classes, as the annual days of work rate reflected, with the 'steady workers' given first crack at every job.⁵⁴ In all the big ports employers responded to labor unrest after the 1890s by extending the discriminatory system. 'Mixed' unions were everywhere successful in defusing agitation by assuring preferential hiring to a quarter to a third of the dockers.⁵⁵ In Marseilles, for example, a quarter of the labor force was in a mixed union which promised that if one employer fired them they would be sent to another company immediately; a mixed union in Antwerp was similarly organized, with two workers on an executive board of twelve men. This system quashed agitation in

Marseilles for more than five years after 1904, in Antwerp for virtually a decade. For the majority of casual workers, whatever the theoretical interest in spreading work, jobs came through foremen or hiring bosses. In the German ports innkeepers served as hiring agents for dockers and especially seamen, taking healthy bribes in the process; British dock foremen were often bribed.⁵⁶ Unskilled work, then, was degrading as well as unpredictable for masses of workers.

The typology of unemployment, even if it cannot be completely filled in, goes some way toward explaining both probable comparative differences among mature industrial countries, and trends in the two decades before World War I. Comparison is of course chancy, given the quality of the statistics, and the main point in all countries is the relatively high incidence of unemployment plus the similarity of patterns among key groups of workers. All the national figures, based on crafts and a few skilled factory sectors, do not adequately take miners into account, though there was some reporting, and they certainly do not embrace the bulk of the unskilled. Most, then, are probably understated for the working class as a whole. But it does seem likely that a country where industrialization was less advanced but where agriculture could no longer sustain a mass of underemployed labor suffered an unusually acute unemployment problem. Add a larger mass of vulnerable domestic manufacturing workers and the probability is enhanced still further. This might, along with admitted differences in the reporting base, explain the significant disparity between France and Britain. For Germany and Belgium, comparison is lost in conjecture because of the underreporting of factory and craft unemployment and the absence of any general information about the unskilled. Possibly the latter were still content to cling to the countryside, so that urban unemployment was really lower than in France. But Germany as well as France could produce the kind of angry rioting by otherwise inarticulate, young unskilled workers that reflects the presence of an excessive casual labor force in the cities. Several strikes called by factory or craft workers in Dusseldorf and Hannover, as in Saint-Etienne and Armentières, turned into violent action by this quite different group of workers.⁵⁷ This was not the case in Britain, where periodic but calm marches by the unemployed reflected greater consciousness of the problem but also, possibly, a somewhat smaller degree of pressure. In any event, it is quite likely that overall unemployment was comparatively severe in Germany as well as in France. In Britain half a century earlier massive movement of the rural poor into the cities had created an unassimilable casual labor force,

Table II Trends in Unemployment

A. National Rates (annual averages per period)

B. Industrial Rates

while unemployment on the Continent remained buried in the countryside; so now changes in agriculture plus, possibly, new yearnings by the rural unskilled imposed a heavy burden on the Continental cities.

And this in turn sets the framework for overall trends during the pre-war decades. A period of fifteen to twenty-five years provides a limited basis for generalization, to be sure, but it is long enough to stake out some conclusions. Despite tempting characterizations of the displacement caused by new levels of mechanization, and the intense and understandable belief in such characterizations on the part of many workers, overall unemployment rates went down in most cases, even as the reporting base improved, making it more likely that unemployment would be detected. Measurement by five-year periods in Britain reveals no upward trend; the three periods in which recessions occurred can be easily compared, while rates in the immediate pre-war years were slightly lower than those in the crisis-free late 1890s. French rates went steadily down, with the final three years reaching unprecedented lows; Belgian rates followed the same patterns. Only German rates rose slightly, but almost certainly because reporting improved right before the war. Where measurement is possible unemployment declined in almost every key industry in France and in many in Britain.(See Table II.)

To be sure, mechanization was felt in some general rates. In machine building and metals there was a secular trend towards increase in most countries. In other industries mechanization may have heightened rates more briefly, only to yield to an adjustment in the labor force or collective efforts to defend jobs. Thus problems in British typography were particularly acute for ten years after 1896, when the introduction of new machines was at its peak. Mechanization could also have shaped rising rates in the construction trades, which began to yield, and incompletely when they are controlled for the absence of recession, after 1911 in Britain. On the other hand, although the process defies exact tracing at this point, increasing mechanization, in opening up jobs to the unskilled, probably cut into rates in this vulnerable group; and overall, as the French and British figures both suggest, growing mechanization was compatible with stable or declining unemployment rates in such trades as printing, food processing, shoe manufacture, and even textiles. Furthermore, despite their persistent suffering, many dockers and other unskilled workers seem to have benefited by some regularization of their work as the period wore on. British docks by the 1890s, in entrusting hiring to employers' agents, were for all their

problems ahead of Continental ports, where hiring agents were under fewer controls. Everywhere the influx of rural workers began to dwindle, while labor agitation, even when abortive, prompted attention to better organization out of capitalist self-interest.⁵⁸ The development of new kinds of work for untrained personnel could only have aided this process. Certainly the trends in French construction and dock work reveal a significant evolution. There is one obvious caveat here: unskilled workers were still pouring in from abroad, as the many complaints of seamen and builders' laborers about foreign competitors attest. The rates of joblessness among the immigrants were rarely measured, because the victims departed when they lacked work. In German Lorraine, for example, up to 25 percent of Italian workers left during the crisis of 1901.⁵⁹

But at least for native workers advancing industrialization reduced irregular employment in normally prosperous years, showing some noticeable effects even within a two-decade span. Recessions were still severe, of course, and aside from this the improvement in average levels was cold comfort to workers in industries where mechanization did cut into jobs or to the masses of workers even in statistically improving situations who faced long stretches of unemployment each year. Even in a prospterous year right before the war, outright unemployment or the fear of it was a major factor in the job experience of all but a few segments of the working class. But the tendency toward some improvement was an important part of workers' experience. It related closely, if ironically, to the workers' own consciousness of their situation.

For beyond the obvious if important statement that workers hated and feared unemployment, the key point in assessing its impact on their consciousness is in terms of the sense of ineluctability. Workers who found the unemployment experience new would most obviously try to react against it. Dockers, as distinguished from the usually more quiescent builders' laborers, may have been roused to their sporadic, massive protests around the turn of the century by their realization that stability had been lost to a horde of new workers. Railroad employees could certainly be spurred by the fear of unemployment, for this challenged one of their main goals in work. In 1909-10 the Prussian railways fired seventy-three workers — a small number, but in context enough to drive masses of their colleagues into the union.⁶⁰ On the other hand, when unemployment was regarded as normal it could sap any desire to seek improvements. Many older workers in Germany and elsewhere were profoundly affected by the searing experience of the

1870s, when they had lacked work themselves or suffered hunger as children because of their parents' unemployment; new recessions in the 1890s and after merely enhanced the desire to cling to whatever work was available. Similarly, high unemployment rates could produce great satisfaction among the minority of workers who were proud to be able to support their families; this was common on the docks, where it reduced ability to take collective action against arbitrary hiring practices.⁶¹ And quite apart from the impact of unemployment as seen in past crises or among one's fellows, many workers brought to their jobs a recollection, if only passed on from father to son, of the long period of underemployment that was part of agricultural life even before industrialization began. This, surely, must explain the mentality found in the Daimler automobile factory, where workers were so ready to claim that they chose their own jobs. There was virtually no unemployment here, yet most workers said they expected to lose their jobs sometime in the future. Yet beyond a sense of concern, even of dread, few could suggest what they would do if this happened. They said they had 'no information' or 'God only knows'; some talked of 'hoping to die first,' even of suicide, while others, less concerned, counted on fate or the divinity. But none had any plans or sense of alternatives, save in a few remarks that the state should increase benefits under existing old age and sickness insurance laws. Most, if pushed beyond general laments, talked only of begging or doing anything that came along.⁶²

This kind of attitude had never characterized the urban crafts, with a tradition of organization against unemployment, and it would decline, with experience, in the factories. This is not to claim that workers had substantial control over the employment situation. Metal workers in Berlin or Britain were extremely conscious of the unemployment problem and consistently tried to remedy it. But direct efforts to limit apprentices had to give way, as we have seen, to attempts to defend job boundaries through demarcation disputes or to pious agreements, such as one reached by the Amalgamated Society of Engineers in 1907, that the employers' federation would urge its members to retain any worker displaced by machines.⁶³ Unemployment mounted anyway, and the growing resentment of its toll simply produced a rising tide of angry protest. But where competitive pressure was less intense, workers' efforts could contribute to the stabilization or reduction of the unemployment rate. Apprentice agreements in printing obviously had an effect; German brewers were also successful in limiting the access of unskilled workers.⁶⁴ On occasion, attacks on competing workers might

help. Unionized workers in many industries divided between factories and domestic production sought to eliminate the latter because of the threat to jobs and pay, most typically through bargaining agreements or organization of the domestic workers, but sometimes through physical assault. Hence after growing periods of unemployment bands of two to three hundred Fougères shoe workers in 1905 attacked the houses of domestic producers and threw their work away.⁶⁵ Countless workers sought to limit production in order to preserve jobs. Parisian masons worked to rule as part of their period of intense unrest, following massive unemployment through 1905. South Wales tinplate workers traditionally responded to overproduction by limiting their effort per shift; in 1913 they even discussed a 'stop week' with employers. Limitation by agreement was actually achieved briefly in some crafts; it was tried in Birmingham brass manufacture in the 1890s, to the detriment of the industry, while precious metal workers set production quotas with their employers in 1903 and 1905.⁶⁶

In no sense were devices or concerns of this sort new, but they were spreading more widely. And at the same time some traditional remedies declined in importance. British engineers, for example, reduced their traveling to seek jobs. Between 1890 and 1910 the use of travel cards from the Amalgamated Society of Engineers declined despite rising unemployment.^{6 7} Here was another sign of the settling-down process which so powerfully affected initial job choice. Traveling was a much more common remedy for workers whose commitment to a profession was limited or who could not really think of unemployment as anything but an act of God; hence its popularity among new miners or metallurgical workers in France and Germany. Increasingly workers strove for collective remedies. In addition to the customary arsenal they sought union shops, which obviously could serve to control job access, and they worked with growing vigor to reduce the hours of work as a means of protecting jobs. Stone cutters in Rouen who struck in 1910 for a ten-hour day and weekly rest defended their goals in terms of their desire 'to avoid too frequent unemployment,' and their thought could be echoed by countless workers throughout the period.^{6 8}

And the growing concern about unemployment was responsible for the repeated claims that the problem was worsening even when this was not in fact true. If engineers had a good case to make in their laments that things had never been so bad, printers did not; there is every indication that printing unemployment had been higher in the 1880s and early 1890s than it was in the two decades before the war, yet complaints of unprecedented difficulty were regularly repeated in

many areas. There is no need to quibble about a lack of historical consciousness or statistical precision, for what was claimed was reality to vast numbers of workers. Growing unwillingness to accept unemployment combined with a sense of precise enemies — machines, in printing, or Jewish workers in British shoe-making or tailoring, or Chinese or Africans in the seaports — to create growing anger.⁷⁰ The anger could lead to significant political or direct action protest; it could also lead to a sense that work had lost its value.

Yet though there was a trend in the consciousness about unemployment, workers were profoundly divided in this matter as in the experience of unemployment itself. Quite apart from the fact that attacks on unemployment often meant attacks on other workers, the gulf between those workers who saw no alternative but to accept unemployment as inevitable, and therefore jobs as a gift, and those who fought for their jobs remained great. Patterns of unemployment in Britain suggested in some ways that the gulf was widening, even aside from the rural workers who could still be found to take casual employment with a sense that any job was a boon. For many workers a few years' experience on the docks produced only a sorrowful apathy; an inquiry among unemployed workers in 1909 revealed a single dominant idea — to get a few days' work. Some workers even abandoned the notion of work itself: almost 10 percent of the York unemployed in 1910 no longer actively tried to get jobs.⁷¹ The very success of many skilled workers in defending their work, save during recession years, expanded the group of long-term unemployed, as we have seen. Their morale sagged and their physical constitution deteriorated. Hopelessness among the unemployed in every country owed much to their inevitable deficiency in diet. Many of the York workers, for example, ate only bread and achieved less than half of their minimum caloric needs.⁷²

But if we must not forget the divisions within the working class concerning unemployment — including, at the other extreme, those industries where unemployment was not a problem — it is important to stress the division that unemployment or its threat created within the typical worker himself. Here was an experience beyond his immediate control. A few—14 percent, in one study — might be voluntarily unemployed, as they sought a better job or merely some leisure time; but the vast majority were under the severest compulsion.⁷³ The damage to their income was immense. A Berlin lock-maker, if unemployed at all during the year, lost on the average a third of the income a full-time worker could earn during a year.⁷⁴ Added to this

was a sense of failure that was hard to shake off. And the experience was profoundly lonely. Unions might offer brief assistance, but it was the worker himself who had to seek another job. Placement agencies existed, of course, and a bitter battles in France by workers such as bakers whose unemployment forced them to request new jobs frequently attested to how important an unbiased job service could be.⁷⁵ But, in part because it was hard to free placement efforts from employer control and the resultant fear of favoritism, few workers actually found their jobs through them; and with rare exceptions trade union efforts to establish competitive services foundered on employer unwillingness to co-operate. Hence, in Germany, less than five percent of all workers found their jobs through the organized placement agencies.⁷⁶ Even collective efforts to protest unemployment had to be indirect. Unemployed workers might vent their anger in violence or in encouraging their fellows to hold out in a strike; they might march. But while they were unemployed there was little they could do to get at the cause of their problem.

Hence the growing resentment of unemployment could be extremely frustrating. Of course it fed the rising tide of labor protest, for with the exception of mining the greatest agitation came from workers in industries with significant but not incapacitating unemployment problems, workers who could articulate the desirability of a different organization of labor. But these were the same workers who felt a sense of choice in their jobs and were increasingly reluctant to change professions. Consciousness of choice and consciousness of unemployment as something other than the hand of God were part of the same mentality, in which each element could feed the other. Here was a tension that embraced one's daily life at work, in which pride and fear regularly battled. In such ambiguity the actual experience of work took place.

Notes

1. Reichsarbeitsblatt, 1903. A claim that a higher percentage of skilled than unskilled workers was unemployed was based on a comparison of reports from the tramworkers' union (unemployment 1%) with printers (8%). Historians, including those most sympathetic to workers' hardships, have not been immune to similar errors. Jurgen Kuczynski, accepting union figures at even more than face value, says they describe the unemployment of the

'mass' of workers, even though they reach a peak of only 7.5% between 1900 and 1914. Die Geschichte der Lage der Arbeiter in England von 1640 bis in die Gegenwart (Berlin, 195 5), 134. In Britain, Norman B. Dearie, Problems of Unemployment in the London Building Trades (London, 1908) contrasts the low jobless rates among unionized (skilled) workers with the much higher but unreported rates of the casual laborers.

2.    Louis Variez, Recensement des chômeurs gantois (Ghent, n.d.), 37.

3.    H. Demain, Les Migrations ouvrières à travers la Belgique (Louvair, 1919), 161. See also B. Seebohm Rowntree, Land and Labour: Lessons from Belgium (London, 1911), who clearly found this one of the things Britain could profitably learn from the Belgians.

4.    F. Hecht, Die Stôrungen in deutschen Wirtschaftsleben (Leipzig, 1903), I.

5.    Dora Lande, Die Arbeits- und Lohnverhàltnisse in der Berliner Maschinenindustrie zu Beginn des 20. Jahrhunderts (Leipzig, 1910).

6.    Alfred Williams, Life in a Railway Factory (London, 1915), 115.

7.    E. H. Phelps Brown, The Growth of British Industrial Relations (London, 1959), 16ff.

8.    The general unemployment rates have been derived from:

Statistisches Jahrbuch fur das deutsche Reich, Bulletin de l'Office du travail (France); Labour Gazette; and the Revue du travail (Belgium).

9.   The Continental figures are simply not good enough even to report for the 1890s. (They are comparable in this regard to the ludicrous material available for the 1870s in Britain; in both cases unions were simply not large enough to report anything useful.) Official German statistics indicated only a 3% unemployment rate during the crisis of the mid-90s, which is absurd. Franz Hitze, Die Arbeiterfrage (Mônchen-Gladbach, 1905). In fact, unemployment in centers like Berlin ranged up to 19% in 1895, reflecting the impact on the unskilled as well as high rates among skilled workers. Alfred Zinn, Die Entwicklung der Industriestandortes Berlin (Berlin, 1955).

10.    André Sayous, La Crise allemande de 1900-1902 (Paris, 1903); Korrespondent fur Deutschlands Buchdrucker und Schriftgiesser, 1908.

11.    Zentralverband der Maurer Deutschlands, Jahrbuch, 19"09. In Paris only 25% of all building workers were reported as working all year round; Amédée Dunois, 'Le Lockout de la maçonnerie,' Pages libres (1908), 54.

12.    Ernest Martel, La Vérité sur le Lock-out des carrières d'Ecausonne (Brussels, 1910).

13.    Franz Habersbrunner, Die Lohn-, Arbeits-, und Organisations- Verhàltnisse im deutschen Baugewerbe (Leipzig, 1903).

14.    Board of Trade, Abstract of Labour Statistics (London, 1915, Cd. 7733), 9.

15.   Journal des Correspondances (Belgium), 1909, 51; Labour Gazette, 1904; M. Devin, 'L'Industrie du cycle à Saint-Etienne,' Revue de géographie alpine (1949), 12.

16.    Board of Trade, Abstract, 10-12.

17.    Ludwig Hagen, Die Lederwaren-Industrie in Offenbach am Main und Umgebung (Karlsruhe, 1905), 33 ff.

18.    Phelps Brown, Growth, 75 ff.

19.    William H. Beveridge, Unemployment, a Problem of Industry (London, 1931), 69.

20.    B. Seebohm Rowntree and Bruce Lasker, Unemployment, a Social Study (London, 1911), 62.

21.    A study in 1895 showed that 40% of all German unemployed were incapable of work, which might help explain the higher incidence of unemployment in Germany than in Britain, though logically it should produce far longer average duration as well. In the York study only 7% were physically handicapped, though another 42% of the unemployed might be described as disadvantaged in terms of age (25% were over 40) or bad character (mainly drunkenness). But the German crafts, where disparity with Britain is marked, were less affected by incapacity (printing 36%, wood 33%) than machines (41%), mining (68%), and textiles (55%). So the explanation cannot be pushed too far; it would not, for example, account¹ for the difference in percentage unemployed at some point annually that stands out so clearly as between German and British printing, Hitze, Arbeiterfrage, passim.

22.    Lande, Arbeitsverhaltnisse. 37% of all locksmiths, 29% of all turners, and 42% of all smiths were thus involved, the latter especially because of illness.

23.    Theodor Leipert, Zur Lage der Arbeiter in Stuttgart (Stuttgart, 1900); Fanny Imle, Kritisches und Positives zur Frage der Arbeitslosenfursorge (Jena, 1907).

24.    Imle, Kritisches; Der Arbeitsnachweiss in Deutschland, December 1913.

25.    Korrespondent fur Deutschlands Buchdrucker und Schriftgiesser, 1908, passim.

26.    20% of a labor force averaging 12 weeks of joblessness would produce an overall unemployment rate of 3.2%.

27.    Without meaning to flog a dead statistician, the German trade union studies, together with other investigations such as Lande, Arbeitsverhaltnisse, make something of a mockery of the officially listed rate during this period, even for skilled, unionized workers alone. The national rate simply cannot be derived from the union reports as published by the government. Thus in 1912, when the national rate was 2%, listed union rates in the government tables were: mines .1%, metals 5.7%, food 3.8%, printing 4.8%, transport 6.5%, textiles 2.7%, leather 7.2%, shoes 7.5%, wood 3.7%, and

bakers 16.5%. If these figures were weighted according to the size of the trade union group in each case, one might produce something like the official pattern, but it would be statistically valueless. There is no question that the official rate is up to 50% too low even in normal years. It would nonetheless remain below the British rate for skilled workers, because of the lower average duration of unemployment. Figures in printing were almost surely correct; in 1907 the 5.3% listed rate coincided with the percentage of union members listed as unemployed at some point, adjusted for average duration. But in 1906 evidence on skilled metal workers (33% unemployment at some point, of 2—4 weeks in Berlin) suggests unemployment of at least 2.2% when the state figure is .9%. Again, the suggestion would be to multiply most figures by two, for comparative purposes. The same probably holds for Belgium; while less information is available, urban rates ranging from 4% to 19% in years when the official rate was under 3.5% suggest some strange manipulations. Variez, Recensement; Statistisches Jahrbuch, 1899-1913.

28.    Richard Calwer, Handel und Wandel in Deutschland, 1910.

29.    Ministère du commerce, Résultats statistiques du recensement général de la population (1901) (Paris, 1904).

30.    Variez, Recensement, passim.

31.    Calwer, Handel, 1910.

32.    The chemicals industry was apparently another haven, though information is scanty. The French censuses listed rated well under 1%, though later trade union returns suggest an increase after 1906.

33.    L. Guérin, 'Les Industries textiles,' Musée social (1912), 179; Variez, Recensement, 13.

34.    Amalgamated Society of Engineers, Monthly Reports, 1906; James B. Jefferys, The Story of the Engineers (London, 1946), passim.

35.    Ministère du commerce, Office du Travail, Rapport sur l'Apprentissage dans les industries de l'ameublement (Paris, 1905), 531 ff.

36.    Fédération française des travailleurs du livre, Compte-rendu du neuvième congrès nationalisons, 1905), 208.

37.    Dearie, Problems, 36.

38.    Fédération nationale du bâtiment, Les 9 heures dans le bâtiment (Paris, 1910), 3.

39.    Archives de la Préfecture de police de la Seine, B/a 13 80. Report of 1905.

40.    Sidney Pollard, A History of Labour in Sheffield (Liverpool, 1959), 202.

41.    Dearie, Problems; Rowntree and Lasker, Unemployment, 62 ff.; Select Committee on Distress from Want of Employment, Report (London, 1896, HC 321), IV, 27.

42.    Royal Commission on Labour, Minutes of the Evidence, Group 'B' (Transport by Water, Transport by Land) (London, 1892, C 6708), I, 292.

43.    William Sewell, 'La classe ouvrière de Marseilles sous la Seconde République,' Mouvement social (1971), 27 ff., describes the exclusivism by which Marseilles dockers defended their jobs.

44.    J. Izart, La Belgique au travail (Paris, 1910), passim.

45.    G. Vaes, 'La Grève du port d'Anvers,' Revue sociale catholique (1907-8), 52.

46.    Select Committee, Report, 27; Beveridge, Unemployment, 93.

47.    André Sayous, Les Grèves de Marseilles en 1904 (Paris, 1904), 62.

48.    Paul de Rousiers, Hambourg et l'Allemagne contemporaine (Paris, 1902), 249 ff.

49.    William H. Beveridge, 'The Problem of the Unemployed,' Sociological Papers (1906), 323-45.

50.    Society of Amalgamated Toolmakers, Monthly Journal, 1913.

51.    Royal Commission, Minutes, I, 75.

52.    London dockers in the 1911 transport strike reflected this problem in asking for three set times of call, at recognized places. Ben Tillett, History of the London Transport Workers' Strike (London, 1911), 14.

53.    Carl Legien, Der Streik der Hafenarbeiter und Seeleute in Hamburg-Altona (Hamburg, 1897); see also Eleanor Rathbone, Report on the Results of the Special Inquiry into the Condition of LabouY at the Liverpool Docks (Liverpool, 1904), 20.

54.    Léon de Seilhac, Les Unions mixtes des patrons et ouvriers (Paris, 1908), 51 ff.

55.    Rathbone, Report, passim.

56.    Niedersàchsiches Staatsarchiv (Hannover). Polizeiakten des Oberprâsidenten der Provinz Hannover (Hann. 122a XI), 70—72, rioting by workers, 1887—1908; Hauptstaatsarchiv Dùsseldorf, Zweigarchiv Kalkum, Regierung Dùsseldorf (Politische Akten), 15919, strikes by non-mining workers, 1899—1921.

57.    German ports, for example, eliminated hiring for short stretches — as little as a half-hour — in favor of a half-day minimum, during the 1890s. Rousiers, Hambourg, 249 ff.

58.    Hecht, Stûrungen, II, passim.

59.    Ludwig Radloff, 'Die Lage der preussischer Eisenbahner,' Socialistische Monatshefte (1909), 1500 ff.

60.    Adolf Levenstein, Proletariers Jugendjahre (Berlin, 1909), passim.

61.    Rousiers, Hambourg, 249 ff.

62.    Fritz Schuman, Auslese und Anpassung der Arbeiterschaft in der Automobilindustrie (Leipzig, 1911), 118 ff.

63.    Jefferys, Engineers, 156.

64.    Martin Weigert, Arbeitsnachweis, Einigungsamt und Tarifgemeinschaft in Berliner Braugewerbe (Leipzig, 1907). A 1903 agreement set the percentage of unskilled workers who could be hired at 15—20%, and this was scaled down in a renegotiation

four years later.

65.    Léon de Seilhac, Le Lock-out de Fougères (Paris, 1907), 104—5.

66.    Albert Hùglin, Der Tarifvertrag zwischen Arbeitgeber und Arbeitnehmer (Stuttgart, 1906); A. Picart, 'Le Lock-out de la maçonnerie parisienne,' Musée social (1908), 70; Arthur Pugh, Men of Steel (London, 1951), 166 ff.

67.    Jefferys, Engineers, passim.

68.    Archives départmen tales, Seine-Maritime, Series M, Commerce et Industrie, report of 22 April 1910.

69.    Korrespondent fur Deutschlands Buchdrucker und Schriftsgiesser, 29 December 1908, for a typical complaint; but 1880s unemployment rates had averaged 9%; Emanuel Baensch, Die Neuerungen in der Tarifgemeinschaft der deutschen Buchdrucker (Karlsruhe, 1908), 17.

70.    National Union of Boot and Shoe Operatives, Monthly Reports, 1892; Scottish Tinplate and Sheet Metal Workers' Friendly and Protective Society, Annual Report and Financial Statement, 1896-7.

71.    Rowntree and Lasker, Unemployment, 62 ff.; Royal Commission on the Poor Laws, Minutes of Evidence (London, 1909, Cd. 4835), IV, 260.

72.    Rowntree and Lasker, Unemployment, 224.

73.    Der Arbeitsmarkt, February 1912; the study dealt with the unemployed in Munich.

74.    Lande, Arbeitsverhaltnisse.

75.    Peter N. Stearns, Revolutionary Syndicalism and French Labor (New Brunswick, 1971), 40.

76.    Heinrich Reichelt, Die Arbeitsverhàltnisse in einem Berliner Gross-be trieb der Maschinenindustrie (Berlin, 1906), 51 ff.

CONCLUSION

Changes in the occupational structure, job choice, and the perception as well as the fact of unemployment constituted vital elements of the framework within which work took place. All suggest important distinctions in the working class, often but not always related. Occupational differences were crucial. Major industries displayed quite different levels of economic vitality. Most of the expanding sectors included many workers who evidenced some sense of job choice; opportunity and choice were obviously related. But expanding industries also drew in many lesser-skilled workers, including growing numbers of women, whose occupational commitment was often vague. And some of the stagnant industries, where a craft background was involved, could positively attract workers still. Obviously, sectors that were declining or barely holding their own normally evidenced high levels of unemployment; this was particularly true of the rural and small-town crafts. But some of the expanding industries were unstable as well. Construction was traditionally notorious in this regard, and it was difficult to see any noticeable improvement. Trades such as engineering and to an extent printing encountered unemployment difficulties for various reasons, but in part because of their attractiveness to new workers. They were vigorous, expanding industries still, but job choice could push more workers toward them than could be readily accommodated.

Occupational differences thus involve, even before we approach the work experience itself, a complicated set of factors. Craft-vs.-factory-vs.-unskilled remains an important tripartite division, but in the cities the craft-skilled factory lines had muddied somewhat as factory workers strove to take on some of the attributes of the artisans, as in the matter of job choice. The unskilled were more clearly discrete, but we have suggested that there was some increase in their job-specificity and even training (in terms of both on-the-job guidance and general educational background). Some halting limitations in their unemployment rates and some opportunities to rise to a definite semi-skilled level brought them more clearly within the basic framework of the working class. Within the tripartite division, other differentiations become obvious: high-unemployment vs. low-unemployment industrial settings

(metallurgy, for example, vs. metals or engineering); rapidly rising industrial populations vs. essentially stagnant (engineering vs. textiles); stable or rising artisanal vs. declining (printing vs. woodwork or leather). Similar distinctions probably existed among the unskilled, as dockers seem to have had more sense of choice, and possibly more clearly declining unemployment, than builders' laborers. Certainly in the crafts and in factory industry, at least a two-fold distinction between sectors with relatively rapid expansion and relatively high levels of job choice is essential.

Partially cutting across this analytical structure are other distinctions. In the crafts and to an extent in the factories, rural-small town vs. big city background and concentration prove consistently important. Sex differences are obviously vital, for the experience of men and women in a given industry differed widely, and certainly the working population as a whole was divided by sex. Age distinctions must also be borne in mind, though they remain difficult to trace precisely. Aging may have been undergoing a twofold process, from the occupational standpoint. Traditional havens such as domestic manufacturing, where workers were somewhat free to set their own pace, were either declining or becoming less palatable. This was clearly of overwhelming importance in countries like Germany and France where industrialization was just becoming fully established, though the virtues of domestic manufacturing should not be exaggerated. Within a more modern or simply more urban setting, the position of older workers may have been improving, at least so long as they could remain on the job. Work hierarchies settled in and could involve some attention to seniority.

In any event, a composite sense of the framework for work should combine the occupational distinctions based on factors of background, age, sex, and residence — a complicated combination but one that can to an extent be charted and, even for those accustomed to finding or wanting their social history relatively simple, not necessarily more cumbersome than the pattern of political factions farther up in society during the same period.

But the temptation to generalization will remain, and a few impressions do stand out from a consideration of the environment for labor. A maturation of the working class is obvious. Fewer outright newcomers were entering the class, an attachment of family to work had developed even in factory settings, and a sense of choosing a job — which meant to some extent orienting oneself within an industrial environment — had spread reasonably widely. Yet maturation

contained some important ambiguities. We have seen that job choice, industrial growth, and employment stability could conflict. Workers could want jobs that were decreasing in economic viability. They could want jobs that were growing rapidly in number but still exposed to massive fluctuations in employment. Both of these clashes could severely reduce the sense of satisfaction enjoyed once on the job.

More generally, the working-class culture that had matured was in important senses conservative. This followed from the continuing ties that segments of the class retained with the countryside, and even more from the contributions of artisanal origins and values. The kind of links that developed between family and work enhanced a conservative approach as well, for a job increasingly was something to be passed along. Conservatism made sense as well when the expansion of opportunities was beginning to shrink. Artisans and textile workers could perceive the slowing of industrial growth directly. So could women in the traditional concentrations, and older workers generally; and these groups were vital in passing job values along to the newer generations. In the expanding industries a defensive mentality could also be produced from the advent both of new numbers and of new kinds of workers — new nationalities, new youth, and possibly more women. Of course some workers, particularly the younger ones, were open to new goals; we have seen signs of this in the periodic defiance of parental direction, when young people entered better-paying industries instead of, waiting for traditional training. But, increasingly, one wonders if many of the most venturesome young workers might more likely have left the class altogether, to enter more rapidly-growing middle-class ranks. To the extent to which this occurred, the conservatism of the working class would only be heightened.

Conservatism may seem too loaded or rigid a term. It is not meant to cover the whole of working-class values; its relationship with obviously radical politics will be explored subsequently. It is certainly not meant in a patronizing sense; most people are in some sense conservative about their jobs or prospective jobs. Nor, finally, should rigidity be over-stressed. The diversity already noted qualifies any sense of a general job conservatism. Workers would still respond differentially to change, and certain kinds of innovation could have obvious appeal. Nevertheless, the contrast with the middle class, when also taken as a whole, was becoming clearer. In the early stages of the industrial revolution the middle class, in its numerical majority, had massive ties to the older order; only the minority was eagerly innovative. Workers, except for the craft elite, were subjected willy-nilly to novelty. Now it

was the working class that was expanding only slowly, open to fewer newcomers than the middle class. Key sectors of the class, beyond the rural-artisanal remnants, which themselves still loomed large, were beginning to close their ranks, some sensing the curtailment of opportunities that would soon lead to absolute decline in several of the leading industries of the early industrial revolution. And the working class had painfully constructed a culture that, reflecting their inability to control their own lot, would protect them against undue change and perhaps provide greater respectability in the relevant society. Defense of job traditions was a major part of this culture, but it increasingly extended to sexual behavior and other matters as well.

The potential clash between working-class culture and its economic setting is already obvious in the contrast between the motivations for job choice, where these could be articulated, and the actual growth patterns of the occupational structure. Change was essential. Parental guidance had to be flexible or to be partially ignored. The sense of job values would be even more severely challenged by changes in the work-place itself, particularly of course in those industries that were expanding most rapidly. For technological and organizational developments were advancing at a rising pace. Workers who had chosen their jobs because of traditional prestige or parental guidance could easily be disappointed. Those who were increasingly conscious that job instability could and should be avoided might feel newly threatened. This does not mean that an anticipation of clash should dominate the consideration of the juxtaposition of actual work situations with working-class values. The values had already proved adaptable. And the class, precisely because it had a work culture, could influence if not deflect the new forces impinging upon it. Obviously, its culture could provide the basis for protest as well.

Part II FORCES FOR CHANGE

CHAPTER 4 TECHNOLOGY

To the extent that workers assigned clear values to work, their culture would be severely tested in the decades immediately before World War I. Three factors demand special attention: technological developments, the new scale of business organization, and the drive for a higher pace of work. All were interrelated, so that in final analysis the assess ment of work must be made against a varied background of change. But each factor can be considered in some isolation, and each illuminates certain aspects of the work experience. For we are concerned, not with the details of inventions or corporate structure, but with the impact on work and on workers' thinking about their jobs.

With an approach to work that was in essential respects conservative, the labor force might be expected to react harshly to what was unquestionably an important new round of technological change. In fact the reactions were subtle, often indirect. They varied with some of the simplest divisions within the working class. Artisans whose skills were threatened lamented loudly, but where their industries were in decline they were powerless to react directly. Perhaps they had been expecting the blow for too long, and were disillusioned by the failure of earlier efforts to confront the industrial wave directly. In some cases, at least, they found refuge in political movements that, in their eyes, seemed to promise a return to an older order. Direct conflict was rare. This was not the case with important groups of unskilled workers, whose growing numbers gave them a sense of strength and who were lacing technological change for the first time. Significant if scattered clashes resulted, for machines were part of the process forcing unskilled workers to rethink their jobs. In the factories, finally, reaction to machines varied with the vitality of the industry and the system under which technological changes were introduced. But on the whole factory workers were least upset by change in this area, if only because of their long experience with the process. Previous exposure to machines, rather than the scale of technology, seemed to determine response. Hence sweeping generalizations on the dehumanizing effect of advancing mechanization would be out of place. Some workers even found new interest in the jobs produced by technological change. Yet there was human cost involved, and one must wonder if acquiescence was not

mainly the product of a sense of helplessness, a belief that this aspect of the job was out of control.

The outlines of technological change can be quickly sketched, for they are familiar in many respects. We have already suggested that the process of change took on some distinctive contours towards the 1890s, for innovation was more spurt-like than continuous and, on the whole, one can view the turn-of-the-century period as the second peaking of a new technology, less radical certainly than the first, early industrial incursion but a significant break with the pace of change in the 1870s and 1880s. Of course this periodization is not equally valid for all industries. The process of speed-up in British textiles, for example, began in the 1870s with the introduction of multiple looms; we pick up only the tail end of the process in the 1890s, and this will prove significant in assessing workers' reactions. Relatedly, the pace of change was greater in the less completely industrialized areas than in established leaders such as Britain. French use of steam engines, for example, doubled in the 1890s and then doubled again between 1900 and 1913; in 1902, 58,700 establishments had steam power, averaging 34.00 horsepower per firm, while in 1912, 63,000 companies averaged 51.3 horsepower. Innovation in Britain or Belgium was much more modest. But there is nevertheless an air of novelty to the 1890s overall, even in Britain. Two main developments can be discerned: the spread of new equipment to branches of manufacturing and goods handling previously untouched by this aspect of the industrial revolution; and the introduction of a new generation of equipment, far more automatic than its early industrial ancestors, to factory production.

The spread of gasoline motors literally revolutionized any work involving the lifting and hauling of materials. Cranes were introduced to all major docks. New stoking machines affected gas workers. Loading devices also gained ground in metallurgy. Hundreds of thousands of workers, who had only their backs and arms to sell, were touched by equipment of this sort.¹

The construction industry was naturally affected by the advent of cranes. Stones and to a lesser extent bricks were no longer lifted by hand. This was an industry that also exemplified the spread of machines into traditional artisanal preserves. Masons and stone-cutters now had machines to cut stones; simultaneously, in that odd accommodation of taste to equipment that has long characterized industrial society, the use of highly ornamented facades receded, so that many types of stone cutting were abandoned outright. French masons complained that even funeral monuments were being turned out in preformed concrete. But

actually carpenters and joiners were more seriously affected by new machines and materials. Some mechanical saws were introduced in woodwork, though mainly in lumbering rather than construction per se. More serious was the growing use of prefabricated wood units and, especially, iron and concrete for buildings. Here was an opportunity for new specialists, and iron carpenters became an important segment of the industry; but many traditional craftsmen could only feel that their trade was declining. The use of asphalt instead of lead and tile affected roofers, and plumbers had new methods to learn as earthware replaced lead piping in sewers. Obviously changes in materials were unlikely to have as radical an effect on skill levels as new equipment, and the construction industry was disrupted rather than transformed. It is also true that changes occurred differentially by area. German and Belgian construction yielded to new methods far more slowly then French or British. And while French masonry and carpentry were touched, roofing, in contrast to Britain, was not.²

Artisanal manufacturing was another story, for here a radical transformation did take place. Throughout industrial Europe the 1890s was a decade of displacement for one rural craft after another. Button machinery spread in France after 1895, while machines to manufacture hats gained ground even earlier. Glass bottling was revolutionized, as new equipment allowed four workers to do the work of 75. In Germany, these devices spread after 1900, though there were only seventeen in the whole country by 1910. Here the slowness of change limited outright disruption, but a traditional trade was being slowly wiped out nevertheless. In leather new equipment was developed, again after 1900, to shave the hair and flesh from skins. The list of trades affected could be expanded almost indefinitely. It was true that the small size of initial units in these craft industries delayed the widespread adoption of new equipment in many cases.³ Power-driven saws and lathes could transform woodwork, for example, and new furniture factories sprang up to apply these devices on a large scale. But still, in Germany in 1904, 82 percent of all companies employing workers in this industry had no power equipment at all, and only 6 percent of these had even hand-foot devices. In Berlin lock manufacture the German lag in artisanal technology was notable even in the cities — the number of companies with motors more than doubled between 1895 and 1907, but from 8 percent to only 17 percent of the total. Even where new machines were employed, the advent of internal combustion and electrical motors could give a new lease of life to otherwise traditional forms of manufacture. The tenacity of home

production in the French clothing industry and the rapid rise of companies employing less than five workers after 1896 was due to the possibility of disseminating new sources of power to operate sewing machines. But where machines existed factories were rarely far behind; in lingerie, for example, home production was almost dead by 1900. And while the gradual character of technological change in several industries must be emphasized, their employees could hardly fail to realize that they were under the gun; in some cases a swift death might have been preferable to the agony of most of the traditional rural crafts.⁴

On the Continent and to an extent in England (where the process had begun earlier) the 1890s saw displacement of most hand manufacture of shoes. The number of hand workers in the German shoe industry, for example, over half the total in 1895, had declined by over 40 percent a decade later. Heeling and finishing machinery spread in the 1890s, while lasting equipment, to nail the parts of the shoe automatically, came in around 1900. And in this industry workers barely had time to adjust to the advent of factories and powered equipment when larger, more fully automated equipment was introduced in the leading plants, most of it imported or copied from the United States.⁵

Textiles, though of course a factory industry for many decades, also found itself opened to new equipment, again often of American origin. Here a distinction must be made between British cotton and the Continental textile industry, with British wool following the patterns of the latter. British cotton was substantially more mechanized then French and especially German. The German spinner operated fewer spindles, the weaver fewer looms; many French and German weavers in fact ran only a single loom, and few operated more than two; while in Britain the norm was four. German equipment ran at least ten percent more slowly than British as well and, in far worse repair, it operated only 80 percent of the working day in contrast to Britain's 92—95 percent.⁶ Hence the most advanced techniques possible in the industry were not the main issue except in Britain. New looms such as the Northropp, which greatly increased the automation of the weaving process by recharging the shuttle with weft if the thread broken ran out and allowed a given worker to run twenty or more looms simultaneously — for a given loom could run untended for two hours — spread in British cotton to a limited extent and, by 1910, they were beginning to receive some attention in France. Even in England the Northropp was more talked about and feared than actually

introduced. In Germany, France, and in the British wool industry the kind of technological change that affected most workers was the steady pressure to adopt two looms instead of one, or three instead of two. Simple multiplication suggests that, if this did not constitute a dramatic improvement in technique, the potential impact on work and on output could be considerable. Concomitantly the amount of new horsepower being applied to the industry rose steadily, in France more than doubling the resources of the industry between 1899 and 1908 (when the process began to slow).⁷

Heavy industry was subjected to less change. Electrical equipment spread in German metallurgy after 1900, and this helped allow a 50 percent expansion in the size of blast furnaces between 1895 and 1913. Everywhere new machinery plus the increased use of Bessemer or open hearth methods displaced traditional metallurgical skills such as puddling.⁸ Coal-cutting machines were introduced first in the 1890s, but they made surprisingly slow headway. By 1913 8 percent of British coal, 10 percent of Belgian, but only 2 percent of German was cut by machines. In Germany the spread of steam engines and even horses for hauling coal affected production more seriously.⁹ Finally, on the railroads the development of larger engines and freight cars seriously affected the nature of work, for the capacity of train engines doubled in the decade after 1895.¹⁰

Of all the traditional factory industries, the manufacture of machinery was most seriously transformed by the new generation of equipment. More specialized lathes accomplished a number of cuts at once, making this process more automatic. Milling machines displaced the chisel-and-file work of the fitters, while grinding machines relieved turners of hand labor with emery tools. Mechanical drills and borers created new skills, or semi-skills, while in shipbuilding the spread of iron ships was accompanied, thanks to new machines, by the creation of teams of riveters and platers. By no means were all traditional skills threatened; fitters, for example, lost some types of work but gained new jobs manufacturing special tools. But the way was clearly open to semi-skilled workers, as both 'handymen' and women gained increasing ground in the industry. Furthermore, for all workers, the use of new steel in machines and the growing application of electric power (by 1914 commanding 14 percent of British machine production) allowed a massive increase in the speed of the equipment generally, in old processes as well as new. For many workers, this industry, like printing where the automatic composing machine massively heightened per worker productivity, became almost unrecognizable within a twenty-

year span.¹¹

The initial introduction of powered equipment in many industries, the growing conversion to automatic processes in others where, after the initial introduction of power, the technical base had remained rather low added up to a substantial increase in the potential productivity of the manufacturing sector and a significant transformation of the conditions of work. And, as suggested for several major industries, the knowledge that processes still more advanced than those actually in use were available heightened the fears and uncertainties of many workers. Hence, if often with some exaggeration, workers themselves voiced the sense that they lived in a new technological age; the refrain, 'never before,' was re-echoed in many industries.

For unskilled workers the whole machine experience was new, yet unlike earlier factory workers few of them operated machinery directly. They lifted and hauled still, but for shorter distances and more rarely up and down stairs and ladders. Some few were trained as crane operators, but for most the nature of work itself seemed to change little. Small wonder, then, that many saw machines primarily as a threat, a real or potential creator of unemployment. A few might add that machines made work more intense, by making workers responsible for stacking, loading and handling more goods arriving at more regular intervals; this was a complaint of British warehousemen in 1896. The spread of steam ships angered many sailors, who felt that the ship had become less their own and who specifically noted that their rise to officer levels was now blocked by a new class of heaters, who came from various backgrounds instead of the traditional sources of the merchant marine.¹² But generally the fear for jobs was overwhelming, in a period when the ranks of the unskilled continued to increase rapidly. Rouen dockworkers, for example, noted that employers drastically reduced work crews whenever a crane was introduced. Small wonder that refusals to work with the new equipment or outright Luddism were reported widely among the unskilled; this was in fact the only group likely to attack machinery directly during a labor dispute. In 1912 dockers in Swansea refused to work on an electric-magnetic loader, until prevailed upon in lengthy negotiations by their own union organizer.¹³ Construction workers in Dortmund threw sand in a crane in a 1900 dispute; Le Havre dockers pushed cranes into the ocean on at least two occasions; unskilled loaders in Lorraine metallurgy destroyed machinery in the 1906 strike. Belgian grain loaders, threatened by the spread of grain elevators, burned one of the structures in 1907.¹⁴

Scattered incidents, to be sure; this was not a characteristic response and the fear of displacement could inhibit violent action as much as it encouraged it. But violence was undoubtedly the tip of an iceberg, for machines caused or enhanced a massive sense of job insecurity among the unskilled. Sardine sorters in Britanny, many of them women and few formally trained, fought sorting equipment for ten years, at one point proclaiming, 'There is nothing left for us except the General Strike if they try to impose the machine on us.' And here we get one of the rare insights into the sense of commitment that even workers we must rate as unskilled could develop for their jobs. Fear of unemployment loomed large in Britanny, and it was certainly realistic, but the machine was resented also because it reduced the human element of work: 'The machine will never manage to do as well as we do.'¹⁵

Objectively, many unskilled workers stood to benefit by new equipment. Physical strain was reduced for many, and on the docks, if not so regularly in construction work, the increase in work to be done — the massive regular growth in tonnage handled, for example - limited the threat of greater unemployment. In fact, we have suggested that unemployment probably declined among this group. Machinery did not cause the decline, but it could promote a regularity of work that was a logical concomitant. German dockers, for example, found the work adjustments induced by investment in capital equipment helped the development of a more stable work-force, limiting the oscillation between short bouts of intense work followed by unemployment. Some also benefited from the substitution of daily pay for the piece rate which had characterized hand work, for they had often been cheated by sub-contractors when paid by the task. Yet all of this is a view from the outside. The unskilled were not normally articulate as yet. They could fear displacement even when it did not occur; they could resent changes in work habits that might otherwise be judged an element of stability. The unskilled did not rise in massive protest against machinery, and since they did express other grievances more vigorously we might assume that most decided they could live with the new equipment, once the fears accompanying the actual introduction of a powered machine died down. All we can be sure of is that they were given little choice; like the more massive Luddite protests of an earlier day, efforts by the unskilled to impede mechanization came to naught. The Breton sardine workers who refused to operate new equipment were simply replaced, sometimes by children. It was no more difficult to find dockers or construction laborers to work alongside the cranes, whatever their views on the

subject.¹⁶

Artisans who saw not only their own jobs but their whole profession being lost were more consistent in their complaints, but of course no more effective. British cabinet-makers proclaimed the downfall of their trade: with machines, apprenticeship was abandoned and boys and Jews were brought in to do the work of Englishmen. Bricklayers correctly noted that the use of steel and concrete threatened their livelihood — buildings were going up in 'half the time' with a third of the traditional number of bricklayers — and so they constantly talked about the need to 'sustain customs' on the job. British construction workers generally blamed their unemployment on the competition of machine-made building materials.¹⁷ But here, far more clearly than with the unskilled, more was involved than the threat to jobs. A French carpenter put it this way: 'With the development of industry, the worker is becoming more and more the complement of a machine. No one wants a "good artisan" any more, no one needs one.'¹⁸ And many French builders used strikes to point out the shoddy materials and sloppy methods their employers had adopted. A glass worker who refused to work on new machinery whose expense forced the establishment of a night shift noted: 'I see no need, for £4 a month, for me or my children to give up our sleep at night or our liberty.'¹⁹ Lock-workers from the Black Country to Berlin fought factory-made locks. A German blasted machine work as the death of art and creativity, claiming that only unskilled workers were stupid enough to endure the machines at all — and even they would find no joy in them, for independent hand work alone offered satisfaction.²⁰

Shoe and leather workers had obvious verbal battles to fight. Introduction of leather-cutting equipment, and unskilled workers to man it, caused craftsmen in Germany to complain of the monotony and tension of machine labor; they particularly lamented the noise, for they could no longer talk to each other: 'When there's good talk along with it, then the work flows gaily along.'²¹ Birmingham shoe workers in 1893 tried, like so many artisans, to point out that machines did not do as good work as hand labor, and 'a class of the workers are dispensed with in consequence.'²² A Parisian worker was more direct, claiming that 'the machine had killed the trade.' 'If I want the revolution, it's not to do any harm to people but to be able to destroy all these machines.'²³

Only one type of artisan was prepared positively to praise the new technology, and this rarely. Home workers might rejoice in the ability sewing machines gave them to stay out of the factories for another

generation. A German rug weaver welcomed the advent of electricity: 'As opposed to work in a velvet factory in Crefeld it's a good exchange.'²⁴

Otherwise the fear of losing one's job or the actual experience of having to transfer to machine labor soured the working lives of countless artisans. Yet, in industries like shoe manufacture, the advance of machinery was inexorable. Wages for hand work fell and periods of unemployment became unconscionably long; many workers had to give up.²⁵ Neither respectful complaints nor revolutionary appeals dented the process. And where this occurred the workers' resentment could be tragically counter-productive. A German union leader noted that the accident rate in the wood industry was almost twice as high on machine as in hand work, a differential he claimed was due as much to the feeling of stress as to actual physical danger.²⁶ Quite possibly many unskilled workers were better adapted to machinery in the newly-converted industries than were trained artisans; certainly it is small wonder that many older craftsmen held out in non-mechanized work, even as their wages fell. How many workers were forced, in mid-career, to shift to a machine we do not know; nor can we do more than speculate about possible resentments which even a teenager fro m a craft background might bring to the machine. There is no question that, in all the industries faced with rapid conversion, the attitude of the established workers was suspicious if not positively hostile — and because of the new scope of technological change this involved a large minority of the whole labor force.

Yet in trades where skills were not actually being extinguished and where growth still occurred, so that a lesser-skilled element could be accommodated without outright displacement, despair was not predominant and some adaptation was possible. Most construction workers did not find their work profoundly altered, and they could fight for their jobs. In Britain the incessant demarcation disputes that developed in the industry represented an attempt to fight on traditional lines; union leaders found it difficult to deal with new processes, insisting that their men should do the work they had always done. The result was the stagnation of union membership and, possibly, an unnecessary shrinkage of the traditional skilled groups as a percentage of the whole. On the Continent, where construction unions were newer and their leaders more radical, the intrusion of new technology less severely limited the growth of skilled categories; French carpenters, for example, might complain about the development of iron work but they organized with their fellow specialists. And, by including unskilled

labor, Continental builders reduced the sense of threat from that direction. We have noted also that in Germany and Belgium, at least, they were not faced with such rapid technological change in the first place.²⁷

Crafts that were more fully mechanized could adapt as well. Grumbling undoubtedly persisted, but if the threat of unemployment was not too great signs of adjustment soon developed. Bakers in Marseilles struck in 1907 when the number of mechanical kneaders spread, not against the machines themselves (which after all lightened their work) but over the rates they were to be paid on them, British glassblowers sometimes talked as if they were ready to resist machines root and branch, but in fact they strove to apply craft procedures to the machines, by limiting access to mechanical training to their own sons; in one case they refused even to let their employer's son learn the trade. Dyers in Saint-Etienne, deciding against frontal resistance to new machines, insisted instead on specifying a large work crew for each machine. The result was a violent strike in which workers who had consented to serve on small crews were attacked; but ultimately an accommodation was reached which minimized the chance of-displacement. German brewers admitted that new machines improved their lot somewhat, for refrigeration permitted production the year round whereas traditionally no malt work could be done during the summer. So they worked for agreements to limit the number of unskilled workers who could be trained on the new machines while seeking reductions in the hours of work and annual vacations — the latter a clear effort to retain, in new guise, something of a traditional break.

In other words few artisans in fields like food processing or construction even thought about acting on a rooted hostility to machines. How much the experience of machines actually bothered them we cannot pretend to determine, but at the worst they were so intimidated by decades of advancing industrialization that they lacked the will to stand their ground. Quite possibly the five hundred Saint-Etienne dyers who threw stones at their colleagues' houses would really have preferred to break the machines directly. But in many job categories the actual displacement of workers - as opposed to intergenerational change — was sufficiently slight and the preservation of a recognizable skill level sufficiently great to make adjustment something more than a pis aller.

This phenomenon can be traced in unusual detail for printers, whose trade encouraged written records. The advent of the composing machine was a severe threat to traditional skills. The number of women

and other semi-skilled workers in the industry actually advanced more rapidly than in many other crafts. It is true that printing had already been partially transformed toward mechanization and large-shop, if not factory, organization. Furthermore the composing machine did not eliminate the need for skills, and one historian has suggeste d that this makes the printers' case unique.²⁹ But the great possibility of using unskilled workers and the obvious preservation of recognizable skills elsewhere indicate that this is not an unusual case. Printers were frightened by the new machines. Their unions discussed them at every congress. But everywhere they opted for adaptation, and in so doing they were able to influence, if not control, the course of mechanization. Their example cannot be generalized too far, but it serves some purpose in illustrating the position of the modernized artisanry generally.

Despite their relatively privileged position as a high-paying, expanding trade, printers were seriously tempted to resist machines directly. Unquestionably their antagonism was greater before the fact; anticipation was worse than reality. Some claimed massive unemployment was inevitable. The British typographers' union noted in 1901 that its members had initially feared the disappearance of compositors, but came to realize that conditions actually improved.³⁰ From France: 'Mechanization pitilessly increases the amount of unemployment, and engenders misery.'³¹ There were claims that mechanization increased nervous exhaustion at work: 'fast-running machinery .. . caused greater mental and physical strain;'³² but it was admitted that this statement, offered in British negotiations over conditions on a Monotype machine, was partly for effect, and that the real goal was an arrangement that would minimize unemployment. Concern about dilution of skill was also expressed. 'The real typographer should be able to judge the engraving ... correct its faults, verify the work and give to the whole an impression of beauty, of perfection that the composing machines will never attain.' Followed brutally by: 'There are two irreconcilable things ... the perfection of the type and speed.'³³ Several strikes occurred against the introduction of new equipment. Lille printers hissed and jostled some of their number who were working a new machine, while also hissing the plant director; printers in Toul fought a union directive that they accept the machines and simply negotiate conditions of their work.³⁴ German printers summed up the whole impulse: 'The composing machines must all be tossed on the junk heap.'³⁵

But this was not the position taken. All the printers' unions

recognized the inevitability of new mechanization, and this helped reduce overt hostility. Workers themselves were able to realize some of their objections in the conditions normally imposed over the machines' use. Training was controlled both to limit new entry to the profession and to maintain traditional skills. Hence a variety of regional and national agreements, in all the industrial countries, reduced the number of apprentices allowed per worker (cutting it by half in Germany, for example, while in London the daily newspapers were forbidden from taking apprentices at all and no apprentice could do machine work until after three years of manual training). Everywhere employers were forbidden to train apprentices just on machines.³⁶ Efforts to reduce hours of work were undertaken with a special eye to the unemployment problem. A poster in Brussels justified a reduction demand as designed 'to attenuate as much as possible that crisis that the inevitable introduction of the composing machine will create.'³⁷ And printers could be tenacious on insisting on nine, even eight hours of work because of their concern over this issue, more than once rejecting compromise proposals.³⁸ There were also some informal efforts to reduce the pace of work on the job; in 1901 English printers were accused of ca'canny on the new presses.³⁹ German unions fought, without much success, to abolish production minimums; the French union accepted them in collective bargaining agreements (agreeing to 3000 letters per hour).⁴⁰ Concern about undue production does not seem to have been widespread, once machines were actually in operation. Far more important, and ultimately the basis for acceptance of the machines, was the insistence that machine work be paid more than hand. From France: in insisting on a raise printers 'wanted mechanization, that killer of hand labor, to be not an instrument of misery but one of liberation; they wanted its intrusion less violent and its employment more remunerative.'⁴¹ From England, in asking for a 10 percent minimum advantage: 'machines being an innovation should be treated as such.. . and [made] as costly as possible to the introducers.'⁴² German agreements set the rate at 25 percent above the minimum for hand wages; Belgian did much the same.

Obviously agreements of this sort at least mitigated the anticipated effects of mechanization. And to an indeterminate degree the strong union approach slowed the process outright, by making it quite expensive; this undoubtedly played a role in the lagging mechanization of German and Belgian printing. Unemployment remained a problem, but no greater than before; the industry still expanded moderately. Admittedly printers, with their special union tradition and special

economic setting, achieved gains that many other workers could not. Their situation contrasts, for example, with that of German bottle-makers; here, machines actually improved health conditions, and they did not result in massive dismissals because employers agreed among themselves on a gradual approach and because the machines continued to require skill. But workers generally felt harmed by mechanization, perhaps fearing more unemployment for the future, given the immense productivity of the new machines, but even more obviously because their weak unions were given no voice in setting conditions at all.⁴³ A sense of control, then, undoubtedly enhanced the printers' adaptation. But their willingness to accept machines, however reluctantly and with important conditions, was probably not distinctive in any industry not threatened with extinction. Here the subsequent experience of printers actually on the machines was particularly informative.

There were some reports of nervousness on the job, but these came mainly early in the period, from Germany where the theme of nervousness was particularly common.⁴⁴ Once the machines were installed complaining came mainly from hand compositors, who resented the competition. British printers criticized their brethren on the machines for their high wages and their desire to go at a piece rate to earn still more. Machine work itself was not unpleasant. Workers reported less satisfaction with the skill involved, although some claimed that working with two hands instead of one proved something of a challenge. A few noted that machine work was too specialized, in that they could not see a whole manuscript through; but others found it interesting to deal with more materials, of greater diversity. Certainly they did not regard themselves as ordinary workers. And they had no desire to go back to hand composing; they admitted they were out of practice, but also mentioned their positive satisfaction with shorter hours and higher earnings.⁴⁵

The example of printers, then, suggests not only that in certain circumstances workers in the craft tradition would win some sense of control over the mechanization process but that they could maintain some pleasure in work. Certainly the biggest objections to the process came from those anticipating it or still outside it. But in this instance new machines also pushed workers toward a more instrumental view of their job; intrinsic satisfactions remained, but at a reduced level, and the advantages of the work were partly defined in terms of benefits off the job.

In so far as many factory workers developed or imported reasons for choosing their jobs that were quite similar to those of artisans, it is not

surprising that their reactions to machinery harbored many of the same ambiguities. But with some important exceptions the shock was less great and the fear of displacement less widespread. In certain situations workers could find positive advantages in increased levels of mechanization, because of previous adaptation to technical change and special features of their industry. A historian of the British pottery industry has characterized the workers' reaction to new levels of mechanization, with attendant introduction of semi-skilled (often female) workers, as one of 'grumbling acceptance,' and it will be hard to improve on this judgement overall.⁴⁶

Complaints abounded, of course. Shoe workers were particularly vulnerable, because many had worked in artisanal shops before and were actually new to machines of any sort. So they criticized the noise and vibration of larger machines, while fearing that the new technology plus the introduction of unskilled labor would create massive unemployment.⁴⁷ Shoe workers in Nancy touched on this latter threat in asking for a reduction of hours in 1913: 'The intense development of mechanization in our industry has led to an unheard-of increase in production.'⁴⁸ There were accusations in Britain that workers went slow on the machines, to limit production, leading employers to oppose a piece rate and the union to strive for 'quantities statements' specifying the type and amount of work each category of worker could do. In the craft tradition the British union also tried to set ratios of apprentice to adult labor, to limit displacement. Indeed in the early 1890s the union had to prod its members to work on the machines at all, lest outsiders take their work away.⁴⁹ Generally, however, shoe workers issued few explicit condemnations of mechanization once they were in the factories; their ire was more likely to be directed against the persistence of home workers, which suggests that they had quickly come to find mechanized work appropriate. Like the printers their concern was to win appropriate rates for work on the new machines. A strike in a Paris factory, when a new heel machine was introduced, said nothing against the equipment; it was directed at a resultant reduction in the piece rate, which workers said lowered their pay. Fougères cutters, stirred by the rumored arrival of new cutting presses, reacted only by insisting on a special rate.⁵⁰

Ironically, this kind of adaptation did not prevail so quickly in the textile industry; but here we can penetrate a bit further into the reasons for worker resistance. Right up to World War I textile workers on the Continent opposed the introduction of multiple looms. In Venders they were unanimous against two looms, claiming that every time they had

tried two looms the employers exploited them and unemployment resulted. Hence weavers quit companies that enforced two looms whenever they could, and a major strike resulted over the issue, forcing the employers to back down. German weavers complained about fatigue on the machine, along with the glare from bright lights. The great Crimmitschau strike resulted from a 60 percent increase in machine speed, which harmed worker health and produced insistence on a shorter work-day. German workers widely claimed that the introduction of multiple looms increased physical effort to traditional manual levels, so overt resistance was not surprising. Strikes in Monchen-Gladbach and elsewhere opposed the introduction of two looms.^5l And such strikes were endemic in France. Twelve thousand Cholet weavers struck when multiple looms were introduced, saying 'the profit will be for our exploiters and the fatigue for us.'⁵² Workers in Hazebrouck struck against new looms, citing existing unemployment in the area and the overwork and accidents that resulted from new machines. Here, deterioration in the quality of product was mentioned as well, for the worker could no longer watch over the entire production process. Roanne weavers agreed to experiment with a higher number of looms, but refused to accept the new level or even repeat the experiment with higher wages (the vote was 2434 to 67). Arrnentières weavers successfully insisted as late as 1913 that a two-loom system would be exceptional.⁵³ Finally, although union policy favored accepting multiple looms so long as pay was maintained (as indeed was the case in France) several strikes by wool workers in Britain opposed the very introduction of multiple looms.⁵⁴

Yet the problem in all these cases was not the principle of technical change itself, for these workers were accustomed to machines and could see some benefits in them under proper circumstances. The only complaints against the whole idea of multiple looms came from some older workers, who did not think they could handle them. Even the fear of unemployment did not come too often. Arrnentières weavers, to be sure, were concerned about the effect of two looms on their sons' opportunities, for apprenticeship would be limited. Hence they won an agreement in 1907 that gave the employers only 10 percent savings in labor costs on the second loom while protecting their older single loom rates. Concern for familial employment was undoubtedly involved in other cases even if not directly expressed. Spinners were thus less successfully concerned about technical change than weavers because their incomes, as wives and children of male workers, were supplementary and their jobs not so important to the family economy.

But two other related factors more generally explain the cases of opposition. New machinery in textiles meant new piece rates, and it was easy to feel cheated (whether this was true or not). Hence the pervasive notion that employers pocketed the whole profit. Even more important was the poor quality of machinery and materials that Continental manufacturers typically provided. Weavers in Ourscamp, refusing to work on three looms, explicitly noted the inadequacy of the machines which, unlike equipment they knew of elsewhere, did not stop automatically when the thread broke. Machines of this sort meant more work and effectively less pay, for production could not increase as much as employers claimed in restructuring the piece rates.⁵⁵ And the invocations of greater physical strain and damage to health have to be taken seriously. A letter from a Verviers worker urged manufacturers to perfect their equipment instead of burdening the worker; for if they provided American-style machines workers would do ten looms without complaint. But as it was, 'the intensity of the work is abnormal, disproportionate to the vital energy of a man.'⁵⁶

Textile workers thus had a traditional basis for assessing what machinery should accomplish. Often in poor health, sometimes, as in Germany, new to the industry, they may have been unusually sensitive to physical strain. But machinery itself was not an issue. Hence, even on the Continent, the opposition to multiple looms had no echo in spinning, where equipment was also changing rapidly. The main reason was simple: there was no way to increase the number of spindles a worker operated without improving the equipment substantially, and the pressure on piece rates was far less severe. A more dramatic contrast results from an examination of British cotton weaving, where multiple looms were accepted readily. Larger companies, with more capital and technical experience and a stronger union structure prevented most of the difficulties that developed on the Continent. There was some fear of unemployment by the 1890s —and indeed levels in the industry were at least as great as across the Channel. And workers' technological openness was not indefinite; the initial introductions of Northropp looms resulted in several strikes, and the workers insisted on a ten-loom limit and a reduction in machine speed lest they be overtired. Employers claimed that the workers slowed up on the machines; workers said that the coarseness of materials provided caused inadequate production and hence inadequate pay.⁵⁷ But generally British cotton workers were not only receptive to more mechanization but could even insist upon it, and the problems with the Northropp loom resulted primarily from the pay rates offered. Multiple looms

were accepted early, and the textile unions were quite willing to reduce piece rates in response: 'our object being to see that the amount of labor performed by our Members is not increased, nor their wages reduced.'⁵⁸ Oldham spinners saw that Bolton workers gained, from the introduction of bigger mules, and agreements in spinning normally specified that manufacturers could not reduce the number of spindles per worker below a certain figure. Piece rates were generally set so that workers believed that they won half the benefit of any technical improvement, while a daily wage minimum was deliberately designed to encourage the manufacturers to introduce new and larger machines since they would be obligated to pay a set amount to each worker regardless of machine size.⁵⁹ A far cry, obviously, from the situation of Continental weavers. Yet while admitting the greater industrial experience and probably better health of British cotton workers, it is obvious that their reasoning did not really differ from that of their Continental counterparts. Machines were fine if they lightened labor and produced some wage benefits for the workers.

The final modern industry that developed significant resistance to new levels of mechanization was, ironically, machine building itself. Here the anguish could resemble that of the artisanal trades, for the good reason that many of the workers were former artisans or, as in Britain, had developed a craftman's stake in their skills. From a letter to the Amalgamated Society of Engineers in 1900: 'Wherever automatic machinery is introduced we find that craftsmanship becomes a thing of the past, [and] cheap labour is introduced.'⁶⁰ From an unemployed iron moulder: 'It almost makes me tremble for the future, especially those of us who refuse to be hoodwinked to the tune of "better times."'⁶¹ And many British workers refused to operate new equipment, claiming it was beneath their dignity. Some tried to go slow on machines, preferring to try to protect their jobs to winning higher pay. Far more refused to train unskilled workers, barred them from the union, and demanded that all machine jobs go to traditionally-skilled labor. As often noted, a host of demarcation disputes resulted, particularly in shipbuilding.⁶² A nine-month strike in Hull fought for the general principle that 'machines which supersede hand-skilled labour should be manipulated by skilled and full-paid men.'⁶³

Some similar sentiments could be found on the Continent. A skilled metal worker in Germany proclaimed his resistance to machine work, saying he preferred to earn lower wages: 'I will not be degraded to a machine.'⁶⁴ Shipbuilders noted that mechanization and greater specialization had made their work duller, but also more tense and

dangerous; the noise of air-hammers was cited particularly.⁶⁵ But generally Continental workers were less vociferous in their objections to machinery than British engineers, less indeed than their counterparts in textiles. They saw that machines could defend their industry's competitiveness; some were aided by Marxism to appreciate the need for technical progress. They had not had the time to become attached to specific skills in the industry, and the generalized hostility which a craft background could produce was no substitute for this. Similarly, in Britain itself, workers in new industries such as electrical equipment did not resist consistent technical change. The absence of skill-specific unions in Continental engineering both caused and reflected greater openness to change; unskilled workers were admitted without difficulty and there was no major effort to limit apprenticeship. Adaptation was aided by the continued rapid growth of the industries, which it also facilitated; here was another difference from the British pattern. In Germany the number of unskilled workers in machine building increased by 187 percent in the decade after 1895, but this barely eclipsed the 145 percent growth of skilled labor. Finally, the relative youth of the Continental labor force contrasted with Britain, where the labor force was older and the voice of senior workers more fully protected. Germany reported a number of disagreements between older and younger workers about machinery. Handworkers in Berlin instrument manufacture argued with their younger relatives who went into machine shops without a qualm, where they easily earned more money than their elders. The new workers operating the riveting machines in the Bremen shipyards were satisfied with their work and their considerable pay advantage, but the workers in the older divisions of the company grumbled and cut their production levels to protect their jobs.⁶⁶ The same division might exist in England, but here the older workers were more numerous and less likely to have been demoted to unskilled jobs.

Unskilled workers generally seem to have found the new levels of mechanization untroubling in machine building, if only because they were able to earn so much more than their traditional wages. Hence French automobile workers were long satisfied despite the growing speed of their machines, with no interest in unionization or in a reduction of hours of work. The unskilled in Germany worked the machines according to rule, with no apparent complaint. But here even skilled workers could find some pleasure in machines. It was noted that they enjoyed tinkering with the equipment if a breakdown occurred, for they felt some personal involvement and challenge.⁶⁷ One skilled

worker, while complaining about the dust and noise in his factory, commented that he was not only used to his job but also capable of understanding his machine — on which he could do repairs — so it seemed his tool rather than his master.⁶⁸

So while machines were rarely welcomed in metal work and engineering, the British recalcitrance was unusual. It played an important role in retarding British industrial development, in combination with the manufacturers' unwillingness to push technical advancement too hard. It was not, contrary to some analysis, part of a general British resistance to new technology, for British textile workers, printers, and many artisans were more open to change than most of their Continental counterparts. Even British metallurgy contrasted with the situation in engineering. The smelters' union supported technical improvements 'provided we get a fair share of the plunder.'⁶⁹ Machines such as new loading devices were welcomed, for they improved the regularity of operations; and generally technology was seen as a means of reducing physical strain in the industry. Fear of unemployment was mild, so that traditional skills such as puddling could be let go without lament. And heavy industry did not have the constant readjustments of piece rates common in engineering, so that a sense of being cheated did not develop. In fact metallurgists had little difficulty winning more favorable rates whenever new machines were introduced. In all of this British metallurgists were not different from their counterparts elsewhere, for acceptance of changing technology was common in the industry — accounting in Germany, for example, for low levels of unionization. Generally there were fewer outcries than in machine construction, for the same reasons as in Britain. All of this merely enhances the realization that British engineering constituted an important exception to the pattern of response to advancing mechanization in factory industry.⁷⁰

Technological change in other industries evoked little comment. A few miners refused to work the new coal cutters, and the German mine union leader cited greater tension and fatigue due to the shaking it involved. But there was little resistance, and many workers welcomed the relief from hand digging so long as they thought their wages would benefit. And of course the introduction of new equipment was gradual.⁷¹ British railway workers cited the increased size of trains in claiming that 'the labour, both physical and mental, of the men . .. has been greatly intensified, and their risks and responsibilities have been increased.'⁷² But generally rail workers were known for their fascination with machines. A German engineer enthused over his sense

of power, even aesthetic pleasure, as he drove through the countryside. Engineers generally were proud of their locomotive, often supervising repairs to make sure they were done right.⁷³ This was unusual attachment, of course, though we should not minimize the number of workers in every industry who thought they could 'feel with their machines.'⁷⁴ It was not impossible to regard machines as new, immensely productive tools — for people whose traditions gave them a special reliance on tools and who might find real benefits in some relief from physical strain. Hence workers who judged machines useful and even took a personal interest in them, though less articulate than the opponents of further mechanization, were not rarities.

And in general, without playing down the immense diversity in reactions to new technology, from individual to individual as well as one industry to the next — which could divide the working class significantly, by background and age as well as occupation — the overriding impression must be one of pragmatism. The working class was not in love with machines, but nor was it hostile to mechanization in principle; only those not yet involved in the process but fearing its advent really fell in the latter camp. To be sure, the most obvious advantage of most of the new machines, the reduction of physical strain, did not usually overbalance the increasingly explicit fear of unemployment. This tempered enthusiasm generally. And, perhaps particularly for workers new to the factory system itself, citations of nervous exhaustion could counter any physical relief; this was a widespread impression in German factories. Workers accepted new machines when their bad effects seemed minimal — hence the obvious difference between industries with high unemployment and those with low levels. Where hostile at first their opposition usually softened quickly, again suggesting the absence of a firmly principled antagonism. Softening was aided by the aging of older skilled workers and the possibility of younger workers to move to other fields, as in the case of Carmaux glass workers. Where machines brought advantages, in terms of greater regularity of work and particularly of higher earnings, they could even be welcomed; this was widely true among the new categories of semi-skilled workers. Certainly generalizations about a sense of depersonalization in the face of advancing technology would miss the mark. The machine could make sense to the worker as an individual; it could fit his goals at work.

This leaves a few questions unanswered, of course. We cannot fully resolve the question of whether workers' pragmatism was due primarily to a sense of helplessness, since the time for direct resistance to

machines had long since passed. Unskilled transport and hauling workers, lacking the experience of earlier frustrations in fighting the machine age, alone tried frontal opposition, but only briefly and sporadically. The declining artisans who hated the machines more knew they were powerless. Opposition was further weakened by the fact that every organized labor movement advised against it. There is no question that workers were frequently more eager to attack new machines than their leaders were. The leaders' ideological belief in progress was often involved here, as was their common-sense understanding that there were advantages to be won from an increase in productivity and that resistance was almost surely futile (and futility damaging to a union or party). The fact that they did not have to work the machines themselves was not irrelevant. So local as well as national leaders worked to dissuade workers from opposing machines — urging them to seek compensatory advantages as well — and not infrequently negotiated agreements that they could barely persuade their constituents to accept. It took years for the French textile union to induce its members even to experiment with two looms, though the union was amenable as early as 1904. Beneath official policy a number of British cotton workers were hostile to multiple looms and co nducted some strikes against them.⁷⁵ But this kind of clash occurred mainly at the first sight of machinery, not as an ongoing tension; and the willingness of workers to challenge their unions more massively on other issues suggests that the gap between official and constituent attitudes was not too great.

Resentment against new technology could burst forth in ways other than direct protest. Strikes against the dismissal of workers could at least implicitly challenge technological displacement; this was the case with some general laborers' strikes in England. Workers troubled by mechanization might simply develop a high rate of agitation, without directing it explicitly at machines. Of all French construction workers carpenters were most seriously affected by new methods; this may well have accounted for the fact that their strikes lasted longest and most often touched more than one company, as well as frequently i nvolving demands for a reduction in hours of work.⁷⁶ In English shoe manufacture, clickers, the last skilled group to be mechanized, seemed to take an undue lead in agitation at the end of the 1890s, as they anticipated the change. So the generalized effects of resentments over mechanization should not be minimized. But we should not neglect the effectiveness of much direct protest over technological change either. In a number of cases hostility to new machines slowed, if it rarely

prevented, their introduction. Workers were not totally out of control of the situation, except in the case of rural craftsmen forlornly viewing the rise of urban factories in their trade. Even British engineers, who lost every direct battle over efforts to limit the entry of unskilled workers, clearly played a role in retarding technological change, to the ultimate disadvantage of the industry. So we need not assume that workers who wished to protest technology would have sought indirect methods alone. Change did Cause uneasiness and raised a host of specific issues, but most workers were able to adjust sufficiently to make massive confrontation unnecessary.

Crucial to the adjustment in many cases, as we have seen, was a willingness to accept higher pay for what may have been a more disagreeable job. It seems undeniable that technological change in this period encouraged workers to take a more instrumental view of their work, seeing compensations off the job as essential to make the job endurable. The role of young workers, eager for higher earnings and not yet tied to specific work methods, was crucial here. But a more instrumental view of the job should not be confused with any sort of thorough-going alienation. Few workers found their sense of skill completely eroded by the new machines, and countless unskilled workers, upgraded to a semi-skilled category, may have found their sense of worth on the job increased — even though we do not hear from them directly. Few of the types of workers who can be studied in some detail emerged from experience of machine work without some belief that the job was still interesting, even if it had deteriorated somewhat. Technological change was sufficiently gradual, and the workers were sufficiently adaptable, to prevent a sense of despair. Where, as was not uncommon, workers could relate the new machines to their own interest in tools and in control over production they might even be stimulated by technological change.

This does not close the subject, of course. The new technology was related to other changes in the workplace, for it often created new requirements of discipline and co-ordination. And these changes, in turn, could antagonize workers considerably. But it remains significant that human arrangements on the job, more than technology directly, roused concern in this period. Partly this was because these arrangements changed more rapidly than technology required, partly because workers could see injustice when human agents were involved that could less clearly be perceived in the case of mute machines. In any event, most workers now viewed technological change as a fact of life; only its rate and the conditions surrounding it aroused persistent concern.⁷⁷

Notes

1.    E. Seidel,        'Der Einfluss der Maschine im Getreideumschlagsverkehr,' Die Neue Zeit (1911-12), 1015; H. A Clegg, General Unions in a Changing Society (Oxford, 1964).

2.    Amédée Dunois, 'Le Lockout de la maçonnerie,' Pages Libres (1908), 52; H. A. Clegg et al., A History of British Trade Unions since 1889 (London, 1964), I, 348 ff.; Raymond Postgate, The Builders' History (London, n.d.), 370.

3.    J. B. Platel, 'La Grève des boutonniers de l'Oise,' Voix ouvrière (1909), 237 ff.; Jean Vial, La Coutume chapelière (Paris, 1941), 241; Cari Ergang, Untersuchungen zum Maschinenproblem (Karlsruhe i.B., 1911); Grenoble et sa région 1900-1925 (Grenoble, 1925), 468-9.

4.    Theodor Leipart, Die Lage der Arbeiter in der Holzindustrie (Stuttgart, 1909); Paul Flattau, Das Schlossergewerbe zu Berlin (Leipzig, 1916); Albert Aftalion, La Fabrique et le travail à domicile dans les industries de l'habillement (Paris, 1906), 7 ff.

5.   Alan Fox, A History of the National Union of Boot and Shoe Operatives (Oxford, 1958), 129 ff.; Friedrich Behr, Die Wirkung der fortschreitender Technik auf die Schuhindustrie (Leipzig, 1909).

6.    Schulze-Gaevernitz, The Cotton Trade in England and on the Continent (London, 1895); W. J. Ashley, The Progress of the German Working Classes (London, 1904).

7.    Albert Aftalion, Les Crises périodiques de surproduction (Paris, 1913), II, 7 ff.

8.    Hermann Muller, Die Uberzeugung im Saarldnder Huttengewerbe von 1856 bis 1913 (Jena, 1935).

9.    A. L. Levine, Industrial Retardation in Britain (London, 1967); Friedrich Schinder, Tradition und Fortschritt; Hundert Jahre Gemeinschaftsarbeit im Ruhrbergbau (Stuttgart, 1959).

10.    Roland Kenney, Men and Rails (London, 1913), 104 ff.

11.    Levine, Retardation, 433 and passim; Frank W. Galton, éd., Workers on Their Industries (London, 1896), 63; James B. Jefferys, The Story of the Engineers (London, 1946), 119.

12.   The International Federation of Ship, Dock and River Workers, To Shipowners (London, 1896); M. Pappenheim, Die Lage der in der Seeschiffahrt beschàftigen Arbeiter (Leipzig, 1902), II, passim.

13.   Dockers' Record, September 1912.

14.   Munster Staatsarchiv, Regierung Arnsberg I 73, Mining agitation 1908—9; Emile Vandervelde, La Belgique ouvrière (Paris, 1906), 242; Léon de Seilhac,' 'Les Grèves du bassin de Longwy,' Musée, social (1906), 390.

15.    Voix du peuple, 18 July 1909.

16.    Seidel, 'Einfluss,' passim. A labor leader in Le Havre also argued that machines represented progress and an improvement in the workers' lot, but here his coal-loader constituency initially disagreed. They did finally consent to work the machines for a one

franc raise, which might suggest they were open to persuasion; but they also agreed among themselves to slow up in their work, in order to prevent unemployment. C. Géeroms, 'L'Affaire Durand,' La Vie ouvrière (1910), 737.

17.    Royal Commission on Labour, Digest of the Evidence, Group 'B' (Transport by Water and Transport by Land (London, 1892, C 6795), 44; Norman Dearie, Problems of Unemployment in the London Building Trades (London, 1908), 53; Operative Bricklayers' Society, Annual Reports, 1908, 1911.

18.    Fédération du Bâtiment, Annuaire (Paris, 1911), 286.

19.    Paul Mantoux and Maurice Alfassa, La Crise du Trade-Unionisme (Paris, 1903), 292.

20.    Arthur Shadwell, Industrial Efficiency (London, 1906), II, passim; Adolf Levenstein, Die Arbeiterfrage (Munich, 1913), 127.

21.    Max Morgenstern, Auslese und Anpassung der industrieller Arbeiterschaft betrachtet bei den Offenbacher Lederarbeitern (Freiburg i. B., 1911), 24.

22.    Webb Collection (London School of Economics), XXIV, 1893 flyer on a dispute over new finishing machines.

23.    Ministère du travail et de la prévoyance sociale: office du travail, Enquête sur le travail à domicile dans l'industrie de la chaussure (Paris, 1914), 369.

24.    Robert Wilbrandt, Die Weber in der Gegenwart (Jena, 1906), 96.

25.    FriedrichiBehr, Die volkswirtschaftliche Bedeutung der fechnischer Entwicklung der Schuhindustrie (Leipzig, 1909). By 1909 machine work was being paid 40% more than handwork in the industry in Germany.

26.    Leipart, Lage, passim.

27.    Postgate, Builders, 317; Raymond Joran, L'Organisation syndicale dans l'industrie des bâtiments (Paris, 1914), passim.

28.    Archives départmentales des Bouches-du-Rhône M6—2037; Archives départmentales de la Loire, 92M/130, report of 1905; Richard Ehrenberg, 'Regenerativ-Ofen und Arbeiterbewegung in der deutschen und englischen Grùnglasindustrie,' Thûnen-Archiv (1909), 18 ff.; E. Backert, Geschichte der Brauarbeiterbewegung (Berlin, 1916); Verband der Brauerei und Mùhlenarbeiter, Zweigverein Hamburg, Unsere Lohnbewegung im Jahre 1911 (Hamburg, 1911).

29.    Levine, Retardation. Printers' employers were of course small on the average, though all craft employers were; this reduced their resistance. Perhaps their unusual contact with intellectuals made them more liberal — hence unusual sensitivity to workers' demands. More relevant is the absence of significant international competition in printing. But all this mainly differentiates printers from factory workers, not from other artisans. Alexander Tille, Lohntqrifabkommen und nationale Arbeit (Sarrbrucken, 1906), passim.

30.    A. E. Musson, The Typographical Association (London, 1957), 102.

31.    Le Typographe lillois, 23 October 1909.

32.    Musson, Association, 243.

33.    Fédération française des travailleurs du livre, Compte-rendu du neuvième congrès national (Paris, 1905), 322.

34.    Archives départmentales du Nord, M621/31, report of August, 1909; Fédération des travailleurs du livre, 8ième Congrès national (Paris, 1900), 19.

35.    Korrespondent fur Deutschlands Buchdrucker und Schriftsgiesser, 29 May 1900.

36.    Max Morgenstern étal., Auslese und Anpassung der Arbeit erschaft (Leipzig, 1912); Board of Trade, Report of Collective Agreements between Employers and Workpeople (London, 1911, Cd. 5366), xxviii; Tille, Lohntarifabkommen.

37.    Archives de la ville de Bruxelles, 197, report on 1900 printers' strike.

38.    Gustav Conrady, Histoire de la Federation locale (Brussels., 1921), 103 ff.

39.    Musson, Association, 202.

40.    Fédération des travailleurs du livre, 8ème Congrès, passim; Korrespondent fur Deutschlands Buchdrucker und Schriftsgiesser, 6 August 1901.

41.    La Typographie française, July 1906.

42.    Musson, Association, 230.

43.    Carl Ergang, Untersuchungen zum Maschinenproblem (Karlsruhe i.B., 1911), Glass workers in Carmaux faced even greater trauma, for this was an industry in which technical change could be startlingly rapid. With craft skills displaced older workers resorted directly to socialism and a major strike (in 1895); this was one case where intergenerational change did not work. But union defenses were inadequate. Older workers could only grumble, or head for other areas, particularly villages, in which manual methods of glassblowing were still possible. But even this instance forms only a brief exception to the theme of adaptation. New workers were recruited for the machines who thought they benefited by their elevation to at least a trace of a skilled craft, while the sons of the older artisans headed toward new fields. Protest, relatedly, dissipated. Sudden incursions of technical change cannot be ignored, but the glass industry was unusual and even its workers, overall, not deprived of possibilities of adjustment. Joan Scott, The Glassworkers of Carmaux (Cambridge, Mass., 1974).

44.    Korrespondent fur Deutschlands Buchdrucker und Schriftsgiesser, 1899.

45.    Musson, Association, 253; Hans Hinke, Auslese und Anpassung der Arbeiter im Buchdruckgewerbe (Berlin, 1910), passim.

46.    W. H. Warburton, The History of Trade Union Organisation in the North Staffordshire Potteries (London, 1931), 213.

47.    National Union of Boot and Shoe Operatives, Monthly Report, March 1897.

48.    Voix du peuple, 10 August 1913.

49.    National Union of Boot and Shoe Operatives, Monthly Report, June 1905, and Conference Report (London, 1906), and Report of the Tenth Conference (Leicester, 1892); Labour Gazette, March 1895.

50.    Archives nationales (France), F7-12766, report of September 1909; Archives de la Préfecture de police de la Seine B/al367, report on the Derréal company, 1900.

51.    Laurent Deschene, L'Avènement du régime syndical à Verviers (Paris, 1908), 235, 376; Levenstein, Arbeiterfrage, passim; Der Crimmitschau Kampf um den Zehnstundentag (Berlin, 1905); Heinrich Brauns, Der Ubergang von der Handweberei zum Fabrikbetrieb in der niederrheinischen Samt und Seiden-Industrie (Leipzig, 1906); Zentralverband christlicher Textilarbeiter Deutschlands, Geschàftbericht, 1910, 53.

52.    Voix du peuple, 4 December 1910.

53.    Achille Maertens, La Grève du tissage Planche d'Hazebrouck (Lille, 1908), 55-9; L'Ouvrier textile, 1 May 1914; Archives nationales (France) F7-13820, report of March 1913.

54.    Labour Gazette, passim; Ben Turner, History of the General Union of Textile Workers (Merkmondwike, 1926); H. A. Turner, Trade Union Growth, Structure, and Policy: A Comparative Study of the Cotton Unions (London, 1962), 258 ff.

55.    L'Ouvrier textile, 1 May 1914; Jules Uhry, 'Les Grèves en 1901,' Musée social (1903), 546—7; Voix du peuple, 23 November 1902; William Reddy, ' Family and Factory; French Linen Weavers in the Belle Epoque,' Journal of Social History (1974), passim.

56.    Le Peuple, 8 August 1904.

57.    Turner, Growth, 258; Roland Gibson, Cotton Textile Wages in the United States and Great Britain (New York, 1948), 70 ff.

58.    Amalgamated Association of Card and Blowing Room Operatives, Quarterly Report, June 1890.

59.    John Jewkes and E. M. Gray, Wages and Labour in the Lancashire Cotton Spinning Industry (Manchester, 1935), 71, 118 and passim.

60.    Quoted in Levine, Retardation, 419.

61.    Associated Iron Moulders of Scotland, Monthly Report, July 1891.

62.    J. C. Carr and W. Taylor, History of the British Steel Industry (Oxford, 1962), 277 ff.; Levine, Retardation, All and passim; Labour Gazette, June 1897; Jefferys, Engineers, passim. The ASE twice pushed for admission of the semi-skilled, but a vast majority of the membership voted this down.

63.    Jefferys, Engineers, 143.

64.    Levenstein, Arbeiterfrage, 51.

65.    Josef Neuman, Die deutsche Schiffbauindustrie (Leipzig, 1910); Carl Severing, Mein Lebensweg (Berlin, 1950), I; Heinrich Herkner, Problème der Arbeiterpsychologie (Leipzig, 1912); Waldemar Jolko, Untersuchungen iiber die Entlôhnungsmethoden

in der Berliner Metallindustrie (Berlin, 1911), 71; Levine, Retardation, 452; Niedersâchsiches Staatsarchiv (Hannover), Polizeiakten des Oberpràsidenten der Provinz Hannover (Hann. 122a XI), 70, report on Bremen, 1903.

66.    Alphonse Loyau, 'La Semaine anglaise en France,' La Vie ouvrière (1912), 379; Walter Timmermann, Die Entlôhnungsmethoden in der Hannoverschen Eisenindustrie (Berlin, 1906).

67.    Le\enstem,Arbeiterfrage, 90.

68.    Levine,Retardation, passim.

69.    Carr and Taylor, Steel, 277; statement by John Hodge.

70.    Arthur Pugh, Men of Steel (London, 1951), 110; John Hodge, Workman's Cottage to Windsor Castle (London, n.d.) 89 ff.;Otto Hommer, Die Entwicklung und Tàtigkeit des deutschen Metallarbeiterverbandes (Berlin, 1912), passim. In contrast, metal workers and engineers were often bewildered by the new piece rates introduced along with novel machines, so that they might have no understanding of what they were earning. Alfred "Williams, Life in a Railway Factory (London, 1915), 183; Generalkommission der Gewerkschaften Deutschlands, Correspondenzblatt, 1911.

71.    Larkhall Miners' Association County Union Reports, 1901 (Manuscript archive); Otto Hué, Die Bergarbeiter (Stuttgart, 1913); J. E. Williams, The Derbyshire Miner (London, 1962), 373; Max Metzner, Die soziale Fursorge im Bergbau (Jena, 1911).

72.    Amalgamated Society of Railway Servants, Minutes of Executive Committee Meetings, (1905), report on a London rally.

73.    Hermann Sinn, Aus den Erinnerungen eines alten Eisenbahners (Leipzig, 1944); Waldemar Zimmerman, Die soziale Verhaltnisse der Angestellten im preussischen Staatsbahnbetriebe (Altenburg, 1902), 9 ff.

74.    Comment by a German metalworker: Levenstein, Arbeiterfrage, 65.

75.    Léon de Seilhac, La Grève du tissage de Lille (Paris, 1910); Shadwell, Efficiency, I.

76.    Municipal Workers' Association, Quarterly Report, June 1911; Peter N. Stearns, Revolutionary Syndicalism and French Labor (New Brunswick, 1972), 127, 132-3.

77.    As against this see the recent assumption, offered with no real effort at proof, that semi-skilled workers on the new machines were 'dehumanized.' Edward Shorter and Charles Tilly, Strikes in France, 1830-1968 (Cambridge 1974), 19, 80-1.

CHAPTER 5 RELATIONSHIPS WITH EMPLOYERS

More obviously than technology, changes in business structure set the decades around the turn of the century off from the preceding period. This was a time of rapidly-growing company size and the proliferation of corporations, trusts, and cartels. Direction of work became more impersonal and workers were herded into larger and larger units of operation. Surely, as even non-Marxist historians have noted, class hostility could only increase and relationships on the job became even more tense. This has recently been invoked as an explanation for the post-1909 strike wave in Britain, while analysis that claims a correlation between political radicalism and the size of units of employment would have obvious applicability to this period in all industrial countries.¹

We can grant the existence of pockets of traditional deference. Elbeuf was described in 1900 as a city in which factories of any size dated from only thirty years before, and many small companies dotted the landscape here. Many workers had held their jobs for twenty years or more and their relations with employers were excellent. A few manufacturers noted a waning sympathy during the 1890s, but most workers trusted the owners to deal with them fairly. Women workers in Germany, often deeply religious, objected to any agitation against the Herr; their loyalty was reinforced with a bonus when they stayed out of a strike. Many workers of both sexes, freshly in from the countryside, maintained similar deference. Catholic union organizers in Germany noted how hard it was to rid Silesian workers of the feeling 'that there are only masters and servants,' inherited from their experience on the landed estates.² Flemish workers, similarly resigned, talked of fearing to offend the 'rich people' were they to strike or join a union.³ But this kind of traditionalism would be quickly eroded by the advance of large units of industry, for the sanctions of deference would not long protect the anonymous boards of directors of corporations. Younger workers, at least, would learn to speak out; and we have already mentioned that even many rural immigrants were sensitized in advance to impersonal controls, for growing formality had driven them away from agricultural employers.

Unquestionably, for large numbers of workers we are dealing with the advent or solidification of big business came as a shock. We know

something of the resentments against anonymity that big business can rouse; surely they would develop most readily when the beast was first encountered. Many of the observers who commented on work during this period noted the feelings of separation that were growing between workers and employers. German machine workers, for example, were assigned numbers by a faceless bureaucracy and complained that they were working for the profit of some unknown millionaire.⁴ A familiar and highly plausible description, which the historian could easily confirm. And the issue of employer-worker relations raises directly the question of the meaning of socialism, though we must disclaim any pretense at a full analysis. Both contemporaries and subsequent analysts have assumed a relationship between employment in big business and the rise of socialism (and later, communism).⁵ Certainly socialist factory workers — and by no means party leaders alone — could pour out a bitter hatred against their employers. Metal workers in Nuremburg spoke directly of the basic relationship during a bitter strike: 'The director of this company still believes he holds the power.'⁶ A Belgian miner saw the forces that controlled him even more abstractly, as he condemned the government for failing to regulate the use of explosives: 'This is gambling with our life and death. Ah! we can feel how the present government supports and protects the shareholders. We would impose other guarantees under a democratic regime ... When will freedom come then, to bring us at least a bit more security in our harsh work.'⁷ Other workers, not oppressed by any particular crisis, related socialism above all to the issue of control at work. A Saxon metal worker interestingly combined signs of a traditionalist deference to social hierarchy with this kind of approach:

You know, I never read a social democratic book and rarely a newspaper. I used not to occupy myself with politics at all. But since I got married and have five eaters at home I have to do it. But I think my own thoughts.... We really do not want to become like the rich and refined people. There will always have to be rich and poor. We would not think of altering that. Eut we want a better and more just organization at the factory and in the state.⁸

And there were violent attacks on the persons of employers. On four or five occasions workers in an isolated factory town in France invaded the manufacturer's mansion in what was really a local revolution against all-pervading company power. Violence was even more common when

the normal resentment against employer control was enhanced by the emotions of a long strike. This accounted for the explosives placed at the manufacturers' homes in Chambon-Feugerolles. In Osnabruck, similarly, metal workers stoned and shot a manager who tried to disperse them as they crowded around a struck plant.⁹

Finally, the scanty available polling material that dealt directly with employer-worker relations produced similar evidence of anger. Adolf Levenstein's queries to several hundred socialist workers produced more complaints about the problem of dependence than anything else; the figures are given some credibility, despite the small sample involved, by the fact that areas of known deference, such as Silesia, listed the complaint least frequently. Thus 39 percent of the miners cited dependence as a leading problem, along with 37 percent of the textile workers; one textile worker noted that his material needs were small, but he resented the necessity of obeying a boss who could fire him so easily. Metal workers, whose industry was increasingly dominated by big business and whose unions were relatedly weak, put dependence at the top of their list in 53 percent of the responses.¹ ° One lamented: 'The consciousness of dependence on the employer embitters me, has made me an irritable man. A gesture of the director... is enough to make my blood boil.'¹¹

None of this is meant as a definitive comment on the meaning of socialism to German or Belgian workers. One can easily find other material suggesting that socialist voting was often seen in more prosaic terms; miners interviewed in the second Belgian general strike most commonly associated a socialist victory with higher wages alone.¹² But the link between political protest and big business is more than plausible. It could certainly help explain the stunting of socialism in England, where trade union power muted the impact of impersonal business forms.

A chapter on worker-employer relations cannot, however, be simply an embellishment on the theme of alienation from big business. Without intending to refute the framework suggested thus far, it must at least be modified. Most workers were not as angry about their undeniable dependence as the ones depicted above. Violence against employers was actually rare, even in France. Verbal attacks were not always meant to be taken literally. German dockers blasted employers in a public meeting, scaring them in the process, but then admitted to a socialist journalist that many were normally friendly and fair. And, though we need more information on the industrial incidence of socialist voting, the association of political protest with subjection to

big business was certainly incomplete. If the special example of England is illuminating, so is that of the United States, where much of the support for political protest came from artisans, not factory workers. Such was indeed the case in Germany and Belgium as well. After the miners, for example, it was Belgian artisans who supported the socialist general strikes most avidly; metallurgical workers lagged. And while artisans unquestionably had organizational problems of their own, which we will discuss shortly, they were not exposed to the leading edge of the rationalization effort.¹³ Quite apart from the thorny question of socialism, it is dangerous to theorize too abstractly about reactions to rationalization and bureaucracy. Workers can be studied as models of a modern accommodation to organizational control at least as fruitfully as rebels against it.

The following discussion will offer several major points in lieu of a simple pursuit of a protest theme. First, predictably, the question of diversity. Not only were pockets of traditional deference large, but the types of employer-worker relations defy generalization. We must discuss the artisan, the unskilled worker, and the factoxy hand separately because their situations differed so radically, though each was touched by some aspects of the rationalization process. Each group had different possibilities of response, which can bring us back to some effort at political correlations. National circumstances differed as well, for rationalizing manufacturers were not faceless types, immune from distinctive cultural and political causation. For all groups, but particularly for factory workers who obviously constitute the most important case, we will offer three other considerations. First, rationalization was not an abstract issue for most workers; it had to be seen in human terms and responses would vary according to the personalities, as well as the systems, involved. This does not necessarily displace the alienation model, but it at least adds depth to it. Second, rationalization proved extremely hard to attack directly, partly because of the weapons big business possessed but partly because of distinctive ways in which most workers judged their employers, large or small. Difficulty in direct attack may bring one back to the question of more general political response, but this is not the only point. For, third, rationalization brought some benefits to some kinds of workers in terms of the framework for their job. We will find ambiguity at many points, as workers resented new, impersonal controls but complained mightily about the personalized abuse to which they had been subject earlier. German miners might, for example, have been somewhat less concerned about dependence in the abstract than metal workers, as

Levenstein's figures suggest, for they were under more individualized direction; but they were far more upset by the results of whimsy. Most workers were dependent, and big business could make them more so; but within a system of dependence new organizational forms could alleviate some of the problems of being directed by other people, which is what the whole subject of 'relationship to employers' meant to workers on a day-to-day basis.

The most general point is this: rationalization probably did not greatly heighten sentiments of dependence. Only among some unskilled was a demonstrably new claim for dignity and freedom being pressed. Many other workers were capable of some effective counteraction against organizational change. Still more widely, business structure was judged mainly in terms of the personalities involved. Intermediaries might mute organizational tensions or draw workers' fire; in either case they distracted from what an outsider might judge to be an increasingly impersonal job structure. For most workers personal dependence was in fact more galling than big business controls, but it was by its very nature extremely variable. Hence a theme of general hatred would be misplaced; among other things personalization meant that all but the most acerbic workers might relieve their tensions by changing companies. Class consciousness was rising in politics, and changes in business forms may have played a role here. On the job, however, it is difficult to perceive any consistent pattern of growing structural tension. Even the personal battles maintained a rather constant rate.

The speed of rationalization can be roughly measured by the growth in the size of units of employment, as far as the impact on workers is concerned; diversity, which takes on the expected contours of craft vs. factory, can be briefly charted as well. We need not linger too long over the inadequacy of the statistics. In Britain numbers of employers and of workers can be compared in key industries, conveying fairly adequately differential rates of growth; but these rates cannot be placed against French or German figures, where the number of companies is known. Belgian figures similarly result from a worker-employer comparison with some predictably misleading results; as the number of managerial personnel outstripped workers in mining, for example, the probable modest growth of the companies in the extractive industries is obscured. Even for France and Germany, where figures are most reliable, exact comparisons are faulted by the fact that France treated independent producers separately; this makes little difference in heavy industry, but inflates French figures vis-à-vis German in the light industries and crafts. Differences in definition of key categories also

mars any comparative effort.¹⁴

All that comparison suggests (Table II) is the essentially common range of the industrial countries. Belgium, as an older industrializer, had much more concentrated firms than Germany in 1896; even the big German surge thereafter did not bring Germany up to the Belgian levels of a decade earlier. The heavy influence of craft structures in Germany is readily apparent; from the working-class standpoint, Germany was not a distinctive home of big industry. Nor, to get at another tenacious stereotype, was France exclusively the haven of small shops. In contrast both to Germany and to Belgium, French mining was highly concentrated, while metallurgy and chemicals were within a common range. German textiles, and quite probably other light industries, were less concentrated than French or Belgian. Printing was remarkably similar, as were the more traditional crafts. The German distribution of workers among sizes of companies probably leaned a bit more towards larger units than did the French, though precise figures on the latter are lacking, if only because of the different balance among industries. But the differences were not great. The example of Belgium suggests that, given the huge influence of crafts and small shops, age of industrialization more than sudden upsurges of concentration in the most visible branches of production explains national differences. Were British figures available they would undoubtedly point to a similar contrast, for the shop form was much less important in Britain than on the Continent.

For the range of company sizes, and worker distributions within the various categories, was considerable in all the Continental countries (and indeed in Britain as well, though the variations were less pronounced given the greater previous penetration of factory organization). Only in heavy industry was there virtually no small-shop sector. Light industries at the end of the period occupied a middle range. Few were so large, for example, as to prevent some contact between employer and those workers who remained with the concern for any period of time. It seems clear that until right before World War I less than a quarter of the work-force (considerably less in France and Germany) was employed in units in which impersonality was dictated by sheer company size. The artisanal trades, some of them huge users of labor, constituted yet another category, in which the average firm rarely employed more than five workers even after 1900. (Printing was an exception, because of the virtual absence of tiny units.) Averages are to some extent deceiving here, but not in terms of overall range. Only in construction did the small average size increasingly conceal a sharp

gap between the majority of big-city workers, employed in units over fifty, and a minority of small-shop or independent producers. In industries such as woodworking or leather, the average company size reflects fairly even gradations of workers from the small shop to the large factory. In the German wood industry in 1907, the average company employed 3.2 workers; the average worker was employed in a unit with a labor force of almost exactly thirty people. This figure in turn is too simple; 11 percent of all woodworkers held jobs in factories with over two hundred employees, another 12 percent in units of over one hundred.¹⁵ But in so far as averages can convey anything, in these industries where a division had opened between shop and factory, it remains fair to characterize the normal experience as one in which the worker held a job in a unit near the borderline between shop and factory. Finally a few large industries were still dominated by shop and home work. Almost 80 percent of all French bakers were employed in units with less than five workers in 1906, and almost half of all clothing workers. The size of firms was increasing; between 1906 and 1911 the number of bakery owners dropped about 3 percent, while the number of bakers increased about 2 percent, but the small firm remained preponderant.¹⁶ For the working class as a whole it is clearly impossible to talk about a common employment experience.

Furthermore, while the evolution of company structure reduced the gaps within the working class — a development of real potential significance — it did not eliminate them. And the differential quality of the evolution raises some additional problems of its own (Tables III and IV). In general, company growth was relatively modest in sectors such as heavy industry, just as it was slower in countries (such as Belgium and probably Britain) in which the big firm had already established itself. Thus except to an extent in Germany, where the mines had been relatively small, the rate of expansion was not great. Metallurgy gained somewhat more rapidly but was rarely a leader. For workers in these industries, especially new entrants, it seems safe to suggest that adjustment to the fact of large units, more than rate of change, would pose an ongoing problem; and indeed there were few expressions of a new kind of concern. In the lighter industries such as textiles and engineering, the advent of the big firm was newer and the rate of change far greater. Machine builders were thus faced with new business structure as well as new techniques, textile workers with bigger companies along with a pronounced decline in the industry's vitality. The experience of key artisanal sectors was potentially at least as troubling, for high growth rates were coupled with a firm traditional

attachment to small units of organization. Printers were relatively immune from change, but for woodworkers, food processers in some areas, and leather workers this was a period of real upheaval. Even more consistently, the construction industry entered a period of intensive rationalization.

Thus the notion of advancing rationalization seems quite accurate in describing the pre-war decades, but its impact on the working class was more subtle, and for this reason more pervasive, than attention to the progress of cartels and industrial finance might indicate. Change was now hitting the middle sectors of manufacturing, and this, more than the solidification of big business in heavy industry, affected a large percentage of the working class. Despite the rapidity of change, however, it can again be noted that the effects on workers tended more to divide generations than to displace individuals. Even in Germany, where the traditional base was substantial and the rate of change considerable, no category actually decreased in size. Most workers employed in units with less than five employees, for example, could expect to remain in such a unit. And in no category did the average size of firms increase massively. What was occurring was a tremendous shift in the relative importance of each type of firm, with the larger units

Table I Differences in Company Size

Table II Comparative Suggestions (German figures: 51+)

Relationships with Employers

Table III Evolution of Company Size

A. Germany

B. Belgium

C. France

Table III (continued) D. Germany

E. Britain

gaining ground most rapidly. In Britain, for example, while workers in factories increased by 21 percent between 1897 and 1907, those in shops still advanced by 3 percent. In important individual cases, such as construction work in Paris or Berlin, workers would clearly have to shift from one form of business organization to another during their own working life. More commonly, however, it was the new entrant to the labor force — particularly among the skilled workers — who supported the rise of the large firm. The persistence of the small company requires some brief attention before we turn to the more obvious sectors of change. And the intergenerational nature of much of the change, the fact that young workers had to move from the kinds of structures in which their fathers had been employed, complicates an assessment of the impact of change.

We can begin to fill in the human outlines of the rationalization process by a brief look at the two most traditional sectors of manufacturing, both declining though in the one case not too rapidly. Domestic manufacturing requires little comment. Its practitioners might have a host of complaints, and their material situation was bad, but a sense of dependence on the employer rarely loomed large. Some were employers themselves, and exploited their apprentices; this reduced any class sense.¹⁷ More commonly they were at the manufacturer's mercy, as domestic industry became steadily less competitive. But the scattered character of the work and its operation, often by women or workers who did not depend on it for their entire support, reduced any ability to articulate resentment; the urban concentration that had produced earlier revolts, such as those of Lyons, had largely disappeared. Hence the home workers studied by public officials usually expressed a sense of personal control over their work, even if they lamented dire poverty. Belgian workers were normally courteous to their employers, who often visited them if they were ill. Their fear of unemployment ironically made them grateful to the manufacturers who deigned to give them work.¹⁸

We can assume that those older home workers who hung on despite the growing competition of factories blamed, not their employers, but outside forces. We might also hypothesize that those sons and daughters directly forced into the factories lamented their loss, having seen an alternative form of employment to contrast to factory life. But if so they are mute on the subject. We do not know how many people were displaced directly, for the shrinking of the domestic labor force was due more to attrition through death or the voluntary retirement of people who could or would not move and for whom this work had always been

supplementary. Ironically those factory workers who were articulate on the subject condemned the system. This reflected their fear of low-wage competition, of course, but also some acceptance of factory employment if not of factory employers. Some might even recall the personal tyranny possible over apprentices and aides within the domestic system, for the control of factory life could seem more remote. Unionized French clothing workers put it this way:

Considering that all home work .. . can only be harmful from all points of view to the working class and especially to our corporation, that it is only in shops that we can attain our goals and profit from the laws regulating our corporation . . . the Syndicat de Bordeaux expresses the wish that each Union issue the greatest possible propaganda to attain the complete suppression of home work and the return to the shop.¹⁹

Here, in one of the only cases in which the domestic system was still spreading, many workers preferred organization to the vagaries of a more individualized economic structure.

Ambiguities within the small-shop artisanry can be explored more fully, though they run along the same lines, almost certainly dividing urban from small-town craftsmen in key trades such as baking, butchering, barbering, tailoring, and even some traditional branches of metal work and woodwork — indeed in all the fields where a vast number of workers labored in units of less than five employees, many serving as the only worker in the shop. The small-shop workers spoke scarcely a word about their lot, but many of them were clearly satisfied by it. In Verviers, 95 percent of all bakers expected to become employers in their own right, and believed that they had essentially the same interests as their bosses. French hat manufacturers worked alongside their employers, whom they respected so long as the manufacturers knew the trade. Signs of solidarity were compatible even with unionization. British brass workers, in a well-known example, co-operated readily with the employers to maintain high quality standards and limit production; ultimately the resultant high costs priced the brass bedstead out of the market. Skilled cutlery workers in Solingen co-operated similarly with employers in keeping wages up as a means of reducing competition in the whole craft. Furth metal beaters banded with their masters to resist the rising prices imposed by the capitalistic manufacturers of the raw materials. In Graulhet the owners of small tanneries encouraged the union as a means of inhibiting the

Table IV Evolution of Company Size: Summary

Annual % Rates of Growth (betweeen census periods: Germany, 1895— 1907; France, 1896-1906; Belgium, 1890-1910; Britain, 1891-1911).

growth of large, mechanized companies.²⁰ Augsburg bakers and their masters formed a single, anti-socialist union, dedicated to loyalty to the master and defense of hand work. Belgian bakers urged their employers to organize, noting that 'Our future ... is intimately linked to that of our employers,'²¹ so that if they controlled the trade more fully workers could only benefit. In a similar display of solidarity, woodworkers in the Taunus region sang the following refrain to the founders of their shops:

Health to the Noble Founder You did a good job of work For us and for our children And for the working man. Hail to the Firm!²²

Surprising sentiments perhaps, but they are worth attention for two reasons. First, although they come from waning manufacturing forms, the types of workers involved, along with their employers, would not yield easily. Birmingham brass workers were headed for near-oblivion, but the small-town baker could hold out for decades. So the numerical importance of this form of work is not only a turn-of-the-century oddity. Second, they remind us of the difficulty of interpreting even highly personalized employment. A gut reaction to the Taunus song could easily be to assume that it was forced sycophancy. It may not have been; gratitude could have been genuine or at least intermingled with disgruntlement. The ambiguity is worth noting because it will recur when we deal with highly sophisticated business organizations.

In France and Germany, the testing ground of reaction to traditional craft norms came over the question of masters' providing board and room to their workers. Only two points are certain. Board and room were surprisingly persistent in these traditional trades. And in contrast the opinion of all articulate workers was extremely hostile. An investigation in Germany, during 1908, revealed that of 5542 bakers, employed in 2111 companies, 3771 were boarded and housed, 810 more housed by their employers — often in very meager conditions. In 416 butcher shops 1194 of 1370 workers were boarded and housed and 106 more just housed. The practice was almost universal in German barber shops, widely noted also among tailors and millers, and rare but not unknown for some smiths, bookbinders, saddlers, cobblers, woodworkers, and transportation workers. In 1904, 47 percent of all German bakers were boarded and housed, while four years later the

percentage may have risen to 60 percent; and many others were tied to partial arrangements. Even socialist observers had to admit that many bakers were contented or at least uncomplaining, for their view of the employment structure was frankly pre-industrial. But the system was gradually decaying, though statistical information is unclear. The German bakers' union, for example, managed to win abolition of board and room for 4000 workers in 1905, and this was not an uncommon strike goal. The picture in France is less clear, for the small size of the French bakers' union suggests an even greater attachment to the old order.²³ More and more workers turned against their obvious dependence, complaining about their inability to own a home or start a family as well as the inadequacy of their material conditions: 'Nowhere are the conditions for the workers so bad as in the baking profession.'²⁴ Verviers bakers, though attached to their employers, decided they wanted the freedom — and sufficient wages — to eat out. And this in an industry whose structure was changing imperce ptibly if at all. For workers of this sort the example of factory employment might have suggested greater freedom. Certainly something was stirring in this traditional setting, which trade unions reflected as well as promoted. For their part many traditional masters found their labor force increasingly difficult. A Rhineland master noted: 'The familial relationship between Master and Journeyman is disappearing.'²⁵ Many of the surrogate guilds formed in these professions in Germany, the Innungen, worked vigorously against the workers' union movement. Perceptions were changing on both sides — indeed the notion of sides was new; and without much alteration of actual practices a sense of constraint gradually developed.

For the bulk of the artisanal trades, as we have seen, structural change was quite real. Far fewer construction workers or metal workers had ever been employed in a one-to-one relationship with an employer, but the experience of some equality was relevant here too. The head of the French typographers' union commented that his workers were accustomed to small shops where they could work alongside their employer, who was sometimes a former comrade. This was pushing things a bit, in terms of the actual structure of French printing, but reflected the possibility of considering even a shop with twenty or thirty workers as falling within the range of a comradely txadition. These were industries also in which the sense of dependence was clearly muted by the fact of having dependents under one's own direction. Skilled construction workers had aides, and they often lorded it over them; German masons for example used the familiar form of address to

their aides, but expected the formal pronoun in return. So where growth in company size was not excessive there may not have been a complete overturning of traditional relationships. And the range of companies in these crafts, combined with the skill of most workers involved, provided the possibility of some choice. Workers in British printing, where over 1700 companies had less than nineteen workers in 1909, compared to less than 120 with more than twenty, or in German woodworking, still had the possibility of opting for the small-shop environment, often at some sacrifice in pay and conditions, if this was particularly important to them. Of course, this was not a free choice, and the rapid rate of concentration in most of these industries steadily limited it. But these were workers long separate from the literal artisanal relationship of master and journeyman, and they experienced this new set of structural changes without.much explicit complaint.²⁶

Only in the construction industry was change so great as to cause particular remark. In the Parisian building trades between 1901 and 1906 the number of companies with fewer than twenty workers declined by 160, while those with over a hundred rose from twelve to twenty-five. This kind of structural change brought a number of innovations in the organization of work. In all the industrial countries the scale of operations in the construction trades, around the major cities, increased immensely; hence a widespread effort to set firm conditions on travel costs. The use of foremen spread widely; by 1899, encouraged by the unions, the system of subcontracting had largely ended in the London building trades, and foremen replaced them. These in turn could be quite nasty to the workers, but here was a case in which rationalization could at least potentially remove a longstanding grievance against more personalized exploitation. French construction workers continued their traditional struggle against subcontractors, whom they accused of forcing the pace of work while trimming the wage payment in order to make good their contract; even in big-city construction, traditional organizational forms did not give way easily. Parisian workers claimed that the number of subcontractors actually increased as companies grew. The advent of big companies also produced a host of complaints, in both France and Germany, about poor training and shoddy planning. The upsurge of protest in the Parisian trades between 1906 and 1909 directly reflected the advent of big employers, and the union made an overt attempt to attack the new capitalists while protecting the smaller units.²⁷

But here, and certainly in manufacturing crafts where structural changes were less great, skilled workers quickly developed or elaborated

a response to new organizational forms. Craft unionism was a direct counter to the decline in personalized employment. It expressed the tensions that the new business forms created, but it also maintained many of the traditions of the craft. Where craft unionism was already well developed it tried to expand its functions to take the new organization of the industry into account. In 1899 the British plasterers' union attempted to force foremen to join the organization, to preserve the mutuality if not the personalization of the industry (Parisian construction workers did the same in the 1906-1909 uprising.) The plasterers had to back down on this demand, but continued to claim coequal rights with employers to determine what work a plasterer should do. And British construction unions attempted, with considerable success, to maintain an elaborate set of work rules which gave them considerable authority in the work site despite the growth in company size. This was in addition to the proliferation of collective bargaining agreements, which the construction unions pushed for vigorously; in 1899, 141 of 173 new agreements signed were in the building trades, which by then had almost 500 pacts in 260 towns and cities. French construction workers signed many local agreements; as in Britain, most contained conciliation and arbitration procedures which maintained some mutual contacts between employers and workers.²⁸

The significance of the craft approach in mitigating a changing industrial structure stood out most clearly in Germany, where these unions were almost alone in winning collective agreements. A Munich building contractor in 1903, noting the rapid rise of bargaining, reflected the relatively conciliatory spirit that normally characterized craft employers as well as workers: 'Worker organizations . . . have become powers that must be reckoned with and must be given a voice in job conditions.'²⁹ Collective bargaining spread quickly in the smaller metal companies, in wood, and in food processing as well as construction. General agreements were won with some of the masters' guilds in metals as in construction, while in contrast company-wide agreements penetrated factories rarely and usually when a strong craft element was present. In 1912 it was estimated that between 70 and 80 percent of all agreements in Germany involved companies with forty workers or less.³⁰

This is not the place to rehearse the detailed history of craft unions, but it seems obvious that they played a major role in the reaction to changes in company size. The German statistics suggest that they were most successful in companies that grew relatively modestly; a few agreements were won in the larger metal, wood, and food processing

plants (far more than in heavy industry) but most workers were not covered. And some of the victories even in the crafts were hard-won, as employers in construction and woodworking periodically mounted counter-attacks. But on the whole crafts unions could maintain a reasonably conciliatory stance and their achievements undoubtedly moderated organizational pressures. Here, then, is a case where change produced a response of great significance, but by that very token avoided complete disruption. An artisanal tradition that might have turned against more advanced forms of organization was instead used, particularly in the medium-sized shops, to elaborate compensatory organizations which preserved some important elements of mutuality. The artisans involved seemed able to adjust to the greater impersonality involved.

In contrast, unskilled workers demonstrated acute symptoms of anxiety over the new business structure in which many of them were now employed. Figures are unavailable on the sizes of companies employing unskilled workers, but obviously changes in the construction industry and in transportation (See Table III D) affected them considerably. The rise of big shipping companies transformed the docks as well as the merchant marine, and it is here that the most anguished reaction developed. Rationalization had a number of potential advantages in this area. When German shippers set up a job office workers welcomed it, despite its unilateral character, for it eliminated the traditional corruption by which jobs were obtained through personal contacts with inn-keepers, often with a generous dosage of bribes. In the 1890s Hamburg sailors had specifically complained about the expense of working through middle-men, who paid no heed to their professional competence but charged them considerable fees and often dictated where they could sleep or buy uniforms. Big companies like the Hamburg-Amerika line, and then a shippers' union, eliminated this problem. But generally attachment to a personalized system was greater than this. British dockers, for example, were often on good terms with their foremen, who gave them drinks or tips if they lacked work. Construction laborers frequently followed a familiar foreman from job to job, hoping to get work. And, with a few exceptions, rationalization seldom quickly eliminated the element of corruption. Foremen for the big companies building the Paris subways required the laborers to buy from their canteens. In the long run the advantages of rationalization may have become more apparent, and it is true that the bitterest battles that big business triggered among the unskilled had given way to more restrained dealings by 1914; but few records exist of a positive

adjustment.³¹

For the unskilled worker new business structures meant closer or at least less sensitive supervision. Many disputes by tram employees broke out when a new administration, bent on rationalizing inherently decentralized procedures, decided to tighten its grip. A strike in Charleroi resulted when inspectors flooded a line without warning — soon after tram lines in the Center had been consolidated; a conductor was accused of falsifying figures on the electricity he used in his daily run, and fired. His reply summed up the outraged dignity that characterized so much of the agitation of the unskilled: 'You not only want to rob me of my bread, but also you want to dishonor me.'³²

The reaction to larger units of organization was particularly keen in the seaports. Sailors were touched not only by big companies but by the increased size of their ships as steam won the day from sailing vessels. French seamen had long been under tight discipline; the state gave captains immense disciplinary authority, including corporal punishment, while desertion or rebellion were punishable by military law. This was not new, but organized resistance was; and the trigger clearly was the new structure on board ship. Particularly in Marseilles, in the period of recurrent agitation on both docks and ships between 1900 and 1904, attacks on strike captains were endemic. Many a captain, accused of strict discipline, had to quit in order to prevent a company-wide strike; ultimately the captains organized to protect themselves. The sailors more generally complained that the 'familial' relations that used to prevail on board had given way to a strict hierarchy of command. The Marseilles outburst was unique, but German seamen similarly agitated against despotic captains. And in Britain big business brought a more specific complaint, particularly in the years immediately before World War I. Medical examinations, an obvious part of rationalized personnel procedures, seemed an intolerable intrusion.³³ Sailors felt 'mauled about and handled as though they were a piece of New Zealand mutton.'³⁴

Clearly the advent of big business, more than any other innovation, caused an extraordinary outburst among many unskilled workers, in word as in deed. Disruption of personalized direction brought an intense desire for greater dignity and liberty. Ironically the advent of big business corresponded with the settling down of many previously nomadic workers; so a traditional outlet for resentments against employers, simply wandering to another area, had less validity — again resulting in the demand for freedom on the spot. Quarry workers in Argenteuil asked for abolition of obligatory company housing and

canteens. Marseilles sailors insisted on abandonment of the 'red book' in which any dereliction of duty was recorded and which could serve as the basis for a general blacklisting. Carters wanted to be treated as something more than 'beasts of burden,' while merchant seamen claimed that they were 'impelled by a sense of dignity' to 'revolt against our tyrants'.³⁵ Sentiments of this sort helped produce in turn the upheavals on the docks (London, 1889; Hamburg, 1896; Marseilles, 1899—1904) that have justly received attention as part of a new kind of working-class unrest.³⁶ Potentially, at least, reaction to big business could impel unskilled workers into the mainstream of the working class for the first time. But insecurity of employment and the power of the big shipping and construction companies reduced the first wave of unrest rather quickly. Shipping companies set up their preferential hiring lists with only a hint of worker representation in their governance or, as in Parisian construction, enforced a lockout, and the wildest effervescence was over. The result was that during most of the pre-war years unskilled workers were exposed to big business without real intermediaries, for their unions had been beaten down. Small ports like Sète or Flensburg saw unions replace traditional hiring corporations and take a major role in dealing with employers; in Sète this even involved a union shop. But in the big ports rationalization proceeded without steady opposition, with a result that workers had to abandon any desire for a return of personalized direction. The fact that rationalization may, after its first onslaught, have brought some benefits could have played a role as well. Certainly when these workers were ready to rise again, as in Marseilles after 1909 or on the British docks in 1911, they were much readier to pit their organization against that of the employers. The bitter, personalized disputes, against an individual hiring boss or captain, gave way to more structured, usually calmer struggles. British sailors in 1911 thus asked not that medical examinations be abolished but that they or the union be allowed to supply the doctor; they sought no return to personalized hiring but wanted control to be taken out of the shippers' office, to be shared at least by the union.³⁷

A similar clash between experience of personalized direction and the reality of new business forms characterized other unskilled or semiskilled workers. Impressionistic evidence suggests that new women workers in the factories were extremely sensitive to personal grievances — more so than to any other problem on the job. We have suggested already that some found the factories a source of freedom, as against parental control or domestic service. But this very sense of

freedom created sensitivities when the factory proved disappointing. Charles Booth found that factory girls rarely complained about their work, pay, or even fines, but excoriated 'the foremen who treat us like animals.'³⁸ A strike in a Nuremburg shoe company resulted when a foreman talked crudely to female workers, while several strikes in French textile companies raised similar grievances.³⁹ Here, of course, it was the quality of personal treatment, rather than business structure, that most obviously triggered agitation. But unskilled workers generally responded more to personal examples of harshness than to the sources that might produce them. It remains true that exposure to new business forms brought large numbers of previously docile workers to a realization of new rights. In the case of women, the desire for individual dignity could turn against labor organizations as well. In 1906 female textile workers in Denton rejected a union membership campaign because the union had repeatedly insulted them and ignored their interests in the past.⁴⁰ And for unskilled workers generally it would take time for the imperatives of organization to be sufficiently accepted to permit a coherent campaign against the new structure of employment. We can glimpse only the outlines of this evolution, precisely because it was incapable of producing elaborate records; but it was almost certainly the most poignant clash between worker expectations and the new world of business.

Factory workers, still most directly under the thumb of big business, had a variety of complaints about the changes in company structure. British miners suffered as their companies grew, for their foremen and managers were now more rigorously controlled by the Home Office. Not only did they work to keep costs down, but also they became less receptive to worker complaints. Young Welsh miners, returning from a year at Ruskin College, were appalled at the arrogance of the managers, for all their newly-polished manners. Foremen now referred workers to the managers, who stalled them in turn: 'What the bloody hell are you bothering me about?'⁴¹ Workers at English Ford resented the tight supervision under which they worked and the requirement that they wear badges. Locomotive manufacturers complained about the new system that compelled them to report to the company doctor and the manager after an illness, feeling this was a threat to their job; some felt that workers regarded as troublesome were dismissed for false medical reasons. And again the report of constant supervision, leaving workers in a state of continual agitation and suspense.⁴² German miners grew more discontented with their piece rate when it was set by a remote central office.⁴³ Grievances of this sort can be multiplied, and they are

undeniably significant. But, impressionistic as they are, they fall short of a general, articulate resentment against new business forms per se. More commonly, workers displayed anxiety lest a new management change their conditions or attacked the personality of the employer directly. When a new combine took over several electrical utilities companies in Paris, workers struck to guarantee maintenance or improvement of conditions. Older miners in the South Ruhr worried about the big companies' insistence on closing unproductive pits, which forced them to move north to keep their jobs.⁴⁴ More directly steel workers in Aachen attacked their multi-millionaire employer as a man who treated workers, 'not as men, but as tools,' while leather workers in Chaumont blasted one new director 'who was only a financier anxious to get rich as fast as possible.'⁴⁵

Clearly, factory workers were troubled by new business structures. But they were divided in their response, to a greater extent than most other groups within the working class. Some of them found advantages in big business. Far more directed their attention to older issues, which rationalization exacerbated but did not cause. Partly because the structural changes imposed on factory workers were less radical than those in the artisanal or unskilled sectors, partly because the factories maintained an unusual repressive power, workers rarely tried to confront big business directly. Issues of principle were seldom posed, for workers judged big business as it affected them on the job, and these contacts were judged mainly by the personalities involved. Change produced some general uneasiness, and unquestionably the development of the labor movement in the factories owed much to this. But aside from some workers who spent much of their free time hating the employing class generally,⁴⁶ the impact of rationalization was assessed mainly through a variety of intermediaries.

Divisions in reaction were partly a matter of personality. Others resulted from important differences among industries. The survey of German socialist workers, while producing a general distaste for dependence, suggested that metal workers were more troubled than their colleagues in textiles or the mines — for the good reason that the factory system was newest in this industry. The loudest laments about impersonality came from workers whose operations had previously been organized on a shop basis — like engineers or leather or woodworkers now forced into factories. A locksmith in Germany, village-trained, recoiled at the very sight of the factory's size: 'I didn't feel really at home in the big hall.'⁴⁷ Others found their foremen, traditionally more like craft masters than employer representatives,

joining them in resisting the systems analysts who began to set standards in some big companies. Metal workers in the Rhineland resented the tone of the educated young bureaucrats who replaced the old masters in their concerns.⁴⁸ British engineers were even more openly hostile, for they had a more precise small-shop tradition for reference. They nearly rebelled against being ordered about instead of being left free to determine their own tasks. Hence the blasts against 'policemen in various shapes and forms and under different names' who forced workers to be 'watched and dogged by a whole army of non-producers.'⁴⁹ Textile workers, though also undergoing a rapid expansion of company size, did not evince this kind of anxiety, because they were already accustomed to a factory system. Change, then, rather than degree of bureaucratization was the key to specific response. Hence metallurgical workers, although their companies were five times as big as those of the machine workers and their treatment far less flexible, evinced little concern about business forms. Still other reactions resulted from differences in specific traditions and circumstances in the industry. Structural changes in the mines were less significant than those in metallurgy, but miners were far more likely to show uneasiness about a loss of freedom. A few even produced pleas for liberty similar to those of unskilled workers; thus miners in Montceau, rising against a local tyrant: 'We ask only our place in the sun.'⁵⁰ Miners had a tradition of protest. They were residentially more dependent on the factory than many metallurgists, who could move more freely. Reaction to the employer depended on a host of factors beyond sheer company size or level of bureaucratization.

Important aspects of the rationalization process reduced tensions on the job, at least for some types of workers; this can help explain the incomplete correlation between business size and worker anxiety. Indirect ownership could reduce the visibility of employer wealth. British miners resented their managers putting on airs. Metal workers in Chambon attacked their employers' new cars. But the only rebellions against the trappings of wealth came in older settings, in factory towns where the manufacturer's mansion was evident for all to see. Absentee ownership caused uneasiness and a sense of loss of control, and unimaginable profits were often evoked in strike meetings; but the most specific targets were now rather remote. Big business could produce more positive satisfactions. Workers could gain prestige from association with a large firm. Many took pride in saying they worked for Zeiss, for example, a 'world-famous firm.' Most large companies had a group of stable, older workers who felt they had shared in the

company's success. Big business meant better-designed factories. British metallurgical workers welcomed the spread of corporations because the factories were better-built. Even in engineering, the fact that German plants had better facilities than their British counterparts helps explain the lower level of German unrest. Nor, of course, was big business necessarily impersonal. Again, German firms may have gained an advantage by treating individual workers, at least on occasion, with a certain dignity. Many trained their officials to take an interest in the workers, whom they called 'Mr.' when they listened to complaints. Of course this system was incomplete — state concerns were reputed to have developed it much more fully than private — and it created its own tensions for those workers who preferred a closer personal touch. But there may be a significant contrast with Britain, where big business tended to sweat a local manager instead of providing him with an adequate staff, leaving him in personal contact with his workers but without the time to know much about them. In Ruhr mines the foreman-to-worker ratio actually increased under bigger business; foremen had 53 workers under their charge in the 1890s and only 49 in 1907, so that in a literal sense personal contacr increased.⁵¹

Certainly rationalization tended to reduce the personal authority of workers' direct supervisers, and it was an interesting test of reactions, to bureaucratization to see if this was perceived as a benefit. German machine plants began hiring engineers to set piece rates, so that 'Calculation Departments' replaced foremen in this operation. A number of complaints resulted. Workers criticized the stiffness of the system, where rates were set by people not actually on the job. Turners in the Borsig company struck in 1897, seeking individual worker-employer agreements on each piece instead of the Kalkultionsburo, and the movement spread to some other companies. But the result was interesting; initial employer resistance gave way, not to a return of personalized direction, but to a modification of the complaint procedure; worker-management committees could now hear grievances over the appropriateness of piece rates.⁵² One new structure thus supplemented another. More generally the limitation of personal initiative on the part of immediate supervisors found fairly wide favor. Krupp had 15 supervisors to make sure foremen actually paid the wage levels the company set, and to prevent foremen from forcing workers to do gardening or other service jobs for them. The piece rate itself was replaced in some British steel plants by divisional determination of pay rates, and the workers seemed to welcome this. The Kalkultionsburos generally reduced complaints about bad treatment

and arbitrary action in the German factories.⁵³

Workers did not of course write hymns to the virtues of big business. Satisfied workers rarely leave records at all, their existence being noted most commonly by self-interested outsiders, so it is risky to emphasize either their number or the extent of their satisfaction. Examinatio n of the kinds of workers most vocally hostile to their employers does suggest that personal tensions, more than impersonal ones, were most productive of grievances and that rationalization, intelligently carried through, could actually answer some of the workers' desire for fair standards of treatment. Tension was greatest, then, where big business grew without changing the chain of command, so that a foreman, still expected to serve as something of a friend and adviser, was turned into a driver of men. Incomplete rationalization, of the sort endemic ia the mines and many engineering firms as opposed to the really large units of employment, caused most disputes.

In fact the most direct clash between workers and employers had little to do with rationalization at all, though its implications may illuminate aspects of the worker reaction. Paternalism continued to characterize the approach many manufacturers took toward their labor force. It was less prevalent in England than on the Continent, and certainly far more powerful in heavy industry than in textiles; but its influence was widely felt. Paternalism involved a network of benefits to the workers. Silesian mines offered canteens and consumer co-operatives, company schools, libraries, and holiday camps, topping it all off with a free 'Arbeiterfreund' subscription and calendar and an annual picnic. French miners received pension programs (mutually contributory), medical service and drugs plus support when ill, and — again the personal touch, with political overtones - twelve francs for each son's first communion and a tip on the Sainte-Barbe holiday. The total cost in 1900 was over 200 francs per worker, a. full sixth of the annual money wage. Company housing continued to spread in the mines (even in parts of Britain) and the metallurgical centers. In 1883, 12 percent of all Ruhr miners were housed by the companies (and provided with gardens and stalls for pigs and goats); in 1900, despite the huge growth in the labor force, the figure was 21 percent. The metallurgical firms in French Lorraine provided reasonably comfortable housing for up to 65 percent of the labor force. Older paternalistic companies, like Le Creusot, added new benefits such as showers and refectories to their long list of programs.⁵⁴

The carrot could not conceal the stick, for the benefits were intended to create a docile, stable labor force. Not only housing but

also pension rights were lost if a worker left his company. At an extreme, the Lorraine metallurgical firms charged their workers so much in dues and fees for the company store that many were in debt to their own employers. Workers in many companies, even in factory towns in England, could not contemplate leaving their firms to defy the manufacturers, for they stood to lose too much. Merely a complaint could bring dismissal; eviction of troublemakers was frequent in British as well as Continental mines. And the paternalistic approach involved outright interference in workers' activities off the job. Some British engineering firms set rules against gambling and swearing. In French factory towns paternalistic employers typically served as mayors, where they could try to forbid political meetings or even ban the entry of union organizers.⁵⁵

Worker reaction in this situation was difficult. Even the basic outlook is hard to assess, for again there were obvious divisions. Some workers, for example, avoided company housing to preserve their liberty, but others welcomed the relief it provided their budget and did not complain. Workers themselves, in memoirs, reported that many of their colleagues judged the paternalistic employers good fellows and professed themselves satisfied. The most elaborate benefit programs reduced job-changing; Krupp, for example, suffered only 60 percent as much worker instability as other firms in the area. Is this a sign of active satisfaction or enserfment? It is impossible fully to decide. Clearly some workers did react less strongly than others, and a certain amount of positive choice was involved, whereby workers most eager for this kind of security would head for the most paternalistic firms. Other workers suffered their dependence keenly but felt powerless to complain, while still others found the means to protest.⁵⁶

Strikes by quarry workers in France and Belgium were often explosions of anger against the dictatorship of a company town. Glass workers struck frequently against traditional employer controls such as a system of payments in kind. Miners still more persistently defied paternalism. The Blanzy company at Montceau was unusually heavy-handed. The company supported a variety of recreational associations and churches as well as the usual benefit programs. It also watched over the workers' habits and organized the labor force on behalf of the employer's political campaigns. Workers were screened on hiring for their political reliability and then watched over by an elaborate secret police. But in 1899 rebellion came, in the form of a massive strike; and from that point onward workers maintained strong interest in union and socialist activities, while radically reducing their rate of attendance

at church.⁵⁷ Risings among metallurgical workers were far rarer. Workers at Le Creusot in 1899 struck for union recognition, as did metallurgists in Lorraine in 1905. Lorraine workers also attacked the system of company stores as a 'perpetual millstone' both for their budgets and for their freedom to organize. In each case the workers were defeated, many evicted and dismissed. But unlike the miners, who also frequently suffered defeat and retribution, their action had no aftermath; no subsequent unions formed, and there was no recurrence of the strike movement. Metallurgical employers thus were almost completely successful in maintaining the paternalistic stance, except in Britain where they did not attempt it seriously in the first place. In Germany as in France the unions were barely able to penetrate the metallurgical companies and never won the right to bargain collectively.⁵⁸

The one concession the metallurgical companies (and also mines in the Ruhr, after 1905) made to possible discontent was the proliferation, of joint committees to deal with grievances. Workers initially responded with considerable suspicion. They feared employer control; even elected delegates often refused to speak, because they were resigned to employer dominance. And they properly sensed that most of the committees would be given no jurisdiction over important matters. In Germany, for example, many were confined to discussion of the benefit programs and sanitary measures, only infrequently being consulted on actual work conditions. Hence elections to the committees rarely drew much interest at first; in the German mines, only 11 percent of the eligible voters came to the polls in 1905. But participation tended to mount; many workers began to ask for committees in German metal works by the mid-1890s. Initially fearful of turning in complaints, workers began to come forward more readily. Even the committees in Le Creusot, though hand-picked by the employer after the great strike, handled several hundred complaints a year over wages and conditions, more often than not ruling in the workers' favor.⁵⁹ Joint committees in a Belgian metallurgical factory reported on several dangerous pieces of equipment, a harsh foreman, the need for a second bassin of water to speed up the cooling of metals, and the desirability of a larger coffee heater in each lounge — prompting one worker to tell an investigator: 'Gradually we perceived we all had the same interest — the good functioning of the enterprise; there resulted a very marked improvement in our relations with management, which can even lead to a m ore complete collaboration in production.'⁶⁰

Clearly, workers were not ready for a complete rebellion against a

barely-modified paternalism. The greater size of metallurgical companies, enhancing their power to control, helps explain the relative docility of workers in the industry in comparison with miners. And many workers, blocked from unionization by employer repression, turned to socialist voting as their only protest outlet.⁶¹ Paternalism unquestionably caused great tension. Efforts to attack it failed repeatedly — hence a high rate of strike defeats in most industries in which paternalism prevailed, including German (but not French) mining; workers could only learn caution from this pattern. But the checkered response to paternalism was not simply the result of repression. Some workers could find more than material benefits from the paternalistic system, for it could correspond to the rather traditional standards of fair dealing and suggestions of some personal interest which they expected from employers. A minority in all the Continental countries joined yellow unions, intimidated by their employers but in some cases also convinced that class conflict was not the right approach. Far more workers, including many who were unionized, were dissuaded by the various aspects of paternalism from direct attack on the bases of employer control in the firms.

Paternalism unquestionably complicated the reaction to business rationalization. It was an old system, dating from the beginnings of industrialization. Growing worker uneasiness resulted less from changes in the system than from increased consciousness of rights. In this sense the ambiguous reaction to paternalism resembles the artisanal response to the traditional masters' controls, simply writ large. Business growth heightened the impact of paternalism. More workers were now grouped under a single paternalistic roof. Employer associations spread the paternalistic principles of control more widely. Manufacturers' organizations worked to break strikes and blacklist troublemakers — in Germany even using employment agencies to weed out dissidents — and while their effectiveness was limited they added to the difficulties of response.⁶² Most important, the persistence of paternalism showed that manufacturers' own attitudes were evolving far less quickly than the actual business structures.⁶³ That they still had the power to maintain this system of controls could not conceal the growing challenge. Innovations such as the joint committees responded in part to new worker demands and even more to the need to compensate for the growing impersonality of management, which paternalism had intended to avoid by its provision of favors and individual supervision. But change was less important than persistence, forcing those workers most concerned about their relationship with management to deal with the

general issue of power rather than the more specific effects of rationalization.

For most factory workers, the actual problems of relationship to management were largely personal. Paternalism may have encouraged this in the bigger plants, because of the stress on the employer's personal responsibility rather than problems of organization of cont rol. In industries such as textiles and even engineering, most companies remained small enough that the personal element could more legitimately seem predominant. Personal issues themselves reflected the changes in industrial organisation to a significant degree, but indirectly and only as filtered by the character of the people involved.

Foremen were the key to the worker's perception of management. General issues of company size or paternalism or even the character of the director or owner normally paled before the question of what the foreman was like. A good socialist in the Ruhr might realize that foremen were themselves exploited by capitalism, but far more justly blamed the individual right above them. And here, precisely because of the personalization of work supervision, a host of reactions was possible. Many workers judged the endurability of their job by the personal atmosphere surrounding them, and the foreman was a key here. A friendly greeting could make all the difference. For most factory workers foremen were the agents of hiring and firing. They set the pay rates and assigned the work, so that many workers believed, with considerable justification, that whether they earned well or not was entirely up to the foreman's assignments and his calculation of pay. In small companies foremen also kept records on the job, informing workers of pay rates only orally. And the worker depended to a considerable extent on the foreman's good opinion for his own judgement of the work, which is why he was so quick to react to insults as well as favoritism and demonstrable bad treatment.⁶⁴

Complaints against foremen vastly outweigh praise in the available records. But individual foremen were greatly admired. Not a few strikes resulted from the transfer or dismissal of a man regarded as fair in his dealings. Even in the larger companies elements of the traditional master-skilled worker relationship remained, so that one observer described the relationship as 'half cordial, half subordinate.'⁶⁵ For some skilled workers the foremen's role was almost a spin-off of their own direction of subordinates, which could still include hiring and determination of pay. The ambiguous position of the foreman was reflected in a number of efforts to keep him in the union. In 1905 nine unionized machine adjusters in a Paris automobile plant struck against a

harsh new foreman, asking initially for no supervision at all; a compromise was finally reached when the employer asked them to select one of their own number. In the British cotton industry foremen were almost invariably chosen from the union in the 1890s. And generally skilled workers wanted to be sure that their supervisor was well qualified. Some might not want the responsibility themselves - this was claimed in German textiles — but for others it was a sought-after chance for mobility and workers generally wanted to make sure the man knew his job and that he rose appropriately from the ranks. Correspondingly many foremen sided with workers against employers, for as a breed they remained uncertain of where their loyalties lay.⁶⁶

This close relationship was the basis for many a dispute. Steel smelters in Gravesend criticized their company in 1914 for bringing a foreman in from outside instead of promoting from the ranks. Urban workers might resent a rural-born supervisor. A foreman, once a colleague, could easily disappoint his fellows by acting high and mighty and siding consistently with the boss. If they served as strikebreakers it would be all the worse. Many were unreceptive to workers' complaints and suggestions, thus demeaning the whole job as well as making it more difficult. Favoritism was often claimed. French workers were particularly vulnerable here, because of the country's political and religious disputes; many felt discriminated against by Catholic foremen in the assignment of work. But the complaint was widespread, particularly in the mines. Insults and beatings were another sore point, for many foremen were harsh and overburdened.⁶⁷ A German mine strike broke out when a foreman berated a worker - 'You idiot, you pitiable scum'⁶⁸ — while in a metallurgical company in Saint-Etienne a series of strikes resulted when a new foreman 'dishonored' the workers by telling them they were too slow.⁶⁹ In the Belgian Hasard mining company workers struck against foremen chosen not from the best workers but according to their sycophancy or their kinship with employers. Women workers were often sensitive; shoe workers in Knighton Fields moved against a 'nagging' forewoman. And direct violence against hated foremen was not uncommon.⁷⁰ For, quite apart from the material aspects of the relationship, foremen could draw upon themselves the full resentment against dependence which many workers felt so keenly: 'I once gave the foreman the pleasure of giving, me notice. I'll never do it again; I never want to give such joy to any foreman again.'⁷¹

Rationalization unquestionably worsened relations between workers and foremen, without removing the highly variable personal factor. The

number of foremen increased, for supervision and record-keeping gained new importance. In industries such as textiles Germany and Belgium had long maintained a higher ratio of foremen than Britain and France, putting one in each department or each room of the factory and leaving workers little initiative, so the change probably had its greatest impact in the latter countries; but it was generally felt. Older British workers particularly resented supervision by interlopers who, they believed, did not really know the work. Foremen were put more clearly under the employer's authority, and were fired when they did not suit. British engineering firms began to insist that foremen not "be part of the workers' union. For their part, many foremen began to view themselves more distinctly as professionals, without the possibility of rising to their own business but with interests quite separate from those of the workers. Increasingly they adopted a distinctive, middle-class style of dress. In all probability fewer were recruited from the ranks or were sent to technical schools for training, which helped create a separate mentality. Rationalizing companies sought more highly edu cated supervisors or brought outsiders in from more advanced competitors. Several strikes in the shoe industry resulted from the introduction of a foreman with experience in engineering shops, for fear that he would force the pace. Above all, foremen were the agents responsible for raising productivity, as they were paid a bonus for output above a certain level. Here, most clearly, rationalization hit workers through personal intermediaries.⁷²

The whole development of rationalization in practice discouraged many factory workers from adopting appropriate counter-measures, which among other things accounts for the lack of correspondence b e-tween the size and sophistication of company organization and the level of formal protest. The personal element loomed too large, so that whether satisfied or not workers had little immediate sense of the changes in business structure that were in fact shaping their lives. Undoubtedly the rise of unionization and strikes reflected more impersonal management, but for the most part indirectly, as workers found it difficult to express grievances in their relationships with employers except on the personal level.

Where there was a craft tradition, as in British engineering, some unions tried to adopt the artisanal approach and set their own rules, to counteract remote management. The effort failed, for the companies were too big to control in this manner. Thus efforts to regulate job conditions by the British boilermakers' and engineers' groups foundered. More logical still was agitation for collective bargaining, as a

counterweight to a powerful management. This did mean the creation of yet another impersonal agency, and workers could easily feel this simply added to their problems. Many British railwaymen, for example, criticized the formal conciliation procedure developed after 1907 because it was far less effective than the older system of direct petitions to management. And the general rebellion against the rigidity and delays in grievance arrangements in Britain after 1910, on the railways and elsewhere, owed something to this sentiment, for attachment to personalized relationships ran deep. But the clearer difficulty with most collective bargaining efforts in factory industry was that they could not influence the actual impact of rationalization within the workplace.

Employers vigorously resisted collective bargaining in most cases, and certainly any intrusion into the determination of conditions on the job. A Lyons manufacturer put his case succinctly: 'There would be two powers in the factory instead of one.'⁷³ Even in Britain efforts to control the work of supervisors normally met with refusal and in some cases outright dismissal. The latter was the fate, for example, of Scottish miners who asked that checkweighers be appointed to review the piece payment determinations of the foremen.⁷⁴

On the Continent collective bargaining was rare in factory industry. Worker interest was present — union recognition figured prominently in the major French metallurgical strikes, for example — but it rarely bore fruit. In Germany by 1913 there were but three agreements in the mines, covering 82 workers, and only 1376 in metals and machine building (covering 207, 472 workers), the vast majority in the smaller shops. Less than 16,000 textile workers were covered, less than 6,000 chemical workers. The pattern is clear. The approach that worked quite well in the crafts had almost no impact in industry. Employer resistance was the most obvious cause, but the difficulty of developing worker unity over the question of power in industry played a role as well.⁷⁵ The situation was less bleak in France and Belgium. Collective bargaining existed in the coal mines and in many textile factories, though metallurgy was rigorously excluded. It could be widely hailed. Verviers workers saluted their first agreement: 'Ended, henceforward, the reign of the master's whim . .. Labor has just won a Parliament at Verviers, and the laws that will be decreed there will be applicable by and for everyone.'⁷⁶ But rarely did collective bargaining in factories include clear grievance procedures. A few individual mining companies in Belgium had arbitration groups, which removed the need to strike against foremen. The Graulhet leather factory established an elaborate procedure in 1910, by which unions and employers would discuss any

grievance case, during which foremen were prohibited from firing any workers involved while workers agreed not to strike. But here the craft tradition gave some basis for the approach. After the 1910 railroad strike French companies set up local joint committees, with worker delegates elected, to transmit grievances, which management hailed as a 'simple and rapid means for workers to express their desires.' Generally, however, collective bargaining meant periodic meetings to discuss wages and hours, not to deal with on-the-job complaints. This was the case in the famous Arras agreements in the French mining industry, for example, and in most textile plants where bargaining was accepted at all. Unions tried to take individual cases in hand, but there was no systematic means to do so. And in contrast to the artisanal situation, there were no union delegates on the spot. German printing shops, for example, were supervised by elected worker representatives (Vertrauensmanner) who dealt with any dispute over the execution of the bargaining agreement and were protected from dismissal if they ruled against the company. Rarely could factory workers claim anything remotely comparable, which meant that they were little better off, in terms of on-the-spot relations with management, than workers in the paternalistic firms who had only a tame committee for appeals. Small wonder that sweeping structural demands, within the company or industry, made little sense and that workers viewed even many tirade union efforts with suspicion.⁷⁷

The situation in Britain was obviously somewhat different. Stronger, older unions and a more pervasive defense of skills within the factory created a situation more analogous to that of the crafts. Fluctuations in union membership, which could mount to over a hundred percent annually in the industrial unions on the Continent, were less great. Formal demands for union recognition were much more common, and agreements covered hundreds of thousands of workers in mining, metallurgy, railroads, and textiles. Of 2,400,000 workers under agreements in 1910, 900,000 were miners, 500,000 railwaymen, 460,000 textiles workers, and 230,000 metal workers — a majority or near-majority in all the industries involved. Relatedly, it may be argued that English employers were somewhat milder than their Continental counterparts. Paternalism was less extensive and efforts t o form yellow-unions less widespread. A German worker living in England commented on the same contrast more personally, noting that English employers regarded workers as human beings, less as commodities. They had learned to accept strikes as a normal part of industrial life, so that retaliation, though not unknown, was less common. Worker initiative

on the job was encouraged, in contrast to Germany where a more rigid hierarchy was held responsible for preventing worker suggestions on the job.⁷⁸ The early experience of British employers, their liberal political culture, perhaps even their waning entrepreneurial zeal, created an atmosphere in which workers could be treated with greater equality, at least in comparison with the authoritarian approach of Continental manufacturers.⁷⁹ The possibility of a distinctive national culture should not be excluded, though it is already clear that it should be applied only to skilled workers; British employers treated unskilled dockers more harshly than was true on the Continent.⁸⁰ But a persuasive argument is still possible concerning the skilled group, which had long been of prime importance in British industry. Lower rates of job supervision as well as the prevalence of collective bargaining show that the argument has some validity. But rationalization was cutting into tolerance, and the very incidence of s-trikes over union recognition should warn against too benign an interpretation. British factory workers were being brought under stricter supervision, even if the style remained somewhat less bureaucratic than that of Germany or Belgium. New controls in many cases were meant to compensate for the country's reduced level of technological change. Hence, while granting the far higher rate of strikes involved, it is worth noting that their rate of success was no greater than those in France or Germany. And collective bargaining agreements, though u ndeniably important, usually did not deal with on-the-job control. The steel workers had a joint committee for disputes. Lancashire cotton workers had widely-traveling union agents to take care of individual grievances. But again, only craft workers had representatives on the spot, as in printing where the 'Chapel Father' served as a recognized spokesman. Other unions worked for similar arrangements, and the growing popularity of the shop stewards' movement right before World War I attested to the desire to flesh out the existing collective bargaining system. But for most British factory workers during most of the pre-war period, collective bargaining meant, as in Continental industry, periodic negotiations on wages and hours. British workers were little more protected against the impact of rationalization; foremen were no more under control.⁸¹

In this situation the symptoms of grievance against management were typically indirect and ineffective. In France, Belgium, and Germany agitation against individuals was the most immediate expression of hostility, and it proved difficult to bring this under control. In both Belgium and France the percentage of strikes asking that foremen, managers (or, more rarely, hated workers) be fired

tended to go up with time, while in Germany it remained stable, despite the fact that these were among the most difficult of all strikes to win. In France, for example, strikes for dismissals failed 52 percent of the time between 1900 and 1913, as against an overall failure rate of 42 percent. Admittedly the strikes were small on the average, and often brief; in Belgium, at least, they showed some tendency to grow smaller after 1900, so that the percentage of strikers involved was more distinctly lower than the percentage of strikes. But generally, apart from the official statistics, a large number of strikes that ultimately produced broader demands were launched because of a personal dispute, and there are few signs of change in this regard.⁸² Strikes strictly over dismissals, which alone can be measured with any precision, clustered in the industries where companies were large and the change in business structure most rapid — in metals above all, then in mining and the lighter factory industries, followed at some distance by the crafts. (See Table V).⁸³ Yet this was a primitive kind of strike precisely because of its dependence on a personal spark. It reflected the tensions rationalization caused, but it did not bring the process itself into basic question. Union leaders of all ideological stripes warred against the personal strike, save where the justification was unusually clear; they recognized how likely it was that these strikes would draw only a handful of workers and then fail. Yet, as the persistence of these strikes demonstrates, workers continued to use the personal attack as a means of expressing their anger. The form was primitive but the purpose was not: outside of the crafts, employer resistance had largely prevented any more structured statement of the kind of rage that arose commonly on the job. Correspondingly, in Britain, where collective bargaining had seemingly penetrated farther into factory industry, the wave of unrest that developed after 1909 in challenging the superficiality of these agreements brought the personal strike to a new level of sophistication. Particularly on the railroads, thousands of workers walked out several times to protect the rights of an individual against management. When a Newcastle driver, not due on duty for thirty hours, was demoted for drunkenness on his own time, 35 00 workers struck despite union opposition. In 1913 a guard refused a foreman's order to add three cars to a train, because he knew full well the train lacked the brakes to handle the load; again agitatLon developed, though this time union backing won a company reversal.⁸⁴ Even here, workers were not directly attacking the business structure, but through the personal focus they expressed much of their frustration.

Table V Strikes for Dismissals (or More General Personal Issues)

A. France

B. France—% over time

C. Belgium—% over time

D. Germany - % over time