Mechanics of the Middle Class: Work and Politics Among American Engineers [Reprint 2020 ed.] 9780520314825

Citation preview

Mechanics of the Middle Class

Mechanics of the Middle Class Work and Politics Among American Engineers

Robert Zussman

University of California Press Berkeley / Los Angeles / London

University of California Press Berkeley and Los Angeles, California University of California Press, Ltd. London, England © 1985 by The Regents of the University of California

Library of Congress Cataloging in Publication Data Zussman, Robert. Mechanics of the middle class. Bibliography: p. Includes index. 1. Engineers—United States. I. Title. TA157.Z87 1985 62O'.OO92'Z 84-8869 ISBN o- 5 20-0 5 1 3 1 -9 Printed in the United States of America 1

2

3

4

5

6

7

8

9

Contents

Acknowledgments

vii

1 . Engineers and the Middle Levels

i

2. Precision Metals and Contronics

17

3. What Engineers Do

33

4. What Engineers Know

59

5. The Division of Labor

75

6. Authority and Participation

ioz

7. Careers

124

8. Collective Organization

160

9. Work, Family, and Community

174

10. The Politics of Work and Residence

192

1 1 . A Working Middle Class

218

Appendix: The Interview Schedule

237

Notes

245

Index

265

Acknowledgments

The engineers discussed in this book are men comfortable in a mechanical environment, at ease with computers and in laboratories and in machine shops, perhaps happiest when tinkering with their cars or around their houses. Yet, they are also quick to acknowledge that the quality of both their work and their homes depends on the help of their colleagues, friends, and, most of all, their families. A sociologist can do no less. I would like, first, to thank the engineers and administrators at "Precision Metals" and "Contronics" who allowed me to interview and observe them. With good grace, they not only tolerated my often naive questions and my sometimes unsettling presence but in many cases were generous enough to invite me into their homes. In return, I can only hope that I have written about them fairly and honestly. A number of friends, colleagues, and teachers not only contributed to the substance of the book but also made the process of writing far more tolerable than it would have been without them. When the book was no more than an unwritten dissertation, Toby Ditz and Mark Martin helped me figure out what the enterprise was all about. Steve Crawford, David Halle, and Peter Whalley were working on similar projects, and conversations with them, although often heated (for reasons now hard to recall), invariably helped me understand better the issues I was trying to address. In addition to wise editorial advice, Magali Sarfatti Larson provided much needed encouragement at a critical moment. Jim Alt, Andy Beveridge, Dan Clawson, Mary Ann Clawson, Kai Erikson, Herb Gans, Wolf Heydebrand, Bob O'Gorman, Marcia Radosevich, A1 Reiss, David Rothman, and Sheila Rothman all read and commented on earlier drafts. Mary Renaud and Sylvia Stein of the University of California Press edited the manuscript with skill and vii

viii

Acknowledgments

thoroughness. Joanne Bodagliacco and Greg Boyle provided valuable research assistance. M y debts to Allan Silver are enormous. From my years as an undergraduate through my years as a graduate student, he showed me what sociology can be as well as what it is and taught me that a respect for ideas implies no disrespect for hands-on research. He not only initiated a larger comparative study of engineers from which this book derives but commented in great detail on every chapter and made funds available for my research (through Grant 74—0594 from the International Division of the Ford Foundation). Without his support—material, psychic, and intellectual—it would have been impossible to begin this book, let alone complete it. M y father, especially in the early stages of the research, explained much about business and engineering that I would not have otherwise understood. M y brother Mark brought his substantial abilities as an editor to bear on the entire manuscript. More important, my mother, my father, and my brother all managed to combine encouragement with an often tested patience. Finally, I would like to thank Naomi Gerstel. She discussed the research from its beginning to its end and read drafts of every chapter, many repeatedly and all with more enthusiasm and insight than I had any right to expect. If, when all is said and done, I exonerate those who saved me from errors of fact and judgment, it is not so much a sign of graciousness on my part as a sign of theirs in encouraging me to take final responsibility for my own work.

i . Engineers and the Middle Levels

Few strata have proven so persistently troublesome to sociological analysis as what I shall call loosely the "middle levels"—the accountants, technicians, officials, administrators, and middle managers of all sorts who work as salaried employees in industry and government. This book is about engineers, a social type distinctive to the middle levels, and is based in large part on interviews and observations in two companies, one in an older industry, the other in an industry technologically more advanced. It is a study of the engineers' work, their careers, and their political beliefs. More generally, it is a study of the labor process, social class, and citizenship. For many years—roughly from the mid-i930s through the mid1960s—sociological treatments of the middle levels were dominated by the concept of "professionalization." Drawing on the historical experience of doctors and lawyers, this tradition emphasized the development among the middle levels of extended professional training, codes of ethics, and professional associations. More important, this tradition viewed medicine and law as examples of occupational groups that stood outside the main lines of class conflict. Because of their long periods of training, their high ethical standards, and their powerful professional associations, medicine and law were taken to represent the triumph of disinterest over selfinterest, of expertise over amateurism, of universalism over particularism. By imagining the middle levels as professionals, or at least incipient professionals, this tradition also imagined a transformation of American society in which the pecuniary values of business would cease to dominate even at the workplace. 1 Yet, a treatment of the middle levels as professionals has never been convincing, for the middle levels are firmly embedded in a workplace and labor process that continues to be organized by 1

2

Mechanics of the Middle Class

the principles of capitalism. As C. Wright Mills and others have pointed out, the old middle class of small entrepreneurs and independent farmers, as well as the classic professions of medicine and law, occupied their middle position because they were located outside industry, apart from either management or labor. In contrast, the middle levels assume their position by virtue of a location within industrial production and insofar as they share the characteristics of both management and labor. 2 Indeed, the continuity of the middle levels with the industrial working class is perhaps stronger than with the independent professional. Neither the middle levels nor the industrial working class owns productive property; both are subject to the discipline of industrial organization; the life chances of both depend on a fate in a labor market; both are involved directly in production. Still, at least in the past, the middle levels have also enjoyed clear advantages over the industrial working class. In varying degrees, they have benefited from considerable wage differentials; they have not only been subject to industrial authority but, in many cases, have also exercised such authority; and, not least, entry into the middle levels has become the most frequent first step in careers leading to the uppermost reaches of corporate management. The tension between their character as labor and their character as management is the starting point for most contemporary discussions of the middle levels. This book shares that starting point but also attempts to move beyond it. T h e Middle Levels At least until recently, the middle levels in general and engineers in particular have been characterized by their easy adaptation to modern industrial organization. Unlike small shopkeepers and independent farmers, they are not victims of rationalized, bureaucratic enterprise but instead owe their very existence to it. Unlike entrepreneurs, they are neither heroic nor villainous representatives of economic individualism but "team players" and "organization men." And unlike industrial workers, they have not been seen as a potential base for opposition to the modern social order so much as acquiescent participants in it. Although the rise and fall of shopkeeper and farmer, entrepreneur and industrial worker—along with the social and political conflict of which they have been a

Engineers and the Middle Levels

3

part—are the very stuff of history, the middle levels have "slipped quietly into modern society." 3 But their silence should not be mistaken for insignificance. The middle levels in general and engineers in particular are a product of the differentiation of management from ownership, first noted by Thorstein Veblen in the early years of the twentieth century and subsequently celebrated as one of the decisive events of American industrial history.4 Attached to the industrial corporation by the promise of stable salaries and stable careers rather than by the proprietary interests of venture capital, the middle levels represent the end of capitalism's entrepreneurial stage and the beginning of its routinization. Charged with the day-to-day operations of the new multidepartment and multidivisional firm, they are both the creation and the chief carriers of the rationalized administrative procedures that characterize much of contemporary work. As managers, the middle levels have also been responsible for the discipline and motivation of the industrial work force. In place of largely autonomous craftsmen who worked at a pace of their own making and immigrant labor accustomed to the more natural rhythms of rural life, they have trained a work force in the stricter schedules of factory and office. Whether as frontline supervisors or from the greater distances of personnel and engineering departments, they have developed the full range of devices—from psychological testing and incentive payment to automated assembly—that constitute the apparatus of modern industrial control. But the middle levels are also themselves labor, albeit typically better paid than most, and a significant part of the total work force. Although any estimate of their number is necessarily rough, depending on who is counted, most would include in the middle levels approximately 25 percent of the contemporary labor force: Engineers and engineering technicians alone account for over two million employees. 5 On the one hand, because the middle levels have typically seen themselves as different from unionized labor, they have tended to modulate industrial conflict. On the other hand, because the middle levels have typically brought to their work higher expectations for autonomy and self-expression, they have tended to raise new issues of industrial integration, particularly over the firm's attempt to limit the very discretion that is often necessary for them to perform their jobs. Finally, engineering, perhaps more than any other occupation, is

4

Mechanics of the Middle

Class

associated with the rise of modern technology. If nineteenthcentury innovation was a haphazard affair—the work of independent, often scientifically untrained tinkerers—today it is more systematic. Beginning in the chemical and electrical industries of the early twentieth century and continuing in the most advanced of contemporary industries, the research department, staffed with university-trained scientists and engineers, has become an integral part of the production process. Innovation, especially technological innovation, has itself become routine. In all this, the middle levels— quiescent and glamorless—have brought to modern industrial life much of its distinctiveness. T h e F o r m a t i o n of an Occupation Like any complex, internally varied occupation, American engineering is woven together from diverse strands, drawn from different times and places. One strand extends down from the canal and railroad construction of the early nineteenth century, another from the machine shops of the mid-century, yet another from the landgrant colleges and science-based industries of the latter part of the century. Civil engineering—as a distinct, publicly recognized occupation—first emerged from the great internal improvement projects of Jacksonian America. A few of the earliest civil engineers were trained in Europe; a few more were trained at West Point; and a few more were trained in the private engineering academies that had begun to appear in the 1820s. Many of these school-trained engineers acted as consultants, following the practice in such better established professions as medicine and law. The school-trained engineers, however, were vastly outnumbered by others whose only training was experience; most of these latter engineers found salaried employment as resident engineers in charge of repairs and operations on canals and railroads. In origin, according to Calhoun, "this civil engineer was the creature of the organization in which he worked" and, in practice, he was " a respectable member of a bureaucracy." 6 American mechanical engineering emerged first from the machine and metalworking shops that produced the earliest locomotives and stationary steam engines. Owing little to either military

Engineers and the Middle Levels

5

engineering or civil engineering, these shops drew primarily on the skills of the millwright and the mechanic, whose apprenticeship the engineer-to-be often shared. Moreover, these shops generated what Calvert calls a "shop culture" that afforded an importance to entrepreneurial activity and respect for the "dignity of hand labor" notably lacking in civil engineering. 7 Yet, despite their distinct origins—and albeit for different reasons—civil and mechanical engineers shared a high social standing throughout most of the nineteenth century. In civil engineering, this standing was assured by a position in an organization: "the Engineer stood next to the Proprietor in a chain of occupations," according to Calhoun, and "by this proximity he was assimilated to the position of the Proprietor." 8 In mechanical engineering, high social standing was assured not only by the engineer's entrepreneurial role but also typically by "upper-class birth," which provided the "extensive social and business connections" necessary to fulfill that role. 9 By the late nineteenth century, however, the high standing of both branches of engineering was threatened by the rise of science-based industry and the concomitant rise of new, state-run engineering schools. In the electrical and chemical industries, companies such as General Electric, Westinghouse, DuPont, and Dow generated an unprecedented demand for technical manpower. Impatient with the longer process of apprenticeship, these companies turned to the graduates of the land-grant colleges, which had developed out of the Morrill Act of 1 8 6 2 and whose engineering enrollments had increased more than fortyfold between the 1870s and the beginning of World War I. 10 As a result, the number of practicing engineers in the United States increased from approximately 7000 in 1880 to well over 1 3 0 , 0 0 0 in 1 9 2 0 . These new engineers were different from the civil and mechanical engineers who had preceded them. In all likelihood, they came from families of lower social standing. But, more important, they followed a very different career path. As Calvert concludes, the college-trained engineers were less likely "to enter situations in which they themselves would be entrepreneurs and were more likely to enter relatively large, bureaucratic corporations where . . . they bore no immediate interest in the profit and loss of the organization." 1 1 Filling staff positions—as research engineers, sales engi-

6

Mechanics of the Middle

Class

neers, or even draftsmen—they no longer "stood next to the Proprietor in a chain of occupations." Engineering had become a mass occupation. The increase in scale and the altered conditions of practice created a crisis in engineering. What had once been an occupation secure in its social standing now manifested symptoms of almost "obsessive concern with social status," including not least a series of attempts to improve that status. 12 If only briefly, in the early decades of the twentieth century, it appeared that engineers might become an independent force in both industry and politics. The engineers' ambitions were largely expressed in attempts at internal reform of technical societies. The American Society of Civil Engineers had been founded in 1 8 6 7 , the American Institute of Mechanical and Metallurgical Engineering in 1 8 7 1 , the American Society of Mechanical Engineers in 1880, and the American Institute of Electrical Engineers in 1884. Before 1900, none of these societies had been particularly concerned with either public policy or the conditions of engineering practice. None of the societies (with the partial exception of ASCE) had even attempted to exclude from membership business people interested in the field but lacking in technical qualifications. However, between 1900 and 1 9 2 0 , all the societies restricted their criteria for membership by emphasizing "professional" standards, and all but the A I M E introduced codes of ethics. So, too, many of the societies began to publish papers on public policy issues such as conservation and industrial efficiency. Finally, the newly formed Federation of Associated Engineering Societies published a report placing most of the blame for industrial waste on management and recommending a shift from the twelvehour to the eight-hour day in many industries. 13 Nowhere, though, were engineering ambitions more explicit than in the work of Frederick Winslow Taylor, once president of the A S M E and the leader of the movement for "scientific management." To be sure, some critics have seen scientific management as an instrument of capitalist domination, essentially antilabor in its attempt to maximize profits by imposing a detailed division of labor over previously autonomous craftsmen. But Taylor himself saw scientific management as part of a comprehensive plan for the reorganization of industry, with control vested in a firm's planning department headed by technically trained personnel rather than by entrepreneurs or owners. Indeed, Taylor was often as critical of

Engineers and the Middle Levels

7

"the indifference of the employers and their ignorance as to the proper system of management" as he was of labor. In this sense, Taylorism was very much an ideology of the middle levels, a claim to a position in industry that would not only be independent of capital and labor, but dominant over them.14 Scientific management promised to bring efficiency to public administration, and this was a promise well suited to the Progressive temperament. One of Taylor's disciples, Morris Cooke, served as director of public works in Philadelphia, where he became a staunch critic of municipal utilities companies, which he saw as a principal source of public corruption. Another, Henry Gantt, helped form the New Machine, an association of engineers concerned with the acquisition of "political as well as economic power." 15 In the context of these developments, Thorstein Veblen wrote the series of essays later collected as The Engineers and the Price System. Impressed by scientific management, Veblen accused the absentee owners of "sabotage"—the conscientious withdrawal of efficiency in the name of profit maximization—and contrasted them with the "technological specialists whose constant supervision is indispensable to the due working of the industrial system, . . . whose work it is to control the strategy of production at large and to keep an oversight of the tactics of production in detail." These technologists, Veblen thought, had become "uneasily 'class conscious,' . . . beginning to take stock of that all-pervading mismanagement of industry that is inseparable from its control for commercial ends." Thus, he continued, although without much optimism, "any question of a revolutionary overturn . . . resolves itself in practical fact into a question of what the guild of technicians will do.'" 6 By the mid-i9zos, however, the engineers' efforts to organize themselves as an independent force had been disappointed. The Federation of Associated Engineering Societies, which had taken the lead in attempts to unify engineering as a profession, disbanded in 1923. It was replaced by the American Engineering Council, which was, according to Layton, "dominated by business interests who exercised an effective veto over its affairs." 17 Although scientific management became standard practice in many industries, it did so as discrete techniques and tactics, not as the comprehensive plan that Taylor had commended. The influence of engineers on

8

Mechanics of the Middle

Class

public policy, never strong to begin with, devolved into "technocracy," a fringe movement that persisted into the 1930s, although with little support from engineers. Even Veblen, retreating from his earlier position, now characterized engineers as a "harmless and docile sort." 1 8 The crisis in engineering had been resolved by an apparently stable integration of the technologist into the business enterprise and by his acceptance of the dominant business ethos. To be sure, engineers in the twentieth century would not be entrepreneurs: Calvert estimated that no more than 1 0 percent of the engineers who graduated from one eastern technical school in 1904 would become proprietors or partners of their own firms. 19 But neither would engineers be independent consultants, organized as a "free" profession like doctors or lawyers; indeed, because they were products of the large-scale business organizations that employed them, they could not be. Within these organizations, though, engineers did have careers leading, if not toward proprietorship, at least toward management and business responsibilities. According to the influential 1 9 3 0 Wickenden Report, sponsored by the Society for the Promotion of Engineering Education, it had become "almost always necessary for an engineer to leave the engineering of materials and enter the engineering of men to become very successful financially and socially." 20 N o w thoroughly dependent on the organization for his social standing, the engineer repaid it with his loyalty. Engineers in A d v a n c e d Industry The attachment of engineers to the business ethos as it developed in the 192.0s was strong, but neither complete nor final. The formation of the National Society of Professional Engineers in 1 9 3 4 , intent primarily on passing registration laws for engineers, revived the attempt at professionalization along the lines of medicine and law. A burst of unionization in the 1940s, although affecting only a small minority of engineers, raised the possibility of an alliance with the labor movement. Most important, the transformation of the conditions of engineering practice, begun in the late nineteenth century, continued in the years after World War II. As the electrical and chemical industries had led the way in the early twentieth century, electronics and aeronautics have led the way since World War II. In these industries, the institutionalization

Engineers and the Middle Levels

9

of research and development, begun in the early years of the century, has become pervasive. As late as 1 9 5 5 , industrial funding for research and development was no more than $2.5 billion; by 1982., it had risen to $38.5 billion, an increase of over 450 percent in constant dollars and over 1 5 0 0 percent in current dollars. 21 In the second half of the twentieth century, as Braverman observes, "the scientific-technical revolution . . . cannot be understood in terms of specific innovations—as in the case of the Industrial Revolution, which may be adequately characterized by a handful of key inventions—but must be understood rather in its totality as a mode of production into which science and exhaustive engineering investigations have been integrated as a part of ordinary functioning." 22 Advanced industries require not only more technical knowledge, but also a different type of technical knowledge. As Bell argues, what has become decisive is "the centrality of theoretical knowledge—the primacy of theory over empiricism and the codification of knowledge into abstract systems of symbols that, as in any axiomatic system, can be used to illuminate many different and varied areas of experience." 23 In this, industry has become more dependent on the university and engineering more dependent on science. The number of engineers has consequently continued to grow, from just under three hundred thousand in 1940 to over one and a half million, over three-quarters of whom were employed in business or industry, in 1 9 8 1 . At the same time, the character of engineering has changed, with a considerably greater emphasis on graduate training: The number of doctorates conferred in engineering, never higher than 1 2 2 in any year before World War II, peaked at over 3500 in 1 9 7 2 and has remained well over 2000 annually since.24 So, too, engineers have been joined in industry by increasing numbers of natural scientists, computer specialists, and engineering technicians, with whom they often work on the complex, "highly engineered" projects characteristic of the advanced industries. These developments have not as yet inspired the same degree of self-reflection and political activism that characterized engineering in the early years of the twentieth century, but they have stirred many speculations about what engineers might think and might do and where they are located in the class structure. To mention only a few: Galbraith, drawing on one strand of Veblen's thought, argues that engineers are leaders of a "technostructure" that has as-

io

Mechanics of the Middle Class

sumed effective control of industrial policy; Mallet and Gorz, drawing on a different strand of Veblen, argue that engineers are in the vanguard of a "new working class" whose "love of workmanship [is] incompatible with capitalist profitability"; Gouldner argues that engineers are part of a "New Class"; Poulantzas contends that they are part of a "New Petty Bourgeoisie"; Barbara and John Ehrenreich include them in a "Professional-Managerial Class." 25 All these speculations have merit, but two broad conceptions have dominated recent discussions: one, that engineers will, at last, "professionalize"; the other, that engineers will be subject to a process of "proletarianization." Professionalization Professionalization in engineering—or in the middle-level "bureaucratic professions" more generally—should not be confused with professionalization as it took place in medicine and law. The "free professions," having successfully linked the right to practice with graduation from a certified training program, are characterized by monopolies over the market for their services. As a result, doctors and lawyers are protected from competition from practitioners in related or emerging fields. The bureaucratic professions enjoy no equivalent protection. Although registration laws exist for many of the bureaucratic professions, they are typically less restrictive than those for medicine and law. In engineering, experience may be substituted for certification by degree, and registration itself is required only of the principal officer of a corporation rather than of all employees of that corporation. For the most part, corporations are at least formally free to hire whomever they like for engineering positions. As a result, neither engineering nor any of the other bureaucratic professions has enjoyed the collective control over entry to practice characteristic of the free professions.26 Engineers, however—like accountants, teachers, or social workers—may still find the free professions an attractive model. Even in the absence of a monopoly over the market for their services, they may lay claim to professional standing in order to distinguish themselves from undifferentiated labor and to share in the generally high prestige reflected from medicine and law. In this sense, professionalization in the bureaucratic professions should be understood, in large part, as ideology. Although professionalism among the middle levels may origi-

Engineers and the Middle Levels

ii

nate in ideology, it is distinctively anti-ideological in rhetoric. The "matter of factness" of engineers, as well as the calculation of accountants, the disinterestedness of teachers, and the objectivity of journalists, represent a claim to the preeminence of reason over self-interest and of universalism over particularism. In this sense, the middle levels equate professionalism with the possession of specialized skills and what Larson calls "monopolies of competence." 27 Although education is not as definitive of position in the bureaucratic professions as in the free professions, many middle-level employees are graduates of specialized programs. This education is not only the basis of the middle levels' claims to professional standing, but also a source of differentiation among them. As Freidson argues: T h e sociological rather than the merely technical or economic significance of a long period of training in putatively complex and abstract skills . . . lies in its tendency to develop institutional commitments on the part of those trained. Such trained workers are inclined to identify with their skill and with their fellows with the same training and skill. 28

Extended training in specialized areas thus produces "not merely general skill-class, or mass solidarity, as is sometimes the case with industrial workers in trade unions, but disciplinary or occupational solidarity." 29 Occupational and professional solidarities among the middle levels are also encouraged by the organization of career lines. Insofar as the skills of engineers, accountants, teachers, and social workers are based on formal education rather than on-the-job training, they are easily transferable from organization to organization. Unlike the industrial working class, whose jobs have "no social or economic foundation for their persistence beyond the plants, agencies, or firms in which they exist," bureaucratic professions can "realistically envisage a career over most of their working years . . . during which they maintain a particular occupational identity and continue to practice the same skills no matter what institution they work in." 30 Such occupational identities may make for tensions with industrial authority—between the "cosmopolitanism" of an occupation practiced in many settings and the "localism" of employers concerned only with their own firms. 31 Or they may occur between

12.

Mechanics

of the Middle

Class

employers attempting to rationalize procedures and the members of an occupation protecting their prerogatives against such rationalization. Moreover, according to some, such tensions are particularly likely to occur in the advanced industries where "what counts is not raw muscle power or energy, but information." 32 There, according to Bell, among "the scientists, the mathematicians, the economists, and the engineers of the new intellectual technology," in the "norms of professionalism" formed by schools and occupational communities, we may find nothing less than a "departure from the hitherto prevailing norms of economic self-interest which have guided a business civilization."33 Proletarianization If speculation about the professionalization of engineering is modeled only loosely on the historical experience of medicine and law, speculation about proletarianization is modeled more closely on the historical experience of the crafts. In this, the indispensable starting point is Harry Braverman's Labor and Monopoly Capital. According to Braverman and the new labor historians, craftsmen retained their traditional skills in the early stages of industrial capitalism. "Spinners, weavers, glaziers, potters, blacksmiths, tinsmiths, locksmiths, joiners, millers, bakers, etc. continue to exercise in the employ of the capitalist the productive crafts they had carried on as guild journeymen and independent artisans."34 These skilled craftsmen did, of course, receive fixed wages for a fixed quantity of production. But whether hired as domestic laborers in a puttingout system, as subcontractors, or even as employees within the factory, they also retained considerable autonomy, with control over their own schedules, methods, system of training (in the form of traditional apprenticeships), and even helpers (often, as in the textile industries, including their wives and children). The proletarianization of skilled crafts began with the replacement of this autonomy by managerial control. The imposition of such control involved the discontinuation of both the putting-out system and subcontracting, the concentration of craftsmen within the confines of the factory, and the enforcement of regular hours of work. Most important was the development of a fixed division of labor, which, by vesting management with responsibility for coordination, divested the craftsmen of knowledge about and responsibility for the overall production process.

Engineers and the Middle Levels

13

The extension of managerial control in the early decades of the twentieth century was realized in part by the widespread adoption of Taylor's program of scientific management. By encouraging analysis and reorganization of the labor process, Taylorism extended managerial control from the simple coordination of already existing skills to the very definition of those skills. It encouraged the destruction of traditional apprenticeships, replacing them with shorter periods of training in limited and specialized techniques. Most important, it encouraged the separation of conception from execution. The attack on craft skills was continued with the introduction of fixed-cycle machinery, which routinized machine tending. Fixedcycle machinery removed responsibility for variations in the speed and quality of production from operators, vesting it instead in those technical departments that design and develop the machinery. By homogenizing skill, it reduced craftsmen to the level of general labor. Engineers played an important part in the proletarianization of the crafts, both as scientific managers and as designers of the mechanical environment of manual work. At first, these developments increased engineers' autonomy by providing a basis for their authority independent of ownership. But in the advanced industries, according to some observers, engineers themselves have become subject to a process of proletarianization similar to that they imposed on the crafts. In the advanced industries, the increase in scale of technical tasks has raised the possibility that engineering, too, may be subdivided and rationalized. The complexity of designs for products with multiple components, or even for entire "systems," has encouraged the imposition of managerial coordination similar to that which limited the autonomy of skilled craftsmen. 35 The "de-skilling" of the middle levels does not, however, proceed exactly as it did among the crafts; the introduction of new machinery has as yet had fewer consequences for salaried employees than it did for industrial workers. Moreover, as Larson points out, threats to the labor market position of middle-level employees often arise as much from the expansion of educational programs and the consequent oversupply of practitioners as from the reorganization of the labor process itself. 36 But, whatever the origin of the threats to the middle levels' market position, the results, according to some, are largely the same.

14

Mechanics of the Middle Class

Like craftwork, the work of the middle levels now tends toward an increasingly rigid division of labor, toward a routinization of previously high-level tasks, and toward an intensification parallel to speedups in manufacturing. According to Aronowitz, The labor of the engineer or scientist who functions as a technician within advanced industry rarely corresponds to the extent of knowledge acquired in college. He has become the new frontline production worker. Typically, his work is so rationalized that its routine character is preponderant, although fragments of creative judgment may be required from time to time. Among those professional workers engaged in the production of goods, the designations "chemist," "metallurgist," and "engineer" have become labels that describe not the old middleclass artisan who was engaged in scientific invention and technological innovation, but a worker whose tasks are often as routine as a machine operator's.37

Proletarianization refers in the first place to the polarization of class structure that may accompany the decline of an independent petty bourgeoisie and the rise of wage and salary labor. However, as Braverman points out, it does not imply that the middle levels are "working class" simply because they do "not own or otherwise have proprietary access to the means of labor and must sell their labor power to those who do." Rather, as Braverman continues, "in the present situation, when almost all of the population has been placed in this situation so that the definition encompasses occupational strata of the most diverse kind, it is not the bare definition that is important, but its application." 38 In short, it is proletarianization in its broader sense—as it refers to a relationship with the means of production—that provides the starting point for speculation about how the middle levels may be following the route of the craftsmen. But it is proletarianization in its more specific sense—as it refers to the reorganization of the labor process—that constitutes the critical test of those speculations.

A New Approach I am, however, skeptical of any dramatic change in the social location or political outlooks of engineers. Speculation, when theoretically sophisticated, is invaluable in raising analytic issues. But, in the case of engineers and the middle levels more generally, spec-

Engineers and the Middle Levels

15

ulation has gone on without a sufficient grounding in empirical research. There are many excellent studies of the relationship between work and politics in American life, but far fewer about white-collar workers than about blue-collar ones and even fewer about the upper reaches of white-collar employment that constitute the middle levels. At the same time, a number of older studies provide important insights into engineering work, but without a corresponding concern with the ideology of engineering. There are also some recent analyses of ideology among engineers, but without a corresponding basis in the detailed investigation of their work. 3 9 In the case of engineers, then, the call for additional research is more than ritualistic. This book, however, does not simply adjudicate among competing theories, for, in certain respects, neither professionalization nor proletarianization provides a conceptual framework sufficiently rich to understand the middle levels. First, whether formulated in terms of professionalization or proletarianization, most recent discussions of the middle levels have tended toward morphology, concerned with classifying them as part of one stratum or another within abstract and formal schemes. Often the development of categories becomes an end in itself: What are the criteria of a profession and does one group or another match these standards? How can we better draw "class maps"? Where do we find the "boundaries" of the middle levels? 40 Too often these categories are imposed on the middle levels; too rarely is there consideration of how the men and women of the middle levels construct their own categories and their own understandings of social structure. In this book, I discuss those understandings as well as the social structure they both emerge out of and help shape. Second, this book is more concerned with careers than are most recent discussions of the middle levels. To be sure, other discussions have said much about the structure of labor markets, but they have said far less about the routes by which individual men and women pass through such structures. Yet, a good deal of research has accumulated demonstrating that the behavior of men and women at work depends as much on background and future aspirations as on present position. Whether they resist authority or reject it, identify with a position or disavow it, look to collective solutions or to individual adaptations depends as much on personal work histories as on workplace organization. This is not to suggest replacing the

16

Mechanics of the Middle

Class

analysis of social structure with the compilation of biography, but to link them: The concept of the career provides that link. 41 Third, most recent discussions of the middle levels have focused almost exclusively on the relations of production at the point of production—on the workplace. As Gagliani puts it in his review of recent controversies: " A t the heart of the problem lies the division of labor." 42 But does it? If we are interested in the middle levels as a class or stratum in the fullest sense—as an actual or potential political actor as well as a structural location in the division of labor—we must recognize that classes and strata are phenomena of national societies, shaped by distinct cultural and political traditions. In the United States, there is a particularly strong tradition of citizen participation, based in the residential community, independent of membership in a class or occupation, and (aside from brief and unsuccessful attempts to establish "industrial democracy") absent from the workplace. The very concept of citizenship, implying a direct and individual relationship to the state, has introduced an egalitarian strand into American life, cutting across conventional distinctions between the middle levels and the working class.43 At the same time, conflicts over the rights and obligations of citizenship have been frequent in American history—over the inclusion of racial and ethnic minorities, the obligations of military service and taxation, the right to education and entitlements. These conflicts, as well as those emerging from the workplace, are critical to understanding the loyalties and sympathies of the middle levels. The empirical core of this book consists of interviews and intensive observation of engineers in two companies, one in an older industry (metalworking), the other in an advanced industry (consumer electronics). A comparison of engineers at the two companies allows us to address directly some of the most important arguments about the transformation of technical work and technical workers.

2. Precision Metals and Contronics

Whether we are interested in the dynamics of proletarianization or the possibilities of professionalism, we must ask what engineers do and under what conditions they do it. More specifically, we must ask how what they do in advanced industries differs from what they did in older industries and whether or not these differences affect the way they think and act. M y strategy for answering these questions was to study engineers at two companies, one in the electronics industry and the other in metalworking. Research Strategy Arguments about the impact of advanced industry on engineers link two levels of analysis. On one level, references to such matters as changing requirements for skill and emerging patterns of control or coordination speak to global characteristics of organizations. On another level, these characteristics of firms are taken as consequential for the orientations or ideologies of individual engineers. Each level of analysis requires its own method of data gathering. Characteristics of firms have been studied with archival records, industrial surveys, and case studies based on observation and interviews with strategic informants. However, the analysis of ideologies has tended to draw on either national sample surveys or depth interviews. 1 This study draws on methods appropriate to both levels of analysis, combining the detailed observation of organizations with intensive interviews of individuals within those organizations. However, some other studies of technical workers based on essentially similar strategies, particularly Mallet's impressionistic study of the "new working class" in a French chemical company and Low17

18

Mechanics of the Middle Class

Beer's more systematic study of technicians in the Italian electronics industry, have been handicapped by their failure to include matching studies of sites outside of "knowledge-intensive" industries. 2 Consequently, a case approach to studying the impact of advanced industry on engineers must include not only a case study of engineers in an advanced industrial setting but also a matched case of engineers in an older industrial setting. This research strategy—combining fieldwork at an organization with interviews of engineers in that organization in both advanced and older industrial settings—offers a number of advantages. M o s t important, the inclusion of two sites rather than one, varying along a conceptually significant dimension but matched as closely as possible in other respects, makes for a strong test of arguments about the impact of advanced industrial production on characteristics of both organizations and engineers. T h e inclusion of both fieldwork and interviews within a limited number of sites enriches the interpretive significance of both. Observations made in the field are an important point of departure for in-depth probes made during interviews as well as an occasional counterpoint to what the respondents say. T h e interview provides both an opportunity to probe in more detail about field observations and, in some cases, a source of information about the distribution of those observations throughout an organization. 3 But such a design also involves limitations. Although the number of individual cases may be large enough to permit quantitative comparisons, the number of industrial settings is limited to two. Consequently, statements about the impact of advanced industry on such global characteristics of firms as patterns of control and coordination cannot be made with as much assurance as statements about the impact of those patterns on individual engineers. M o r e over, the samples of engineers are drawn from organizations within particular sectors rather than from the national population of engineers. Although the strength of the design is the comprehensiveness of the comparisons it allows between a firm in an older industry and one in a more advanced industry, there is a complementary weakness in the ability to extrapolate from those comparisons to national distributions. We must also avoid identifying the concrete research sites with the concept of advanced industrial society in general, a concept that characterizes aspects of national societies, many of which are not

Precision Metals and Contronics

19

amenable to analysis at the level of firms or production sites. For example, the research design does not address such issues associated with the emergence of advanced industrial societies as the role of science in national politics or the sources of expansion in higher education. Many of the effects imputed to advanced industries, even at the level of the firm, may well permeate sectors of industry adjacent or even peripheral to advanced industry, thus blurring the comparison between old industry and new.4 What is lost in scope is compensated for in the precision with which the design addresses a more limited number of issues. Many of the arguments concerning both the situation of the middle levels in general and of engineers in particular refer to the proximate effects of advanced industry, at the level of the firm, on the organization of work and the values and orientations associated with that work. This particular research design is most sensitive to propositions of this type. Old Industry, Advanced Industry In large measure, the success of the research design depends on the selection of sites embodying the most significant variations between older and more advanced industrial production. Interindustry comparison as a strategy for the study of social change is perhaps most familiar to sociologists from the work of Blauner and Woodward. 5 Both Blauner, in his study of alienation among bluecollar workers, and Woodward, in her study of organizational characteristics, distinguish among industries on the basis of production technologies. They rank these technologies "in order of chronological development and technical complexity," distinguishing among craft, machine tending, mass assembly, and continuous process. 6 But such a procedure, and these categories in particular, present a number of problems for our purposes. First, there are internal inconsistencies in the use of an industrial classification ranging from craft technology to continuous process technology as a means of specifying old and advanced industry. The identification of continuous process with advanced industry is seriously misleading. Although continuous process technologies are surely distinctive to modern industry, they are by no means the only production technologies to be found in settings that are, by more intuitive criteria, clearly representative of those industries whose

20

Mechanics of the Middle

Class

greatest growth has come since World War II. In fact, continuous process technologies are suited only to those liquid products, such as some chemicals, that literally and physically flow.7 By contrast, advanced production technologies applied to solid products typically result in mass assembly production. At the same time, aeronautics and electronics, both appropriate examples of distinctively modern industries, are often characterized by unit and small batch production of technologically complex products designed and assembled on special order. Thus, continuous process technologies or high levels of mechanization might disqualify a company from consideration as an older industry, but no particular production technology is an essential criterion for the selection of an advanced industry. Second, although production technologies directly shape the work of the men and women who operate the machines and equipment constituting those technologies, their impact on the work of the engineers who design the machinery is less direct, shaping the specific content of their work but not necessarily its organization. 8 In this context, what Perrow calls "knowledge technology"—"the characteristics of knowledge used in the work flow"—is more important than production technology.9 Here, the critical distinction is between those industries in which scientific and technical knowledge is applied routinely and systematically to the core problems of production and those to which such knowledge is incidental. In more operational terms, this is a distinction between industries in which research and development is a regular part of the product cycle and industries in which it is not. And, most important, this distinction plays a prominent causal role in conceptions of both proletarianization and professionalization. Third, we should not imagine that in comparing industries we are comparing only technologies. Particular industries experience spurts of organization and growth at different times: agriculture and construction before the nineteenth century; textiles and woodworking in the early nineteenth century; metal making and metalworking in the middle and late nineteenth century; electronics, aeronautics, and chemicals in the twentieth century. As Stinchcombe argues, the organizational features that characterize these industries are conditioned not only by the technologies they use but also by the resources available at the time of their formation: by the availability of capital, state support for research and develop-

Precision Metals and Contronics

zi

ment, and particular types of labor. Moreover, industries tend to retain, by force of tradition and through the vesting of interests, the features of their formative period. 10 Consequently, age of industry is itself a conceptually rich criterion for distinguishing between older and advanced industry, capturing an archeology, as it were, of industrial organization. I settled on two primary criteria for the selection of sites: age of industry and knowledge technology. In addition, I eliminated industries, such as textiles and woodworking, dating to the very earliest stages of industrial production because they lack sufficient density of engineering staff for the purposes of my research. These criteria constrained the choice to either metal making or metalworking as representative of old (or mature) industrial production and to aeronautics, electronics, or chemicals as representative of advanced industry. In addition, the particular firms or division of firms chosen as sites should themselves exhibit characteristics typical of the industries of which they are a part: In the case of age of industry, by a date of foundation similar to that of their industry as a whole; in the case of knowledge technology, by a research and development effort roughly equivalent to their industry as a whole. Not only must the two sites differ along dimensions sensitive to arguments about advanced industry, but they must also be matched as closely as possible in other respects. Otherwise, we could not confidently ascribe observed differences in how engineers think and act to differences between old industry and advanced industry. Although difficulties of access required a relaxation of selection criteria in this regard, the initial search for appropriate sites began with three criteria of comparability, (i) Research suggests that company size is related significantly to organizational characteristics and patterns of labor relations. 11 Because there was no reason to assume that larger or smaller size is distinctive to either old or advanced industry, I attempted to find two sites of roughly comparable size. (2.) Particular regions can be identified as distinctively advanced industrial (for example, Route 1 2 8 around Boston and "Silicon Valley" just south of San Francisco) and others as distinctively old industrial (for example, western Pennsylvania). Consequently, there was some temptation to look for sites within those regions, thus expanding the scope of issues addressed by the research. However, this strategy would risk confounding the effects of work experiences with regional variations that cannot be attrib-

2z

Mechanics of the Middle Class

uted to the direct effect of differences between old and advanced industry. In the United States, where regional differences are particularly strong, this consideration is decisive for a strategy in which sites are matched as closely as possible by region. (3) The advanced industries are characterized by considerably greater reliance on the state as a major consumer than is the case for older industries. Moreover, at least a few observers have suggested that this relationship to the state is consequential for the organization and orientations of the work force. 1 2 These are important considerations. If resources had permitted an extension of the research to an additional industrial site, one appealing strategy would have been to include a company clearly dependent on the state—for example, an aeronautics company with large defense contracts. But to maintain the precision of comparison between sites, and given both a limitation of resources and the difficulties of access (compounded in the case of companies engaged in defense work), I decided to limit the selection of sites to companies, or divisions of companies, with customers primarily in the private sector.

T h e R e s e a r c h Sites Precision Metals The company chosen to represent old industry—pseudonymously called "Precision M e t a l s " — w a s founded immediately after the Civil War and is part of the metalworking industry. At the time of my research, Precision Metals had recently been acquired by one of the hundred largest manufacturing corporations in the United States. However, the company retained an almost entirely autonomous organization, setting its own production and personnel policies. Aside from an occasional truck parked in a loading area and bearing the name of its parent company, there was little to suggest that Precision Metals, with its twelve thousand employees worldwide, was not an entirely independent company of medium size. Precision Metals includes a number of divisions, each specializing in a separate product line. M y research concentrated on the largest of those, Division A , responsible for the design, sale, and manufacture of a line of small metal parts that Precision Metals developed and began manufacturing during the 1 9 2 0 s . This line was an improvement on a similar one that Precision Metals and a

Precision Metals and Contronics

number of other companies had been manufacturing since the nineteenth century. The parts, which vary in size, shape, and materials, are produced in both a commercially available standard line and on special order to industrial customers for use in a wide range of mechanical products. Although particular steps and sequences of production depend on specifications of material and design, in most cases, that production involves a variety of metal-treating, machine tool, and assembly processes. The engineering staff of Division A, one of its manufacturing facilities, and company headquarters are all located in an old New England town of just over thirty thousand people. Division A also includes three manufacturing facilities in southern states and shares three manufacturing facilities in Europe with other divisions of the company. Unless otherwise specified, Precision Metals will refer only to Division A. The engineering functions are divided among four departments. Two of them, including the largest number of engineers at the New England facility (forty-one), are responsible for product design. The smaller of these departments handles the standard line of products, modifying and updating specifications for the items listed in the company's hundred-page catalogue, as well as a variety of special projects on new products requiring special manufacturing techniques. The other product design department handles special orders, working out rough designs with potential industrial customers and then, internally, more detailed specifications of designs, materials, and manufacturing processes. Both product design departments also monitor the first few cycles of the manufacturing process of those products for which they are responsible. A department of mechanical design (with fourteen engineers) is responsible for the actual equipment, mostly machine tools, used in the manufacturing process. Occasionally, the department designs new machinery; more often, it modifies machinery purchased from outside suppliers to the specific needs of Precision Metals. A manufacturing department (including approximately thirty-five engineers divided among the four U.S. plants, with thirteen in New England) is responsible for "getting the product out the door." Unlike the engineers in other departments, the engineers in the manufacturing department have, or are likely to have at some point in their careers, direct line responsibility for the manual labor force.

14

Mechanics of the Middle Class

In addition to the four engineering departments entirely within Division A, including approximately ninety engineers, there is also a research and development department, located in the same building complex as Division A but outside the divisional structure. Although roughly half the work of the sixteen engineers in this department is devoted to product or process development for Division A, this work is not closely integrated with the regular product cycle. Rather, it is of exactly the incidental type previously described as characteristic of old industry, resulting in the occasional introduction of a patentable product improvement or process modification, but by no means essential to the design and production of most items included in the standard line or available on special order. Neither does the developmental work of the research and development department—there is no basic research—represent a qualitative break with the techniques or personnel of other departments. Although there is less time pressure in research and development than elsewhere, engineering techniques are no more "scientific" than in other departments. Indeed, many of the research and development engineers continue to depend, for much of their work, on the time-tested techniques of machine tool work as they are practiced in the machine shop, which is an adjunct of the department. These engineers are often recruited from within the company after a fairly long period of employment, a policy that emphasizes the importance of experience rather than schoollearned, "theoretical" knowledge. Contronics The company chosen to represent advanced industry—called "Contronics"—was founded immediately after World War II. Company headquarters, the engineering staff, and a manufacturing facility are located about twenty miles from Precision Metals in the suburb of a small New England city. The company also has one manufacturing facility in the South and others in the Far East. Total employment, worldwide, is approximately twenty thousand. From its beginnings, Contronics established its reputation and market position by automating the production of the mass consumer item of which it is a leading producer in the United States. It has also, for many years, designed and manufactured a line of scientific instruments, primarily on government contracts. However, within the previous five years, the consumer item that is its

Precision Metals and Contronics

2-5

primary product had been revolutionized by developments in integrated circuitry and semiconductors. As a result, in order to maintain its position in an unstable and highly competitive market, Contronics established a third division, the Electronics Group, to begin production of a product line whose design and manufacturing processes are highly innovative. The items in this product line include a number of components, each of which is designed and manufactured separately. My research concerns the Electronics Group; unless otherwise specified, "Contronics" will refer only to it. The Electronics Group, the smallest of the three divisions at the time of my research, includes headquarters, an engineering staff, and a pilot production line at the New England site as well as a manufacturing facility in the Far East. The group had also recently acquired from a large electronics company a facility and engineering staff in New Jersey for the design and manufacture of an electrochemical component. However, because Contronics management was concerned about the effects of this recent acquisition on the engineering staff there, they refused me access to it. The engineers in the Electronics Group at the New England site are divided into three departments: manufacturing and process design (twenty-one engineers), product design (ten), and research (twelve). The manufacturing department is responsible both for the design and operation of the pilot production line maintained at the New England site and for the eventual transfer of technologies developed and tested there to full production facilities in the Far East. This broad function includes, for the process design engineers, the development of processes used in the manufacture of the two components made by Contronics itself; the development of assembly and packaging processes, a particularly critical element in the manufacture of products using semiconductor devices; the design of equipment used in both manufacturing and assembly; and the organization of labor processes for the manual workers who perform the tasks of manufacture and assembly.* In addition, three individuals with four-year engineering degrees were employed as foremen on the pilot line; these are rough equivalents of manufacturing en* Although Contronics had hired an engineer to begin manufacturing semiconductor devices at Contronics, at the time of my research, that component and most others used in the final product were purchased from outside suppliers.

2.6

Mechanics of the Middle

Class

gineers in the sense of that term as it is used at Precision Metals. The product design engineers are responsible for the overall design and testing of the product as well as writing specifications for components purchased from outside suppliers. For many years, the research department of the entire company had been housed separately from the rest of the engineering staff, about fifty miles from the main New England facilities. However, a few months before my research began, the research department of the Electronics Group was relocated at the main facility with the intent of integrating its work more closely with the day-to-day operations of product and process design. Unlike Precision Metals, research and development at Contronics is a regular part of the product cycle: Contronics introduces a new model on its pilot production line approximately once every six months. While the product design department was responsible for overall product design and the manufacturing department for the production and assembly of components, the research department was responsible for basic modifications of those "high-technology" components, whether made at Contronics or by outside suppliers, that distinguished one model from another. The Sites Compared Precision Metals is an older company (one hundred ten years) in an older industry. It is characterized by stable technologies in which research and development plays a minor role. Contronics is a relatively new company (thirty years) in a relatively new industry, consumer electronics. It is characterized by innovative technologies in which research and development plays a central role. Although exact employment figures were not available for individual divisions at either company, my rough estimates support a characterization of the Electronics Group at Contronics as "knowledge intensive": Of a total labor force of five hundred in the United States, approximately one in eleven were engineers. In Division A of Precision Metals, of a total U.S. labor force of three thousand, only one in thirty were engineers. Although Division A at Precision Metals is considerably larger than the Electronics Group at Contronics, the size and structure of the companies as a whole are broadly comparable. Both maintained an engineering staff at a central location with some facilities in the United States as well as outside the country. The main facili-

Precision Metals and Contronics

ties of the divisions in which I conducted research were only twenty miles apart. However, the companies did differ in the organization of their product markets, the significance of which to engineering tasks I had not fully recognized at the time of site selection. Precision Metals manufactures a product for sale to industrial customers; Contronics manufactures a mass consumer good. Consequently, both the relationship of engineers to the product market and the distribution of their tasks differ in ways that bear no explicit relationship to the impact of advanced industry. Many of the Precision Metals engineers meet directly with representatives of their industrial customers, a direct and personal experience of the product market; the Contronics engineers have no direct contact with the individual consumers who buy the product they design. At Precision Metals, the modification of standard orders to special orders assumes particular importance; at Contronics, the mechanization of production for a limited number of models assumes greater importance. Throughout the analysis of the two companies, distinguishing the effects of variations in the product market from those resulting from more basic attributes of old and advanced industry will be important. Detailed Procedures I conducted research over a period of six months: February, March, and April of 1 9 7 7 at Precision Metals and May, June, and July of 1 9 7 7 at Contronics. I first contacted the two companies by letter.* I then met with company representatives to explain the proposed research in greater detail, set conditions under which I could conduct the research, and make arrangements for my initial introduction to the engineering staffs. The research then proceeded in three overlapping stages: first, fieldwork at the work site and, less intensively, at the homes of individual engineers; second, interviews with forty engineers at each company; and third, throughout the period of research, discussions with nonengineering and manage*The letters to Precision Metals and Contronics were only two of approximately fifty I sent. Most companies I contacted did not respond; others responded only to deny access. A few others were willing to permit my research but did not meet site selection criteria as well as Precision Metals and Contronics.

28

Mechanics of the Middle

Class

rial officers, primarily in personnel departments, to gather information relevant to the companywide context. The Research Agreement: Conditions of Access Although interpreted somewhat differently at the two companies, it was a basic condition of the research, pressed both by me and by my "contacts" in the personnel offices of the companies, that participation be entirely voluntary on the part of individual engineers. Although no sanctions of any sort were applied for failure to cooperate, a few managers at Precision Metals mildly encouraged some engineers to participate. At Contronics, to the best of my knowledge, there was neither encouragement nor discouragement. In addition, although I agreed that aggregate findings would be made available to the company, discussions with individual engineers, whether during interviews or during the fieldwork stage, would remain entirely confidential. At Contronics, these arrangements were explained to the engineers in a memo from the personnel department and again by me in a meeting to which all the engineers had been invited at the beginning of the research. At Precision Metals, the agreements (and introductions) were explained in brief meetings with the managers of engineering departments and then, more casually, to individual engineers by those managers. Fieldwork Engineering work (indeed, most higher-level, white-collar work) is not easy to observe. Unlike manual labor, which is often limited to a small number of operations repeated in a regular sequence within an area circumscribed by a machine or group of machines, engineering is characterized by a greater variety of operations, sequenced irregularly, and performed in a number of different locations. Moreover, manual work consists of the physical manipulation of material objects and is, therefore, concretely observable, but much of engineering is desk work based on thought processes whose character is manifest only partially in the various pieces of paper that are its immediate product. Finally, engineering work is often highly differentiated, with what are often unique job titles reflecting the unique character and combination of an individual's responsibility.

Precision Metals and Contronics

Some aspects of engineering work, especially its technical content, resist observation, but other aspects, particularly the social relations in which it is embedded, are rather easily observed: the organization of work groups (whether an engineer works alone or in teams, whether he works with other engineers, technicians, manual workers, or nonengineering, white-collar workers); the range, if not the frequency, of particular activities (talking on the phone, inspecting machinery, doing lab work, performing calculations, writing, attending meetings, socializing); and, perhaps most important, some aspects of the control to which engineers are subject (the physical presence or absence of supervisors, the frequency and content of meetings or discussions with supervisors, physical freedom of movement). Observation, in a strict sense, is only one part of fieldwork. Equally important are casual conversations, which help familiarize the researcher with the history and structure of the organization, and the technical content of work itself. I began the fieldwork by following individual engineers, chosen as representative of functions in the product cycle and of different levels of responsibility, as unobtrusively as possible ("like a shadow," according to one engineer) over the course of the workday. Later, as I became a familiar figure to most of the engineers, my main technique was simply—no more elegant phrase is as precise—"hanging out." At this point, usually after I had begun the interviews, observation bordered on participant observation: Although I was neither an employee of the company nor an engineer, I was one more "science worker" doing his job at the company, someone to have lunch with, talk to during coffee breaks, and, occasionally, take home for dinner. But the casual quality of this phase of the fieldwork should not detract from its importance; it was in these relatively relaxed and informal moments that the engineers were often most open about their beliefs and attitudes. The Interviews I conducted all but six of the interviews myself. They typically lasted from two to three hours and took place both at the work site (either during or after working hours) and at the homes of engineers, the time and location determined by the engineer. Occasionally, especially for the longer interviews conducted during working hours, it was necessary to break the interview into two sessions.

3 ^ 3 rs| Û-3 W . _: . >o o ° "o 2 > gg-S 5 o •• S o\ • 2

o « S X» " M K 5 CN O U

¡

3 tx Os rn O lo so 00 00

s

•a

I s •Sí"? O c/j -O ¡o

^ u £ .SP -= g» ¡2

m -

ci -ü

¡C i5 -ï^ ~ ÜS u. a

s e - s S•aS «i 3s a, J j §•«

o oo os

£

S S j U •S-S-SO ->. -». ->.2S M -O -O M) c B§J .2 .2 -2 "S m s « ¡g

s 5ò c Üj

«S o PO (X IO