1—
Introduction:
The Task of Explanation

Do international differences in culture create and sustain decisive, systematic divergences in the formation of manufacturing practices and of industrial relations? In this study I begin with a controlled test to judge with fresh reasoning and evidence whether the answer has been "yes." I endeavor to show that as factory systems took shape in nineteenth-century Europe, contrasting techniques of manufacture emerged in similar economic settings as a result of the cultural premises structuring the producers' conduct. These assumptions concerned an enigmatic transaction whose original strangeness now eludes us: the sale of human labor as a commodity.

In the long and difficult crossing from the feudal and corporate organization of work to a liberal commercial order in Europe, labor became more than an expression of human diligence or a means of generating prosperity. Labor came into public sight as a crude ware. It was objectified as a disposable material with a metric value. To characterize this outcome by means of a generic model of "capitalist development" is to mistake a token for the object. I try to show that during the fateful transition to the new commercial order, a different apparition of labor as a commodity took hold in each of the leading economies of Western Europe. In each country a different solution prevailed for determining just how the precious but subtle thing called labor could be calibrated and transferred from hired hands to the employer in the workshop. How could workers sell their ephemeral activity to an employer if the realized output alone bore an exchangeable value? Just how should the "quantity" of labor be measured? At what moment was the labor considered to have been conferred upon the employer? British workers and employers resolved these questions differently than did their German and French counterparts. The diverging specifications of labor as a commodity, inherited

from peculiarities in each country's initial transition to a formally free market regime, became entrenched instruments of practice in the succeeding age of the factory. Labor did not serve only as a tool of production; in each economy its symbolic calibration organized in a distinct fashion the experience of industrial work and the use of time and space in the production process.

To isolate the industrial effects of diverging cultural definitions of labor as a commodity, this inquiry starts with a comparison of shop-floor activity in the wool textile mills of Germany and Britain during the nineteenth century. These cases provide an ideal comparison of enterprises that developed under similar circumstances in the early tide of industrial change. German and British wool textile mills developed contemporaneously, installed similar kinds of machines, and competed in the same markets. Such uniformities help to rule out conventional economic explanations for the emergence of divergences in shop-floor procedures. Instead, the fundamental similarity in the immediate economic and technical setting for this branch of textiles allows us to highlight culture as the structuring principle of national differences in factory practice. Differing conceptions of labor as a commodity gave rise to national contrasts in methods of remuneration, calculation of output and costs, disciplinary techniques, rights to employment, and even mill architecture. The dissimilarities pervaded industrial experience, for they were contained within each of the nationally prevailing definitions of the valorization of labor. To speak of "the rise of market culture" or the commodification of "labor" without contextualizing their definitions falsely objectifies our terms of understanding. It might seem that the expressions labor and capital , as elemental and necessary constituents of commercial bourgeois culture, would naturally take on the same meaning throughout industrializing Europe. But they appeared in varying guises and signified disparate features of human endeavor within the German and the British economies.

Cultural conceptions of labor as a ware do not only illuminate the fixed structures of early factory life. They also aid us in appreciating the strategies and demands of labor movements. Until the classical period of laissez-faire industrialization came to an end in the First World War, German and British workers enacted strikes and protests with different beliefs about what comprised the withholding of the commodity of "labor." They articulated different responses to identical workplace challenges, such as employers' imposition of disciplinary fines. Finally, German and British workers arrived at different explanations for their exploitation based upon their per-

ception of labor as a commodity. In the decades leading up to the First World War, the German labor movement proved more receptive to Marx's analysis of exploitation than did its British counterpart, even among British workers who were convinced of the necessity of dramatic social transformation. I suggest that for each country, the schemas encoded in silent practices within the private factory lent workers the concept of labor they used to voice demands in the public sphere.

If reliance upon a vision of labor as a commodity cast both factory practices and workers' responses in a distinctive image, why did a different apparition of labor prevail in each country? I try to show how conjunctural differences in the timing of the recognition of formally free markets in finished goods, in the abolition of feudal dues in labor, and in the breakdown of guild supervision over urban labor established different motivating conditions for the definition of labor as a commodity. To dissect the combinations of factors that inspired varying cultural outcomes, I investigate the transition to a formally free market in labor not only in Britain and Germany but in France and, more briefly, in northern Italy. The forces at work in these cases show that the understandings of the labor transaction that prevailed in Germany and in Britain, the two chief cases for analysis, resulted from opposite journeys among an array of developmental pathways to wage labor in western Europe.

The discovery that the world of concrete procedures on the shop floor was systematically structured by cultural specifications of labor as a commodity—by practical "theories" about labor, if you will—may open a new avenue of research into correspondences between nineteenth-century practice and the postulates of political economy reigning in that age. I endeavor to show that the contrasts in the apprehension of labor marking the German and British variants of classical political economy matched the differences in the theories about labor operating in German versus British manufacturing. Adam Smith's portrayal of the exchange of labor products recovered the presumptions about the transfer of labor that governed British industrial procedure in his day and long after. Karl Marx's celebrated reflections upon labor power, it turns out, replicated the definition of labor contained in German treatises composed at the beginning of the nineteenth century. Marx's formal economic analysis eerily borrowed from the established cultural schema distinctive to the workers and business people of Germany. Each country's intellectual representatives brought the implicit theory embedded in the quotidian practices of manufacture into the explicit theory of political economy.

In brief, this work examines the national origins of cultural definitions of labor as a commodity, the installation of these specifications into procedures on the shop floor in Germany and Britain, and the ideological consequences for the labor movements of such culturally structured forms of industrial practice. The work's range is broad but its analytic focus precise: it portrays the development of manufacturing to shed light on the explanatory significance of popular understandings of labor as a commodity. My comparative perspective focuses upon the responses of German and British workers that typify the overarching differences in the definition of labor as a ware in each country. Given this cross-national perspective, only occasionally do I dwell upon more specific differences in workers' responses within each country based on occupational, gender, and regional identities. If I manage to encourage further reflection upon the practical effects of labor's reification as a commodity, I will have accomplished my task.

The Initial Test Cases

Comparative history succeeds when the grounds for juxtaposing cases are specified with precision. My examination of the German and British wool textile industries attempts to single out the effects of culture upon the workplace by providing approximate controls for the confounding effects of differing economic and technical trajectories of development. Britain's reputation for having had a unique experience as the textile pioneer rests on intense, in some respects excessive, attention to the precocious development of cotton factories in Lancashire at the end of the eighteenth century. In Yorkshire, however, the most important center of the country's wool trade, power looms in weaving sheds did not prevail until after the middle of the nineteenth century—by which time the woolen and cotton mills in Germany had also begun to mechanize.^[1] Whereas the British enjoyed a head start of half a century in the mechanized treatment of cotton, their lead in the technology for wool was minimal. This consideration simplifies the task of presenting a cultural account of the differences in factory practices that emerged. It helps to exclude explanations of differences that appeal to the timing of development or to the world industrial environment prevailing at the inception of a factory system.^[2]

[1] Georg Brodnitz, Vergleichende Studien über Betriebsstatistik und Betriebsformen der englischen Textilindustrie (Halle: Habilitationsschrift Universität Halle-Wittenberg, 1902), pp. 15–16.

[2] John Child and Alfred Kieser conducted an Anglo-German comparison of managers' authority which revealed that, taking into account the size of firms, decision-making was morecentralized in contemporary German than in British companies. Although this finding tallies with stereotypes about German culture, the results could also be explained as adaptations to circumstances under which those firms were founded or to the present conditions of doing business in the branches of manufacture to which these firms devoted themselves. My comparative design takes these alternative economic explanations into account. John Child and Alfred Kieser, "Organizational and Managerial Roles in British and West German Companies: An Examination of the Culture-Free Thesis," in Cornelis Lammers and David Hickson, editors, Organizations Alike and Unalike: International and Interinstitutional Studies in the Sociology of Organizations (London: Routledge and Kegan Paul, 1979), pp. 252–271. The most recent influential cross-national study which omitted controls for technology and path of development is that of Gary Hamilton, Nicole Woolsey Biggart, and Marco Orrù, "Organizational Isomorphism in East Asia," in Walter W. Powell and Paul J. DiMaggio, editors, The New Institutionalism in Organizational Analysis (Chicago: University of Chicago Press, 1991), p. 366. In an important comparison of workers' attitudes in the French and British oil refining industries, Duncan Gallie set up controls for the contemporary "level of technological development" in these enterprises but not for their developmental trajectory in each country. Social Inequality and Class Radicalism in France and Britain (Cambridge: Cambridge University Press, 1983).

Why did the mechanized production of wool cloth appear later than that of cotton in the global textile industry? It was a quirk of nature that placed its mechanization on a deferred time scale. Wool fibers proved more recalcitrant to mechanical handling than cotton. Although some enterprises for the power weaving of cotton succeeded in Britain during the 1790s,^[3] experimental power looms for weaving in woolens did not evolve in Britain until the 1830s.^[4] By the 1850s these looms could outperform hand looms, and the construction of mechanized weaving mills was begun in earnest.^[5] Thereafter change was swift. Employers in the Yorkshire woolen trade had nearly completed the shift to power weaving by the start of the 1870s.^[6]

[3] Neil J. Smelser, Social Change in the Industrial Revolution (Chicago: The University of Chicago Press, 1959), p. 147.

[4] Frederick James Glover, "The Rise of the Heavy Woollen Trade of the West Riding of Yorkshire in the Nineteenth Century," Business History Volume 4, No. 1 (December 1961), p. 9.

[5] In 1856 total employment in woolen and worsted textile mills exceeded the number of handweavers in those branches, but statistics for the mills included auxiliary workers. Brodnitz, op. cit., pp. 15–16. In 1860, Samuel Jubb reported that wages for hand weavers in the Batley district had suffered no decline and remained above those for power-loom weavers. The History of the Shoddy-Trade (Batley: J. Fearnsides, 1860), pp. 68–70. One-quarter of the looms in Huddersfield in 1868 were still worked by hand. J. H. Clapham, "The Decline of the Handloom in England and Germany," Journal of the Bradford Textile Society Volume 11 (1905), p. 45.

[6] Handweaving remained an important enterprise at least into the 1880s around Kirkburton, Skelmanthorpe, and Shelley. Textile Manufacturer , June 15, 1884, p. 237, and October 15, 1886, p. 447. For an instance of a firm using sixty handlooms in 1891: Yorkshire Factory Times , July 10, 1891, Heckmondwike. For other British handweaving towns in the 1890s, see Gerhart von Schulze-Gävernitz, Der Grossbetrieb: Ein wirtschaftlicher und socialer Fortschritt (Leipzig: Duncker & Humblot, 1892), p. 272. The Yorkshire worsted textile industry, concentrated in Bradford, used specially combed wool fibers. Worsted weaving was mecha-nized about two decades before woolens. Since the cultural principles in the treatment of labor in worsted mills were the same as in woolen mills proper, I include evidence from the worsted branch in my grouping of Yorkshire mills. E. M. Sigsworth, "The Woollen Textile Industry," in Roy Church, editor, The Dynamics of Victorian Business (London: George Allen & Unwin, 1980), p. 193.

In Germany the transition to power looms for woolens occurred more unevenly. A few entrepreneurs experimented with power looms during the 1830s in Berlin and Saxony.^[7] By the 1860s, the mechanization of wool weaving was fully underway. In the preeminent textile centers of northwestern Germany, such as Elberfeld and Rheydt, the mechanization of weaving with wool materials was nearing completion by 1875—at almost the same time as in Yorkshire.^[8] Small wonder Karl Marx commented that "whenever one travels through the Prussian Rhineland and Westfalen, one thinks of Lancashire and Yorkshire."^[9] Towns in Saxony, the other leading industrial region in German textiles, lagged about a decade behind.^[10]

To be sure, Germany had outlying areas such as upper Lausitz, where isolated hand weavers survived even into the 1890s.^[11] Perhaps the decisive condition is that in wool technology, unlike cotton, the German and the

[7] Horst Blumberg, Die deutsche Textilindustrie in der industriellen Revolution (Berlin: Akademie-Verlag, 1965), p. 89.

[8] Wolfgang Hoth, Die Industrialisierung einer rheinischen Gewerbestadt, dargestellt am Beispiel Wuppertal (Köln: Rheinisch-Westfälisches Wirtschaftsarchiv, 1975), p. 200; Alphons Thun, Die Industrie am Niederrhein und ihre Arbeiter. Part Two: Die Industrie des bergischen Landes (Leipzig: Duncker & Humblot, 1879), p. 197; Horst Matzerath, "Industrialisierung, Mobilität und sozialer Wandel am Beispiel der Städte Rheydt und Rheindahlen," Probleme der Modernisierung in Deutschland (Opladen: Westdeutscher Verlag, 1979), p. 23. A useful summary of the mechanization of wool production in northwestern Germany appears in Kathleen Canning, "Class, Gender, and Working-Class Politics: The Case of the German Textile Industry, 1890–1933," Ph.D. diss., Johns Hopkins University, 1988, p. 73.

[9] Quoted in Rolf Dlubek and Hannes Skambraks, editors, "Das Kapital" von Karl Marx in der deutschen Arbeiterbewegung: 1867 bis 1878 (Berlin: Dietz Verlag, 1967), p. 23.

[10] By 1880, in such Saxon centers as Glauchau and Gera power looms prevailed even in the fancy goods market. Louis Bein, Die Industrie des sächsischen Voigtlandes. Zweiter Theil: Die Textil-Industrie (Leipzig: Duncker & Humblot, 1884), pp. 444–445; Enquête-Commission zur Untersuchung der Lage der Glauchau-Meeraner Webwaaren-Industrie, "Bericht der zur Untersuchung der Lage der Glauchau-Meeraner Webwaaren-Industrie berufenen Enquête-Commission," 1881, Stadtarchiv Glauchau, W Abt. 1, Nr. 41; for weaving in Gera, Gera-Greizer Kammgarnspinnerei, 50 Jahre Gera-Greizer Kammgarnspinnerei 1890–1940 (Gera: Karl Basch & Co., 1940), p. 78. In Zittau, the production of half-wool goods was mechanized by 1860. Brigitte Wrobel, Geschichte der Arbeiterbewegung des Kreises Zittau: Chronik 1830–1945 (Zittau: Kommission zur Erforschung der Geschichte der örtlichen Arbeiterbewegung, 1972), p. 3.

[11] See Jean Quataert, "Workers' Reactions to Social Insurance: The Case of Homeweavers in the Saxon Oberlausitz in the Late Nineteenth Century," Internationale Wissenschaftliche Korrespondenz zur Geschichte der deutschen Arbeiterbewegung Volume 20, Number 1 (1984). In 1875, only 34 percent of looms in Germany as a whole in the wool branches were mechanized. Horst Blumberg, op. cit., pp. 89 ff.

British engineers were equals from the start.^[12] From the first days of mechanization, German wool equipment was usually of domestic design and manufacture.^[13] As early as 1855, at the Paris Exposition, the Germans competed on an equal footing with the British in the design of textile machines for wool.^[14] The German weavers themselves often preferred their own country's equipment over the British versions.^[15] The technical literature shows that in the decades before the First World War, German and British looms from the woolen branches typically ran at similar speeds, as measured by the number of crossings the looms' shuttles could finish each minute.^[16] On the eve of the war, a German business journal even boasted that, ranked against British factories, "the German woolen industry in its technical and organizational institutions can be considered in many respects superior."^[17]

A focus on wool textiles as a test comparison also simplifies the task of explanation because it offers basic parallels between the niches in the world market occupied by the producers. Business journals from the nineteenth century confirm that German and British fabrics made of wool and wool mixtures were often of very similar design.^[18] The markets within the British Empire, which were protected for British manufacturers, absorbed primarily cotton, not wool, manufactures.^[19] In the wool

[12] Herbert Kisch, "The Crafts and Their Role in the Industrial Revolution: The Case of the German Textile Industry," Ph.D. diss., University of Washington, 1958, p. 35.

[13] Textile Manufacturer , March 15, 1881, p. 95. Artur Peltzer, "Die Arbeiterbewegung in der Aachener Textilindustrie von der Mitte des 19. Jahrhunderts bis zum Ausbruch des Weltkrieges," Ph.D. diss., Universität Marburg, 1924, p. 9. Domestic machinery played a role in the mechanization of wool weaving in Saxony as early as the 1830s. Horst Blumberg, op. cit., p. 89; Gewerbe-Blatt für Sachsen , June 21, 1838, pp. 197–198.

[14] Alfred Schröter and Walter Becker, Die deutsche Maschinenbauindustrie in der industriellen Revolution (Berlin: Akademie-Verlag, 1962), p. 186. In worsteds, it seems, German weavers in the Wuppertal did import looms from Bradford. See Centralblatt für die Textil-Industrie Volume 15 (1884), p. 233. In worsted spinning the Germans disseminated their own machine designs in the 1860s. F. Orth, Der Werdegang wichtiger Erfindungen auf dem Gebiete der Spinnerei und Weberei (Berlin: Verein Deutscher Ingenieure, 1922), p. 81.

[15] Textile Manufacturer , September 15, 1883, p. 391; Centralblatt für die Textil-Industrie Volume 19 (1888), p. 435.

[16] Royal Commission on Labour, PP 1892 XXXV, p. 207. HSTAD Regierung Aachen 1634, report of February 4, 1899.

[17] Seide , February 1914, pp. 85 ff.; Die Textil-Zeitung , Feb. 16, 1897, p. 895.

[18] Das deutsche Wollen-Gewerbe , August 10, 1884, p. 1285. The product market was extremely fragmented, however. Sigsworth, op. cit., pp. 185, 190. German wool manufacturers, who began with a reliance on the luxury goods market, pursued a strategy of diversifying their output. They competed in both fancy and plain styles. D. T. Jenkins and J. C. Malin, "European Competition in Woollen and Cloth, 1870–1914: The Role of Shoddy," Business History Volume 32, Number 4 (October 1990), p. 82.

[19] British wool fabric exports were focused on Europe. Gary Firth, "The Bradford Trade in the Nineteenth Century," in D. G. Wright and J. A. Jowitt, editors, Victorian Bradford(Bradford: Bradford Metropolitan Council, 1982), pp. 30–31. Schulze-Gävernitz aptly described the extent to which the British enjoyed a protected market for cotton staples: "It is therefore correctly stated in Lancashire that the demand from India year by year for certain cotton stuffs is as sure as that the English nation requires every year a certain quantity of wheat." The Cotton Trade in England and on the Continent (London: Simpkin, Marshall, Hamilton, Kent and Co., 1895), p. 69. At the top end of the scale for fine yarns, British cotton spinners in Lancashire hardly competed with foreign enterprises, for their expertise lent them a virtual world monopoly. John Jewkes and E. M. Gray, Wages and Labour in the Lancashire Spinning Industry (Manchester: Manchester University Press, 1935), p. 48.

trade, however, the British exporters had to compete to a greater degree in noncolonial markets, where entry for British goods was no easier than for German ones.^[20] In the wool business, therefore, differences in the market demands satisfied by German and British manufacturers can more easily be discounted as an essential cause for divergences in factory customs.

The German and British wool textile industries resembled each other another way: in both countries, the majority of factories in this branch operated under the principal ownership of family partners.^[21] In cotton, by contrast, joint stock undertakings prevailed in some British towns, in particular within Oldham's spinning trade.^[22] The sizes of firms in the two countries varied greatly by market specialty, by region, and even by town,^[23] but in national comparisons their average sizes were not far apart. In both

[20] For comments on German competition in the American market for woolens and worsteds, see Textile Manufacturer , June 15, 1884, p. 244. On German competition in other markets, see Erich Thal, Die Entstehung und Entwicklung der Halbwoll- und Wollindustrie im M.-Gladbacher Bezirk bis zum Jahre 1914 (Mönchengladbach: W. Hütter, 1926); and Jenkins and Malin, op. cit., p. 69.

[21] Textile Mercury , March 28, 1914, p. 253.

[22] von Schulze-Gävernitz, Der Grossbetrieb , pp. 69, 91; Patrick Joyce, Work, Society and Politics (London: Methuen, 1980), p. 340. Karl Emsbach, Die soziale Betriebsverfassung der rheinischen Baumwollindustrie im 19. Jahrhundert (Bonn: Röhrscheid, 1982), pp. 407–411. Mike Holbrook-Jones, Supremacy and the Subordination of Labour (London: Heinemann Educational Books, 1982), p. 36. Even limited liability firms were sometimes managed by members of a single family. For an example from textiles, see Augustus Muir, The History of Bowers Mills (Cambridge: W. Heffer, 1969), p. 35, and, for national overviews, David Landes, "The Structure of Enterprise in the Nineteenth Century: The Cases of Britain and Germany," in Comité International des Sciences Historiques, editor, Rapports V: Histoire contemporaine (Uppsala: Almwvist & Wiksell, 1960), pp. 115, 121, and T. R. Gourvish, "British Business and the Transition to a Corporate Economy: Entrepreneurship and Management Structures," in R. P. T. Davenport-Hines and Geoffrey Jones, editors, Enterprise, Management and Innovation in British Business, 1914–1980 (London: Frank Cass, 1988), p. 26.

[23] For example, enterprises in the old textile center of Bocholt were much smaller than elsewhere in the Münsterland. Bernhard Bergmeyer, "Das Baumwollgewerbe im Münsterlande," diss., Bonn, 1921, pp. 87 ff. On the eve of the First World War, the Rhineland had more looms in place than did the Münsterland, but its cotton mills averaged only half as many looms each.

countries, the weaving departments of woolen mills at the close of the century typically employed about sixty looms.^[24]

Finally, in wool factories the development of textile workers' unions and collective bargaining show fundamental similarities in the two countries. As is well known, Lancashire's cotton towns sponsored the development of strong and enduring craft unions for spinners and weavers back in the era of artisanal production.^[25] This unusual legacy of early organization presents a striking structural difference in an Anglo-German comparison of cotton mills. A comparison of regions with wool mills eliminates this complexity. From an analytic standpoint it is fortunate that unions in Yorkshire for factory weavers and spinners did not become full-fledged standing organizations until after 1881 in the Colne Valley, and not until the 1890s in other localities.^[26] The rise of the union movement in Yorkshire therefore coincided with the emergence of formal organization in Germany, where factory textile unions experienced a take-off in membership during the 1890s. In Yorkshire the major union for textile workers embraced both weavers and spinners, establishing another parallel to the German case.^[27] The similarities in the timing and structure of collective organization among British and German wool factory workers help us to assess alternative explanations for divergences in shop-floor customs in this branch.

History never duplicates its creations to order. Yet the basic similarities in technology, the timing of mechanization, product lines, proprietorship, and the structure and procession of workers' unionization allow a focus on

[24] J. H. Clapham, op. cit., p. 133. Textile Mercury , April 8, 1911, p. 271. Germany, Statistik des Deutschen Reichs , Volume 214 (Berlin: Kaserliches Statistisches Amt, 1910), p. 303. Brodnitz found close parallels in the average sizes of the work forces of German and British worsted and woolen weaving mills excluding hand shops. Brodnitz, op. cit., p. 31.

[25] Isaac Cohen summarizes the evidence in American Management and British Labor (Westport, Connecticut: Greenwood Press, 1990), p. 15.

[26] Ben Turner, Short History of the General Union of Textile Workers (Heckmondwike: Labour Pioneer, 1920); Ben Turner, A Short Account of the Rise and Progress of the Heavy Woollen District Branch of the General Union of Textile Workers (N.p.: Yorkshire Factory Times Press, 1917). Prior associations for weavers in Yorkshire, whose by-laws were printed as early as 1824, were short-lived. Report from the Select Committee on Combination Laws , PP 1825 (437) IV, p. 27.

[27] Yorkshire Factory Times , May 26, 1893, p. 5; Joyce, op. cit, p. 76. The independent local Yeadon textile union, too, emphasized recruitment across professional divides. Archive of General Union of Dyers, Bleachers, and Textile Workers, Bradford, Union minutes, May 13, 1896, and Oct. 28, 1903. For Germany, see Klaus Schönhoven, "Localism—Craft Union—Industrial Union: Organizational Patterns in German Trade Unionism," in Wolfgang J. Mommsen and Hans-Gerhard Husung, editors, The Development of Trade Unionism in Great Britain and Germany, 1880–1914 (London: George Allen & Unwin, 1985), p. 229; R. M. Dehn, The German Cotton Industry (Manchester: Manchester University Press, 1913), pp. 73, 78.

the German and British wool mills to approximate an ideal comparison. This industry also affords a weighty, if relatively neglected, body of evidence. At the turn of this century, the Yorkshire district counted about 180,000 woolen and worsted textile workers.^[28] In Germany, there were at this time as many weavers in wool as in cotton.^[29] For the German case, much of the primary evidence comes from northwest Germany and Saxony, where mills were renowned for the stiff competition they offered Yorkshire. British technical journals in the nineteenth century selected the German towns of the Wuppertal and of the lower Rhine, such as Düren, as their chief competitors in the woolen market. They also singled out the work forces in these German towns for their ability to rival British textile workers in technical expertise.^[30]

Of course, a comparison limited to wool textile factories within narrow geographical regions renders suspect any allegation that findings result from nationally prevailing cultural differences. Where my explanation relies upon the influence of culture, I have an obligation to demonstrate the generality of the outlooks that I hold responsible for divergences in factory customs. In addition to my evidence on the wool mills, which serves as the decisive test example, I therefore include many examples from factories in the major branches of textiles in other regions of each country. At this point I can take advantage of a supplementary kind of comparison. In spite of differences in the timing of mechanization and in the labor and product

[28] Data for 1901 from J. H. Clapham, The Woollen and Worsted Industries (London: Methuen & Co., 1907), frontispiece. To put Yorkshire employment in perspective, the cotton industry of the time in Lancashire retained about half a million workers. Sydney J. Chapman, The Lancashire Cotton Industry: A Study in Economic Development (Manchester: Manchester University Press, 1904), p. 179. The German spinning and weaving mills of the wool branches in 1907 employed over 180,000 workers. Germany, Statistisches Jahrbuch für das Deutsche Reich (Berlin: Kaiserliches Statistisches Amt, 1909), p. 80.

[29] In 1895, 30 percent of German weavers labored in the wool trade, and 29 percent in cotton; in 1907 the figures were nearly the same. See summary statistics in Karin Zachmann, "Der Mechanisierungsprozess in der deutschen Textilindustrie im Zeitraum von 1870 bis 1914," Beiträge zur Wirtschaftsgeschichte , Number One (Dresden: Technische Universität Dresden, 1988), p. 29. In all, the German and British textile workforces were of similar size, each totaling about one million workers. Textile Mercury , 1909, p. 330.

[30] For the Wupper Valley and Krefeld, see Textile Manufacturer , June 15, 1883, p. 217. For Düren, see Textile Mercury , August 30, 1890, p. 139. For Mönchengladbach, see Textile Manufacturer , February 15, 1906, pp. 37–38, and Textile Mercury , December 15, 1883, p. 510. As early as 1828, William Radcliffe focused on Elberfeld as a powerful adversary in the weaving business. Origin of the New System of Manufacturing Commonly Called "Power-Loom Weaving" (Stockport: J. Lomax, 1828), p. 92. Although not so famous a pioneer as the Wupper Valley, Düren was distinguished for its centralization of woolen manufacture in large manufactories during the first half of the nineteenth century. Joachim Kermann, Die Manufakturen im Rheinland 1750–1833 (Bonn: Ludwig Röhrscheid, 1972), p. 155.

markets of the various branches of textile production within each country, the practices on the factory shop floor within each country tend to display similar traits. This extension increases my confidence that nationally dominant cultural assumptions represent the source of similarity in outcomes within each country.

Even when this study incorporates evidence from cotton, silk, and jute mills to help establish the generality of cultural differences, the comparison of the textile industries as national wholes rests on a prudent criterion. With the maturing of textile machinery in the late nineteenth century, Germany and Britain comprised the premier exporters in the world textile market. As early as the 1880s the professional textile periodicals in Britain focused on Germany, not France, as Britain's most important challenger.^[31] By the start of the First World War, British managers complained that the Germans utterly controlled the trade in certain fancy weaves.^[32] "We have been lamenting or resenting 'foreign competition,' " the Textile Mercury said upon the outbreak of the war, "—meaning by that term almost exclusively German competition in the outside markets of the world."^[33] In cotton, to be sure, the British retained a substantial edge in efficiency. They operated their cotton weaving equipment at a speed perhaps 30 percent higher on average than that of German competitors.^[34] Yet even in cottons, weaving mills in the two countries were of similar sizes.^[35] Outside of wool textiles, the German and British textile producers did not stand at equivalent levels of technical and

[31] Textile Manufacturer , July 15, 1883, p. 272. For other examples of the British preoccupation with German competition, see Textile Recorder , October 15, 1892, worsted branch; Textile Manufacturer , September 15, 1889, p. 436; Yorkshire Factory Times , February 15, 1912, speech by Swire Smith to the Bradford Textile Society. British textile employers told their workers that German competition in particular prevented them from granting wage increases. Bradford Labour Echo , January 28, 1899.

[32] Journal of the British Association of Managers of Textile Works, 1918–1919 (Volume 9), p. 108.

[33] Textile Mercury , August 22, 1914. The American textile industry hardly competed in the world market. For comparative estimates of national exports of wool manufactures, including those from the United States, see D. T. Jenkins and K. G. Ponting, The British Wool Textile Industry 1770–1914 (London: Heinemann Educational Books, 1982), p. 294. For a discussion of the lack of American competitiveness in exports of cottons, see Gary Saxonhouse and Gavin Wright, "Stubborn Mules and Vertical Integration: The Disappearing Constraint?" Economic History Review 40, n. 1 (February 1987), p. 93.

[34] von Schulze-Gävernitz, The Cotton Trade , pp. 107–108.

[35] Ibid., pp. 67, 79. German cotton spinning mills were typically larger than those of the British, whereas the British cotton weaving factories were larger. See employment figures (excluding self-employed) in Brodnitz, op. cit., pp. 23–24.

professional advance, but at least they occupied closely related stages of development.^[36]

At junctures when I consider alternative explanations for factory customs based on market adaptation or "rational choice," I usually must return to the main comparison motivating this study, that of the wool branch, for the most effective controls on sources of variation. Although my comparative framework at these points allows me to consider and reject specific noncultural explanations for differences in factory customs, I never treat culture as a residual category. That is, in no instance do I assume that a practice unexplainable by economic principles is attributable to culture by default. Nor do I suppose that culture clarifies only the variation that remains after applying economic reasoning. The logic of isolating an important cultural cause of differences in outcomes by considering alternative, economic sources of differentiation in no way implies that culture serves only as a supplement for explaining what is left over.

Although the strategy for ruling out alternative explanations for differences in outcomes follows basic comparative logic, the design of my model follows a line of reasoning specific to cultural analysis. Let me preview some findings to illustrate. German owners and workers viewed employment as the timed appropriation of workers' labor power and disposition over workers' labor activity. In contrast, British owners and workers saw employment as the appropriation of workers' materialized labor via its products. These divergent assumptions led to differences in the definition of wages, the calculation of costs, rights of employment, disciplinary fines, and the design of factory buildings. Since the manufacturing practices in each country formed a meaningful constellation, my positive argument is configurational, attached to an overarching pattern of techniques rather than to

[36] By the late nineteenth century the Germans also became the chief rivals of the British in the export of the textile machinery itself. The Textile Manufacturer of Manchester reported that German textile machine makers before 1914 had "built up an export trade of considerable dimensions which meets us in the neutral markets of the world." Textile Manufacturer , February 15, 1917, pp. 60 ff. True, in the early decades of mechanization after midcentury, the Germans remained dependent on British machine designs in the cotton branch. Textile machine makers in Germany began modifying British makes by the 1860s. F. Orth, op. cit., p. 81. By the turn of the century the Textile Manufacturer , in a report on the looms crafted in Chemnitz by Louis Schönherr, said that "although this country [Britain] still holds the premier position in the building of looms  . . . there are many instances where foreign ingenuity has got slightly ahead of us." Foreign loom makers, the journal concluded, "are working abreast of us—not behind us." Textile Manufacturer , November 15, 1901, pp. 375–377. The Germans thought even more highly of their equipment. At the turn of the century, the German Textil-Zeitung said, "We cannot learn anything from the British, for German loom construction absolutely nothing. We are superior to them." February 16, 1897, p. 126.

a simple outcome. This challenges rival explanations to account for an equally broad range of details in German and British factory customs.

Configurational analyses of factory practices are best executed through comparisons of single industries. By this means the investigator may scan the entire breadth of the practices of production to discern the significance of consistencies which would otherwise go unnoticed. This strategy also permits the researcher to contrast solutions to technical problems that are particular to each branch of capitalist enterprise. As signifying practices, manufacturing techniques create a system of signification from the fixtures specific to each kind of commercial undertaking. It is useless to hold culture constant while varying the economic and technological circumstances—say, by comparing textile and metal factories within the same region—for culture cannot yield uniform effects across industries. For instance, a difference between the fining systems in Germany and Britain was unlikely to appear in industries where a multi-stage production process made the assignment of responsibility for faults impossible or where routine channels for customers to bargain over the price of damaged goods were lacking. In short, the investigator searching for generalities cannot extrapolate from textiles to make inferences about the design of German and British factories across industries as one would extrapolate from a statistical sample. The shop floor furnishes a literal example of a social institution that can be viewed, as Jean Comaroff once expressed it, as a "meeting ground  . . . of two distinct orders of determination—one material, the other semantic."^[37] In this study I begin generalizing my findings about textiles to characteristic practices in other labor processes, such as coal mining and iron casting. But I do not look for uniformities in production methods across businesses in each country. Instead, I search for meaningful analogies in practices within each country, considering the technical environment peculiar to each kind of industry. The parallels across economically dissimilar branches of enterprise rebut many economic explanations for nationally prevailing routines.

Comparative analyses of the influence of culture on factory organization have to date chosen contrasting cases with an eye to maximizing the cultural differences in the cases under review. The first landmark study to compare the effect of national traditions on the development of factory systems, Reinhard Bendix's enduring Work and Authority in Industry , took Victorian Britain and Tsarist Russia for its primary cases. This gave Bendix a

[37] Jean Comaroff, Body of Power, Spirit of Resistance (Chicago: University of Chicago Press, 1985), p. 4.

pairing of sharply contrasted ideologies used by entrepreneurs to legitimate their authority, and he highlighted this variable to explain differences in the evolution of their systems of industrial relations. Bendix did not assess the independent contribution of cultural or ideological traditions, as opposed to purely economic and technological variables, in the creation of factory institutions.^[38]

In recent years the unmatched performance of Japanese firms in the world market has intensified research into the historical origins of their system of industrial relations. The pioneering analyses, such as those by Ronald Dore and Robert Cole, have suffered from the same inability to disentangle cultural and economic influences. They have compared Japan with structurally dissimilar cases such as Britain and the United States.^[39] In Work, Mobility, and Participation , Cole examines the legitimation of authority with a logic paralleling that of Bendix. In his comparison of Japanese and American employment practices, Cole argues that Japanese employers required a tradition of group loyalty to establish their system of permanent company employment. They used this ideology at the beginning of the century as a means of parrying workers' objections to lifetime dependence on a single employer. In the absence of a comparison with an

[38] Bendix conceived his work only with the aim of depicting the interdependencies between the development of legitimating cultures and the play of economic interest. Work and Authority in Industry (Berkeley: University of California Press, 1956), pp. 442–443. His selection of cases invites a host of economic explanations for differences between British and Russian factories, based on severe contrasts between their business and technical environments—such as the availability of skilled labor or reliance on a permanent urban work force. Bendix himself lends credence to conventional economic deductions about the pattern of industrial relations when he acknowledges that the Russian company towns, from which he draws much of his evidence, closely resembled communities the British employers established under similar economic circumstances in isolated villages of dependent colonies (p. 183 note).

[39] See Ronald Dore, British Factory, Japanese Factory (Berkeley: University of California Press, 1973), pp. 401–402. Dore's original study constructed a dual set of arguments to explain the differences between British and Japanese forms of industrial enterprise in the electrical industry. On the one hand, it attributed their differences in work relations and methods of supervision to dissimilarities in the countries' cultural emphases upon individual versus group achievement, to variation in familial values, and to the extent of popular acceptance of the authority exercised by economic elites. On the other hand, Dore connected the differences in work organization to contrasts in the world-economic epochs from which they emerged. By his account, late-industrializing Japan institutionalized its factory practices under economic circumstances that objectively favored the initiation of its lifetime employment systems, for example. After noting some of the congruences between his two lines of causal reasoning, Dore declined to evaluate the relative weight or respective usefulness of the cultural and economic modes of analysis. The same comparative strategy and indecision between these two lines of argument recurs in his more recent book, Taking Japan Seriously (London: Athlone Press, 1987), pp. vii, 94–95.

economically similar case, however, even this shrewd view of culture as a legitimizing tool cannot judge whether the institutions would have evolved differently had it not been for the stock of traditions. From Cole's evidence an analyst could conclude just as readily that the needs of capital reinvigorated and sustained an older ideology as that the ideology steered the direction of institutional development.^[40]

Since publication of these benchmark comparative studies, advances in comparative description of manufacturing institutions have not taken on the task of delineating with precision culture's separate, systematic effect in the development of workplace organization.^[41] The research team of Gary Hamilton, Nicole Woolsey Biggart, and Marco Orrù has forcefully exhibited distinctive national patterns of organizing the financing of manufacture and the exchange of goods among concerns in Japanese, Taiwanese, and South Korean enterprises.^[42] But such recent studies have not resolved the

[40] Robert Cole, Work, Mobility, and Participation: A Comparative Study of American and Japanese Industry (Berkeley: University of California Press, 1979), Chapter One. A more recent study by Shojiro Ujihara likewise outlines the economic preconditions for the vivification of paternalistic tradition. Shojiro Ujihara, "Essai sur la transformation historique des pratiques d'emploi et des relations professionnelles au Japan," Sociologie du travail Volume 33, Number 1 (1991), p. 23. Japanese scholars have also stressed the extraordinary economic conditions in Japan as a cause by itself of the development of a distinctive work ethic. Inagami Takeshi, "The Japanese Will to Work," in Daniel Okimoto and Thomas Rohlen, editors, Inside the Japanese System (Stanford: Stanford University Press, 1988), p. 33.

[41] For a global survey of cultural findings, however, see Raghu Nath, editor, Comparative Management: A Regional View (Cambridge, Massachusetts: Ballinger Publishing Co., 1988). Marc Maurice recently outlined an admirably sophisticated strategy of cross-national comparison. Maurice seeks to identify a national influence upon the construction of work practices but defines this effect primarily in terms of the context of societal institutions for certification and training. Marc Maurice, "Méthode comparative et analyse sociétale: Les Implications théoriques des comparisons internationales," in Sociologie du travail Volume 31, Number 2 (1989), p. 184. The preliminary results of his research agenda appear in his The Social Foundations of Industrial Power (Cambridge: M.I.T. Press, 1986). A more recent examination of the effect of educational systems on workplace organization, following Maurice's lead, appears in François Eyraud and Frédérique Rychener, "A Societal Analysis of New Technologies," in Peter Grootings, editor, Technology and Work: East-West Comparison (London: Croom Helm, 1986), pp. 209–230.

Howard Kimeldorf's comparison of the dock workers' union movements on the East versus the West Coast of the United States is founded on similar work institutions but contrasting economic settings, given the differences in the supply of labor. Reds or Rackets? (Berkeley: University of California Press, 1988), p. 39.

[42] The authors acknowledge that their selection of countries for comparison is not designed to discount or to isolate the effects of contrasting economic trajectories or of differing industrial specialization upon the crystallization of organizational patterns in each society. Hamilton, Biggart, and Orrù, "Organizational Isomorphism in East Asia," op. cit., p. 365. A. Budde and Geert Hofstede have both measured important national differences in the exercise of authority and in the principles of interpersonal conduct in formal organizations. A. Budde, J. Child, A. Francis, and A. Kieser, "Corporate Goals, Managerial Objectives, andOrganizational Structure in British and West German Companies," Organization Studies , Volume 3 (Berlin: W. de Gruyter, 1982), pp. 1–32; Geert Hofstede, Culture's Consequences: International Differences in Work-Related Values (Beverly Hills: Sage Publications, 1980).

question that Michel Crozier's brilliant study of French organizational culture, The Bureaucratic Phenomenon , brought to light three decades ago: how do we show that differences in culture do not merely reflect economic constraints upon the development of institutions but also shape that development? In France, Crozier emphasized, individuals' preference for avoiding face-to-face authority relations and their reticence about creating solidary peer groups correlated with an emphasis in French bureaucracies on the indirect exercise of highly centralized authority through impersonal rules. Whether the organizational structures adapted to the economic context and then created this distinctive culture of interpersonal interaction, or whether they reflected this culture from the start is a question Crozier never tried to resolve.^[43]

Culture in Labor History

The key issues that must be resolved to specify the effective role of culture have been debated most sensitively in the fast-developing field of labor history. To illuminate the creation of new institutions of work and the development of workers' collective movements, labor historians have devoted increasing attention to the face of culture among both workers and employers. Yet in the main their strategies of research are not designed to respond adequately to the question addressed by this book: whether we can demonstrate and specify culture's independent effect upon the construction of factory practices.

The inextinguishable starting point for pondering culture's effect remains E. P. Thompson's The Making of the English Working Class. This work, which once served as a charter for cultural inquiries, demonstrated that workers did not acquire a shared class consciousness in early nineteenth-century Britain only in response to the degradation of labor and the rise of factories; workers also depended upon the peculiar legacy of Radical political discourse, carried originally by middle-class shopkeepers and small tradespeople.^[44] In The Making , the economy moved with a dynamic of its

[43] Michel Crozier, The Bureaucratic Phenomenon (Chicago: University of Chicago Press, 1964), pp. 220–224. Crozier emphasizes at some points that organizational structures adapt to economic circumstances. The pattern of bureaucracy in the United States, he claims, "corresponds to a large extent to the general evolution of industrial society." See pp. 232, 296.

[44] E. P. Thompson, The Making of the English Working Class (New York: Vintage Books, 1963), pp. 197–198.

own. It established the foundation of change to which workers responded. Culture—in this instance primarily meaning the legacy of political ideas—intervened to mediate workers' reactions to capitalist development. Thompson's argument rested on circumscription: he showed that new economic conditions, typified by the steam engine and textile mill, did not suffice to explain the emergence of class consciousness. Having limited the domain of economic explanation, he celebrated the mysterious indeterminacy of human "agency," for he believed it sufficient for his purpose that culture serve as an indispensable ingredient in workers' responses.

This approach in The Making , even if it served at moments only as a device for framing the narrative, has fallen to an objection in principle: it implicitly assumes that workers have an anterior experience of socioeconomic conditions to which popular culture and political discourse respond. The powerful critiques of Gareth Stedman Jones, Patrick Joyce, and Joan Scott have made it commonplace to emphasize instead that culture and language are constitutive of and, in this sense, prior to social and economic experience.^[45] From my perspective, Thompson's initial position offers an ineffective defense of the centrality of culture for a very different reason: it does not respond adequately to social investigators who doubt that culture can be called upon to develop rigorous explanatory arguments. In any sequence of change, the number of causes that are necessary for an outcome considered in all its concreteness is unlimited. The issue is not whether cultural components represent necessary ingredients, for almost everything is worthy of that designation; it is, rather, whether cultural elements have an independent and specifiable contribution apart from the influence of other factors. Do they carry a strong, systematic effect which justifies concentrating on them in their own right? Analysts who discount the prominence Thompson lent to culture may justifiably contend that if he probed economic or demographic variables more deeply, the indeterminacy in workers' responses, which he attributed to community culture, would taper off.

Of course, Thompson's own evidence implies that the economy becomes an historical force only as it enters into human experience. He shows that the earnings of the proud artisans, the prices of tools and bread in the countryside, and even wage differentials in the new mechanical

[45] Joan Wallach Scott, Gender and the Politics of History (New York: Columbia University Press, 1988); Patrick Joyce, Visions of the People: Industrial England and the Question of Class 1848–1914 (Cambridge: Cambridge University Press, 1991); Gareth Stedman Jones, Languages of Class (Cambridge: Cambridge University Press, 1983).

industries conformed to community expectations and notions of social honor.^[46] In this sense, the economy itself operated through cultural standards. But in this line of reasoning, too, culture appears as an ingredient whose independent, structuring influence is undemonstrated. The underlying forces of market and technological development might still carry the exclusive principles configuring social change or the form of stability; after all, the "moral economy" of the community eroded as required for the furtherance of capitalist development.^[47] As Thompson tells us, lofty artisanal standards suffered earthly degradation: "The form and extent of deterioration relates directly to the material conditions of the industry—the cost of raw materials—tools—the skill involved—conditions favouring or discouraging trade union organisation—the nature of the market."^[48] When custom survived, it might do so only as it was selectively appropriated and shaped as a resource by the active, selective logic of market and technological forces.^[49]

[46] Thompson, op. cit., pp. 235–237.

[47] In my view, Thompson's discerning portrayal of the "moral economy" to which crowds appealed at times of food shortage in the eighteenth century illustrates this possibility of the explanatory adequacy of market adaptation alone. The ideal of a "moral economy" was revived periodically so long as it fulfilled a strategic function: in times of crisis it facilitated price bargaining among the common people, the gentry, traders, and local authorities. At the beginning of the nineteenth century, the confluence of interests supporting the fiction of a moral economy disappeared. E. P. Thompson, "The Moral Economy of the English Crowd in the Eighteenth Century," Past & Present Number 50 (February 1971), pp. 126, 129. For a discussion of Thompson's more recent work, see p. 36, below. Food rioters requested in advance official permission to fix prices by riot, an occurrence which makes the riot appear as a controlled and institutionalized bargaining strategy. John G. Rule, "Some Social Aspects of the Cornish Industrial Revolution," in Roger Burt, editor, Industry and Society in the South-West (Exeter: University of Exeter, 1970), p. 93.

[48] Thompson, Making , op. cit., p. 258.

[49] Patrick Joyce's exemplary study of the paternalist regimes built upon the full-grown textile mills of the late nineteenth century illustrates the same theoretic issue. Joyce shows that workers in the textile communities of the north of England embraced factory life by identifying with their employers. Workers subscribed to folk stories about the family owners, shared membership in religious organizations with employers, and saw the mill and the collective celebrations it sponsored as the epitome of the community. The traditions of deference, religious association, and local attachment called into play for this accommodation in late Victorian Britain were invigorated and manipulated to suit the needs of capital. Despite the richness of his cultural portrait, Joyce's evidence in Work, Society and Politics could support the view that community culture had a coherence of its own while it remained subservient in practice to economic requirements. No wonder Michael Burawoy uses Joyce's evidence to emphasize the subjection of culture to the structure of the labor process. The Politics of Production (London: Verso, 1985), pp. 97–99. In his more recent examinations of nineteenth-century British workers' representations of the moral community, Joyce shows that popular concepts are not deducible from the logic of capitalism and do not "reflect" an anterior reality (Visions , op. cit., pp. 9, 333). Even if these symbols emerge through a distinctive discursivelineage and draw upon themes unrelated to the economic categories of capitalism, they may nonetheless be selected, maintained, and indirectly appropriated by the supposedly "instrumental" logic of the marketplace.

More recently, William Reddy has transformed the debate on culture's influence by tracing the development of market orientations themselves as cultural forms. In pathbreaking investigations focused upon French textile production, Reddy has demonstrated that dynamic networks of production and distribution in prerevolutionary France promoted the growth of the industry without a model of free market exchange. Only after the Great Revolution did the ideal of pure market transactions, promulgated initially by intellectual elites, gradually became part of economic agents' self-understanding.^[50] The new market model was unrealistic. It ignored overwhelming rigidities in the merchandizing of labor power, and it excluded the human interest in honor and autonomy which could not be extinguished in the production process. Yet the model became an effective prescription. It led employers to oversimplify points of contention with workers into plain monetary exchanges and thereby complicated the resolution of labor conflicts.^[51] Reddy's The Rise of Market Culture is profoundly subversive: rather than treating culture exclusively as a "tradition" separate from and opposed to the market, it turns the market regime itself into a cultural project.

The present study maintains Reddy's emphasis on the cultural construction of economic categories but fully historicizes these forms of practice and experience. In Reddy's narrative, at moments of crisis employers are forced to adopt the postulates of "market culture" to improve production. For example, to cope with mounting commercial challenges in the first half of the nineteenth century, they imagined that they appropriated, not simply a worker's output, but a labor service over which they claimed jurisdiction. When they imposed more exacting rate schedules on mule spinners to gauge labor effort, the design was allegedly determined simply by a need to exploit improved machinery.^[52] Reddy inadvertently offers a new cultural teleology: employers acquire market categories through a learning process, but in the end there is only one kind of market culture, and one definition of labor as a commodity, which they are destined to adopt.^[53] He collapses

[50] William Reddy, The Rise of Market Culture (Cambridge: Cambridge University Press, 1984), pp. 66–67. Reddy also stresses, however, that commercial change in the eighteenth century stimulated the development of the cultural model of market society (p. 61).

[51] Ibid., p. 324.

[52] Ibid., pp. 124, 213, 215.

[53] Ibid., p. 251: "Here, finally, a step was made toward paying for labor, rather than for its outcome" (p. 124). But, after all, the book does not bear the title "The Rise of a Market Culture." Reddy misleadingly portrays Adam Smith as a theorist whose premises about work-ers' conveyance of labor to employers are prototypical and universal to market models (pp. 65, 85).

market categories as real forms of experience and as schemata for the use of technology into the generic analytic model of "market society." As an issue of history, this effaces actual cross-national diversity in Europe; as a matter of theory, it reduces the explanatory power of culture. If we rest a cultural argument on the metathesis that the most general building blocks of market-industrial society, such as cost-accounting and the maximization of returns on investment, are cultural creations,^[54] this does not enable us to explain variation in realized capitalist practice.

The deciding question is not whether market conduct is culturally acquired and reproduced; the purest economic theorist is justified in ignoring this issue as a philosophical point about the origins of the "capitalist" system or its broadest parameters. The true issue of contention is whether cultural forms of explanation account for variation in historical outcomes on the shop floor better than alternative approaches do. In Reddy's narrative, "market culture" germinates as an intellectual project but disseminates out of practical necessity.^[55] From this viewpoint it is all too easy to rest the case for culture's importance upon the comfortable supposition that the most general parameters of conduct are culturally fixed, allowing historical narratives to present as adaptations to economic requirements the specific design of the institutions of work. The comparative strategy of the present study, by contrast, does not merely assert but demonstrates exactly how the cultural construction of economic concepts configured even inconspicuous parts of instrumental practice by symbolic principles that varied in this study's primary cases of Germany and Britain, as well as in Reddy's case, France.^[56]

Where, then, may we turn for the theoretical tools to handle such a case demonstration of culture's formative logic upon practice in the factory? In my view, the specification of culture's independent role in the capitalist labor process remains an open problem in contemporary social theory. The most promising theories on the scene that accept the challenge of demonstrating culture's effect, rather than (unconvincingly) taking its influence as an a priori necessity, conceive of culture as a practical schema for organizing

[54] Ibid., p. 70.

[55] See above, footnote 53, as well as ibid., p. 99.

[56] Rather than contrasting "market culture" with the pursuit of nonmonetary rewards or with less calculative varieties of economic enterprise prior to the rise of market culture, the present study contrasts different incarnations of the fundamental capitalist category of labor as a commodity.

activity.^[57] Within this general approach it may be helpful to group into three families the leading attempts to specify the influence of culture upon economic conduct. If we examine each in turn, we may clarify the conditions that must be met to demonstrate satisfactorily the independent, constitutive influence of culture upon the organization of practices at the point of production in capitalist society. Culture's influence is contested in social inquiry in part because the leading cultural theorists have not appreciated the challenge before them.

The Ambiguity of Practice Theory

There is, first, the school of analysis that I will call, for lack of a more widely employed term, cultural practice theory , in whose development Pierre Bourdieu has played a celebrated role. Although Bourdieu has scarcely applied his approach to the analysis of the capitalist factory, he has established a baseline for discussion of the symbolic and material dimensions of economic conduct. It is therefore incumbent upon me to suggest why his approach does not address the guiding question of this book—and why; perhaps, it should.

Bourdieu's work is intended to overcome the contest between cultural and purely utilitarian accounts of the development of social institutions which divides contemporary theory. The utilitarian approach, consecrated anew in the currently fashionable theories of "rational choice," would explain the visible social order as the outcome of the well-considered activity of individuals pursuing their interests as best they can.^[58] Of course, supporters of cultural approaches have long accepted the viewpoint that people conduct themselves as strategizing agents. But adherents of cultural forms of explanation insist that the process by which agents pursue their interests must be situated within a broader perspective upon the operation of human agency and reason. Before people set out in pursuit of their interests, they require an order of cultural symbols that establishes for them a relation to the world. The concepts on which agents rely to accomplish this are an historical product whose constitution and development follow a discipline of their own. Cultural forms of explanation need not exclude the play of

[57] Ann Swidler summarizes the reasons for the shift to practice-centered cultural theory in "Culture in Action: Symbols and Strategies," in American Sociological Review Volume 51, No. 2 (April 1986), pp. 273–286.

[58] By way of illustration, see James S. Coleman, Foundations of Social Theory (Cambridge, Massachusetts: Harvard University Press, 1990).

utilitarian calculation, but they are inclined to emphasize that collective concepts give shape to individuals' percepts.

Thus is initiated the cycle of debate between cultural and utilitarian varieties of social explanation. For just as culture can inaugurate the terms for the exercise of instrumental reason, so instrumental reason can establish the conditions for the development of culture. The rational choice theorist may admit that the horizon for agents' conduct is momentarily fixed for them by collective traditions. The question then becomes, what are the forces that lend such a system of shared insights and concepts its distinctive shape? Its formations, too, may follow from simple strategic logic, and its defining features may represent a convenient adaptation to the circumstances of action.^[59]

Bourdieu tries to overturn several of the distinctions on which this debate between cultural and purely utilitarian modes of explanation has been founded. Like rational choice theorists, he underscores the agents' unceasing manipulation of their symbolic and material environments. But he contends that agents' strategies are not purely means chosen for the pursuit of interests. The strategies are patterned by implicit principles governing perception and action that are transmitted to the agents by their prior life circumstances in society. The long-term acquisition of these skills enables the agents to compete against others, but in so doing the agents do not rationally follow preestablished interests. They are guided by implicit know-how, and they find themselves dedicated to the very practices through which the competition takes place.^[60]

[59] This is the vision of hard-headed anthropologists such as Marvin Harris, Cultural Materialism: The Struggle for a Science of Culture (New York: Vintage Books, 1980), pp. 56–58. The wide variety of ecological explanations are explored in John G. Kennedy and Robert Edgerton, editors, Culture and Ecology: Eclectic Perspectives (Washington, D.C.: American Anthropological Association, 1982). Political scientists, ever the philosophers of cynical reason, have also treated culture as a byproduct from the rational pursuit of self-advantage. They have emphasized that political entrepreneurs may manipulate forms of cultural identity in order to create and sustain political alliances between disparate groups. Their stress upon the malleability of culture in the pursuit of self-interest results in a less naturalistic appreciation of culture than that of reductionist anthropologists such as Harris. These political scientists do not treat culture as if it were an adaptation to the physical environment or to the true state of things-as-they-are, but their moral for human society remains the same: culture represents a dependent tool of utilitarian practice. See, illustratively, Abner Cohen, "Symbolic Strategies in Group Organisation," in his Two-Dimensional Man (Berkeley: University of California Press, 1974). For a survey of the application of rational choice theory to the formation of political identities, see David D. Laitin, Hegemony and Culture (Chicago: University of Chicago Press, 1986), pp. 99–102.

[60] Pierre Bourdieu, The Logic of Practice (Cambridge: Polity Press, 1990), p. 290.

Bourdieu's insistence that agents organize manufacturing and other kinds of practices in accordance with acquired schemata seems congenial to cultural forms of explanation. But he also suggests that these acquired schemata are "durably inculcated by objective conditions."^[61] The agents' accumulated know-how "organizes perception of the world and action in the world in accordance with the objective structures of a given state of the world."^[62] What room, then, does Bourdieu leave for the symbolic mediation of social conditions as agents acquire their social skills? Bourdieu adds a proviso that the agents' dispositions do not mechanically mirror social structures. Rather, the agents' prior locations in the social structure decide how they will appropriate and respond to structural conditions of the moment.^[63] The historical filtering of "objective" structures does not offer a positive theory for culture's systematic influence. What is more, since Bourdieu views culture as a creation of practice, he insists that it has only partial coherence as a system of meaning. To his mind, cultural principles exist only in the process of getting things done. Their operation appears fuzzy and inarticulable in the light of contemplative reason. Bourdieu's emphasis on culture's inextricability from the ongoing life of practice enjoins us against representing culture as an intellectually coherent structure with a systematic effect of its own.^[64]

Yet Bourdieu's refusal to define culture's own structural effects leads him in his histories to embrace economistic explanations that he denies in his theories. In Distinction , his wide-ranging investigation of contemporary tastes in France, Bourdieu takes care to show that the dispositions of persons in the working class appear to follow a popular logic of their own but actually reflect the force of economic necessity. He claims, for instance, that "it is possible to deduce popular tastes for the foods that are simultaneously the most 'filling' and most economical from the necessity of reproducing labor power at the lowest cost which is forced on the pro-

[61] Pierre Bourdieu, Outline of a Theory of Practice (Cambridge: Cambridge University Press, 1977), p. 77.

[62] Logic , p. 94. Likewise, see Pierre Bourdieu, "Scientific Field and Scientific Thought," in Comparative Study of Social Transformations Working Paper Number 32, University of Michigan, November, 1989, p. 92.

[63] Pierre Bourdieu and Loïc Wacquant, An Invitation to Reflexive Sociology (Chicago: University of Chicago Press, 1992), pp. 135–136.

[64] Bourdieu, Logic , pp. 90–91, 261–262. Neil J. Smelser discusses the implications of practice theory for the investigation of cultural systematicity in "Culture: Coherent or Incoherent," in Richard Münch and Neil J. Smelser, editors, Theory of Culture (Berkeley: University of California Press, 1992), p. 16.

letariat as its very definition."^[65] Such shifts to reductionist forms of explanation are probably unavoidable if Bourdieu wants to account for—rather than merely redescribe—social practices in contemporary societies. In his model, only economic or institutional circumstances (or the agents' transversal of such circumstances over time) offer a specifiable foundation for explanation. Culture is a marker, often misrecognized, of the true arrangement of things. It serves as a model of society, not as a model for society's creation.^[66]

In sum, Bourdieu's work grants culture a prominent but analytically dependent role. To be sure, in Distinction Bourdieu makes the survival of the capitalist system dependent upon culture's ability to mystify and legitimate inequality. But if culture serves as a conduit for the expression of economic power, it does not thereby gain independent influence upon the development of institutions or upon historic change.^[67] To address the question of whether culture donates a separate constitutive logic to the formation of institutions, we may still preserve one of Bourdieu's insights: namely, culture can be conceived provisionally as the schema that agents employ to orchestrate their instrumental strategies, rather than as a set of revered values.^[68] If the principles of a culture are thereby conditioned by the

[65] Pierre Bourdieu, Distinction: A Social Critique of the Judgment of Taste (Cambridge, Massachusetts: Harvard University Press, 1984), p. 177. Likewise, the culinary styles of other class strata develop from the commercial value of the food preparers' time and from the ethic of deferred or immediate gratification that is instilled by their occupational position.

[66] "On Symbolic Power" in his Language and Symbolic Power , edited by John Thompson (Cambridge: Polity Press, 1991), p. 169. Logic , p. 94; Outline , p. 77.

[67] In Distinction , Bourdieu's exclusive attention to France for illustrations of contemporary capitalist society is integral to his explanatory strategy: it sustains his maneuver of reducing the specificities of French cultural life to a revelation of the generic logic of mature capitalism. He can thus avoid pondering the character of the fit between French culture and the economy, of why certain symbolic goods and practices prevail in France, whereas different ones arise in countries with analytically similar economies. Bourdieu's cross-societal comparisons are limited to gross contrasts between traditional and modern social formations, which protects the implicit reduction of culture to social structure.

[68] What other research falls into the subdivision of cultural practice theory? Certainly Ann Swidler's presentation of a new research agenda in her incisive essay "Culture in Action." Swidler grants culture "an independent causal role" because it shapes agents' competencies at assembling enduring strategies of action. Unlike Bourdieu, Swidler does not insist that culture corresponds to the objective institutional environment. But neither does she demonstrate that different cultural complexes can survive in similar structural settings. To the contrary, she emphasizes that institutions distribute resources to decide which cultural ideas will be sustained in the body social. What is more, her definition of culture as a set of skills separates the effectivity of culture from the conventions of an autonomous system of signs. In her view, the capacities that comprise culture are learned rules of purposive action. As for Bourdieu, so for Swidler it is not essential to focus on the mediation of the environment by a set of symbols: the cultural competencies may represent a common-sensical correlate to the surroundings, asis demonstrated in Swidler's exposition of the culture of poverty. For these reasons, her framework makes it difficult to isolate the independent influence of a symbolic schema upon instrumental conduct (p. 275). Sherry Ortner's initial formulation of practice theory also emphasized this approach's relative disinterest in isolating culture as an analytically separable domain of social life. Sherry B. Ortner, "Theory in Anthropology Since the Sixties," Comparative Studies in Society and History Volume 26, Number 1 (January 1984), p. 148. Her recent work, however, returns to the issue of culture's own identifiable influence upon action. See "Patterns of History: Cultural Schemas in the Foundings of Sherpa Religious Institutions," in Emiko Ohnuki-Tierney, editor, Culture Through Time: Anthropological Approaches (Stanford: Stanford University Press, 1990).

ongoing logic of practice, we can still employ a comparison to search for the means by which culture partially shapes practice into a consistently meaningful structure. If a degree of cultural coherence obtains, it must be identified initially by comparing practices themselves—in our case, everyday solutions to similar manufacturing challenges—rather than by comparing discourse about practice.

Taxonomies of Production

A second family of theory that attempts to integrate the meaningful and pragmatic dimensions of economic life while preserving the causal autonomy of culture is that of cultural structuralism. The proponents of this approach share an emphasis on culture as a set of signs whose meanings are fixed only differentially, that is, by their relations to all other signs in a hypothesized system. Starting with the premise that meaning inheres in a coherent, overarching structure of signs, these practitioners tend for the sake of analysis to abstract culture from its contextual uses and to think of it as a formal, systematically interrelated series of terms. Marshall Sahlins is among the distinguished investigators who have applied this specification of culture to the analysis of economic institutions.^[69]

In Sahlins's Culture and Practical Reason , culture appears to intervene in the same mode in capitalist as in kinship-based social orders: out of an inchoate environment it creates for agents a meaningful order. Only the content of the cultural forms and the site for the invention of the integrative forms of culture—commercial production versus kinship—appear to vary between these social orders.^[70] Yet in practice, Sahlins does not faithfully

[69] Mary Douglas is also a member of this family of investigators, although her extensive publications reveal important differences from Sahlins. Douglas emphasizes that the classificatory principles that make up a culture reflect social morphology, but she lends culture a measure of autonomy by insisting that diverse symbolic schemata suit the same social structure. Implicit Meanings (London: Routledge & Kegan Paul, 1975), p. 314.

[70] "The uniqueness of bourgeois society consists not in the fact that the economic system escapes symbolic determination, but that the economic symbolism is structurally determin-ing." Marshall Sahlins, Culture and Practical Reason (Chicago: University of Chicago Press, 1976), p. 211.

apply his theory to capitalist manufacture. The deviation becomes apparent if we compare his dissection of economies integrated by kinship with those held together by the cash nexus. In traditional Moalan civilization in Fiji, Sahlins shows, the production process itself incarnates an overarching cultural scheme founded on a series of isomorphic, binary distinctions: land/sea, inside/outside, female/male. For instance, men busy themselves in the extremities of the high seas and the distant bush, whereas women work in the interior lagoon and within the village. By the same scheme, villages where the people are designated as belonging to a "land" group do not angle even when they have access to fishing grounds. These binary distinctions governing production are reiterated in the codes of governance, domestic furnishings, and myth.^[71]

The logic by which Sahlins demonstrates the economy's dependence upon a symbolic order changes, however, when he turns to "market-industrial society." Here he applies the interrelated distinctions of his formal schemata only to items of consumption such as food and clothing. No application of Saussurean principle emerges for the living execution of production itself. The focus now is on the operation of culture at a remove, as the agents' application of a cultural code identifies for them the kinds of goods worth manufacturing.^[72] We observe a cultural structure for the production process, not in it. In Sahlins's portrayal of the kinship-based society, the divisions of the symbolic order constitute, not just perceptions of production or of goods, nor merely the distribution of particular agents among economic roles, but—the very methods and organization of the production process. What generates this shift in the way Sahlins attempts to demonstrate culture's effect? Is it attributable merely to the misapplication of an adequate theory? Or is capitalist production resistant in principle to this variety of cultural analysis?

[71] Ibid., pp. 38–41.

[72] The conceit that Western society alone is constructed pragmatically arises from ignorance of the symbolic determination of the concrete ends of production: "Because it appears to the producer as a quest for pecuniary gain, and to the consumer as the acquisition of 'useful' goods, the basic symbolic character of the process goes on entirely behind the backs of the participants—and usually of economists as well, insofar as the meaningful structure of demand is an exogenous 'given' of their analyses. The differentiation of symbolic value is mystified as the appropriation of exchange-value." At this moment of theoretic synthesis, the symbolic determination of the instrumentalities of production, given a cultural selection of the goods worth producing, recedes from Sahlins's view. Ibid., p. 213.

For one question immediately arises: once capitalist manufacture takes account of the cultural valuation of goods, may it follow an unmediated economic logic in their production? In his more recent studies of Polynesians' contact with Europeans, Sahlins emphasizes that the chiefs' demand for finery by which they could denote their mana obviously helped articulate the Polynesian and European economic systems. The demand for particular European goods and therefore their prices were set by native conceptions of mana.^[73] Yet this acknowledgment of culture's presence does not maintain its explanatory significance. It is theoretically deficient for social investigators who underscore the importance of culture to contend only that culture comprises a necessary ingredient in reconstructing a concrete historical situation.^[74] As we know, in the recreation of a sequence of change the number of components whose presence is indispensable for the outcome is inexhaustible if one attempts to appreciate events in all their concreteness. Culture may still operate as a necessary element in a course of change whose fundamental, underlying logic is that of a pristine "capitalist mode of production."

If we oversimplify the task of dissecting the independent contribution of culture to capitalist practice, we subvert the enterprise by making its accomplishment trivial. Even if an investigator must refer to the categorical distinctions of a culture in order to explain agents' conduct, this set of distinctions may change principally in response to the active, directing logic of market and technological pressures. It is not enough to contend that the distinctions themselves can neither register directly nor reflect the bare material logic of economic circumstance. For the forces driving—and necessitating—the redefinition and realignment of cultural categories may still be those of economic imperatives, however much the culture registers the

[73] Islands of History (Chicago: University of Chicago Press, 1985), pp. 155–156. "To the extent, then, of the Hawaiian market, the European mode of production and trade in the 1820s was organized by the Polynesian conception of mana." Historical Metaphors and Mythical Realities (Ann Arbor: University of Michigan Press, 1981), p. 31.

Sahlins's shift from culture at the site of production in kin-based societies to culture in the sphere of exchange in capitalist societies is typical of interpretive analysis. In parallel fashion, Mark Gottdiener has asked whether "ideological mechanisms exist at the foundation of capitalist processes or if they are produced only secondarily by the relations of production to promote accumulation." Gottdiener intends to show how ideology constitutes capitalism and does not simply grow out of it. But he focuses only on consumption, the circulation of capital, and the reproduction of labor power outside work—not upon the site of manufacture itself. "Ökonomie, Ideologie und Semiotik," in Zeitschrift für Semiotik Volume 10, Numbers 1–2 (1988), pp. 21–23.

[74] Sahlins is not the first to locate the effect of culture in the outputs for consumption. See Mary Douglas, The World of Goods (New York: Basic Books, 1979).

changes after the fact in modified terms. Few social theorists care to assert that culture has no influence whatsoever upon the concrete events of history. Just as the cultural theorist may accept the actuality (though not the unmediated presence) of material constraint and instrumental adaptation in history, so the rational choice theorist may acknowledge that the cognition of the maximizing agent depends in part upon the orienting assumptions of a culture. Although Sahlins's solution is inadequate for the capitalist production process, he has correctly posed the challenge: the issue is not whether culture represents a social force but whether it bears a constitutive and identifiable logic of its own.^[75]

So long as culture is separated from the construction of capitalist manufacture, it surfaces as a decorative frill on the fabric of capitalist development. For if the categorical distinctions of a culture are employed merely to distinguish worthy goods or to assess the course of events, culture recedes to the sphere of contemplation. Ought we to suppose that the techniques of machino-facture in the capitalist factory are subordinated to an elegant system of conceptual correspondences such as guides the ritualized procedures of the Moalan economy? The thinking subject under capitalism may impose and reimpose such a comprehensive order upon the free and expressive realm of consumption—and in theory could even draw upon a prior experience of production to do so. But the practitioners of cultural structuralism have not yet shown in a particular setting how the ever-changing, efficiency-driven practices of the capitalist workplace itself correspond to an overarching taxonomy of reiterated classifications. The obstacles to doing so are ones of principle.

In capitalist societies integrated by the mechanisms of commodity exchange, the agents in the labor process have no need to ensure that the categories employed to execute work and to organize social relations at the point of production are aligned with those in other institutional domains. Nor does such a requirement arise at the level of the collectivity. If, as in Moalan society, the institutions of agriculture and manufacture can be contemplated as the fulfillment of a global cultural design, not unlike an intentionally created piece of art, then one can both conceive of culture as a constellation of interrelated categories and see it as constitutive of production. In this setting the process of executing practice also calls that globally integrated culture into everyday experience. But this felicitous application of structuralist theory results from its coincidental

[75] Culture and Practical Reason, p. 14.

fit with the operation of kinship-based societies. Social reproduction in kinship-based societies depends upon an integrative, overarching cultural schema: or, more exactly, the isomorphism of symbolic relations is a concomitant of the reproduction of institutions that are all centered on kinship relations.^[76]

In a society integrated by market relations, by contrast, organizations fulfilling diverse tasks will articulate with one another not by taxonomic design but by commercial function. It causes no surprise, then, that on the single occasion when Sahlins refers to the intervention of culture into the capitalist labor process, he offers only a dissipated form of his characteristic argument: the general economic conditions under which production will proceed always leave meaningful particularities of the manufacturing process unspecified. "An industrial technology in itself," Sahlins contends, "does not dictate whether it will be run by men or women, in the day or at night, by wage laborers or by collective owners, on Tuesday or Sunday, for a profit or for a livelihood."^[77] These specificities make of production a realized human endeavor. At this moment, culture illuminates the residual: whatever objective constraints, rational choice, or conventional economic logic leaves undetermined. A system of binary distinctions no longer constitutes the basic form of practice.^[78] Advocates of utilitarian explanation could justifiably claim that Sahlins's exposition at this juncture sets up an artificial contrast between the generic and the particular: who would suppose that all the details of production could be explained by introducing only the most basic of economic constraints? By this contrary line of reasoning, the persistence of a residuum is only to be expected, for the conditions for responding to the environment, the appropriation of cultural categories for utilitarian purpose, and the opportunities for profitable adaptation have not been filled in with sufficient detail. Even accepting that inexplicable details are always left

[76] Ibid., pp. 39, 41. Similarly, it is no accident that Bourdieu undertakes a study of the schemata used in the production process only for "archaic" societies. For capitalist society, he, like Sahlins, shifts the study of categorical oppositions to the domain of consumption (in Distinction ) or of personal outlooks upon practice (in Pierre Bourdieu and M. de Saint Martin, "Le Patronat," Actes de la recherche en sciences sociales Number 20–21 [March-April 1978]), but abandons analysis of the form of concrete manufacturing procedures themselves. Cf. Logic , op. cit., pp. 214–217; "Le Patronat," e.g., p. 56.

[77] Culture and Practical Reason , p. 208.

[78] By Sahlins's own reasoning, the relations between oppositions must be outlined to identify culture in the first place. "It is not that, as some have believed, we have a 'need' to classify. Formal classification is an intrinsic condition of symbolic action." Islands of History , op. cit., p. 146.

over, however, small ground remains for attributing to culture the consistent and identifiable effect of a structure or of making it the guiding object of study. For the residuum is precisely that—a haphazard deposit, rather than an ordered constellation.

Cultural structuralism cannot be moved from its original home in kin-based society to the capitalist labor process. If culture is conceived as a global system of categorical distinctions, it cannot order and organize the capitalist labor-process from within. To be sure, interrelated sets of categorical oppositions can be discovered in the agents' outlooks, which in turn influence the way agents interpret the operation of the production system. But can we take this as a starting point for the ultimate destination, that is, for elucidating culture's systematic constitution of material practices at the point of production?

The masterful work of Paul Willis inadvertently exposes this path as a blind alley. In his classic study Learning to Labor , Willis discovers the classificatory distinctions that organize young working-class men's perception of jobs in Britain. These workers seek their place in the world of labor through a series of parallel cognitive oppositions: manual versus mental exertion; free versus conformist activity; productive versus impotent work; and masculine versus feminine actions.^[79] The lads see manual labor as a realm of independence, since it leaves their thoughts unsubordinated, and they esteem it as a demonstration of their manliness. Willis concludes that the culture acquired by the boys in rebellion against the intellectual rigors of school also generates their tragic commitment to a life of toil. If culture thereby serves as a conduit for the reproduction of practices in the capitalist workplace, it does not structure them by its own logic. According to Willis, the commemoration of physical exertion infuses production with an imported meaning that is, in his words, "no part of its intrinsic nature." Willis's emphasis on culture as a superimposition allows him to see it as a distinct component of social reality. It also limits culture's effectivity. The lads' indifference to the particular manual occupation they find and their assumption that work itself is meaningless apart from the detached attitude they adopt in its execution adapt them to any given work environment. Culture is identified merely with the subjective response and adaptation to already given institutions of work.^[80]

[79] Willis calls these parallel oppositions "sets of divisions." Learning to Labor (New York: Columbia University Press, 1977), p. 171.

[80] Quotation from ibid., p. 150. Willis's positioning of culture resulted from theoretic choice rather than from the thematic commitment of an opus that bridges school and factory.This is demonstrated in his essay devoted solely to the site of production, Human Experience and Material Production: The Culture of the Shop Floor (Birmingham: Centre for Contemporary Cultural Studies, 1975), pp. 1–3. Here, too, Willis associates culture with the "experiential conditions" of manufacture rather than with the structuring of practice. In the abstract, of course, culture always plays a "role in maintaining the conditions for continued material production in the capitalist mode"—for it sustains the living capacity to work (Learning to Labor , p. 171). The parallel distinctions of working-class culture—such as manual/mental, productive/impotent—are not deducible from the logic of capitalism, but they may be used and held in place by that logic all the same. If culture serves as capitalism's transmission belt, even imperfectly, it becomes an ancillary component, not an independent principle in its constitution. Of course one can maintain a hope for radical social transformation by romanticizing the uncertainty of cultural reproduction and by recalling at the last moment its reliance on the human element, as Willis does. But should the capitalist structures fail to reproduce, this may be attributable to their own disarticulation—or even to indeterminacy—rather than to the order of culture by default. Contrast with Learning to Labor , pp. 171–172, 176.

If the general approach of cultural structuralism fails to demonstrate the cultural constitution of practice at the site of capitalist manufacture, it nonetheless offers important resources for this task. In contrast to the practice theorists, the cultural structuralists do not center their theory upon the confrontation between a singular agent and structural conditions, but upon the invention and use of a cultural schema as an accomplishment of a collectivity. By making the creation of meaning dependent upon the conventionally understood distinctions that stand above the immediate context for action, the cultural structuralists block the reduction of culture to social structure. Yet in their hands this very separation isolates culture and thereby makes it difficult to identify its formative effect. The challenge remains of reattaching culture to the living execution of work while preserving its autonomy. The present study is not engineered to show how diverse concepts articulate and form a kind of architecture in the agents' minds; rather, it is designed to show through comparison that a single concept, that of labor as a commodity, was consistently incarnated in a field of practice.

Practice and Subjective Meaning

In classic sociological theory the paradigmatic demonstration of culture's impact on production is Max Weber's The Protestant Ethic and the Spirit of Capitalism. This work founded a third major family of cultural studies of economic practice, one centered upon individuals' understandings of the meaning of their social action. When the cultural structuralists made good on their promise to bridge culture and the shape of institutions in kinship-based societies, they did so by investigating the cultural template embodied

in the very form of those institutions. By contrast, Weber sometimes emphasized the disjuncture between culture and organizational forms.

When Weber traced the influence of ascetic Protestantism upon the early development of the Western variety of capitalism, he emphasized that the understanding of the world's significance that Protestant doctrine lent its adherents was subjective, in the sense that agents located in business ventures with the same organizational structure could be motivated by decisively different outlooks about the final import of their actions. This premise guides Weber's comparisons. For example, after Weber examined the enterprises that evolved in Renaissance Italy, he concluded that ascetic Protestantism was not a necessary precondition for the development of "capitalist forms of business organization."^[81] Prior to the Reformation, entrepreneurs in Italy established the same capitalist business associations and procedures as their dour successors to the north. Therefore the advent of stern Protestantism could not have been necessary for changes in commercial organization. Neither was it sufficient. The rise of this-worldly asceticism may have given its adherents a novel appreciation of the ultimate significance of their actions, but it did not as a matter of course lead them to build new institutional structures. In Weber's depiction, entrepreneurs who ran the putting-out system for the weaving trade and whose conduct was motivated by the Calvinist world view did not have to reshape the organization of their enterprises. They reduced turnover time and introduced a more dynamic pace to the putting-out networks, yet they might well leave the forms of that system intact.^[82] Therefore Weber's own evidence identifies a difficulty: an individual's interior assessment of the personally most significant implication of economic practices does not specify with precision how those practices should be outwardly organized. To distinguish the causal influence of culture upon the form of the collective practices of production, it is simpler and more effective to consider the publicly discernible principles immediately organizing their execution.

No wonder theorists who seek to specify and illustrate culture's contribution to social organization have turned away from Weber's concentration on subjective meaning and the agents' evaluation of the ultimate significance of their own existence. As we have seen, theorists have converged

[81] Max Weber, "Kritische Bemerkungen zu den vorstehenden 'Kritischen Beiträgen,' " in his Die protestantische Ethik II , edited by Johannes Winckelmann (Gütersloh: Gerd Mohn, 1982), p. 28.

[82] The Protestant Ethic and the Spirit of Capitalism (New York: Charles Scribner's Sons, 1958), p. 67.

from different directions onto a conception of culture as the useful set of categories and distinctions on which people rely to organize instrumental practice. Jürgen Habermas has attempted to combine this appreciation of culture with Weber's original concern for analyzing forms of social action and social order in terms of the fundamental types of subjective orientations adopted by the agents. Habermas's imposing philosophy provides a touchstone for metatheoretical reflections upon the relation between culture and capitalist production. Can it provide the tools not just for conceiving but for convincingly isolating culture's historical effects?

In The Theory of Communicative Action Habermas uses the term culture to indicate "the stock of knowledge from which participants in communication supply themselves with interpretations as they come to an understanding about something in the world."^[83] His choice of metaphor—here, culture as a reserve—is precisely motivated. In his view, agents rely upon culture as an inexhaustible and inescapable store of background assumptions that allow them to familiarize themselves with the world. From this starting point, Habermas assigns culture a more powerful inaugurative force than the representatives of practice theory and structuralist theory grant it. He does not think of culture as the set of schemata people employ to order both their conduct and the elements of the world. Rather, culture comprises the surrounding environment of implicit, undissectable assumptions that allow people to employ such schemata.^[84]

From Habermas's perspective, instrumental action in the workplace is just as embedded in this indiscernible knowledge as is ritual action at a place of worship.^[85] Since he identifies culture with the linguistic resources that erect a stage for action, rather than with the principles that the agents follow in executing their action, Habermas avoids the challenge of specifying the contributions of culture to variation in the relations of production. The influence of culture becomes problematic when he emphasizes that the

[83] The Theory of Communicative Action (Boston: Beacon Press, 1987), Volume Two, p. 138.

[84] "Language and culture neither coincide with the formal world concepts by means of which participants in communication together define their situations, nor do they appear as something innerworldly. Language and culture are constitutive for the lifeworld itself." Ibid., pp. 124–125.

[85] Culture plays an equally formative role whether the agents are engaging in purposive-rational action—based on the pursuit of specified goals by the most economic and efficient means—or communicative action, which seeks mutual agreement about binding norms. Theory and Practice (Boston: Beacon Press, 1974), pp. 158–159. Habermas's most recent formulations respond to the criticism that he separates symbolic communication from work. Cf. John Keane, "Work and Interaction in Habermas," Arena Number 38 (1975), p. 65.

mechanisms of the market and of bureaucratic administration have replaced communicative action as the integrative principle of society. The symbols people use in their lifeworlds are subordinated to the functional requisites of the marketplace or to other mechanisms for the impersonal exercise of power. In this instance, cultural resources are constitutive of the lifeworld only in the sense that they represent prerequisites on the level of metatheory for the construction of social institutions; they no longer subject the particular shape assumed by those institutions to a determinant cultural logic.

The disadvantage of Habermas's approach for our purpose is that it defines the role of culture in the organization of production by philosophical decree. It does not contest the predictive power of so-called rational choice theory. It disputes only the conceit that this theory truthfully stipulates the foundations for the agents' execution of their strategies. Surely we must seek a framework that does not simply postulate an overarching cultural dimension for all forms of practice, as Habermas's does, but provides the tools for demonstrating culture's effect. Otherwise we will not be able to battle alternative explanations for institutional practices. Unless allegiance to a theoretical agenda can be debated and disputed in the light of research, the theories with the most commonsense resonance will command the field in much of social research. In this instance, rational choice theory, which suits the ruling power of commerce, will surely become the accepted coin of the realm.^[86]

Is it a mistake to let the abstract issues established by theoretical debate shape our own agenda for historical research? E. P. Thompson has suggested that it may be wiser to abandon attempts to make an analytic distinction between practices shaped purely by utilitarian economic strategies and those guided by culture.^[87] Thompson's final research into the operation of eighteenth-century law and moral economy in Britain points to the indissoluble concrescence of cultural practices and economic logic.^[88] After all, the categories of "the cultural" and "the economic" are imposed by the investigator, not inscribed in the nature of things. Why not acknowledge the palpable merging of their operation and carry on with the investigation of the

[86] Habermas does not see his role as that of providing a research agenda that will compete with established programs of theory. Op. cit., p. 375.

[87] E. P. Thompson, The Poverty of Theory and Other Essays (New York: Monthly Review Press, 1978), pp. 292–295.

[88] E. P. Thompson, Whigs and Hunters: The Origin of the Black Act (New York: Pantheon Books, 1975), pp. 261 ff.

concrete conditions under which the institutions of manufacture and trade developed?

This solution is deficient because, rather than transforming the field of theoretic debate, it quits the arena altogether. The grand thinkers who champion the self-sufficiency of economic reasoning will always proceed to formulate models founded on the pristine "laws" of necessity within a given mode of production or on individuals' elemental pursuit of self-advantage. These theorists need not assert that culture is absent or irrelevant—only that their hypothetical model captures those fundamental processes about which we can feasibly generalize. Historical investigators who in response draw a series of vignettes showing the intervention of culture cannot dislodge competing heuristic economic models. The historians' scenarios may identify the limitations of a deliberately simplified representation, but they do not displace it. No theory is abandoned in the face of limited discrepancies. No contest can develop as a two-sided struggle between a theory and the facts, only as a many-sided comparison between the evidence and rival theories.^[89] If the champions of cultural study wish to engage in debate about the causal intervention of culture, they must support culture as a separate analytic category and specify its structure and effects. Otherwise culture becomes a vague residual category—exemplified in Sahlins's localized retreat from a taxonomy for capitalist culture in the labor process—or, alternatively, culture becomes an undissectable background condition whose influence is pervasive in principle but undemonstrable in detail.^[90]

The clash between cultural and purely utilitarian forms of social explanation amounts to much more than adjudicating between academic hypotheses. It determines our understanding of how we make ourselves in history. If the cultural order is dismissed as a mere reflection of individuals' utilitarian adaptations to the environment, it is no longer a fateful realm of human contingency—and those who wish to understand and ultimately contribute to the project of our collective self-creation will have to look elsewhere. The comparative logic of this study shows that culture exercised an influence of its own but not completely by itself. The

[89] Of course this is the lesson of Imre Lakatos, "Falsification and the Methodology of Scientific Research Programmes," in Imre Lakatos and Alan Musgrave, editors, Criticism and the Growth of Knowledge (Cambridge: Cambridge University Press, 1970), p. 119.

[90] The culture/economy divide, haughtily dismissed by some, will have been transcended, if this is possible, only when parsimonious models that go beyond this opposition outdo their rivals in explaining, and not just commenting upon, a range of practical outcomes. All else is merely a play of words.

power of culture arose from its inscription in material practice, a finding which suggests that the influence of discursive struggles upon historical development, often taken for granted in cultural studies, is pivotal but conditional.

A Look Ahead

My approach, which will unfold in the course of this historical study, identifies culture in the order of practice at a single locus, the point of production, not in practices reaching across many institutional domains and not in an overarching world view encompassing social conduct at large. For, in contrast to many other cultural studies, mine does not begin with the philosophical premise that culture enabled people to create a meaningful order. Rather than take such imputed mental processes for granted, I proceed merely from a controlled comparison of outcomes: economic agents in Germany and Britain constructed different techniques for carrying out the same tasks of manufacture under similar business circumstances. These manufacturing procedures tended to be arranged in a consistent constellation around a nationally dominant specification of labor as a commodity. The cultural definition of labor as a commodity was communicated and reproduced, not through ideal symbols as such, but through the hallowed form of unobtrusive practices.

The comparison unfolds in three stages. Part One introduces the differences between the German and the British concepts of labor as a commodity and describes the diverse ways in which the instrumentalities of production communicated these ideas on the shop floor. Part Two considers the historical genesis of the divergences in the cultural specification of labor as a commodity in western Europe. In Part Three the emphasis shifts from settled institutions to movements for change. This last section traces the ways practice on the shop floor provided the templates for workers' understanding of exploitation and choice of tactics for resistance.

The selection of Germany and Britain as the primary countries for comparison grew out of the logic of finding economically parallel cases, but in the end the rewards of study come from the profound differences between the German and the British cultural heritages. This opportunity for comparison was not lost on Thorstein Veblen, who made use of it for his classic Anglo-German study Imperial Germany and the Industrial Revolution. The Germans, Veblen observed, lacked "three or four hundred years' experience" of the free play of individuals in commercial intercourse which the

British had gained through their earlier advance to a market economy.^[91] Veblen called attention to the Germans' unusual combination of feudal tradition and modern technology:

The chief distinctive characteristic of the German culture being a retarded adherence to certain mediaeval or sub-mediaeval habits of thought  . . . this variant of the Western civilization is evidently an exceptionally unstable, transitory, and in a sense unripe phase. Comprising, rather than combining certain archaic elements—e.g. its traditional penchant for Romantic metaphysics and feudalistic loyalty—together with some of the latest ramifications of mechanistic science and an untempered application of the machine industry,  . . . it makes for versatility and acceleration of change.^[92]

Veblen's language is antique, but his choice of cases for comparison remains as compelling as ever—and just as rich for renewed excavation. In the end we will be in a position to appreciate the wisdom behind Veblen's suggestive remarks: in comparison with British development, the merging of feudal and machine-age culture in Germany indeed created a potential for a fateful "acceleration of change."

[91] Thorstein Veblen, Imperial Germany and the Industrial Revolution (New York: Macmillan & Co., 1915), p. 90.

[92] Ibid., pp. 230–231. My excerpt merges two paragraphs from the original.