1. I have developed this theoretical approach at length in a work on the evolution of political culture in nineteenth-century Germany (Eder 1985). The continue

idea of social groups that select for competing cultural ideas and images t thus goes beyond both the classic approach of a mere history of political ideas and the theory of modernization.

2. For a parallel account of this discussion refer to Stephen Kalberg in this volume. Kalberg takes the most popular lines of this discussion and accounts for them by adopting a sociology of knowledge approach. That such arguments should arise within the German discussion might be due to the specific German cultural tradition. But this tradition is, typically, less monolithic and more open to discussion than many cultural sociologists (especially the Weberian ones) assume. Maybe Kalberg's perspective in this discussion should be incorporated in a sociology of knowledge analysis of specifically American ways of seeing German culture. Nonetheless, Kahlberg's contribution is important as an instance of self-reflection of sociological analyses and explanations. These become more important as more fields of analysis are ideologically and culturally at stake. For the German discussion of the Krise der Arbeitsgesellschaft see the collection of papers presented at the Meeting of the German Sociological Association 1982 in Bamberg (Matthes 1983). One of the organizers, Claus Offe, in fact argues against this notion.

3. This theoretical proposition draws on the actionalist sociology of Touraine (1977, 1985a, 1985b). The capacity for historical action is the key idea in his approach. It is used here as the starting point for reconstructing a sociological theory of crisis.

4. One of the appealing aspects of this specifically German discussion is the opportunity it provides to develop a new theoretical perspective on different strands of sociological theory and research within the fields of industrial relations and class analysis. For an overview of the broad range of discussions in this field see Pahl (1988). The German discussion also overcomes some other more nominalistic discussions about a postindustrial society and the new and fashionable discussion about postmodernity where nominalism seems to reemerge.

5. By treating the idea of a Protestant ethic as a model and distinguishing between it and the social practices that use it as a reference, I have sidestepped the discussion about the historical adequacy of Weber's theory. We can bypass any historical claim and simply start with the assumption that there is something called the Protestant ethic . To find out whether or not the elements attributed to it are there we have to look at group-specific practices in the work sphere. Which aspects of this ethic are found among business executives or workers is an empirical question.

6. Bell's (1973) and Touraine's (1969) earlier conception of a postindus-trial society starts by assuming the increasing primacy of scientific work over industrial productive work. The changes we can observe are merely exaggerated generalizations (Bühl 1984). For a reorientation of this discussion see Braverman (1974), Brandt (1981), Gershuny (1983), Benseler et al. (1982), Gorz (1983), and Bonß and Heinze (1984).

7. See Mickler (1981) and Kern and Schumann (1984). For an extensive account of the new type of dependent, but highly skilled, work see Hermann and Teschner (1980), Teschner and Hermann (1981), or Baethge (1983). These studies contradict the well-known analysis of modern work by Braver- soft

man (1974), who defends the theory of a continual downgrading of qualified work.

8. It is important to notice that this type of structural differentiation alters the traditional class structure of industrial societies. But instead of proclaiming the end of class society, I prefer the research strategy of looking for the class structure that has replaced the old one. There are enough indicators of social differences on the level of culture, as well as on the level of work, to justify such research. Bourdieu has been the foremost theoretician to follow this line of reasoning. For a discussion of his views and its implications for class theory and class structure see my contributions to a volume on Bourdieu's theory of class (Eder 1989a, 1989b).

9. See data comparing the 1980s with the 1950s: people start their occupational careers two years later (at age seventeen); they retire three years earlier (at fifty-eight). Given the condition of scarce work (Dahrendorf 1983), the politics of working time takes on a new importance beyond the function of psychic and physical relief for the workers. See also Offe et al. (1982).

10. The relationship between biography and time spent working has been examined by Brose (1983, 1984). The type of worker who works irregularly ( Gelegenheitsarbeiter ) seems to have become a possible model, especially among adolescents (Pialoux 1979). But the conclusion that the value attributed to work will therefore diminish has so far found neither a logical nor an empirical basis.

11. These survey data relate to the question of whether work is still a central life interest. For a classic statement of this type of research see Dubin (1955). In recent research, the increase in the number of people who respond negatively and are undecided is interpreted as a decline in work-related values, for example, achievement orientation. Because of the semantic instability of such questions over time and because of changes in income that have also made consumption a central life interest, this proposition was attacked with fatal consequences to the scientific reputation of its authors. It did not, by the way, hinder its success in public discourse! A good critique is to be found in Vollmer (1986).

12. This is not apparent as long as all work outside wage labor is ideologically interpreted as leisure (" Frei"-zeit ), as nonwork. But this ideology is effective, as seen in the demand for more leisure time. What really happens is a revaluation of work outside wage labor. See Rosenmayr and Kolland (1988).

13. Chronic unemployment disproves some of the central assumptions of theories on postindustrial society. The service sector--which was supposed to absorb a growing and highly skilled labor force--is no longer growing. The expansion of the service sector had already reached its limits before we even entered postindustrial society. Thus unemployment is a structural feature of both postindustrial society and industrial society. For this discussion see Mutz (1987).

14. This hypothesis has been taken by Kalberg (in this volume) as the central discussion point within the German crisis of the Arbeitsgesellschaft .

15. This relationship has been confirmed several times. See, in addition to Bourdieu (1984), the empirical research by Blossfeld (1983, 1985a, 1985b) and Windolf (1984). break

16. For a general discussion of the de-coupling debate see Kalberg in this volume. It is important to distinguish between its structural and cultural aspects: the de-coupling of work from status and the de-coupling of work from self-realization. Only the latter process seems to be specifically German. For self-realization in work is something that is constitutive for the Lutheran work ethic.

17. In my discussion of cultural changes in the sphere of work, I refer to some of the research in the area of value change insofar as it is work related. I also rely upon unpublished qualitative research (including some of my own) concerning the cultural orientations of workers whose occupational careers have been interrupted by phases of unemployment. For the latter see also Brose (1986). For a more general discussion of work-related culture see Habermas (1968), Honneth (1982), Bahrdt (1983), Clausen (1981), Jahoda (1983), and Garfinkel (1986). For the empirical discussion see Cherns (1980), Hostede (1980), Hoffman-Nowotny and Gehrmann (1984), Klipstein and Strümpel (1985), Strümpel and Scholz-Ligma (1988), and Pawlowsky (1986). For critical accounts see Reuband (1985, 1989) and Gehrmann (1986). The research upon which Noelle-Neumann relies and which has fueled the German discussion is the comparative research on "Jobs in the 1980s and 1990s" (Yankelovich et al. 1985). See also, as a reference point for interpreting changes in the work ethic, the more recent research comparing value orientations with respect to work in Europe (Harding, Philips & Fogarty 1986:150ff.).

18. Some historical studies are pertinent to the specific German tradition of old and new virtues. See, as an example of classic secondary virtues (secondary to the principles of an abstract morality or ethic), the work of Münch (1984). See also the section on Lutheranism in Kalberg (in this volume). Also relevant in this context is the discussion about the new "reflexive" identity formations that foster a kind of rational planning of the meaning of one's life by permanent rational reflection of one's biography. This new "habitus" is widespread among the new middle classes, as Bourdieu argues using French data. Qualitative analyses of such identity formations in Germany point in the same direction. See among others Oevermann (1985, 1988).

19. This discussion has received renewed attention in recent research. See above all Rosenmayr and Kolland (1988). For a critical discussion of supposed instrumentalism see Knapp (1981), and for the famous hypothesis of the dual relationship of workers to work, that there is simultaneously an instrumental and a substantive relation to wage labor, see Kern and Schumann (1980).

20. Hedonism applied to the work culture is one of the favorite targets of cultural pessimists. Hedonism is an instrumentalist work ethic that has spread from the proletariat to the middle classes. At least this is the interpretation offered by Noelle-Neumann (1980).

21. The literature taking up this discussion is increasing. Most interesting is the redirection from mere quantitative reasoning to qualitative analyses of the work ethic. Researchers are apparently trying to escape from the sterility and futility of those quantitatively oriented discussions where the data are not adequate to the questions. An important alternative, or complement, has become continue

the "biographical" approach to the culture of work. See, among many others, the volume edited by Brose (1986) that contains empirical as well as methodological and theoretical contributions to this critical turn in the analysis of work culture.

22. The relation between youth and change in the work culture is one of the most interesting contemporary research areas. Recent studies show that differences correlate strongly with gender. Another important variable is the development of a youth culture with value orientations not related to work. Whether a delayed socialization into a work culture hinders the adoption of a work ethic is not clear, although research shows it probably does not. For a recent critical discussion of the available longitudinal data see Brock and Otto-Brock (1988).

23. Such an alternative is, for example, the "cyclical theory of value change." For this new approach in the interpretation of value change see Namenwirth and Weber (1987) and further research by Bürklin (1988).

24. For an excellent discussion of the genesis of an inner-worldly work ethic within Protestantism, see Soeffner (1988). He makes it clear that the Lutheran (as well as the Calvinist) ethic is intended to control the work force rather than intrinsically to motivate workers. A possible conclusion is that the Lutheran and the Calvinist variants point to national differences that have influenced the social use of the Protestant ethic. This conclusion allows us to address the question of the specificity of the German debate of the Arbeitsgesellschaft within the modern work culture, and it relates to Kalberg's argument (in this volume), which tries to relativize its relevance.

25. I will not go into the validity of Weber's theory that the spirit of capitalism has something to do with Calvinism. For a recent discussion of the historical relevance of Weber's hypothesis, see Pellicani (1988), who concludes that the historical origins of what we call the Protestant ethic have nothing to do with Calvinism. But I do accept Weber's analysis of the Protestant ethic as the reconstruction of the model of a work ethic that has been constitutive for the development of capitalism. The model is based on the assumption of a highly internalized ascetic attitude toward work--on the acceptance of work without external force. There may have been social groups that were close to this model, but such a relationship is already socially specific and has to be explained by specifying the social position of the group in question.

26. In qualitative interviews (conducted in a research project by the Münchner Projektgruppe), these problems manifest themselves as the dominant ones, contrary to the evidence of those who look for the psychic stress of unemployment. This psychological approach blocks recognition of the implicit cultural codes that demand and structure the formulation of a consistent work biography even under stress. Unemployment creates an ambivalent experience for those who still rely upon the Catholic ethic. For it destroys access to the collectivist alternative characteristic of the Catholic work ethic; or rather, it individualizes it. Those who still adhere to some Calvinist element in their work ethic can therefore cope with unemployment more easily. This alternative is structurally open to the skilled workers, less so to the unskilled and semiskilled. break