Dual Labor Market Theory
Introduction
There have been a few attempts to explicitly investigate the relationship between the monopoly/competitive fractioning of the capitalist class and segmentation within the working class. It is efforts of this kind, rather than analyses which remain solely on the level of intra-class competition, which illustrate the full importance of the monopoly/competitive distinction. In fact, efforts to link capitalist and working class segmentation can potentially furnish us with powerful explanations for changes in the social structure of a concrete formation, as well as for the nature of the struggles which take place in it. We will review those theorists who attempt to understand these important linkages in the section immediately following the present section. But, before we do so, we wish to examine theories which have concentrated solely on explaining segmentation within the working class itself. These, we classify under the rubric of dual labor market theory.
Dual Labor Markets
The dual labor market tradition is one of the most recently formulated attempts to explain segmentation in the working class – in what we will call both economically determined fractions and politically and ideologically determined social categories (race and sex distinctions). Some writers view these segments as specific to the current phase of the capitalist mode of production and no attempt is made to exprapolate the specific segmentation back to previous times or across to different contemporary societies (See, for example, Doeringer and Piore, 1971). For others (Edwards, Gordon and Reich, 1975) the theory is informed by historical analysis of the process through which this segments have been produced and have achieved primacy.
The starting point of dual labor market theory is the observation that conditions of employment within the working class differ widely. The literature attempts to explain this observation theoretically by conceptualizing it in terms of two different job ‘sectors’ of markets grounded both in the economic system and in race/sex discrimination and job typing. Two aspects of a job and their interrelation as seen as important – (a) the employment conditions, e.g., wages, benefits, stability and (b) the labor market from which workers are selected into the jobs, e.g., the female sector, or the ghetto. The dual labor market literature relies on both of these aspects of job structure, sometimes alternatively, sometimes in conjunction, and often without being theoretically precise about the relationship of the two factors.
The motivation of this theoretical approach has come less from within the Marxist framework than from a reaction against how poverty and inequality are treated by orthodox economic theory. The dual labor market literature struggles to understand poverty, wage differentials and general economic inequality in terms other than individual qualities and capacities. This struggle has led to historical analysis of the economic system (Stone (1974), Marglin (1974), to an analysis of the role of the state under monopoly capitalism (Bluestone, 1970) and to the analysis of the different types of exchange relationship between employer and employee involved in different categories of jobs (e.g., wage competition versus job competition, see Thurow, 1975). Throughout, dual labor market theorists remain tied to the original problematic of economic and non-economic differentials between economic fractions and social categories of workers. This has been both a strength and a weakness. How social categories and class fractions overlap is of much theoretical and practical interest because the overlapping of these groups and interests has great potential for revolutionary change. Unfortunately, the dual labor market literature rarely deals explicitly with such groups as potential class actors. More frequently the analysis reverts to simply describing structural ‘immobilities’ whereby social categories become trapped in economic positions and therein become the ‘working poor’ or ‘super-exploited ghetto workers’ or ‘secondary’ workers.
Besides low wage the problem of job instability appears as the most consistently emphasized feature of the secondary sector. This is largely a reaction to earlier analyses focusing on unstable work habits of the poor. As Gordon says, “instability of the work force seemed not only to be accepted by employers in these job categories but often encourage by them…. There seemed to many to be a critical interdependence between the characteristics of ‘secondary’ workers and ‘secondary’ jobs” (1972:45). We can understand this analysis as indicative of a transition in the dual labor market theory tradition. Basically the dual labor market writers were reacting against individualistic interpretations of poverty by analyzing the interaction of worker and job. Later they will come to find job structures themselves interesting in their own right but will still retain their concern for the mapping of social categories into these differentiated job structures.
Gordon notes several problems which are apparent in the tradition as formulated to this point (1972: 50-51). First, job stability is forced to serve double duty both as an index of the distinction between labor markets and as a theoretical source of the differentiation. Second, the match between social categories and job sectors is not as straight forward and clean cut as is often implied in the literature and is often sloppily handled. Third, little historical research or analysis has accompanied the hypothesized historical process of segmentation between the two segments.
In an effort to remedy this lack of historical analysis Edwards, Gordon and Reich (1973, 1975) locate the genesis of dual labor markets at the point of transition between the competitive and monopoly phases of capitalism around the turn of the century. “During the preceding period of competitive capitalism, labor market developments pointed toward the progressive homogenization of the labor force, not toward segmentation…. The increasingly homogenous and proletarian character of the work force generated tensions which were manifest in the tremendous upswing in labor conflict that accompanied the emergence of monopoly capitalism… the new needs of monopoly capitalism, control were threatened by the consequences of homogenization and prolaterianization of the work force…. To meet these threat employers actively and consciously fostered labor market segmentation in order to ‘divide and conquer’ the labor force… (Reich, et al, 1973: 361).
The manipulation of the relations of production by the capitalist class to produce differentiated job sectors took on several forms. “The restructuring of the internal relations of the firm furthered labor market segmentation through the creation of segmented ‘internal labor markets’” (Reich et al, 1973: 362). Workers were also tied to specific enterprises by firm specific benefits and industry wide privileges. Industrial trade unions played into the hands of the capitalist class on this issue by operating only within industries for specific benefits and thus furthering job segmentation between industries. Finally, “employers also consciously manipulated ethnic antagonisms to achieve segmentation” (Reich, et al. 1973: 362). Sex typing of great numbers of job positions in contemporary capitalist production can easily be added to this list. The basic argument here is that “Labor market segmentation arose and is perpetuated because it is functional that is , it facilitates the operation of capitalist institutions” (Reich, et al, 1973: 364). In addition to conscious manipulation, systematic forces also promoted labor market segmentation.
The larger, more capital-intensive firms were generally sheltered by barriers to entry, enjoyed technological, market power and financial economies of scale and generated higher rates of profit and growth than their smaller, labor intensive competitive counterparts…. Along with the dualism in the industrial structure, there developed a corresponding dualism of working environments, wages and mobility patterns (Reich, et al. 1973: 363).
Reich, Gordon and Edwards have developed a believable theory of the generation and maintenance of labor market segmentation, located largely at the economic level but also bringing in ideological and political factors as mobilized by the capitalist class. They correctly noted that the capitalist class in the monopoly phase of capitalism is able to analyze and act on its class interests to a degree never before possible. This is achieved both through direct monopoly control of the production apparatus and through the monopoly capitalist’s control of the state. The capitalist class does not possess the same possibilities for class organization and capacities as the proletariat (except perhaps under Fascism) but, nevertheless, its organized class conscious action affect us all daily.
On the negative side, the historical analysis of Reich, et al. has the weakness of being one-sided in its emphasis and relies too heavily on this conspiratol theory of the origins of labor market segmentation. Their analysis of the systematic structural factors promoting labor market segmentation under monopoly capitalism is developed only weakly. In particular the connections between the structural developments in capitalism (the dynamics of accumulation and uneven development) and the effects of these developments on the structure of labor are underanalyzed. We must often assume causal mechanisms by coincidence in the absence of theoretically specified mechanisms of effect and determination. For example, the need for more work on the mutual determination between the technological and social relations of production is immediately apparent.
In this context we must emphasize that working class segmentation under advanced capitalism represents more than intentional capitalist class diffusion of worker resistance. Structural changes in the organization of production provided resources for class struggle not only for the capitalist class but for workers as well. Therefore, working class segmentation also reflects how the new forces of production induced by higher organic composition of capital provide workers certain capacities for class struggle. Worker interdependence, the lack of measurability of tasks, the need for workers to pass on their skills to new workers and to accept technological innovations all result from this higher organic composition of capital (concentrated in the monopoly sector) and contribute to differentiated resources for worker organization from which labor market segmentation partially results. Likewise, we must better understand the impact on the social relations of production of trade union movements oriented to the struggle for interests relizable within the framework of capitalist social relations.
I have seen that the dual market literature approaches the social totality with a very mixed bag of concepts. The most common focus is on unstable and undesirable jobs, often seen as being produced by a combination of industrial segmentation or differentiation and the mapping of certain social categories of individuals into these differentiated positions. The analysis takes place most typically at the level of monopoly capitalism as an abstract phase of the capitalist mode of production. In this phase ideologically and politically determined social categories are imperfectly mapped into economic class fractions (labor markets). This is an illuminating formulation but often left theoretically underspecified. Much remains to be done in the historical analysis of the forces producing this segmentation. The historical work so far has both illustrated the process whereby labor market segmentation came into being and at the same time has demonstrated the need to understand the sectors as determined by the dynamics of monopoly capitalism, and not as determined by remnants of earlier production relations.
On occasion writers in this tradition have mixed their analysis of abstract economic fractions and social categories with entities and issues whose existence is tied to a specific concrete social formation. Ghetto employment and the problem of mobility out of poverty belong to this latter category. The theoretical understanding of this latter issues does not belong on the same level of abstraction as the understanding of the class fractions and social categories of monopoly capitalism. More precision must be used in differentiating the various levels of abstraction involved in the analysis of split labor markets. Again, this problem results from the tradition’s early and persisting concern for the concrete problems of poverty and the working poor in the United States.
Perhaps the most telling criticism of the dual labor market literature is that it stops short of an analysis of the interests of the working class fractions and social categories out of which it builds its descriptions of labor market segmentation. These entities therefore appear to us as simply effects, never as potential class actors with class interests and class capacities. The convergence and/or contradiction of the immediate interests of these class segments with the more fundamental interests of the proletariat as a whole in a socialist society is the stage out of which class struggle arises. Without an understanding of these dynamics and conflicts the entire point of our analysis is lost.
The Articulation of Segments of Capital Within the Working Class
Introduction
If we are to achieve understanding of working class segments as actors in struggles which take place in concrete social formations, we must have some understanding not only of how segmented labor markets historically came to be, but of how such labor markets are today dynamically linked with the development of fractionings within the capitalist class. Only by linking the notion of the basic contradiction between labor and capital with the segmentations which are produced within each camp can we begin to get a feel for the class and class fraction interest which are potentially activated in struggle.
Capital and Working Class Fractions
Bluestone (1970, 19) and O’Connor (1973) have made explicit attempt to investigate and understand the manner in which working class fractions correspond to the segmentation of capital into monopoly and competitive elements. Bluestone attempts to describe the relationship of sectors of capitalism to sectors of the working class in what he calls tripartite economy. He distinguishes two industrial sectors and an irregular sector, and their corresponding working class fractions. For Bluestone, the two industrial sectors have become differentiated through a process of uneven development.
[The core economy is] entrenched in durable manufacturing, the construction trades, and to a lesser extent, the extraction industries, the firms in the core economy are noted for high productivity, high profits, intensive utilization of capital, high incidence of monopoly elements, and a high degree of unionization. What follows normally from such characteristics are high wages…. [The peripheral economy is] concentrated in agriculture, non-durable manufacturing, retail trade, and sub-professionals services, the peripheral industries are noted for their small firm size, labor intensity, low profits, low productivity, intencive produce market competition, lack of unionization, and consequently low wages. (Bluestone, 1970: 24).
A managerie of different conditions and factors define the two sectors which are, in a sense, constructed almost as idealtypes – few establishments fill all the conditions of one sector or the other. Thus, a series of dichotomous empirical differentiations of industries are used to define the outline of the two sectors of the economy.
To this part industrial division Bluestone adds a third sector the ‘irregular economy’ which is roughly identified as ‘ghetto employment’ day contract work, illegal economic activity, petty entrepreneurial efforts, hustleing and so on.[1] we believe the existence of such a sector clearly separate from the peripheral sector is questionable. The positive value of what Bluestone suggests concerning the irregular economy may not be so much in delineating a productive sector as in designating a group of workers permanently on the fringes of the working class and with perhaps the possibility of generating a force for revolutionary social change. Bluestone also indentifies the state as contributing to the process of labor market segmentation through two activities: (1) as a consumer of goods and services, the state leans heavily toward purchases from the core sector (e.g., aerospace and construction); (2) as a regulator of economic activity, the federal government has promoted segmentation by industry labor legislation, differential industry tax rates, tax loopholes, depreciation credits and so on.
Bluestone’s major contribution is his suggestion of the congruence between sectors of capital and labor markets with different relations of production. The industrial core sector corresponds to primary labor market employment; the competitive sector corresponds to secondary employment; the irregular economy corresponds to tertiary employment.
The major weakness in the points covered by Bluestone is the lack of theoretical linkages in understanding the process of sector differentiation. The Marxist theory of capital accumulation and the literature on uneven development are left relatively untapped. Bluestone provides an ideal-typical description of the three capital sectors but suggests little about their historical generation and their concrete manifestations in which sectors of capital may cross cut fractions of labor. Consequently, a second weakness is his forced one to one allignment of industrial sectors with labor markets. As we shall see, the fractioning of capital into core (monopoly) and peripheral (competitive) sector determines only a tendency towards the fractioning of the working class into primary and secondary positions. The fact is that primary and secondary positions exist in each of capitalist sectors.
O’Connor (1973) extends the analysis of the correspondence of working class fractions to segments of capital in two ways. First, like Bluestone, O’Connor elaborates three sector of the economy which, however, are different from Bluestone sectors. Second, O’Connor theoretically speculates about possible shape of the class struggle as it might emerge from the present structure of class fractions and capitalists sectors.
We believe O’Connor’s designation of state, monopoly and competitive sectors is a more accurate representation of the structure and dynamics of advanced capitalism than Bluestone’s designation of core, peripheral and irregular sectors.
For O’Connor the internal dynamics of each productive sector as well as the relationship between them determine the conditions which differentiate working class fractions as they correspond of these capitalist sectors. For example, in the monopoly sector “wages are relatively high (and)… the demand for labor is relatively stable and work is available on a full-time, year round basis” (15-16). A different set of working conditions correspond to the competitive sector of the capitalist economy. “unstable and irregular product markets and unstable and irregular labor markets go together in competitive industries. Employment in the competitive sector tends to be relatively low paid and causal, temporary, or seasonal. Workers who want and are unable to find full-time, year-round, well-paid work in the monopolistic or state sectors will accept employment in the competitive sector on almost any terms” (14). In the state sector wages are relatively high but have a ceiling lower than that of highly paid jobs in the private sector. This is because state sector wage depend on tax revenues whose collection and allocation is limited by political constraints. Stability of employment and labor immobility also characterize state sector employment in sharp contrast to competitive sector employment.
O’Connor elaborates the interrelations of classes, fractions, and sectors of capital. He spells out these interrelations in his discussion of how particular fractions of capital and labor would be aligned either for or against the development of a capitalist “social-industrial complex.” The social industrial complex “consists of the transformation of social expenses into social capital by mounting socioeconomic opportunities for monopoly capital and to ameliorate the material impoverishment of the surplus population.” (O’Connor, 1973: 221).
In our terminology, O’Connor provides a projection of how class fractions would be mapped into positions of class struggle on the immediate order to rationalize capitalism and preserve it from socialist revolution. While O’Connor would accept the abstract thesis – to be developed in a latter section – that all workers share a fundamental interest in the arrival of socialism, he suggests what actual alliances of fractions of classes would occur if workers and capitalists sought a solution short of socialism to the anarchism of the fiscal crisis of the state.
O’Connor speculates that the fractions allied in support of the state directed reform of capitalism (the social-industrial complex) are (1) sections of monopoly capital not presently wound up in the military industrial complex; (2) the surplus population – clients of the welfare state who would receive the majority of new jobs; (3) low-paid monopoly sector workers; and (4) state workers who would comprise “the small army of technologists, administrators paraprofessionals, factory and office workers, and others who plan, implement, and control the new programs in education, health, housing, science, and other spheres penetrated by social-industrial capital” (22). Aligned against this reform are those who have least to gain from the restructuring of the state’s productive and consumption priorities: (1) competitive capital; (2) capital comprising the military-industrial complex; (3) organized labor in both the competitive and monopoly-sector (who would gain few new jobs and no increase in wages but whose tax burden would increase to finance the reform).
It is not our purpose here to evaluate the validity of O’Connor’s speculation about the potential evolution of capital and the alignment of classes and class fractions. Rather, we wish to point out that O’Connor’s discussion of the social-industrial complex demonstrates a valuable mode of analysis. O’Connor elaborates not only abstract capital sectors and classes of the phase of state directed monopoly capitalism, but goes further and sets out how concrete fractions of capital and labor are aligned in pursuit of their immediate interests – interests which can be realized within capitalist relations – in a specific social formation. In addition, his analysis of the role of the state and of class conflict in a social formation recognizes and takes into account the political import of what we will term social categories such as the “surplus population of state clients.”
Conclusion
Thus, we have seen that a consideration of the two basic classes of the capitalist mode of production, that is labor and capital, along with analysis of the segmentations of these basic classes can potentially provide us with very useful tools for understanding concrete class struggles in specific social formations. However, there are positions in the class structure of advanced capitalist social formations which cannot be subsumed under discussion of capitalists, workers, and/or fractions of these classes. These positions as well have implications for concrete class struggles.
In the first four major sections of this review and critique of existing literature, I considered approaches to the class structure as a whole which viewed such “middle” positions in terms of a polarization view of the social formation. It is correct to view any social formation dominated by capitalist social relations as polarized in the sense that the principle structural contradiction is embodied in the relationship between capital and labor power. However, there are class positions in such social formations which are neither working class nor capitalist class and yet cannot be viewed as either outside the class structure (e.g., middle strata) or as transition classes (e.g., a transition petty bourgeoisie). It is to a reconsideration of such “middle classes” and to arrive at a correct theoretical specification of these positions that we now turn. From both the structuralist and the contradictory position approach to class structure, in conjunction with the previously developed notions of the basic classes and their class segments, we draw the major elements which allow us our understanding of the totality of class structure in an advanced capitalist social formation.
Polulantzas and the Structuralist Approach to Class Structure.
Introduction
Structuralist Marxism has its origins in the work of Allthusser. While we do not intend to draw or debate the major lines of structuralism, we do contend that its application to the arena of class structure has offered some important insights. In this section, we consider the work of the pioneer in structuralist interpretations of class formation, that is, of Nicos Polulantzas. (1973a, 1973b, 1973c, 1975). We have already considered Poulantzas’ contributions to an understanding of fractions within the capitalist class. In classes in Contemporary Capitalism (1975), in addition to discussing such fractioning, Polulantzas present a solution to the “middle class’ problem which uses both economic and ideological criteria to demarcate the boundaries of the working class and the capitalist class. These criteria lead him to view contemporary French class structure as consisting of a working class, capitalist class and both a new and traditional petty bourgeoisie. In Political Power and Social Classes (1973a), Poulantzas approaches the subject of class at a theoretically higher level of abstraction. In this book, whose main topic is the elaboration of a theory of the capitalist state, Poulantzas also offer a capsilized general theory of class formation. While Poulantzas’ analysis propositions and their specification to French social structure provide some illuminating insights.
General Approach
Among the issues Poulantzas addressed in his general statements relevant to class formation, several are critical for our own approach. The concept of a mode of production and its distinction from the social formation is one such issue. For the author, a mode of production is both a totality of forces of production/relations of production, and an abstract articulation of economic, political and ideological institutions and relations, with the economic “determinant in the last instance”. Such economic determination is not expressed in mechanistic determination of political and ideological superstructure by economic base. Rather, the relations of production of a given mode determine whether it is economic, political or ideological relations which play the dominant role in structuring the social life of that mode (1973a: 15, 70). For example, Poulantzas argues that in the feudal mode, religious and political ideology dominate social life because of the requirements of the economic organization of feudalism. It follows that classes in a mode of production are not simply a function of – or expressed in – economic organization. Instead, “they reveal the effects of the global structure (the ensemble of economic, political and ideological structures) in the field of social relations.” Classes in the feudal mode “already manifest themselves as particular economics political cautes.” (1973a: 63, 68).
Whereas the mode of production is a theoretical abstraction, the social formation is a concrete historical reality. Like Pierre-Philippe Rey (1973), another French Marxist trained in the Althusserian tradition, Poulantzas conceives of the social formation as the terrain in which different modes of production come together to produce real historical societies.
The mode of production constitutes an abstract formal object which does not exist in the strong sense of reality. Capitalist feudal and slave modes of production which equally lack existence in the strong sense also constitute abstract formal objects. The only thing which really exists is an historically determined social formations, i.e., a social whole in the widest sense, at a given moment in its historical existence, e.g., France under Louis Bonaparte, England during the Industrial Revolution. But a social formation, which is a real concrete object and so, always original because singular, presents a particular combination of specific overlapping of several ‘pure’ modes of production. (1973: 15).
In the unique configuration of overlapping modes comprising the social formation, one mode of production is dominant. (1973a: 15) this dominant mode both structures the articulation of the other modes as well as the expression of a principal contradictions inherent in the social formation. For Poulantzas, the implications for class structure in advanced capitalism of this view of the social formation are the following. At the level of the capitalist mode of production, there are the two classes of capitalists and workers. However, in a social formation dominated by capitalism, there are both capitalists and workers, representing the principal contradiction, and additional classes which result from the presence of the formation of dominated modes of production. Some of these classes can be identified theoretically in the abstract ‘pure’ form of the dominated modes. For example, the traditional notion of the petty bourgeois producer in advanced capitalism comes from the conceptualization of simple commodity production. Other classes and fractions cannot be identified in an abstract mode. Rather, they emerge only in the complex real articulation of multiple modes of production in the social formation. “…the effects of the concrete combination of the respective instances of the modes of production, effects of combination which are present in the effect of the structures of a social formation on its supports, (that is to say, in the social classes of formation) give rise to a whole series of phenomena of splitting, dissolution and fusion of classes, in short of the overdetermination and underdetermination of classes, of the appearance of specific categories, etc. these can not, however, always be located by the examination of the pure modes of production entering into combination (1973a: 73).
This in turn implies that while all advanced capitalist formations will have a capitalist class and working class, the totality of class structure in such formations is variable and to be given detailed content only by examining the specific formation.
Nevertheless, Poulantzas’ view of class structuration in a social formation is not voluntarist. It is not ideology and politics as subjective consciousness in concrete struggles which determines classes. If classes are determined politically and ideologically as well as economically, they are nonetheless structurally determined. Therefore, the political and ideological criteria, as well as the economic criteria, used to delineate classes must be found in objective institutionalized arrangement. This will be elaborated further when we consider the specific criteria Poulantzas uses to delineate French class structure
It should be noted that the logic of multiple modes of production comprising the social formation is not inherently incompatible with a more voluntarist view of class formation. In fact, as Rey have shown in Les Alliances de Classes (1973), this logic is quite amenable to a more “struggle-ist” perspective, which sees class formation as a variable dependent upon the character of the specific struggles which occur. One can place the emergence of classes and class fractions not only in the structural articulation of modes of production, but in the varying struggles which such complex articulation permits. The implication here is that while some generalizations are possible – for example, all advanced capitalist social formations will have a capitalist and working class-classes, the form and content of their organizations and struggles will vary among advanced capitalist formations.
Before moving to Poulantzas’ specific application of his more general view of class formation, one further issue warrants discussion. This is the author’s identification, on an abstract level, of the concept with which to order the terrain of class formation. We have already defined that the notion of class in Poulantzas’ schema. In addition to the classes of a social formation, the author identifies class fractions, social categories and strata.
Class fractions are segments of classes located primarily at the economic level. For example, industrial, commercial and banking capital are fractions of the capitalist class. While fractions, like classes themselves, are effects of the totality of social relations and can therefore be located at the political and ideological levels as well, they are always distinguished from social categories which are the “specific effect” of political and ideological structures. “While political and ideological criteria can intervene in a more or less important fashion in determining fractions and strata, in the determination of social categories they have the dominant role.” (1973a: 87) Intellectuals and the state bureaucracy are important social categories for Poulantzas. Social categories may draw their members from more than one class, yet they have a structural unity of their own. Like fractions, they are “primary effects” of the articulation of modes in a social formation, and like fractions, social categories may become social forces. It should be stressed that despite their structural unity and consequences for struggle, Poulantzas is careful to place social categories in their proper context in the class structure as a whole. Neither are they outside classes, nor do they occupy a theoretical status parallel to the notion of class. For Poulantzas, “the division into classes forms the frame of reference for every social stratification” both in the theoretical and in the concrete historical arenas. (1975a: 199)
Continuing the author’s descriptions of the social terrain, strata, like fractions are located in the ensemble of economic, political and ideological relations. But unlike both social categories and fractions, strata are “products of the secondary effects of the combination of classes, categories and fractions, which can, without being social forces, exert an influence on the political practice of these forces”. (1973: 85) The working class aristocracy, a class segment, as well as the summits of the state bureaucracy, a category segment, are examples of strata.
The distinction of social segments which have primarily in economic basis – class fractions – from those which are located primarily in political and ideological processes – social categories – seems easily applicable and useful in ordering the terrain one wishes to analyze. The distinction between those segments which may become social forces – fractions and categories – and those which may not – strata – is more problematic. By applying such phrases as “capable of constituting social forces” and “substratum of eventual social forces” to fractions, (1973: 84) the author implies one may distinguish fractions from strata before examining the struggles in a particular historical setting. If this is so, the social force criterion is again relevant. But in its application, the analysis of fractions categories and strata in an historical setting becomes voluntaristic and lacking in structural basis. Moreover, in different historical context, similarly structurally determined segments could be differentially labeled as fractions or strata, creating semantic and analytic confusions. For these reasons, Poulantzas’ distinction between fractions and categories, and strata according to a social force criterion appears misguided.[2]
Specific Application: Political and Ideological Criteria.
Having outlined the analytic categories of class formation in Political Power and Social Classes, in Classes in Contemporary Capitalism, Poulantzas gives these categories content in French situation. Contrary to the PCF anti-monopolist view of the “middle strata”, Poulantzas claims there are no segments which exist outside the class structure. Rather, as noted above, the author conceives French class structure to consist of capitalist and working classes, as well as a petty bourgeoisie composed of new and traditional segments. The problem becomes identifying criteria which locate the boundaries of the petty bourgeois segments vis-Ã -vis both capital and labor. In this task, a certain continuity with other European writers is evident.
Poulantzas claims we must recognize the petty bourgeoisie from the point of view of the polarization of capital and labor in a social formation dominated by capitalist social relations. The first criterion for recognizing members of the petty bourgeoisie is that they are neither workers nor capitalists. They do not produce surplus value for capital as workers must. Neither do they simultaneously hire and exploit labor power and enter into the relations of real economic ownership or possession. These are the criteria which, for Poulantzas, delimit the boundaries of the capitalist class.
The traditional petty bourgeoisie is composed of artisans, small shopkeepers and independent professionals who are petty bourgeois not only because they operate with limited capital outlays, but because they do not hire and exploit labor power. This creates a qualitative, and not, as the PCF would have it, only a quantitative boundary between the traditional petty bourgeoisie and small capitalists.
The new petty bourgeoisie is initially defined as unproductive wage and salary labor. While classes for Poulantzas are effects of the totality of social relations, adherence to the idea of “economic determination in the last instance” would seem to indicate ultimate reliance on economic criteria to define classes. In accordance with this logic, which he insists upon, the author has so far defined both segments of the petty bourgeoisie in terms of economic criteria. However, he overrides his unproductive labor criterion for the new petty bourgeoisie to state that technicians, through productive labor, should not be considered working class because of ideological criteria which in their specific case dictate they belong to the new petty bourgeoisie. Poulantzas further argues that political and ideological criteria are more important in defining the petty bourgeoisie that in delineating capitalists and workers, who form the classes of the ‘pure’ capitalist mode of production. This enables him to take two segments which have different positions in the relations of production, i.e., the new and traditional petty bourgeoisie, and unite them in one class, the petty bourgeoisie, on the claim that their divergent economic positions have similar political and ideological effects. If certain groupings which at first sight seem to occupy different places in economic relations can be considered as belonging to the same class, this is because these places, although they are different, nevertheless have the same effects at the political and ideological level (1975: 205).
To better grasp some of the problems which emerge when one looks at class structure using Poulantzas’ criteria, it is worth going into more detail on the structural determination of the new petty bourgeoisie. First, let us consider the unproductive labor criterion. While some of the writers discussed under the rubric of polarizations views used this criterion, and while both Marian Nicolaus in “Proletariat and Middle Classes in Marx: Hegelian Choreography and the Capitalist Dialectic” and John Urry in “Toward a Structural Theory of the New Middle Class” base theories of the middle class in unproductive labor, Poulantzas attempt to use this criterion is the most sophisticated.[3] Poulantzas recognizes the surplus labor furnished by all wage and salary employees of capital, yet claims that since the extraction of surplus value is the specific form of exploitation under the capitalist mode of production, only wage labor which is productive of surplus value can be considered working class.
It should be mentioned that Poulantzas, rather arbitrarily, employs a very restrictive notion of productive labor as that which creates surplus value in the production or material goods. It seems ill advised to use either this definition or the more common definition of productive labor as that which creates surplus value for capital to define the working class. As we noted in discussing polarization views the use of the productive labor criterion requires decisions as to what positions in production are actually productive. A lively debate within Marxist theory focuses on precisely this problem. (See, for example, Gough 1972; Baran, 1957; Morris, 1958; Blake, 1960; Gillman, 1965; Mandel, 1971; O’Connor, 1975.) Even if we could all agree on what constituted productive labor, it would seem a rather useless criterion for delineating class structure. If the point of such a delineation is to derive interests which may become activated in struggle, it is difficult to see that the productive/unproductive labor distinction would today produce divergent interests. Rather, since it is in capital’s interest to dominate and extract surplus labor from productive and unproductive workers alike, we should expect that both categories of labor processes of segments of productive and unproductive workers to produce common interest. Conversely, if sufficient differences in conditions of work within both the productive or unproductive labor camp exist, we should expect that both categories interests. Specifically, it would seem that the proletarianization of commercial labor, discussed by Braverman and by many of the French and German writers, would create many new positions which share the interest of a productive working class in a transformation to a socialist society.
If the criterion of productive labor is viewed historically, we can see that one hundred years ago, the distinction between productive and unproductive labor would have been a more meaningful one to tap class boundaries. At that time, the number of unproductive employees was quite small. Commercial employees, for the most part male clerical laborers, generally occupied a privileged position close to the small competitive capitalist. As Braverman has noted, such clerical employees were paid well, taken into the confidence of the capitalist, and could expect they might become a partner in a boss’s business and/or marry the boss’s daughter. Under these circumstances, because of its superimposition on criteria of autonomy in the labor process as well as, in some cases, participation in real economic control of the enterprise, the productive/unproductive labor distinction was consequential for class structure. It was, in fact, a proxy for the autonomy and economic control criteria which were operating to insure that clerical labor had interests which differed from those of the Industrial proletariat. Now the unproductive labor is no longer coterminous with these criteria, the productive/unproductive labor distinction has lost its usefulness for theories of class structure.
Moving to a consideration of the role played by political and ideological relations in Poulantzas’ schema, we have already noted that political and ideological determinants of class position are important for the new petty bourgeoisie as a whole. For technicians and engineers involved in the production of material goods, these criteria override the economic criterion of productive labor. While Poulantzas refuses to consider technicians and engineer to be working class, neither are they members of the capitalist class. Unlike managers, they do not enter into relations of possession. They only execute the orders of top management, and are exploited by capital.
Technicians and engineers are members of the new petty bourgeoisie because of their function as supervisory labor, itself exploited, but exploiting as well in its role of extracting surplus value from the working class. The exploiting supervisory function of technicians dominates their productive function as simple coordinators of labor power in a technique of production. Coordination signals the technical aspect of the division of labor, supervision, the political and ideological aspect.
For Poulantzas, as for Braverman and the French writers we have considered, supervisory labor is bound up with the political control of the working class, and with the ideological support of this control, that is, the separation of mental from manual labor. Such political and ideological relations are as much a part of what goes on in the capitalist enterprise as are economic relations. Scientific applications of technology are, for Poulantzas as for Braverman, the materialization of political and ideological domination. Only in capitalism does science, which is based ultimately on the experience of the direct producers, take the form of the strict separation of mental and manual labor. In words echoing Braverman, Poulantzas states “This division is thus directly bound up with the monopolization of knowledge, the capitalist form of appropriation of scientific discoveries and of the permanent exclusion of the subordinated side by those who are deemed not to ‘know how’.” (1973: 237) Technicians and engineers, as bearer of such political and ideological domination in the work place, can not be working class.
The new petty bourgeoisie as a whole is a bearer of technocratic ideology. Simultaneously, the mental/manual distinction and the domination/subordination authority distinction mark the boundary of the working class from the new petty bourgeoisie of commercial and service employees, teachers, bureaucrats, technicians etc. These distinctions are usually coterminous with unproductive labor as Poulantzas has defined it, but as we have seen, it is ultimately upon the political and ideological criteria themselves and not upon productive/unproductive labor that Poulantzas rests his case for the new petty bourgeoisie.
Poulantzas recognizes that there are many new petty bourgeois employees – keypunchers, file clerks etc. –who do not directly supervise and exploit others. However, he argues that even when not directly dominating workers, new petty bourgeois employees, because of the bureaucratization of their jobs, reproduce within themselves the relations of domination and subordination which technocratic ideology claims are necessary. The author argues further that bureaucratization is the deformed expression of the political domination of the bourgeoisie, and that the working class does not reproduce such bureaucratization within its ranks. Like Braverman, Poulantzas signals continual separation of conception from execution in the skill hierarchies of the working class, but he claims that hierarchies of authority are the badge of the new petty bourgeoisie.
Such bureaucratic hierarchies produce vertical fractions within the new petty bourgeoisie. In addition to signaling extensive fragmentation due to bureaucratization, Poulantzas emphasizes the presence of some fractions which are potential social forces polarized toward the working class. They are, unsurprisingly, the proletarianized wage earners in the commercial sector, particularly those in mechanized offices; low level functionaries in public and private bureaucracies; clerks, waiters and waitress, busboys etc., in large restaurants and department stores, and technicians and engineers who are directly implicated in the production of surplus value.
Not only does Poulantzas ultimately define the new petty bourgeoisie according to political and ideological criteria, but, as has been noted, he claims the new and traditional petty bourgeoisie may be considered as one class on the basis of their shared ideology. This common petty bourgeois ideology is not autonomous and ‘pure’ as is that of capital and labor. Rather, as the heart of the terrain of class struggle, the petty bourgeoisie incarnates elements of both working class and bourgeois ideology, combining them in its own unique way. Petty bourgeois ideology is marked principally by the following traits: (1) a reformist anti-capitalist penchant – in the new petty bourgeoisie this take the form of piecemeal economistic reforms and demands for more ‘expert’ participation in government, while in the traditional petty bourgeoisie it is expressed in adherence to ideologies of free enterprise – (2) individualism – for the new petty bourgeois this takes the form of upward mobility career aspirations, whereas the traditional petty bourgeois hopes to move from petty bourgeois status to the position of small capitalist – (3) a tendency to conceive of the class as offering a third alternative to government by either a working or capitalist class – the new petty bourgeoisie sees the state as both its own, and a neutral arbiter among conflicting forces, while the traditional petty bourgeoisie sees itself as a neutral class between labor and capital.
While we appreciate Poulantzas’ attempt to outline class segments which may potentially ally with the working class as he defines it, there are some serious problem with his use of political and ideological criteria to delineate the petty bourgeoisie. First, as Wright has pointed out, the different economic positions of a new and traditional petty bourgeoisie do not really produce a convincing similarity of ideological effects. (Wright; 1976: 13) Poulantzas’ own specification of new and traditional petty bourgeois ideology better illustrates the differences between them than it does their convergence.
Second, as Wright also notes, the use of political and ideological criteria to override the economic, both in the definition of the petty bourgeoisie as a whole, and in the placement of technicians and engineers in the new petty bourgeoisie, is inconsistent with Poulantzas’ statement that economic criteria should be primary in the structural determination of class. It seems capricious at best that political and ideological relations dominate to make productive technicians petty bourgeois, whereas unproductive employees who supervise no one and, in Braverman’s terminology, execute rather than conceive, are nonetheless not considered working class. Poulantzas would argue that such unproductive labor is different from the working class according to political criteria since it reproduces capitalist relations of domination and subordination within itself, whereas the working class does not. Yet we have seen that in the extreme view, the working class, when not in direct revolt against capital, may be seen as aiding in the reproduction of both capitalist political and ideological relations by its very submission to its lot. Thus, Poulantzas’ argument stressing the objective political differences of the working class from such proletarianized unproductive labor is weak.
Poulantzas’ discussion of petty bourgeois ideology and of the ideologies opf fractions of the new petty bourgeoisie illustrates another danger in relying too heavily on political and ideological criteria to define class. There is the ever present danger that objective and structurally determined political and ideological criteria such as authority or mental/manual distinctions may degenerate, as in practice they often do for Poulantzas, into the subjective politics and ideologies of class segments. Classes in Contemporary Capitalism is thus open to Poulantzas’ own critique of voluntarism made in Political Power and Social Classes.
Moreover, as Wright has cogently argued, the major political criterion used by Poulantzas and most others who argue for “political” determination of a “middle class” – that is, the criterion of domination and subordination in hierarchies of authority – can be viewed as essentially economic. Following this logic, the autonomy inherent in some positions in the labor process can be conceived of as an economic criterion as well.
Following Poulantzas, supervision can be conceived as the direct reproduction within the process of production itself of the political relations between the capitalist class and the working class.’ Alternatively, supervision can be seen as one aspect of the structural dissociation between economic ownership and possession at the economic levels itself. That is possession, as an aspect of the ownership of the means of production, involves (to use Poulantzas’ own formulation) control over the labor process. In the development of monopoly capitalism, possession has become dissociated from economic ownership. But equally, possession has become internally differentiated, so that control over the entire labor process (top managers) has become separated from the immediate control of labor activity (supervision). Unless possession itself is so be considered an aspect of political relations, there is no reason to consider supervision a reflection of political relations within the social division of labor rather than a differentiated element of economic relations (Wright, 1976: 19, 20).
Finally, despite a necessity to locate classes in the totality of social relations, unless some kind of primacy is given to economic criteria, we lose a fundamental Marxist insight into social life. While Poulantzas does not offer us a way to use political and ideological criteria and yet to retain the primacy of the economic in defining class, the work on contradictory class positions, discussed below, moves us closer to this goal.
Contradictory Class positions.
Introduction
The work of Wright (1967, 1977) as well as that of Cardhedi (1975a, 1975b) offers a major theoretical breakthrough in the analysis of class structure. Whereas many of the writers we have discussed have argued that monopoly capitalism produces some positions which take their characteristics from both the working and the capitalist class, Carchedi and Wright formalize this argument into the theory of contradictory class positions. Such positions are not ambiguous or simply difficult to pigeonhole into the capitalist class or the working class: they are precisely and exactly contradictory.
Both Carchedi and Wright formulate contradictory class position theory as a largely economic solution to the problem of the “middle classes”. Their work seeks to theoretically “find out what is the limited yet fundamental (determining) role of the economic in the definition of classes”. (Carchedi, 1975a: 1) the development of capitalism as a mode of production is analyzed in order to ascertain the production relations relevant for determining class positions and contradictory class positions; and, the abstract laws of monopoly capitalism provide the framework within which the analysis of contradictory class positions for the most part takes place.
We will know highlight the key points of Carchedi’s and Wright’s analyses. The major thrust of both writers’ discussions is very similar though they start with the specification of different factors as determinants of ‘pure’ and ‘contradictory’ positions within class relations.
Carchedi: A Functional Definition
Carchedi cites three relations of production as relevant for determining class positions. These are ownership, expropriation of value, and function. Ownership is real economic ownership and is distinguished from both legal title and possession (i.e., “the ability to set in motion and govern the means of production” Carchedi, 1975: 363) which under monopoly capitalism belongs to the collective worker.[4] Expropriation refers to both the expropriation of surplus value and surplus labor time and therefore, both productive and unproductive workers are exploited and are included in the working class on this criterion. The function of the capitalist is supervision and management of the production process.[5] The function of labor is the production of use values.
The correspondence of these three relations defines for Carchedi the positions of collective worker and alternatively the global capitalist. The collective worker is the non-owning, expropriated, producer of use values. The global capitalist is the owner-expropriator who supervises and manages the labor process in order to extract surplus value.
Under monopoly capitalism the capitalist function of supervision and management becomes separated into a) control and surveillance and b) co-ordination and unity of the labor process. “To perform the function of capital basically means to carry out the work of control and surveillance” (Carchedi, 1975a: 24). The function of co-ordination and unity of the labor process reverts to the collective worker under monopoly capitalism.
The new middle class “performs both the function of the collective worker and the global function of capital” (Carchedi, 1975a: 4) in that it is engaged in the control and surveillance as well as the coordination and unity of the labor process. This class embodies contradictions on several levels of production relations: “ (1) It does not own either legally or economically the means of production (2) it performs both the global function of capital and the function of the collective worker (3) is therefore both the labourer (productive or unproductive) and the non-labourer and (4) is both exploiter (or oppressor) and exploited (or oppressed) (Carchedi, 1975a: 51).
The objective economic relations of production of which such positions consist are contradictory – the positions embody both capitalist and working class relations. The wages of the new middle class are “actually made up of both wages and revenue” (Carchedi, 1975a: 56). The more a contradictory position performs the function of the collective worker, the greater is the weight of the wage share. Whereas Braverman carefully documents the historical process of proletarianization of United States clerical and service labor, Carchedi’s formulation allows us to understand this process of proletarianization with greater theoretical precision. “Proletarianization is the limit of the process of devaluation of the middle class labour power, i.e., the reduction of this labour power to an average, unskilled level coupled with the elimination of the global function of capital”. (Carchedi, 1975a: 65)
Wright: Contradictory Class Locations Within and Between Modes of Production.
Wright’s analysis of contradictory class positions is very similar to form and content to Carchedi’s but has some differences, and more interestingly, some elaborations. Wright, like Carchedi, identifies three relations of production for defining class positions, but they are different relations than the ones Carchedi has identified. Wright’s definitional criteria derive from relations formulated in terms of aspects of control of the production process.
It is possible to isolate three central processes underlying the basic capital-labour relationship: control over the physical means of production; control over labour power; control over investments and resource allocation…when we speak of the fundamental class antagonism between workers and capitalisms, what we mean is that these two classes are polarized on each of these three underlying processes: the capitalist class has control over the entire apparatus of production, over the authority structure as a whole, and over the overall investment process: the proletariat is excluded from each of these (Wright, 1976: 30).
For Wright, control over investment and resource allocation constitutes real economic ownership. Control over the physical means of production and control over labor power are the two aspect of the relation of possession. Contradictory positions between the capitalist class and the proletariat are positions which are partially excluded from these relations of control and partially included. The form of the argument is identical to Carchedi’s analysis of the congruence and noncongruence of class defining relations of production.
Wright’s framework proves particularly useful for analyzing different positions within the category of contradictory locations. Top and middle managers, technocrats, and foreman and line supervisors all participate in varying degrees in the three different relations of control of production processes and relations. The specification of each category of contradictory locations in terms of its degree of participation in the relations of control of the productive process produces a detailed and revealing analysis of the relations and interests of these positions. We will return to a further specification of this argument in the second major part of this paper.
At this point Wright adds a major elaboration to the theory of contradictory positions within class relations. The conjuncture of previous and current modes of production adds the class category of petty bourgeoisie to the schema and adds contradictory positions within a mode of production there exist contradictory positions between classes from different modes of production. Chart 1 despicts this elaboration. (Wright, 1976: 27).
Small employers occupy a contradictory position between the petty bourgeoisie and true bourgeoisie. Semi-autonomous employees occupy a contradictory positions position between the petty bourgeoisie and the proletariat. Wright introduces this elaboration to bridge what he considers a major cap in Carchedi’s schema – the inability to deal with positions outside the simple proletariat – manager – bourgeoisie relationships.
Wright also links discussion of contradictory class locations to his distinction between fundamental and immediate interests. Class interests which involve the retention or transformation of a given system of production are fundamental interests. Immediate interests exist within the limits of a given mode of production. Immediate interests constitute interests within a given structure of social relations; fundamental interests center on interests which call into question the structure of social relations itself. That is, immediate interests are interests defined within a given mode of production, while fundamental interests are defined between modes of production (1977: 4).
When existing mode of production is capitalism, fundamental interest are defined in reference to a socialist transformation of capitalism, whereas immediate interests involve reforms leaving the basic structure of capitalist relations of production untouched. Note, that both fundamental and immediate interests are objectively determined by the relations of production. Contradictory locations are contradictory precisely because they entail fundamental interests of more than one class.
A final and particularly important aspect of Wright’s analysis is his attempt to specify more precisely the meaning of economic determination of classes ‘in the last instance’, and the way in which political and ideological relations enter into the determination of class structure. Wright agrees with Poulantzas that economic relations are primary in the determination of classes, but that nonetheless political and ideological relations have their effect. However, he criticizes Poulantzas’ failure to theoretically specify this effect since this failure at times leads Poulantzas to allow political and ideological relations to supercede economic relations. Wright therefore suggests the following: “The extent to which the political and ideological relations enter into the determination of class position is itself determined by the degree to which those positions occupy a contradictory location at the level of social relations of production.” (Wright, 1976: 40) This implies that economic relations are the sole determinants of Wright class un-ambiguous class locations and that contradictory class locations are first and foremost the result of economic processes. Political and ideological relations such as the mental/manual distinction act on contradictory class locations which are predetermined at the economic level to push these locations closer to or farther from specific unambiguous class locations.
Evaluation and Conclusion
While both Wright and Carchedi have developed the notion of contradictory positions, their exact theoretical specification of the relations defining these positions differs. The global function of capital and the collective worker form the bulwark of Carchedi’s analysis. Wright’s criteria are relations of control – of resources and investments, of the physical means of production, and of labor power.
In general, concepts grounded in ‘relations’ are preferable to concepts grounded in ‘functions’ because relations are an abstraction from concrete man in his concrete activity, whereas functions are an abstraction from theories of man and society. More specifically, Carchedi’s analysis lends itself well to understanding the proletarianization of the middle class as a process whereby the interests of his class become closer and closer to the fundamental economic interests of this class into one or another alliance in a concrete social formation. Wright’s work goes beyond that of Carchedi in that it adds the conjunction of classes from previous modes of production. It also allows us to make a qualitative distinction between certain “middle layer” positions. For instance, top corporate executives who, despite their different juridical status, share full real economic ownership and possession with more traditional capitalists, are part of the capitalist class. The rest of the positions in the management hierarchy are identified, according to Wright’s criteria, as contradictory class locations.
Wright’s concern with the specification of the way in which political and ideological relations enter into the determination of class structure is praiseworthy. Such a specification is vital to a correct theory of class structure. Wright’s formulation is both appealing in its simplicity, and, we feel, basically correct. While much of the literature we have reviewed would suggest that the effect of political and ideological relations on class structure is greater than that suggested by Wright, part of the difference of viewpoint is semantic rather than substantive. That is, as noted in the discussion of Poulantzas, Wright maintains that the process resulting in the hierarchy of supervisory and management positions is an essentially economic process, rather than the political and ideological process others have termed it. The hierarchy of fragmented “conception” as Braverman would term it, with the differentiated authority attached to it, then becomes an economic effect. As such, it leads to economic criteria to define both classes and contradictory locations.
Implicit to Wright’s usage of economic relations is its equivalence with the totality of social relations of production. We agree with the major thrust of Wright’s argument on the role of political and ideological dominate labor are simultaneously economic, political and ideological. In these social relations of production are found Wright’s criteria for defining both ‘pure’ classes and contradictory locations. Hence, as an aspect of the social relations of production, political and ideological relations in the production process enter into the creation and definition of classes and contradictory locations. For example, despite Wright’s contrary statement on the mental/manual division, this distinction acts in the creation of contradictory class locations. The degree of conceptualization versus execution attached to jobs does not only enter after the fact to push contradictory class location closer to or farther from a class boundary. Rather, it acts simultaneously with the other aspects of positions of managerial, supervisory and semi-autonomous employees, in the specification of the positions themselves. Since both classes and contradictory locations are defined by position in the social relations of production, that is, in the economic realm, the economic determination “in the last instance” of class is satisfied.
As a parting note we believe it is worth the effort to be careful with terminology when discussing contradictory positions. Wright’s phrase of ‘contradictory locations within class relations’ seems to us to be sufficiently clear. These positions are, in a sense, in two classes defined at the level of fundamental economic interests. Phrases such as the ‘contradictory position of the new middle class’ are to be avoided for they imply that these positions define a unique class which has ambiguous or ill-defined fundamental interests.
[1] The identification of such sector is similar to what some Latin American theorists call the ‘marginal sector’ of employment. However, Bluestone’s irregular economy has a different basis from the Latin American ‘marginal sector’. The irregular economy of an advanced capitalist system is comprised of positions at the lowest levels of the competitive sector. In contrast, the marginal sector of dependent economies is comprised of positions engaged in petty bourgeois as well as wage labor positions both of which are generated by a system of simple commodity production coexisting with more advanced capitalist production. Labor market relations are structured differently in the marginal economy of dependent nations and in the irregular economy of advanced capitalist nations. In the marginal sector the market logic of simple commodity production enables workers to create their own positions through craft production and sale of goods. In the irregular sector the logic of competitive market relations obtains. This means that while both positions and workers may be irregular or unstable the quantity and wage rate of labor are determined by the traditional market relations of supply and demand. As consequences in the marginal sector workers have the possibility of creating their own positions, either by taking up craft production or by engaging in various forms of petty commercial trade. In the irregular sector workers supply labor for sale on the fringes of the competitive market where the demand for labor is concomitantly a function of employer decisions to create positions.
[2] However, a notion of strata which avoids voluntarism is potentially useful. Poulantzas does mention a structural criterion for the designation of strata: this is the status of strata as “secondary effects” of the combination of fractions, classes and categories, themselves primary effects of the combination of modes in the concrete social formation. It is difficult to know precisely what is meant by this notion of secondary effect. Logically, it seems to mean that strata as segments within or cutting across classes, fraction, and categories are dependent on the prior existence of those classes, fraction, and categories. A clearer structural picture of why such secondary effects can not constitute autonomous social forces in struggle might be gotten by retaining Poulantzas’ concepts, but explicitly invoking the criterion of direct “rooted ness” in the relations of production for fractions. Class fractions could then be seen as being coterminous with objective immediate political and ideological interests, all of which could be potentially activated in struggle. The objective immediate interests coterminous with a stratum like the working class aristocracy could not be abstracted from the differing working class fractions, directly rooted in production, which this stratum cut across, they could not, in Poulantzas’ terminology, constitute autonomous social forces. Although we do not retain the notion of strata in our discussion of the most important elements comprising the structuration of the social formation, we use and develop the above logic in our discussion of class fractions in the second major part of this books.
[3] Erik Wright in “Appendix A: Alternative Conceptions of Class in Contemporary Marxist Theory” of his dissertation Class Structure and Income Inequality, gives excellent brief reviews of the arguments made by Nicolaus and Urry.
[4] Note that this definition of possession is quite similar to that of Poulantzas who defines possession as “the capacity to put the means of production into operation”. (Poulantzas, 1973c: 28) However, for Poulantzas, possession is a function belonging to the capitalist class.
