1. Among the book-length surveys, I recommend Karsz 1974 (excellent for placing the whole range of Althusser's writings in the context of his later emphasis on class struggle); Callinicos 1976 (excellent on structural causality, weak on and hostile to Althusser's concepts of philosophy, science, and ideology); Benton 1984 (a judicious overview of Structural Marxism with a good discussion of Althusser and his critics); and Elliott 1987 (solid account of Althusser's theory in light of his politics, but occasionally reducing the former to the latter). Among the critiques, Rancière 1974 remains worth reading, as do Hirst 1979 and A. Glucksmann 1978. E. P. Thompson 1978 is as uninformed as it is hostile, but it has achieved something of a cult status and thus a life of its own. See also Vincent et al. 1974. For a full bibliography of Althusser's publications, see Elliott 1987.

2. For Derrida's reflections on Marxism, history, and Althusser, see Derrida 1981; for an attempt to integrate Marxism and deconstruction, see Ryan 1982. Rosalind Coward and John Ellis (1977) provide a first-rate synthesis of Althusserian, Lacanian, and semiological perspectives (from Barthes to Kristeva). Juliet Mitchell (1974) brings Althusser and Lacan to bear on the relationship between feminism and psychoanalysis; see also Barratt 1980. Callinicos 1982 contains an excellent account of Deleuze and Foucault from a perspective informed by Althusser. Pierre Vilar (1973), a member of Annales , has written a lengthy article on Althusser that I discuss in chapter 1, note 1; see also D'Amico 1973. From the camp of the Habermasians, attempts to assess Althusser have been disappointing: Schmidt 1981 is superficial; John Thompson 1984 is much broader and more detailed, but stubbornly obtuse with respect to the explanatory power of the concepts he is attempting to critique. Perry Anderson (1980)

Page 366

responds to E. P. Thompson's polemic against Althusser; see also Nield and Seed 1979 and Benton 1984.

3. The literature on Western Marxism is overwhelming. As a very selective list for those seeking a comprehensive introduction to the major figures and currents, I recommend the following works: Jay 1984; P. Anderson 1976a; Howard and Klare 1972; Poster 1975 (an invaluable account of French developments); Stedman Jones et al. 1978; D. Harvey 1982; and Carnoy 1984. For classic Western Marxist interpretations of Marx and Marxism, see Lichtheim 1965; Avineri 1968; and Gouldner 1982. For an excellent Structural Marxist account, see Therborn 1976.

4. For Althusser's increasing frustration with the Party bureaucracy throughout the seventies, see Althusser 1978a; see also Althusser 1978 and 1977. Althusser's letters on the events of May 1968 may be found in Macciocchi 1973. Althusser must be seen as seeking a "third way" on the question of party reform, opposing both the existing Stalinist organization, which subordinated mass initiative and participation to the interests of the Party apparatus, and the conversion of the PCF into a reformist, parliamentary party. Rancière 1974 and Elliott 1987 are valuable on Althusser's political evolution.

There was a nuanced opposition within the Althusserian camp between more Leninist and more Gramscian views expressed with respect to the elimination, in 1976, of the slogan "dictatorship of the proletariat" from party canon. Etienne Balibar (1977) argued against elimination on the grounds that, whatever its Stalinist perversions, the term focused political attention on the class nature of existing parliamentary democracy and on the problems of the transition to socialism—problems that, in Balibar's view, were being dangerously ignored by the PCF as it blindly pursued a strategy of alliance with the Socialists. Balibar's arguments are not superficial; they stress important differences between capitalist democracy, in which politics is controlled by the wealthy, and socialist democracy, in which the people would actually have power. Balibar is concerned, rightly enough, that the role of the state as an instrument of class struggle during the transitional period between capitalism and communism not be forgotten. The post-revolutionary state must not only smash the old elitist institutions but actively organize and promote new popular democratic forms. However, Balibar remains caught up in a Leninist view of state power and Lenin's untoward confidence in the capacity of the post-revolutionary state to resolve the problems of pre-revolutionary society. The potential authoritarian dangers of Balibar's faith in the primacy of revolution are foremost in the mind of Nicos Poulantzas (1978). In contrast to Balibar (and many other Althusserians), Poulantzas took a very critical stance toward the PCF. He was a strong defender of Eurocommunism and fully endorsed the rejection of the dictatorship of the proletariat. From a Gramscian perspective, Poulantzas was concerned with pre-revolutionary rather than post-revolutionary events and with ensuring the democratic nature of post-revolutionary society—preparing for democratic socialism through ideological and political struggles whose purpose is not merely to promote a revolutionary crisis but to extend and preserve existing political liberties as well.

The political differences between the Leninist and Gramscian orientations

Page 367

within the Structural Marxist camp with regard to Eurocommunism and reform within the PCF are outlined cogently in Benton 1984. A useful triangulation on Eurocommunism is obtained from Mandel 1978 (a Trotskyist critique); Claudin 1978 (a Left-Eurocommunist perspective); and Boggs 1982 (a post-Marxist account). For the French context and the PCF, see Ross 1982 and Jenson and Ross 1984. Debates within the PCF are surveyed in Kelly 1982 and Molina and Vargas 1978. For events of the eighties, see Ambler 1985; Cerney and Schain 1985; Ross and Jenson 1988; and Singer 1988.

5. For critical discussions of the neo-liberals, see Callinicos 1988; Tucker 1980; and Levine 1988. For Rawls, see Buchanan 1982; for a persuasive critique of Habermas, see Roderick 1986. I discuss Foucault and Deleuze in chapter 4; see also Callinicos 1982; and Resch 1989. Stoianovich 1976 remains a perceptive critical introduction to the Annales school. The disparate approaches of anthropologists Victor Turner, who is interested in the disruptive gaps between functionally ordered symbolic systems, and Clifford Geertz, a Parsonian disciple defending the primacy and unity of the cultural-symbolic, are straightforwardly presented in their own collected essays (Geertz 1973; Turner 1974). Baudrillard may be sampled in his selected writings (Baudrillard 1988); see also Douglas Kellner's thoughtful study of Baudrillard (Kellner 1989).

6. My thinking on postmodernism has been inspired by the seminal essays of Fredric Jameson (1984; 1984a; 1987) and by David Harvey's brilliant and comprehensive study (Harvey 1989). Both Jameson and Harvey focus on the political economy of postmodernity, and Harvey, in particular, does a thorough job of relating the cultural logic of postmodernism to the dissolution of national Fordist modes of production and the globalization of capitalism. For the inability of dissident postmodernism to comprehend its own existence, see Soja 1989 (and my review essay, Resch 1992). For cultural-aesthetic discussions of postmodern, see Eagleton 1986; Foster 1983; Huyssen 1986; Krauss 1985; McHale 1987; Owens 1980; and Ulmer 1985.

7. The class position of the professional middle class-new petty bourgeoisie is best developed, in my opinion, in Erik Olin Wright's reformulation of the concept of social class (Wright 1985; see also Wright et al. 1989). Wright's work has been influenced by Poulantzas and by the "labor theory of exploitation" developed by John Roemer (1986; 1988). Unlike his fellow "analytical" Marxist, Jon Elster, Roemer has contributed positively to the development of Marxist social theory through the use of game theory and mathematical model building. However, his contribution remains limited by its ahistorical and individualist assumptions regarding rationality. Ultimately Roemer is unwilling or unable to build the historical-intransitive dimension into his models—neither the ideological complexity of social subjectivity and habitus nor the matrix effect of the structured whole on the effectivity of individual structures. The analytical Marxists oscillate between an interesting critique of Marxist concepts of history and a neo-positivist attempt to subordinate the science of history to philosophy, specifically to certain premises of empiricist philosophy of science, namely, nominalism and methodological individualism. This move may permit a certain radicalism; Roemer, for example, takes neo-positivist methods

Page 368

and attempts to turn them against conservatives. More representatively, Jon Elster (1985) uses methodological individualism and nominalist empiricism to attack Marxism and to defend a philosophical position virtually indistinguishable from neo-liberalism. For a critical discussion of analytical Marxism see Ware and Nielson 1989; Callinicos 1989; and Levine, Sober, and Wright 1987.

Poulantzas 1975 explains why the "new petty bourgeoisie" of professionals and managers should be viewed as an ally of capital, not the working class (see chapter 6, note 12). The alternative, post-Marxist concept of the "professional middle class" is indebted to the opposite notion, fundamentally wrongheaded in my view, that because professionals are not owners of the corporations they manage they are somehow opposed to capitalism and somehow undermining the power of the capitalist class that does own or effectively own these corporations. Ultimately, the autonomy of the professional middle class is posited in order to refute the Marxist view of economic determination and to justify the "end of ideology" ideology of post-industrial or post-capitalist society. This technocratic view of a "managerial revolution'' has a long history, but it is perhaps most succinctly stated in Gouldner 1979 and thoroughly discussed in Walker 1979, a collection of essays on the concept of the professional middle class. The illusion of professional middle-class independence and anti-capitalism are finally succumbing to reality, however; see Barbara Ehrenreich's stimulating account of the inner life of the American middle class during the eighties (Ehrenreich 1990). Despite her many profound insights, Ehrenreich still cannot (or will not) see capitalism as determining the struggle between liberals and conservatives within the professional middle classes, a struggle she stubbornly persists in viewing as autonomous.

8. My views of the structural dynamics of perestroika have been informed by Post and Wright 1989; Lane 1990; Mandel 1989; Kerblay 1989; Lewin 1988; and Alec Nove 1983. Nove's pragmatic attempt to define a "feasible" market socialism that might avoid the mistakes of Bolshevik-type economies without abandoning the goal of socialism or leaping beyond the objective possibilities of the present conjuncture should be read in conjunction with Carens 1981 and Miller 1989. Nove's ideas have been criticized in Mandel 1986 and 1988. Nove replies to Mandel's first article in Nove 1987; see also Elson 198.

9. The totalitarian school of Soviet studies has concerned itself almost exclusively with moral condemnation, body counts, and a concept of party power and oppression detached from any social cause or explanation. It was never much interested in the class forces at work in the Russian Revolution or in the popular class struggles and political difficulties that transformed Bolshevism from a dictatorship of the proletariat to a dictatorship over the proletariat. Even Trotsky's interpretation of the "revolution betrayed" tends to ignore class analysis in favor of an explanation based on the "evil genius" of Stalin and a totalitarian conception of party power. Structural Marxists offer a more complex explanation emphasizing first the Stalin deviation, which substituted the development of the productive forces for the development of popular democracy, and second the class struggles within the Soviet Union, which ended in the triumph of a new technocratic ruling class, the state bourgeoisie (see Althusser 1976, 78-93; and Bettelheim 1983; 1982; 1978; 1976). These efforts paral-

Page 369

leled those of revisionist Anglo-American scholars seeking finally to break free from the dogmas of totalitarianism. For some recent fruits of this revisionism, see Fitzpatrick 1982; Getty 1985; Viola 1987; and Kuromiya 1988. These works build, of course, on the monumental achievement of the original and long-isolated revisionist E. H. Carr, who has lucidly condensed the results of his ten-volume History of Soviet Russia in Carr 1979.

10. My emphasis on the conjunction of realism and formalism in modernist culture is indebted to John Berger (1985), who argues brilliantly the thesis that cubism represents the truly original component of modernism and a new "syntax" for the modern experience. My distinction between realist modernism of the Left and the irrationalist and elitist movements of the Right may seem untenable at first, but I call the reader's attention to Peter Bürger's provocative attempt to define the avant-garde in terms of the social status and function of art (Bürger 1984) and John Willett's fine survey of the politics of modernism from 1917 to 1933 (Willett 1978) in support of my position. The politics of art cannot, in any case, be read off from formal criteria alone, a fact that postmodernists carefully avoid in their ahistorical condemnation of the failures of "high modernism" and their tendency to obscure profound differences in the social and political context of European art during the interwar years and the Cold War. For the aesthetic Right, see Jameson 1979; Kaplan 1986; and Herf 1984. And for the capitalist assimilation and domestication of modernism after World War II, see Guilbaut 1983 and Allen 1983. For a more inclusive conception of modernism than the one defended here, see Calinescu 1987; and for the sweeping changes in technology and culture that created new modes of understanding and experiencing time and space between 1880 and World War I, see Kern 1983.

11. For the social and political context of the scientific revolution and its philosophical and religious consequences, the works of James R. Jacob and Margaret C. Jacob are innovative as well as enlightening; for a general survey of the historical process by which scientific knowledge became an integral part of Western culture and a convenient synthesis of their approach, see Margaret Jacob (1988).

12. For useful accounts of Structuralism, see Hawkes 1977; Jameson 1972; Culler 1975; Coward and Ellis 1977; and Merquior 1986. For poststructuralism and the broader philosophical and literary context, see Anderson 1984; Dews 1987; Descombes 1980; Eagleton 1983; Lentricchia 1980; and Gasché 1986 (the best study of Derrida and his philosophical project).

13. This is the fatal flaw of post-Marxists Ernesto Laclau and Chantal Mouffe's concept of "democratic revolution" (Laclau and Mouffe 1985). Laclau and Mouffe define political and ideological discourses as free-floating, autonomous systems unrelated to economic determination and class positions and defend a neo-Crocean view of history as the "story of liberty" driven by a desire for democracy inherent in human nature and independent of social determinations. The concept of "democratic revolution" is, for Laclau and Mouffe, a Sorelian myth that can and should be constructed "autonomously" by ideological and political means without reference to the economy or social classes. Such an irrationalist view of discourse is not only unable to explain why it is

Page 370

we have no democracy (in the only meaningful sense of the word: popular control over the means of production and its distribution) but is also unable to move beyond the level of postmodern populist sloganizing or even to begin to articulate the material conditions for an alternative vision. Populism, Laclau argues in an earlier book (Laclau 1977), is a "popular-democratic" discourse that pits the interests of "the people" against those of the "power bloc" without reference to concepts of class. However, because Laclau stubbornly rejects the notion that political practice is assigned its "relatively autonomous" place and function by the matrix effect of the mode of production (a position Laclau dismisses as "reductionism''), he cannot accept the fact that the categories of populist discourse are determined, in the last instance, by economic structures and relations. Populism is always irrationalist insofar as class struggle is never explicitly present in populist discourse but rather masked behind collective political categories such as "the people." But populist irrationalism may be grasped rationally insofar as populist discourse is an indirect effect of economic class determinations refracted through ideological and political structures.

Populisms, democratic or otherwise, are politically unstable (and dangerous) precisely because they lack the very grasp of economic determination and class power that Laclau and Mouffe seek to replace with vulgar pluralism, relativism, and individualism. As Ellen Meiksins Wood points out (Wood 1986), Laclau and Mouffe cannot tell us who, in particular, might want or need democracy, whether some kinds of people might want or need more—or different aspects—than others do, how a social force capable of bringing it about might come into being, or indeed why there should be any difficulty or conflict about it. That we need a new, hopeful vision of democratic socialism is undeniable; that we will achieve it by a post-Marxist, postmodern lobotomy performed on the critical faculty of social theory is ludicrous.

Laclau and Mouffe, along with Barry Hindess, Paul Hirst, and others, make up a post-Althusserian variant of post-Marxist postmodernism that has rejected Structural Marxist concepts of structural causality, relative autonomy, and social class in order to embrace political voluntarism and philosophical irrationalism (see my discussion of Hindess and Hirst in chapter 1). The most thorough-going critic of the post-Althusserians and their concept of democracy is Ellen Meiksins Wood (1986). For a brief, lucid, and devastating critique of representative democracy under capitalism, see Macpherson 1977 and 1973; see also the powerful arguments for socialism put forward by Levine (1988; and 1987); Miller 1990; and for equality put forward by Baker (1987).

1. The reception of Althusser's ideas within the historical profession has frequently been negative and ill informed. See E. P. Thompson's self-indulgent polemic (Thompson 1978) for an unfortunately representative example. Thompson's diatribe should be read along with Perry Anderson's patient rejoinder (Anderson 1980). It is ironic, to say the least, that Thompson's own formidable historical research can be effectively marshaled against his theoretical humanist philosophy of history—and in support of the very Althusserian con-

Page 371

cepts that Thompson so rabidly condemns (see Anderson's examples, 31-49, 69-71). For a comparison of the Thompsonian and Althusserian schools, see Nield and Seed 1979. Hirst 1979a is a thorough critique of Thompson from within the post-Althusserian camp; a defense of Thompson's conception of class consciousness is offered in Wood 1982. For a critique of the rejection of class struggle by post-Marxist historians, see Resch 1989a.

Also of interest is the important article by Pierre Vilar (1973), one of the few Marxist historians currently working within the Annales school. Vilar's article is disappointing, however, because it engages in the dubious tactic of arguing against Althusser from a position that is nebulous and largely unspecified, a "black box" that Vilar passes off as somehow self-evidently real history. One gets the feeling that included among Vilar's unspoken premises are many that he would rather not put forward openly, perhaps because Althusser so devastatingly criticized them. In addition, Vilar's interpretations of Althusser's ideas are often dubious. I will confine myself to the central section of Vilar's essay, where he defends Ernest Labrousse and Fernand Braudel and their usage of historical time from Althusser's criticism. Althusser's "criticism" is largely a positive acknowledgment of the fact that these historians were beginning to observe differential temporalities in history. He is critical, however, of the fact "that they do not pose them explicitly as a function of the structure of the whole" (Althusser and Balibar 1970, 96). This is in fact the case. The three economic cycles put forward by Labrousse in his famous study of the French Revolution are undoubtedly important structural factors. There is nothing in them that Althusser would have any quarrel with, although this is implied by Vilar's misreading of Althusser's concept of differential time. The problem, for Althusser, is not so much what Labrousse has demonstrated but simply that he doesn't go far enough, that he fails to situate his three differential times in relation to the overall structure in dominance within which they occur. It is not that such an extension is impossible or even that Labrousse's work is not an important initial step in achieving such a task. It is the fact that he doesn't do it that constitutes the problem, which is all Althusser ever said. Braudel's monumental work on the Mediterranean region has often been faulted even by traditional historians on similar grounds: its three strata never achieve any significant synthesis into a structure. For further discussion of these matters, see Stoianovich 1976; and for a comparison of the Althusserians and the Annalistes , see D'Amico 1973.

2. Perry Anderson points out that the term social formation marks a space between the term society , the standard concept of classical sociology, and the Marxist concept of mode of production . The former term is unsatisfactory because it suggests a coherence and unity that reduces social contradictions to "dysfunctions" from functional norms, while the latter term has been too often used to justify the reduction of social contradictions to reflections of economic phenomena, a tendency that ignores the relative autonomy not only of other social practices but also of the economy as well; see P. Anderson 1980, 67. For a masterful comparative account of the development of classical sociology and historical materialism from a Structural Marxist perspective, see Therborn 1976.

3. See P. Anderson 1976a, 64-66; Patton 1978; and Elliott 1987. These

Page 372

misunderstandings persist despite Althusser's own comments on his relation to Spinoza (Althusser 1976, 132-41, 187-93) and, in the case of Elliott's more recent book, Macherey 1979. My own understanding of Spinoza is indebted to Hampshire 1987 and Curley 1988, as well as Macherey's interesting, if controversial, interpretation.

4. This view originates with A. Glucksmann 1978; see esp. 286-90.

5. Lenin (1975, 648-51) introduces the phrase "unity of opposites" but fails to distinguish sufficiently between the Hegelian term identity and the Marxist concept of unity . Althusser defines the "identity of differences" in terms of uneven development, an anti-Hegelian conception shared by Godelier (1972, 86-92) who also rejects the internality of contradictions in favor of their externality—contradictions between rather than within structures.

6. The term overdetermination is, of course, borrowed from psychoanalysis. Freud uses it to express the fact that unconscious formations can be attributed to a plurality of determining factors. For Freud, the term has two senses: first, it indicates that the formation in question is a result of several causes, since one alone is not sufficient to account for it; second, it expresses the fact that the multiplicity of unconscious elements that make up a formation may be organized in different meaningful sequences, each having its own specific coherence at a particular level of interpretation. Althusser was "not particularly taken by this term overdetermination" but used it "in the absence of anything better" (Althusser 1969, 101). Presumably, Althusser was attracted by Freud's effort to avoid extremes of reductionism and pluralism in his analysis of the unconscious and the drives. For a brief comparison of Althusserian concepts and those of Freud, see Miriam Glucksmann 1974, 99-103. Unfortunately, Glucksmann fails to consider the work of Jacques Lacan and its important influence on certain of Althusser's ideas, most notably the latter's theory of ideological interpellation but also the concepts of displacement and condensation. My account of Freud's use of the term relies on that given in Laplanche and Pontalis 1980, 292-93.

7 For these debates and their context, see L. Rosen 1971; Kurzweil 1980; Descombes 1980; Hirsch 1981; and above all, Poster 1975, which, despite its idiosyncratic attempt to marry Althusser and Sartre and its dubious assessment of the impact of the events of May 1968, remains the best survey of French intellectual developments between 1945 and 1968. For Althusser and the debates over humanism within the communist party, see Kelly 1982. Althusser's position is thoughtfully defended in Mepham 1985.

8. For critical discussion of Althusser's framework, see Buci-Glucksmann 1976; Renaut 1983; and Gerratana 1977. Grahame Lock's introduction to Althusser's Essays in Self-Criticism elaborates and extends Althusser's conception. Bettelheim (1975; 1976; 1978; 1982; 1983) provides a historical and theoretical analysis of Soviet society influenced by Althusser. See also the interesting discussion of the "technological voluntarism" of "proletarian biology" in Le-court 1976. It should be mentioned that the Althusserian School did not originate the "state capitalist" critique of the USSR. Credit for this critique, I believe, goes to British ex-Trotskyist Tony Cliff (1974; first published in 1948) and to the various American writings of C. L. R. James and Raya Dunayevskaya, also former Trotskyists, dating from the late forties and early fifties.

1. For important debates on the concept of a mode of production, see Hindess and Hirst 1975; Wolpe 1980; Foster-Carter 1978; and J. Taylor 1979. For discussions of slave-tributary mode of production (not discussed here), see P. Anderson 1974a; Saint-Croix 1981; Wood 1988; Garlan 1988. For the Asiatic mode of production, see Krader 1975; Bailey and Llobera 1981; Godelier 1978; and P. Anderson 1974.

2. For the primacy of "technological determinism" see Cohen 1978; for the primacy of "class struggle" see Robert Brenner's articles in Aston and Philpin 1985. Both arguments reduce the forces and relations to technological development and class struggle in order to separate them and privilege one over the other. Once it is made clear that it is the unity of the forces and relations that constitutes the class struggle at a given level of economic development, then it also becomes clear that Cohen's defense of technological determinism establishes not the primary of the forces over the relations of production but rather the dominance of a more advanced mode of production over a less advanced mode. Similarly, it becomes clear that Brenner's defense of the primacy of class struggle implies the dominance of relations of production and the class exercising economic ownership within a given mode of production and not the primacy of the political over the economic function nor the freedom of ruling or exploited classes to leap ahead of the existing forces and relations that as a unity constitute their powers and define their interests.

3. Conceptualizing the labor process in this way opens up its political and ideological dimensions, what Michael Burawoy has aptly called "the politics of production," for investigation; see Burawoy 1979; 1983.

4. See Dupré and Rey 1980 for a penetrating critique of "formalist" and "substantivist" perspectives within economic anthropology. Formalists envision primitive societies in terms of neo-classical economic concepts derived from capitalist societies (marginal utility, maximization of scarce resources, and so forth). Substantivists, such as Karl Polanyi and his followers, insist on the anachronism of formalist methodology and restrict the term economics to capitalist market societies. For the substantivists, kinship, religion, and politics must be employed to explain the "economies" of pre-capitalist societies. Structural Marxists, such as Dupré and Rey, contend that both the formalists and the substantivists deal only with surface phenomena ("the market," kinship, religion, and so forth) and fail to explain pre-capitalist societies in terms of underlying functions and structures, most fundamentally, the relations of ownership and labor process that define their modes of production. For Dupré and Rey, exchange can be understood only relative to the mode(s) of production involved in the exchange.

5. Althusser has been frequently, and rather uncritically, lumped together with Structuralists such as Lévi-Strauss. While Althusser acknowledges a "flirtation with Structuralist terminology" in his early works (1976, 126), the comparison can easily obscure more than it reveals, as is often the case with Miriam Glucksmann's comparative study of Althusser and Lévi-Strauss (Glucksmann

Page 374

1974). For other useful accounts, see Jameson 1972; Coward and Ellis 1977; and Merquior 1986. For a Structural Marxist critique of Lévi-Strauss, see Godelier 1972, vii-xlii; 1977, 15-98. See also Sebag 1964 and the uncompromising, if perhaps excessively naturalist, critique of Structuralist anthropology in Harris 1979, 165-215.

6. Balibar's self-criticism goes so far as to throw out the baby with the bath-water, rejecting along with rationalism the utility of any cross-cultural concepts, including the concept of a mode of production:

In such a perspective [the concept of a mode of production], the very designation of "instance" in the social formation can only lead to the designation of further elements , invariant essences of historical analysis . . . pre-existing the process of their historical transformation. . . . This means . . . that the term "economic" will have the same meaning in the feudal or capitalist mode of production, and in fact in any mode of production whatsoever. In short, it is the risk of a return to the ideological presuppositions of political economy and bourgeois historiography. (E. Balibar 1974, 230-31)

Such a dramatic retreat into nominalism and historicism ignores the fact that concepts pitched at different levels of generality are equally valid—"abstract" concepts of general functions (economic, political, and ideological) produce knowledge in exactly the same way as do "concrete" concepts of particular social formations (French capitalism during the Second Empire). The general concept of the economic function and the particular concept of the economy of the Second Empire are both concepts; neither is more "true" or more "real" than the other. We must, of course, avoid treating ''the economic" as an idealist essence existing apart from concrete social formations and imparting to them its reality in an approximate and imperfect form. However, rejecting idealism does not oblige us to abandon general concepts of social functions actually existing in every human society albeit in qualitatively distinct structures and institutions.

Balibar's born-again enthusiasm for the particular is inseparable from the voluntarist espousal of "class struggle" embraced by him and Pierre Macherey during the early seventies. Inspired by a distorted understanding of Maoism, such voluntarism bears within it the seeds of a further move, toward either neo-liberalism (e.g., Glucksmann, Hirst) or postmodern dissidence (Rancière, Laclau). The question of general concepts raised by the "new historicism" of Balibar and Macherey is discussed in chapter 5.

7. On the "theoreticism" of Hindess and Hirst, see John Taylor's two-part review of Pre-Capitalist Modes of Production (Taylor 1975, 1976). For a brief discussion of the post-Althusserian view of the autonomy of discourse and its relation to democracy, a view most fully developed in the work of Ernesto Laclau and Chantal Mouffe (1985), see "Introductory Conclusion," note 13.

8. We cannot trace the evolution of Marxist anthropology in France or even the various divisions within the ranks of the Structural Marxists themselves. Although one must avoid attributing too much homogeneity to the latter, it is hard to avoid the conclusion that most of the (often nasty) disagreements between them are motivated by concerns other than those of science. Meillassoux and Godelier initially shared a view of "primitive" societies as economic sys-

Page 375

tems to which ahistorical, neo-classical economic concepts were inapplicable, and both have been reluctant to associate their work with that of Althusser despite the obvious influence of the latter on their work. Meillassoux remained suspicious of the "theoreticism" of Althusser even as he came to take a position progressively less distinguishable from Althusser's own. Godelier initially shared the basic Althusserian framework of structural causality, mode of production, and so on, but he retained a positive view of Lévi-Strauss as well. In his later work, Godelier moves away from the concept of the social formation as an articulation of distinct structural functions and toward a more essentialist view of the structured whole within which functional distinctions have no meaning and thus no hierarchy. Godelier also places greater value on mental structures than do other Structural Marxists, although he continues to insist that such mental structures adapt to changing modes of production. Godelier's earlier views are presented in Godelier 1977, and his more recent development is evidenced in Godelier 1986. For an excellent overview of Marxism and anthropological theory, see Bloch 1985. For French developments in particular, see Kahn and Llobera 1981, 263-329; and Jean Copans and David Seddon, "Marxism and Anthropology: A Preliminary Survey," in Seddon 1978. Valuable collections of essays, in addition to those edited by Kahn and Llobera and Seddon, include Bloch 1975 and Friedman and Rowlands 1977. See also Dialectiques 21 (1977), a special volume devoted to Structural Marxist anthropology.

9. Meillassoux 1978; 1964; see also 1980.

10. I concentrate on Marxism and "Primitive" Societies in order to contrast Terray's initial defense of the primacy of the forces of production, a view that has had considerable impact independent of Terray's later retractions, with the rival views of Rey (emphasizing the relations of production) and Meillassoux (emphasizing reproduction as well as production). The subsequent development of Terray's views on dominance and subordination of articulated modes of production, in the context of the relationship between trade and the constitution of political authority in West Africa, can be followed in Terray 1974; 1975; 1977; 1985.

11. For further discussions of Rey, see Foster-Carter 1978; Bradby 1980; and Brewer 1980. The literature on Rey suffers from the fact that critics tend to concentrate on one aspect of his work—the lineage mode of production, imperialism, the transition from feudalism to capitalism, and so on—without demonstrating an awareness of the existence of the others or of the overall position that integrates all of them into a single problematic.

12. I do not wish to exaggerate the opposition between modes of production analysis and the global system approach of dependency theory. The problematic I am defending here has a place for both levels of analysis; indeed, despite important and obvious differences between national, regional, and global structures, this approach insists on the necessity of analysis of each of them for exactly the same reasons it insists on different structural levels of analysis within individual social formations. The literature on dependency and the global capitalist system is, of course, enormous. It is perhaps most helpful to refer the reader to Brewer 1980, who cogently compares the positions of André

Page 376

Gunder Frank, Immanuel Wallerstein, Ernesto Laclau, Giovanni Arrighi, Arghiri Emmanuel, Robert Brenner, and Samir Amin and provides a useful bibliography as well. Stern 1988, a critique of Wallerstein, surveys an enormous amount of recent work from a mode of production perspective.

13. The following discussion draws on debates over the nature of the feudal mode of production and the transition from feudalism to capitalism inaugurated by Maurice Dobb's pioneering mode of production analysis (Dobb 1963). For major collections of articles pertaining to these debates, see Hilton 1976; Aston and Philpin 1985; Aston 1965. For their special relevance to the application of the concept of mode of production to feudalism and the transition to capitalism, see Kula 1976; Kriedte, Medick, and Schlumbohm 1981; de Vries 1976; Wallerstein 1974; Wallerstein 1980; Wallerstein 1983; Wallerstein 1989; Porchnev 1972; Ladurie 1969; and Lublinskaya 1968.

14. For a critical discussion of Rey's analysis of the transition, see Cutler and Taylor 1972. Robin 1970 and Postel-Vinay 1974 attempt to apply Rey's framework to eighteenth-century France. Comninel 1987 criticizes Rey, Robin, and Postel-Vinay in a work flawed by the author's misunderstanding of the Structural Marxist concepts. Comninel's revisionist criticism—namely, that traditional Marxist interpretations of the French Revolution apply the term capitalism too loosely to forces and relations that are still feudal in character—has some validity against Rey (an Africanist after all) but misfires when applied to Robin and Postel-Vinay, who are, in fact, quite careful about such matters. In any case, this is an empirical, not a theoretical, question, and its resolution has no impact on the validity (or invalidity) of the Structural Marxist conception of articulation and historical transformation, about which Comninel hasn't a clue. Finally, Comninel's assertion that the "ancien régime . . . shows no sign of either developed capitalism or its emergence" (1987, 192) is an indefensible overstatement. Not only does Comninel fail to shake Postel-Vinay's persuasive interpretation of the capitalist nature of the relations of production introduced by the fermiers , but he also ignores the breakthrough of capitalist relations initiated by nobles in Toulouse and other regions. Comninel likewise ignores the undeniable proto-industrialization of the countryside of northern France and, even more seriously, fails to take into account the work of Bois, Kriedte, and others regarding the logic of feudal accumulation and the emergence of capitalism. My own account of these positions, whatever its shortcomings, is sufficient to refute the theoretical criticisms that Comninel directs at Structural Marxism. Furthermore, I would argue that it is precisely from the Structural Marxist position of Rey, Robin, and Postel-Vinay that a new framework for "rethinking" the French Revolution in light of the revisionist challenge is in the process of emerging.

15. Robert Brenner has engaged Bois in yet another chicken-egg controversy over "politicism" and "economism" in articles published as part of the so-called Brenner Debate (and collected in Aston and Philpin 1985). In his response to Brenner's critique of the neo-Malthusian interpretation of the crisis of feudalism from a "class struggle" perspective, Bois accuses Brenner of "politicism" and of ignoring the structural dynamics of a feudal mode of production. Brenner responds by consigning Bois to eternal damnation for the sin of

Page 377

"economism." Brenner's arguments against Bois are considerably weaker than those he marshals against his other critics, not least because of the straw-man tactics he employs. Brenner consistently refuses to acknowledge Bois's recognition of the "political" nature of the feudal levy and the inherent class struggle over it, as well as Bois's arguments regarding the possibility, indeed, the inevitability, of aristocratic political centralization as an outcome of this class struggle. Brenner's own emphasis on seigneurial political power and organization, although much more nuanced in his final essay (largely because he takes over much of Bois's own position, as well as that of Perry Anderson, who is not even cited by Brenner) continues to reify political power. Brenner, following a path already taken by Rey (who is also not cited), treats class power as if it were somehow independent of the forces and relations of production. How can class be distinct from production when it is the forces and relations of production that constitute classes? Brenner makes an interesting and persuasive contrast between France and England as if he were thereby refuting Bois's general concept of feudalism. All Brenner has really accomplished, however, is an identification of the degree of variation of the English and French cases, that is, their uneven and combined development within the logic of feudalism, something Bois himself explicitly recognizes. Brenner's central argument—that because different outcomes can be demonstrated from similar feudal conditions, the independent variable must be class power, not the mode of production—is simply specious. If English lords had significantly more class power than did their French counterparts, they had it simply because the forces and relations of production in England were significantly different from those of France.

16. The debate has many dimensions; the central division, however, is between the Weber-Pirenne-Polanyi-Wallerstein position (which asserts first the inherently capitalist nature of cities, merchants, and markets and second the primacy of cities and trade in the evolution of capitalism) and the Marx-Dobb-Bois-Brenner position (which argues, if not for the primacy, at least for the importance of agrarian transformation and which denies the ahistorical view that all economic exchange is capitalist in nature). These debates, of course, echo those between formalists and substantivists, Marxists and Weberians, mode of production analysis versus dependency theorists, and so on. Kriedte follows Dobb in emphasizing the transformation of urban feudal classes (merchants and artisans) into capitalist classes (owners and workers) in articulation with agrarian transformations from seigneurs and peasants to landowners, tenants, and wage laborers.

1. The argument of the present chapter is deeply influenced by Bhaskar 1978; Bhaskar 1979; Ruben 1979; and Newton-Smith 1981. Each of these works provides a searching critique of contemporary philosophy of science, and all, by different yet convergent paths, attempt to defend scientific realism and the rationality of science without recourse to epistemological absolutism. Although none deals explicitly with Althusser (Bhaskar being a closet Althusserian), I am encouraged to find considerable similarities between the argu-

Page 378

ments advanced in these works and the positions elaborated earlier by Althusser. Although I dissent from the stronger claims made by Bhaskar for his "transcendental realism" (I find the arguments made by Newton-Smith and Ruben more persuasive), I know of no better philosophical elaboration of the strengths of Althusser's philosophical position than the one implicit in Bhaskar's Possibility of Naturalism (1979). Since he expresses Althusser's position better than Althusser himself does, I have applied Bhaskar's terms ontological realism and epistemological relativism to Althusser's materialist theses (which are obviously the source of Bhaskar's original inspiration anyway). For Althusser's philosophy of science, I am also indebted to two excellent articles, O'Hagan 1981 and Gordy 1983; and, of course, to Lecourt 1975, a brilliant Structural Marxist survey of Bachelard and Canguilhem, which is equally informed with regard to Althusser's philosophical position. Finally, I am indebted to Kolakowski's lucid history of positivism (1968) and Novack's trenchant critique of Dewey's pragmatism (1975).

2. Even the best discussions of Althusser fail to explore satisfactorily the connections between his early and later works. Benton 1984 glosses over the implications of Althusser's new view of philosophy for the concept of science; Callinicos 1982 simply dismisses the problem by asserting that Althusser's final position (which Callinicos sees correctly enough as subordinating epistemology to historical materialism) is incoherent because it continues to employ a science/ideology distinction. Callinicos does not seem to see that Althusser's final assertion of a science/ideology distinction has dropped all pretence of epistemological absolutism, nor does he grasp the import of Althusser's distinction between the substantive concepts of science and the logical categories of philosophy. Oddly enough, Callinicos proceeds to develop his own solution (based on Lakatos), which defends historical materialism on the grounds of its productivity, precisely the move already made by Althusser, but he lacks Althusser's awareness of the need to take a materialist and realist position that explains and defends such knowledge effects in the Kampfplatz of philosophy.

3. Balibar provides an excellent discussion of this, marred however by his contention that historical materialism can have no "general concept" of an epistemological break (E. Balibar 1978, 231). We will deal with Balibar's critique of general concepts (an excessive reaction to Althusser's rejection of theoreticism) in chapter 5.

4. Most controversial of these accusations is the charge of Stalinism. E. P. Thompson accuses Althusser of "an attempt to reconstruct Stalinism at the level of theory" (Thompson 1978, 323), while Alvin Gouldner sees him "providing a theoretical storm-cellar for Stalinism" (Gouldner 1977, 450). Jacques Rancière, one of the coauthors of Lire le Capital , maintains that Althusser's initial project, to undermine the Party bureaucracy and its stranglehold on theory, dissipated itself with the debates over "humanism" wherein Althusser's defense of the autonomy of theory, in a backhanded fashion, ended up serving the interests of the Party leadership; see Rancière 1974. A more restrained critique of the political implications of Althusser's theory may be found in Gerratana 1977 and 1977a; see also the wide-ranging discussion in the special Althusser issue of the French journal Dialectiques, 15-16 (Autumn 1976). For

Page 379

a defense of Althusser against the charges of Stalinism, see P. Anderson 1980, chap. 4; and Benton 1984, 14-23.

5. This position is most intelligently presented in Callinicos 1976. For Callinicos, the solution is to abandon once and for all the notion of ideology as the other of science and to retain only the concept of ideology as the outcome of material practices. This is ultimately an irrationalist position, denying the concept of science and the specificity of scientific discourse (or any other discourse for that matter). Such a move also ignores the fact that philosophical debates over science and Marxism do take place and that without a concept of science as distinct from ideology/error a realist and philosophical position cannot be formulated. Callinicos argues that we should pursue Althusser's insights into the materiality of ideology and eliminate the "epistemological relics" of his early works. But where does this leave Marxist philosophy? Ideology is what social agents "think in" or "represent themselves in," but it is the economic and the political that they think about . Because Marxism defines the economic and the political in a different way, the distinction between science and ideology cannot be dispensed with theoretically; Marxism not only explains ideology as socially produced ideas and beliefs but also combats those ideas and beliefs as erroneous (ideology/error). Thus it is not, as Callinicos seems to think, a matter of simply "leaving behind" the relationship of ideology and science and the functional distinctions between them, for Marxism cannot conceptualize science without ideology and without ideology/error, nor can Marxism philosophically defend its own knowledge effects without taking a position defending the category of the scientific.

6. Descombes, in a popular survey of the French intellectual scene, argues that in his later works Althusser "did away with Althusserianism and restored the priority of the political over the theoretical instance" (Descombes 1980, 134). Descombes cites the discussion of Althusser's "Preface to Capital" (Althusser 1971, 71-101), in which Althusser attempts to explain why intellectuals have difficulty understanding Capital while workers do not. Althusser writes: "The former [intellectuals] are blinded by bourgeois ideology which does everything it can to cover up class exploitation. The latter [workers], on the contrary, despite the terrible weight of bourgeois and petty bourgeois ideology they carry cannot fail to see this exploitation since it constitutes their daily life" (Althusser 1971, 100). Descombes concludes from this statement that "in returning in this way to experience and the 'lived through' Althusser abandons the attempt to endow Marxism with an epistemological foundation and reverts to the phenomenological foundation which had previously been thought good enough" (Descombes 1980, 135).

Descombes is guilty of two errors of interpretation here. First, he seems to imply that in rejecting theoreticism in his later work Althusser is abandoning the relative autonomy of theory, a view that is demonstrably wrong. Second, the implication that Althusser is returning to phenomenological Marxism cannot be sustained even within the text Descombes has chosen to prove his point. Descombes fails to mention that, in the passage cited, Althusser is referring to the class instinct of the proletariat and not their class position, an important distinction since, for Althusser, the former is ideological (referring to lived ex-

Page 380

perience) while the latter is theoretical (referring to a concept). While class instinct is certainly an important factor in the receptivity of working-class subjects to Marxist theory (and an obstacle for bourgeois subjects), it is not a guarantee either that the theory will be received (or rejected) or that it will even be created: "Class instinct is subjective and spontaneous. Class position is objective and rational. To arrive at proletarian class positions the class instinct of proletarians only needs to be educated; the class instinct of the petty bourgeoisie, and hence of intellectuals, have, on the contrary, to be revolutionized" (Althusser 1971, 13). Descombes has simply missed the fact that the terms class position and class instinct are perfectly consistent with Althusser's (unchanged) position regarding the different modalities of ideology and science. The working class experiences exploitation but can attribute it to many different causes: the will of God, bad luck, natural scarcity, as well as capitalism; what they know about it, however, is something completely different.

7. The usual tactic is to dismiss Althusser's thought as no more than the reflection of a political line. We have already referred to this tactic in reference to the charge of Stalinism leveled against Althusser, but it is also evident in Elliott 1987, who mechanically plots the course of Althusser's intellectual development in terms of political phases—Leninism, Maoism, Eurocommunism, and finally the demoralizing break with the PCF. While there is no denying the general accuracy (and usefulness) of Elliott's political chronology, nor even the fervor of Althusser's faith in communist internationalism and the intensity of his disillusionment with the PCF leadership, there is no justification for reducing theory to politics as Elliott occasionally does. Althusser's political commitment to the PCF and to global communism always existed in an uneasy relationship to his theoretical enterprise. His political faith in the popular masses and his desire for a revolutionary party organization capable of imaginative leadership and developing a "mass line" was never easily reconciled with a scientific application of his own problematic to existing capitalist and communist social formations and the global economy. Perhaps, as Elliott maintains, Althusser blamed first the ossified Stalinist bureaucracy of the PCF and second his own theoretical enterprise for the party's failure to activate the "popular masses," whose existence he posited as an article of faith rather than a scientific hypothesis. Perhaps also the widening gulf between political faith and objective reality contributed to Althusser's personal tragedy. Even if such speculations were to be true, however, they would have no relevance to a theoretical assessment of Althusser's work or to theoretical debates over historical materialism.

8. Althusser's periodization of Marx's intellectual development is elaborated in Althusser 1969, and summarized in the introduction to that work. Althusser returned to the problem in light of his reformulation of the concept of philosophy in a 1970 essay, "Marx's Relation to Hegel" (in Althusser 1972, 163-86). For excellent accounts of Marx's work sympathetic to Althusser's point of view see Therborn 1976, 317-413, and Callinicos 1983, 26-60; my summary of Althusser's interpretation of Marx's development is much indebted to them. For another interesting assessment of Althusser's intentions in relation to Hegel and to the scientificity of Marxism, see Levine 1981.

9. Lecourt 1975, written from an Althusserian perspective, contains one of

Page 381

the best evaluations of Foucault's pre-Nietzschean works (which depend positively on Canguilhem and negatively on Althusser). For Bachelard in relation to Anglo-American philosophy of science, see Bhaskar 1975. For the relationship between Althusser and Bachelard, see Brewster 1971 and Balibar's more self-serving account, Balibar 1978. For an interesting application of the Althusserian problematic to the history of mathematics, see Raymond 1978; 1977; 1973. For another interesting application, this time focusing on the relation between Stalinism and science, see Lecourt 1977.

10. Thus the oft-repeated claim that Althusser's project was to create an "epistemological Marxism" is mistaken. In fact, Althusser attempted to rid Marxism of the onus of grounding its knowledge in a philosophical discourse outside itself. André Glucksmann 1978 (an article that originally appeared in Les temps modernes in May 1967) anticipated Althusser's self-criticism of theoreticism without acknowledging the realist dimensions of Althusser's initial position. Given the trajectory of Glucksmann's intellectual career, which culminates in neo-conservatism, such a one-sided critique is not surprising. It is rather more surprising to find the "epistemological Marxism" designation in Smith 1984. Smith's discussion, unfortunately, deals only with For Marx and Reading Capital , and then only with Althusser himself. I must also dissent from the interpretations of Althusser's epistemology presented in Elliott 1987 and Patton 1978.

1. On this point Althusser owes a great deal to Spinoza's concept of experientia vaga . Althusser credits Spinoza with the "first theory of ideology ever thought out in terms of (1) its imaginary 'reality'; (2) its internal inversion; (3) its 'center', the illusion of the subject" (Althusser 1976, 135).

2. Hirst 1979 criticizes Althusser for positing an impossible essence (the human individual) prior to the interpellated subject and for equating the concept of a social subject with a single human individual. With regard to the first point, Althusser's distinction between individuals and interpellated subjects is simply a heuristic device expressing the process by which human beings are constituted as social beings. Althusser has always insisted that the individual is always already a subject. What Hirst really objects to, and this is the thrust of his second point, is Althusser's claim that it is only human beings that are the locus of interpellations. Hirst rejects this "theoretical humanism" on Althusser's part and insists that economic enterprises are also socially constituted subjects, each having its own individual existence, rights, and powers. I would not wish to understate the reality or the power of corporations. However, I would argue that their social "subjectivity" is of secondary, not primary, significance: only human agents act.

3. The literature on Lacan is immense. Useful introductions to Lacan are Lemaire 1977; Bevenuto and Kennedy 1986; and Macey 1988. One should also consult two major collections of articles by or about Lacan published in Yale French Studies , volumes 48 and 55-56 (1972, 1977). Of particular importance in the context of the present work are Jameson 1977; Coward and Ellis

Page 382

1977; and Badiou 1982. Althusser's own support for Lacan's project is recorded in "Freud and Lacan" (in Althusser 1971, 189-219). Althusser became progressively estranged from Lacan's posturing, however, and finally denounced him as a "magnificent, pathetic Harlequin" (see Clément 1983, 20, 206-7 n. 14). My summary of Lacan follows that of Lowe 1982.

4. In The Dominant Ideology Thesis , Abercrombie, Hill, and Turner (1980) attack Althusser for this monolithic view of ideology. However, they completely miss Althusser's emphasis on the non-normative aspect of ideology, which opens the way to the subjection/qualification contradiction developed by Therborn. As for the dominant thesis of The Dominant Ideology Thesis —that ideology is not important, that it has never assimilated the exploited classes of any social formation, and that under capitalist modes of productions it is not even essential to the reproduction of the social relations of production (since capitalism encourages diversity and relies on economic compulsion to keep it within "safe" limits)—it is at least arguable that the valorization of the "free individual" by the book's authors (who see it as symptomatic of the decline of capitalism) is in actuality a manifestation of a dominant ideology rather than a refutation of its existence. For an extended critique of Abercrombie, Hill, and Turner, see Therborn 1984. For a careful elaboration of the dominant ideology of individualism and its contradictions within the present American context, see O'Conner 1984.

5. Althusser himself concedes this point in a personal communication to Therborn in which he states that he is "no longer prepared to defend the theoretical necessity of the notion ideological State apparatus," maintaining only that "the crucial point is to grasp the intrinsic link between the ideological apparatuses and the State" (Althusser, cited in Therborn 1980, 85).

6. For the New Philosophy, see A. Glucksmann 1980 and Levi 1982. See also the special issue of Chicago Review (volume 32, number 3 [1981]) devoted to the New Philosophy. For critiques to which I am especially indebted, see Dews 1985; Lecourt 1978; and Poulantzas 1978. It is important, I believe, to distinguish the gauchisme of Rancière from the neo-conservatism of the New Philosophers. While Rancière 1985 makes some cogent criticisms of Althusser's science/ideology distinction (as developed in the context of Althusser's essay "A propos de l'article de Michel Verret sur 'mai étudiant'" [1969a]), for the most part they are vitiated by Rancière's extreme voluntarism and irrationalism. Rancière rejects any attempt to defend the relative autonomy of scientific practice and instead collapses scientific practice wholly into ideology. Whereas Althusser attempts to distinguish between the technical and social divisions of labor within the university apparatuses—defending the value of the former while condemning the deformations of the latter—Rancière maintains that the university must be viewed as an oppressive unity, the object of class struggle, not as a contradictory apparatus internally divided by class struggle. While Althusser may or may not be guilty of understating the class bias of the university apparatus, there is no reason to follow Rancière and reduce science to ideology (to be replaced by what? proletarian science?). Rancière continues to privilege ideological struggle and working-class voluntarism in The Nights of Labor: The Workers' Dream in Nineteenth-Century France (Rancière 1989), a work interesting for its iconoclastic attack on predominant interpretations of

Page 383

working-class consciousness (its supposed desire to remake the world according to a working-class image rather than to change their lives in order to escape from a working-class existence); however, this effort—fully consistent with an Althusserian concept of ideology, by the way—is flawed by Rancière's unrepentant contempt for structural determination and scientific realism.

7. The French Nietzscheans Deleuze, Foucault, and Lyotard are critically discussed in Descombes 1980 and Callinicos 1982 (my account of Deleuze is based on that of Callinicos). For Foucault's appreciation of Deleuze, see Foucault 1980a, 165-96. For Deleuze's appreciation of Foucault, see Deleuze 1972. Donzelot 1979, one of the best works inspired by Foucault, amply demonstrates the affinity of Foucault and Deleuze.

There is a vast literature on Foucault. The "authorized" commentary is Dreyfus and Rabinow 1983; another sympathetic and thorough account is that of Foucault's translator Alan Sheridan (1980). More critical is Merquior 1985, which, unlike Dreyfus and Rabinow, gives full attention to the neo-anarchist Nietzschean element of Foucault's work. Merquior also provides a useful survey of the controversies that have emerged over Foucault's scholarship. For the relationship between Foucault and the New Philosophers, see Dews 1979 and 1984; Lecourt 1978; and Poulantzas 1978. For my part, I am building on the arguments of Dews and Poulantzas regarding the ontologizing and demonization of Power in Foucault's works of the seventies; see also Merquior 1985. On Foucault's early period, see White 1978 (my summary of Foucault's The Order of Things follows White's) and Dominique Lecourt's chapter on Foucault in Lecourt 1975. For Foucault's later work, see White 1979 and Megill 1979. For the influence of Deleuze on Foucault, see Callinicos 1982. Megill 1985 situates Foucault in the tradition of Nietzsche and Heidegger but is insensitive to the uniqueness of the French neo-Nietzschean Left represented by Deleuze. Both Deleuze and Althusser are, characteristically, absent from the pages of Rabinow and Dreyfus. Indeed, Foucault's negative dependence on Althusserian concepts (see Lecourt 1975) is a well-kept secret among Foucault enthusiasts in the United States. Sheridan acknowledges the possibility of Althusser's influence but angrily (if feebly) rejects it by repeating Foucault's own disclaimers (see Sheridan 1980, 214). On the postmodern Left, the uncritical acceptance of Foucault advances furthest in Poster 1984, where Foucault is depicted as the discoverer of a "mode of information" that has allegedly replaced the mode of production as the relevant critical concept for contemporary social theory. Without denying the importance of the technological revolution in information processing and control, I fail to see why it vitiates the concept of a mode of production any more than did the steamship revolution or the telegraph revolution. For another post-Marxist, postmodern interpretation of Foucault, see Smart 1983. Finally, for Foucault's relation to the tradition from Sade to Bataille, see Stoekl 1985 and Lash 1985.

1. My interpretation is directly opposed to that of Bennett 1979 and more or less compatible with Sprinker 1987; see also Kavanagh 1982. Macherey (1982) has come to acknowledge grudgingly the necessity for some concept of

Page 384

literary practice, while Eagleton waffles back and forth on his commitment to "Althusserianism"; see the preface to Eagleton 1986 and the interview (Eagleton 1982).

2. For an expanded critique of contemporary linguistics, see Pêcheux 1969; Pêcheux and Gadet 1981; Ducrot 1972; Henry 1977; and Faye 1972. All of these works share and develop a Structural Marxist approach to discourse and language that I am introducing through Pêcheux alone. For additional discussions of Pêcheux, see Macdonell 1986; Macabe 1979; and John Thompson 1984, a work flawed by its author's failure to acknowledge the common ground shared by Pêcheux, Faye, Bourdieu, and Althusser (Thompson self-servingly identifies Althusser solely with the work of Hindess and Hirst). Thompson deprecates, where he does not ignore entirely, the comprehensiveness and explanatory power of the problematic within which the particular works he discusses are situated in order to inflate the significance, in the minds of uninformed readers, of authors who reflect his own eclectic hodgepodge of phenomenology and critical theory.

3. On Gadamer and Reception Aesthetics, see Hoy 1978; Holub 1984; and Hohendahl 1977 and 1983.

4. It is more plausible that the historical variation of the "categories of literary reception"—the fact that certain texts are born literary while others become or cease to be so—indicates the presence of a certain type of objective, determinate cognitive activity, something like ideological production, around which aesthetic ideologies of reception and consumption constantly draw and redraw ideological lines of demarcation. Structural Marxism is able to conceptualize both the specific literary effectivity of the text (its autonomy relative to other signifying practices) as well as its determinate conditions of production and reception (its dependence on a variety of historically specific ideological practices). While such knowledge cannot produce an absolute or definitive interpretation of the meaning of the text in the objective idealist manner of Ingarden, Poulet, and Iser, it also cannot be mystified in the subjective idealist manner of Heidegger, Gadamer, and Jauss. Structural Marxism successfully avoids the aporia introduced by positing an "incommensurability" between production and reception in the manner of the so-called New Historicism of Greenblatt, Montrose, and company (see Veeser 1989). Against the New Historicism, we must protest that it is not enough to "juxtapose" text and context, literary and non-literary phenomena (a process that Dominick LaCapra aptly characterizes as weak montage), nor is it legitimate to hide inadequate explanations behind theoretical fig leaves such as the "irreducible complexity" of history, nor, finally, is it justifiable to parade an ideological desire to dissolve the distinction between knowledge and art—why? in order to sustain the notion of the critic as genius? to evade the political consequences of taking a firm position?—under the banner of textualizing history. The New Historicism may indeed answer such burning questions as how a professional middle-class academic from Berkeley might feel if he suddenly found himself in the body of Thomas More, but beyond legitimizing such thought experiments it is difficult to find any theoretical value in the New Historicist enterprise. (For an alternative, "Cultural Materialist'' view of the English Renaissance, a view informed

Page 385

by Structural Marxist concepts, see Dollimore 1984; Halpern 1991; and Dollimore and Sinfield 1985.) Nor, finally, is there much to recommend a Marxist version of reception aesthetics, which simply replaces the crude materialist objectivism of reflection theory with an equally crude materialist subjectivism of a historicist reception theory.

It is interesting how a general concept of literary production keeps popping up, even in the works of its avowed enemies. Raymond Williams (1977), for example, begins by attacking the validity of a concept of literature by virtue of its historicity but ends up extolling the practice of literature as the "human essence of creativity."

5. My résumé of Renée Balibar's books is much indebted to Bouché 1981.

6. Fredric Jameson (1981) has convincingly shown that semantic elements (which he calls "ideologemes") perform an essential function in the realization of the literary styles of Balzac, Gissing, and Conrad. I have not included an analysis of Jameson's important book here despite the strong influence of Althusserian concepts on it—not primarily because of space limitations, although these are prohibitive, but because of the many other influences on Jameson's work. For perceptive evaluations of Jameson, see Eagleton 1986 and Dowling 1984.

1. Poulantzas found in Althusser's concept of structural causality the key to the problems of political power and hegemony raised by Lenin, Gramsci, and Weber. Take away Althusser's problematic and Poulantzas's entire work becomes incomprehensible. The tendency of certain commentators (Carnoy 1984; Jessop 1985) to depict Gramsci and Foucault as the "true" references for Poulantzas—thereby reducing his Althusserian "phase" to a temporary aberration of no real significance—is a preposterous misrepresentation and insupportable from Poulantzas's own remarks (see Poulantzas 1980; 1976a; 1979). The names of Gramsci and Foucault have become traces of the erasure of Althusser's impact on social theory—and Poulantzas, unfortunately, has been used frequently as an instrument of this process. As I have already argued, Foucault is something of a bastard child of Structural Marxism; Poulantzas accepted in Foucault only what they both took over from Althusser. We have also seen how Poulantzas wholeheartedly rejected the neo-Nietzschean, post-modern, post-Marxist tendencies of Foucault. I am not attempting to minimize the importance of Gramsci for Poulantzas (or for Structural Marxism generally, for that matter). For the important interaction between them, see Buci-Glucksmann 1980; Macciocchi 1974; Mouffe and Sassoon 1977; Anderson 1976. For Poulantzas on Gramsci, see Poulantzas 1965; and on Althusser in relation to Gramsci, see Poulantzas 1966. The importance of Weber for Poulantzas has been overlooked in the secondary literature. There is a need for a systematic comparison of Althusser and Poulantzas in relation to Weber.

2. The subtlety of Poulantzas's concept of the matrix effect of a mode of production—the fact that it denotes the intransitive moment of structural cau-

Page 386

sality in opposition to the transitive moment of a particular practice—eludes most of the commentators, who throw up their hands in despair over the "circularity" of Poulantzas's thought (see, for example, Milliband 1973; Connell 1979). Ernesto Laclau's writings on Poulantzas (Laclau 1977) are superior to the rest of the literature in this respect, but even Laclau sees Poulantzas's framework as "unilateral," thus grossly misrepresenting the subtle indirect determination of the matrix effect as "class reductionism" and willfully dismissing the distinction between intransitive and transitive moments as merely "abstract formalism." While Laclau has a point—Poulantzas, like Althusser, assumes the determinant place of the economic function in all social formations and the dominant role of the instance which exercises the ownership function within the mode of production—to call this reductionism or formalism makes sense only if one is seeking to defend pluralist indeterminacy, irrationalist relativism, and political voluntarism. Laclau's hidden agenda, revealed by his subsequent intellectual development, is a post-Marxist, postmodernist attack on scientific realism and economic determination. Laclau's views on Poulantzas are taken over and developed by Jessop 1985.

3. See, for example, Bridges 1974 and Milliband 1973. This criticism misses the central point, namely, the fact that the limitation of the political by the economic is both historical and structural (the matrix effect of a mode of production and the concepts of social class and political practice) as well as conjunctural. The well-known exchange between Milliband and Poulantzas regarding the class nature of the capitalist state is rather sterile since Milliband's subject-oriented view of class power and Poulantzas's structure-oriented approach lack any common ground, neither party being able (or willing) to bring up the concepts of ideological interpellation, pertinent effects, or social class, which might bridge the distance between their respective positions. For the exchange, see, in addition to Milliband 1973, Poulantzas 1969; Poulantzas 1976; and Milliband 1970.

4. Briefly, Poulantzas argues that the accumulation of contradictions in Germany and Italy accounts for the emergence of fascism. Within the power bloc of Germany the primary source of contradictions was the rapid expansion and concentration of capital in a country where political hegemony still resided with the landed aristocracy—the Prussian Junkers—who carried out the Prussian "bourgeois" revolution of the nineteenth century. German monopoly capital required mass state intervention in its favor in order to compensate for the disproportionate political weight of the Junkers, yet the structure of the power bloc and the relative strength of the various non-monopoly groups within it were obstacles to such intervention. In Italy the situation was even more accentuated. The power bloc consisted of the industrialists of the north and the landowners of the Mezzogiorno, with the former establishing their hegemonic position by maintaining the feudal character of southern agriculture.

The rise of fascism also involves a political confrontation between the forces of the working class and the bourgeoisie, of course, but it is the petty bourgeoisie that plays the essential role in the coming to power of fascism. Political and economic crises dissociate the petty bourgeoisie from liberal capitalism, and "status quo anti-capitalism" becomes a dominant oppositional ideological

Page 387

theme. This ideological sub-ensemble "replaces" the dominant bourgeois ideology and "cements" the social formation back together (this is the decisive element of a fascist takeover, as opposed to Bonapartism or military dictatorship). The function of fascism, according to Poulantzas, is to bring into existence a form of the state capable of establishing and organizing the hegemony of monopoly capital in this particular set of historical circumstances and political-economic crises. Under these particular crisis conditions, the petty bourgeoisie is able to climb to the highest levels of political life, yet the crisis is finally resolved only by the neutralization of the petty bourgeoisie and the establishment of the hegemony of monopoly capital—the latter being the essential component of the nationalist military power desired by the petty bourgeoisie themselves.

According to Poulantzas, the formal separation of the state and the economy is characteristic of all capitalist social formations, not simply parliamentary democracies. Poulantzas has written two books about "exceptional" or non-parliamentary forms of the capitalist state, including military dictatorship in The Crisis of the Dictatorships and fascism in Fascism and Dictatorship (Poulantzas 1976a; 1974). He examines a third form, "Bonapartism," in Poulantzas 1973. For the best discussions of Poulantzas's views of fascism, see Laclau 1977; Jessop 1985; and Faye 1973; see also Caplan 1989. Abraham 1986 is a suggestive and stimulating attempt to extend Poulantzas's line of investigation into the political dynamics of the rise of German fascism. The revised edition corrects certain controversial errors of quotation and includes a retrospective defense of the original argument and a response to critics. See also the important collection Dobkowski and Wallimann 1989. From an immense literature, I will mention only two brilliant works of synthesis that independently corroborate and complement Poulantzas's analysis, Wehler 1985 and Broszat 1981.

5. Poulantzas's work only suggests the fruitful potential of an application of Althusserian concepts of ideology, ideological interpellation, and habitus to the study of nationalism. They assimilate and transcend the cultural functionalism of Geertz 1973 and Gellner 1983. See, in particular, B. Anderson 1983; Nairn 1981; Howell 1986. For overviews of traditional Marxist views of nationalism, see H. Davis 1980, 1967.

6. For the creation of the post-war "American Century"—the European recovery, the Cold War, the Fordist state, and so on—in relation to the development of global capitalism, see Kolko and Kolko 1972; Block 1977; Pijl 1984; Mandel 1986; Mandel 1970; and Schurmann 1974.

7. Aglietta, Christian Palloix, Alain Lipietz, and others are all French economists and members of the so-called école de régulation concerned with capitalist regimes of accumulation and the global economy. Students of Althusser who have been significantly influenced by the Structural Marxist concept of a mode of production, the authors of the Regulation School are gaining increasing recognition as an alternative to the banalities of postmodernism and the shortcomings of dependency theory. In addition to Aglietta 1979, see Palloix 1972; Palloix 1976; Lipietz 1977; Lipietz 1983; Lipietz 1987; and Aglietta 1982. For a critical evaluation, see Brenner and Glick 1991.

8. For global restructuring and its effect on the uneven development of me-

Page 388

tropole and periphery, see Mandel 1975; Lipietz 1987; Warren 1980; Amin 1980; and J. Kolko 1988.

9. For a lucid and systematic analysis of the present condition, see J. Kolko 1988. Lash and Urry 1987 is illuminating in its historical survey of the "organized" capitalist states, but its Habermasian thesis that capitalism is becoming "disorganized" is unconvincing; more persuasive is D. Harvey 1989, which analyzes postmodern culture in relation to global economic restructuring and the transition from a Fordist to a "flexible" regime of accumulation. Mention may perhaps be made here of the influential work of Structural Marxist urban geographers such as David Harvey and Manuel Castells, who have done much to clarify the importance of urbanization for capitalist accumulation and the significance of social space for capitalist domination; see, in particular, D. Harvey 1989a; and Castells 1977.

10. Of course Poulantzas's work hardly exhausts the field of Marxist and neo-Marxist analysis of the state in contemporary capitalism. At the very least, mention must be made of Wolfe 1977; O'Conner 1984; Offe 1984; Offe 1985; and Gross 1982. For the important work of Joachim Hirsch and the German "state derivation" ( Staatsableitung ) debates of the seventies, see Holloway and Picciotto 1979. Useful surveys of new theoretical developments include Carnoy 1984 and Jessop 1982.

11. The phrase is Stuart Hall's. Hall 1988 and Leys 1989 are superb analytical accounts of the ideological and social-structural basis of Thatcherism. For Reaganism, see the equally perceptive analysis of Mike Davis 1986.

12. Poulantzas justifies this designation by arguing that the "middle class" of white-collar workers, technicians, supervisors, and civil servants is a recently emerged fraction of the petty bourgeoisie. His argument is based on the concept of social class—the fact that today's class relations bear the mark, via the matrix effect, of yesterday's ensemble of political, ideological, and economic relations. He argues that "certain groupings which at first sight seem to occupy different places in economic relations can be considered as belonging to the same class [the petty bourgeoisie] . . . because these places, although they are different, nevertheless have the same effects at the political and ideological level" (Poulantzas 1975, 205). In effect, Poulantzas maintains that despite having different positions in the forces and relations of production, the new traditional petty bourgeoisies have similar political and ideological positions, and these similarities justify defining them as fractions of the same social class.

At first sight, Poulantzas appears to have reversed the causal relationship between social classes and their pertinent effects in order to assert the predominance of political and ideological effects over economic relations. While similar ideological and political positions today will participate in the constitution of the matrix effect of tomorrow, Poulantzas has apparently lost sight of the fact that these political and ideological positions will not constitute the matrix effect by themselves, but only in their articulation with economic relations whose modes of determination are dominant within the capitalist mode of production. The apparent confusion of pertinent and matrix effects is corrected in a later article in which Poulantzas forcefully reasserts the primacy of economic relations: "What are social classes in Marxist theory? They are groups of social

Page 389

agents . . . defined principally but not exclusively by their place in the production process , i.e., by their place in the economic sphere. The economic place of the social agents has a principal role in determining social classes. But from that we cannot conclude that this economic place is sufficient to determine social classes . . . the political and the ideological also have an important role" (Poulantzas 1973a, 27).

Poulantzas's attempt to incorporate political and ideological relations into the matrix effect constituting social class should never have been construed as a denial of the primacy of economic relations for the simple reason that such a reading renders the entire body of his work incomprehensible. Unfortunately, however, it is precisely such a reading that opened the door for a post-Althusserian, post-Marxist misappropriation of the concept of social class such as that of Laclau 1977, who criticizes Poulantzas's "economic reductionism" while developing a concept of social class based on the autonomy of ideology and the primacy of politics with respect to economic relations. This approach reaches its logical conclusion in Laclau and Mouffe 1985, who treat history as a "story of liberty" with the discourse of "democracy" as its historical motor. Laclau and Mouffe advocate a new ideological offensive "cutting across" class lines and deploying "democracy" as a neo-Sorelian, populist myth of "liberty as well as equality." Laclau and Mouffe assert an irrationalist individualism combining the most specious aspect of Saussurean linguistics (there is no social reality, only the differential reality of discourse) with a perverted form of Lacanian psychology. Lacan's concepts of the Real (desires, needs, and feelings of the individual), the Imaginary (the psychological process of identification), and the Symbolic (the discursive codes and practices of society) refer to an objectively existing tension between the psyche of a determinate individual and a determinate social reality. Laclau and Mouffe transmogrify Lacan's problematic in order to assert a free-floating relationship between the Symbolic and the Imaginary from which the Real of the individual psyche and the reality of social structures are equally absent. From here it is a simple matter for them to reduce political theory to an arbitrary dispersion of equally inadequate ideological positions and political power to a random pattern of "nodal points" condensing willy-nilly in social space. Thus the "socialist strategy" of Laclau and Mouffe is little better than a defense of Madison Avenue huckstering, and their view of democracy reduces it to yet another commodity to be mass-marketed without regard to its substance.

Poulantzas compounds his problems by attempting to distinguish the "new petty bourgeoisie" from the working class by means of the distinction between productive and unproductive labor. Productive labor is defined as "labor which produces surplus value while directly reproducing the material elements that serve as the substratum of the relation of exploitation: labor that is directly involved in material production by producing use-values that increase material wealth " (Poulantzas 1975, 216). Insofar as I understand the distinction in Marxist theory between productive and unproductive labor (see Gough 1972), it seems to be increasingly unhelpful when large proportions of "unproductive" workers sell their labor power to capitalist firms and when the production and realization of surplus value increasingly depend on the integration of produc-

Page 390

tion and distribution, on scientific and technological services, and on a vast array of governmental activities. In any case, Poulantzas never successfully reconciles his ad hoc use of economic criteria as a principle of distinction in one case and political and ideological criteria in another.

These errors, I hasten to add, do not vitiate the general thrust of Poulantzas's argument. A common ideological "sub-ensemble" (reformism, individualism, power fetishism) does exist between the traditional and the new petty bourgeoisie and for precisely the reasons Poulantzas specifies: both fractions are caught in a contradictory class position between the hegemonic class interests of the capitalist class and its antithesis, the class interests of the working class. Given the primacy of the capitalist-working class contradiction, the position of the new petty bourgeoisie is necessarily contradictory and oscillates with varying degrees of instability between capitalism and socialism. One can hardly argue with Poulantzas's contention that historically the new petty bourgeoisie has, with rare exceptions, gravitated toward the capitalist pole of the ideological spectrum. Even less can one refute his contention that the mere fact that the new petty bourgeoisie and the working class are both employed by capital has produced little ground for a viable anti-capitalist political alliance.

I believe that the confusion engendered by Poulantzas's attempt to deploy the concept of social class for purposes of clarifying the class position of the new petty bourgeoisie can be salvaged by the rather obvious device of considering the distinction between mental and manual labor from an economic rather than an ideological perspective. Poulantzas, of course, recognizes the vital significance of the mental-manual labor distinction, but he assigns it to the realm of ideology and not to the forces and relations of production. I would argue the reverse: credentials, degrees, organizational positions, and skills are personally owned economic assets, and as such they correspond, roughly, to the personal property of the traditional petty bourgeoisie. The new petty bourgeoisie can never be hegemonic in any capitalist social formation—functionally, their assets, like the personal property of the traditional petty bourgeoisie, are subsumed by the requirements of monopoly capital—but they can and do wage a fierce struggle to preserve their symbolic capital from devaluation by capitalists (deskilling, mechanization) and by the working class (equal opportunity, equality).

Thus similar economic relations (personal ownership of economic assets) account for similar pertinent effects (a common ideological sub-ensemble) shared by the traditional and new petty bourgeoisie. While recognizing the effectivity of political and ideological positions within the ensemble of social relations that constitute the matrix effect and social class, we must also acknowledge the primacy of economic relations within the ensemble. Such an approach demonstrates the essential validity of Poulantzas's contention that the "middle class" is a class fraction of the petty bourgeoisie. But even here the significance is not so much in the name as in the careful elaboration of the phenomenon.

Finally, the Structural Marxist view that the political spectrum is defined, in the last instance, by its capitalist and working-class poles and not the contradictory class position of the petty bourgeoisie is certainly more persuasive than

Page 391

the universe of "free-floating" signifiers devoid of class determinations and class values posited by postmodern, post-Marxist social theorists. This later position is at the very least guilty of peddling wish fulfillment to the middle classes; at most it may be justly condemned for facilitating the ideological legitimation of global capitalism and relegating the Left to the degrading and hopeless position of "loyal opposition" and chief whipping boy of the New Right.

My own views on the problem of the new petty bourgeoisie are heavily indebted to Wright 1985 and Larson 1977. For alternative perspectives from within a Structural Marxist perspective, see Baudelot, Establet, and Malemort 1981 and Carchedi 1977. For a brilliant sociological analysis of ideological distinctions between the various class fractions of the bourgeoisie and the working class using the concepts of habitus and symbolic capital, see Bourdieu 1984. For discussions of Poulantzas's conception of the new petty bourgeoisie, see Wood 1986; Connell 1979; Ross 1979; and Jessop 1985.