structuralisme.fr

Richard P. Appelbaum

Born-again functionalism? A reconsideration of Althusser’s strucuralism


    Less than a decade ago, for those of us in the United-States who experienced the sixties as much in the streets as in the lecture hall, Marxism first emerged as a serious intellectual alternative to conventional social science. Not that Marxism was totally new to us : lost of us as read the Communist Manifesto in some introductory philosophy course, learned that Marx believed all societies had to pass through certain necessary stages (and had thereby erred in predicting the downfall of capitalism), and had studied the « plain Marxism » of C. Wright Mills. But about the beginning of the seventies, for many of us, a new, sophisticated Marxism was discovered. This Marxism was concerned with alienation, with the reification of thought (how perfectly that single word seemed to capture all American ideological productions, from pop culture to the truth of social science), and with political practice (or, seemingly more significantly, praxis) To a generation of humanistically-oriented, leftist-learning young academics, the new Marxism seemed to promise everything : a world view that made sense of the changes all round us; a philosophy capable of situating and criticizing (even negating) the dry, irrelevant, and often mindless empiricism with which we had been educated ; a new and complex language with which to intellectually pillory our teachers; and a notion of politics which somehow suggested - the linkage was never quite clear - that theory and practice were intertwined (again, praxis!) and that therefore our privileged academic status was not totally without redeeming political value. No one seemed to symbolize the excitement and promise of Marxism more than Georg Lukacs - theorist no sociologist read in graduate school in the sixties - who, in a few hundred dense pages, managed to synthesize much of western philosophy and sociology into a compelling critique that introduced us to reification and then showed the way to its inevitable dissolution.
    It was therefore, perhaps not surprising that a generation with virtually no philosophical training whatsoever, whose education seldom - if ever - raised epistemological questions, sought to become instant Marxist philosophers. We graduated from Fromm and Marcuse to Lukacs and Hegel, and then - with perhaps a brief detour in Frankfurt - took up Habermas, whose writing often made Lukacs seem elementary by comparison. We learned that the evils of capitalism had been superseded (dialectically, to be certain) by the technocracy with its legitimating sciences, and that even Marx himself (and certainly Engels) was not immune to certain empiricist tendencies. We learned to selectively read the Marxist classics, to seek out the rational kernel in the early writings, in selected parts of Capital, and in the Grundrisse; we ignored Marx’s expressed concern with developing a science, and became at once critical theorists.
    But at the same time, some of us felts uneasy with this sophisticated Marxism. It appeared to be, for one thing, profoundly anti-empirical: it dealt with ideas, not things, and seemed far more suited to the classroom and the library than the « outside world » of people and institutions and political events. Related to this was another problem: this new Marxism was, in the last analysis, philosophy; and most of us had been trained as empirical social scientists, not as philosophers. Fortunately, one possible route out of these dilemmas was being opened for us - a new continent of Marxism was being charted that promised to satisfy both our empirical and philosophical inclinations. This new continent was being opened on a very old one - Europe, and mainly France - and its terrain had actually been undergoing exploration fir quite some time; but English translations were slow in making their way across the ocean and into our classrooms, so that the land had been well worked over by the time we first entered it. The new continent had first been discovered by yet another Marx — Marx the scientist whose work somehow managed to be scientific without lapsing into an untheoretical empiricism which simply reflected current conditions. Here, then, was real promise: the promise of a safe return to empirical work, uncontaminated by the mindless empiricism we had previously associated with all research. This new approach to Marxist social science contained a damning critique of those philosophical realms into which we had so recently ventured; and, while we might have rankled initially at the highly polemical and dogmatic quality of that critique we also felt ourselves — to paraphrase Lukacs — confirmed by it: because ultimately most of us were far more comfortable (and certainly more skilled) at being sociologists or political scientists or economists than at being philosophers. This new approach was expressed in a language at once more arcane and less elegant than even that of Marxist philosophy — yet it was somehow a familiar language: while it spoke in terms of historical conjunctures, of levels and regions and instances and specificities, of relative autonomy and contradiction and determination in the last instance (which never cornes) and overdetermination, somehow — it was difficult to specify exactly how — we knew we were on familiar terrain. For, given a certain interpretation, was it not possible that what was really being talked about were structures and functions, multivariate models of causation (with, of course, appropriate factor weightings), feedback loops, economic and political systems: in other words, an updated leftwing amalgam of Parsonian systems theory, Durkheimian structural-functionalism, and pluralist political science? An amalgam which came complete with its own epistemological self-reflections — a developed theory of knowledge (termed, of course, "dialectical materialism") which appeared to distance it from these seemingly closely-related cousins?
    To answer these questions I shall reopen the explorations on structuralism's new continent, first examining the world as the structuralists view it, then their theory of knowledge, and finally their notion of history. I will then return to the question of the relationship between structuralism and conventional functionalist theory, before finishing with some persona) speculations concerning what is useful in the structuralist approach, and how that approach might be fruitfully wedded to the insights gained from Marxist philosophy.

THE STRUCTURALIST WORLD

    Althusser has set himself a major task: to rescue what he regards as Marxist 'science" from those ideological elements which have inhibited its full development. Such a task is not easy, for Marxism has fallen into many heresies, principal among which is Hegelian idealism and the anti-scientific humanism which attends it. To rescue Marx, one must read carefully: the classics are limited to the works of Marx, Engels, Lenin, and Gramsci, and then only certain works are admissible; within those works, what is absent often counts for more than what is present, so that one's reading must be cautious and highly selective. Such reading is termed symptomatic, alter Freud; just as the psychoanalyst shuns an "innocent reading" of the patient's symptoms for a theoretically-informed one, so must the dialectical philosopher cull the classics for the scientific structure behind the ideological representations. Althusser believes a science is founded on prescientific or ideological elements, and Marxism is no exception; the "epistemological break" with the old ideological problematic2 — anthropological humanism — came with the Feuerbach theses in 1845, but traces of it can be found throughout Marx's writings until the very end.' This symptomatic reading is, of course, far from arbitrary; it is informed by the scientific knowledge available to the reader at the time, a sort of hermeneutic reinterpretation of particular statements in light of one's knowledge of the overall organization and content of the argument.
    What does such a symptomatic reading of the classics reveal? Despite the complexity of the structuralist language — a language which, at the pen of a writer such as Poulantzas, manages to purge any hints of subjectivity almost entirely — the basic ideas are simple and familiar enough; the most nove) and complex ideas, as we shall see, are correspondingly the most ambiguous. First, let us consider the malin genie which lies at the root of structuralist skepticism, the primary source of error in all innocent readings of Marxism: Hegelian humanism. Hegel —and, more to the point, the young Lukacs — are deceitful precisely because they appear so close to the mark: the notions that the elements of a society are interlinked in a totality; that the totality is marked by contradiction; that such contradiction provides necessary movement and autogenesis towards ends which are immanent in the totality; that knowledge of such processes is power and that therefore a theoretically-informed praxis can enable people to make history within circumscribed limits: all of these notions are seductive and not unrelated to the truth. But, for Althusser, the production of knowledge from such allegedly ideological notions requires a theoretical labor; it does not corne automatically, no simple inversion will reveal a "rational kernel.'4 Consider the Hegelian totality. It is, as Althusser often tells us, a merely expressive totality: a totality whose elements are phenomena of an underlying essence, which they all express. In Hegel, the essence was the Idea, manifested phenomenally in nature and history; all movement in the latter realms could be traced to movement in the Idea, to the primary contradiction which produced the necessary self-movement of thought. In the Marxist inversion of Hegel, the same basic mode) is retained: only now the economy is viewed as essence, and all else as phenomenal superstructure. The primary contradiction — now between forces and relations of production — is manifested at all levels, and propels the totality towards its inevitable end. This process is compellingly described in Lukacs' essay, "Reification and the Consciousness of the Proletariat," where the totalizing subject-object of history is located within the necessary self-development of thought, itself firmly grounded on economic developments whose telos is quite simply assumed. For all of the elegance of Hegelian Marxism, its notion of systemic autogenesis remains a highly simplified one: everything is related to everything else, organized around a single fundamental contradiction, with materialism being salvaged by the assertion that the economy occupies a privileged position in the totality5 Such a position lends itself to economism alongside of voluntarism: economism in that all elements are held to reflect the economic substructure, whose position is therefore totally strategic; and voluntarism in that autonomous, revolutionary activity is inserted into interstices formed by economic contradictions, ruptures in economic production and reproduction during which time things are up for grabs. All of these errors are seen as having been made — repeatedly — throughout Marxist thought; and, in this regard as others, history once again repeats itself with the resurgence of Hegelian Marxism in academic circles.
    To eliminate Hegel from Marxism, a complete theoretical reworking is necessary. What Marx really meant, according to Althusser, runs like this: social formations' — that is, concrete, empirically-existing societies — consist of many overlapping, intersecting structures.' Three principal structures, or regional instances, can be identified as common to all social formations: economic, political, and ideological structures. These structures are relatively autonomous from one another, in that each possesses its own internally structured contradictory elements, revealing its own pace or rhythm of development ("uneven development"). Yet this autonomy is only relative, because the structures are complexly interdependent at the same time: in fact, the nature of each structure — its definition, one might say — is given only within the complex totality of which it is a constituent element.8 This totality, moreover, is not a simple one of co-equal parts; such a pluralism (equivalent to some Hegelian readings of Marx, which similarly see the whole in all the parts)9 is avoided by the assertion that the complex totality is a totality structured in dominance. This means that among the various structural instances, one instance mll have the dominant role: contradictions at other levels will find themselves displaced '° to this instance (thereby averting a revolutionary rupture) or many contradictions may become fused or condensed" in this instance (producing the possibility of a revolutionary rupture). The dominant instance will vary according to the social formation, but in all cases its role is determined — in the lest instance — by the economy; or, in other words, the economy (which is ultimately determinant) often exercises its effects indirectly, displacing its contradictions, by determining the specific efficacy of other instances.12
    Which instance will be determined by the economy to occupy this dominant role? That depends on the mode of production, a concept which — unlike that of social formation — is a purely formal abstraction." According to Balibar (RC: 212-216), a 'scientific' reading of Marxism reveals that all historical modes of production are constituted on the basis of five elements: three elements of production itself (workers, means of production, and appropriators of surplus labor), and two relations that combine these elements (property connections and material appropriation connections.)" The various combinations of these elements determine the mode of production; conversely, the mode of production determines the content of each.15 A social formation, in turn, consists of various overlapping modes of production, in which one will dominate; the remaining modes must not be regarded as historical survivals, but rather as structures which contribute to the complexity of the total formation through their interdependence /autonomy. In the capitalist mode of production, the economy — always determinant in the last instance — assigns to itself the dominant role; in other modes of production, this is not necessarily so (under feudalism, for example, ideology — particularly religion — is assigned this role). Further, under capitalism the political and economic are specifically autonomous — in contrast to the "mixedness" of instances under feudalism (see Poulantzas, 1975:127). Within a regional instance there are various functions and corresponding (sub)- structures — for example the state will have economic functions, ideological functions, political functions; and here, also, the dominant function will be determined by the economic: the political function being dominant in the "liberal" state, the economic function in the interventionist state of monopoly capitalism (Poulantzas, 1975).
    What precisely is the meaning of "relative autonomy," in light of the fact that the economy is determinant "in the last instance?" Althusser — who makes a great deal of the fact that Marx provided the "two ends of the chain" which must now be interlinked — provides little help at this point.16 Poulantzas' discussion of political practice is, however, illuminating. He notes that the intervention of one instance on another "consists of the limits within which one level can modify the other. These limits are the effect both of the concrete matrix of a formation and of the respective specific structures at each level, which are themselves determined by their place and function in this matrix" (Poulantzas, 1975: 94). The limits of variation — i.e., the relative autonomy of the instances — are thus determined by all the structures, individually and in concert (the interdependence of the two terms being recognized). Dominance in this respect, involves the setting of limits to the relative autonomy of the instances: that instance which is dominant in a given formation possesses a greater limiting ability vis-avis the other instances.
    Within such a totality — complexly structured in dominance — no simple notion of causality will suffice: causality is neither linear-transitive, nor expressive after the fashion of Hegel (lieras, 1972:75-76). Rather, it is metonymic — a term borrowed from Lacan to denote the multidimensional causality of a structure of quasi-autonomous yet interrelated substructures, in which each element reflects the greater or lesser causal influence of the others and of the totality itself. Under such circumstances we may speak of overdetermination" — of the fact that each element reflects the causal efficacy of the mutual determinations of a structure in dominance. Causality is treated synchronically —that is, as an aspect of a particular structural configuration at a point in time — rather than historically: the codetermination of the structures, each relatively autonomous and therefore exhibiting uneven development with respect to the others, determines the concrete development of a social formation18 Historicism is thereby ruled out: no notions of telos or immanence propel social systems to predetermined ends. The uneven development of the instances, in their complex interdependence in dominance at a particular conjuncture,19 determines the periodization of a social formation — its particular configuration at a particular point in time. lnasmuch as each formation is unique in this fashion, it is inappropriate to refer to "models" (e.g., the "typical mode) of the bourgeois revolution"): although various formations undergoing the same transition may possess certain common features, the concrete form "depends on the conjuncture of a formation in its historical individuality, original in every case" (Poulantzas, 1975:169).
    Although Althusser claims to be opening new continents, the terrain is familiar enough: Althusser's world is one we have traversed before, and if it appears strange at first it is because through his eyes the commonplace becomes refracted and renamed. Through a careful reading, however, it is perhaps possible to retrieve those familiar elements which lie in the background — a task not so difficult, since these elements are in fact not so deeply hidden. We have first a simple borrowing from chemistry: a periodic table of the constituents of modes of production, whose three elements combine via two relations to exhaustively determine the possible pure types." As in the chemists' version, the elements are conceived as structured entities: an atom of hydrogen has emergent properties that distinguish it from either an electron or a proton, and when atoms of hydrogen combine with atoms of oxygen the resulting compound modifies the structures of both and superficially resembles neither. This is precise, and it is above all 'scientific'.21 Next, we have some borrowings from systems-theory, cross-fertilized with modeling procedures derived from multivariate analysis in sociology and economies — most notably path analysis with feedback loops. This is not so readily apparent, because these adaptations are disguised in the language of psychoanalysis; yet what else is the displacement of a contradiction from one regional instance to another, if not systemsmaintenance by political or ideological subsystems, of disequilibria generated within the economy? What does the condensation of contradictions entait, beyond the multiple causation of several variables on a single one? Does the mutual definition of structures really involve anything more than a codetermination of parameter values after the fashion of the feedback models of modern cybernetic modeling (e.g., the Club of Rome studies)? Is Althusser's formulation of the determinacy of the economy in the last instance simply a nove) way of stating that the economy exerts its effects indirectly as well as directly? One suspects Althusser of being a leftwing amalgam of Forrester, Duncan, Parsons, and Weiner, and there is little in his writing to dispel such suspicions." This is not altogether bad: empirically-oriented Marxists would do well to consider some of the advances in mainstream social science during the past two decades, and Althusser has re-directed our attention fruitfully. But then the central question must arise: what distinguishes Marxist structuralism from an up-to-date version of structural-functionalism? One answer, of course, is obvious: the latter deals only with the existing world of reified social structures (i.e., appearances); the former with the underlying, hidden structures which presumably are ultimately causal (i.e., reality). Yet this difference is itself a complex one raising epistemological issues inadequately dealt with by the structuralists. To understand its significance, it is necessary to first consider the structuralist theory of knowledge and notion of history; we shall then return to a reconsideration of the central question of this paper.

THEORY OF KNOWLEDGE

    Structuralism adheres to a realist epistemology, according to which the underlying structures, although invisible, are conceived as real entities. These underlying structures constitute the building-blocks of the empirical world — the atomic particles, or, to employa more appropriate metaphor from structural linguistics, the syntax of transformation" out of which the empirical world is constituted. Yet given the invisibility of the structures, our knowledge of them depends upon a theoretical production: we can know them only through their effects, and since effects conceal and deceive as well as reveal, only a theoretically-informed reading of effects will result in knowledge.
    We have seen how Althusser seeks to rescue Marxism from its Hegelian origins; he devotes nearly equal attention to countering its tendencies towards what he views as empiricism» According to Althusser, the empiricist problematic, which involves counterposing a perceiving subject to an externat object, merely Ieads back into the age-old irresolvable struggle between materialism and idealism (LP:54-5). Rather, in Althusser's view, the production of knowledge should be regarded as occurring entirely within thought: there is no externat concrete, thing-in-itself that constitutes the raw material of scientific knowledge. Empiricist versions of science attempt to begin with the real-concrete: the tangible object, infinitely rich in its determination, outside of human cognition. The task of science, according to this view, becomes to somehow abstract from the concrete object its essence — that set of features or qualities which enable us to recognize it as a member of a class of objects, as opposed to the host of accidenta) qualities which comprise its uniqueness in time and space. Knowledge is then hetd to consist of these essential attributes, derived inductively: the movement is from real-concrete to abstract-in-thought, the latter representing a greatly simplified version of the former. In rejecting this view, Althusser begins with the assertion that all practices (economic, political, ideological, theoretical) are structured homologously as a production process, wherein a raw material is transformed by a means of production into a final product. 'Scientific' practice begins with a raw material that consists of concepts: ideological concepts derived from pre-'scientific' (i.e., untheorized) problematics; partially 'scientific" concepts; and fully-elaborated "scientific' concepts, themselves the result of prior 'scientific' practice. No 'science' begins with the real-concrete, but rather wth pre-existing concepts which it refines — i.e., 'science' begins with the abstract-in-thought.' This raw material is then transformed by the means of 'scientific' production — "That corpus of concepts whose more or less contradictory unity constitutes the theory' of the science at the (historical) moment under consideration" (FM: 184)...' This transformation then produces a final product, knowledge, which is the concrete-in-thought: concrete because it is the "concentration of many determinations, hence the unity of the diverse" (Marx, 1973:101). The movement of knowledge is thus precisely the reverse of what the empiricists daim: from abstract to concrete (and again it must be emphasized: entirely wthin thought). Althusser terms these three stages in the production processes Generalities, to empha size the fact that they all occur within the process of thought; Generality I ref ers fo the raw material (pre-existing concepts), Generality II to the means of production (corpus of theory), and Generality III to the final product (scientific knowledge).
    This process of knowledge production is regarded as iterative. Generality III consists of ideas which are themselves destined to become the raw material of future 'scientific' practice. these ideas also serve to modify the corpus of existing 'scientific theory and methods, eliminating inconsistencies and contradictions, and thereby refining the means of scientific' production. >Science' therefore proceeds by a series of successive approximations, continually improving its theory. methods. and knowledge of the structures. 'Scientific' practice —within a problematic — is thus regarded as cumulative: Althusser adheres to the continuity view of scientific progress, according to which each scientific advance comprehends all preceding unes. eventually converging on truth — the real abject - which itself remains unaltered by the conceptual refinements of the knowledge process:

the process of production of a knowledge necessarily proceeds by the constant transformation of its (conceptual)object. that it is precisely the effect of this transformation which is the same thing as the history of knowledge, that it produces a new knowledge (a new object of knowledge)which still concerns the real object knowledge of which is deepened precisely by this reorganization of the object of knowledge. As Marx says profoundly the real abject of which knowledge is to acquired or deepened remains what it is after as before the process of knowledge which involves it (cf the 1857 Introduction). if therefore it is the absolute reference point for the process of knowledge which concerned with it — the deepening of the knowledge of this real object is achieved by a labour of theoretical transformation which necessarily affects the object of knowledge since it is only applied to the latter

In this process the object is ultimately determining. for t determines both tself and our knowledge of it.
    Thus, in Althusser's epistemology, material reality is preserved intact, unalterable, in-itself: it does not disclose itself to us, for our knowledge is concerned with conceptual objects atone. The empiricist problematic is thereby sidestepped — there is no need to deal with the paradoxes implied in the extraction of an essence from real objects — but at a rather major cost: the relation between real and conceptual object is left untheorized, removed, by fiat, from philosophical discourse. Althusser distinguishes the "process of the real" from the "process of thought:" yet somehow there is a correspondence between the two: knowledge depends on realty, and reality is revealed only through knowledge (Glucksmann, 1972: 73-74). Since an isomorphism between object and knowledge is tacitly assumed truth will win out, somehow. given sufficiently rigorous theoretical practice — one need not concern oneself tao much over the status of structuralist facts. If Balibar identifies three elements and two relations as constitutive of all possible modes of production. Why then there are exactly three elements and two relations: these are, after all, revealed by a symptomatic reading of Marx and Engels, and one can only marvel at the mathematical ingenuity required to construct such a periodic table on the basis of a half-dozen historical instances. If Althusser tells us that there are three regional instances, or four practices, or that all practices are homologous — why it "just happens" (Glucksmann, 1972: 83) that this is so. Althusser arbitrarily elevates concepts derived from the observation of the capitalist mode of production to the status of universals, in what one critic has referred to ironically as a form of conceptual positivism (D'Amico, 1973: 94); such a characterization, as we shall shortly see, is by no means unfair.

    Why should knowledge necessarily correspond to external reality, given the assumed epistemological break between the real-concrete and the thought-objects which constitute the raw materials of knowledge production? Althusser may choose to remain silent on this issue, but others calling themselves structuralists have been embarrassingly forthright in asserting the one obvious solution to the problem: the hidden structure of the mind imposes its form on externat reality. Thus, according to Levi-Strauss:

    While the neo-Kantian character of structural linguistics, anthropology, and psychology is readily acknowledged by the practitioners of these disciplines, Althusser can hardly own up to such a position: the idealism implied by such a retreat must certainly outweigh the benefits to be achieved by finally by-passing Hegel altogether. And so we encounter a silence on the part of Althusser — one which must be broken by others."
    The troubled isomorphism between reality and knowledge is further threatened on another front: the fact that reality is deceptive. The Marxist concept of Darstellung (representation) occupies a central position in Althusser's epistemology. This refers to the fact that the underlying structures exist only in and through their surface effects; yet these effects distort our view of the structures, which they simultaneously reveal and conceal. Generality III would thus appear to be twice-removed from the real structures which are ultimately the object of thought: first, because of the epistemological gap between real-concrete and Generality I; and, secondly, because Generality I itself consists of concepts which are at their core deceptive. How can one know whether one's knowledge productions (Generality III) indeed approximate reality, since the ability to judge the deceptiveness in the Darstellung is inhibited by the epistemological gap? In Althusser's solution, all judgments are made entirely internal to the science: Generality III is assessed in light of Generality II, which itself must be reinterpreted in light of Generality III, and so on. A science thus becomes its own arbiter of truth; internal consistency, the convergence of the iterations, constitutes the basis of the verdict.

   

Since any discussion of the relationship between knowledge and concrete reality is, according to Althusser, part of the empiricist problematic, there is no external basis for judging the adequacy of theory; science judges itself, and finds itself adequate. This self-serving formulation is known by another narre: dogmatism.
    'Science,' for Althusser, is thus doubly privileged: not only is it entirely in the objective, dealing as it does with facts, but it is its own arbiter of truth in judging the degree to which its knowledge corresponds to reality. In Althusser's original version of this vision, 'scientific' practice was to be regarded as totally autonomous from the other practices (economic, political, ideological): Marx, Althusser originally asserted, never (except perhaps in the "pre-Marxist" 1844 Manuscripts) included 'science' in the superstructure: "science can no more be ranged within the category "super-structure" than can language, which as Stalin shows escapes it" (RC:133)." lnasmuch as 'science' was thereby to be distinguished from ideology as being concerned only with knowledge for its own sake, Althusser's position was surprisingly close to that of Mannheim: detached from any interested social base, 'science' was regarded as free to develop in an ever-closer approximation to the truth. This formulation served to insulate Marxist theory from the internal politics of the Third International following destalinization and the wave of liberalizing tendencies which characterized the decade after the Twentieth Congress of the CPSU; but it was a formulation that brought AIth.usser under increasing criticism from the French Communist party (Singer, 1975:224-229). By the end of 1968, Althusser had denounced his earlier insistence on the primacy of theory over practice as a "theoreticist deviation;" theory, he now argued, reflects class struggle, with philosophy being, "in the last instance, class struggle in the field of theory" (ESC: 37: emphasis removed). In a series of recent writings," Althusser has argued for a more "conjunctural" view of knowledge: "without the proletariat's class struggle, Marx could not have adopted the point of view of class exploitation, or carried out his scientific work" (LP: 9). l he task of philosophy thus becomes that of "taking sides" in the struggle between a 'science' and the ideological elements that continue to permeate it in class society: philosophy is regarded as a battleground for the contending ideas which are spawned by class struggle, ideas which eventuate in a 'science' only as philosophy ferrets out and destroys their false elements. Historically, it was therefore necessary that Marx's "philosophical evolution (be) based on his political evolution; and that . . . his scientific discovery (`the break') (be) based on his philosophical evolution" (ESC: 69).3' Althusser has thus reestablished the political significance of philosophy — Marxist philosophers are warriors on the battlefield of ideas — but at the expense of rendering ambiguous the presumed objectivity of theory, which now must be interpreted within the framework of the sociology of knowledge.38
    Althusser's epistemology, developed explicitly in opposition to empiricism and positivism, remains within the French positivist tradition inaugurated by Comte — and with precisely the same political implications: Althusser, like Comte bef ore him, delivers truth into the hands of the chosen functionaries of his positive polity. Let us review. Facts are facts, yet are somehow revealed only through 'scientific' inquiry, itself regarded as capable of neutrally transcribing the factual reality which is its object into a set of concepts which are both universal and ahistorical (Brown, 1975: 51; Geras, 1972: 810-882; D'Amico, 1973: 94; Glucksmann, 1972: 81-82, 87-88; Glucksmann, 1974: 135-137). Scientists constitute the elite cadre who atone have privileged access to these universals, since truth is sufficiently hidden and distorted by the play of daily events that it can only be retrieved by a h ig hly technical (if not outright arcane) theoretical labor. ln further keeping with the Comtean tradition, science is regarded as simultaneously radically detached and engaged (Glucksmann, 1974: 120-122; Geras, 1972: 85-86): detached because it contemplates its object without touching it or altering it, and engaged in that it serves and advocates the interests of a particular social group.39 Science thus becomes the legitimating mantle in a crusade to realize objective history, with scientists the secular priesthood arbiting between truth and falsehood in the holy struggle. This technocratic vision, with which Comte inaugurated sociology (Lenzer, 1975), becomes all the more dangerous in proportion to its possibility of fulfillment in today's adminstered societies, socialist as well as capitalist.
    Some seventy years ago, pursuing a course charted by Comte and Mach, philosophy set forth in quest of the positivist dream: the development of an inductive logic which would permit the theoretical terms in any science to be derived from purely observational ones, according to explicit correspondence rules. Such a quest, if successful, would guarantee the objectivity of scientific knowledge: all untestable assumptions, all metaphors or analogies, all theoretical statements adopted heuristically for the purpose of organizing perception would be purged from a neutral scientific language consisting of statements about observations, deductions from such statements, and the logical operators connecting the two. Scientific statements, in short, were to be limited to true statements of fact. The belief that such was indeed possible guided developments in the philosophy of science for a half-century, through logical positivism and what eventually came to be known as the dominant or "Received View" (Suppe, 1974). Althusser's quest is a parallel one: he also seeks after objectivity and truth, after self-subsistent facts which are unambiguously represented in scientific concepts. Unlike the adherents to the "Received View" of the philosophy of science, he does not seek to develop an inductive logic — much, it would appear, to his detriment, since were he to have done so he would have had to take into account the failures that followed fifty years of effort to provide a philosophic basis for scientific certitude. Rather, Althusser paints with a broad brush, offering generalities about Generalities embedded in a production metaphor: and if there appears to be something compelling about his argument, it is precisely because the argument is so unspecified.
    The early positivists knew what they were after, and their successors sought to develop a method of scientific practice which would guarantee rigorous, absolute certainty. They were unable to do so; and their failure should provide a caution to Althusser, whose objectives are the same; but it does not.°" "The Received View" foundered on the basis of its own contradictory results; its most prominent critics, during the past fifteen years, have been united in rejecting its fundamental premise — the belief that facts are capable of being constituted purely objectively, that is, prior to all conceptualization. Facts, it is recognized, do not speak for themselves; rather, the language through which they gain a voice is one which has been previously constituted, and, most significantly, whose meanings cannot be directly derived from the facts themselves. It is in this spirit that Weltanschauungen (Suppe's (1974:125-221) term) philosophers such as Kuhn, Hansen, and Toulmin — and in more extreme form Feyerabend — have argued for a radical pluralism of scientific theories, on the ground that scientific "facts" are constituted in a dialectic entailing scientific practice on the one hand, and the objects of scientific inquiry on the other. The production of a particular knowledge is therefore in part circumscribed by conventions and relevances internal to the world views of particular scientific communities: the choice of paradigm cannot ultimately be referred to an external totalizing viewpoint. Althusser, who does not wish to be considered any variant of positivist, offers a parallel formulation in which facts are regarded as theoretical constructs resulting from a social process: class struggle in the production of knowledge. But the resulting knowledge, to the extent that it is "scientific," is cumulative and converges on rock-hard reality. Scientific theories, for Althusser, do not entai) Weltanschauungen elements nor admit of paradigmatic shifts; they are simply true (George, 1971:82; Zimmerman, 1976:83; Glucksmann, 1974: 152-152). The philosophy of science, unwilling to deny epistemology, systematically addressed the issue of the correspondence between knowledge and its externat object: and the consequence has been a radical assault on the belief in the possibility of objective certitude. Althusser, for whom epistemology has no place, readily finds scientific certainty — but only through an unacknowledged neoKantianism which is inconsistent with his dialectical view of knowledge as a concrete production. °'
    The danger in all this is self-evident: as Lichtheim (1971: 144) puts it, structuralism is dogmatism -a "learned and sophisticated variety," to be sure —but dogmatism nonetheless. The Party scientists and bureaucrats alone possess truth — they alone have symptomatically read the works of Marx, Engels, Lenin, and Gramsci — and they alone understand history: they reveal the hidden facts and the underlying laws, the masses submit.


FREEDOM/DETERMINISM: HISTORY AND PRAXIS

    Given structuralism's rejection of the subjectiobject problematic, it is not surprising that the freedom /determinism antimony is likewise rejected as ideological — as outside the concern of 'science.' Poulantzas, who provides the fullest discussion of political practices in relation to the social structures, conceives of class practices as structured systems of relations occurring within and reacting on the limits set by the "complex unity in dominance" of the three structural instances (economic, political, and ideological). The limits on political practice are thus "raised to the power of two": "political practice is conducted within the limits set by the other practices and by the global field of class practices (the economic, political, and ideological struggle), insofar 'as this field is itself limited by the effects of the structure as limits" (Poulantzas, 1975: 95).' Political power, in this formulation, is clearly no individual attribute (Weber's notion that a social actor will be able to exercise hisiher will despite resistance), nor even an attribute of a role: it is a characteristic of the struc ture, specifying "as limits the effect of the structure in the relations between the various practices of the classes in conflict" (Poulantzas, 1975:108; emphasis removed).
    History, in such a view, is truly blind (Althusser variously refers to "immense machines," "plays without authors," and "processes without subjects"). Human beings are 'scientifically' conceived as the incumbents of structurally-determined roles —as mere supports (Trager) for the structures: any concern with human agency reflects the ideological legacy of bourgeois society, with its Benthamesque notions of autonomous individuals maximizing utility through free choice." Struggle occurs not between exploited people or even classes; it is rather a property of a "system of relations between atomic terms which are invested with the force of labor or property" (George, 1971: 95). What is falsely termed "history" is nothing more than the arrangement and rearrangement of the atoms according to the laws embedded in their very structures; that is, there is no history, as Althusser takes great pains to argue throughout his writings.

    Here structuralism has at last arrived at its ultimate objective: it has achieved 'scientificity' by obliterating the conscious individual. Alongside a Lacanized Freud and a Leninized Parsons we may now perceive a third major family resemblance in the behaviorism of B.F. Skinner, who informs us that social science cannot advance until we eliminate the old teleological notions of purpose and volition and even consciousness, recognizing instead that human behavior is selected by environmental contingencies. To be certain, Althusser would not likely assimilate Skinner uncritically — a symptomatic reading of behaviorism must necessarily reject its atomistic social psychology along with its epistemological commitment to observable events — but once it is recognized that environment must be conceived structurally, Althusser and Skinner merge with disquieting ease. Both believe that ideology deceives us into thinking we are subjects, whereas human relations are 'scientifically' conceived as entirely in the objective, decodable only by highly trained specialists whose privileged understanding results from their 'scientific' training. Both argue for a 'scientifically'-grounded technology of behavior control — although whereas Skinner places such control in the hands of prison authorities and hospital administrators who are concerned with (mis)- behaving individuals, Althusser's technocrats are Party functionaries preoccupied with 'scientifically' guiding the working class."

    History has corne full circle: if, in 1845, Marx criticized the philosopher's division of society into "educators" and "educated," Althusser today reaffirms and celebrates that rupture, unconcerned about who will educate the educators: certainly not the masses, since theory is never tested or con
firmed in practice. The Party knows history, the rest of us only experience ideology: we will be educated if only we will listen in silence, although it doesn't really matter: our practices are objectively structured in any event. In announcing the death of man, Althusser merely completes the positivist programme inaugurated by Comte; we must not be deceived by the fact that positivism has switched sides over the past hundred-odd years. To observe that Althusser's anti-humanism is profoundly anti-Marxist is beside the point.'" What is important, however, is that its reactionary character also be recognized. To deny the efficacy of human agency is to deny any motive for political action: whether behavior is regarded as environmentally selected or as supportive in the great drama of the structures, any consistent determ in ism must be antipolitical. The fact that neither Skinner nor Althusser corne to this conclusion merely testifies to the truth of Emerson's aphorism: their minds being large, they are not troubled by the hobgoblins of foolish inconsistencies. Althusser's structuralism celebrates the reification of contemporary society, elevates it to the level of a 'scientific' truth: in it we find ourselves confirmed in our alienation, counseled by scientific authority to expunge even the term 'alienation' from our vocabulary. At a Lime when science is one of the principal legitimating props of the social order, an uncritical belief in science is urged upon us: discourse is cut off from the left as well as the right, dogmatic authority is reasserted on all fronts.

CONCLUSION: STRUCTURALISM'S UNFULFILLED PROMISE

    To appreciate the extent to which Marxist structuralism constitutes an advance on conventional structuralism-functionalism, it is necessary to focus on two claims of the former: first, that it deals with hidden or underlying rather than surface structures; and second, that its structures are characterized by internai contradiction rather than externat relationship. Each of these claims will be taken up in turn.

   

    Althusser — indeed, all structuralists, including Levi-Strauss — take pains to distinguish their structures from those of the Anglo-American empiricist tradition deriving from the works of Durkheim, Radcliffe-Brown, Merton and Parsons. The difference, we are told, resides in this: the structures of the latter are taken as given, directly accessible to empirical inquiry, while those of the former are hidden, retrievable only by means of a theoretical labor of reconstruction. I have criticized some of the difficulties with structuralism's position in the first two parts of this paper: its ontological realism, its faillire to deal adequately with the relationship between 'real' and theorized structures, its inadequate treatment of the problem of verification (compounded by the fact that surface events simultaneously reveal and distort their underlying structures) — in a word, its refusai to take seriously the epistemological difficulties implicit in its formulations. Yet these criticisms do not constitute an indictment of the fundamental insight, which seems to me central to any Marxist approach: rather, they are an objection to structuralism's dogmatic refusal to admit epistemology into its lexicon, a refusal which can only impede the development of that which is useful in its approach.
    Concerning the hidden nature of the structures, what strikes me as useful is this: the recognition that the underlying logic of a system must be theoretically reconstituted if surface events are to be rendered intelligible. In this sense, as Althusser argues, structural analysis must precede historical analysis. Nor can underlying structure be deduced from surface event in any direct fashion, but rather as Godelier puts it, "it is never directly visible and accessible at the empirical level but has to be discovered through theoretical research, which will involve the setting up of hypotheses and models."" Conventional functionalist theory implicitly recognizes that it cannot move solely in the realm of surface institutions and events, and so has been compelled to develop concepts which tacitly acknowledge the presence of underlying structures. Merton's (1957: 19-84) notion of latent functions (unintended /unanticipated consequences), for example, suggests that institutions often work "behind our backs" — although it does so in an inVerted fashion which focuses attention on the lack of imagination and insight on the part of social actors, rather than on underlying system requirements themselves." Similarly, the notion of universal functional prerequisites (see e.g. Parsons, 1951, 1960, 1966; Levy, 1965, 1967) suggests that certain underlying structures must be common to all societies, but stops short of theorizing the necessary interrelations among such structures: rather, the functional prerequisites can presumably be satisfied in any number of empirical ways, and so the resulting social systems are treated as essentially creative responses to a set of common underlying problems. Structuralism, on the contrary, holds that beneath any manifest set of social relationships lies a system that is complexly interdependent in determinate ways — a structured unity which accounts for the seemingly accidentai qualifies of concrete historical  forms.

    This leads us to consider the second daim of structuralism — that, unlike conventional functionalism, its structures constitute a "unity in dominance" whose motive force consists of contradictions circumscribed within their vert' nature. What precisely does this mean? Within the Hegelian dialectic the function of contradiction is clear enough —contradiction is inherent in thought and language (according to Hegel), entailing a process in logic whereby the terms which constitute a conceptual unity are regarded as mutually negating, propelling the Idea towards ts inevtable end. Althusser has
performed a service to Marxist sociology by demonstrating that no simple inversion will permit the adaptation of such a notion to the social world: that because Hegel's dialectic necessarily resolves all of thought and history at a given instant into the two moments of a single contradiction (of which all else is pure phenomena), Hegel's dialectic can never serve to capture the structured complexity of concrete social history. But it is in offering a materialist version of the dialectic that Althusser runs into his deepest difficulties: for what remains when Hegel is excised appears to be little more than a pluralist functionalism recast in Marxian categories. Since this is a major criticism, it will be explicated at some length, although aspects have been touched on throughout this paper.
    In his essay "Contradiction and Overdetermination" (FM: 89-128) Althusser distinguishes Marx's dialectic from Hegel's by pointing to its sociological nature — that is, the tact that the Marxian contradiction entails an "accumulation of 'circumstance' and 'currents' " (FM 99) and is therefore complex and overdetermined in a way that the Hegelian contradiction is not. The Marxian contradiction is specified by historically concrete conditions, including those characterizing the superstructure (the state, the dominant ideology, religion, politically organized movements), the internal historical situation (state of the bourgeois revolution, remnants of the feudal past, customs, national traditions, political 'etiquette' ), and the externat historical situation (most notably the extensiveness of capitalism as a world imperial system). Lenin's treatment of the Soviet revolution is taken as illustrative of the "accumulation and exacerbation of all the historical contradictions then possible in a single State" (RC: 95-99: emphasis removed). What is most notable about Althusser's enumeration of the numerous 'contradictions' which rendered Russia the 'weakest link in the imperialist chain' — an enumeration which runs on for almost two full pages — is that it is purely descriptive: it points to a set of historical circumstances to be sure, but just why these constitute 'contradictions' is unclear. One could just as easily substitute 'tensions' or 'conflicts' for 'contradictions' in the text, which would then hold few surprises for more conventionally-minded political historians.- Althusser recognized this difficulty when he later notes (FM: 107) that he has thus far used the term only descriptively — that ne nas not yet shown the necessary link between the structure of 'contradiction' and Marx's theory of history and society. He then seeks to establish the linkage by demonstrating that Marx's dialectic radically alters both the terms and relations implied in the Hegelian mode! of society, rather than merely inverting it. Thus, the basic terms of Hegel's model — civil society and state — are abandoned in favor of such terms as forces /relations of production (replacing civil society), state as an instrument of the ruling class (replacing the simple state), and class (which mediates between the two). Similarly, the relations no longer are simply those of Hegel's essence /- phenomena (or the simple Marxist 'inversion': base-/superstructure), but rather entai) the 'two ends of the chain' — determinism by the economy in the •- instance at one end, and relative autonomy ispecific, effectivity of the structures at the other. It is clear that this formulation takes us from Hegelian philosophy to a pluralist sociology employing Marxian categories. But how Althusser differs from an updated pluralism remains unclear, for he has not yet fulfilled his promise of demonstrating how Marx's 'contradiction' is internally linked to Marx's theory. As I have previously noted, all that has been said so far —including statements concerning the famous twoended chain — can be recast in terms of a multicausal model in which the economy exercices a disproportionate amount of direct and indirect influence.
    In his later essay "On the Materialist Dialectic" (FM: 161-218), Althusser returns to this issue, poing the question: "What, then, is this 'specificity' of contradiction?" (FM: 193). This time he turns to a text of Mao, the essay "On Contradiction," for guidance. What unites Mao's distinctions (primary vs. secondary contradiction, primary vs. secondary aspects of a contradiction, the uneven development of contradictions) is "the complexity of the process at (their very) heart . . . every 'simple category' presupposes the existence of the structured whole of society simplicity is merely the product of the complex process" (FM: 194, 196). Furthermore, Althusser notes, the dominance of a single contradiction is circumscribed within the unity of the structured whole, in that whole and part condition one another: "secondary contradictions are essential even to the existence of the principal contradiction" (FM: 205). Thus, "overdetermination designates the following essential quality of contradiction: the reflection in contradiction itself of its conditions of existence, that is, of its situation in the structure in dominance of the complex whole" (FM: 209); it is this character of overdetermination "which gives Marxist contradiction its specificity" (FM:210). Althusser summarizes his "theoretical results" in the following "schematic" form:

The specific difference of Marxist con tradition is ifs "unevenness." or "overdetermination." which reffects in its conditions of existence. that is. the specific structure of unevenness (in dominance) of the ever-pregiven complex whole which is its existence. Thus understood. contradiction is the motor of all development. Displacement and condensation. with their basis in its overdetermination. explain by their dominance the phases (non-antagonistic. antagonists. and explosive) which constitute the existence of the complex process. that rs. "of the development of things."

Restated in more familiar terms, a society consists of interrelated structures (Marx's categories are of course employed), each one reflecting the influence of the others, and each one rent by conflicts or tensions which themselves reflect the influence of the others. Some of these conflicts or tensions are temporarily resolved through their displacement from one structure to another; others become critical as they corne together at a single location. Throughout this process the effects of one structure will be particularly decisive for the articulation of the whole.
    Thus stated, Althusser's formation is itself a formal inversion of structural-functionalism, perhaps not so distant from what is sometimes termed "conflict theory" (Dahrendorf, 1959, 1964; Turner, 1973, 1974, 1975a, 1975b; see also Appelbaum, 1978a: 72-74). The ubiquity of contradiction is asserted, but its necessity is never demonstrated. Hegel was able to demonstrate such necessity because for him contradiction was a characteristic of logic; but why such necessity should be circumscribed within the character of things remains a mystery, the vert' mystery with which we began. Is there a way out of this dilemma? Perhaps. Godelier provides a clue when he observes that "in showing that the mechanism of the production of surplus value is the common origin of the visible forms of capitalist profit .. Marx successfully analyzes the articulation of the internal structure of the system to (its) visible forms . In his analysis of the capitalist economic system, Marx demonstrates how in a class society based on the extraction of surplus value under the 'givens' of capitalism (e.g., private ownership of means of production: competition; formally free wage labor; etc.), the economic structures entail inherently contradictory requirements. Whether Marx is looking at long-term tendencies towards instability (e.g., the declining rate of profit) or short-term recurrent ones (e.g., crisis resulting from overproduction / underconsumption), he is able to demonstrate the structural necessity of such tendencies in the economic sphef e." But this is because the concepts Marx has developed in Capital to analyze contradictory economic system requirements follow directly from his effort to theorize a concrete economic formation. Althusser's writing, on the other hand, remains by comparison overly formalist: it does not deal with particular structures in particular societies, except in isolated cases where the treatment is highly abstract (tor example, the essay "Ideology and Ideological State Apparatuses" in LP: 127-186). It is my belief that Marx was able to give the idealist notion of contradiction its materialist form only because of his specific economic theory, and in particular the central rote within that theory of the concept of surplus value extraction in economic production. Althusser has rightfully sought to redirect Marxism away from economism, and by so doing to g-3neralize Marx's insights to other spheres of social life. But he has not developed a parallel set of concepts with which to theorize noneconomic forms of production; and consequently he has not been able to demonstrate the inherent instability of noneconomic structures, a demonstration which is essential if Marxist structuralism is to be fully distinguishable from its more bourgeois relatives.
    Althusser has advanced us in drawinq attention to both the "hidden" nature of social structure, and its complex and interdependent quality. He has thus managed to combine Hegel with sociology — or at least Levi-Strauss with sociology — and th is is perhaps no small achievement. But his key concepts will remain unspecified until those who work within this framework ground their theorizing in the analysis of concrete social, political, ideological, and economic structures, and their interrelations within particular societies. Whether current efforts in this direction (for example, Poulantzas in political theory, Castells in urban sociology) will produce a theoretical development comparable with that of Marx remains, of course, to be seen.

A MODEST PROPOSAL

    The shortcomings of structuralism result from its scientistic claim to truth; its anti-humanist denial of praxis and history; and its ontological commitment to the reality of invisible underlying structures, which leads it into epistemological difficulties that are resolved only with a tacit (if denied) neo-Kantianism. While these shortcomings may seem to add up to a great deal, it is my belief that they do not amount to a fundamental denial of structuralism itself. Althusser's attempt to reintroduce sociology into Marxism, and in particular his criticisms of Hegelianized versions of totality and immanence as applied to social formations, seem to me to be telling criticisms: they provide a useful corrective to the anti-empirical influence of philosophical Marxism. What is necessary, therefore, is a carefu I — perhaps even symptomatic — rereading of structuralism, in light of the criticisms that have been raised. What follows is intended to suggest the contours of what a Marxist structural sociology might entail, absent some of Althusser's more excessive and inconsistent claims. The essential features of such a structuralism would include the following:

    1. Social relations are conceived in Althusserian terms as a "complex structured unity in dominance," the economic being determinant in the last instance. Practices are aimed at the limits or interstices provided by the particular ("conjunctural" ) configuration of the structures, and are capable of reacting back on the structures and modifying them and their interrelations. The precise characterization of the structures — in Althusserian terminology, the numbers of "regional instances," the nature of the substructures, the elements and forms of modes of production, etc. — will reflect a theoretical labor which is as yet incomplete; provisionally, the characterizations offered by the Althusserian school are retained at least as concerns capitalist society."

    2. The structures are real in their consequences (i.e., are experienced in their consequences as "social facts" in the Durkheimian senses of exteriority and constraint), but they are not thereby accorded ontological facticity. Whether or not there are universal, invariant, ever pre-given structuresin-themselves is at best moot, following Hegel's criticism of Kant; the "in-itself" can only be known as conceptually constituted, and therefore totally objective being should not be posited. Equally significantly, the structures themselves, insofar as they are objects of praxis, are modifiable: the neo-Kantian solution is therefore rejected as well. Levi-Strauss' characterization of social structures as reflecting universal and invariant mental or linguistic forms does not appear to me to be a fruitful avenue for a Marxist sociology to pursue: any approach which hypostatizes universal categories can only lead away from a belief in the malleability of social institutions.

    3. The structures, although hidden, are not thereby "invisible." This is because they are not real objects which somehow exist in a timeless and spaceless Platonic superstrate, impervious to direct observation or other means of perception; but rather because they exist for us only as theoretical constructs. We do not "see" them, because they require a prior theoretical labor to "recognize;" and white we may assume the existence of real structures as the ultimate ground of such theoretical productions, we do so instrumentally — to ensure a conce r with some form of objective empirical confirmation —rather than dogmatically. Consistent with Althusser's epistemology of the Darstellung, underlying structures are both revealed and disguised through their phenomenal effects; contrary to his intentions, Althusser's notions of structural causality and "absent presence" seem to me most useful in comprehending and penetrating reified social forms, rather than describing an underlying factual reality.

  

    4. The foregoing considerations imply what Suppe terms a Weltanschauungen philosophy of science, after the fashion best known through the writings of Thomas Kuhn (e.g., 1970). That is, theoretical productions are recognized as resulting from conventions internat to the practice of a scientific community, as much as from a commitment to observational objectivity. In this view, objects and theory are mutually constituted in a scientific practice; the practice itself is subject to sociological interpretation. lt follows, therefore, that significant theoretical terms as well as canons of scientificity are always and necessarity adopted by convention, in accordance with the accepted outlooks of the community of scientists. While this position seems to be the most defensible one in light of the history of the philosophy of science over the past seventy years, it is arguable on purely practical grounds as well: surely it is better to err on the side of humility than dogmatism, if one assumes (as Althusser apparently does not) that the existing state of Marxist science is far from perfect, that the political and human stakes are sufficiently high so as to warrant a highly skeptical attitude towards all proferred claims to truth in any event, and that a principal task of Marxists should be a frontal assault on reified social forms and the uncritical acceptance of any authority whatsoever.

 

    5. Social formations are characterized by alternating periods of crisis and normalcy, as contradictions which arise within or between instances are more-or-less successfully displaced;" major crises involve the accumulation (in Althusserian terms, "condensation") of contradictions at a single location. Crises provide major occasions for organized political practice — i.e., class struggle —not only because they represent an objective deterioration in the materiat conditions of life for large numbers of people, but also because they are crucial in revealing the underlying structural mechanisms of society. Thus, both scientific knowledge and mass consciousness depend in part on the occurrence of crises, albeit to differing degrees.55 An analogy with Kuhn's notions of "normal" and "revolutionary" periods in scientific practice is instructive. During the "normal" phase, things run relatively smoothly; empiricist versions of science prove highly successful, both because phenomena operate with Iaw-like regularity (given the relative absence of accumulated contradictions at the structural level during such periods), and because people have so internalized society's norms that they indeed become mere supports for predetermined roles. Both ideological and scientific productions reify existing social relations; people are, in C. Wright Mills' phrase, reduced to the status of "cheerful robots," exhibiting moreor-less predictable behavior. During the crisis, however, anomalies arise which are inexplicable in terms of the taken-for-granted ideological and scientific productions; these anomalies provide occasion for scientific advance and the subsequent reformulation of ideology along more scientific lines. Scientific advance is made possible because the underlying structures, which are the objects of scientific theorizing, can only be known through their effects; and the workings of the structures become transparent to the extent that their effects assume nove) or unexpected forms." Ideological reformulation will follow a crisis as a direct consequence of the class struggle: various efforts will be made to rationalize or explain the crisis in ways compatible with the interests of the various classes and class fractions involved, and the extent to which the crisis permits significant reeducation along scientific (i.e., Marxist) as opposed to ideological Unes will reflect the adequacy of Marxist theory and the relative influence of Marxist ideologies.

    Crisis thus paves the way for de-reification, and makes possible self-conscious, theoretically-informed action. The effect of such action depends, in turn, on where it is directed. If aimed blindly at surface events, the contradictions giving rise to the crisis will be readily displaced, the crisis repaired, and a return to normalcy and renewed reification will rapidly follow — until the contradictions reaccumulate and another crisis ensues. if action is directed at the source of the contradiction itself (at the underlying structures) on the basis of adequate theoretical understanding, significant structural change can result; for this reason, efforts will be made to divert attention to more trivial events. It therefore becomes extremely important to distinguish fundamental structurai change from surface repairs that leave the underlying structures largely unaltered — i.e., to distinguish revolutionary from reformist political practice. The adequacy with which this distinction can be drawn will in turn depend upon the adequacy of theoretical understanding concerning the underlying structural mechanisms.