Learning Journal essay 500 words

10.4324_9781315820033-10.pdf

Home >Political Science homework help >Learning Journal essay 500 words

Structuralist Marxism and development

A renewed interest from the late 1960s in "unorthodox" approaches to Marxism paved the way for an alternative to mainstream develop- ment and underdevelopment theories in the 1970s. Modernization theory was criticized for its evolutionary approach, while under- development theory was rejected because it focused on trade rela- tions. These theories were contrasted to a "scientific" method that took the mode of production as the determining concept. Although there were some divisions within this school of thought, the most in- fluential thesis argued that Third World social formations were char- acterized by an articulation of modes of production.

This chapter provides a critical analysis of the modes of production school, and assesses some of the suggested alternatives to it. In mak- ing this assessment, I enter into a direct dialogue with those writers who have identified the impasse in development theory. In doing so, I argue that their identification of an impasse is essentially correct, but I do not share all of their arguments concerning the way out of it. This will also become dearer in the second part of the book.

The chapter is divided into three sections. First, I examine the theory of articulation of modes of production. Then I provide a cri- tique of this theory, and show its similarities with the development theories discussed in the previous chapter. I extend the critique in the third section, which discusses the relationship between capitalism, unfree labour and the state, which in turn leads me back to an exami- nation of the impasse, and especially of the relationship between theory and method. It is at this point that I return to the impasse, and express some reservations about some of the "post-Marxist alterna- tives" to the impasse.

STRUCTURALIST MARXISM AND DEVELOPMENT

The articulation of modes of production

Modes of production theory attempted to overcome the basic prob- lems of the development theories discussed in the previous chapter. Modernization theory and orthodox Marxism were rejected for their "evolutionism" and their monocausal explanations, while under- development theory was rejected for its "mirror image" of evolution- ism and its focus on trade relations rather than the relations of production (see Laclau 1971/1977, Taylor 1979). In terms of its chal- lenge to evolutionary theory, the new approach was an attempt to move "beyond the sociology of development" (Oxaal et al. 1975).

The theory emphasizes that capitalist development differs in both time and space, and so rejects models that argue that it uniformly de- velops or underdevelops the periphery. It is also argued by this school of thought that an emphasis on the determining role of the mode of production leads to a greater understanding of the mechanisms of "underdevelopment" (or, more precisely, uneven development). This is said to be the case because trade relations presuppose what they are supposed to explain - the production of a surplus product and the di- vision of the world into core and periphery (see Laclau 1977, Dore & Weeks 1979: 65). The mode of production, on the other hand, takes as its point of departure the production of the surplus product and so is able to move to an explanation of the division of the world between core and periphery based on modes of production rather than trade relations per se (MandIe 1972: 53). The core therefore coincides with the capitalist regions of the world, which are largely based on free wage labour (see Chapter 2), while the periphery was incorporated into the world economy on the basis of unfree relations of production (that is, non-capitalist modes of production), which prevented an unprecedented accumulation of capital. Unequal trade relations were therefore not unequal trade relations per se, but in fact reflected un- equal relations of production. It is for this reason that the "advanced" capitalist countries were able to dominate other areas of the world where non-capitalist modes of production existed (Bettelheim 1972).

Modes of production theory does however recognize that this rela- tion is not static, and that capitalist relations of production have emerged in the periphery. Nevertheless, it is a specific kind of capital- ism, one that is qualitatively different from its form in the core coun- tries. Capitalism in the periphery is different because it is combined

THE ARTICULATION OF MODES OF PRODUCTION

with non-capitalist modes of production - in other words, capitalism coexists, or "articulates", with non-capitalist modes (see Post 1978: 27). Non-capitalist production may be restructured by imperialist (that is, "core capitalist') penetration but it is also subordinated to it by its very "conservation" (see Bettelheim 1972).

At the heart of the theory is a distinction between social formation and mode of production. The social formation refers to a combina- tion of economic, political and ideological practices or "levels". The mode of production refers to the economic level, which determines which of the different levels is dominant in the "structured totality" that constitutes the social formation. The economic level sets limits on the other levels, that carry out functions which necessarily repro- duce the (economic) mode of production. These non-economic lev- els therefore enjoy only a relative autonomy from the mode of production (see Althusser & Balibar 1979: 178-80, and glossary, 319).

Within the periphery, any social formation (or "economic system" - see Laclau 1977) may be constituted by more than one mode of pro- duction (Taylor 1979: 105-42, Wolpe 1980: 19-27). Such an articu- lation does not lead to a new version of dualism because the different modes of production are closely linked within this social formation. Thus, for Laclau (1977: 35) an economic system is based on "the mu- tual relations between the different sectors of the economy, or be- tween different production units, whether on a regional, national or world scale" and can include "as constitutive elements, different modes of production". Thus in an economic system the capitalist mode of production may be dominant and other modes of produc- tion are subordinated to it. Therefore, where the capitalist mode is dominant, the preservation of non-capitalist modes is explained by reference to the needs, or reproductive requirements, of the dominant capitalist mode. As Bradby (1975: 129) states:

Capitalism has different needs of pre-capitalist economies at dif- ferent stages of development, which arise from specific historical circumstances, e.g. raw materials, land, labour power, and at times of crisis, markets.

Although some proponents of this theory also analyze the "laws of motion" of the non-capitalist mode (especially Rey, see Foster Carter

STRUCTURALIST MARXISM AND DEVELOPMENT

1978), the main focus of this school is on articulation where the capi- talist mode dominates, and so the needs of the capitalist mode are paramount. Some examples illustrate their argument. In the case of various agricultural modes of production, the "peasant-worker's" continued access to land enables the capitalist to pay a wage lower than he or she might otherwise have to, and the peasant is exploited by the capitalist who buys the peasant's produce cheaply and sells it more expensively in a wider market place (Cliffe 1982). In the case of the so-called "informal sector", a "petty commodity mode of produc- tion" serves the requirements of the capitalist mode by providing a re- serve army of labour, and by providing cheap goods to workers in the capitalist sector, which in turn reduces the reproduction costs of workers, thus allowing capitalists to again pay cheaper wages than would otherwise be the case (Moser 1978).

Perhaps the most interesting example, though, is provided by the work of Claude Meillassoux. He argues that imperialism conserves the "domestic mode of production" as a means of preserving high im- perialist profits through the exploitation of female labour, which car- ries out domestic tasks. As he states (1981: 95):

It is by establishing organic relations between capitalist and do- mestic economies that imperialism set up the mechanism of repro- ducing cheap labour power to its profit - a reproductive process which, at present, is the fundamental cause of underdevelopment at one end and of the wealth of the capitalist sector at the other.

Wolpe (1972; see also Wolpe 1980 and 1988) has utilized this theory to explain a specific form of capitalism in what became South Mrica. Instead of an unambiguous break-up of non-capitalist modes in the late nineteenth century, the Mrican labour force was dispos- sessed from its land (see Bundy 1979), but was effectively relocated on to 13 per cent of the land. The result was that the labour force working in the mines still had direct access to the means of produc- tion, which enabled capitalists to pay workers lower wages than would have been the case in a "free labour market". As Wolpe (1972: 429) argues:

The state has been utilized at all times to secure and develop the capitalist mode of production. Viewed from this standpoint, racist

THE ARTICULATION OF MODES OF PRODUCTION

ideology and policy and the state now not only appear as the means for the reproduction of segregation and racial discrimination gen- erally, but also as what they really are, the means for the reproduc- tion of a particular mode of production.

Modes of production theory is also often closely linked to the notion that capital has become more international, as it crosses the bounda- ries of different nation-states. An articulation may occur in that "na- tional capital" articulates with international capital, and vice versa, or international capital may have the effect of conserving non-capitalist modes within Third World social formations. In either case, it is as- sumed that the interests of capital will be realized (see Palloix 1975, Cypher 1979, Browett 1985).

So, proponents of the theory of articulation of modes of produc- tion argue that capitalism in the periphery "coexists" or "joins to- gether" with non-capitalist modes of production to form a social formation. "Underdevelopment" is therefore not a product of trade relations but a result of the preservation of the non-capitalist mode of production. This preservation prevents the general emergence of free wage labour and therefore generalized commodity production, and so the development of the productive forces is not as rapid as in those (core) countries (or regions) where the capital-labour relation is gen- eralized. The two modes are not however "separated" (as moderniza- tion theory contends), but are closely linked to the limits of the pre-capitalist mode (where this mode is dominant), and the needs of the dominant capitalist mode of production (where this is dominant). So, the paradox is that the preservation of the non-capitalist mode is actually in the interests of the capitalist mode, at least when the capi- talist mode is dominant. As Taylor (1979: 227) states:

This articulation of one practice within another is governed both by the reproductive requirements of the capitalist mode, and by the restrictions placed on this articulation either by the limits within which the penetrated instance can operate, as set by the non-capitalist mode of production, or by the continuing reproduc- tion of elements of the non-capitalist mode.

Finally, and related to the previous point, this theory contends that the mode of production and the articulation between different modes

STRUCTURALIST MARXISM AND DEVELOPMENT

of production determine the "phenomenal appearances" of the social formation. Once again Taylor (1979: 274) makes this clear:

economic features [such] as urban unemployment, combination of different types of labour, and accumulation of indigenous capital ... [are] analysed as forms whose determinants - the changing re- productive requirements of the industrial capitalist mode and the level of resistance of the non-capitalist mode or division of labour - were necessarily absent in their phenomenal appearance.

A critique

Functionalism

Modes of production theory has two potential strengths compared with modernization and underdevelopment theories. First, it avoids (at least to some extent) the problem of conceptualizing a supposedly "normal" and ahistorical path of capitalist development. Secondly, it potentially leads to a politics that recognizes subordination within the international political economy and exploitation within the nation- state. In other words, modes of production theory represents the most sustained attempt to transcend the "internal-external" di- chotomy that plagued the modernization-dependence debate. The result is, or at least could be, "a politics sensitive to local opportuni- ties and struggles". This is preferable to the logic of Frank's and Wallerstein's positions, which "deny the very possibility of meaning- ful political action in Third World countries" (Corbridge 1986: 61).

Nevertheless, these would-be strengths are never completely real- ized because modes of production theory is weakened by a function- alist (and, it is argued below, evolutionist) methodological approach. This is because the theory explains social change as a product of a necessary logic of capitalism. As the quotations from Taylor, Wolpe and Bradby above make clear, the continued existence of non-capital- ist modes is explained by their necessary function of cheapening the reproduction costs of the dominant capitalist mode. But this begs the question of why non-capitalist modes broke down in any society. This leads to the inconsistent and circular reasoning that characterizes all functionalist methodologies. Corbridge's (1986: 63) basic criticism of this approach addresses the problem:

A CRITIQUE

If the PCMP [pre-capitalist mode of production) survives (as in the Bantustans) then that is evidence of its functionality for capitalism; and if it does not ... then that too is evidence of capitalism's func- tional requirements. (See also Booth 1985: 774-6)

The implication of this theory is that any social change can be ex- plained as the product of the "reproductive requirements" of capital- ism. This tendency to "read off" changes in the periphery in terms of the requirements of capital is not, of course, exclusive to this theory. The previous chapter showed that mainstream development theories also explained social change in terms of the needs or actions of the metropolis, so that modernization or underdevelopment were ulti- mately caused by the Western world.

The roots of this functionalist methodology can be traced back to the original Enlightenment project, but it was the development of Althusserian Marxism in the 1960s and 1970s that had the more im- mediate impact on development sociology (see for example Hindess & Hirst 1975, Taylor 1979, Meillassoux 1981, Peet 1991, 1992). A re-examination of this methodology (see also Chapter 2) is impor- tant, and not only for the obvious reason that it has directly influ- enced modes of production theory. Perhaps more importantly, though also more controversial, an examination of the problems of Althusserian Marxism is useful because its methodological errors are indicative of the problems of development theory in general, particu- larly radical development theory. Underdevelopment, world systems and dependency theory share similar methodological weaknesses, even if they do not explicitly acknowledge the Althusserian influence. So, an examination of this methodology is useful, as a critique both of modes of production theory and of other radical approaches. In a sense then, in discussing the methodology of Althusserian Marxism, the methodology of all radical development theory is being put under scrutiny. This discussion in turn leads back to a discussion of the root causes of the impasse in the sociology of development.

Althusserianism and the impasse in the sociology of development

The work of Louis Althusser and his followers should be seen in the context of debates in the French Communist Party and its attempts to move beyond Stalinist orthodoxies (for a commentary see Benton

STRUCTURALIST MARXISM AND DEVELOPMENT

1986: ch.1). Althusser'smain concern was to break with the "econom- ism" of orthodox Marxism and therefore reject base-superstructure and stagist models (see Chapter 2). More relevant to this discussion however is that he was also equally hostile to empiricist methodolo- gies, which simply took the "facts" as given. For instance he wrote (Althusser & Baliber 1979: 105) that:

we must once again purify our concept of the theory of history, and purify it radically, of any contamination by the obviousness of empirical history, since we know that this "empirical history" is merely the bare face of the empiricist ideology of history.

His former British followers, Barry Hindess and Paul Hirst (1975: 312), are even more explicit:

Marxism, as a theoretical and a political practice, gains nothing from its association with historical writing and historical research. The study of history is not only scientifically but also politically valueless.

The "empiricism" of history is replaced by the construction of con- cepts, which "are formed and have their existence within knowledge" (ibid.: 1). Theories and concepts are therefore a priori constructs, which are derived from "theoretical practice" (for an extended defi- nition, see the glossary in Althusser & Balibar 1979: 316) that is com- pletely separated from the empirical world. The concept that is given determining status is the mode of production, which is defined as "an articulated combination of relations and forces of production struc- tured by the dominance of the relations of production" (Hindess & Hirst 1975: 9, see also Balibar in Althusser & Balibar 1979: part III). It is this concept that explains social change. Political action may have a "relative autonomy" from the (economic) mode of production, but is always "in the last instance" explained by it.

It is precisely this methodology that articulation theorists share with other theories in the sociology of development. Changes in the Third World are "read off" from the reproductive requirements, or needs, of an a priori conception of (global) capitalism. This concep- tion is used to explain the preservation of non-capitalist modes (as in Meillassoux 1981), the destruction of non-capitalist modes (Warren

A CRITIQUE

1980), the development of the periphery (Warren 1980, Rostow 1971), or the underdevelopment of the periphery (Frank 1969a,b, Wallerstein 1974). It also subordinates agency to structure, and as- sumes that social phenomena are explained by their functionality for capitalism, rather than by the actions and struggles of human beings themselves. Added to this is a spatial dimension, in which the struc- ture is determined by the needs of Western capitalism or the capitalist mode within a peripheral social formation. This approach therefore rules out, on a priori grounds, any meaningful political action by so- cial actors within the Third World, because they are, by definition, passive actors, directed by the whims of the capitalist mode of pro- duction.

Therefore, the "articulation" of modes of production should not be assumed away as part of a "logic of capital", but should be grounded in a historical analysis of social struggles, in a particular location, at a particular time. To return to the example of South Africa, Michael Burawoy (1976; see also Cohen 1987: 105-6) has argued that the cheapening of labour costs under segregation and apartheid was far more contradictory than Wolpe implies. He asks the following ques- tions: in what respects is labour cheap, and who benefits from it? Cer- tainly it was cheap for South African mine-owners, but the costs of policing the system of migrant labour were actually very expensive for the South African state. Moreover, the process of labour recruit- ment to the emerging capitalist mode of production in late nine- teenth-century Transvaal and twentieth-century South Africa was far more problematic than modes of production theory allows. This is because people are not as passive as the theory assumes, and there was widespread (and ingenious) resistance to labour recruitment by the mine-owners (see for example Delius 1980).

Similarly, "economic" relations cannot be given theoretical priority over "non-economic" ones. Class (which, I stress below, is not a purely economic relation) is a very important concept in Marx's thought, but this does not mean that other social factors are reduced to "it". Any suggestion that this is the case implies that class is a "thing" and not a social relation, and it therefore leads back to the economism that post-Marxists so rightly criticize. To take one exam- ple, Meillassoux's (1981) attempt to explain the oppression of women as a "mode of reproducing" the labour force should be re- jected because it does not explain why it is the man who works in the

STRUCTURALIST MARXISM AND DEVELOPMENT

labour market and the woman in the subsistence sector (see Mackin- tosh 1977, Bozzoli 1983). Such a functionalist "logic of capital" ap- proach has an economistic starting point, which is separated from the real world and which therefore attempts to explain the oppression of women in terms of its concept of capitalism.

Structuralism therefore replicates the errors of the theories dis- cussed in the previous chapter. It takes for granted that which is sup- posed to be proved, and assumes that capitalism automatically secures its own reproductive requirements. The real history of human societies is far more complex and contradictory than this, as the South African example above illustrates. To recognize the existence of "phenomenal forms" in capitalist society is not the same thing as assuming that these forms automatically serve the needs of capital. As Corbridge (1986: 67) argues,

there is nothing in the concept of capitalism itself which should lead us to expect that it must have X, Y or Z development (or un- derdevelopment effects). Such contingencies are not forged at this macro-theoretical scale. The reproduction of capitalist relations of production clearly presupposes the existence of definite condi- tions of existence (private property relations, for example, and free wage labour), but it tells us nothing about whether they will be secured, or about the all-important forms in which they are made flesh. (See also Mouzelis 1980)

Contrary to the arguments of many Marxists, both Althusserian and orthodox, Marx (1989: 48, the second emphasis is mine) was equally clear on this issue:

To present the laws of the bourgeois economy, it is not necessary to write the real history of the production relations. But the correct analysis and deduction of these relations as relations which have themselves arisen historically, always leads to primary equations ... which point to a past lying behind this system. These indica- tions, together with the correct grasp of the present, then also offer the key to the understanding of the past - a work in its own right.

At this point we arrive back at the extended discussion of Marx that I undertook in Chapter 2. It should be clear that, in separating struc-

CAPITALISM, UNFREE LABOUR AND THE STATE

ture and struggle and in proposing a view of capitalism in which its needs are automatically met, structuralist Marxism is guilty of the fet- ishism of social forms that Marx so convincingly criticized. This fet- ishism and functionalism are rooted in modes of production theory's account of the relationship between theory-building and empirical research. So far this point has been only implicit, but in the next sec- tion I hope to make my point more clearly. This section examines the relation between capitalism and unfree labour, and discusses the state, and in so doing leads us back to a discussion of the wider rela- tion between theory and empirical research, and structure and agency. This in turn provides a convenient route back to a further dis- cussion of the impasse in the sociology of development.

Capitalism, unfree labour and the state: the relationship between theory and method, and the impasse in development theory

Capitalism, unfree labour and the state

As I showed in Chapter 2, Marx argued that in capitalist society the wage-labourer is "free" in two senses. On the one hand, she is free to sell her labour power as a commodity to any potential employer - that is, the worker in capitalist society differs from the peasant in feu- dal society in that the former has a choice of to whom she will sell her labour power. On the other hand, freed or separated from direct ac- cess to the means of production, the worker is compelled to work for a wage in order to obtain the means of subsistence (food, clothing and shelter). So, to over-simplify, the peasant is compelled to work for the landlord through laws, customs and statutes. In capitalist society, the worker is compelled to work for the capitalist, not for legal or politi- cal reasons, but because of "the dull compulsion of economic rela- tions" (Marx 1989: 85).

However, as the example of South Africa above makes clear, capi- talism does coexist with unfree labour. Moreover, there are many ex- amples that reinforce this observation. Unfree labour continues to exist in many parts of the Third 'World, and indeed the state in metro- politan countries has regulated the capital-labour relation - the guest worker system in Germany and the Youth Training Scheme in Britain immediately spring to mind.

STRUCTURALIST MARXISM AND DEVELOPMENT

Of course one explanation for this persistence, at least in the case of the periphery, is the notion of articulation, but, as I argued above, there are problems with this conception. Others imply that Marx was simply wrong (see Cohen 1987). Still others (for the best exposition, see Miles 1987: 196-222), influenced by the articulation school, ar- gue that capitalism and unfree labour are compatible but that unfree labour is an anomaly (albeit a necessary one). The problem with this argument is that it falls between the two stools of evolutionism and functionalism. Unfree labour is an anomaly because capitalism is based on free wage labour (evolutionism), but it is a necessity because it is based on the needs of capital at a particular time (functionalism).

Once again, the problem with this approach is that it conflates theoretical construction or the elaboration of models with the em- pirical world, and tries to impose the former on the latter. Marx's analysis of the historical origins of capitalism is not a universal model or a general theory of the capitalist mode of production, as I showed in Chapter 2. Instead, it is an account of how certain social forms, such as the commodity, appear as "natural" but are in fact the product of historically constituted social relations (Williams 1978b). There- fore, just because "free" labour was the "norm" in England (and even this assertion is questionable - see Corrigan 1977), it does not follow that all "capitalisms" must follow the English path. So, although Miles is at pains to separate his analysis from a unilinear theory of his- tory, his assertion that unfree labour is an anomaly leads him back to precisely the trap of regarding England as the norm and all others as anomalies (see Brass 1988: 185). Indeed, this evolutionism is re- peated by articulation theory, which argues that "normal" capitalism (of an English type) is different in the periphery, and so it therefore must have "articulated" with non-capitalist modes. In the real world, all are in a sense the norms and all are anomalies. The forms of labour characteristic of particular capitalisms are likely to, and indeed do, vary. Marx (1989: 27-8) argued along these lines when he stated that:

The specific economic form, in which unpaid surplus labour is pumped out of direct producers, determines the relationship of rulers and ruled, as it grows directly out of production itself and, in turn, reacts upon it as a determining element ... This does not pre- vent the same economic basis - the same from the standpoint of its

CAPITALISM, UNFREE LABOUR AND THE STATE

main conditions - due to innumerable different empirical circum- stances, natural environment, racial relations, external historical influences, etc., from showing infinite variations and gradations in appearance, which can be ascertained only by analysis of the em- pirically given circumstances.

This leads on to a second, and equally important, point, which takes us back to the heart of the impasse. The arguments that capital- ism is based on "purely economic" forms of labour control, and thus that unfree labour is an anomaly, falsely separate different levels in a social formation. The state is relegated to a derivative, political level, and class relations are located purely at the determining economic level (see Poulantzas 1973: 191-4). Such an approach downplays the prospect of any systematic analysis of the state, because it is said to be derived (albeit in the last instance) from the economic level. This is of direct relevance to any discussion of the impasse because the state in peripheral (and, I would argue, all) capitalist societies cannot be as- signed a derivative or superstructural role - which, critics of the im- passe have argued, is one example of Marx's "economic determin- ism" (Mouzelis 1988: 35-40, ibid. 1990). The state should not be derived from a determinant "economic" sphere, but should instead be conceptualized on the basis of the historical emergence of capital- ist social relations. As Corrigan and Sayer (1981: 27) state:

Marx's argument is not that law and politics are separate from production and its relations as such, but rather that in commodity production the social relations of production themselves take the forms of apparently exclusive political and economic spheres. "Polity" and "economy" are in other words but different facets of one and the same set of production relations.

In other words - and this has great implications for "post-Marxism" (and neoliberalism, which is discussed in Chapter 6) - a separate "economy" or "political system" are forms of existence of capitalist social relations. Marx was clear on this and argued that the state was not part of a "superstructure" in capitalist society. For example, he documented the role of the state in the "primitive accumulation" of capital in England, and argued that "force" (a supposedly "political" relation) "is itself an economic power" (Marx 1976b: 916).

STRUCTURALIST MARXISM AND DEVELOPMENT

"Back to the impasse"

The discussion of this particular debate, namely the relationship be- tween capitalism, unfree labour and the state, once again leads us back to the impasse. Contrary to the claims of structuralist Marxism, the political state does not simply serve the needs of capital, nor is it part of a so-called superstructure. Similarly, unfree labour is not sim- ply an anomaly, but is a constitutive part of "actually existing capital- isms". It is these kinds of reductionist assumptions that have been forcefully challenged by post-Marxist writers in the development field, and so it is now appropriate directly to confront debates around the impasse, and the ways in which Marxism has contributed to it.

The writer who first identified an impasse was David Booth (1985). The impasse was seen as "the result of a generalized theoretical diso- rientation affecting in different degrees all of the main positions in the radical development debate" (ibid.: 776). Underdevelopment and related theories are rejected on similar grounds to those outlined in the previous chapter, but what is more important here is his similar rejection of Marxism. His main contention is that Marxism has a "metatheoretical commitment to demonstrating that what happens in societies in the era of capitalism is not only explicable, but also in some stronger sense necessary" (ibid.: 773). So, in the two examples that I have discussed in this chapter, the preservation of non-capitalist modes of production and unfree labour relations are "explained" simply by the "needs of capital". Empirical research is subordinated to this a priori theoretical construction (see also van der Geest & Buttel 1988, Sklair 1988). For development sociology to move be- yond this stagnation it "must be freed ... from Marxism's ulterior in- terest in proving that within given limits the world has to be the way it is" (Booth 1985: 777).

As an explanation for the impasse, Booth's paper is a powerful one, not least because it shouts what others have only whispered. What is less clear are his suggestions for a way out of the impasse. For exam- ple, he argues that some Marxist concepts are still useful (ibid.: 777), but does not specify which should be dispensed with and which should be retained. It has been left to others to make explicit sugges- tions of ways of transcending the impasse. A common argument is that development sociology should focus on empirical and compara- tive research and construct "middle-range" theories out of this work

CAPITALISM, UNFREE LABOUR AND THE STATE

(Mouzelis 1988, Evans & Stephens 1988, van der Geest & Buttel 1988, Sklair 1988). I share these views and expand on them below (and register my concern about how this might be carried out), and in Part II of this book.

But, for the moment, I want to concentrate on the suggestions of one writer who has developed Booth's arguments and suggested some ways of moving out of the impasse - Stuart Corbridge. He ar- gues that the way out of the impasse is the development of a "post- Marxism" that rejects the functionalism and determinism of modes of production and other development theories (1990: 628). This is most clear in his discussion of the work of Hindess and Hirst, who broke with the "high Althusserianism" of their early work, discussed above, and laid the framework for a "post-Marxist" alternative (see Corbridge 1986: 66-7, ibid. 1990: 629, Hindess & Hirst 1977, Cut- ler et al. 1977, 1978). In their post-Marxist work, the conceptual framework of Marxism is said to be tautological and economically reductionist because it is assumed that the economy can secure its own "non-economic" conditions of existence. To return to the three examples in this chapter, this would include the preservation of non- capitalist modes, of unfree labour, and of a state that served the inter- ests of capital. As Hirst (1977: 131) has argued,

Either economism, or non-correspondence of political forces - that is the choice which faces Marxism.

Similarly, Hindess (1978: 96-7) has argued that,

The choice for Marxism is clear. Either we effectively reduce po- litical and ideological phenomena to class interests determined elsewhere (basically in the economy) ... Or we must face up to the real autonomy of political and ideological phenomena.

What is clear from these two quotations is that Hindess and Hirst give only two choices - either general determinism (by the concept of the economic mode of production), or absolute autonomy from this determinism. So, to return to one example above, the state is either determined by the economy or autonomous from it. The first propo- sition is wrong because it is based on a theory in which the determin- ing role of the economy is built into the discourse itself (Corbridge

STRUCTURALIST MARXISM AND DEVELOPMENT

1990: 629). Therefore, according to the logic of the theory, the sec- ond proposition must be correct. Although the move away from theoreticism and structural-functionalism is welcome, there is a great danger in this "either/or" approach. For, despite Corbridge's protes- tations (1990: 628-9, 1986: 66-7), this dualism has led Hindess and Hirst to move away from a theoreticism divorced from the empirical world to an empiricism that takes for granted, and therefore natural- izes, historically constructed social forms. Hindess and Hirst's post- Marxism is therefore as guilty of fetishism as the structuralist theory it leaves behind. This is hardly surprising given that the two views, based on a rigid dichotomy between the theoretical and empirical world, are in effect a "mirror image" of each other. As Meiksins Wood (1990: 128, see also Meiksins Wood 1986) points out,

It soon turned out that Althusserianism had simply replaced - or supplemented - the old false alternatives with new ones. Marxists had in effect been offered a choice between structure and history, absolute determinism and irreducible contingency, pure theory and unalloyed empiricism. It is not surprising, therefore, that the purest theoreticists of the Althusserian school became the most unalloyed empiricists of the post-Althusserian generation.

What this discussion suggests is that functionalism, economism and theoreticism should all be rejected as Booth, Corbridge and others suggest. However, they should not be replaced by an empiricism that naturalizes social forms or autonomizes social practices. The notion that "the economic determines the political" is replaced in Hindess and Hirst's (and their collaborators') auto-critique by the notion that "the economic does not determine the political" (see, among others, Cutler et al. 1977: 222,224,226,227). Booth and Corbridge both correctly reject the economism of the first premise, but say nothing about the second. However, as the discussion of the state above made clear, it is the common starting point - that is, the separation of dis- crete, separate "levels" - of both contentions that is inadequate. The implications of this argument for social analysis are great, because it suggests that factors such as "class", "gender" or "state" cannot be segregated to their own particular "sphere" or "level". Class is there- by not a purely "economic" relation, just as the state is not a purely "political" relation. By separating social phenomena into compart-

CAPITALISM, UNFREE LABOUR AND THE STATE

mentalized units, Hindess and Hirst (marks I and II), automatically take "things as they are" and so reproduce the "phenomenal forms" of social relations. That is, they are guilty of fetishizing capitalist so- cial relations.

This is not to suggest that the new "post-Marxists" in development theory have followed this road, but there is a danger that they may replicate the fetishized categories of Hindess, Hirst and others. To take another example, Booth is rightly scathing of grand theories of development but he takes a rather eccentric view of one work that is similarly critical. This is Robert Brenner's (1977) justly famous article on the origins of capitalist development. In this work Brenner ex- plains the dynamic nature of capitalism in the "core" as a product of the historical emergence of a capital-free wage labour relation and of the generalized commodity production and "free market" that this gives rise to. Booth (1985: 770) rightly describes this essay as bril- liant, but also complains that it "does not give us what many people have looked for in the mode of production literature, namely, a genu- ine third position in the debate over colonial and contemporary de- velopment in the Third World". I think that Booth is undoubtedly correct, but that is one of the great strengths of Brenner's work - it is not a search for a "new position" because such a position, forged at the level of "high theory", will automatically subsume practice to theory and agency to structure. Booth (1985: 770) assumes that Brenner shares Warren's optimism concerning capitalism in the post- colonial period, but the whole thrust of Brenner's paper is a challenge to the "developmental" logic that assumes that the effects of capital- ism can be examined independently of the actions of human beings. In other words, Brenner's work reconstitutes the unity of structure and agency that development and modes of production theory sepa- rate. As Ruccio & Simon (1986: 216) state, for the Brenner approach "it is not a matter of a mechanistic, deterministic process" (see also Foster-Carter 1978: 77, for a "Brenneresque" feminist critique of modes of production theory, see Bozzoli 1983).

Booth's ambiguity about theory/practice and structure/agency is also apparent in his conclusion when he recommends the preserva- tion of some lower-order concepts derived from Marx, but also sug- gests that there is a case for "a relative (and hopefully temporary) shift of emphasis in the sociological development debate from theory to metatheory" (1985: 777). The problem with this proposal is that it

STRUCTURALIST MARXISM AND DEVELOPMENT

proposes no way out of the structure/agency dichotomy, and is likely to replace one "grand theory" with another.

Finally, Booth and Corbridge both caricature Marx's work and so, once again, there is a distinction to be made between Marx's method and the method of his followers. Althusserian Marxism, and there- fore modes of production theory, begin with the concept of the mode of production, which determines non-economic institutions and the social practice of human beings. Marx, on the other hand, did not be- gin with a priori dogmas. Instead, Marx & Engels (1982: 42, my em- phasis) stated that:

The premises from which we begin are not arbitrary ones, not dog- mas, but real premises from which abstraction can only be made in the imagination. They are the real individuals, their activity and the material conditions under which they live, both those which they find already existing and those produced by their activity. These premises can be verified in a purely empirical way.

This point is consistently made throughout Marx's work, and is not part of Marx's humanist youth, as Althusserians contend. For exam- ple, Marx (cited in Sayer 1979: 92) explained his methodology in Capital in the following way:

I do not start from "concepts" ... What I start from is the simplest form in which the labour product is represented in contemporary society, and this is the "commodity". I analyze this, and indeed, first in the form in which it appears ... Thus it is not I who divide "value" into use-value and exchange-value as oppositions into which the abstraction "value" divides itself, but the concrete social form of the labour product.

This famous - but strangely neglected - passage from Marx, de- scribing his methodological approach, leads us nicely back to the questions of articulation theory and unfree labour. The fact that dif- ferent capitalisms utilize different forms of labour control does not mean that the concept of the capitalist mode of production has to be abandoned. Once again, structuralism presents us with a false dichotomy - either a capitalism based on one form of labour control (wage labour), or the abandonment of the concept of capitalism (as a

CONCLUSION: SOCIOLOGY, DEVELOPMENT AND THE IMPASSE

social form) itself. In other words, the choice is between a "pure" cap- italism and purely contingent methods of labour control. But there is a third choice, which is based on different capitalisms, based on differ- ing methods of labour control, whose commonalities and differences are explained through comparative and historical analysis.

Marx's critique of political economy is thus not a grand theory of a pure capitalism but a "relational analysis. It allows us to ascertain the conditions of existence of particular kinds of experiences, but it tells us nothing at all about how these conditions originally came about. It is not, in other words, a causal analysis. It merely reveals internal rela- tions: those 'conditions and relations' of particular modes of produc- tion" (Sayer 1975: 789, see also Sayer 1987, and Ollman 1976).

Conclusion: Sociology, development and the impasse

Modes of production theory constituted the most sustained attempt to break with the evolutionism of development theory after 1945. It rightly pointed to the existence of different forms of labour control in "actually existing capitalisms", and provided a convincing explana- tion for the reality of uneven development and the different forms this had taken in various parts of the world. Moreover, in placing some emphasis on the internal structure of peripheral social forma- tions, the theory also potentially gave some considerable weight to the salience of local struggles. However, this potential was never ful- filled because the theory repeated the errors of previous development theories, albeit at a lower level of abstraction. The major problem was the theory's functionalism, in which changes in the Third World were "read off" from the needs of the dominant capitalist mode of production. This was not so very different from (Marxist and non- Marxist) modernization and underdevelopment theories: the former explained development as a consequence of "Westernization", while the latter explained underdevelopment as a consequence of the domi- nance of Western capital. In each of these theories, the struggles of so- cial actors were reduced to the overriding requirement that the structural needs of the system be reproduced.

The result of this excessive structuralism was that empirical re- search is carried out on the basis of fitting into a preconceived theo- retical model. There was actually very little need for empirical

STRUCTURALIST MARXISM AND DEVELOPMENT

research because the results of, and explanations for, changes were built into the theoretical system. This in turn led to a dogmatism that rejected the empirical findings of scholars who lacked the "correct" a priori method (see for example Peet 1989, 1991). It is for this reason that the sociology of development largely isolated itself from the im- pact of the counterrevolution in development theory in the 1980s.

The growing awareness of this impasse has led to the suggestion that the discipline should adopt a "post-Marxist" approach, which would utilize some of the non-functionalist work of radical social theory. This might include regulation theory, post-imperialist theory or structuration theory (see Corbridge 1989,1990, Spybey 1992). Some of these ideas are discussed in more detail in Part II, but it should be clear from the above discussion that there are some grounds for ques- tioning the utility of at least some post-Marxist approaches. This does not mean that one simply "returns to Marx" to show that he had all the answers. This is precisely the kind of ahistorical dogma that I re- ject. What is useful in Marx's work is a historical and materialist method, grounded in empirical research, that shows that phenomenal forms are historically and socially created. Marx's concern was there- fore to build theory out of empirical research. This approach differed from an empiricism, that took the "facts" as given, precisely because of its recognition that such "things" are neither universal nor natural.

The impasse is therefore best overcome - and here I agree with the general arguments of those who have recognized the stagnation in the field - by a renewed commitment to empirical research, and the abandonment of the search for a grand theory (see Mouzelis 1988, van der Geest & Buttel1988, Sklair 1988). This does not mean that theory per se should be abandoned, for this would lead back to an em- piricism that is simply the other side of the theoreticist coin, and in which "things" are "taken as they are". It is in this way that theory can be built a posteriori, on the basis of the recognition that the struggles and actions of "social actors" are important, but also that there are structures that impede their action. These structures might even give rise to certain "laws" (see the next chapter for an example), but these are neither universal nor separable from the action of human beings.

This leads to my final point, which concerns the way out of the im- passe and the distinctiveness of sociology. Development sociology must transcend its theoreticism and give proper weight to the interdisciplinary nature of development studies. I also believe that it is

CONCLUSION: SOCIOLOGY, DEVELOPMENT AND THE IMPASSE

precisely at this point that it may also preserve its specific character and at the same time move beyond its theoretical stagnation. The discipline should draw on the best empirical work in the field, and use this as the basis for the development of a "grounded theory". There is a wide scope for work on the basis of "suggestive contrasts" (Runciman 1983: 168-93): for example, between the successful development trajectories of east Asian newly industrializing countries compared with the less successful examples in Latin America (Evans 1987,jenkins 1991, and the next chapter), or between state formation in Africa and east Asia (see Sandbrook 1986).

The impasse is therefore best transcended by a method that is historical, empirical and materialist. I hope that it is clear, however, that any work that does not utilize this method cannot simply be dismissed on a priori grounds; it must instead be demonstrated by a direct confrontation with, rather than dismissal of, the discourse itself. I therefore stress again the a posteriori nature of theory build- ing. In Part II of this book I expand on these arguments through a dis- cussion of some of the recent changes in the global political economy and some of the political implications of these developments.

This page intentionally left blank