Capital & Class. - 1978. - Issue 6

EDITORIAL

This issue marks the completion of the first two years of Capital and Class,in which time the CSE has grown far beyond its origins as an organisationof academic economists to become a much more diffuse organisation thatembraces a wide range of intellectual workers and political activists . TheCSE is correspondingly less and less a forum for academic discussion, andmore and more an arena within which socialists of all political tendenciesand of all theoretical persuasions come together to discuss crucial questionsof political strategy in a comradely and non-sectarian way . There is somefeeling within the CSE that Capita/and Class lags behind this development,partly because of the long gestation period of material that is published inCapital and Class, and partly because of editorial difficulties involved inreconciling demands for theoretical originality and rigour with those ofpolitical immediacy and relevance .

This year's annual conference provided clear signs that theoreticalwork within the CSE is beginning to match the development of the organ-isation . The conference was attended by almost 500 people, includingabout 100 from overseas . The conference was notable for a number offeatures . Firstly, it was a participatory conference. A large proportion ofthose attending were actively involved, directly or through CSE workingand local groups, in preparing material for the 100 working sessions .Secondly, discussion took place in small workshops and was, by and large,comradely and constructive despite the existence of a wide range of dif-ferent points of view . Thirdly, the form of work developed in CSE localand working groups, bringing together those primarily engaged in intellect-ual work and those who come to CSE through involvement in particularstruggles, is finding fruition in an embryonic fusion of theory and practicein work produced within CSE . This fusion was reflected in many of thepapers presented to the conference as links were made between theoreticalanalysis and political strategy in two directions : by those engaged in part-icular struggles in various sectors theorising their experience, and by those

engaged in theoretical work drawing out the practical political implicationsof their analyses .

The conference was notable not only as an intellectual and politicalevent, but also as a social event . The relaxed atmosphere provided a basisfor the renewal of old friendships and the making of new friends, breakingthrough geographical, political and institutional barriers . Finally the con-ference was notable in that for the first time a creche was organised thatattracted 24 children of all ages and contributed both to the increasedparticipation of women in the conference and to the comradely spirit ofthe occasion .

In the next year it is to be hoped that the CSE can develop in thedirections indicated by this year's conference . We need to develop theorganisational framework of CSE activities in the CSE working and localgroups, and members are encouraged to use the Newsletter to establishcontact with existing groups and to form new ones . We also need to in-crease the participation of those who are not professional intellectuals inthe CSE by further breaking down the barriers between intellectual workand political activity . In this the activity of CSE working and local groups, andthe work of the Education Committee is vitally important . We can thenhope that next year's conference, the theme of which is "Crisis of WorldCapitalism and Working Class Strategy: The Transition to Socialism", willdevelop the positive tendencies discernible in this year's gathering, andthat the contents of Capital and C/ass will increasingly reflect thesedevelopments .

If you are not a member of the CSE and would like to get involved inCSE activity you should join by subscribing to Capital and Class now .Details can be found inside the front page . Membership of the CSE notonly gives you Capital and Class, the Newsletter and the right to participatein CSE activities, including the conference, but also gives you privilegedaccess to the CSE Book Club that offers books published by the CSE andby commerical publishers at substantial discounts .

TRADE UNIONISM AND THE STRUGGLE

FOR LIBERATION IN SOUTH AFRICA

David Henson

The political and industrial organization of the black working class inSouth Africa has always been central to the struggle against exploitationand racial oppression . This paper examines the relationship between massstrikes, trade unionism, and revolutionary strategy in the liberation struggle .The 'stay at home' strategy of the 1950s has grown in the period of massresistance to apartheid into mass strikes and insurrectionary action byworkers and students. While black trade unionism is not openly revolut-ionary, trade unions are essential in defence of black workers, in supportingstrike action, and in advancing the demands of the workers .

Following the decade of relative quiet and economic growth based onpolitical repression and a massive inflow of foreign capital, South Africais now caught in the grip of a crisis of accumulation and political authority .The current rate of growth of the national economy has dropped to thelowest level since the Second World War, black unemployment has beenestimated at two million, and the rate of inflation continues to increase .The continuing industrial and political resistance following the mass strikesof 1973 and unprecedented resistance to the apartheid state has profoundlyaffected the possibility of capital regaining control through reforms .These struggles have raised acutely the position of black workers and thecharacter of the struggle for liberation . Given the long history of workingclass struggles and the importance of the black proletariat which includesover six million workers, the relationship between class struggle as a wholeand the position of organised workers is a crucial issue in the developmentof working class resistance to apartheid . Capitalism in South Africa is rela-tively advanced and has brought into being a working class possessing in-dustrial skills and political consciousness . What has been the role of theworking class (particularly organized black workers) in the struggle for

2

CAPITAL & CLASS

liberation? Have trade unions of black workers been able to survive theintense repression of the 1960s without becoming compromised andsubordinate organisations? What has been the response of organisedworkers to the mass struggles against apartheid, particularly during theSoweto uprising which followed the shooting down of demonstrators on16 June 1976?

Despite the intensity with which strikes by African workers have beensuppresscd, the immense problems in organizing African workers, and theextent to which the state has taken action against trade unionists organ-izing the mass of unorganized workers in South Africa, there is considerabledisagreement about the real value of trade unions of African workers inSouth Africa . The questions arise at two levels : firstly the relationship be-tween class struggle which is fundamental to all capitalist societies andracial oppression which is developed to an unprecedented degree in SouthAfrica. An analysis which concentrates on racial legislation and race rel-lations is central to the politics of liberalism which has deeply influencedopposition to apartheid both internally and internationally . The liberalanalysis denies the objective existence of classes in South Africa in prefer-ence for racial categories and sees political life being centred on the rel-ations between races . An analysis of class relations in South Africa posesa deeper problematic and a different political practice and has been in-corporated in the socialist movements which have grown in South Africaand increasingly within the national resistance . Secondly questions arise atthe level of the relationship between trade union action and politicalstruggle . Different views prevail about the relative importance of workingclass organization, and the relationship between working class strugglesand the seizure of political power . Studies in this area have questionedtrade unionism as the form of organization of the working class and haveevaluated the practice of trade unions from the perspective of potentialrevolutionary action .

The black working class in South Africa is brutally exploited and op-pressed under conditions which differentiate African workers from otherworkers and heighten divisions even among African workers (e .g . the dis-tinction between urban and contract African workers) . African workers,and to a lesser degree Indian and Coloured workers, are subjected to raciallegislation which defines their position in the labour process, to pass-lawswhich control mobility, and to service contracts which carry severe penal-ties for breaking contract with employers . African workers are deniedfamily housing in industrial centres except under the most stringent regu-lations. Legislation is now being introduced which will make unemploy-ment for African workers a crime, which provides for deportation fromthe urban centres, and forced labour colonies . All these measures aredesigned to reproduce a high level of exploitation and a particular form ofproletarianization : contract migrant labour. Through a long process ofclass struggle, African workers have fought against this subordination,utilized the contradictions within capital which require a 'committed'urban labour force, and have won limited and tenuous urban rights .

While the urban section of the black proletariat constitutes the van-guard of the working class in South Africa, millions of black workers (both

TRADE UNIONISM IN SOUTH AFRICA

3

men and women) who are denied urban rights and forced to maintainhouseholds on minute plots in the Bantustans should not be considered'peasant-workers' . It is undoubtedly true that imperialism in Africa hasproduced the phenomenon of 'oscillating workers' in an attempt to under-mine the political impact of proletarianization, but an uncritical descrip-tion of African workers as 'continually migrating peasant-workers' (Woddis,1960, p .46) reproduces the idea of an unstable and only partially devel-oped proletariat . As the struggle of the black working class has sharpenedin the urban areas the strategy of the ruling class of providing land grantsto separate 'nations' is being accelerated as a response to the crisis ofemployment and as a means of disorganizing working class resistance . Theresult of the process is that increasing numbers of African workers arebeing classified as 'foreign' to the centres of accumulation, urban rights arebeing undermined, and deportations to the rural areas are intensified . Thesquatters' movement which has grown in the urban centres is fighting des-perately against this trend of policy and consciously rejects any connect-ion to the land .

CLASS AND RACE OPPRESSION AND INDUSTRIAL RELATIONS

The disruption of the social cohesion of the African working class isclosely related to the internal structuring of the working class as a whole .Industrial legislation, in particular the Industrial Conciliation Act of 1956,coupled with the repressive labour lesiglation applicable to African wor-kers, maintains and extends the division of the South African workingclass to the extent that apartheid society has been described as having twolabour codes : a legalistic formal guarantee of certain industrial 'rights' toWhite, Coloured, and Indian workers ; and a repressive labour regime forAfrican workers (Hepple, 1971, p.1) . (1) African workers are restricted inupward and geographical mobility by industrial legislation and the oppres-sive pass laws, the state is firmly opposed to the uncontrolled unionizationof African workers, and strikes are prohibited under heavy penalities . Allforms of industrial action by African workers are prohibited under the de-finition of a strike which includes the refusal or failure to continue workor resume work or accept re-employment or to comply with the terms orconditions of employment applicable to African workers . Shop flooractivity such as 'go-slows' is prohibited : "The retardation . . . of the pro-gress of work or the obstruction by them of work" by any combination,agreement, or understanding . Mass resignations are also prohibited if "thatrefusal, failure, retardation, obstruction, breach or termination" is to in-duce any person to agree to demands for the changing of conditions of em-ployment or to re-employ any person (2) . The dichotomy between thetwo labour codes should not be extended too rigorously ; strikes (whichare similarly defined for non-African workers) are prohibited for allworkers unless exacting legal formalities are undertaken although the pen-alties for African workers are considerably more severe . The dichotomyshould, however, be a warning against automatically assuming the unityof all black (African, Coloured, and Indian) workers on the basis of com-

4

CAPITAL & CLASS

mon racial oppression . Industrial legislation which permits the registra-tion of trade unions having White, Coloured, and Indian workers as mem-bers and incorporates these unions in the system of collective bargainingmakes the simple division between workers on the basis of colonized andcolonizers made by Fisher inadequate (1977 :311). The complex frac-turing of the South African working class leads to the general indifferenceof most registered trade unions to the struggles of African workers be-cause of the positions occupied by their members in the labour processor through fear of losing their legal status and financial position . The rel-ationship between registered and unregistered trade unions organizingAfrican workers can in many ways be compared to that of craft and in-dustrial unions in the United Kingdom : initial opposition or superioritywhich becomes transformed by organization of new areas of trade union-ism into forms of competition to protect bargaining areas .

Through their participation in industrial councils (corporate bodieson which organized workers and employers are represented), registeredtrade unions are able to consolidate the position of their members, negot-iate for wages increases, and extend a wide variety of benefits only formembers of the registered trade unions. As the bargaining procedures havebecome routinized, so over the years the state has made relatively fewerofficial interventions in the racial allocation of labour ; the enforcement ofinformal job reservation (which increasingly protects Indian and Colouredworkers in secondary industry) is negotiated discretely at meetings of in-dustrial councils (3) . The industrial councils represent a highly developedindustrial relations system based on the social partnership of registeredtrade unions and employers and over one million workers of all races arecovered in the most strategic industries . Wages in the less significantindustries, excluding African miners and agricultural workers who aretotally unorganized, are revised periodically by the state Wage Board . To alarge extent the expansion of industrial councils in particular helps to ex-plain the declining militancy of White, Coloured and Indian workers .

The advanced nature of the industrial relations system in SouthAfrica poses important questions for trade unionism among black wor-kers. These trade unions are totally excluded from deliberations over jobclassification, wage rates, hours and overtime, benefits, complaints andgrievances and yet their members' jobs, wages, and working conditions aredetermined by this process . The industrial council agreements are highlytechnical documents: in a recent court case they were described as sub-ordinate domestic legislation and not a contract between union membersand employers (4) . All aspects of wages and conditions in key industriesare determined by a complex legalistic procedure which is impenetrable byunregistered trade unions. These agreements provide in minute detailcontrols over employment: for example, maximum hours in industry arelimited to 46 per week, child labour is prohibited, workers are given pay-slips, overtime rates are laid down . This formal superstructure of controlover production processes is an attempt to strengthen bourgeois ideologyamong all workers, to limit competition among capitals, and to reduceeven registered trade unions to the level of benefit funds . The completeexclusion of unregistered trade unions raises the question of their rel-


5

ative incorporation in collective bargaining procedures and their ability tostrengthen bouregois ideology .

There is at the moment an active debate taking place within the rulingclass in South Africa on the appropriate form of trade unionism for Africanworkers and the question of 'deracializing' labour relations . The issueshave come into prominence because of the crisis of accumulation (therehas been no real expansion in the gross domestic product over the past twoyears), the rising demands of black workers, and the demands of theinternational trade union movement for trade union rights in South Africa .Capital requires a decisive transformation of the labour process with rapiddeskilling and job fragmentation to restore a higher rate of return on in-vestment and make South African products more competitive .

POLITICAL STRUGGLE AND WORKING CLASS ACTION

The central question to the reforms being planned is the political con-sciousness of the black working class . The concessions which may beoffered (recognition for official African trade unions under state control,an end to job reservation, etc) are predicated on a belief in the underdevel-opment of class consciousness among African workers and a determinationto suppress class conscious leadership . The potential and significance ofblack working class action is seen fundamentally differently by revolution-ary and bourgeois theorists .

It has been argued that black workers are subjected to the most ex-acting exploitation and oppression by capital in the form of racial legis-lation . The intensity of exploitation with a penal labour code and wide-spread forced labour (particularly prison labour) and the form which thelegalized oppression takes (racism) has led to the concept of the super-exploitation of African workers arising from the `colonial' nature ofapartheid society . The superexploitation of African workers is more thansimply the fact of a higher rate of exploitation of the working class thanthat in advanced capitalist countries. "African workers", it is argued, "areexploited both as a class and as a race . . . national consciousness super-cedes class consciousness, but serves the same purpose . National antagonismis a form of class antagonism in South Africa " (Braverman, 1974, p.57) .This statement equates nationalist ideology with socialism as a form of`proletarian nationalism' and makes the development of classconscious-ness unproblematic .

While the idea that 'nations' may be exploited must always remainproblematic in Marxist theory (witness the debate about underdevelop-ment and unequal exchange), this statement is a definite attempt to linkclass struggle to the popular consciousness of the oppressed masses inSouth Africa. Black workers, from this analysis, suffer a two-fold oppres-sion; economic and national oppression . "The two are so interwoven thatthe worker has to play a dual role ; he has to fight for his economic rightsand simultaneously for his national rights" (Sechaba, November 1969,p.15) . In this context African workers develop, through their struggles, not'pure' class consciousness but revolutionary nationalism, which arises as

6

CAPITAL & CLASS

the antithesis of national oppression . Class interests from this analysis tendto become reduced to economic interests, and political practice is notbased on a working class strategy . The concept of national oppressionwhich implies an absolute control over the lives of African workers by theapartheid state provides a complete fusion of political (national) and ec-onomic (class) struggles. Concretely, all forms of social action by blackworkers are seen as being fundamentally political and potentially revolu-tionary, and industrial strikes then automatically become political actions ."Because of the setup in South Africa strikes organized by Africans growfrom economic struggle into a political struggle immediately the workerswithdraw their labour from any particular industry because they are illegal .Also in terms of the pass laws they are immediately open to arrest becausethey have broken their contracts by withdrawing their labour" (Sechaba,November 1969, p .15). The politics of African working class action isimmediately related to the totality of state repression and racial legis-lation . From the perspective of double exploitation (both as Africans andas workers) and national oppression every strike can be compared to arevolt of the oppressed, outright defiance of the racist state, and virtuallyan insurrectionary act. "The working class movement, because of theessential peculiarities of the African situation under semi-colonial and cap-italist exploitation, can only have one tendency - to develop into a des-perate all-out struggle, a struggle for complete victory over the forces ofwhite oppression and exploitation" (Magubane, 1975, p .24). The possib-ility of reforms confusing working class action or of trade union practicebecoming economistic is immediately negated by the revolutionary pot-ential of African working class action .

Bourgeois theorists of industrial relations and social change, to thecontrary, present the black working class in South Africa as disorganizedand incapable of determined political action . One of the few thoroughstudies of industrial relations in South Africa concludes that black workersare not fighting militantly for higher wages and better conditions and thatSouth African has "a remarkably low incidence of industrial strife" (Clack,1963, p .94) . Clack distinguishes between industrial strikes, which he des-cribes as of limited size and short duration, and large-scale work stoppageswhich are "only incidently strikes, having been political demonstrationsagainst general rather than industrial disabilities" (Clack, 1963, p .101) . Asother bourgeois theorists of industrial relations he insists on dividing in-dustrial from political action and avoids any discussion of the nature ofexploitation in apartheid society .

It has also been argued that strike action is a 'blunt weapon' in thestruggle against apartheid, and that industrial action cannot easily beturned into an insurrection . The black working class is presented as incap-able of disciplined national collective action : "at present, the black labourforce in South Africa is politically too isolated, too heterogeneous incharacter, too poorly organized and occupationally too unstable ." Strikeaction, argues Gann (1973, p .154) can always be neutralized by the"great reserve army of migrant labour" from the former British High Com-mission Territories, the countries along the northern border, and theBantustans with the country's own borders . Working class action is presented


7

as grossly economistic and easily diverted from political resistance. "TheDurban strikes of 1973 were representative of the determination ofAfricans and others to attain increased wages and economic benefits . Al-though Africans, much as workers elsewhere, are engaged in a continualquest for money, the additional amount desired is not at any time verygreat. Such demands as those articulated during the course of the Durbanstrikes (98 per cent for higher wages) were easily manageable and were infact quickly dealt with in the various industries by the White industrialmanagement" (Petryozak, 1976, p.542) . Bourgeois theorists devalueworking class action and present the black proletariat as capable of onlysporadic action and presenting demands which can easily be met byemployers or effortlessly destroyed by the state . They are determined toremove the revolutionary content from working class action .

TRADE UNIONISM AND WORKING CLASS ACTION

The political significance of trade unionism among black workers hasbeen questioned from a number of perspectives . The most advanced blacktrade unions have been described as inherently economistic, and doubt hasbeen thrown on the idea that working class militancy necessarily developsrevolutionary consciousness . In an article in the African Communist Davisis critical of all educational institutions and trade union organisations with-in the country as at best militantly economistic, or at worst operating todisplace SACTU (the South African Congress or Trade Unions) as theinternational representative of the South African working class (Davis, 1976) .At a deeper theoretical level is has been argued within the movement thattrade unions as such are not productive of a socialist consciousness andretard the fulfilment of the primary conditions of revolution . As a form ofworking class organization trade unionism is limited and counter-pro-ductive to revolutionary action. Even in the context of a capitalist crisis,trade unionism does not necessarily develop revolutionary consciousness,crucial to which is the pre-existence of the revolutionary party and therelative strength of the hold of bourgeois ideology over the proletariat .Bourgeois ideology, it has been argued, is strengthened by trade unionismand in this way trade unionism retards the development of the party .

From another perspective there is scepticism about open workingclass organization which stems from a view that the apartheid state is all-powerful and that the only form of struggle appropriate to the strengthof the state apparatus is armed struggle . Class-conscious workers, it isargued, should abandon the struggle to establish trade unions and acknow-ledge the primacy of political struggle by leaving the country to get profes-sional military training . The issue of working class rights thus can only besolved through military struggle ; in a national democratic state tradeunion rights will be granted to workers . This interpretation of the weak-ness of potential working class organization and action within the countrygiven the massive repression leads to an understanding that the apartheidstate will not tolerate any effective challenge to its rule and that any sur-viving organization of black workers is necessarily suspect. From this

8

CAPITAL & CLASS

point of view the geo-political developments in southern Africa, the liber-ation of Angola and Mozambique in particular, are creating the bases formilitary struggle which makes all other forms of struggle irrelevant oreven obsolete .

In this article the development of the labour movement among blackworkers will be related to the growing number of strikes, uprisings in thetownships and mines, and the general level of activity of the black workingclass . What were the conditions which have transformed South Africa froma relatively strike-free country to one in which workers are increasing thelevel of their organizations and increasingly taking industrial and politicalaction? To what extent have trade unions dampened or advanced revol-utionary consciousness among black workers? Is any effective black workingclass organization possible within the present repressive economic andpolitical structure, or are the only alternatives faced by working classorganization to be accommodated within these structures or be destroyed?Is it true, as a working party of the Ruskin Students Association haveargued, that trade union organization within the country must be prac-tically impotent until the present apartheid system is crushed? (RuskinStudents Association, 1975, p .5) . What has been the relationship betweenthe trade unions and the uprisings after Soweto? More generally, to whatextent is class action by black workers political and a challenge to theapartheid state?

It will be argued that trade unionism among black workers is notrevolutionary by nature, but only in relation to the development ofworking class struggle, and by creating bases for decisive action by blackworkers . Operating, as they are forced to do, outside the system of col-lective bargaining or functioning unionism, trade unions of black workersare forced through their weakness and lack of disciplinary powers torelate to the fundamental questions raised by the working class or bedestroyed . Their ability to advance working class consciousness is depen-ent on their response to the working class movement which is deepenedby the growth of an underground revolutionary party . It will be arguedthat trade unions, as open and legal organizations, are not the vanguard ofresistance, for resistance must be spearheaded underground during theintense repression now experienced by the black working class .

WORKING CLASS STRATEGY

The key issues in working class strategy in relation to practice inSouth Africa are those of the growth of mass working class resistance andstrike action and the development of underground working class organi-zation . These issues will be developed before moving on to an evaluationof South African experience of working class struggles . Despite theirimportance as a weapon in the armoury of mass struggle, political strikeshave not been considered inherently revolutionary. Despite providing acoherence to working class action the question of the purpose of masspolitical strike action remains: are they to be the basis for demandingpolitical reforms, for deepening the working class movement, or creatingthe preconditions for an insurrection?


9

In revolutionary theory mass strikes are necessary in creating the pre-conditions for seizure of power, but the numerous mass political strikeswhich have taken place in Europe (particularly Italy) and in Latin Americancountries (until recently) have not necessarily led to conditions for theseizure of power . In part this has been because of the failure of mass poli-tical strikes to achieve the disorganization of the state which was anti-cipated, but in many cases there was no conception of transforming generalstrike action into insurrection . A strong current of socialist theory hastransformed the Marxist theory of revolution into an ideology of inte-gration . German Social Democrats and trade unionists were, for example,hostile to the 'anarchistic' advocacy of general strikes as opposed to the"painstaking day-to-day work of strengthening the organization of theworkers". Party leaders such as Bernstein conceived of the general strikeas a means of accomplishing the proletarian revolution .

These views of the incorporation of the working class through politicalstrikes were vigorously opposed by Rosa Luxemburg who argued that massstrikes were the "method of motion of the proletarian mass" brought aboutthrough revolutionary struggle (Luxemburg, 1925, pp.44, 49) . She arguesagainst parliamentary fixation, the subordination of all political and massaction to parliamentary politics, and in particular against "practical poli-ticians" who couple the mass strike with the issue of universal franchise :"it follows that they can believe two things - first, that the mass strike isof a purely defensive character, and second, that the mass strike is evensubordinate to parliamentarism, that is, has been turned into a mereappendage of parliamentarism" (p.20) . Despite evoking the spontaneityof the working class, she does argue for the most responsive interactionbetween mass strikes and the party which is called upon to take up thepolitical leadership of the strike movement to bring about a revolutionarymoment .

Basic to Luxemburg's conception of the mass strike is the productiverelation between political and economic struggles, even to the extent ofturning the more conventional idea of economic strikes being transformedinto political strikes upside down : "the purely political demonstrationstrike plays quite a subordinate role" . Both forms of struggle are necessaryto each other : "Each great political mass action, after it has attained itspolitical highest point, breaks up into a mass of economic strikes" Be-tween political and economic demands there is "the most complete re-ciprocal action ." (p.46-47) . While mass strikes are led by the more con-scious workers, the crucial aspect of this form of struggle is the recruit-ment of the widest possible proletarian layers for the struggle, such asunorganised workers (pp.58, 67) . Through mass strikes there is a definiteadvance in class consciousness, in the material position of the workers,and even in attaining elementary trade union demands previously denied(pp.36, 59) . Far from being destroyed by participating in the mass up-surge, Luxemburg argues that it is only by seizing the moment to advanceindustrial and political demands that the trade unions could be transformedfrom collective bargaining agencies into mass proletarian organisations .

10

CAPITAL & CLASS

The subordination of mass proletarian activity to parliamentarystrategy was also firmly opposed by Lenin who was more concerned withtransforming the strike movement into a political and revolutionary strugglewhich faces the task of insurrection and the seizure of power . In this con-text he theorizes the proletariat producing leadership through organizationand action, containing its own reserves of energy in less advanced workers,and having interacting groups of workers within itself. Contrary to Luxem-burg he argues that organization is central to mass strikes : the leadingworkers are the most organized and it is this vanguard which develops thepolitical slogans and the longest fighting strikes .

In a crucial article which combines his theory of the working classvanguard with statistical analysis, Strike statistics in Russia (1910), Lenindevelops the concept of an active working class putting forward demandsand develops strike committees and distinguished between differentlevels of consciousness in the working class . The advanced workers drawin the 'average' and 'backward' workers and provide the political leader-ship in the struggle . (CW 16, pp.393-421) . From an analysis of the 1905revolution he distinguished three phases in revolutionary strike action :•

The bold political demands of the vanguard at the inception ofmass strikes

•

Support provided for this leadership by economic strikes•

As the vanguard finds its position weakening, support for its politicaldemands comes from 'average' workers who encourage the vanguardto maintain its stand

The advanced workers, made up particularly of metal workers, were thebest organized in the largest industrial plants and were the first and mostpersistent in striking for purely political objectives . They had a crucial rolein mobilizing the mass of workers to action through advanced slogans andaudacious strike demands which spurred on the backward sections into themore modest, and often economic, demands . The wave of largely economicstrikes in turn sustained and gave force to the emphatically political strikesof the advanced sections .

Each forward move encouraged new, unorganized sections of the pro-letariat to seize the chance to improve their conditions, and these claimswere generalized and radicalized in the demands of advanced workers . The'average' workers can then make the transition to political demands asoutlying areas of the country and unorganized sections of the class aredrawn into battle . This theory of revolutionary waves of mass strikesdepands on the closest connection between political and economic de-mands as the most unorganized sections can be aroused only by "the mostextraordinary accentuation of the movement" and by economic demands .

"The interdependence between the economic and political strike isthus quite obvious: no really broad, no really mass movement ispossible without a close connection between the two ; the concreteexpression of this connection consists, on the one hand, in the factthat at the beginning of the movement, and when new sections arejust entering it, the purely economic strike is the prevalent norm, andon the other, in the fact that the political strike rouses and stirs the


11

backward sections, generalises and extends the movement, and raisesit to a higher level ." (CW 16, p .414)

Lenin argues, in short, that the political demands can only be sustainedwhen backed by economic strikes, and that the movement as a whole canonly be developed by drawing in the most unorganised workers througheconomic demands. It is this constant interaction and transformation ofdemands and strikes which makes possible the telescoping of evolutionarysocio-political development into revolutionary advance .

The upsurge of mass strike action may not have been precipitated bythe immediate commands of reolutionary organization, but the fact ofmass working class action creates the basis for wider organization of theworking class. The new leadership, new militant groups, new tactics, whichare created at the moment of mass action draw upon the revolutionaryexperience of previous resistance ; the history and experience of workingclass struggles which bring forward coherent working class demands . AsLenin has argued, there are degrees of 'spontaneity' and political maturityin the working class. During an upsurge in class action the secret and under-ground organization of workers takes decisions on the timing of strikes,formulation of demands, and action against informers . (5) It is this acti-vity which creates the bases and reserves both for the legal trade unionmovement and for the revolutionary party . It is during mass action thatthe concrete connections are made between organized revoltionaries andthis underground : the hundreds of revolutionaries are turned into thou-sands of leaders of mass proletarian action . This concept of a working classmovement relies on a developed network of social relations which formthe basis for working class solidarity within the docks, factories, plant-ations, offices, and other working places .

As the working class advances through struggle and develops class aspir-ations which are first manifest in demands for control over production,capital responds with more refined forms of bourgeois ideology which at-tempt to win over the organic (unofficial and unelected) leadership of thework groups in production . The organizational strength of the workingclass depends, therefore, on the links of the revolutionary movement withthe organic leadership of workers in production and the overall develop-ment of revolutionary ideology . In most capitalist countries these devel-opments lead to an open trade union movement within a socialist move-ment. In highly repressive states such as South Africa the immense pres-sure of class and racial oppression has virtually eliminated organizedworkers as a group during crucial phases of accumulation .

While a party constructed on the principles of 'democratic centralism'in the Leninist mould may find it possible to survive under severe repres-sion as a highly integrated, secret group organized on the basis of indepen-dent cells linked horizontally to other cells, a trade union is difficult to con-tinue as an effective mass organization . The effectiveness of trade unionorganization is closely related to the mass movement of workers and to thegrowth of broad and popular organization. The problems of an 'under-ground' trade union are immense . Nevertheless they have been concep-tualized as illegal trade unions following a section in What is to be done? in

12

CAPITAL & CLASS

which Lenin argues that a small compact, propaganda group could per-form the functions of a trade union organization .

A small, compact core of the most reliable, experienced and har-dened workers, with responsible representatives in the principal dis-tricts and connected by all the rules of strict secrecy with the organ-zation of revolutionaries, can, with the widest support of the massesand without any formal organization, perform all the functions of atrade union organization, and perform them, moreover, in a mannerdesirable to Social Democracy . (Lenin, 1902, p .459)

During the intense repression of working class action in the 1930s thispoint was taken up in the socialist movement in response to those whoargued that illegal trade unions were impossible to organize . The argumentcentred around the purpose of organizing; either to win the right to openexistence as a trade union, or to develop the level of action by the workingclass which would make illegal organizations possible . The issues whichconfronted revolutionaries at the time were whether the illegal tradeunions would be reduced to 'narrow sectarian groups' out of touch withthe mass of workers or would be able to continue as a means of developingthe links between revolutionary organization and the workers . The stra-tegy evolved by the theorists of the illegal unions was that of a secret leader-ship operating through factory and residential associations directing themass struggle on the basis of economic and political demands and partic-ularly of the right to organize .

The illegal unions were to be founded in the strike committees andassociated with party nuclei, they would not have membership cards orsubscriptions, and would struggle to be as open and influential as possible(in distinction to the underground party) . There existed two concepts ofillegality in trade union organization : functioning openly despite stateorders and not submitting to any attempts to close down organization, oraccepting that fundamentally new conditions existed and moving on to aconsciously underground basis (Wojtkiewicz, 1930 ; Santo, 1930) . Throughmass organization and militant action the whole structure of repressionwould be broken down and the illegal unions emerge into the open fromthe factories, sustained by a network of factory committees and the fulluse of ancilliary organizations .

The discussion of illegal trade unionism has a direct relevance fortrade union organization in South Africa in which the most extrememeasures of repression have alternated with a highly legalistic approach tothe problems of organization . The question of illegal trade unions isreflected in the concept of the underground in South Africa ; permanentcentres of resistance within the country and the articulation of workingclass demands . In South Africa the mass party (the African NationalCongress) which organizes African people on a territorial basis advancedits relationship to working class demands through the adoption of theFreedom Charter which calls for the nationalization of monopoly capital,mines, and banks . This relationship has been developed systematicallythrough the adoption by SACTU (South African Congress of Trade


13

Unions) of the Freedom Charter and through its participation in theCongress Alliance which organized resistance to apartheid during thephase of legal extra-parliamentary opposition in South Africa . In this sensethe movement of national resistance came to relate to organized workersboth programatically and organizationally through the trade unions whichmade up SACTU. By the late 1950s the formulated relationship was thatthe African National Congress was the 'spear' of the African people whileSACTU was the 'shield' . Members of each organization were encouragedto become members of trade unions and the political organizations whichmade up the Congress Alliance . On its prohibition in 1960 the links be-tween the industrial and political organizations became cemented under-ground .

TRADE UNIONS AND THE STRUGGLE FOR DEMOCRACY

This history of the working class in South Africa includes the his-tory of continual struggle to expand trade unions among African workers .At no time since 1919 (the year in which the Industrial and CommercialWorkers Union was established) has there been a period in which tradeunionism has been completely suppressed among African workers . Butthere have been widely fluctuating cycles of growth, disorganization, re-pression, and revival related in a complex manner to phases of capitalaccumulation and the force of state repression . Mass unionism flourishedin the 1920s with the expansion of the Industrial and Commercial Unioninto a political movement . The first major phase of industrial trade union-ism came during the Second World War which was marked by a massiveupsurge in strike action . Organization and strike action stimulated a pro-longed debate within the ruling class over the advisability of recognisingor registering trade unions of African workers . The temporizing of theAfrican petty bouregoisie in the Native Representative Council and thedebate about incorporation of African workers into collective bargainingwere brought to an abrupt halt by the mass strike of African mine workersin 1946 (O'Meara, 1975) . New forms of struggle were evolved which com-bined working class action and national political campaigns . A study ofthese developments is crucial to an understanding of the strategy of resist-ance as it was the formative period in the development of the politicalprogram and mass organizations of oppressed people in South Africa .

The assimilation of the mass strike into the experience of the blackworking class arose out of the struggle against the suppression of com-munism and socialist ideas in South Africa . The Communist Party ofSouth Africa and the African National Congress called a national strike on1 May 1950 which was 80 per cent successful in the industrial Witwaters-rand. Police action resulted in 18 blacks killed, more than 30 wounded andwidespread victimization of participants, particularly by municipalities .(Guardian, 4 May 1950) In response the black political movements calledworkers to repeat the strike on a national basis on 26 June, in remembranceof the dead and wounded . This form of protest was particularly successfulin Durban, Cape Town, and Port Elizabeth, (the Witwatersrand was

14

CAPITAL & CLASS

apparently less responsive), and 26 June became an annual occasion forcampaigns and political strikes .

The inception of mass strikes on a national level raises the obviousquestions of the objective of the struggle, the relationship between tradeunions and the political party, and the ideology of resistance . The politicalcampaigns and national strikes were aimed at a defence of existing demo-cratic rights and a struggle against new forms of racist legislation . Thecrucial features of the period were the use of the 'stay-at-home' as a tech-nique of mass resistance, the holding of the Congress of the People whichadopted the Freedom Charter, the formation of the South African Congressof Trade Unions (SACTU), the legal prohibition of the African NationalCongress (ANC), and the formation of the armed underground . Thesedevelopments were compressed in the years between 1950 and 1961, atime of increasing repression and resistance culminating in the massacreat Sharpeville and the legal suppression of the African nationalist move-ments .

The idea of a national strike of black people to deal a decisive blowagainst the white government was not altogether new, and is a syndicalistcurrent running through the ideology of African nationalism (similar in asense to the Irish 'rising in the night') . In African nationalism the strategyis based on the argument that the African people was the only productiveclass . Prior to 1950 the idea was used in a rhetorical way by Kadalie,leader of the Industrial and Commerical Workers' Union, as a means ofdemanding political reforms . In practice Kadalie renounced the strikeweapon and ignored the struggle of ICU members in the coal mines, dia-mond fields, and docks, and his method of posing the problem of articu-lating the power of the black working class never reached the level oforganizing national political strikes . The political strikes of the 1950s werewell advanced in the conception of collective resistance, in the sloganswhich were advanced, and in their organization . The problems of depth ofsupport, the relationship between political and economic demands, thesynthesis of struggle in the townships and in the factories, and the isolationof mining and agricultural workers remained as crucial weaknesses .

The stay-at-homes were not intended to become prolonged strikeaction (the Unity Movement from a syndicalist position felt they shouldbe indefinite to be sincere), on the contrary they took a deliberately limitedform (in any year never more than three days) so as not to exhaust thepolitical capacity of the workers . In this form the stay-at-homes (in a sim-ilar manner to the May Day strikes in Europe) were political demonstrationstrikes designed to put forward the political demands of black people tothe state rather than mass strikes which combined the struggles in thetownships with those in the factories and directed workers attention to thecentres of political control . In some literature sympathetic to the politicalresistance they were described as political demonstrations in a mannersimilar to other campaigns of defiance against unjust laws and againstBantu Education .

The perspective of the debate around the combination of political andeconomic demands, the relation of advanced workers to unorganizedworkers, political demonstrations to mass strikes, and the transformation


15

of mass strikes into revolutionary action provide a basis for an analysis ofpolitical strikes in South Africa. Following the political strikes in oppositionto the implementation of the Suppression of Communism Act, the AfricanNational Congress with other mass organizations advocated a widespreadcampaign of defiance of unjust laws in 1952 . This call was supported bythe trade unions of African workers, and large numbers of workers wereinvolved in political action similar to that organized by blacks in the UnitedStates which led to imprisonment . The greatest advance in the ideology ofresistance was made at the Congress of the People, the most representativeassembly of the South African people, held in 1955 . The political goals ofthe popular movement were enshrined in the Freedom Charter drawn upat this assembly which called for universal franchise, a division of the land,and the nationalization of monopoly industry, banks, and mines . TheSouth African Congress of Trade Unions (SACTU) formed in the sameyear adopted the Freedom Charter as its own program . From its inceptionon 5 March 1955 SACTU recognised that the industrial struggles ofworkers were "linked inextricably with their struggle for political rightsand liberation from all oppressive laws" (Levy, 1961, p .36) . SACTUjoined the National Consultative Committee which coordinated theactivities of the black political movements in joint campaigns and politicalstrikes .

The 'stay-at homes', called by the Congress Alliance in support of thedemands of the Freedom Charter and specific campaigns, became theprincipal form of struggle in the late 1950s . The turn towards mass actionby the black working class on the part of the African National Congressoccurred because of the rising level of working class action in the factories,but even more dramatically in bus boycotts and resistance to the stricterimplementation of the pass laws. The term 'stay-at home' reflects some ofthe ambiguities of mass action during this period. The focus for action wasin the townships rather than the factories, and the form of action carriedimplications of being less than militant political action. The fact that thepolitical strikes were carried out in support of national political slogans didnot immediately link the struggle in the townships and in the factories norlead to the formation of industrial strike committees to press the demandsat the point of production . The problem remained of transforming thepolitical strike action (which, in Luxemburg's langauge, took the form of ademonstration strike) into a mass strike which fused the political and eco-nomic not only in slogans but which also focused demands on employersand in the urban political centres .

Given the conditions of urban segregation in which black townshipsare deliberately sited away from the industrial areas and centres of statepower, the 'stay-at-homes' tended to direct political attention away fromthe point of production and the central organs of the state . The success ofthe stay-at-homes in 1957 on the Witwatersrand, in 1958 nationally, andthe massive strikes in 1960 and 1961 were more dependent on the devel-opment of an element of popular power in the black townships throughstreet organization, roadblocks, and general control by pickets of the out-lets to the industrial areas than on a high degree of industrial organization .The comparative advantages of community organization (usually supported

16

CAPITAL & CLASS

on the streets by the lumpenproletariat) has to be set against the fact thatthese townships have been precisely designed for external control by thepolice and army . Through an overwhelming presence, cordons and searches,the state forces can match popular local power and then come to dominatethe area through physical presence or gunfire . The other problem of theconcept of popular power in townships is that it can degenerate intodemands for self-government on the basis of segregated areas (a tendencyencouraged by the black petty bourgeoisie) rather than leading to a con-frontation with state power. Ironically the most decisive moments ofpolitical struggle in the 1950s occurred when black workers marched intothe city centres in Cape Town and Durban in 1960 and the state wasforced into making temporary concession on the primary issue of struggle :the pass laws . The police's role was then completely reversed : from forcingblack people out to work, to guarding the outlets to stop any marches onthe centres of political control . The acceptance of conditions of urbansegregation in a strategy of resistance implied by the 'stay-at-home' inthese cases became transformed into potentially insurrectionary action :'March on the city!'

Despite these objective problems with the stay-at-home as a form ofpolitical strike based in the township, the support given to the strike move-ment by SACTU through its component unions widened and deepened thesupport given by workers to the political strikes . While statistics of thestay-at-homes were generally given by area rather than industry, it is sign-ificant that industries even only partially organized by SACTU had a sign-ificantly higher level of participation . This point confirms the Leninistargument that an organized vanguard is crucial in providing leadership tothe less advanced workers . As the SACTU unions were concentrated inlight manufacturing and services (e .g . textiles, dry-cleaning, food andcanning, etc) support by heavy industry, power, communications, andtransport was more problematic . In mining and agriculture insulated fromthe political centres, support for political strikes was impossible (the dif-ficulties were made even greater by the fact that black workers in thesedifficult sectors were predominantly housed in compounds under thedirect control of their employers and the police and were relatively or ab-solutely isolated from trade union or political organization) . These con-siderations demonstrate how difficult it was for mass political action topenetrate key productive sectors and the limits on the stay-at-home as amethod of disrupting production and disorganizing the state . An import-ant problem in disrupting communications, railways, and the core of themetal industry has always been the existence of white workers who areprepared to act as strike breakers and emergency workers .

The growth of political strikes in the late 1950s was largely due to theincrease in working class action reflected in the '£1 a day' campaign ofSACTU which aimed to mobilize the mass of black workers into tradeunions. It was this slogan which developed the momentum of the 1957stay-at-home which was successful on the Witwatersrand and which led tothe Workers' Conference in 1958 which called for mass strike action insupport of demands to abolish the pass laws and to secure a living wage .The Conference's call for mass strike action caused a split in the leadership


17

of the national movement as it raised the question of the leadership ofthe mass resistance to apartheid : was it to come from the trade unionsand workers in the political movements, or directly from the politicalparty as the supreme authority? As posed by the Africanists ('purified'African nationalists) the question was whether the African national move-ment could be dictated to by non-Africans (trade unionists and Marxists) .

Following a series of discussions in the Congress Alliance it wasdecided that the stay-at-home should proceed but that it should coincidewith the date of the national election and should centre around the slogan'Nats must go!' (Afrikaner nationalists must be voted out of government) .These rearrangements, plus the hostility of the Africanists and the intenseforce of police repression, disorganized the strike movement . There weredeeper problems in the subordination of proletarian action to parliament-arism (by the mass strike supporting indirectly the white oppositionparties), which tended to widen rather than narrow the separation be-tween political and economic demands . Since a significant section ofcapitalists in manufacturing and commerce supported these oppositionparties the division between the struggle against the state and the struggleagainst employers was widened . A writer sympathetic to the national move-ment wrote that black workers tended to tell employers they had been'intimidated' into staying away from work . "It was an easy excuse to giveyour employer - you passed the buck on to some force greater thanyourself" (Drum, Agust 1957, p .21). Strategems such as these tended tovitiate the political significance of strike action. The 1958 stay-at-homewas conceded as a failure after the first day by the organ of the CongressAlliance . (New Age, 17 April 1958) .

While it was true that the national strike was not a success, the com-bination of political and economic demands did bring about importantindustrial strikes (in both cases with workers housed in compounds) in theDurban docks and in a glass factory (to name two examples quoted inthe press), which showed that the political strikes could detonate industrialstrikes around the demands of the political movement and provide a basefor further political action . The inability to develop a sustained mass char-acter and base for political demands through economic strikes limited bothindustrial and political struggles to manageable proportions for the state .The stay-at-homes, from the viewpoint of the ruling class, could be con-trolled through raids and arrests of the political movements and tradeunions and by occupying townships on the days of action, and industrialstrikes isolated and broken by baton charges and arrests .

The apparent contradiction between a high level of political activityby the black working class and the relatively low strike rate in industrywas a result of the difficulty in establishing recriprocal integration betweenmass political campaigns and industrial action . Despite a mass organiza-tional drive by SACTU and the popular slogan of f1 a day, the number ofstrikers increased only slightly in 1955-60 above the figures for the early1950s .

This slow expansion of strike action in the factories can be explainedpartly by the accent on political struggles, the low rates of growth inmanufacturing industry, the stricter application of pass laws, and by poli-

18

CAPITAL & CLASS

tical and industrial repression . The campaign for higher wages and boycottof buses caused employer associations to phase in wage increases graduallythrough increasing productivity and thus increase the differentiationamong black workers. In the crucial manufacturing sector employment ofAfrican workers grew slowly ; only some 5,397 new jobs for Africanworkers became available between 1955-56 and 1959-60 (6) . During thisperiod control over the movement of African workers was drastically in-creased and extended to women workers . Through the extension of labourbureaux the state intensified pressure on 'redundant' workers within theurban areas and between 1956 and 1963 there were almost half a milliondeportations from the urban areas .

The high level of organizational support provided by SACTU unionsto strikers led to a strong repressive policy by the state against workstoppages of all kinds. In describing black trade union organization duringthis period, Levy writes that strikes resembled "small-scale civil wars" .Strikes by black workers met with "lorry loads of police armed with batons,sten guns and tear-gas bombs, great pick-up vans arrive and all the strikersare arrested" (Levy, 1961, p .39) . All attempts to get higher wages and bet-ter conditions, for the smallest advance in factory conditions or reinstate-ment of unjustly dismissed fellow-workers were immediately met by thefull force of the state .

While the overwhelming number of strikes took place on the initiativeof the workers themselves (many strikes took place in unorganized fac-tories), the resources of the SACTU unions and the Congress Alliance (insome notable struggles in the textile and tobacco industry) were thrownbehind the workers. Between 1955 and 1960 when organizational supportwas provided by SACTU, 421 strikes took place involving 34,854 blackworkers. Of these strikers 3177 were arrested and charged by the police(many more were arrested and then released), some 11 per cent of thetotal number of strikers (7) . These official statistics reflect only the sur-face of repression, employers themselves took decisive action to smashindustrial and political strikes . Often these counter-measures were theresult of close collaboration between employers, the police, and govern-ment departments . Strikes by migrant workers (the most vulnerable sectionof the African working class) were met with mass dismissals and deport-ations from the urban areas . Migrant workers on strike, for example, inthe stevedoring industry in Durban in 1959 were summarily deported fromthe urban area . (Hemson, 1976) .

Despite the tremendous handicap of intensified suppression of poli-tical and industrial organization in 1960 when mass political movements inSouth Africa were prohibited, the trade unions associated with SACTUput up a determined struggle for survival . Finally the full force of the Sup-pression of Communism Act ; arrests, detention, and bannings, was usedagainst all working class organization . Between 1960 and 1966 over 160SACTU office holders were banned from taking part in SACTU or anyother trade union activity . (SACTU, 1976a, p .21) Throughout the 1960sthe level of strike action undertaken by black workers fell considerablybelow that of the 1950s undoubtedly because of the intense repression ofthe time, despite a rapidly expanding economy . While there were few


19

major strikes over the decade 1960-1970, strike activity persisted, part-icularly among migrant workers (especially the Durban stevedoring workers)and in the expanding and low-wage textile industry . The small number ofstrikers (there were only 1,708 black workers on strike in 1969) led somewriters to conclude that strike action was falling irresistably(Gervasi,1970, p .32) .

These years were a period of setback for the political resistance,marked by drawn-out political trials and long prison sentences for theleadership . Many working class militants left the country and joined themilitary wing of the African National Congress, and intense repressiondrove the SACTU unions underground . These developments were accom-panied by a hesitant organizational program by reformist trade unionswhich met with state opposition and was finally suspended in 1969 . Theseyears were also marked by the state policy of Bantustans : the consolidationof reserves into regional political units on an ethnic basis, designed todivide the African people by offering some advantages to the African pettybourgeoisie and consolidating controls over the African working class . Anew phase in resistance to apartheid was marked by the development ofblack consciousness, an ideology of black awareness and psychologicalliberation from the categories of racism, and the opening of armed strugglesby ANC guerillas in Zimbabwe .

MASS STRIKES

While agitation on issues such as poverty wages in 1972 helped tocrystallize working class action, the conditions for mass action wereripening in terms of crippling price increases which rapidly lowered thereal wages of black workers . These price rises were a reflection both ofthe growth of inflation among all capitalist countries and the crisis of ac-cumulation in South Africa, marked particularly by the inability ofSouth African manufactured commodities to compete on internationalmarkets. The increases were particularly steep in the cost of essentialitems such as food, clothing, and transport . Calculations of the povertydatum line made over a number of years give an idea of the escalation ofprice increases affecting black workers .

Mean poverty datum line index, Durban, 1958-1973 (8)

The data show that it took 13 years for basic prices to rise 40 per cent up

Year Mean PDL (R) Index 1971=1001958 41 .05 60.51971 67 .87 100.01972 74.66 110.01973 95 .26 140.4

20

CAPITAL & CLASS

to 1971, but the same rise in basic prices took only two years thereafter .The price rises between 1972-73 (some 30 per cent) were extraordinary inthe experience of the black working class . The rising level of working classactivity which accompanied these price increases accelerated into a seriesof mass strikes unparalleled in South African history .

It is estimated that some 100,000 African workers were involved instrikes in 1973 which originated in Durban and spread during the yearthroughout the country . In the Durban industrial areas is concentrated thelargest number of African industrial workers in the country, even morethan in Johannesburg : 165,000 and 159,000 African factory workersrespectively. (9) The industrial workers in Durban tended to be concen-trated in larger factories (both in the textile and metal industry) and thestrike movement swept rapidly from one industrial area to another . Themass strikes in the Natal province demonstrated the unity of the blackworking class across the divisions between migrant and urban workers,between different industries, and even between industrial and agriculturalworkers. Contrary to bourgeois theorists the strike movement provided adirect challenge to employers and the state by demanding a doubling inwages which could only be met by threats of mass dismissal backed bypolice action . Police were flown in from Pretoria to patrol the townshipsand industrial areas.

The organization and development of the strike movement has to berelated to the historic experience of the black working class in the stay-at-homes, although as a sustained form of proletarian action independent ofexplicit centralized organization the strikes indicate a growing maturity ofworking class consciousness . There appears to have been some debateamong the workers whether there should be a boycott of the train services(which raised fares early in the year) or strikes against employers. Thewidespread rumour of a bus boycott caused the state to intervene massivelyin the townships to crush any picketers . The mass strikes at the point ofproduction certainly disorganized the state's response and created a gapthrough which the strike movement could spread . Initially the police werereduced to dashing from the townships to one strike, and then to otherstrikes in different industrial areas .

Surveys of the mass strikes (Douwes Dekker, 1975 and Boulanger, 1974)argue that the strength of the strike action lay in the low-wage, large-scale factories characterized by oppressive management, frequent dismis-sals, and victimization of potential spokesmen . The textile industry(which is highly concentrated in the Durban area in particularly largefactories) produced some of the most prolonged struggles of the time,although the metal industry experienced a higher number of strikes(IIE, 1974, p .29). Striking workers called on neighbouring factories tosupport their demands, mass pickets gathered outside the factories to stopany scabs . Virtually without exception the strikers refused to put forwardrepresentatives, aware that there were massive powers of victimizationavailable to the management and police and wary of the compromisedleadership of the employer's favourites . The 'negotiations' took the formof workers shouting demands for between R20-R30 and refusing to electleaders or return to work . Employers and the police responded by threat-


21

ening mass dismissals and prosecution and eventually offering some minorwage increases . The workers then shouted down the offer, but were forcedthrough necessity eventually to return to work . As one factory returned towork, another came out on strike; for a number of days there were 20,000workers on strike from a number of industries and services, reaching apeak of 50,000 strikers at a single moment . The momentum of the strikewave was ultimately broken through the growing presence of the policein army uniforms flown in from other centres, and the lack of strike funds .Street demonstrations by workers which threatened to draw all the strikerstogether, call out non-striking factories, and make the strike general wereattacked and dispersed by the police . Despite the relatively small wageincreases (between R1 and R2,50 per week) the consciousness of the wor-kers had advanced considerably beyond the relative caution of the periodof repression before the strikes . The overwhelming number of strikes hadbeen successful (in the sense that wage increases had been won and strikersnot dismissed) because the strikes were able to take on a mass form . Inthe period from January to the end of March 1973 there had been morethan 160 strikes involving 61,410 workers (10) .

The mass strikes were an explosion of working class energy ; in thesethree months more black workers were involved in strike action than inthe previous twelve years after Sharpeville . Faced with this unprecendentedupsurge and the defiance of apartheid legislation prohibiting strikes andmaintaining black workers in subordination, the ruling class was placed ina quandry . Was it possible to make mass arrests of strikers? Would not thisinflame the workers to new heights to class activity and deepen the poli-tical content of the strikes? Given the mass character of the strike move-ment the police were reduced to dispersing marches through the centre ofDurban and using teargas on strikers in Hammarsdale and Richards Bay . Inthe face of united mass action the police were reduced to saying: "Thepolice have nothing whatsoever against people demanding higher wages -provided they do not break the law", a statement which amounted to asurrender to proletarian action (IIE, 1974, p .20) Black workers werelosing the fear of police retaliation in acting illegally .

Of the 98,029 black strikers in the whole of 1973 relatively few wereprosecuted for striking : 207 or 0 .2 per cent of the total . These figures canbe compared to the 822 out of 3,462 workers prosecuted for striking in1959, 24 per cent of the total . Partly this can be explained by the impot-ence of the police when faced with mass proletarian action centring on thefactories, and by the changes which had taken place in the composition ofcapital in the 1960s towards more capital intensive industry in whichwages formed a smaller component of total costs . A more capital-intensiveindustry in conditions of reasonable protitability could afford to increasewages and not be as dependent on police repression as labour-intensiveindustry, for instance, the stevedoring industry . Wages could be raised im-mediately within certain limits, the total wage bill could be held constantby insisting on higher productivity and firing less productive workers .

The strikes brought forward the political as well as the economicdemands of the black working class. The level of wages demanded by the

22

CAPITAL & CLASS

workers could only be brought about by a transformation of apartheidsociety . The discipline and united action of the workers raised a funda-mental demand for the right to associate and organize freely . The massdefiance of legal subordination and control showed a complete rejection ofthe idea of a passive working class accepting employment as a privilege andunsure of its own potential . Within the resistance to apartheid the massstrikes demonstrated the leading position of the black working class in thestruggle against apartheid : the "largely illiterate and semi-literate" in thewords of a black consciousness writer .

The state was well aware of these issues and responded with apparent'liberalism' to draw African workers more firmly into the state systemof labour representation. The Bantu Labour Relations RegulationAct (No 70 of 1973) brought African workers into line with White,Coloured, and Indian workers in respect of the legality of strike actioneven though the penalties for striking were more severe for African wor-kers. All workers in South Africa are prohibited from striking during thecurrency of an industrial agreement, when employed in essential services,or in other specified industries . Despite the amended legislation, legalstrike action for African workers is only possible after the completion of ahighly legalistic procedure involving a series of compulsory delays . (Horner,1976a, pp.1 7-18) Even granted these procedures, legislation such as theRiotous Assemblies Act and other security legislation also makes strikes byAfrican workers illegal .

The other important feature of the revised 'Bantu labour relationssystem' is the provision for liaison and works committees on a more de-fined statutory basis . The liaison committee, made up of equal represent-ation by workers and management, is particularly appropriate for the ex-tension of a human relations ideology in industry which has been import-ant throughout capitalist economies for dampening class conflict . Thelegislative reforms, worked out in close collaboration with employer assoc-iations, marked an important anticipation of the changes in managementof labour which would be needed in a period of increasing working classaction. In the words of the SACTU Secretary General, John Gaetsewe :"The net effect of these new measures . . . is not to strengthen the hand ofAfrican workers, but to strengthen the power of government bureaucrats"(Ruskin Students Association, 1975, p.20) .

TRADE UNION ORGANIZATION

The mass strikes in Natal, involving a successful defiance of repressiveleglislation, created a new mood of confidence in the black working classand an eagerness to build working class organization . The strikes them-selves were only possible through the leadership of the underground,whichexists as the groupings of workers acting consciously, illegally and secretlyto carry forward working class struggles . While the mass strikes wereundoubtedly spontaneous, in the sense of not being planned from apolitical centre, they were not unorganized, although this organization wasnot open or unified in an overall strategy . The strikes were not merely aseries of heightened social interactions nor a result of communication be-


23

tween social groupings in bus queues, but collective action by workersguided by working class leadership .

While underground working class organization and strikes withouttrade unions may be difficult for industrial relations specialists to under-stand these are real phenomena in working class struggles in South Africa .Decisions on strike action are actually taken among workers, and leader-ship comes into being through groupings in production . The fact thatworkers refuse to elect representatives during strikes is not so much anindication of a 'leaderless' situation, as an astute assessment of the powerof the employers and the state to smash the hard struggle to build organi-zation and leaders .

One of the most illuminating examples of the underground is providedby the action of the Durban dock workers in demanding increased wagesin 1972 ; an anonymous but well drafted letter laid out the workers' de-mands to the Wage Board . In other instances the demands of the workershave been set out and put on a noticeboard with the time limit for manage-ment to respond, 'round robins' and petitions have been organized, andsecret strike committees have been set up . This level of organization comesout into the open only when strikes are reported or in the trials of workersaccused of striking, but point to the existence of growing undergroundactivity by workers not only at the point of production but also in thetownships . Independent trade unionism in South Africa among blackworkers is only possible with the support of the workers' underground forthe union officials and, in turn, the unions' ability to protect and extendthe organization of the underground. The success of independent tradeunionism is thus dependent upon the support of the organic leadership inthe factories.

Black trade unions differ significantly in their attitudes to workscommittees, the state, responses to strike action, registered trade unions,and Bantustan governments. All of them do, however, have to struggle forexistence within the context of the state 'Bantu Labour relations system'which is designed to frustrate industrial organization of black workers andundermine their trade unions ; in the words of the Minister of Labour in1953, to "bleed the unions to death" . This policy was not modified afterthe mass strikes in Natal, despite some division among employers on theappropriate response to working class organization . The Minister ofLabour, Marais Viljoen rejected any form of accommodation with blacktrade unions, despite the demands of black workers, early in 1973 : "Iwish to state quite unequivocally that the government will not considerthe recognition of Bantu trade unions or their organisation or affiliation ina way which is tantamount to recognition . (11) The Minister hoped byforcing the incorporation of workers' representatives into the committeesystem and through repression to make the unions irrelevant .

Despite this official hostility, the banning of trade union officials,and the suppression of strikes, trade unions among black workers havedeveloped as a result of the growth in working class action . Following astrike of transport workers in Johannesburg a transport union was formedin 1972 . In the buoyant atmosphere of post-strike Natal, the membershipof the General Factory Workers' Benefit Fund (a proto-union) expanded

24

CAPITAL & CLASS

rapidly and on 28 April 1973 the first trade union of black workers inNatal was formed in Pietermaritzburg by metal workers . Trade unions ofclothing, textile, chemical, furniture, and transport and general workersfollowed. In comparison with this rapid growth there was reportedlyrelatively slow advance made by unions associated with the Urban TrainingProject (an educational labour group), which concentrated on the devel-opment of the state system of works committees . At a comparativelyearly stage of development of trade unionism among black workers in theaftermath of mass strike action, certain clear tendencies between thevarious groups became evident and then more pronounced . It is possibleto distinguish between the following tendencies in the open trade unionmovement :

1 . The 'non-aligned' Urban Training Project tendency which includesthe Engineering and Allied Workers Union, Laundry and Dry CleaningUnion, Transport and Allied Workers Union, African Chemical WorkersUnion, and other unions which have developed subsequently . From theirinception these unions have been characterised by an explicitly 'non-political' stance and have sought accommodation within the works com-mittee system which is considered a useful basis for black trade unions .

2. The 'subordinate' union tendency of Johannesburg black tradeunions including the well publicized National Union of Clothing Workers,but also tobacco, leather, commercial, and banking workers . These unionshave been established under the direction of the registered trade unions inthese industries and their continued existence is totally dependent on sup-port from the parent union .

3. The 'mobilizing' Trade Union Advisory and Coordinating Council(TUACC) including the Natal unions mentioned and later a section ofworkers in the Transvaal province organised in the Industrial Aid Society .TUACC was initially developed in cooperation with the KwaZulu Govern-ment when Barney Dladla, the Minister of Community Affairs, gave sup-port to the strike movement . After vigorous protest from employers inNatal, Dladla was removed from this post, and the connection with thetrade union movement is now non-existent .

4 . The 'nationalistic' Black and Allied Workers Union which is theworkers' arm of the Black Peoples' Convention (now legally prohibited)and expresses black consciousness ideology and an ambiguous attitudetowards capital .

These various groupings are not only distinguished by strategy andpolicy but also by geography ; the non-aligned, subordinate, and national-isitic tendences are based in Johannesburg and the TUACC unions are theonly ones of significance in Natal. These tendences within open tradeunions are supplemented by the Western Province Workers Advice Bureauin Cape Town which favours the formation of works committees as pos-sibly democratic forms of worker representation and is critical of industrialunionism among black workers . Workers associated with SACTU are alsoregrouping underground to discuss and develop SACTU policy and draw inpolitically conscious workers to train them in the ideas and work of theworkers' movement. SACTU supports independent trade unions "in as faras they advance the workers' struggle ." Part of the work of the under-


25

ground groupings of SACTU is "to guide and influence, firmly but care-fully, the work of the open trade unions, etc, to which the workers belong,so that the errors of the union leaders and officials may be corrected bythe rank and file, and the organizations kept on the right course andstrengthened ." (Workers' Unity, 5, September 1977, p.6) The links whichexist between SACTU and open trade unions are not formal, given themassive repression surrounding the political organization of workers, butare decided underground .

Without specifying the membership of each trade union (by 31 August1975 there were 24), it is useful to classify the tendencies according tomembership :

Given the fierce recession of the economy and intensified repression, thesefigures reflect more or less the position in 1978 although the Black andAllied Workers Union has been destroyed by banning or detaining officials .According to figures tabulated by the South African Institute of RaceRelations there were 59,550 members of black trade unions in 1975, thelarge majority of the subordinate unions being made up of the NationalUnion of Clothing Workers membership of 23,000 workers . Trade unionmembership now compares with the situation in 1961 when there wereapproximately 52,800 members, an indication of the resilience of theblack working class (Hemson, 1973, Annexure B) . This development hasnot been homogenous as the differentiation of trade unions implies and itis appropriate to discuss the different strategies and policies behind themembership figures rather than the development of each union .

An evaluation of trade unionism among black workers necessarilyinvolves a discussion of their practice in the face of repression and agrowing working class movement . Since the mass strikes in 1973 therehave been many protracted strikes over union recognition and a number ofserious strikes which have involved black workers in confrontation with

Membership 1975 Number of UnionsNon-aligned

Urban Training Projectand unaffiliated unions 12,000 10with similar policies

SubordinateNational Union ofClothing Workers andsimilar parallel unions 29,120 8many affiliated to TUCSA

MobilizingTUACC and IAS,Johannesburg unions 15,620 5

NationalisticBlack and AlliedWorkers Union 2,700 1

26

CAPITAL & CLASS

the state . The mass strikes of 1976 and 1977 in protest against the shootingof students and workers in Soweto and in support of demands for therelease of all political prisoners have posed the question of the relation oftrade unionism to mass political resistance .

RESPONSES TO STATE POLICIES

Black trade unions have had a varied response to the reforms intro-duced by the 1973 legislation which provided for a more developedsystem of plant committees . Some favoured a 'stages of development oftrade unionism' approach whereby works committees would be used as`stepping stones' towards industrial organization, and others denouncedthe system as a threat to their existence. Differences between the 'non-aligned' and 'mobilizing' trade unions rapidly became evident in responseto the growing number of works and liaison committees stimulated bystate officials and employers .

The Urban Training Project until mid-1973 concentrated its activityalmost exclusively on the development of works committees, to the legaldefence of works committee representatives, and to the publication ofworkers' calenders calling on workers to form works committees . With theupsurge in working class action, particularly in Natal, the Project modifiedits policy to encourage the growth of black trade unions from establishedworks committees .

This 'stages in the growth of trade unionism' approach found its mostcomplete expression in Douwes Dekker's writings (particularly DouwesDekker, 1973) . In this pamphlet he suggests a combination of works com-mittees (whose constitutions could be modified to include recognition of atrade union) and trade unions of black workers, a view which even the sub-ordinate National Union of Clothing Workers found unsatisfactory . Thestrategy which he offered encourages a particularly limited form of tradeunionism which blends in with, rather than challenging, the state appara-tus. The strategy possibly found its most complete expression in the En-gineering and Allied Workers Union which claimed that black trade union-ism had become a part of the works committee system in the Transvaalengineering industry and had been used to create a working relationshipbetween management and the unregistered trade union . "Without a unionthese works committees will never work", the Secretary, Jane Hlongwanesaid at a meeting of the Progressive Party (12) . In this way, by educatingworkers to use works committees and absorbing them into unregisteredtrade unions, the organizers disturb neither management prerogatives norstate policy . The workers' dependence on substitutes for independenttrade unionism is strengthened .

A more radical approach is advocated by two differing tendencies :the Western Province Workers Advice Bureau and the mobilizing unions .The former urged the formation of works committees because of thepotential security the legislation provided for workers' leaders and as abasis for fundamental workers' education . The formation of trade unionswas rejected as it was felt that industrial trade unions tended to reinforce


27

existing divisions among workers, that they were inappropriate for themass of black labourers in Cape Town, and that they were institutions thattended towards bureaucracy. The mobilising unions, in contrast, decidedin October 1973 that works committees of all types were part of theapartheid labour system and should be opposed as such . The workers'representatives on works committees, it was argued, tended to become cutoff from the workers and to oppose trade union organization, whichthreatened their interests . The dependence of other tendencies and unionson the anti-victimization clause in the Bantu Labour Relations RegulationAct was seen as misplaced ; protection of union members depended basicallyon the unity of the workers backed by the Wage Act which protects unionmembers organizing in factories whether they belong to registered or un-registered unions (although the possibility of enforcement was consideredremote) .

The mobilizing unions which grew in the wake of the mass strikes inNatal have considerably more experience of strike action than othergroupings . Many union organizers are former shop stewards who have beenvictimized for union activities and for being at the centre of informalleadership in factories which had been decisive in strike action . Organizerssuch as these are particularly alive to the political nature of the struggle formass organization and said so . After the house arrest of four leading tradeunionists in Natal a shop steward of the Metal and Allied Workers Unionsaid that action by the state "symbolizes our strength and shows we arebecoming stronger and stronger". An organizer of the textile workersshowed his contempt for state authority and his belief in the futurepower of the workers : "Even though they (the Security Police) are standingthere at the gate we are not scared of them because they will have to cometo us and beg . . . if this government is trying to cow us I warn them : wewill drive them into the sea!" (13)

Given this level of consciousness and conception of organization it iswrong to assume that union organizers have had nothing to do with strikeaction by their members or even by unorganized workers . According to arecent review of trade unionism in South Africa the mobilizing unions"found themselves forced into the position of playing a subsidiary role tothe spontaneous action of the members themselves ." It is further arguedthat at factory after factory it was the workers themselves who took thedecision to strike and only contacted the union afterwards (Davis, 1976,p.97) This is a serious criticism of the mobilizing unions and (with theother criticisms which are made) is an attempt to deny these unions anycredibility. These criticisms are partly valid in the early phase of tradeunion organization in Natal during the mass strikes, when the benefit fundattempted to function as a trade union . Even here, however the textileunion was well aware of the growing resistance of the workers and hadsubmitted a lengthy statement of their demands to management shortlybefore the strikes in the textile industry broke out (IIE, 1974, p.30) . Anylags which subsequently occurred between strike action and union res-ponse have to be understood in terms of the perspective of strikes as auniversal phenomenon in capitalism . In discussing these allegations thefollowing points have to be made :

28

CAPITAL & CLASS

1 . That spontaneous strikes are the most authentic means of struggleunder the violent suppression of working class action ; black workers pre-fer strikes to appear and actually be spontaneous to avoid reprisals tothemselves, the factory leadership, and the trade union ; and

2 . That 'spontaneous' strikes (i .e . those directly organized by the shopfloor) are the norm in advanced capitalist countries . In the vast majorityof cases in the United Kingdom trade unions are not aware they havetaken place. Spontaneous strikes are an indication of the workers' strengthand are an immediate challenge to management ; waiting on unionofficials to declare strikes 'official' before coming out on strike wouldcertainly weaken the working class movement in the United Kingdom (14) .

These important considerations apart, it is ridiculous to deny that themobilizing unions have either been instrumental in organizing strikes orhave given their fullest support to strikes whether they have been strikes ofunion members or not . The unions have paid a heavy price for the highlevel of support which has been given to the strike movement . Between1 January and the end of August 1974 these unions supported strikeaction by some 12,520 workers in a variety of industries . These strikesranged from the large-scale strike in the cotton textile industry in Pine-town involving 8,000 black workers in January to the 300 black workerswho struck work at Hypack in July . In these and subsequent disputes infactories such as Heinemann, where workers demanded recognition of theMetal and Allied Workers Union in 1975 and were repressed with a batoncharge, the charge that the mobilising unions were reduced to processingindustrial complaints (Daivs, 1976, p.97) is shown to be completelyfallacious .

Union organization in these struggles was significant in raising theissues from the immediate concern over wages and conditions to demandsfor the destruction of all state committees within the factories whichhindered union development . The unions were crucial in regrouping thestrikers after they had been terrorized and dispersed by police dogs andarmed police in the cotton strike in J anuary 1974 and eventually a qualifiedvictory in the form of higher wages was achieved . Following this strike, thefour trade unionists involved in varying degrees in the organization of tex-tile workers were house arrested (15) . In the subsequent strikes the unionswere well aware of the issues leading up to the strike action, providedwhat support was possible before and during the strike, and legal defencefor those strikers who were prosecuted . A high proportion of those accusedof striking illegally have been successfully defended by lawyers paid bythe mobilizing unions. The question of whether trade unions should leadthe workers into strike action (which is implied by 'left' criticism) is amatter of serious debate; it would certainly be adventurist for open organ-izations to call strikes during a time of fierce repression and mass un-employment. The union organizers of the mobilziing unions have not seenit their role to try to initiate strike action from 'the office' (many strikesare lost with severe repercussions for the workers) ; this has to be decidedupon by the workers themselves in consultation with the union officials .The strategy, timing, and organization of the strike obviously has to takeplace underground to protect the actual leadership . While the strikes


29

which have been supported and led by the mobilizing unions are notexpressly political, in all cases they have either challenged the prevailingstate wage control system, the 'Bantu labour relations institutions', orinvolved a demand for the recognition of independent trade unions orblack workers . They have all involved a defiance of the security legislation(which defines strike . action as a threat to the security of the state) and theBantu labour relations system which provides for complicated proceduresfor legal strike action .

Despite a higher level of repression of strike action in a climate ofeconomic crisis accompanied by a decline in production and persistentinflation, the number of black workers on strike remains high in comparisonto the 1960s . In 1974, the year in which the most rapid expansion ofblack trade union membership took place, there were even more strikesthan in 1973 although the number of workers involved declined to 58,975 .In contrast to the mass strikes of 1973, which were overwhelmingly suc-cessful in achieving higher wages, the industrial and factory strikes whichhave followed, given the changed economic and political conditions,have less chance of success. The strength of the working class movement isshown by the fact that industrial strike action persists despite the greaterchance of failure and retaliation, and that political strikes have carried themovement into direct challenge to the state. The increasing politicalactivity of the black working class has been accompanied by an emphasisin industrial strike action on demands relating to working conditions andunion recognition. Despite a decline in industrial action by the blackworking class considered as a whole, the level of strike action by unionmembers has remained at a high level . Unorganized workers on strike haveincreasingly sought support and protection from the mobilizing unions asdefenders of the black working class.

MASS RESISTANCE AND TRADE UNIONISM

The discussion of the organization of the black working class has, upto the present, centred on the black workers in industry and in certainkey industries in particular : garment, textile, and metal . The crucialmining sector in South Africa (gold, coal, and base minerals) is sealedoff both geographically, as the mines are usually situated outside theurban centres, and physically, in compounds which are segregated fromthe wider currents of the working class movement . The overwhelmingmajority of the workers, until recently, have been foreign, recruited fromthe 'hostage' dependent territories surrounding South Africa . Despite thefactors of migrant contract labour, brutal discipline underground and con-trol off shift, and a sophisticated system of exploitation, the mines havebeen wracked by some of the most explosive confrontations betweenworkers and capital . In a series of strikes and uprisings, workers havedemanded higher wages, an end to the compound system, and in manycases immediate repatriation. It is estimated that from the beginning ofthe mine uprisings (at Western Deep Levels on 11 Septebmer 1973) up toSeptember 1975 some 50,000 black workers broke contract with themining companies, a 'crime' in terms of the penal labour code .

30

CAPITAL & CLASS

Although the policy of the mining companies and the apartheid statehas been to seal off black mine workers from communication with thewider social and political movements in South Africa, the growing strengthof the mine workers' action predated to a considerable extent the devel-opment of mass resistance to apartheid in the black townships . The char-acter of the uprisings, many of which have involved a direct attack on theoffices of management, destruction of documents, and combat with thepolice, has demonstrated the capacity of the black working class for in-surrectionary action . The workers' resistance has been met with violentretaliation : between 1972 and June 1976 it was estimated that 178workers had been killed and 1,043 injured (Horner and Kooy, 1976) .Because of the developed system of ethnic groupings in compounds someof this resistance to capital was diverted from attacks on the whole systemof control and exploitation into conflict between different nationalitiesand ethnic groups . Despite this factor, a committee appointed to investi-gate mine uprisings concluded that the workers are well aware of howvulnerable the mining industry is to strike action and are strongly at-tracted to socialist ideology (Report of Inter-Departmental Committee,n .d .)

Growing working class resistance on the mines and the violent sup-pression which was the response of mining capital and the state had animportant effect on trade union struggles in industry. These changes canbe related to the crisis of accumulation in South Africa characterized bymassive unemployment among blacks which was estimated at 2 million bySimkins (June 1976), declining demand, a negative real growth in grossdomestic product, and increased inflation . In this economic and politicalcrisis, the pressures on black workers and their trade unions have inten-sified sufficiently to cast doubt on the view that independent trade union-ism was the object of official 'repressive tolerance' . A much tougherofficial attitude has been taken to strike action, most strikes have beenunsuccessful, and at the same time the unions and their members havebeen raising important structural questions: those of management of thefactory floor, unqualified union recognition, and challenges to redund-ancies. Union support for strike action has deepened, and the strikeshave become more protracted despite the discipline of mass unemploy-ment and harsher repression .

Far from the union of black workers becoming more detached fromthe spontaneous working class movement, the most recent period hasseen bitter struggles for trade union recognition . Workers who struck atNatal Cotton and Woolen Mills in Durban in support of their demandsfor the removal of the personnel manager (who is closely connected tothe secret police) were all dismissed after a strike lasting 12 days . Thestrike was led by union members in the factory, had the full support ofthe union officials, and could only be sustained by this organizationalbacking. During this strike some 7,200 striker-days were recorded, a figurealmost 40 per cent of the total striker-days in 1975 . In the Transvaalworkers at Heinemann Electric who signed a petition stating : "We wantthe union (Metal and Allied Workers Union) to represent us and not aworks or liaison committee" were baton-charged on 29 March 1976 andmany injured . This onslaught on the militant black working class organized


31

in the metal union made a tremendous impact on working class conscious-ness and was described as a 'mini-Sharpeville' . The workers who reappliedfor employment after the mass dismissal were forced to fill in a question-naire stating whether they were union members and whether they accepteda liaison committee . Union organizers who were injured in the batoncharge were found guilty of inciting the strike . In his judgement themagistrate said that organizing an unregistered trade union was inherentlya political act given the existence of the Bantu labour relations system .These struggles were hardly an indication of the mobilizing unions reachingthe dead end of the road to economism (16) .

Even the non-aligned trade unions of black workers in the Transvaalhave found their struggles within the law for trade union recognition andagainst redundancy raising decisive issues of struggle. Despite the pains-taking legality of organized strike action against Pilkington's ArmourplateSafety Glass in the Transvaal which started on 6 September 1976, allnegotiations with the union were refused by management and policeaction was taken against picketers . Pilkington management called in thepolice to deal with the strikers, and arbitration through the Departmentof Labour was turned down (17) . Pickets were organized which called onreplacements to join the struggle, and the strike demonstrated a cautiousbut determined resolve to combat management prerogatives and thepolice. Workers on the Rand raised funds to pay out strike pay, and thestrike turned into the most prolonged strike by black workers for a longtime ; the 205 workers held out for an almost unprecendented 10 weeks,causing some 11,070 striker-days lost time to the company .

This increased combativeness among organized workers has to beseen in the context of the growing political resistance to apartheid through-out South Africa which has been brought into motion through mass workingclass struggles and is now taking the form of protracted resistance andarmed action by the underground . The use of the stay-at-home weapon hasincreased beyond all previous experience ; there were three massive generalstrikes before the end of 1976 - the first involving an estimated 100,000black workers, the second 132,000, and in September some 500,000African and Coloured workers in Johannesburg and Cape Town (18) .This tremendous outburst of activity by the black working class raises thewhole question of the relationship between wage demands, `structural'demands (such as the abolition of liaison committees and trade unionrecognition), and mass political action . It also raises the question of therelationship between organized workers, their trade unions and massstruggle .

There is undoubtedly a connection between the increased combativityof organized black workers, violent resistance on the mines, and the un-precedented support given to the massive stay-at-homes, themselvesmore prolonged than in previous South African experience (the stay-at-homes in August and September lasted three days, many of those in the1950s were for shorter periods) . There is definite evidence that organ-ized workers gave greater support to the stay-at-homes than other workers .(Organized workers obviously are not limited to those in trade unionmembership, factory and residential groups may also be developed under-

32

CAPITAL & CLASS

ground by politically advanced workers . It is very difficult to assess theeffect of the latter groups, and the question here is of the position oftrade unions as open and public organizations in relation to mass politicalaction) .

Despite appeals from trade union leaders in the garment industry totheir members not to damage the industry by withdrawing their labour,for instance, workers in this industry are specifically mentioned as bringingthe industry to a halt in the August stay-at-home . The garment industry inthe Cape, in which the workers are represented by a reformist registeredtrade union, also took the lead in the September stay-at-home (19) . Sincethe garment industry in the Transvaal has almost 100 per cent membershipof African workers in the subordinate National Union of Clothing Workers,the full support given to the stay-at-home can only be explained in termsof the independent activity of the 300 African (mainly women) shopstewards . African women, who are most vulnerable to racial legislation andare also subject to discriminatory legislation which classifies them as per-petual wards, have an advanced political consciousness . The same argu-ment about the independent movement of shop floor leadership applies tothe registered trade union in the Cape garment industry which representsvirtually all Coloured workers (again predominantly women) . This pointraises the issue of the effectiveness of reformist control over organizedworkers during periods of mass working class activity. Research wouldprobably reveal that class conscious workers in black trade unions took aleading role in getting decision by factories to join the stay-at-homes.

CONCLUSIONS

While liberal social science has argued that strikes in South Africa area 'blunt weapon' in the struggle against apartheid, the evidence has shownthat mass strikes have at times paralysed state repression and have shownthe potential of developing into insurrections . These strikes are not thecomplete expression of resistance to apartheid (the problem of devel-oping mass struggles together with armed struggle has been faced by theunderground since 1961) but no decisive challenge to the state is possiblewithout this proletarian form of action to disorganize production and thestate. Industrial and political strikes have taken place despite the massiveindustrial reserve army marshalled by the recruiting agencies and thelabour bureaux, testifying to the relative development of the black workingclass and the growing coherence of the political and economic aspirationsof black workers .

The mining industry which has been based on foreign contract labour(and workers' resistance is weakened by the mass of unemployed wor-kers in 'hostage' territories) has experienced the most determined andviolent resistance in strikes and uprisings. Despite the disorganizing effectsof contract labour and the imposed Bantustans, migrant workers have inmany cases taken a leading role (in the 1973 mass strikes) or joined in thepolitical resistance (despite points of disjuncture caused by state inter-vention) . The strikes and uprisings of migrant workers on the mines


33

indicate that not even the most severe restrictions on movement and work,as in the case of contract workers housed in compawnds, can restrain thedevelopment of class demands . With the growth of thonopoly capitalismand the concentration of production in large-scale, highly mechanized,factories and 'industrial mines', basic production is carried out by a mas-sified black proletariat neither differentiated by traditional skills nor havingexperienced the benefits of reform. These are the conditions for a rapidadvance in class consciousness as the political resistance to apartheid gainsmomentum .

The independent trade unions have arisen from these conditions ofstruggle although they by no means reflect the totality of working classresistance, for example, struggles in mining . But these unions have de-fended and advanced the position of workers involved in crucial struggleswhich are an essential component of the overall working class movement .Independent trade unions have deepened the struggles of the blackworking class by raising working class issues and 'structural' demands(such as the struggle for union recognition, against shop steward victim-ization, in defence of strikers, and for the permanent organization ofworkers) . Through organization within trade unions, black workers haveexposed the role of foreign capital in South Africa (particularly in thefight for union recognition at Leyland, in Glacier Bearing, and in the tex-tile company of Smith and Nephew ; in strike action at Pilkingtons, and inthe strike for the reinstatement of victimized shop stewards at US-ownedHeinemann Electric) . Foreign capital, which claims a liberalizing functionin South Africa, is shown to rely on state and managerial repression tosecure high profits in South Africa . These struggles have opened up newareas for international solidarity with the fight for elementary politicaland trade union rights in South Africa .

By linking workers in a wide variety of factories in industrial unionsthe independent trade unions are developing an important level of blackworking class unity ; most notably the distinction between migrant andurban workers which has come to the fore in recent political struggles, hasnot been reproduced within the trade unions . Independent trade unionsare carrying out important defensive functions, basically in defence of theinitiative and action of the black working class, through provision of legaldefence for their members and other workers . Defence of advanced workersis essential during a time of economic crisis in which workers are morevulnerable to the dictatorship of management and prosecuted morefrequently for striking . In 1975 the police intervened in 61 cases of strikeaction and 503 workers were prosecuted for striking illegally, for 'publicviolence', 'illegal strike and continuation of strike', 'refusing to obeylawful command of employer', 'malicious injury to property', 'breach ofcontract', 'arson', 'incitement', and 'intimidation' (20). Increasing num-bers are being defended (in many cases successfully) by independent tradeunions .

Despite the extent of repression of black trade unions in SouthAfrica, this alone does not make them uniquely centres of working classaction in contrast to the reformism of trade unionism in advanced capit-alist countries. But this is not to argue that independent trade unionism

34

CAPITAL & CLASS

does not raise important political questions from a working class view-point which are missing from a nationalistic perception of struggle againstracial oppression . Within the country the demands of the SACTU under-ground express a developed program of trade union demands, linked to theoverall demands contained within the Freedom Charter (which includefundamental working class aspirations such as a national health service andthe nationalization of monopoly industry, mines and the banks) .

The political nature of the open, independent, trade unions is notsomething that can be determined apart from an examination of theirtendencies which are subjected to the most powerful pressures to take a'non-political' stance. Reformism in black trade unions is not so much areflection of the personality of the leadership (as it is often presented), asmuch as a response to massive repression by employers and the state andthe international forces pressing on union leadership : such as the materialsupport given by the British TUC to a conception of non-political, 'bread-and-butter', trade unionism . It is debatable whether the ideology of tradeunionism in the 'unionism only' form is the dominant ideology of unionmembers, much less that of the black working class as a whole .

Bourgeois ideology in the working class cannot be generated ab-stractly, but only materially through trade union recognition, the growthof collective bargaining, substantial increases in wages and decisive miti-gation of the oppressive conditions of wage labour under apartheid . Yet itis precisely in these areas that the state finds it difficult to make conces-sions without enabling the advanced forces within the black working classto seize these opportunities as a platform to demand more fundamentalconcessions . Against the tendencies towards reformism within independenttrade unions has to be set the working class movement whose direction andenergy has been so clearly evident both in the 1973 mass strikes and in thepolitical strikes of 1976 and 1977 . These mass movements have generatedthe support which does exist for independent trade unionism in the fac-tories, have limited the tendency towards reformism, and have providedthe bases for a revolutionary party . It is in this context that reformistleadership within the subordinate and non-aligned unions has been discip-lined, and in one case a union secretary dismissed, for not carrying out thepolicies of the membership .

Various tendencies within the independent trade union movement areproviding a direct challenge to the Bantu labour relations system and areleading struggles against management dominated factory councils and fortrade union recognition . There are now more than half a million blackworkers covered by these official factory systems which are designed toeliminate the demand for independent trade unionism among blackworkers. As management in South Africa, particularly management ofmultinationals, is using the works and liaison committee system to isolateblack workers from wider industrial and political movements (Legassickand Hemson, 1976), the struggle against the state forms of factoryorganization becomes ever increasingly important . The inaugurationof the Wiehahn commission of enquiry into industrial legislationpresages the state's offensive against independent forms of organizationand the further subordination of trade unionism among black workers to


35

factory councils and registered trade unions . It seems likely that the Com-mission will recommend that the activity of independent trade unions behighy restricted or eliminated altogether and something approximating toan official company union be recognized as a collective bargaining unit ateach factory .

For the independent trade unions to survive and grow under a moresophisticated form of repression, they have to be broad organizations ableto unify advanced organized workers and the mass of unorganized workersby taking up economic demands and relating them to the system of op-pression in South Africa . Lenin argued that trade unions should be organ-ized by industry, that they should be broad organizations, as open as pos-sible to be able to mobilize workers effectively . The aim of the unions isdirectly connected to their mass character : their object would be unattain-able if they failed to unite all who understand the necessity of strugglingagainst employers and the state . It is for this reason that Lenin argued thatcommunists should support non-party unions, and distinguish the partyfrom the trade union even in conditions of extreme repression .

By arguing that the most advanced black trade unions are militantlyeconomistic, Davis falls into the trap of failing to distinguish betweenopen and underground organization, and fails to acknowledge the tacticsadopted by black militants in working class struggles . He fails to recog-nise the distinction between the organization of workers and the organ-ization of revolutionaries, between the necessarily all-embracing and opentrade unions taking the issues as far as possible within the context of sup-pression of working class action, and the underground revolutionary partywhich has to be as compact and secret as possible, taking up each andevery issue in the struggle against apartheid . It is the revolutionary partywhich can make the linkage between the particular forms of oppression intrade union struggles and all forms of struggle against apartheid, and thedemand for a socialist society to end all forms of oppression .

While the task of the open trade union is to advance the most generaldemands of the workers as far as it is able without destroying the basesfor its existence and by defending all the actions of union members andother workers, the task of the underground party is to transform the"trade union strike movement into a political and direct revolutionarystruggle of the masses" (Lenin, CW 13, p .61) . By attacking every tradeunion tendency within the country as politically underdeveloped, Davis isdiscrediting all open forms of working class organization which blackworkers have struggled to develop since 1972, including those with whichSACTU workers have been associated . His arguments tend towards a re-jection of work in existing trade unions but fail to develop any alternativeconception of revolutionary work within working class organization . Byattempting to distance himself from trade union struggles without showinghow the work of underground organization could otherwise be developedDavis misunderstands the importance of trade union organization to theblack working class .

Lenin describes trade unions as a "tremendous step forward for theworking class" and argues that the proletariat cannot develop withouttrade union organization (CW 39, pp .SO-51) . The work of the trade unions

36

CAPITAL & CLASS

is to educate and give all-round development and training to workers, andwhen workers are becoming organized Lenin argues it is fundamentallyincorrect "even criminal" to be "critical" and disparaging . "The task devol-ving on Communists is to convince the backward elements, to work amongthem, and not to fence themselves off from them with artificial and child-ishly `Left' slogans" (CW 39, p.54) . When revolutionaries are forcedthrough repression not to declare their intentions openly, work in openand legal (even "downright reactionary") organizations is absolutelynecessary . "Inexperienced revolutionaries often think that legal methodsof struggle are opportunist because, in this field, the bourgeoisie has mostfrequently deceived and duped the workers . . . while illegal methods ofstruggle are revolutionary" (CW 39, pp .96-97) . The task of revolutionariesis not to adopt one form of struggle and to reject the other, but to com-bine illegal forms of struggle with every form of legal struggle . It is insuf-ficient to build a satisfactory critique of trade unions of black workerswithout posing the real problems of working class organization : thesupport for working class struggles, development of underground tradeunion organization, and defence of the advances which have been made .While the trade unions are not the vanguard organizations of the resistanceto apartheid (from which they are excluded by their vulnerability as openorganizations) because of this they are in no way an impediment to thegrowth of a revolutionary party . Both the party and the trade union areintegral to the development of a revolutionary working class movement .

The question of the relation between the trade unions and the revol-utionary party cannot be reduced to mechanistic and inappropriate form-ulae such as the trade union being a transmission belt to the party, anauxiliary body linking the party to the class . (Stalin, 1940, p .78) Theparty has to be based immediately and vitally within the working classthrough the development of revolutionary theory and practice . The tact-ical questions of the organizational links between a revolutionary partygrounded in the black working class and the open trade unions can only bedecided underground where the relative significance of open organizationin the trade unions (which Lenin endorsed) can be assessed in terms of thenecessity for determined illegal political struggle by the black workingclass . The danger is that the very intensity of repression dissolves all nec-essary distinctions between the organization of workers and the organ-ization of revolutionaries, leading both to ineffective mass organization,careless underground work, and the subordination of essential industrialorganization relating mass working class action to other political work .

The struggle to seize state power through armed struggle does notrelegate the concrete struggles of the working class to insignificance . Leninargued that the party should assist workers in giving precise and definiteexpression to demands, in promoting the organization of workers, and inexplaining the real aims of the struggle . (CW 2, pp.1 14-116) This work isdoubly important when the overwhelming mass of the black proletariat,particularly on mines and in plantations, is unorganized . The consciousand devoted participation of the party in the daily struggles and disputesof the exploited against the exploiters, was seen by the Third Congress ofthe Comintern in the theses on the method of work of communist parties,


37

as a prerequisite for the seizure of power . "it is the greatest mistake forcommunists to remain passive and disdainful or even hostile to the presentstruggles of the workers for small improvements in their working conditionsby appealing to the communist programme and the final revolutionaryarmed struggle" (Degras, I, p.263) . The reformist leadership of the sub-ordinate trade unions cannot be transformed through institutional re-organization, as suggested by Davis, in forming a single black union fed-eration functioning under the overall umbrella of SACTU (Davis, 1976,p.103), which is a blandly utopian suggestion, but through the devel-opment of a revolutionary base in the working class which supportsthese unions . Abstract formulations which do not include the dialecticalrelationship between working class struggles, the formation of open tradeunions, and the development of a revolutionary party, tend only to leadto disorganization and despondency .

Since trade unions of black workers have not been brought within theambit of the developed system of industrial relations in South Africa, theyare struggling to survive, to protect their members, and engaging in strugglesfor recognition . The appointment of two government Commissions ofEnquiry into labour legislation points to the crisis which is being faced inintegrating the black working class into the subordinate structures of in-dustrial relations in South Africa . The crisis of accumulation, marked bydeclining rates of profit, intense resistance by the black working class, andmassive foreign indebtedness, leaves less room for the apartheid state tomanouvre . Bourgeois reforms which attempt to stem the rising tide ofresistance come too late and only accelerate demands for more funda-mental concessions .

Faced with this situation the state is attempting to introduce minimalreforms in the wake of massive repression, hoping to neutralise workingclass action by eliminating revolutionary leadership . The process of en-couraging reformism by waves of repression is graphically described in arecent issues of Workers Unity, the organ of SACTU, which argues thattrade unions are forced to take up a public posture of being non-politicalto avoid being smashed .

Within them and among their leaders, various tendencies are to befound. There are, of course, not a few reformists, opportunists, andeven collaborators - but there are also many who walk a tightrope ofpersonal danger in truly serving the struggle of the working class . . .These organizations are forced by the repression to keep themselvescut off from the liberation struggle as a whole, but we do not opposethem. Our policy is to fight for independent unions and to give thesenew organizations our support - in as far as they advance the workers'struggle . (No 5, September 1977)

This even-handed formulation avoids claiming that the demands for tradeunion rights are inherently revolutionary because of the state of nationaloppression of black workers, or the contrary position of dismissing alllevels of working class struggle as irrelevant to armed national struggle . Ithas been argued that trade unionism is not a 'predetermined phenomenon'

38

CAPITAL & CLASS

outside of the strength and will of the members to impress a policy and anaim which defines it, and the growth of black trade unions in South Africais far from reproducing an ideology of trade unionism in the black workingclass .

A crucial aspect of the political and ideological crisis in South Africais the relative underdevelopment of social reformism . The fact that blackworkers do not clearly distinguish between employers and the governmentcreates difficulties in the development of more sophisticated bourgeoisideology and hampers the development of economism. The capital relationin South Africa exists in a weak form because it has to be constantly re-inforced by the state to an unparalleled degree . Given the well developednature of capitalism in South Africa, and the dependence of capital on theapartheid state for high rates of exploitation, the struggle against apartheidhas to raise the question of a socialist society in South Africa .

FOOTNOTES

Daivd Hemson, research officer of the Textile Workers IndustrialUnion in South Africa before being banned in 1974, is a postgraduatestudent at Warwick University in the Department of Sociology .

1 The tenuous nature of the formal industrial rights was shown duringthe struggle of white mine workers for a five day week . The threat tostrike was met by a proposal to amend industrial legislation to pro-hibit all strikes in the mining industry . Financial Mail, 22 April 1977.

2 Section 18 (5) of the Native Labour (Settlement of Disputes) Act48/1953 which has now been amended under the title : the BantuLabour Relations Regulation Act . The definition of strike action isaltered only to the extent that in a highly restricted context, with duenotice, strikes are not explicitly prohibited . These strikes do, however,remain prohibited in terms of security legislation . For a description ofthe amended legislation see Horner, May 1976 .

3 The latest agreement for the furniture industry restricts artisan andcertain other categories to 'members of registered trade unions', inthis case Coloured and Indian workers . Similar disguised job reservationapplies in the metal and other industries . African workers are not per-mitted to belong to registered trade unions . The crisis in South Africahas brought employers to demand massive job fragmentation (part-icularly in the metal industry) and the removal of all 'restrictive prac-tices' on the free deployment of black labour .

4 State vs Prefabricated Housing Corporation (Pty) Ltd, 1974 (1) SouthAfrica, 535, Appelate Division .

5 During periods of low levels of working class action, this leadership isoften dormant unless brought into direct contact with revolutionaryorganization. Working class leadership exists in groups in factorieswhich attempt to control production e .g. piece rates, and build upsolidarity e .g . make collections for the families of deceased workers .These groups were 'discovered' by bourgeois social scientists


39

in America with the Hawthorne experiments, and the 'human relat-tions' approach (now the position of avant garde management inSouth Africa) is an attempt to penetrate groups of workers withmanagerial ideology . The stress of all communist parties of the ThirdInternational which distinguished them fundamentally from bour-geois parties was the attempt to build up revolutionary groups inproduction as the basis of a political proletarian movement .

6

Statistical Year Book, 1964, H-25 .7 Figures calculated from Horrell, 1969 : 73 and from Statistical Year

Book, 1964, H-43. Figures of black workers prosecuted should behigher as statistics on prosecutions in 1956 are not available, only thenumber of convictions .

8

Pillay, P.N. 1973. A poverty datum line study among Africans inDurban . Department of Economics : University of Natal, Durban, p .22 .

9 Report of Department of Labour, 1973 . RP 33/1975 .10 House of Assembly Debates, 1973 . No 11, Cols 687-690 .11 People and Profits, August 1973 : 14 .12 'Black trade unionism part of system', Daily News, 7 July 1975 .13 Meeting of workers held 9 February 1974 to protest banning of four

trade unionists in Durban.14 For a discussion of spontaneity of strike action see Richard Hyman,

1972, Strikes, Fontana, pp .41-43 and for the response of trade unionsto strike action the Donovan Commission . In comparison with manyBritish trade unions, black trade unions particularly the mobilizinggroup, are highly activist .

15 On this important strike see Sechaba, April 1974, pp .2-4 .16 See report in Star, 1 November 1975, and Financial Mail, 14 November

1975 : 646. There were 18,720 striker-days in 1975 (Bulletin ofStatistics, September 1976, 2 .43) . Pamphlets distributed by studentsin Soweto mentioned the Heinemann strike as a reason for workerscoming out on strike during the August stay-at-home .

17 Financial Mail, 8 October 1976 .18 Figures calculated from percentages of workers in Johannesburg

participating given in Anti-Apartheid News, October 1976 and JohnGaetsewe's speech in 1976 to AAM labour movement conference .

19 Daily Telegraph, 24 August 1976 reported that some industries'notably clothing' came to a complete halt .

20 Hansard, 11 February 1976, Minister of Police .

BIBLIOGRAPHY

Boulanger, M ., 1974, "Black Workers and Strikes in South Africa", Race,xv, 3, pp.351-9 .

Braverman, R.E ., 1974, "The African Working Class : Recent Changes,New Prospects", African Communist, 59, pp.48-60 .

Clack, Garfield, 1963, "Industrial Peace in South Africa", British Journalof Industrial Relations, 1, 1, pp.94-106 .

40

CAPITAL & CLASS

Davies, Rob, 1976, "The Class Character of South Africa's IndustrialConciliation Legislation", SouthAfrican Labour Bulletin, 2, 6, pp .6-20 .

Davis, David, 1976, "African Unions at the Crossroads", African Com-munist, 64, pp .93-104 .

Degras, J ., 1966, The Third Communist International 7919-7943 . OUP,London .

Douwes Dekker, L . et al, 1975, "Case Studies in African Labour Action inSouth Africa and Namibia (South West Africa) ", in Richard Sandbrookand Richard Cohen, The Development of an African Working Class,Longman, London pp.205-238 .

Douwes Dekker, L ., 1973, Are Works Committees Trade Unions? SouthAfrican Institute of Race Relations, Johannesburg .

Feit, Edward, 1975, Workers Without Weapons: The South African Con-gress of Trade Unions and the Organisation of the African Workers,Archon, Hamden .

Fisher, Fozia, 1977, "Class Consciousness among Colonized Workers inSouth Africa", in Perspectives on South Africa, African StudiesInstitute, University of Witwatersrand, Johannesburg, pp.300-353 .

Gann, Lewis H ., 1973, "Southern Africa - No Hope for Violent Revol-ution", South African International III, 3, pp .147-158 .

Gervasi, Sean, 1970, Industrialisation, Foreign Capital, and Forced Labourin South Africa, United Nations, New York .

Glass, Y ., 1973, "Industrialisation of an Indigenous People", SouthAfrican journal of Science, 59, pp .386-94 .

Hemson, Daivd, 1973, Black Strikes, Prices and Trade Union Organisation,1939-73, mimeo .

Hemson, David, 1976, Dock Workers, Labour Circulation and C/assStruggles in Durban, 1940-1959 . Paper presented to the Institute ofCommonwealth Studies .

Hepple, Alex, 1971 . South Africa : Workers under Apartheid, Defence andAid, London .

Horner, Dudley, 1976, African Labour Representation, SALDRUWorking Paper 3 .

Horner, Dudley, and Kooy, Alide, 1976, Conflict on South AfricanMines 1973-1976, SALDRU Working Paper 5 .

Horrell, Muriel, 1969, South Africa's Workers: Their Organisation and thePatterns of Employment, South African Institue of Race Relations,Johannesburg .

International Labour Office, 1966, Apartheid in Labour Matters, ILO,Geneva .

International Labour Office, 1965-, Speical Reports of the Director-General on the Application of the Declaration Concerning the Policyof Apartheid of the Republic of South Africa, ILO, Geneva .

Institute of Industrial Education, 1974, The Durban Strikes - 1973, IIE,Durban .

Jordaan, Ken, 1974, "Trade Unionism versus Revolution in South Africa",Race Today, 6, 3, pp .76-80 .

Legassick, Martin and Hemson, David, 1976, Foreign Investment and theReproduction of Racial Capitalism in South Africa Anti-Apartheid,London .


41

Levy, Leon, 1961, "African Trade Unionism in South Africa", AfricaSouth in Exile, 5, 3, pp .32-43 .

Luxemburg, Rosa, 1925, The Mass Strike, The Political Party, and TheTrade Unions, Merlin, London .

Mhlongo, Sam, 1964, "Black Workers' Strikes in Southern Africa", NewLeft Review, 83, pp.41-49 .

Mugabane, B., 1975, "The Continuing Class Struggle in South Africa",Studies in Race and Nations, Denver, 6, 3/4 .

Mthethwa, Alpheus and Mfeti, Pindile, 1975, "Report on Leyland MotorCorporation and the Metal and Allied Workers Union", South AfricanLabour Bulletin, 2, 5, pp .36-47 .

O'Meara, Dan, 1975, "The 1946 Mine Workers' Strike and the PoliticalEconomy of South Africa", journal of Commonwealth and Com-parative Politics, XI 11, 2, pp. 146-173 .

Petryszak, Nicholas, 1976, "The Dynamics of Acquiescence in SouthAfrica", African Affairs, 75, 301, pp.444-462 .

Report of the Interdepartmental Committee of Inquiry into Riots on Minesin the Republic of South Africa, n.d ., (1975?) .

Ruskin Students' Association, 1975, Ruskin College and the Institution ofIndustrial Education, Mimeo .

Santo, Bella, 1930, "Problems of Illegal Trade Union Activities", The RedInternational of Labour Unions, 2, 2 and 3, pp .28-33 .

Simkins, Charles, 1976, Employment, Unemployment and Growth inSouth Africa 1961-1979. SALDRU Working Paper 4 .

South African Congress of Trade Unions, 1976a, Babebetsi Mekoting:Mine Workers' Conditions in South Africa, SAC I U, London .

South African Congress of Trade Unions, 1976b, Workers in Chains,SACTU, London .

Stalin, J ., 1940, Leninism, Allen and Unwin, London .Trade Union Congress, 1973, Trade Unionism in South Africa : Report of

a Delegation from the TUC, TUC, London .Various, 1975, "Women Workers in South African Industry", South African

Labour Bulletin, 2, 4 .Woddis, Jack, 1960, The Roots of Revolt, Lawrence and Wishart, London .Wojtkiewicz, M ., 1930, "Problems of the Illegal Trade Union Movement",

The Red International of Labour Unions, 2, 1, pp .542-547 .

LABOUR POWER AND THE STATE

AbooT. Aumeeruddy Bruno Lautierand Ramon G. Tortajada

The Marxist theory of the state has developed in three main directions : thestudy of the relation between the bourgeoisie and the state ; the study ofthe functioning of the state ; and the study of its role in the valorisation ofcapital . However, it seems to us that the essential question is that of thefoundation of the state under capitalism . This foundation must be soughtin the wage-relation, whose reproduction is never given a priori. Hence thenecessity for a study of the relation between the state and labour-power .

INTRODUCTION

The debate on the nature and function of the state under capitalismtakes two forms. The first type of debate is concerned mainly with the'intervention' of the state in the field of the production and circulation ofcommodities. More precisely, it is a debate about the way in which, oncethe social conditions for the production of value have been achieved, thestate intervenes in the way this value is divided up. This activity affects thedivision between wages and profits as much as the division of profit be-tween fractions of capital . This activity implies, among other things, theregulation of money .

The second type of debate is about the very nature of the state, andabout the links it maintains with the class structure of society ; in particular,it is concerned with whether or not the state establishes the central socialrelation of capitalism, the wage-relation .

We do not deny the importance of the first type of debate . On thecontrary, in our view 'economic policy' is more than just a series ofmeasures intended to restore the rate of profit or to ensure the enforce-ment of the rules of competition . The state is distinct from a collectiveorganisation of capitalists not only in the sense that it alone has the power

LABOUR-POWER AND THE STATE

43

to make the collective interests of capitalists prevail in the face of theirindividual interests, but also in the sense that it alone is able to maintainthe existence of a kind of production that rests on private and mobilecapitals .

But we must note that the emphasis put on 'economic policy' overseveral decades has helped to conceal the second type of debate . The latterhas only emerged historically in periods of revolutionary crisis . Marxopened it after 1848, then after 1871 . Lenin, Rosa Luxembourg, Panne-koek, Gramsci started it again before and after 1917 . Then it got lostin the fog. Even as the European Communist Parties incessantly repeatthat capitalism is entering an unprecendented crisis, the debate over thenature of the state is conjured away . For example, one of Marx's principalconclusions on this question, the idea that one cannot break with capit-alism without the 'dictatorship of the proletariat' was abandoned almostwithout open debate (apart from the statements of Etienne Balibar avail-able outside the Party) by the French Communist Party at its XXI I Congress .

The contemporary emphasis on the 'economic intervention' of thestate therefore contrasts sharply with historical studies analysing theperiod in which capitalism emerged, the period known as that of 'primitiveaccumulation'. While Marxist historians concede that in this period thestate played a fundamental role in the constitution of the proletariat, thatrole disappears as soon as one gets beyond this 'stage' . The relation be-tween social classes seems to reproduce itself on its own or, at least, by theaction of 'economic' compulsion alone .

A few works, in recent years, have tried to reopen the debate on thenature of the state, and these have immediately put into question the pol-itical strategy of the Parties which have aimed to take control of the state .

Among those authors who have contributed most to reopening thedebate there seems to be a consensus, beyond their differences, on oneessential point about the role of the state :

"The state is an instrument of domination at the service of the dom-inant class ." (1)

But in most of these works this role of the state, as well as the analysisof its violence, whether overt or unobtrusive violence, is treated as self-evident. That is to say that once these obvious points have been explicitlyrecognised, this role of the state is not further considered in the develop-ments that follow .

The obvious thus functions to obscure . Its role is to dismiss that fund-amental, open, violence that is the attribute of the state, to refer it else-where, to something unsaid, something implicitly recognised by theorists .But this violence is fundamental in the sense that, as we shall see, it is theprecondition for the other interventions of the state .

The aim of this article is precisely to underpin this obviousness theor-etically, which means that one cannot start from the functioning of thestate - defined at the level of its 'organization' or its 'interventions' - oreven of its 'functions', defined as the search for remedies to 'dysfunctions'external to the state .

44

CAPITAL & CLASS

One cannot grasp the state as a 'regulator' (or even 'arbiter') betweendifferent fractions of capital, or as a (transitory) substitute for privatecapitalists in the event of a crisis, unless one has already established thefoundation which permits it to impose political and social norms while atthe same time being an integral part of capitalism as a whole. Our aim isnot, however, to produce an ontologising analysis, but to study what est-ablishes the state as "a separate entity, beside and outside civil society"according to the formulation of Marx and Engels in The German Ideology(p .78) .

It is the externality of the state in relation to 'civil society' that seemsto us to be the fundamental characteristic that distinguishes the state ofthe capitalist mode of production . It rests on the two central features ofcapitalism :(1) The production and circulation of commodities does not automatically

reproduce that which is external, and at the same time essential to it :the existence of labour-power as a commodity .

(2) 'Labour-power', the commodity specific to capitalism, in the sensethat its existence is the "unique historical condition" for the existenceof the capitalist mode of production, cannot be reproduced as a capit-alist commodity .It is because the conditions of reproduction of labour-power are ex-

ternal to the process of production of commodites that state interventionis necessary in order to :(1) 'produce' this specific commodity, i .e. to constrain the 'bearers' of

labour-power to enter into the wage-relation . It therefore amounts toassuring the conditions for the wage-relation . This is carried out inthe 'developed' countries today (where precapitalist forms of pro-duction have been almost completely destroyed) by the direct under-taking by the state of part of the reproduction of the workers . Butthis is by no means a necessary condtion for the reproduction of thewage-relation : for a century the reproduction of wage-labour inEuorpe was effected on the basis of the pure and simple destructionof the bearers of labour-power (2), while, up till the present day incolonial and neocolonial countries reproduction on an exclusively'domestic' basis alternates with the destruction of workers (3) .

(2) assure, in the course of its use, the reproduction of this commodity .The state is at the heart of the 'fragmentation of the collective lab-ourer', of the introduction of contradictions between wage-labourers,although they are 'objectively' reunited by the fact that they con-tribute to the production of a single mass of value .

PART ONE : THE STATE AND THE WAGE-RELATION

A . Externality of the reproduction of labour-power in relation to the pro-cess of production of commodities .

Marx's elaboration of the concept of labour-power is the result ofmany years of research during which he struggled with multiple contra-dictions .


45

On the one hand, Marx, who never abandoned certain Hegelian form-ulations, tried to provide a theoretical foundation for the fact that capitalcan only realise itself as such socially in the face of not-capital ; but, con-tradictorily, this "not-capital" cannot be labour, since the latter appears,within the production of commodities, only as a moment of capital .

On the other hand, Marx tried to provide atheoretical foundation for -and not merely to postulate, as Ricardo did - the existence of profit,without, however, violating the rules of commodity equivalence .

The 'invention' of the concept of labour-power is at first sight thesolution to this double problem : on the one hand, labour-power is definedindependently of capital, even if Marx recognised that it is no more than a"possibility" (4) ; on the other hand, thanks to 'labour-power', one can atlast name, encapsulate, that commodity "whose use-value possesses thepeculiar property of being a source of value, whose actual consumption istherefore itself an objectification of labour, hence a creation of value"(Capital, I, p .270), the commodity without which the whole theory ofsurplus-value would collapse .

But it is "within the sphere of circulation, on the market" thatlabour-power appears . If it is sold, there must be a' seller, and the sellercannot be confused with the commodity he or she sells . Two possibilitiesare open : either the seller is a person other than the bearer of labour-power, and so sells both 'bearer' and labour-power at the same time : thiswould be a relationship of slavery; or else the seller is the bearer of labour-power him or herself, free and equal - during the period of the exchange -with his or her co-contractor . Such an exchange corresponds to the gen-eral characteristic of commodity exchange : the commodity sold, labour-power, is a 'use-value' (5) for its buyer (thus defined as a capitalist) and anon-use-value for its seller, defined as a wage-labourer (note that it is non-use-value for him or her because they have been expropriated from themeans of production, because they cannot reproduce themselves by usingthis labour-power themselves) .

Labour-power clearly exists within circulation ; it is bought and sold,it is a commodity . But that in itself does not tell us what governs the con-ditions of this commodity exchange . The definition of the rules of ex-change by political economy, as much as in Marx's critique of it, post-ulates that it deals with a socially specified exchange : whether one post-ulates the rules of exchange as involving the equalisation of the rate ofprofit, the equality of marginal cost and marginal revenue, or the equalityof quantitites of abstract labour, the seller on whom these rules are im-posed is a capitalist, who is 'valorising' a capital .

There are therefore two possibilities : either the seller of labour-power is a capitalist. He or she is conflated, as an individual, with thecapital whose temporary disposal he or she sells . We fall into theoriesbased on the idea of 'human capital' . Or else the seller is not a capitalistand the relation entered with the buying capitalist is a simple com-modity relation . There is no reason to presume that this relation willfollow the rules defined on the basis of the case in which the seller is acapitalist .

Two types of relatively distinct problems follow: on the one hand, theproblem of the conditions of the exchange between the worker and the

46

CAPITAL & CLASS

capitalist, and, in particular, that of the determination of the price oflabour-power . On the other hand, the problem of the reproduction of theseller of labour-power. The distinct character of the two problems derivesfrom the fact that the worker and the capitalist do not seek the same thingin this exchange ; the workers seek to reproduce themselves as socialisedindividuals, and the wage exchange is the necessary,but not sufficient, con-dition for this reproduction . The capitalist seeks only to reproduce hisability to dispose of a labour-power concretely specified on the basis ofthe structure of the collective labourer ; to give the worker the means ofhis or her reproduction is neither sufficient (the worker still has to be intro-duced into the labour process under conditions imposed by the capitalist,which implies disciplinary practices at the level of the wage-exchange andin the labour-process itself), nor necessary in every case (the capitalist maybe able to find on the market workers reproduced outside any wage-rel-ation, and use their labour-power without, to that extent, giving them themeans to reproduce themselves) .

One source of the incoherence of the neoclassical theory of distri-bution lies in this confusion between the problem of the determination ofthe wage-level and that of the reproduction of the labourer ; but manyambiguities in the contemporary Marxist perspective have the same basis :indeed, it is often said nowadays that the wage is merely the 'price-form'of the 'value of labour-power', or the expression in money of the latter .From which is derived the idea, to be found in all the Marxist-inspiredtext-books, that the value of labour-power is determined, "like that of allother commodities", by the quantity of labour socially necessary to pro-duce it. But, over and above the question already raised of the failure totake account of the labour directed at the reproduction of labour-power,two points raise problems :(a) Firstly in relation to the determination of the quantity of labour

'socially necessary' . One may try to define this quantity outside thewage-relation, in which case one must establish a list of 'needs' . How-ever, this leads to two possible, but equally paradoxical, solutions :- either the wage-exchange conforms with the rule of determinationby the 'socially necessary labour-time', and every 'need' is necessarilysatisfied ;- or one tries to show that the needs are not satisfied, and the wage-exchange violates the rules of commodity exchange .The other solution is to define this quantity on the basis of the wage-exchange itself. But then one has a simple equality in terms of ex-change-value and not of value : the exchange-value of the commoditiesbought with the wage is equal to the exchange-value of labour-power .One cannot immediately discover any determination one way or theother from the observation of this rather tautological equality .

(b) if this determination of a 'socially necessary quantity of labour'creates problems it is because it leads back to another question: thatof the legitimacy of the use of the concept of 'value' itself in relationto labour-power. Certainly Marx adopts this usage time and again,even going so far as to make a direct connection between the cond-itions of production of consumption goods and the 'value of labour-


47

power'. However, when Marx defines value, he does so with respect tocommodities which are produced, and which are produced in acapitalist framework . (Although Marx poses the problems of thesubstance and of the measure of value before that of the valorisationof capital, this 'capitalist framework'is, at this level defined by the factthat the products of social labour appear, in a generalised manner, inthe form of commodities) . The use of the concept of value with res-ect to labour-power is only legitimate if the commodity 'labour-power' is of the same nature as the commodiites on which the defin-ition of the concept of value is based .To define the value of a commodity, one must first think of con-

crete labours ; as the commodity passes through circulation the ex-changers abstract from the concrete characteristics of the labour toretain only one common quality : the time spent . This is a prelimaryto the realisation of the value, that is to say to the fact that it maybecome real in being socially recognised . Nothing in this scheme cor-responds to what happens in the case of labour-power : concrete labour(of the housewife, of the teacher, etc . . . .) is not recognised : there isno process of abstraction, and one starts from a/ready realised values(those of the commodities bought by the worker) to define the valueof labour-power . This leads to the concealment of a very importantphenomenon : it is the worker him or herself, and not the labour-power, that is reproduced on the basis of the consumption of com-modities and of a labour process situated outside the process of pro-duction of commodities (and one does not know, a priori, whether ornot that process is subject to the constraints imposed by the needs ofthe production of value) .The capitalist form of production is thus characterised by a situationthat appears at first sight paradoxical :

(a) On the one hand, we find the production of commodities bearing avalue : that is to say the labour-time necessary for the production ofthese commodities can and must make itself recognised sociallythrough the intermediary of the market. Once the value of these com-modities is defined one can show, following Marx, that the process offormation of value implies that it is not just any labour that is recog-nised in this way : it is above all wage-labour, wage-earners being thecounterpart, the reverse of the social relation that is capital (self-val-orising value which poses as the subject of production) .

But if there are to be wage-labourers, it is necessary for the worker toenter commodity exchange as a free subject . If he or she is to be a freesubject, and is to be renewed as such, it is necessary that he or sheshould have a commodity to sell, and that is the condition for the re-production of capitalism . At first sight it matters little to the capitalistwhat process brings an individual, who is not preordained to do so, topresent him or herself as a partner in exchange : it is enough that thisindividual should actually be there, to sell the capitalist what he needs .What the capitalist needs (the specific commodity that only has use-

value for the capitalist) cannot be called 'labour' ; in fact, if capitalimplies the wage-earner, the existence of the wage-earner, in turn,

48

CAPITAL & CLASS

implies that the direct producer has been dispossessed of the sociallyrecognised means of production . There cannot therefore be any'labour' before the producer and the means of production come to-gether. Since the capitalist's partner in exchange encounters the cap-italist within circulation, prior to production, he or she can only sellthe capitalist a 'potentiality', a labour 'power' : it is up to the capitalistto make it a reality .

(b) On the other hand, alongside the production of value, we find all thesocial relations that determine that the capitalist actually finds beforehim the 'bearers of labour-power' that he needs . In contrast to what theviewpoint of the individual capitalist might lead one to suppose, thereis nothing spontaneous about this confrontation of the exchangerswith one another . Two types of condition have to be satisfied :-The reproduction of the potential force of production that is the

wage-labourer requires a consumption. Since the wage-labourer isdispossessed of the socially recognised means of production, it is nec-essary that he or she should either be able to find his or her means ofconsumption outside the framework of this production (family gar-dens, domestic labour, non-commodity education and health systemetc) or should be able to consume some of the commodities producedwithin the capitalist framework (which assumes that the wage-earnerhas access to the universal equivalent) . There is no process of surplus-value extraction whatever during the labour associated with this con-sumption (6) . Labour within the family unit does not produce value .The consumption carried out by the workers is a consumption of pureuse-values, those which are assigned to them by the social norms . Thisconsumption is not a production of new values, nor even conservationof value . It is an unproductive consumption (7) . But this unproductiveconsumption is nevertheless the mark of capital . It is only possible ifthe worker has previously inserted him or herself into the capitalistproduction process .But the bearer of labour-power, once his or her consumption is com-

pleted, is still not going to present him or herself spontaneously to thecapitalist, in the conditions required by the latter . Certainly, the factthat the capitalists as a whole command the commodity means of con-sumption is an element of the balance of forces that favours them ; butit has never been enough . In fact, wage-labour implies that the wage-worker must be a social subject who is partially autonomous : at themoment of exchange he or she must be free ; the capitalists as a wholemust ensure that this autonomy is limited in such a way that, despitehim or herself, the wage-worker will duly come and re-engage underconditions that they fix (not only conditions of the wage, but alsounder concrete conditions of labour) and within the framework of thedivision of labour that they try to impose .But the capitalists cannot control the reproduction of the bearer of

labour-power and of the conditions of wage-exchange from within thefield of value . Nor can they ensure from within this field that the potentialworkers should be adequate to the requirements of the labour process, ex-cept at the risk of a disorganisation of production during the whole timethat any inadequacy might last (8) .


49

It is these two types of condition that underlie the externality of thereproduction of the labourer in relation to the production of commodities .(In order to simplify we call the bearer of labour-power who actuallycomes to be inserted in the wage-exchange the 'labourer' ; strictly speakingthis is somewhat inaccurate: this bearer of the 'potential' to work onlyreally becomes the labourer within the process of production, the potentialdisappearing, immediately it is recognised, to give way to reality) . Outsidecommodity production we thus find :

-labour processes, in the general sense of the term : domestic labour,teaching labour, health care labour, administrative labour (social sec-urity, institutions connected with teaching etc) . Social relations arereproduced within these processes . (The term 'social relations' is usedhere in a general sense, and does not only describe the social relationsof production) .- control processes, whether this control is effected through overtviolence (police, army), through masked violence (administrative con-trol), or through indirect procedures of imposition of norms (throughthe family, teaching, health)(cf. Foucault 1976, esp. pp.152f) . Thesecontrol processes are also modes of reproduction of social relations .-other 'activities' (for want of a better general term), reducible toneither of the above, in which another set of social relations is repro-duced: affective, sexual, authority relations ; while these relations donot have a homogeneous theoretical status, they are certainly socialrelations: the individual is not able to choose whether or not to enterthem, but in entering them he or she reproduces them (9) .'Labour-power' cannot be produced as a capitalist commodity . Its

production and its reproduction cannot be the occasion either for the val-orisation of capital or for the production of value (10) .

There is thus a real 'externality' of the reproduction of labour-powerwith respect to the process of production of commodities . This externalityis situated at two levels :

-The 'bearers of labour-power' (potential labourers) are not producedwithin the framework of commodity production .-Their transition from the condition of 'potential seller' to that of'real seller' is not determined merely by the production-circulation ofcommodities .

B The Basis of State InterventionIt is the externality of the production-reproduction of labour-power

that is the basis of state intervention . The state intervenes-firstly so as to 'produce' this specific commodity : setting it to work,forcing its insertion into commodity relations .-then, in the conditions of sale ; the latter can only take place for a"limited time, or for a given task" . This is one of the contradictions ofbourgeois nations : the development of capitalist relations of pro-duction depends on liberation from serfdom and from individualslavery, and the constitution of an individually free labourer ; but col-lectively, the collective worker appears as a collective slave, not ofthe capitalists, but of capital . The state is at the heart of this contra-

50

CAPITAL & CLASS

diction : it must maintain the mass of wage-workers within a set ofnorms, certain of which are codified by the law, others not (cf . Pashu-kanis, 1951, for the analysis of the norms . c f. Foucault, 1967, for theanalysis of the origins of notions of 'extra-norms'), without at thesame time directly regulating their reproduction as individuals .-finally, the state must constantly intervene in the social relationonce it is established in order to ensure the historical conditions for itsreproduction .This ensemble is often described by the term economy policy : but

this economic policy is not as coherent and autonomous as its descriptionmight lead one to believe . It is essentially a case of managing the conjunc-ture, of a day-to-day response to the modifications in the power relationsbetween capital and the collective labourer (de Brunhoff, 1976 pp .4 ff) .

The place assigned to the state in this social relation is dual andcontradictory :(1) It is on the one hand a matter of reducing, or minimising the cost of

reproducing the social relation, which takes the form of the repro-duction of labour-power, as much at the level of each particular cap-italist as for the mass of capitalists . In fact, in a society in which allproduction is commodity production, in which therefore all expendi-ture is a process of value creation, the reproduction of labour-poweris the only true expense for the mass of capitalists . In other words, itis the only expense which is a non-production of value, i .e. the only'real' expense .

(2) It is on the other hand a matter of reproducing not labour-power, butthe conditions of existence of labour-power . The fundamental role ofthe bourgeois state is above all to "guarantee the existence of the classof wage labourers as the object of exploitation . . . Capital itself. . . isnot able to produce these foundations" (Altvater, 1973, p.99) .The dual aspect of the reproduction of labour-power does not always

appear clearly, to the extent that what is usually studied is the reproductionof the labourers . In other terms, one takes the wage-relation as given, asalready constituted once and for all . However the facts contradict thissimplistic view. The existence of the wage-relation demands overt violenceevery time such violence is thought necessary (11) .

Nevertheless this overt and explicit violence, which is unlimited butsporadic, is itself the condition for another continuous violence, latent,day-to-day and limited . But we are dealing with different sorts of violence :the object of the first is to underpin or re-establish a social order whichis endangered ; the object of the second is to negotiate the rate of exploita-tion . This difference in nature is often translated at the institutional levelinto a certain specialisation of different bodies, the sporadic violence ofthe army being directed (among other things) at the subordination at allcosts of the working class, to confine it within wage-labour ; while the dailyviolence of the police is situated, alongside other institutions, within theframework of the legal discussion of wages and of conditions of work (12) .

From this it follows that, like labour-power, the state is 'external' tothe process of production of commodities . But this externality has a dif-ferent status in each case, except for one point : the externality of the stateenables it, like labour-power, to be a 'realiser' of value . On the other


5 1

points, the externality of the state is dependent on that of labour-power,to the extent that it is only 'from the outside' that the state can assert,but also guarantee, the 'freedom' that is the reverse of the worker'sdependence .

This does not exclude the state from participating directly in the pro-cess of production as well : in the construction of 'industrial infrastructures',the control of credit, levying of customs duties, etc . It is indeed the repre-sentative of the mass of national capitalists, charged with ensuring thattheir collective interest prevails in the face of their divergent individualinterests . But, as something 'internal' to the process of production, it isnot necessarily different from those collective capitalist organisationswhich, generally, coexist with it or have even preceded it (chamber ofcommerce, trade associations etc .) .

The expression 'State (Monopoly) Capitalism' characterises a 'stage' ofcapitalism by these interventions of the state in the distribution of profitamong capitalists, in the devalorisation of capital, etc . In this sense, it aimsto specify both this 'stage' and the nature of the state by something whichis neither essential nor specific . However, what is essential - the redirectionof the wage-relation and its maintenance ) interferes with what appear tobe the properly 'economic' tasks of the state, situated within the field ofvalue, because they cannot be carried out except by virtue of an impositionon surplus-value .

This maintenance of the existence of the wage-relation is translatedinto a whole series of expenditures that we will describe as the costs ofexistence of capitalist production . These expenditures, far from beingthe faux-frais of production, are literally essential to it . They are the socialcost of capital . Each particular capitalist, as much as the mass of capitalists,aims to limit the extent of this cost, while recognising its social and polit-ical importance . The management of these expenditures is thus external tocapitalist logic at the same time as being the condition of its existence .

So it is the state that has to take on the role of sharing out the costsof maintaining the particular social relation, confused with the costs ofmanaging society ; the state which is the institution that must manage theinterests of the dominant class as the interest of the whole of the societyit dominates. But the state has no relevant criteria for calculating any kindof an 'optimum' in the distribution of costs or in the choice of the area inwhich the means of finance shouldd be levied . It is up to the administration(in the broad sense) to adjudicate the conflicts between the differentfractions of the bourgeoisie (the judicial apparatus only intervenes in thedetails, overall decisions always being taken at the political level), and thechanges in the alliances between bourgeois strata are the source of modi-fications in the circulation of surplus-value .

It is only once the maintenance of the conditions of existence of thewage-relation is assured that one can take into account the management ofthis relation . It is then time to 'manage' the reproduction of the labourersas a fraction of capital (variable capital), or as participants in the collectivelabourer. This management of the collective labourer is not done primarilythrough violence . Since the latter is omnipresent, it is a question :

-on the one hand, of making the social order accepted .

52

CAPITAL & CLASS

-on the other hand, of ensuring that these workers are adequate tocapitalist needs .

PART TWO: THE STATE AND THE 'MANAGEMENT' OF THE COL-LECTIVE LABOURER

In reproducing the wage-relation, the state recreates not only the pri-mary condition of existence of the bourgeoisie, but also, through its actionon the modes of distribution of the mass of wages, on the conditions ofreproduction of the wage-workers, it challenges the unity of the 'collectivelabourer' that is formed in the production of value .

A. The Collective Labourer (13)Most analyses of the collective 'labourer' start from the labour process

and, more particularly, from the division of productive labour (cf . in part-icular Nagels 1974 ; Berthoud 1974 ; Gouverneur 1975) : the individualworker does not produce a commodity only, while the production of com-modities requires not only workers acting directly on the matter trans-formed, but equally workers at the level of planning, the preparation oftasks, etc . The collective labourer is thus first of all connected with thedivision of labour, then with machinery to the extent to which it is thelatter that separates 'manual labour' and 'intellectual labour' (14) (and theanalysis of the collective labourer as a result makes it possible to reorientatethe debate on the definition of productive labour, the analysis thus nolonger being conducted at the level of the labour process alone, but of theimmediate process of production, so that it becomes possible to state thebases of an 'objective solidarity' between technicians, staff, etc . and opera-tives all defined as productive) .

Without denying the interest of such a direction of research, it seemsto us that the fact of defining the collective labourer at the level of the im-mediate process of production alone particularly limits the scope of theconcept and does not make it possible to put forward the basis of a theoryof education, of wage differentials, of qualifications, etc . But before goingfurther it must be noted at once that the 'collective' character of the lab-ourer who produces a commodity appears as the product of the specificallycapitalist (immediate) process of production (real subsumption) . Howeverthe latter is the (contradictory) unity of the labour process and the valor-isation process .

The collective labourer (in particular in Nagel's work) is generally onlydefined at the level of the labour process . However in the capitalist pro-duction process, there is not firstly a labour process producing use-values,in which the collective labourer appears (co-operation in the division oflabour) and subsequently a valorisation of capital on the basis of the pro-duct becoming a commodity . In fact the workers who co-operate are al-ready, from the moment the collective character of labour is established,"incorporated into capital" (15) . It is as a fraction of a value valorising it-self that they participate in the production of the commodity . It is not thecase that labourers produce collectively according to technical deter-


53

minations subsequently followed by the appropriation of the product bythe capitalist, because it is basically the (real) subordination to capital inthe production process and (from capital's point of view) the reduction ofthe mass of labourers producing a commodity to the status of variablecapital, that make possible the existence of the productive force of the col-lective labourer (a force of which the latter is, moreover, dispossessed) .

On this basis one can therefore define an initial structure of the col-lective labourer at the level of the immediate production process . Thisstructure (and this is the reason for its inadequacy) rests on the doublecriterion of concrete labour and of valorisation, without introducing thearticulation of production and circulation, in which the abstraction oflabour is realised ; and it is above all the fact of introducing valorisationbefore circulation which is at the root of this inadequacy . (16) . This struc-ture can be called the 'technical' structure to the extent that it does infact concern the cohesion-division of the concrete labours co-operating inthe production of a product . But this product is a commodity ; the co-herence and division that produce this initial structure have not emergedfrom just any technical rationality, but from the requirements of valori-sation (17) . Even at the level of the definition of the individual labourswhich (cohesive and divided) form the collective labourer (which onlyexists because it is subordinated to a single capital), this 'technical' struc-ture of the collective labourer can only be defined in its relation to capital .This initial structure of the collective labourer is that of the concrete(cohesive and divided) labours that it brings about (and not that of thelabourers who form it) and, at the same time, that it defines . The recog-nition of the use-value of the individual labour power only takes placewithin this collective labourer, that is to say to the extent of their beingadequate to the concrete tasks . The collective labourer produces a deter-minate use-value (value does not appear at this level) . But the character-istics of this use-value, as of the labour process itself, are determined bythe requirements of valorisation (18) . This explains why, even at this levelof the collective labourer, the concrete labours of command and controlappear together with a rhythm of labour that is the result of a compromisebetween the requirements of valorisation and workers' resistance .

The realisation of value, that is to say the social recognition of thequantum of labour incorporated in the commodities, is carried out throughexchange. But exchange itself rests on a relation of equivalence, and thisrelation of equivalence rests on the real abstraction of the labours that co-operate in the production of the commodity (19) . The definition of valueas a determinate quantity assumes that all the concrete qualities of thelabour expended are not taken into account .

The value of a commodity expended is not defined at the level of asingle production process (nor one carried out under the domination of aspecified fraction of capital) .

At the level of the totality of production processes of a given product,a collection of technical norms of production are defined, which are them-selves the product of the history of past techniques and of the competitionbetween fractions of capital valorising themselves by means of this deter-minate production . But these technical norms do not directly determine

54

CAPITAL & CLASS

the socially necessary labour-time, it is necessary to be able to think of theproduct as a commodity; and to think of a commodity (and of its value)one must simultaniously think of the totality of commodities .

At the level of the totality of production processes leading to the pro-duction of different products that are exchanged for one another, the pro-blem of equivalence appears, which in turn raises that of value . The hetero-geneity of the commodities exchanged is at the root of the social and realprocedure of abstraction of labour . It is not at the level of concrete labourthat a quantity of abstract undifferentiated labour emerges . However homo-geneous production norms may be, concrete labour in weaving cannot butbe differentiated from that which takes place in metal-working : thequantity of abstract labour necessarily emerges in circulation and makesthe concrete labour disappear (on all these point cf . Tortajada 1974) .

Already in the simple forms of value (under the aspect of the relativeform), exchange assumes that the values of the commodities exchanged(20 yards of linen and a coat, to take Marx's example) are two fractions ofa social expenditure of labour, and the exchange is the social procedurethrough which the quantity of socially necessary abstract labour is recog-nised . It is not a matter of a producer exchanging a 'good' with a constit-uted value against another 'good' with the same characteristic, since theabstraction of labour is the product of exchange at the same time as itscondition .

If one moves on to the general form of value (with multiple simul-taneous relations of equivalence), all the commodities exchanged partici-pate in the definition of the unity of abstract labour at the same time as inthat of the quantity incorporated in each good . It follows that what firstappears is the "labour-power of society" (20), and individual labour-powersare only realised (only become socially real) in appearing as part of thisglobal labour-power, and thus in negating their own concrete characteristics .

The totality of workers producing value under capitalism thereforecomprise a single collective labour, that is to say the value of each com-modity cannot be thought of as the product of an autonomous group ofworkers. The value of a commodity can only be thought of as a portionof the value created globally by the totality of workers confrontingcapital .

In circulation, the abstraction of labours therefore creates the undif-ferentiated character of the labourers (21) . The sanction which is providedby the determination of the socially necessary labour-time is not the recog-nition of the collective labourer who appeared at the level of concretelabour; on the contrary, this determination is made on the basis of a globalmass of abstract labour-time and of a confrontation between all the com-modities seeking to achieve social validation (recognition as values) .

The concrete characteristics of the labour process disappear in cir-culation . The first result of this is "the fetish character of the commodity" .The second result is the social negation of the specific use-value of eachlabour-power (a use-value which is nevertheless the basis of the contractualrelation between capitalist and labourer) . The third is that the commodityis constituted as capital (self-valorising value) in constituting the collectivelabourer as not-capital, which makes possible the reproduction of the


55

process of production, since this itself depends on the labourer being not-capital (22) . The commodity is not a 'thing', a 'product', or even an 'ex-change-value' : it is at the heart of the reproduction of social relations .

The opposition between capital and labour is not the product of acontractual relation (the immediate form of wage-labour) or of the (moreor less hierarchical) modalities of the labour process, even though the op-position is expressed in these places . The basis of this opposition is themechanism that makes each labourer the means (object) of the reproduc-tion of the whole of capital while excluding him or her from capital .

The production-circulation of value thus creates a unity of the collec-tive labourer which, certainly, is the basis of the 'objective' unity of theproletariat, but which is not a product of the revolutionary practice of theproletariat (23) . The 'management' of the collective labourer describes thecollection of practices by which this unity of the mass of wage-labourersis broken at the same time as the labourers are reproduced as a sum of in-dividuals offering the characteristics required by the production process .This 'management' is essentially conducted through the wage, in itscontradictory nature :

-from the point of view of the wage-earner, it reproduces the workeras an individual . The process of reproduction of the wage-workers canbe individual or collective and, in the latter case, takes place throughthe state .-from the point of view of the individual capitalist, the wage is themeans of reproducing 'his' collective labourer or, more exactly, thefraction of the collective labourer that enters into a relation with hisfraction of capital . The management of the mass of wages is thus, bythe diversification of grades and the opposition it introduces betweencategories of wage-earners, the means of isolating the workers andintroducing competition amongst them .-finally, from the point of view of the collectivity of capitalists, thewage is one of the means of reproducing a working class subordinatedto capital .

B. WagesThe action of the state in relation to wages is marked by the need to

reconcile the different objectives stated above, while guaranteeing the'confrontation' of 'dead labour' and 'living labour' (specified according toage, qualification) that this requires .

Moreover, state policies in relation to wages are not homogeneous,they are adapted to the wage form itself and so are i nscribed . i n a typologyof wages which it is worth spelling out . (The argument of this sectionderives from Lautier and Tortajada 1976) .

I . Direct wage and indirect wageThis distinction rests on an empirical approach, and is, moreover,

confused . In fact by direct wage is meant the price that is appropriate tothe labour contract . It is thus the price of the reproduction of the sub-ordination of the individual worker to the capitalist . It expresses the costto the individual capitalist of setting labour-power to work, but it is not

56

CAPITAL & CLASS

the price of the commodities brought by the worker for his or her repro-duction, since contributions, taxes, etc . are deducted .

The indirect wage is not itself a price . In fact it combines monetarybenefits (social security, allowances, etc) and the use of collective facilities(a use which is not matched to price) .

If the notion of the direct wage has a certain degree of relevance, thatof the indirect wage does not, since it is not defined in the same conceptualfield as the former .

2 . Private wage and social wageWhile the idea of the direct wage expresses the relation of constraint

and subordination that characterises the wage, this distinction underpinsits aspect of being the means of reproduction of the labourer . The privatewage is the price of the quantity of exchange value over which the wage-earner has effective control for his or her 'private' reproduction (essentiallywithin the framework of the family) . It is thus exchanged by the wage-earner him or herself for the commodities, the latter being consumed, butalso transformed, in a place external to the capitalist production process,but immediately controlled by the wage-earner him or herself.

The 'social' wage is the price of the mass of commodities consumedcollectively by the labourers, in places (whether institutionalised or not :school, health service, collective facilities, etc .) equally external to thecapitalist production process, in which the socialised reproduction oflabour power is effected .

This distinction therefore expresses the situation of the worker in thecourse of reproduction (as such the 'indirect' forms of the wage such asunemployment benefit, family allowances, etc, are nevertheless 'private'forms of the wage) . But it also expresses the modalities of the attempt bycapital, as a whole, to 'manage' this reproduction .

3 . Individual wage and collective wageThe capitalist does not 'manage' the wages of an isolated individual .

For him the wage is an outlay. The individual wage (the amount of moneynecessarily laid out to set the labour power of an individual to work, underthe form of 'direct wage', taxes, contributions, etc .) is only one element ofthe collective wage, both at the level of quantity and at that of the struc-ture of the collective wage .

As a quantity, the collective wage expresses the constraints imposedon the capitalist by valorisation . In its structure, it expresses the repro-duction of the hierarchical structure of the collective labourer, the latterbeing guided not only by political requirements, but also by the simul-taneous need to reproduce its technical structure .

4. Variable capitalVariable capital, for the individual capitalist, is a fraction of the cap-

ital advanced, specified in its employment (24) .As a fraction of capital, variable capital (which, when spent, is laid out

in the form of the collective wage) is certainly a value (in the process ofvalorising itself) . For the individual capitalist, that is from the point of


57

view of the formation of value (25), variable capital is value being ex-changed against labour-power, just as the commodities consumed by theworkers are values being exchanged against the wage . But this does notpermit one to postulate that labour-power has a value, because it is notproduced under capitalist conditions . To speak of the value (and not ofthe exchange-value) of labour-power is in fact to abstract from the ex-ternality of the reproduction of labour-power and to see in the wage onlythe price of the commodities consumed, the intermediary in the exchangeof one value (capital) against another capital (these commodities) andnothing else (26) .

If variable capital is the part of capital that 'varies', it cannot be as-similated to the totality of the outlays agreed to by a particular capitalistto reproduce 'his' labour-power'. A part of the variable capital is used in a'private' manner by the labourer to reproduce him or herself, another is'socialised' (whether laid out directly by the capitalist - employers'expenses connected with the payment of wages, etc - or whether itpasses through the wage-earner who pays contributions and taxes him orherself) . But a part of the socialised reproduction of labour-power is fin-anced on the basis of taxes or contributions not linked to the wage. Forthe individual capitalist it is not then a matter of variable capital, but of'faux-frais' (analagous to the faux-frais imposed by the requirements ofthe reproduction of the universal equivalent : they are necessary to thereproduction of capital as a whole, but not of particular fractions ofcapital) .

Nevertheless, if one considers the capitalists as a whole, these 'faux-frais' are a necessary expense (levied on capital) to set the collective lab-ourer in motion (27). The socialisation of capital implies the socialisationof the wage-relation and also the duplication of the concept of variablecapital : for the individual capitalist only the 'collective wage' (direct wagesand linked charges) is variable capital ; for capital in a given social space(defined for the moment as the socio-political space in which the social-ised reproduction of labour-power is effected), variable capital is formedby the totality of the outlays undertaken to set in motion collectivelabour-power, which is doubly structured (technically and hierarchically) .From this point of view, but only from this point of view, variable capitalfinances both expenses which are at one and the same time expenses forthe 'mobilisation of labour-power' and for the reproduction of the lab-ourers (like expenditure on education), and also expenses which onlyhave the former aspect (such as expenditure on the police etc .) .

5 . Relative surplus-valueThe unity of the capitalists in the 'management' of the reproduction

of the workers and in that of the mobilising and structuring of their labour-power, is not only expressed in the fact that they have to finance and'manage' the processes of the socialised reproduction of the workers andof the mobilisation of their labour-power . In fact, the mass of capitalistsare unified in the matter of the 'management' of the production of relativesurplus-value .

58

CAPITAL & CLASS

Before analysing the movement of relative surplus-value it is worthmaking it clear that, while labour-power does not have a value, surplus-value is well defined within the field of values . It is the difference betweentwo homogeneous quantities : on the one hand, the value produced at thesocial level by the collective labourer ; on the other hand, the quantity ofvalue that leaves capitalist circulation, that is realised in becoming non-value, unproductive consumption (29) .

Relative surplus-value is not sought for itself ; what a capitalist inDepartment II seeks is to increase his share of surplus-value . He increasesthe 'productivity' of the labourer, that is the possibility of reducing labour-time below that socially necessary, of producing extra surplus-value, or,more exactly, a surplus profit which appears at the level of the individualcapitalist .

But if, as a result of the generalisation of this movement, the globalvalue of consumption goods is reduced, relative surplus-value will not berealised by the capitalist who has achieved the increases in productivity,but by the mass of capitalists who buy labour-power (to the extent, al-though this is by no means a mechanical phenomenon, that the wage isreduced when the value of consumption goods falls) (30) .

What distinguishes relative surplus-value from 'extra surplus-value' orfrom 'absolute surplus-value', is essentially the location of the circulationin which it is realised (31) .

The 'management' of the cost of reproduction and mobilisation oflabour-power thus implies the united control by all the capitalists of theprice of consumption goods, whether the latter are produced in specificallycapitalist conditions (industry) or not (agriculture, certain 'services') .

What is called 'management of the reproduction of labour-power',with respect to the management of its cost, implies the control of thewhole of the production of Department II and implies consideration ofthe contradictions between the interests of capitalists of this departmentand the interest of the capitalists as a whole .

C. Socialised management and state management of the reproduction oflabour power.The need for socialised management of labour-power rests on a dual

foundation :1 . First on the triple meaning of the wage as a price . For the workers, it

is the means of their reproduction as a species (which implies the re-production of one generation after another). For the individual cap-italist, it is the means of mobilising labour-power in relation to therequirements of the structure of 'his' collective labourer . For capitalistsas a whole, it is the means of reproducing and mobilising a mass oflabourers defined socio-politically and technically . The dislocationsbetween these three meanings impose socialised management, forexample : - the education system must respond not to the immediateneeds of this or that capitalist, but to the needs of capital as a whole .The individual capitalist, in fact, will not finance the training of free

labourers if he is not assured that the latter will remain 'his' employees(32) while the existence of an education system is a need of capitalists


59

in general (to develop training appropriate to the process of production,but also a 'general' training, which is the ideological condition for thereproduction of the 'mobilisation' of labour-power and of the insertionof the labourers into social relations) . The education system is equallyrequired by the workers who come into conflict with the collectivecapitalist not over the principle of generalised education (33) but overthe content and extent of education .However the capitalists do not spontaneously demand the repro-

duction of those 'workers' who no longer have anything to sell :invalids, the retired, etc . It is only political conflict with the workersthat establishes the modalities of this reproduction .-In the case of the unemployed, the objective of the reproduction

of the workers comes into direct conflict with that of the reproductionof the 'mobilisation' of labour-power .While the maintenance of the unemployed depends largely on the

mechanisms of family solidarity, individual capitalists enter intodirect conflict with the general interest of the capitalists in the deter-mination of the payments made to the unemployed, not only becauseof such payment itself, but also because of the fact that 'excessivebenefits' have implications for the level of wages .But the reduction of unemployment benefits, which strengthens the

position of the individual capitalist, is only possible if the globalpower relation between capital and labour improves in favour of theformer .-Finally, capital 'manages' the production of relative surplus-value .

In this area the labourers are concerned not with exchange-value butwith the use-value of the consumption goods acquired by means ofthe wage .Once the latter has been negotiated, it becomes a question of con-

trolling prices, and beyond that, the development of productivity inDepartment II .The growing part played by the 'socialised' wage in relation to the

'negotiated' wage not only has the effect of increasing the role of thestate in wage management, but also of reducing, of transforming, themanner of wage negotiations with workers' organisations . The latter,in this respect, find their importance at the workplace reduced .

2 . The state is thus charged with creating a unity out of three aspects ofthe collective labourer :

-The collective labourer structured in terms of concrete labours, asthe latter are defined on the basis of the labour process . It must, des-pite the contradictions between individual capitalists, ensure thatspecific labour-powers are effectively sold with the qualifications nec-essary for the reproduction of the production process .- The collective labourer structured hierarcho-politically, that is to

say divided-The collective labourer structured hierarcho-politically, that is to

say divided according to three demarcations :*demarcation in terms of wage-level*demarcation in terms of the hierarchy of authority in the strictsense

*demarcation into "true and false wage-earners" (cf . Magaud 1974),

60

CAPITAL & CLASS

or, more exactly, between workers covered by collective agree-ments that guarantee a certain amount of employment and con-tractual negotiations over security, etc, and the others(cf. Aumeeruddy 1977) .

But the true cement of the unity of the collective labourer remainsthe production of value, which at the same time assigns a place to workerswhose function is the realisation of this value, or even the reproduction ofthe labourers . But any recognition of unity on this basis can only be an im-mediate recognition both of the unity and of the revolutionary characterof the proletariat .

The role of the state appears as a succession of discrete, fragmented,interventions precisely because it cannot explicitly base its practices onwhat really determines them . The imperative need to divide the workingclass demands that it 'manage' the latter in fractions, that it negotiatescattered wage 'privileges', as to the duration of labour, the age of retire-ment and the level of pension, unemployment insurance, forms of training ;all in a differentiated way . Certainly, the state is the place in which classconflicts are partially resolved, but this is only possible because theseconflicts are expressed in a displaced, indirect way. And this indirectnessis particularly expressed in the fact that the proletariat never appears assuch at the level of the state, but only appears as an aggregate of frag-mented categories. This means that the state, properly speaking, while itmanages the fragmentation of the collective labourer, does not directly'manage' the collective labourer as such .

CONCLUSION

The development of the debate over labour-power is not of purelyacademic interest. It is linked to problems posed at the level of politicalpractice, including the problem of domestic labour. But this debate isequally linked with the question of the strategy for a break with capitalism .Although it is rarely explicitly posed as such, the question of the abolitionof wage-labour remains at the heart of the definition of socialism .

What we have tried to show is that this question is not independent ofanother, which is equally central in the definition of a political strategy,the question of the state . The state under capitalism cannot be defined ex-cept by reference to the wage-relation .

The recent years of crisis have brought back to the fore the fact that'economic policy' is above all the management by the state of the divisionsin and the cost of reproduction of the collective labourer . Certainly thiswas a bit blurred by the three post-war decades, but that is nothing new :immediately after the armistice of November 1918, the German employersand unions established a system of collective agreements ratified by thestate, and it was Roosevelt, in 1935, who got the Wagner Act passed . How-ever, this 'management' itself rests on the status of labour-power undercapitalism, forced as it is to reproduce itself outside the process of com-modity production yet without being able to do so adequately for theneeds of capital .


61

However, even if the crisis shows that the state is partially unable totake over this management completely, it is more than risky to concludefrom this that we are witnessing a 'decomposition', if not a 'witheringway' of capital's state that would make it possible to avoid a revolutionarysituation ; as the pillar of the reproduction of the wage-relation, the state isalways in place, and nothing allows us to assume that the violence that ithas always shown will disappear of itself .

NOTES

This article is the result of collective work carried out at the universityof Grenable (France) from 1976 to 1978 . Previous drafts have been pre-sented at different meetings (Annual Conference of L'A .C .S .E .S ., Nice,September, 1976, Annual Conference of the C.S.E ., Bradford, July, 1977,The Society of Socialist Economists, University of Sussex, February,1978). Our thanks to all those who have discussed, criticised and encouragedour work, specially to Pierre Eisler (Grenoble), Simon Clarke, OlivierLe Brun and Sol Picciotto (Editorial Board of Capital and C/ass) . Respon-sibility remains ours . Criticisms and comments are welcome at the fol-lowing address :

Bruno Lautier, 20 Galerie de I'Arlequin, (3407), 38100 GRENOBLE,FRANCE .

1 . Cf., among others : "the state is the instrument of capital's dominationover the class of wage-labourers" (Altvater 1973, p.98). "The differentforms of the state apparatuses should not hide its universality asinstrument of the power of the dominant class . . . (Drugman 1973) .

2 . The reports on the condition of the working class dating from themiddle of the nineteenth century insist on this fact (the life expec-tancy of workers had fallen to 21 years) . The best known in France isthe 'Villerme Report' of 1840 (Tableau de l'Etat Physique et Moraldes Ouvriers . . ., republished by 10/18) .

3 cf. on this point, despite some theoretical confusions about the con-cept of 'labour-power' : Comite d'Information Sahel 1975 ;Meillasoux 1975 .

4 This is what Marx develops, although he does not use the word"labour-power", as early as 1857 in the Grundrisse, p.267 :

"The use-value which the worker has to offer to the capitalist,and he cannot offer anything else, is not materialised in a product,does not exist apart from him at all, thus exists not really, but onlyin potentiality, as his capacity . It becomes a reality only when ithas been solicited by capital, is set in motion . . ."

(N.B . Nicolaus's translation has been slightly modified to make itcorrespond to Dangeville's French translation) .

5 The term 'use-value' is used here in the usual sense : something thathas a certain usefulness, and not as the criterion for distinguishingbetween Departments I and II in the framework of the reproductionschemes, as Benetti 1977 suggests .

62

CAPITAL & CLASS

6 Contrary to what is the underpinning of a certain number of feministpublications. For a formalisation see Harrison 1973. For the critiqueof this position see Lautier 1974, pp .76 et seq .

7 Value only becomes socially 'real' in negating itself as value, in leavingthe circulation of value, just as capital only becomes capital in theface of its opposite . The realisation of value thus pre-supposes this un-productive consumption, which can only be carried out outside pro-duction, either in the consumption of the wage-earners, or in that ofthe capitalists, or in a destruction organised by the state .

8 This 'risk' was nevertheless effectively taken for over a century inFrance, up to the Astier law (1919) ; however, during this period, thereproduction of the collective labourer rested to a large extent on theinflow of workers formed within the artisanat .

9 We have found it useful to distinguish among the totality of socialrelations between those that are the result of the process of commodityproduction and those that are set outside the field of value . It is clearthat social relations at school, in the family, etc ., do not escape thecapitalist mode of production . The distinction is nevertheless neces-sary so as not to fall into the reduction of social reality to the pro-duction of commodities and social relations to the social relations ofproduction alone .

10 Lautier and Tortajada 1977, pp.265 et seq . As Marx suggests when heassumes a society in which capital has taken possession of all produc-tion : "Therefore only the capitalist is the producer of commodities(the sole commodity excepted being labour-power) ." (TSV, I, p .158)At the same time the "humanist' attacks on the Marxist theory, thatclaim that the validity of the theory of surplus-value depends on theproduction of human beings according to the rules of capitalist rat-ionality are invalidated (Schumpeter 1967, p .650 - "a special ob-jection", said he! Schumpeter 1943, pp.27-8) . By contrast, those whoextend the rules of capitalist rationality to human beings are in factcertain contemporary neo-classical economists such as Becker 1964,Friedman 1962 .

11 The litany of massacres perpetrated by the bourgeoisie against 'its'national working class, whether killing hundreds of thousands ofpeople (Indonesia) or in using napalm bombs (Bolivia), to say nothingof Argentina, Chile, etc ., may appear exotic . However, without goingback to the Paris Commune, it is as well to remember that Nazism aswell as fascism cannot be explained by the tendential fall in the rateof profit alone .

12 Our aim is not to analyse all the 'functions' of the army and of thepolice, notably in crisis periods .

13 This section derives from an earlier work of B . Lautier 1976, pp .13-19 .14 It is worth noting, in most of these analyses, a deformation of the

thought of Marx, because his "real subsumption of labour undercapital" is reduced to machinery. cf. :

"With the development of the real subsumption of labour undercapital, or the specifically capitalist mode of production, the reallever of the overall labour process is increasingly not the individual


63

worker . Instead, labour-power socially combined and the variouscompeting labour-processes which together form the entire pro-duction machine participate in very different ways in the im-mediate process of making commodities, or more accurately in thiscontext, creating the product . Some work better with their hands,others with their heads, one as Manager, engineer, technologist, etc .the other as overseer, the third as manual labourer or even drudge"(Capital I, pp.1039-40) .

15 "Being independent of each other, the workers are isolated . They enterinto relations with the capitalist, but not with each other . Their co-operation only begins with the labour process, but by then they haveceased to belong to themselves . On entering the labour process theyare incorporated into capital . As co-operators, as members of aworking organism, they form a particular mode of existence of capital .Hence the productive power developed by the worker socially ("trav-ailleur collectif" in French) is the productive power of capital .(Capital I, p.451)

16 In the same way, introducing valorisation before circulation (whetherin studies conducted in terms of 'branches', 'industries' or 'sectors'),makes it impossible to give any theoretical status to the 'quest for themaximum rate of profit', but rather to make it an exogenous datum, a'psychological disposition of the entrepreneur' .

17 Analyses of this question have become more widespread since thestart of the 1970s, tending to show the impossibility of analysing con-crete labour (and, in particular, the productivity of labour) withoutreference to the problem of valorisation and to the real domination ofcapital over labour in the production process . See in particular Gorzet al . 1973, (especially Gorz's article "technique, technicians et luttesde classe), Coriat 1976 .

18 "The process of production is the immediate unity of labour processand valorisation process, just as its immediate result, the commodity,is the immediate unity of use-value and exchange-value . But thelabour process is only the means whereby the valorisation process isessentially the production of surplus-value, i.e . the objectification ofunpaid labour" (Capital I, p .991)

19 "Equality in the full sense between different kinds of labour can bearrived at only if we abstract from their real inequality, if we reducethem to the characteristic they have in common, that of being ex-penditure of human labour-power, of human labour in the abstract .And it is only the exchange of products which performs this reductionby confronting products of different labours on an equal basis"(Capital I, p .166)(N .B . Ben Fowkes' translation of Capital has been slightly modified tomake it compatable with Joseph Roy's French translation - revisedby Marx -.)

20 " . . . the labour that forms the substance of value is equal humanlabour, the expenditure of identical human labour-power . The totallabour-power of society, which is manifested in the values of theworld of commodities counts here as one homogeneous mass of

64,

CAPITAL & CLASS

human labour-power, although composed of innumerable individualunits of labour-power . Each of these units is the same as any other, tothe extent that it has the character of a socially average unit of labourand acts as such . . ." ( Capital I, p.129, our emphasis) .The concept of "abstract labour" has been very little discussed in theMarxist literature. Although Rubin posed in explicitly in 1928 it isonly with the work of Napoleoni, Colletti and, more recently, Arthurand Kay that we have seen a renewal of the 'debate' .

21 A lack of differentiation of a completely different character from themovement generally analysed on the tendency of 'deskilling', of gen-eralisation of simple labour, which is situated at the level of concretelabour.

22 "The use-value which confronts capital as posited exchange-value islabour . Capital exchanges itself, or exists in this role, only in con-nection with not-capital, the negation of capital, without which it isnot capital ; the real not-capital is labour" (Grundrisse, p.274)

23 Here we cannot accept the conclusion of the Brighton comrades :"The revolutionary task of the working class is the reconstruction ofthe collective worker with the objective of Socialist accumulation".(Brighton Labour Process Group, 1976, pp .91-2)

24 "The means of production on the one hand, labour-power on theother, are merely the different forms of existence which the value ofthe original capital assumed when it lost its monetary form and wastransformed in the various factors of the labour process . ( . . .) On theother hand, that part of capital which is turned into labour-powerdoes undergo an alteration of value in the process of production . Itboth reproduces the equivalent of its own value and produces an ex-cess, a surplus-value, which may itself vary, and be more or less ac-cording to circumstances . This part of capital is continually beingtransformed from a constant into a variable magnitude. I thereforecall it the variable part of capital, or more briefly, variable capital ."(Capital I, p.317)

25 cf. Same reference : "The same elements which from the point of viewof the labour process, can be distinguished respectively as the objectiveand subjective factors, as means of production and labour-power, canbe distinguished, from the point of view of the valorisation process, asconstant and variable capital ."

26 And it seems to us that Marx does this especially in Capital I pp.273ff .27 Not in the sense of the sum of the workers co-operating in the pro-

duction of a commodity but co-operating in the production of itsvalue, that is to say participating in the process of production of a//commodities .

28 It seems that Brunhoff calls capital from the point of view of thecapitalist the "day-to-day value" of labour-power, and variable capitalfrom the point of view of capital-in-general the "reproduction value" .But then, if variable capital really is initially value, labour-power is notand does not have to 'validate' its 'value' socially, as all other com-modities have to . Thus "the problem of the relation between the day-to-day value and the reproduction value of labour-power is presented


65

anew, not at the level of principles (of the reproduction value beingsocially validated), but at the level of the arrangement of its financing ."

29 For obvious logical reasons, it does not seem to us that it is possibleto define surplus-value as "the difference between the use-value oflabour-power and its value" (Cartelier 1976, p.261). To be able toestablish a difference, the two elements must initially be homogeneous .

30 Marx clearly postulates such a 'mechanistic' relation :"The value of commodities stand in inverse ratio to the productivityof labour . So, too, does the value of labour-power, since it de-pends on the values of commodities ." (Capital I, p .436)

31 cf. K. Marx, Oeuvres I, p .1679 where M. Rubel quotes the lines, cutfrom the final edition of Capital : "Surplus-value is absolute, becauseit implies the absolute lengthening of the working day beyond thelabour-time necessary to enable the worker to live . Absolute surplus-value is relative, because it implies a development of the productivityof labour which allows the limitation of the necessary labour-time to apart of the working day . But if we consider the movement of surplus-value, this apparent identity disappears" .

32 A situation that cannot generally exist in contemporary capitalism .33 It is therefore only at first sight surprising to state that the laws

dealing with education under the capitalism of the Third Republicshould have been taken up almost word for word certain of the reso-lutions of the Paris Commune .

BIBILOGRAPHY

Altvater, Elmar, 1973, "Note on Some Problems of State Interventionism",Kapitalistate, 1, pp .96-108 . Reprinted in J . Holloway and S. Picciotto,ed, State and Capital (Arnold, 1978) .

Arthur, Chris, 1976, "Abstract Labour" CSE Bulletin .

Aumeeruddy, Aboo, 1977, La politique d'emploi des entreprises et lesformes d'emplois (Mimeo, Grenoble) .

Becker, G.S ., 1964, Human Capital (NBER, New York) .Benetti, C ., 1977, Marx et l'Economie Politique (Maspero, Paris) .Berthoud, A ., 1974, Travail Productif et Productivity du Travail chez Marx

(Maspero, Prais) .Brighton Labour Process Group, 1976, "The Production Process of Capital

and the Capitalist Labour Process, CSE Conference Paper, Coventry(Mimeo) .

Brunhoff, Suzanne de, 1976, Etat et Capital (PUG/Maspero, Grenoble) .Cartelier, J ., 1976, Surproduit et Reproduction (PUG, Grenoble) .Colletti, Lucio, 1972, "Bernstein and the Marxism of the Second Inter-

national" in From Rousseau to Lenin (NLB, London) .Comite d'Information Sahel, 1975, Qui se Nourrit de la Famine en Afrique

(Maspero, Paris) .Coriat, B ., 1976, Science, Technique et Capital (Seuil, Paris) .

66

CAPITAL & CLASS

Drugman, B ., 1973, Etat, lutte de classes et reproduction elargie du capit-alisme (Mimeo, CERES, University of Grenoble) .

Foucault, M ., 1967, Madness and Civilisation (Tavistock, London) ., 1976, La Volonte de Savoir (Gallimard, Paris) .

Friedman, M ., 1962, Capitalism and Freedom (Chicago UP) .Gorz, A ., 1973, Critique de la Division du Travail (Seuil, Paris) .Gouverneur, J ., 1975, "Le travail 'productif' en regime capitaliste",

Insthut des Sciences Economiques de Louvain, Working Paper 7503 .Harrison, John,1973, "The Political Economy of Housework", CSE Bulletin

Winter, pp .35-52 .Kay, Geoff, 1976, "A Note on Abstract Labour", CSE Bulletin, V, 1 .Lautier, Bruno, 1974, La Reproduction de la Force de Travail, These

d'Etat, University of Paris IX .

, 1976, "L'analyse du travailleur collectif", Grenoble .Lautier, Bruno and Tortajada, Ramon, 1977, "La force de travail comme

marchandise particuliere", in Sur l'Etat (Contradictions, Louvain) .Magaud, J ., 1974, "Vrais et faux salaries", Sociologie du Travial, ! .Marx, Karl, n .d . Theories of Surplus Value, Volume 1 (LawrenceMarx,Marx, Karl, n .d . Theories of Surplus Value, Volume 1 (Lawrence and

Wishart) ."

, 1965, German Ideology1965, German Ideology, (Lawrence and Wishart) .1973, Grundrisse (Penguin) .1976, Capital, Volume 1 (Penguin) .Oeuvres, Vol 1, (NRF, Paris) .

Meillassoux, Claude, 1975, Femmes, Greniers et Capitaux (Maspero, Paris) .Nagels, J ., 1974, Travail Collectif et Travail Productif dans /'Evolution de

/a Pensee Marxiste (University of Brussels) .Napoleoni, 1975, "Abstract Labour, Exchange and Capital in Marx", in

Smith, Ricardo, Marx (Blackwell, Oxford) .Pashukanis, E ., 1951, "The General Theory of Law and Marxism", in Babb

and Hazard (eds), Soviet Legal Philosophy (Boston) .Rubin, I ., 1972, Essayson Marx's Theory of Value (Black and Red, Detroit) .Schumpeter, J ., 1943, Capitalism, Socialism and Democracy (Allen and

Unwin) .Schumpeter, J ., 1967, History of Economic Analysis (Allen and Unwin) .Tortajada, Ramon, 1974, La reduction du travail complexe au travail

simple, these complementaire .

THE ANOMALIES OF CAPITAL

Susan Himmelweit and Simon Mohun

INTRODUCTION

Ever since the Conference of Socialist Economists was founded, con-troversy among its members has surrounded the fundamental concepts ofMarxian economics . Debate has been heated and vehement, confused andrepetitous, dogmatic and assertive, and productive of little agreement be-tween participants, except an agreement to disagree abusively . While thedebate has touched virtually all areas of Marxian economics, it basicallyconcerns the nature and status of value theory and its relation to the every-day concepts of wages, prices and profit . Positions on debates in otherareas, such as productive and unproductive labour, or whether one canmake statements about tendential movements in the rate of profit, cannotbe taken prior to an understanding of the basic debate about value . Whatthis paper attempts to do is to show on what issue an understanding of theconcept of value turns, and what the theoretical implications are for dif-ferent understandings .

In the process of developing the argument, we will survey the basicissues according to our estimation of their intrinsic importance . Clearlythis may differ significantly from the weight various issues have been ac-corded by past authors or by our current readers . Consequently, while wehope to summarise as clearly and succintly as possible the issues that havebeen raised, we do so within a framework about which there is no agree-ment, and we hope that those who disagree will bear with the argument tothe end rather than adopting the entrenched dismissive positions so com-mon in the past. Indeed, we would stress that our argument is addressedprecisely to the problems which are central to the neo-Ricardian critiqueof Marx's theory of value . While our discussion of these problems attemptsan answer to this critique, it is not thereby concerned with an uncriticaldefence of the letter of Marx's writings . Accordingly, we do think that wehave something new to say .

We have had some difficulty with terminology . One of the main pointsthat we shall make in this paper is that Marx made a separation of certaincategories that for others are not differentiated . Thus for example Marxdistinguishes exchange-value from value, while Ricardo tries to makevalue do for both . In general, we have tried to use the terminology of theauthor under discussion, but clearly we have had to move into Marx'susage when making specific criticism of the elaboration of the categoriesthemselves. We hope that this will not cause any confusion . In the interests

68

CAPITAL & CLASS

of (at least intial) clarity, we will give our understanding of Marx's usagehere. That of other authors we hope can be understood from the context .

For Marx, value is the property of commodities that differentiatesthem from products of labour in general (i .e . from things not producedspecifically in order to be exchanged) . Exchange-value is the proportion inwhich commodities exchange for one another in the market. Commoditiesare bought and sold for money, and the quantity of money for which theyexchange is called their price . Given a theory of money, the determinationof exchange-value is immediately the determination of price, and there-fore any theory of one is automatically a theory of the other . Accordinglywe use the phrases "theory of price" and "theory of exchange-value"interchangeably .

Another pair of categories which Marx distinguishes but Ricardo doesnot are those of abstract and concrete labour . So when we use "labour"when referring to Ricardo, we mean neither abstract nor concrete labourbut an undifferentiated category. This is important, for it is easy to slipinto viewing Ricardo as merely having failed to make Marx's abstraction,and thinking that what Marx called "concrete labour" was simply Ricardo's"labour" . We shall show in the paper why we consider this view to bemistaken .

It is important to note that in general "abstract" does not mean "theor-etical"; concrete categories are also theoretical . Indeed, "facts" are butabstractions themselves. However, a mere consideration of facts (surfacephenomena) leads us nowhere . Abstractions are validated as starting-points for analysis if and only if their elaboration produces an under-standing of more concrete categories . This seems to us to be what Marxmeans when he writes,

"The concrete is concrete because it is the concentration of manydeterminations, hence unity of the diverse . It appears in the process ofthinking, therefore, as a process of concentration, as a result, not as apoint of departure, even though it is the point of departure in realityand hence also the point of departure for observation and conception ."(Marx, Grundrisse , 1973, p.101)

A final terminological point : we occasionally talk of theories and ab-stractions as unscientific . We mean this in the minimal sense that theyhypostasise some cateogry that is historically specific, that is, that theytake a category that needs explanation as a facet of capitalist productionrelations, and eternise it, taking it as a given, whether of man, nature ofof production in general .

The plan of our paper is as follows . The first section considers theRicardian labour theory of value and the fundamental contradiction inwhich it is enmeshed . The second section considers Marx's theory of value,and shows how this theory, while located within a totally different method,incorporates, reformulates and can thereby resolve Ricardo's dilemma . Thenext section outlines Marx's critique of Ricardo, making more explicit themethodological differences between them, and concludes with a taxonomyof current positions on the basis of their stance with respect to Ricardo'sdilemma. The following two sections examine the current indictment of

THE ANOMA L IES OF CA PI TA L

69

value on charges of redundancy and inconsistency ; the alternative under-standings of value are shown to determine what case there is to answer .

Issues of correct and incorrect technical solutions to Marx's so-called"transformation problem" are related but not central to the substance ofour argument . We have therefore relegated them to an appendix . Whileour paper should be comprehensible without reading this, those who wishcan pursue the implications of our position into this area .

SECTION 1

RICARDO AND EMBODIED LABOUR

The modern criticisms that have been directed at Marx's theory ofvalue have their origin in Ricardian value theory . Consequently it is nec-essary to examine the latter in order to situate subsequent discussion sothat first, we can determine the precise significance of what constitutesMarx's advance on Ricardo, and second, we can show that the epithet neo-Ricardian, while often employed as a term of abuse, does have a certaindescriptive meaning and validity .

Ricardo developed his theory as an attack upon Adam Smith's cost-determined theory of price . As Ricardo saw it, Smith's view was that in asociety characterised by private property, the price of a commodity wasdetermined by the sum of its three constituent parts, wages, profit andrent; or as Smith put it,

"wages, profit and rent, are the three original sources. . . of all ex-changeable value ." (Smith, 1904, p .57)

Nevertheless, Smith did recognise that,

"in that early and rude state of society which precedes both the ac-cumulation of stock and the appropriation of land, the proportionbetween the quantities of labour necessary for acquiring differentobjects, seems to be the only circumstance which can afford any rulefor exchanging them for one another." ([bid, p .52)

But he found such a "labour theory of value" to be insufficient with theemergence of private property and the accumulation of means of pro-duction, for it takes no account of rent and profit ; accordingly Smithabandoned his labour theory in favour of what Marx later dubbed the"trinity formula ".0)

In this context Ricardo's achievement was to generalise Smith'slabour theory from Smith's "early and rude stage of society" to a worldin which production involves the use of accumulated stock (capital) .Ricardo rejected any notion that the value of a commodity was determinedby the remuneration due to the owners of the labour, capital and land neces-sary for its production . Instead, he considered that the value of a com-modity was determined by the quantity of labour necessary for its pro-

70

CAPITAL & CLASS

duction - both direct labour, and indirect labour embodied in the meansof production . Thus Ricardo's labour theory of value is essentially an em-bodied labour theory . And Ricardo is very clear on how he differs fromSmith :

" . . . Adam Smith thought, that as in the early stages of society, allthe produce of labour belonged to the labourers, and as after stockwas accumulated, a part went to profits, that accumulation, necessarily,without any regard to the different degrees of durability of capital, orany other circumstance whatever, raised the prices or exchangeablevalue of commodities, and consequently that their value was nolonger regulated by the quantity of labour necessary to their pro-duction. In opposition to him, I maintain that it is not because ofthis division into profits and wages, - it is not because capital ac-cumulates, that exchangeable value varies, but it is in all stages ofsociety, owing only to two causes : one the more or less quantity oflabour required, the other the greater or less durability of capital :-that the former is never superceded by the latter, but is only modifiedby it ." (Ricardo, 1951, pp .XXXVI-XXXVI I, emphasis in the original)

In this letter to James Mill (of 28 December 1818) Ricardo raisesissues which are germain to current controversies . For on the one hand hehad posited a theory of the value of products, independent of and prior totheir division between classes ; that is, there is in commodities an inherentvalue, to be subsequently allocated according to the claims each class hasby its role in production . This is in contrast to Smith's view that the sumof demands for remuneration arising from production relations (i .e . costs)constitutes the value of the commodity itself . But on the other hand forRicardo, although he was not always consistent, "exchangeable value" and"value" were generally synonymous concepts, and therefore although herejected Smith's view of value as being determined by the sum of costs, hestill required their equality . For in any theory of "equilibrium price", theterm that modern vulgar economy (neoclassical economics) gives to "ex-changeable value", this price must tautologically equal total costs paid,because the money paid for a commodity must be remuneration for some-one and a fully inclusive definition of costs is just total remuneration (2) .

This is why Ricardo was forced to recognise that the durability ofcapital modified the determination of value by the quantity of labour em-bodied . For if commodities do actually exchange in proportion to theirembodied labour times, the rate of profit per year of a capitalist whosecapital is required to be tied up for a longer period will be less than that ofone whose capital is used faster, even if the total amounts of embodiedlabour in the two cases are the same . This is clearly not how prices willremain : the price of the more slowly produced commodity must in equil-brium be higher than that of the other, otherwise no capitalist will advancecapital for its production . Since not all commodities do involve the tying-up of capital for equal lengths of time, it follows that if value is to beidentified with equilibrium price (exchange value) it cannot be determinedby embodied labour time without modification .


71

Ricardo made two very closely related points with respect to thedurability of capital . He defined capital to be circulating capital if it isturned over at least once in the time period under consideration (i .e . istotally consumed during that period), and to be fixed capital if it isturned over in a longer period (i .e . is not totally consumed in the period inquestion) . Now consider two capitals embodying the same total labour .First, they may be very differently divided into fixed and circulating com-ponents . Second, they may be identically so divided, but their fixed com-ponents may be of different durabilities . Either way, although total em-bodied labour is the same for each capital, the commodities produced byeach capital must differ in (exchange) value, because of the differenttimes for which the various components of each capital are tied up . This isnecessarily the case in an economy in which competition equalises the rateof profit on each capital over the time period being considered . Since thedistinction between fixed and circulating capital merely has reference tothe time period chosen, the two cases can be subsumed under the genericpoint that because in equilibrum each capital must receive the same rateof profit, then different turnover times must modify the determination of(exchange) value by labour embodied .

It cannot be stressed too strongly given their contemporary currencythat these results are not new ones ; they emerge out of the first chapter ofRicardo's Principles . . ., and indeed gave Ricardo a great deal of trouble .Thus he wrote to McCulloch in June 1820,

"I sometimes think that if I were to write the chapter on value againwhich is in my book, I should acknowledge that the relative value ofcommodities was regulated by two causes instead of by one, namely,by the relative quantity of labour necessary to produce the com-modities in question, and by the rate of profit for the time that thecapital remained dormant, and until the commodities were brought tomarket." (Ibid, pp .XXXIX-XL)

But he persisted with his embodied-labour theory of value, writing fourmonths later to Malthus,

"You say that my proposition 'that with few exceptions the quantityof labour employed on commodiites determines the rate at whichthey will exchange for each other, is not well founded' . I acknowledgethat it is not rigidly true, but I say that it is the nearest approximationto truth, as a rule for measuring relative value, of any I have everheard." (Ibid) .

Yet Malthus was correct within this framework . For with the formation ofa general rate of profit in the economy, exchange-values of commoditiesdo vary with both the quantity of embodied labour and the temporalstructure of how that labour is embodied in capitals of different durabili-ties . If these structures differ, then so do relative (exchange) values fromratios of embodied labour. Ricardo was reduced to saying that embodiedlabour was the most important element in the determination of value . But

72

CAPITAL & CLASS

within a theory which is only concerned with the explanation of the mag-nitude of exchange value, this is an arbitrary assertion .

Thus Ricardo refused to concede that embodied-labour time mightnot be the predominant determinant of exchange-value ; but to justify thisrefusal he should have set himself the problem of deriving the necessarydifferences between value and price from the determination of value it-self. However, posing the problem in this way demands a distinction be-tween categories of value and exchange-value not consistently present inRicardo's work . The recognition of a contradiction between a labour-embodied theory of value and a cost-summation account of price marksthe limits of Ricardo's scientific achievement (3) .

SECTION 2

MARX AND ABSTRACT LABOUR

Two possible resolutions of Ricardo's contradiction exist . One wayout of the dilemma is to abandon the first approximation of the labour-embodied theory of value in seach of some other account of the magnitudeof exchange-value. Such a path historically comprised the retreat fromscience to vulgar economy . The other possible resolution involves the com-plete reconceptualisation of value, a recasting of the theory of value as anabstraction, rather than an hypostasised assumption, wherein its signifi-cance and status is such that its apparent inconsistences can be recreatedas the expression of the real contradictions of capitalist society . This wasMarx's project, the development of a theory of value as the specific ap-plication of his method of historical materialism in order to analyse theproduction relations of capitalism .

The method of historical materialism defines the differentia specificaof epochs of histroy by class relations established between those whoproduce and those who appropriate the surplus . That is,

"the essential difference between the various economic forms ofsociety . . . lies only in the mode in which this surplus-labour is in eachcase extracted from the actual producer, the labourer ." (Marx, 1938,Capital I, p .200)

More generally,

"The specific economic form, in which unpaid surplus-labour ispumped out of direct producers, determines the relationship of rulersand ruled, as it grows directly out of production itself and, in turn,reacts upon it as a determining element . Upon this, however, is foundedthe entire formation of the economic community which grows up out ofthe production relations themselves, thereby simultaneously itsspecific political form . It is always the direct relationship of theowners of the conditions of production to the direct producers - arelation always naturally corresponding to a definite stage in the devel-opment of the methods of labour and thereby its social productivity -which reveals the innermost secret, the hidden basis of the entiresocial structure . . ." . (Marx, 1972, Capital III, p.791 )


73

The application of such an understanding to capitalist societies, then,requires first of all investigation of what is specific to the form of pro-duction in these societies ; that is, that it is the production of commodities .For without this, the specific form of surplus extraction that defines thecapitalist mode of production could not be described . This is because cap-italist exploitation is the form of surplus extraction that requires for itsdefinition only relations of commodity production and exchange, includingcrucially those of labour-power. For the extraction of surplus-value,which distinguishes capitalism from other class modes of production, isitself a commodity relation . Hence, while Marx had many different starting-points in his analysis, he settled on the commodity to begin his finalexposition (4) .

The distinguishing feature of commodities is that they are producedfor exchange . In such an exchange process, two commodities are measuredagainst each other in determinate proportions, yielding a relation ofequivalence between a certain quantity of one commodity and a certainquantity of the other. There is therefore an equalisation of the com-modities as far as their exchange-value is concerned, and a differentiationwith respect to their use values (the effects to be gained from subsequentconsumption of the commodities) . Exchange value is thus merely thatwhich is posed as equal in the pure act of exchange, something manifestedby commodities in exchange, after their production . But since what is atissue is the production relations of capitalism, rather than abstracting fromthe production process itself the analysis has to investigate what it is thatdifferentiates the production of commodities from that of products ingeneral .

Since production of a commodity is both the production of a use-value,and the production of an exchange-value,we can make a similar conceptualdistinction between the labour which produces the one aspect and thelabour which produces the other . The labour which produces the individ-uated useful properties of a commodity is useful or concrete labour ; suchlabour produces products, but it is only in certain sorts of societies thatproducts become commodities . For in these societies, in addition to theaspect of labour which produces use-values, there is another aspect oflabour which produces use-values as commodities, which is abstract labour .Marx's "value" is the product of abstract labour . Accordingly, value is acategory of commodity production, whose form is exchange-value ; whatgives commodities exchange-value is the labour that remains on abstractionfrom the labour that produces use-value . Now in a sense such an answer tothe question of what gives commodities exchange-value is already pre-empted by that very question, for all that Marx has to do is to specify thenature of commodity-producing labour. So to talk of the proof of such aspecification is meaningless . But, within the framework of historicalmaterialism, that Marx's answer is not an arbitrary one is ensured by thenature of the exchange itself, since it carries on as a real process thecommensuration of the products of labour under commodity production .

Therefore there can be no a priori determination of abstract labour,for not until commodities are actually exchanged on the market can theproducts of individual producers satisfy the needs of others . It is the pro-

74

CAPITAL & CLASS

cess of exchange on the market which manifests the social character ofindividual labours, establishes the social connections between independentcommodity producers, and which thereby determines that the valuerealised in exchange (exchange-value) is the form of appearance of thatlabour, and only that labour, which is socially necessary to the productionof the commodity in question . Hence value is measured not in units ofembodied labour time, but rather in units of "socially necessary labourtime." Thus the reduction of labour to abstract labour is something thatcan only be done by the market ; the value of a commodity has to beexpressed, and then only after the event, in the use-value of anothercommodity .

Eventually, one commodity becomes that against which the value ofall other commodities is expressed ; one commodity becomes universalequivalent, that is, assumes the money-form .

"Commodities, first of all, enter into the process of exchange just asthey are. The process then differentiates them into commodities andmoney, and thus produces an external opposition corresponding tothe internal opposition inherent in them, as being at once use-valuesand values. Commodities as use-values now stand opposed to moneyas exchange-value . On the other hand, both opposing sides are com-modities, unities of use-value and value . . . These antagonistic forms ofcommodities are the real forms in which the process of their exchangemoves and takes place ." (Marx, 1938, Capital I, pp.77-8)

So now the separation of the expression of the value of a commodity fromthe commodity itself is complete . The value of a commodity has no ex-pression except as exchange-value, commensuration of itself againstanother commodity in the market . But this exchange-value itself has noexpression except against one particular commodity, money . Price is thissole expression of value (and exchange-value) . There is no manifestation ofvalue in terms of its substance, abstract labour, nor of its measure, sociallynecessary labour time . The only form in which value appears, and the onlyway it can appear, is in terms of the money commodity (gold, for example)and its quantitative measure (weight, for example) . This is what theprice of a commodity is, the quantity of the money-commodity for whichit will exchange. It is the only form in which exchange-value is expressed,and clearly therefore the only form for the expression of value too .

Marx summarised his method as follows :

" . . . I do not proceed on the basis of 'concepts' hence also not fromthe 'value-concept', and I do not have the task of 'dividing' it up inany way, for that reason . What I proceed from is the simplest socialform in which the product of labour in contemporary society mani-ests itself, and this as 'commodity' . That is what I analyse, and first ofall to be sure in the form in which it appears. Now I find at this pointthat it is, on the one hand, in its natural form a thing of use-value,alias use-value, and on the other hand that it is bearer of exchange-value,and is itself an exchange-value from this point of view. Through


75

further analysis of the latter I discovered that exchange-value is onlyan 'appearance-form,' an independent mode of manifestation of thevalue which is contained in the commodity, and then I approach theanalysis of this value ." (Marx, 1976a, Marginal Notes on Wagner,p.214. Emphasis in original )

So his starting point is not an arbitrary assumption, but a reality, the com-modity, considered as the social, historically specific form of the product .And the abstraction which allows this consideration of the commodity-form is a real one. The process of the theoretical discovery of abstractlabour is not merely a process of mental generalisation, but has a realexistence in the reality of the exchange process . The equalisation of pro-ducts of labour on the market occurs every day, standardised by money,the universal equivalent of value . Since individuals alienate their productsas commodities in exchange, so too do they alienate the labour producingthose commodities . Abstract labour is a real activity, a social reality,whereby individuals alienate their labour-power from themselves . Collettiputs the point well :

"This in turn implies that in a society in which individual activitieshave a private character, and in which therefore the interests of in-dividuals are divided and counterposed, or, as we say, in competitionwith one another, the moment of social unity can only be realized inthe form of an abstract equalization, ignoring the individuals them-selves ; hence, in this case, as a reification of labour-power - a labour-power which is said to be equal or social, not because it genuinelybelongs to everyone and hence mediates between the individuals, butbecause it belongs to nobody and is obtained by ignoring the realinequalities between the individuals." (Colletti, 1972, p .87. Emphasisin original)

Hence the analysis of the commodity-form reveals exchange to bemore than the equalisation of products . For equality of human labour canonly take the form of a quantitative value-relation between the productsof labour; and the social realtions between the producersof labour; and the social relations between the producers can only take theform of social relations between their products . This of course is central toto Marx's theory of value, for

" . . . the labour of the individual asserts itself as a part of the labourof society, only by means of the relations which the act of exchangeestablishes directly between the products, and indirectly, throughthem, between the producers . To the latter, therefore, the relationsconnecting the labour of one individual with that of the rest appear,not as direct social relations between individuals at work, but as whatthey really are, material relations between persons and social rela-ions between things." (Marx, 1938, Capital I, p.44)

From this understanding of commodity fetishism, Marx develops hisanalysis to reveal behind the equality of the exchange relation, the inequal-

76

CAPITAL & CLASS

ity of capitalist production relations . As soon as the direct producers'ability to work (labour-power) assumes the commodity-form, then theelaboration of the contradictions immanent in such a form constructs themode of surplus extraction in capitalism . For once the conditions underwhich labour-power becomes a commodity are realised, then its uniqueuse-value (realised in its consumption in the labour process) of creatingmore than its own value is sufficient to explain both the production of thesurplus and its extraction as value by the capitalist. Through this under-standing of capitalist production relations, the extraction of surplus-value can be understood without recourse to theories of unequal exchange .

This marks a crucial difference of Marx's value theory from an em-bodied-labour theory of value : the latter cannot correctly theorise theclass relations of capitalism . For without an understanding of the pro-duction relations of capitalism, the exchange relation between capital andlabour can only be treated as of equal significance with those betweendifferent fractions of each class . That is, within Ricardian theory, com-petition between capital and competition in the labour market are logicallyas fundamental as the exchange between capital and labour itself . Whetherthe proponents of such theory recognise the more fundamental signifi-cance of the latter, that recognition is a purely arbitrary choice, followingnot from their theory but from a superimposed ethical position . When allparts of the whole are analysed as exchanges, there is no way in which theexchange between a worker and a capitalist is necessarily exploitative, un-less that between, for example, one capitalist and another is potentially soas well . To overcome the symmetry of the exchange relation and to un-cover the production relations that constitute the basis of capitalist ex-ploitation requires a recognition of the commodity as the specific form ofthe product under capitalist relations . To do this requires the formulationof a completely new set of categories which cannot be situated in theembodied-labour theory of Ricardo. And all this, despite the seeminglysmall change which Marx makes :

"The value of every commodity . . . is determined not by the nec-essary labour-time contained in it, but by the social labour-time re-quired for its reproduction ." (Marx, 1972, Capital III, p .141 . em-phasis in original)

However, this is only the first step . For the formal subsumption of labourto capital on the basis of previous techniques of production gives way toits real subsumption as capitalist relations affect every aspect of the labour-process and "revolutionise" the mode of production . The laws of motionof capitalist development can then be elaborated . But capital is here stillconsidered as capital in general,

"the incarnation of the qualities which distinguish value as capitalfrom value as value or as money . Value, money, circulation etc .,prices etc ., are presupposed, as is labour etc . But we are still con-cerned neither with a particular form, nor with an individual capital asdistinct from other individual capitals etc . We are present at the process

THE ANOMA L I ES OF CA PI TA L

77

of its becoming . This dialectical process of its becoming is only theideal expression of the real moment through which capital comes intobeing. The later relations are to be regarded as developments comingout of this germ ." (Marx, 1973, Grundrisse, p.310 . Emphasis inoriginal. See also pp.449-50)

It is only through the development of these later relations that we canrecognise that not only is production under capitalism production for(surplus) value rather than for use-value, but capital itself is value in pro-cess; a process in which a given value takes the form of different use-valuesin the process of its self-expansion . It is only through this separation ofvalue from use-value, a social separation and not merely an analytic one,that the exploitative relation that is capital functions . On this basis, capitalin general can be individuated intodifferent capitals through the synthesis ofproduction with the circulation of different use-values ; but now commod-ities can no longer be taken to exchange at their values.

Marx derives this result by showing initially how the logic of his anal-ysis leads to the presumption of different rates of profit in different in-dustries ; and this for two reasons . First, only living labour produces value,but the rate of profit is defined across total capital . Thus for a given rateof surplus-value, a greater proportion of capital represented by labourpower (variable capital) rather than means of production (constant capital)will produce more surplus-value than one with a higher ratio in valueterms of means of production to labour power . This ratio, the value com-position of capital, determines for a given value of capital and rate of sur-plus-value, the quantity of surplus-value produced . If we could work outthe rate of profit in value terms, it would be the total surplus-value pro-duced per unit of total capital, and would thus vary, for a given rate ofsurplus-value, according to the value composition of capital .

Secondly, if two capitals, again with the same rate of surplus-value,and of equal value, have the same value composition but different turn-over times, then the capital which turns over faster (has the shorter turn-over time) has the higher rate of profit, because surplus-value is producedeach turnover yet the capital advanced remains unaltered . But havingstated the problem with respect to different turnover times (a statement inwhich the common basis with that of Ricardo's stumbing block is mostobvious (5) ), Marx proceeds to concentrate on the problem with respectto different compositions of capital .

The rate of profit under discussion is the "value rate of profit", or therate of profit in value terms upon each capital . Algebraically, this is givenby

r= s

= svc+v

c/v+1where c, v and s are respectively, the constant and variable parts of, andsurplus value produced for, an individual capital . The rate of exploitationis then s/v, while c/v gives the value composition of capital . The effect ofcompetition between capitals is to create a tendency for the rates of profiton different capitals to be equalised, and hence a tendency towards the

78

CAPITAL & CLASS

formation of a general rate of profit . Now on the assumption of a uniformrate of exploitation s/v -. (6), but no such assumption of uniformity for c/v,the "value rate of profit" will vary across industries and cannot thereforebe this general rate of profit .

Marx therefore implicitly dismisses the "value rate of profit" at leastfor an individual capital when he says :

"differences in the average rate of profit in the various branches ofindustry do not exist in reality, and could not exist without abolishingthe entire system of capitalist production . It would seem, therefore,that here the theory of value is incompatible with the real phenomenaof production, and that for this reason any attempt to understandthese phenomena should be given up ." (Marx, 1972, Capital III, p .153 )

But of course he does not give up his attempts to understand the real phen-omena. Recognition of the incompatibility implies recognition that com-modities necessarily cannot exchange at their values ; this as a direct resultof the extension of value analysis to its logical results in capitalist com-petition . For while value analysis implies the commensuration of sociallynecessary labour-times through commodity exchange, capitalist competitionimplies the commensuration of paid labour-times through the formation ofan average rate of profit . These two commensurations have been shown tobe in contradiction . The latter is a consequence of the former, since it isonly as a consequence of the commensuration of commodities in exchangethat labour-power can become a commodity and capitalist competitionresult . Thus, though competition's commensuration is a consequence ofthat of socially necessary labour-times, it modifies the operation of thelatter .

This modification is expressed in the transformation of values intoprices of production . This, the "transformation problem", is therefore anecessary result of the contradictory nature of capitalist production rel-ations, and not at all a problem with Marx's theory, which recognises thisresult . Indeed, we could rather say that Ricardo's problem is that by failingto recognise the contradiction between the two commensurations, he doesnot pose the transformation problem - he just requires its solution. Theimportant point, then, is not the technicial correctness of Marx's solution(we leave discussion of this to the Appendix), but that the neccesity forthis transformation of values into prices of production has been derivedfrom his understanding of value itself .

SECTION 3

METHODOLOGICAL DIFFERENCES

The previous section situated Marx's theory of value within this prob-lematic of historical materialism ; the present section requires however amore explicit presentation of the methodology of his (abstract) labourtheory of value, partly in order to elucidate further the theoretical process

THE ANOMA L I ES OF CA PI TA L

79

of abstraction, and partly to situate Marx's criticisms of Ricardo's (em-bodied) labour theory of value . The inclusion of the latter is justified bythe dismal fact that many today continue to work in the Ricardian frame-work and that indeed much modern "Marxism" has not only failed to gobeyond Ricardo's theory but has indeed retreated from its very sub-stantial achievements.

Marx's fundamental critique of Ricardo, as the best exponent of clas-sical political economy, is that he

"never once asked the question why labour is represented by the valueof its product, and labour-time by the magnitude of that value."(Marx, 1938, Capital I, p.52)

and so for capitalism

"does not examine the form - the peculiar characteristic of labourthat creates exchange-value or manifests itself in exchange-values -the nature of this labour ." (Marx, 1969, Theories of Surplus Value11, p .1 64 . Emphasis in original .)

Since the question at issue is why under capitalism labour and labour-time take these particular forms, a failure to resolve this question sim-ultaneously fails to identify the historical specificity of the capitalistmode of production. Thus Marx remarks that

"It is one of the chief failings of classical economy that it has neversucceeded by means of its analysis of commodities, and, in particular,of their value, in discovering that form under which value becomesexchange-value . Even Adam Smith and Ricardo, the best represent-atives of the school, treat the form of value as a thing of no import-ance, as having no connection with the inherent nature of commodities .The reason for this is not solely because their attention is entirely ab-sorbed in the analysis of the magnitude of value . It lies deeper. Thevalue form of the product of labour is not only the most abstract,but is also the most universal form, taken by the product in bourgeoisproduction, and stamps that production as a particular species ofsocial production, and thereby gives it its historical character . If thenwe treat this mode of production as one eternally fixed by nature forevery state of society, we necessarily overlook that which is thedifferentia specifica of the value-form, and consequently of the com-modity-form, and of its futher developments, money-form, capital-form, etc ." (Marx, 1938, Capital I, p.52, n .2. Emphasis in original)

Now there are two senses in which the analysis could fail to identify thehistorical specificity of capitalism. In a weak sense, the analysis might justfail to explain what is historically specific . In this sense Ricardo's analysiscan be judged wanting, for while he is undoubtedly talking about capital-ism, he fails to explain its historical specificity . Because of this failure,Ricardo's analysis is enmeshed in a contradiction irresolvable within his

80

CAPITAL & CLASS

framework . The resolution in the direction of vulgar economy leads to thestrong sense of the lack of historical specificity : historically specific cate-gories are rendered universal and hence natural . It follows that any analysiswhich uncritically employs these categories will always tend to ascribeasocial, natural, even eternal qualities to what is socially specific to capital-ism (7) . For economic categories are but thought-constructs which aredeterminate abstractions of the prevailing relations of production ; that is,

"The categories of bourgeois economy . . . are forms of thought ex-pressing with social validity the conditions and relations of a definite,historically determined mode of production, viz., the production ofcommodities ." (Ibid, p .47)

Without this recognition, for Marx, no analysis can be scientific .Ricardo's "great historical significance for science" was to show that

"The basis, the starting point for the physiology of the bourgeoissystem - for the understanding of its internal organic coherence andlife process - is the determination of value by labour-time . Ricardostarts with this and forces science to get out of the rut, to render anaccount of the extent to which the other categories - the relations ofproduction and commerce - evolved and described by it, correspondto or contradict this basis, this starting-point ; to elucidate how far ascience which in fact only reflects and reproduces the manifest formsof the process, and therefore also how far those manifestationsthemselves, correspond to the basis on which the inner coherence, theactual physiology of bourgeois society rests on the basis which formsits starting-point ; and in general, to examine how matters stand withthe contradiction between the apparent and the actual movement ofthe system. " ( Marx 1969, Theories of Surplus Value 1 1, p .166, firstemphasis in original, second added .)

But his procedure was misguided :

"Ricardo's method is as follows : he begins with the determination ofthe magnitude of the value of the commodity by labour-time and thenexamines whether the other economic relations and categories contra-dict this determination of value or to what extent they modify it . Thehistorical justification of this method of procedure, its scientificnecessity in the history of economics, are evident at first sight, but sois, at the same time, its scientific inadequacy . This inadequacy notonly shows itself in the method of presentation (in a formal sense),but leads to erroneous results because it omits some essential links anddirectly seeks to prove the congruity of the economic categories withone another ." (Ibid, pp .164-5 . Emphasis in original)

But there can be no proof of "the congruity of the economic cat-egories with one another" in the way in which Ricardo attempts . ForRicardo begins with the existence of commodities, wages, capital, profit,


81

the general rate of profit, the various forms of capital in circulation (fixedand circulating) and so on . In other words he begins with the whole ofcapitalist production and a complete theorisation of the relation betweenwage and profit rates . Far from postulating the existence of a general rateof profit,

"Ricardo should rather have examined in how far its existence is infact consistent with the determination of value by labour-time, and hewould have found that instead of being consistent with it, primafacie, it contradicts it, and that its existence would therefore have tobe explained through a number of intermediary stages, a procedurewhich is very different from merely including it under the law ofvalue." ([bid, p .174. Emphasis in original )

The lack of mediation of categories in Ricardo's work proves to be aninsuperable barrier within his frame of reference . In the face of the contra-diction between a labour-embodied theory of value and a cost-summationaccount of price, Ricardo's scientificity consisted in his obstinacy in main-taining the former in the face of the latter, rather than dissolving it in thevulgarity of passive reflection upon appearances . Yet consider the cate-gory embodied labour . A sort of abstraction is certainly being made, for inso valuing a commodity, various aspects of the labour performed in itsproduction are ignored, such as who did it, where it was done, and what itproduced . But the content of this abstraction is purely that of renderingphysically heterogeneous objects commensurable . Embodied labour is thusa means of aggregation, and there is nothing to restrict its application toany particular sort of society . (8) The recognition simply of commensur-ability is insufficient to make the concept historically specific to capitalism,which is to say that embodied labour is not abstract labour . The point isnot that no abstraction is involved in the concept of embodied labour ;rather it is not a social abstraction corresponding to a particular historicalprocess, but it is arbitrary, a mental convenience: an assumption thatlabour is homogeneous when it is plainly not . Indeed it is becauseRicardo's concept of labour has no historical specificity, since it is not theproduct of an historical process of abstraction, that his theory as a wholefails to explain what is historically specific .

Ricardo's theory is, then, a model built upon assumptions rather thanthe theorisation of a real world process by means of abstraction . Assump-tions are thought-constructs which have no real existence but are inventedin order to simplify and to structure the complexity of the analysis (con-sider, for example, the assumptions of neo-classical perfect competition) .Assumptions are of course designed in order to be able to say somethingabout the world, but to the extent that they do this (Ricardo's assumption,for example, of the tendency towards equalisation of the rate of profit indifferent sectors across the economy) they do so purely by virtue of beingimposed upon empirical 'facts' in order to render appearances coherentand plausible . Hence they are expressions of surface phenomena which seein such immediate forms the whole nature of the phenomena in question .But descriptions of surface phenomena exclude the possiblity of necessary

82

CAPITAL & CLASS

contradictions as the determinants of the motion of these immediate forms .For at the level of the particularity of phenomena, appearances either areor are not contradictory with each other, and contingently so. Hence therecognition of any determinate contradictory reality is ruled out . And itfurther follows that nothing can be deduced from assumptions which isnot already entailed by those assumptions ; hence theory becomes taut-ology, the deduction of conclusions from assumptions .

By contrast, Marx's method of abstraction, his dialectical method, isprecisely concerned to show that the forms of appearance of capitalismare just that, forms of appearance . Through his critique of these forms,and their reflection in bourgeois thought, Marx shows first, that thesereflections fail to identify the social forms that they express but treatthem rather as natural phenomena (commodity fetishism) ; and second,that the forms of appearance of capitalism are always transitory in thatthey constitute barriers to their further development . This is no mysticalidealism, but a materialist recognition of the process of motion throughcontradiction which constitutes the process of human history . Such adialectic is "in its essence critical and revolutionary" ; it

"includes in its comprehension and affirmative recognition of theexisting state of things, at the same time, also, the recognition of thenegation of that state, of its inevitable breaking up ; because it regardsevery historically developed social form as in fluid movement, andtherefore takes into account its transient nature not less than itsmomentary existence." (Marx, 1938, Capital I, pp .xxx-xxxi)

The process whereby both the existence of something and its impermanencecan be captured is by the method of abstraction . Thus abstract labour isderived by an abstraction inherent in the exchange process in order tocapture the moment of the social fabric of capitalist society in its simplestform . The immanent development of the pure value-form into its specificcapital-form is immediately posited, (just as historically commodity pro-duction preceded and evolved into capitalist production), and the contra-dictions constituting the most abstract moment are elaborated throughmediation, suspension and their reappearance in successively more concretemoments. The result is that the historical reality of capitalism is grasped inthought as a part of the process of the transformation of that reality . ThusMarx describes how the method of inquiry

"has to appropriate the material in detail, to analyse its differentforms of development, to trace out their inner connection . Only afterthis work is done, can the actual movement be adequately described . Ifthis is done successfully, if the life of the subject-matter is ideallyreflected as in a mirror, then it may appear as if we had before us amere a priori construction . " ( I bid, pp .xxix-xxx. Emphasis added .)

But what the "a priori construction" will show is value as the constitutiveform of the social relations of capitalism, from which the analysis proceedsthrough a hierarchy of abstractions in order to comprehend in thought the


83

real world . Only once the basic capital/labour relation has been graspedcan

"the various forms of capital . . . approach step by step the formwhich they assume on the surface of society, in the action of differentcapitals upon one another, in competition, and in the ordinary con-sciousness of the agents of production themselves ." (Marx, 1972,Capital I11, p.26)

Thus considerable mediation is required before competition can beexplained on the basis of abstract labour . Of course,

"Competition merely expresses as real, posits as an external necessity,that which lies within the nature of capital ; competition is nothingmore than the way in which the many capitals force the inherentdeterminants of capital upon one another and upon themselves ."(Marx, 1973, Grundrisse, p .647 . Emphasis in original )

Nevertheless such mediation is essential in order to explain the forms ofappearance which are necessarily adopted by social relations in the capit-alist mode of production . For all the various phenomena arising from theprocess of competition

"seem to contradict the determination of value by labour-time asmuch as the nature of surplus-value consisting of unpaid surplus-labour. Thus everthing appears reversed in competition . The finalpattern of economic relations as seen on the surface, in their realexistence and consequently in the conceptions by which the bearersand agents of these relations seek to understand them, is very muchdifferent from, and indeed quite the reverse of their inner but con-cealed essential pattern and the conception corresponding to it ."(Marx, 1972, Capital III, p.209 emphasis in original .)

Marx's account of the formation of a general rate of profit in competition,his transformation procedure, is precisely the transition from the "innerbut concealed essential pattern" to the "pattern of economic relations asseen on the surface" . It is a movement required, a procedure of transform-ation, between abstractions of different orders of conception . What it isnot is a process of redistribution of aggregate surplus-value . It is oftenclaimed that aggregate surplus-value is redistributed such that capitals donot share in it in accordance with the amounts of money-capital ex-changed for labour-power, but rather that they share in it in accordancewith the amounts of capital exchanged for both labour-power and themeans of production ; this redistribution is supposed to occur throughdifferences between values and prices of production (9) . But there is noreal world state which exists prior to such redistribution . Of course com-petition distributes aggregate surplus-value according to total capital ad-vanced, but there is no redistribution . The process of redistribution is nota real world process, but a conceptual one which is symbolic of the theor-etical transition required between concepts of a different order . The pro-

84

CAPITAL & CLASS

cess takes place wholly in thought, for the process is one of conceiving thevalue produced by production first in abstraction from competition, andsecond by allowing for the effects of competition. This transformationbetween abstractions of different orders of conception is rendered per-haps more comprehensible without resort to parables of redistribution .

We can now bring together the threads of the discussion in order todraw some conclusions about Ricardian and Marxian value theory, and thedifferent methodologies employed . First, if one adopts a Ricardian conceptof value as embodied labour, one is trapped in irreconcileable contra-dictions . Prices are and must equal costs, therefore cost-determined"theories" of price are correct (and contentless) . But it follows that onecannot assign embodied-labour values to commodities in general in any waythat preserves a meaningful functional correspondence between embodied-labour values and equilibrium prices, or between their rates of change .However, this has long been known - it is one of the reasons why neo-classical economics is not Ricardian . That Marx made mistakes in his owntransformation procedure (to put it least charitably- is no reason to fallback on a Ricardian understanding of value . For Marx's transformationprocedure is a deduction of the forms of appearance of values which isbased on a full elaboration of the contradictions immanent in the realabstraction that is value itself. The capital-relation is a value-relation, butvalues as quantities of abstract labour can exist only in commodity-form,and commodities are purchased and sold on markets for prices . In this waythe fundamental contradiction between value and use-value finds its mostcomplete expression in the fact that it is through competition that thelaws of motion of capitalist development are expressed . For quantities ofvalue are expressions of social relations, the fact that there is a necessaryconnection between a certain article and that fraction of aggregate sociallabour-time required to produce it . It is the market that makes this con-nection, and necessarily only in distorted form . Only market processesrealise the quantitative expression of abstract labour, and this quanti-tative expression only has a price-form . As we have seen, the abstractionthat is value yields the price-form directly, but as soon as competition isaccounted for, then abstract labour cannot directly be assigned to com-moditites . I n this sense, all of Marx's numerical examples are potentially mis-leading in their assignment of numerical values to commodities - thesenumerical exercises must be seen as didactic and expository devices ratherthan analytical explanations. Otherwise, there is no complete escape fromRicardo's problemic .

It was Ricardo's failure to recognise the historical specificity of capit-alist production, and thus of the value-form, which led him to demand forhis (embodied) labour theory of value a direct bearing on the determinationof price, a direct bearing which neither his theory nor the contra-dictions of capitalism could support . The reason for the inclusion of thisapparent digression into the history of political economy, and Marx'scritique of it, is that it is very common to find modern work which whileclaiming to employ a Marxist understanding of value has not in fact ad-vanced upon Ricardo's . Indeed, we can use the contradiction between alabour-embodied theory of value and a cost-summation account of price,


85

the contradiction that beset Ricardo, to classify such work in modernpolitical economy (10) . Under the heading of "Ricardian" come thoseanalyses which remain enmeshed in Ricardo's problem and do not advancebeyond it (11) . "Sraffian" is a term that applies to those which "solve"the problem by clinging to one horn of the dilemma, abandoning altogetherthe category of value, thus repeating, over a century later, the post-Ric-ardian retreat from political to vulgar economy (12) . In spite of this dif-ference (that Sraffian analysis forms a part of vulgar economy while Ric-ardian does not) the two together have generically been called "Neo-Ricardian". Indeed they do have certain features in common . For in both,labour is neither abstract nor concrete, value (whether consequently re-jected or not) is embodied-labour, capitalism is not rendered historicallyspecific, and occasionally a sociological theory of class struggle is appendedas guarantee of "Marxist" authenticity and rectitude (as if the class strugglewere discovered by Marx, when it is plainly present in Ricardo and manyothers.) As we shall see below, such work is enmeshed in all the contra-dictions which beset Ricardo, which is not really terribly surprising .

Another school, called by some "fundamentalist ",does not recogniseRicardo's problem and its consequences (13) . While the term "fundament-alism" may be appropriate to their apparent desire to show every word ofMarx's to be correct, more importantly they miss the fundamental issue .While adopting an apparently Marxist understanding of value, and recog-nising the necessity of accounting on this basis for the appearances of thereal world, their insistence on the technical correctness of Marx's solutionto the transformation problem implicitly assumes, despite their explicitmethodological promises to the contrary, that value is a category of directapplicability as an analytic tool at successively lower levels of abstraction .Thus they want appearances and reality, if not, like Ricardo, to coincide,at least to be immediately accessible to the same tools of analysis withouttransformation . The issues of this paper then are of considerable relevanceto a critique of "fundamentalism", but rather than continue to criticise aschool whose practice is to dissolve the real world into an epiphenomenonof capital in general, we shall continue to focus on those issues directly .

On the basis of this understanding, we now turn to an examination ofcertain modern "puzzles" in value theory to demonstrate how the analysisof these within an embodied-labour framework provides no escape fromRicardo's problematic save that of .vulgar economy. The reason is simple :these puzzles are of the same genus as the transformation "problem"itself. By contrast, a Marxist analysis of them will serve to develop theunderstanding of abstract labour and price which we have establishedabove .

SECTION 4

THE MODERN ASSAULT ON VALUE

An interpretation of Marx's transformation procedure as a "problem"provides the starting-point for much of the modern criticism of Marx's

86

CAPITAL & CLASS

value theory . Such modern criticism, whose ancestry goes back at least asfar as von Bortkiewicz and Bohm Bawerk, is largely based on the work ofSraffa. As such, it is precisely subject to the strictures which Marx con-sidered applied to Ricardo's work . For the introduction of competition atthe outset of the analysis leads to the same problems ; problems which havedevastating consequences for what is interpreted to be Marx's value analysis .Indeed many would go as far as to say that

"the project of providing a materialist account of capitalist societies isdependent on Marx's value magnitude analysis only in the negativesense that continued adherence to the latter is a major fetter on thedevelopment of the former ." (Steedman, 1977, p.207) .

In its strongest statement this claim has two foundations -- eachapparently equally damaging to value analysis ; together, invincible . Thefirst is that value as a concept is internally inconsistent, the second that itis redundant. The claim is that

"it has been proved that Marx's value reasoning is often internallyinconsistent, completely failing to provide the explanations whichMarx sought for certain central features of the capitalist economy . Bycontrast, these same features can be given a coherent explanation interms which make no reference whatsoever to any value magnitude ."(Ibid, pp.206-7)

i) The redundancy of valuesWe will take the charge of redundancy first . To avoid any misunder-

standing, we will read the charge, from the most recently published workof one of the least circumspect of the accusers, Steedman :

"If there is only one available method for the production of eachcommodity - and thus the value of a commodity - is determined_only one product then :i) the physical quantities of commodities and of labour specifying the

methods of production, together with the physical quantities ofcommodities specifying the given real wage rate, suffice to deter-mine the rate of profit (and the associated prices of production) ;

ii) the labour-time required (directly and indirectly) to produce anycommodity - and thus the value of any commodity - is determinedby the physical data relating to methods of production ; it followsthat value magnitudes are, at best, redundant in the determinationof the rate of profit (and prices of production) ;

iii) . . . The traditional value schema, in which all the constant capitaland all the variable capital elements in a productive activity aresummed and represented by a single "C" and a single "V" figure, isnot adequate to the determination of the rate of profit (and pricesof production .)" (Ibid, pp.202-3) (14)

To summarise, the rate of profit, prices and values can all be calculatedonce the methods of production and the wage are specified ; on the other

THE ANOMA LIES OF CA PI TA L

87

hand, none of these other variables can be calculated from a knowledge ofvalues alone. Thus values are but statistics derived from the methods ofproduction, a compression of the full data, and a compression that losesmuch that is of importance .

Now the Sraffians are in their own terms clearly correct . A full spec-ification of physical input-output requirements does give sufficient inform-ation to calculate exchange ratios and rates of profit . (15) Certainly no lesscan be demanded of a cost-determined "theory" of price . But further,such a full specification also gives sufficient information to calculateamounts of labour embodied, were such information required . But in nosense is it, since knowledge of the amounts of labour embodied is notsufficient to calculate exchange ratios and rates of profit . This for themclinches the argument that "values" are redundant: they are derived fromthe phsycial input-output specification, not vice versa . They cannot beused by themselves to derive exchange ratios and rates of profit, whileanyway the physical input-output specification can do that directly .

It should of course be clear that the "value" being talked about here istotal labour embodied, the Ricardian concept . For how else could (as in(ii) above) the value of a commodity be determined by the physical datarelating to methods of production rather than vice versa? Why else wouldone elaborate the determination of value rather than the determination byvalue? The former means the mere functional determination of onequantity (values) by other quantities (input coefficients and labour inputvectors) . The latter encapsulates the method whereby the relations of com-modity production are such that input coefficients and labour input vec-tors can be specified . For it is only through the exchange of products thatindividual labours are commensurated and socially necessary labour timesare established . And this is critical . For what is being counterposed here ison the one hand an understanding of values as mere derivates of physcialquantities required for production, and on the other hand an under-standing of the social quantification of production requirements positedon the value abstraction .

How then has Steedman advanced on Ricardo? For as we have seen,Ricardo recognised the lack of functional dependence of prices and therate of profit on embodied labour times . The point is that Steedman,following Sraffa, has solved this problem of Ricardo's by showing us uponwhich variables prices and the rate of profit do functionally depend . Indoing so, however Sraffa and Steedman are treading a well-worn path -the path that leads to vulgar economy and a cost "theory" of price . Insolving Ricardo's problem, they have simply stated a series of tautologies,for that, after all, is all a proof of functional dependence can be . Haltedby the contradiction between a labour-embodied theory of value and a cost-summation account of price, Sraffians are thus led to jetison the former asredundant .

While Steedman emphasises the logical purity of his conclusions, it is infact in Sraffa's own work (Sraffa 1960) that the argument is clearest. ForSraffa never saw his task as specifically one of criticism of Marx, and thushe found it unnecessary either to introduce a concept of value as distinct

88

CAPITAL & CLASS

from price or even to consider value as labour-embodied . What Sraffa didwas to show that with a full specification of the input-output requirementsof each commodity (i .e . a list of the amounts of all commodities, includinglabour-power, required to produce a given commodity) then the price ofeach commodity can be calculated as a sum of its costs of production to-gether with the profit accruing to the capitalist on whose account the pro-duction of the commodity is taking place . It is important to recognise herethen that Sraffa's famous dated-labour analysis, in which the price of acommodity is broken down into the sum of labour-input costs incurred inall periods from the present backwards, is just precisely that : an expressionof price in terms of costs . Sraffa therefore implicitly rejects any theory ofvalue as a property of the labour employed in the production of a com-modity (whether "embodied" or "abstract") in favour of an account ofprice as determined by costs of production . That the coefficients involvedin the dated-labour equation are quantities of labour embodied in no senserelates to Ricardo's labour-embodied theory of value and in no sense re-moves Sraffa from the school of vulgar economy . For Sraffa, and forvulgar economy, prices are made up of costs and costs, of course, are butprices . (16)

Sraffian economics is thus clearly different from Ricardian . It istalking about the physcial product and its division between classes insteadof the total labour of society and its division . Nevertheless, both Sraffiansand Ricardians work on the assumption that the specification of inputrequirements (whether in terms of physical inputs of particular use-valuesone. This does not mean, as neo-Ricardians would claim, that the specific-ation of input requirements is simply determined by the class struggle andby profit maximisation . What it rather means is that the specificationof-what is produced (the composition of output) and the techniquesby which it is produced (the technical coefficients of production) is mean-ingless in abstraction from the way in which the labour process is organisedand from the way in which production, as a social activity through themarket's universal commensuration of what is produced, determines bothwhat it produced and how it is produced . Hence neither of the latter canbe taken to be a given .

ii) The inconsistency of valuesThe other criticism that the Sraffians level against Marx's value

theory is that there are internal inconsistences in the concept of valueitself. The argument can be dressed in more or less technical clothing. Herewe shall give it in as uncomplicated a way as we can ; and will show whythe sheep's clothing hides nothing but a sheep after all . Not surprisingly thewolf which appears to some finally to have gobbled Marx will be revealedto have had a most Ricardian supper.

The claimed inconsistency in the concept arises when establishedmethods of calculating values give either indeterminate or negative answers .The former can arise when there is a choice between two equally profitabletechniques of producing the same commodity, and the latter when twodifferent commodities are the simultaneous result of the same productionprocess and a question arises as to how to allocate the total embodiedlabour time between them . Basically, both problems arise from the same

THE ANOMA L IES OF CA PI TA L 89

source : capitalists, who control what is produced and how, make theirdecisions on the basis of maximising their profit, which, for theproduction of a given commodity means minimising their costs . But costsare prices, not values, and the minimisation of one may not, and in generalwill not, be the minimisation of the other and thus anomalies may arise .

Let us take the two cases in turn . The first arises when there are twomethods of production of the same commodity, that are (in price terms)equally cheap and cheaper than all others for producing that commodity .These methods are clearly the only ones that will be used by efficientcapitalists, but because there is nothing to choose between them on thecriterion of profit maximisation, some capitalist will use one methodand some the other . Now there is no reason why these two methodsshould use the same quantities of total embodied labour . Where such achoice of techniques exist, which method of production determines thevalue of this commodity?

The second anomaly arises when we have joint production of twocommodities . Joint production is considered to be an important problembecause this is the standard way of incorporating fixed capital into Sraffiananalysis. The inputs to such a production process are a machine of a cer-tain age (thus productive capacity), other means of production, and labour ;the result is a commodity that will be sold plus a slightly more usedmachine which will be employed in the production process next period .The machine will continue to be used each period until, because it is nolonger profitable to use it, it is scrapped . This way of looking at productionprocesses involving fixed capital, together with the existence of clear realcases of joint production,points to the problem of how in ajoint productionprocess to allocate the embodied labour between the two commoditiesproduced in order to determine the values of each commodity (17) .

If some other process exists and is in use for producing one of thecommodities alone, we could determine the labour embodied in thatcommodity from this non-joint process and determine the labour em-bodied in the other commodity by subtraction from the total quantityof labour embodied in the joint process . But the total labour embodiedin the joint process may be less than that in the non-joint one, leaving - anegative "values" for the second commodity after subtraction . It is possiblethat a process which used more labour to produce only one product thananother did to produce two would remain in use because, as we have seen,capitalists decide which production method to use on the basis of profitcalculations based on prices, not on the basis of embodied labour calcul-ations, and the former process might be as profitable as the latter thoughclearly more "wasteful" of embodied labour. This anomaly thus arises forthe same fundamental reason as the other, that capitalists make their dec-isions on the basis of prices rather than on the basis of embodied labourvalues. The problem however appears in a different form : not that "values"may be merely indeterminate, but that they may even be negative .

Various solutions have been suggested to these problems, primarily bythose who want to rescue the Ricardian concept of value, but all solutionsproposed involve some modification of the concept .

One approach (Armstrong, Glyn and Harrison 1978) to the problem

90

CAPITAL & CLASS

of the indeterminacy of the "value" of a commodity which has twomethods currently in use for its production is to take the value of the com-modity to be the average "actual labour-time" embodied in it . By "actuallabour-time" is meant the total labour time spent on production of thatcommodity divided by the quantity produced of the commodity. Thiscertainly gives an answer, given the required information, but the requiredinformation is much greater and specifically more concrete than thatpreviously required . It begs the question of why "labour embodied" is aworthwhile quantity to calculate even more strongly than before, if tocalculate it requires a full specification not only of methods of productionbut also of the extent of their use . It is now a purely descriptive statistic,stripped of any analytic content ; and in any case such an approach isforced to concede the indeterminacy of embodied labour values of individ-ual commodities in the case of joint production, the other anomaly . Thusthe price of such a rescue attempt is the reduction of Ricardo's analyticprocedure to the empirical inspection of actual labour times in cases inwhich it is hoped that joint production is not significant .

A considerably more robust approach is that of Morishima 1973,1974 . Following von Neumann, he reformulates his theory of value interms of inequalities rather than equations . This means that the value of acommodity is the minimum total labour-time required to produce a netoutput that includes one unit of the commodity . So in the case of twoequally profitable methods of producing the same commodity, its value isgiven by the method that uses least labour-time . Joint production doesnot affect this ; if x units of total labour time are needed to produce com-modities A and B jointly and there is no other way of producing com-modity A, then x is the minimum total labour needed to produce A andthus the value of A . If there are other methods of producing A, each willrequire a certain amount of labour, and then the value of commodity Awill be the smallest of these requirements . This way clearly avoids theproblem of indeterminate or negative values, for every process uses a det-erminate positive amount of labour . The value of the commodity is justthe smallest of these determinate, positive amounts. On the other hand, adifferent problem arises, that of non-additivity of values . The value of twocommodities jointly produced, such as A and B above, will not in generalequal the sum of their individual values . Indeed, in the case above, if com-modity B also cannot be produced by an alternative method, the values ofcommodity A alone, of commodity B alone and of a bundle consisting ofA and B together will all be the same. This non-additivity poses problemsin calculating such variables as the rate of exploitation, since the sum ofnecessary and surplus labours may not equal total labour performed . Never-theless it does capture the labour-embodied concept of value and does notlead to nonsensical results . In so far as this is Morishima's purpose he hasclearly and elegantly succeeded .

To summarise then, the attack on value as both a redundant conceptand an inconsistent concept is well-founded within a Ricardian framework .Attempts to salvage the concept of embodied labour from the attacks onits consistency by some modification of the concept must fail, on the samecount as Ricardo, to provide a basis for a theory of price . Yet within this


91

framework, the calculation of embodied labour times, in whatever form,has no other justification . The problem of redundancy - already presentin Ricardo's own work - is thereby rendered even more acute .

SECTION 5

REDUNDANCY, INCONSISTENCY, AND MARX'S THEORY OF VALUE

In this section we return to Marx's theory of value in order to examinewhether the charges of redundancy and inconsistency have any validitywhen directed against the category abstract labour . As a preliminary wecan note that the answer to the charge of redundancy has already beenmade ; value theory is not redundant in Marxism because it is necessary tothe very specification of the production relations of capitalism . The redun-dancy of values cannot therefore be "proved", were it to be possible, byshowing that values can be derived from a specification of productionconditions, by input-output data . For the law of value itself is the processby which those production conditions are such that they can be technol-ogically represented by an input-output specification at all . It is valueanalysis that reveals what is historically specific to the capitalist mode ofproduction, and the burden of our argument has been to show that valueshave a real, social existence . To render them an arbitrary piece of theor-etical baggage, to be taken on board and jetisoned at will, is precisely tothink of values as a mental construct, an assumption from which to makedeductions to be compared with the facts, rather than the necessary resultof a real process of abstraction . The reality of values stands or falls withthe reality of capitalism . Dispense with the one and you dispense with theother. While as a political programme this may be admirable, as a theor-isation of current reality it leaves something to be desired . Occam's razorcannot therefore be used to excise the category of abstract labour ; thecharge of redundancy must be rejected .

The second charge is that of inconsistency . We have seen that thecalculation of quantities of embodied labour involves insuperable dif-ficulties unless the concept be redefined in a programming framework, asMorishima. The question is whether any of the same difficulties are in-volved in the calculation of quantities of abstract labour, in the calcul-ation, that is, of socially necessary labour times .

To answer this question we must reexamine the separation of ab-stract from concrete labour as it applies to the anomalous cases describedin the last section . In these, the problem arose either because one com-modity was being produced by two different production processes orbecause two different commodiites were being produced by one produc-tion process . Now two commodities can differ qualitatively only in theiruse-values . Therefore the identity of the commodities produced by twoproduction processes or the difference of two commodities produced byone, in the anomalous cases, is at first just a question of use-value .

Marx used the differences between use-values to differentiate con-crete labours . Provided we consider only use-values that are produced

92

CAPITAL & CLASS

singly with only one technique in use at any point in time, this allows aone-to-one correspondence between use-values and concrete labours . Theproblem with the anomalous cases is that this one-to-one correspondencebreaks down . For joint production involves one type of concrete labour toproduce more than one use-value, and different techniques are differentconcrete labours producing the same use-value .

These peculiarities arising from the use-value aspect of the commoditywould be of no significance for the value aspect were it not that the dif-ference or identity of use-values is manifested in their exchange . For thedifferent use-values produced by a joint production process will be sold attheir own, in general different, prices . Conversely, if two different pro-duction processes produce identical use-values, they must be sold at thesame price . But value is the category which links production to exchangein the single product single technique case. In the anomalous cases, theone-to-one correspondence between production processes and the processof exchange of their products breaks down . For since concrete labourscannot uniquely be allocated to use-values, the abstraction from concretelabours to abstract labour also performs no unique determination. Abstractlabour is still being performed in each of the production processes; that itcannot be allocated to the specific use-values produced merely implies thatthe "commodity" cannot now be both the result of the production processand that which is exchanged on the market . The link between the two isvalue of course: value is produced in the labour-process, but only has aprice-form. However, the specification of that linkage is always a contra-dictory one, counter-posing value as an attribute of what is produced,reflected in the commodity, with its manifestation in the commodity-form and more specifically upon its exchange.

This relates back to the necessity to derive the contradiction be-tween value and its form, exchange-value, from the unfolding of the valueabstraction itself. For until this unfolding leads to the level of abstractionat which capitalist competition is introduced, there is no need for the dis-tinction between values and prices of production, and the anomalous casesdisappear. At this level, the market allocation of labour to different pro-duction processes is not in conflict with the law of value ; cost minimisationis value minimisation. Therefore any individual commodity produced willhave a determinate, positive value. For at no level would commodities beproduced unless their exchange-values were each positive and at this levelthis occurs if and only if their values are each positive . (18)

But the development of the commodity-form through exchange entailscapitalist competition. As soon as this is taken into explicit account, pricesof production must in general differ from values . So we can retain valuesas attributes of production processes, but cannot allocate values to indiv-idual commodities produced in joint production processes, not a uniquevalue to the indistinguishable product of two different production proc-esses. Prices of production remain but the expression of value . However,and this is the crucial point, they are the only expression of value. Thusthe uniqueness and separability of the former is essential, while that of thelatter is not .


93

Marx himself, of course, did not consider these problems specifically .But our treatment here is quite consistent with Marx's own elaboration ofhis categories, albeit in different contexts. Thus when he considers theresults of the production process as a mass of commodities whereby thetotal capital advanced reproduces itself together with a surplus-value, heremarks that,

"The labour expended on each commodity can no longer be cal-ulated - except as an average, i .e . an ideal estimate . The calculationbegins with that portion of the constant capital which only enters intothe value of the total product in so far as it is used up ; it continueswith the conditions of production that are consumed communally,and ends with the direct social contribution of many co-operatingindividuals whose labour is averaged out . This labour, then, is rec-koned ideally as an aliquot part of the total labour expended on it .When determining the price of an individual article it appears as amerely ideal fraction of the total product in which capital reproducesitself." (Marx, 1976b, Capital I, p .954 . Emphasis in original)

Individual values become "ideal estimates ",fractions of the total, whichmay or may not be realised in the price-form depending on the quantitiesof the commodities actually sold . And in his discussion of the time-chit-ters, he writes :

"Every moment, in calculating, accounting etc ., that we transformcommodities into value symbols, we fix them as mere exchange values,making abstraction from the matter they are composed of and alltheir natural qualities. On paper, in the head, this metamorphosisproceeds by means of mere abstraction; but in the real exchangeprocess a real mediation is required, a means to accomplish this ab-straction ." (Marx, 1973, Grundrisse, p.142. Emphasis in original)

Such a real mediation is the real contradiction underlying all theproblems discussed in this paper. The anomalous cases discussed are nomore anomalous than the solution of the transformation "problem", forin each case the anomaly arises from the contradiction between the fund-amental concept of socially necessary labour time and the development ofits full consequences in capitalist competition . For the former has implicitwithin it the operation of the law of value, the minimisation of values,while the latter is expressed through the formation of the general rate ofprofit, and the minimisation of costs, i .e . prices . It is this contradiction it-self which renders the assignment of numerical values to commodities im-possible as soon as explicit account is taken of competition . It is thiscontradiction itself which renders an understanding of the transformationproblem as a real world procedure of the redistribution of surplus value,mistaken ; what is required is the dialectical development of categories inorder to appropriate a real contradictory world in thought .

It follows then that because the law of value operates through thedistorted form of capitalist competition, the capital which sets in motion

94

CAPITAL & CLASS

some production processes which are "wasteful" of total social labour maystill be validated by that competition, and hence produce a portion of thetotal surplus value . Such "wasteful" processes are those in which the timespent in the production of a given use-value is longer than that spent inthe production of that same use-value by another process (one anomalouscase); or those which produce one use-value where another can producethe same use-value plus something extra for the same or less expenditureof human labour-power (the other anomalous case) ; or those whose con-tinued use is validated by the market in competitive capitalist conditionsrather than by the "pure" operation of the law of value . The last, of course,is the general case of capitalist production, which is itself the ultimateanomaly .

CONCLUSION

In this paper we have attempted to explain and extend Marx's valuetheory. In order to provide a framework in which much of the debates onthe nature and significance of value theory can be situated, we made adistinction between a Ricardian embodied-labour theory of value and theMarxian theory of value based on the category of abstract labour . Whilethe former is intended immediately to be a theory of price, the latter is soonly after several mediations . These mediations are critical to the fund-amental difference in method between the two theories .

Ricardo made no distinction between value and exchange-value,demanding unsuccessfully of his theory that it be an immediate explan-ation of surface phenomena . He failed because capitalist relations are in-herently contradictory, and the forms they take constitute a set of appear-ances that do contradict their fundamental determinations . Two responseswere possible to Ricardo's failure . One, that of vulgar economy, was toabandon the search for explanation, seeing appearances as the wholephenomena in question, and limiting the scape of any particular theory bythe requirements of internal consistency . The other, that of Marx, was torecognise capitalist reality as contradictory, and to attempt to structuretheory in such a way as to elaborate within it the contradictions of thecapitalist world . To do this, he used a method of abstraction which recog-nised the historical specificity of the capitalist mode of production andhence its transience . We have shown how this differs from a method ofbuilding theory on assumptions, a method common to the defeatedRicardo and to vulgar economy . And by this method of abstracting valueas that which is specific to capitalist products, Marx not only had a just-ification for the distinction between value and exchange-value, but couldderive the very necessity for that distinction from the category of valueitself.

Hence Marx's transformation procedure is not, as it would be forRicardo, an attempt to correct an unfortunate disjuncture between an em-bodied-labour theory of value and the requirements that the equalisationof the rate of profit makes of prices . Rather, that disjuncture is recognisedas the necessarily contradictory link between value, as the explanation of

THE ANOMA L IES OF CAPITAL

95

capitalist production relations, and its expression as exchange-value inprices . Hence it is not surprising that, when competition is accounted for,the one-to-one relationship of values to exchange-values disappears .

We then considered certain modern objections to values, and showedthat when brought against Ricardo (often addressed under the pseudonymof Marx) they have validity . But the substance of the objections is notnew, in the sense that they are objections to the results of the distinctionof exchange-value from value . For this reason, the same objections have norelevance to Marx's theory of value, for the contradictions between valueand price, and the anomalies that thereby arise, are explained within thetheory itself as consequences of the recognition of value as an attribute ofthe production of commodities which is only validated by their exchange .Again, it should not be surprising that, when capitalist competition isaccounted for, the one-to-one relationship of values to exchange-valuesdisappears .

The anomalies are therefore all of the same sort . They arise out of theelaboration of the contradiction between value and use-value in order toaccount for the phenomena of capitalist competiton . We have shownhow this contradiction is contained in the distinction between abstractlabour (which produces value) and concrete labour (which produces use-value) . The former is commensurated through the exchange of com-modities . But there is a second commensuration in capitalism : the equal-isation of the rate of profit consequent upon the purchase and consump-tion of labour-power in definite amounts of time . The commensuration ofthese quantities of times through the equalisation of the rate of profitcontradicts the commensuration of socially necessary labour times throughcommodity exchange . The burden of this paper has thus been to show thatboth commensurations are fundamental to an account of capitalism whichreproduces the contradictions of capitalist reality .

For fundamentalists, values are sacred ; the commensuration of paidlabour times, while recognised, is allowed no effect on socially necessarylabour times. For Sraffians, values are irrelevent ; the only commensurationis that of capitalist competition, and production is a black box technologyof input-output coefficients . For Ricardians, there is no realisation ofRicardo's contradiction ; commensuration is a muddled one of actuallabour times, and analysis cannot be developed since contradiction cannotbe contained within Ricardo's framework .

What this paper has attempted is an approach which develops the con-tradiction between use-value and value such that the contradictory com-mensurations of capitalism are integrated within a methodology of histor-ical materialism. But this is only a beginning in the understanding of thecomplex reality of the capitalist mode of production .

APPENDIX

We saw in Section 2 that the formation of a general rate of profit isincompatible with the exchange of commodities at their values . Put an-other way, the amount of profit accruing to each capital must in general

96

CAPITAL & CLASS

differ from the amount of surplus-value produced for that capital . Ourconcern here is to capture the existence of surplus value in its more con-crete individuated form of profit . This capturing of something in a moreconcrete form could in general be called a process of transformation . Theparticular transformation at issue here is that which displays surplus-valueas profit and, in so doing, reveals the contradiction between value and itsforms, exchange value (expressed as price) .

Now this transformation has a quantitative dimension . Values, ex-change-values and prices are all quantities, as are rates of exploitation andof profit . If there are use-values to which both value and price can beassigned, then we should be able to write down the relation between thesetwo quantities. Similarly, if consequently rates of exploitation and ofprofit can be defined for the same economy, we should be able to expressthe quantitative transformation of the one into the other in algebraicform. As we have seen in Section 3, these transformations of quantities arenot the whole content of the transformation procedure, for they only en-capsulate one aspect of the procedure : that concerned with quantities . Never-theless, given the above proviso (and Section 5 shows why it is a meaningfulproviso), we can consider a quantitative transformation procedure .

This for Marx is one which accounts for "prices of production" asdifferent from values, the former being defined to equal total capitaladvanced plus an amount of profit proportional to that capital ; that pro-portion being given by the rate of profit (in value terms) for the economyas a whole. Then capitals with higher than average value-composition willsell their commodities at prices of production greater than their values,and conversely for those with lower than average value-composition . Al-gebraically, Marx's solution is to let the average rate of profit be R, where

R= SC+V

where S, C and V are now defined for total social capital .That is, summing over all commodities,

S = -i si, C = 1~ i ci, V =

i vi .

Then for an individual commodity, its price (of production) is given by

wi = ci + vi + R(ci+ vi) _ (ci + vi) (I + R)compared with the value wi of that commodity,

wi=ci+vi+siCertain properties of Marx's solution should be noted at this point . It evi-dently follows that total price of production equals total value :

Y- i -'i = Ei withat total profit equals total surplus-value :

R( 2. i[ci +Yi vi) = 2~iCsi,and that R is the average rate of profit in the sense of an average of allindividual value rates of profit, each one weighted by the value of thecapital advanced :

R = ~i (si/ ci + vi) (bi+ vi)C+V

Two related objections have been raised to Marx's procedure . Thosewho raise these criticism focus on the inadequacy of Marx's formula

THE ANOMA L I ES OF CAPITAL

97

for the determination of prices of production, and generally proceed fromsuch weaknesses to question his value theory as such . We have been con-cerned with the latter in the main body of this paper, and it is to the pro-cedural criticisms in their own right that we now turn .

First, in the equations, prices of production are calculated on thebasis of the means of production and labour-power consumed beingassessed as capital in value terms. This clearly means that prices of pro-duction cannot be any representation of exchange-value ; otherwise wewould have the absurd situation that the money paid for a commodityconsumed in production by its purchaser is different from the moneyreceived by its seller . There is plenty of textual evidence that Marx wasaware of this (Marx, 1972, Capital III, pp .160, 164-5, 206-7), and that toaccount fully for the formation of a general rate of profit, not onlyoutput values but also input values require transformation . Second,Marx's procedure is indicted for inconsistency. The rate of profit measuredin value terms is used to prove the necessity for deviations of prices ofproduction from values . But such deviations imply that the ratio of theaggregates of total surplus-value extracted to total value advanced ascapital, cannot in fact be taken to be the rate of profit, since the point ofthe procedure is precisely to transform value magnitudes in order to derivethe existence of an average rate of profit .

Both indictments of Marx's procedure are clearly well-founded, andto render Marx's account both complete and consistent is straightforward .Many different formulations have been proposed, all correct solutionsbeing formally identical . Seton's starting point (Seton 1957) however isclosest to that of Marx, and so we reproduce it here . Algebraically, letcij = value of commodity j transferred to commodity i in order to produceone unit of commodity i ; and vij = value of commodity j, necessary toreproduce the living labourer, required in order to produce one unit ofcommodity i . (i,j = 1,2N, where there are N commodities .)

Hence, adding necessary subscripts, we get_ Zcij and vi = Evij

J

jand so

wi = E(cij + vij) +si1

Let xi = wi/wi be the transformation factor which transforms the value ofcommodity i . Then

W, = xiwi = xj (ij + vij) (1 + t)where r is the (price)rate of profit on an individual capital, i .e . the generalrate of profit, and is to be found from these equations .

These N equations can be solved for the N + 1 variables, x1, x2, . . . .x Nand F. There will be up to N possible solutions for the rate of profit, butonly one of these will give a meaningful (i .e . non-negative) solution for theset of relative "prices of production ." (19)

Relative to Seton's "solution", Marx's procedure can be regarded asmerely incomplete . We can take Marx's procedure as an algorithm at eachstage of which thus far transformed values for outputs are used to valueinputs, to calculate the rate of profit and thus obtain a further transform-ation of the value of outputs . Provided that this iteration converges, the

98

CAPITAL & CLASS

limit of the sequence of transformed values and transformed rates of profitwill be identical to that obtained from the solution to the full set of sim-ultaneous equations given above . Indeed the evidence used to support thethesis that Marx knew he should have transformed input values couldequally well be adduced to support the idea of an iterative procedure .(Marx, 1974, Capital III, pp .164-5 especially ) (20)

It is important to note that this procedure is not the same as anyhistorical transformation procedure (21) . The buying and selling of labour-power is a later historical development than that of many other com-modities. It therefore makes sense to talk of the commensuration ofsocially necessary labour times, that takes place through the exchange ofcommodities, occurring before the other commensuration that occursthrough the equalisation of the rate of profit. But, and this is critical, it isonly through the development of the wage-labour relation, which sim-ultaneously brings into play this latter commensuration, that socially nec-essary labour times are brought into any equivalence with hours of theclock . In the abstraction of simple commodity exchange, there is no reasonwhy the time that a tailor takes to produce that which is exchanged withthe produce of one hour's labour of a carpenter, should be one hourmeasured by the clock . As socially necessary labour times these are equiva-lent. But clock-time is only introduced with the wage-labour relation whenlabour-power is sold for specified amounts of time . Thus while it is truethat, historically prior to the development of capitalist competition, com-modities exchange at their values which are measured in units of sociallynecessary labour time, the development of capitalist competition not onlyintroduces a deviation of prices from socially necessary labour times, butalso gives those socially necessary labour times an independent quantitativeaspect that did not exist before. This process is certainly not captured inany iteration procedure like that above .

Related, but not identical, to the historical transformation procedure isthe idea that the transformation procedure is a process of redistribution . (9)This is very much the way in which Marx alludes to the process, but wefeel that such allusions are mistaken . Surplus-value is not redistributedbetween capitals so as to equalise the rate of profit because there is no statefrom which this redistribution occurs. At no stage in the circuit of capitalis surplus value attributed to capitals proportional to the labour-powerthey consume. A parable of the sale of the commodity leading to a re-distribution until each capital's share of surplus-value is proportional tototal capital advanced, is as misleading as parables of a mere redistributionthrough history . Redistribution is only meaningful if one can specify astate from which redistribution occurs and a state after such a redistri-bution .

Both parables are misleading therefore because they spuriously at-tribute an identity of substance to surplus-value and profit . Surplus-valueand profit are one and the same thing in that the one is the form of theother. But to proceed in analysis from surplus-value to profit, rather moreis involved than mere redistribution : that, after all, is the burden of themain body of our paper .

Hence, in the corrected solution, that total value does not equal total

THE ANOMA L I ES OF CAPITAL

99

"price of production", that total surplus-value does not equal total profit,and that Marx's average value rate of profit R does not equal the price rateof profit T, is irrelevant . The insistence of fundamentalists that theseequalities must hold shows that their understanding of the transformationprocedure is as a mere quantitative procedure . They do not recognise thatthe movement from surplus value to profit, from values to prices, is a totaltransformation that cannot be expressed as a quantitative redistribution .That Marx chose to stress these aspects of his solution only shows howeasy it is to fall into the redistribution trap .

Finally, as regards the algebra, we can note that the solution to thefull set of simultaneous equations yields only relative "prices of produc-tion". If we are interested in money prices, and, as we have seen, theseare the only phenomenal expression of exchange-values, then the exchange-value of all other commodities must be expressed in terms of the money-commodity . (22) Thus we must specify the price of a unit of the money-commodity to be one . (23) In so doing, we have changed our units ofmeasurement from those of socially necessary labour time to those of themoney-commodity. Labour time is expressed in terms of gold, and thisstage in the construction of the world of appearances is now completed .

NOTES

Susan Himmelweit teaches Economics at Birkbeck College, Universityof London. Simon Mohun teaches Economics at Queen Mary College,University of London .

We would like to thank the encouragingly large number of CSEcomrades from the UK, the USA and Holland who took the time andtrouble to read earlier drafts of this paper and to send us commentsupon them. We would like to emphasise that we do not see this draftas a definitive statement, and that we hope that its publication willencourage wider debate .

1 Smith's discussion is in Smith 1904, ch .6. Marx's appreciation ofSmith is in Marx 1968, Theories of Surplus Value 11, p.165, and hisdiscussion of the "trinity formula" is in Marx 1972, Capital 111,Part VII .

2 The point here is that a tautology is not a theory (though the state-ment of a tautology may well have ideological content) . Prices are asum of costs, but this is not an explanation unless we have some in-dependent explanation of costs . There are different ways of evadingthe need to explain costs . Smith appeared at times not to try at all .Sraffa takes one cost (the wage, or the rate of profit) as given, derivesthe remaining ones from this, and sums to derive prices . The un-explained part is clearly the initial "given" cost . Neoclassical eco-nomics derives costs from initial endowments and individual prefer-

100

CAPITAL & CLASS

ences, which are themselves unexplained . (Indeed, the latter are onlyideological constructs .) Sraffians sometimes use class-struggle in thesame way. All such theories do is to reflect upon appearances ratherthan attempt to explain those appearances ; for this reason they can beclassified as vulgar . Within historical materialism, labouring activityprovides the basis for explanation . So from this perspective onlyRicardo and Marx escape from this classification because their theoriesof value have an independent basis in the production process . It wouldnot be inconceivable to base an explanation of value upon remuner-ation, but remuneration itself would then have to be explained onsuch an independent basis .

3 This formulation of Ricardo's contradiction is to be found in Fine1975, 1977 .

4 While we will follow Marx's own elaboration only up to a certainpoint (our aim is not to reproduce the three volumes of Capital) itshould be emphasised that the starting-point can only be justified tothe extent that the appearances of capitalism can be understood andthereby reproduced in thought by such an elaboration . This latter isnot, however, a purely passive process ; the end of the analysis of cap-italism is the removal of this contradictory reality, (the revolutionaryend of capitalism), and it is in these terms that the starting-point mustbe assessed .

5

Marx's identification of the problem with respect to different turnovertimes seems to take us straight back to Ricardo . But Ricardo's capital,as money laid out on inputs, can only vary in the length of time forwhich it is tied up - hence the distinctions made between fixed andcirculating capital, and within fixed capital, purely on the basis ofthe time for which such money is tied up . For Ricardo, it is only

"a matter of transferring given, advanced values to the productand of their replacement by the sale of the product . The differencenow depends only on whether the transfer of value, and conse-quently the replacement of the value, takes place piecemeal andgradually, or in bulk . By this means the distinction between thevariable and constant capital, which decides everything,is blottedout, hence the whole secret of the production of surplus-value andof capitalist production, the circumstances which transform cer-tain values and the things in which they present themselves ascapital, are obliterated . All constituent parts of capital are thendistinguished merely by their mode of circulation (and, of course,circulation of commodities concerns itself solely with alreadyexisting given values) . . ."(Marx, 1970, Capital 11, pp.222-3)

If we are concerned with market phenomena "rather than the internalmachinery of the capitalist process of production" (ibid . p.220),then different turnover times and different value compositions areequally important. But Marx focuses on the latter since it is the pro-duction of surplus value that is critical, not the manner in whichparticular values are transferred . Emphasis on the latter merely

"brings to completion the fetishism peculiar to bourgeois PoliticalEconomy, the fetishism which metamorphoses the social, economic


101

character impressed on things in the process of social productioninto a natural character stemming from the material nature ofthose things." (Ibid . p .229)

6

Marx justified this assumption by claiming that"although the equalising of wages and working-days, and therebyof the rates of surplus-value, among different spheres of produc-tion . . . is checked by all kinds of local obstacles, it is neverthelesstaking place more and more with the advance of capitalist productionand the subordination of all economic conditions to this mode ofproduction. The study of such frictions, while important to anyspecial work on wages, may be dispensed with as incidental andirrelevant in a general analysis of capitalist production . In a generalanalysis of this kind it is usually always assumed that the actualconditions correspond to their conception, or, what is the same,that actual conditions are represented only to the extent that theyare typical of their own general case ." (Marx, 1972, Capital IIIpp.] 42-3)

Whether or not we accept this as an accelerating tendency of capitalistdevelopment, it is a valid assumption . Even if there do exist differencesin rates of exploitation, we have no reason to assume these to bebalanced by compensating differences in value compositions . (Indeedthere is no reason to make any assumption about value compositions .)The value rate of profit on individual capitals will still not be equalised .

7 As long as these categories define the frame of reference of the anal-ysis, then in so far as political conclusions can be deduced from theemployment of these categories, such conclusions will tend away fromthe advocacy of revolutionary change . Thus there resulted, in theearly stages of capitalism, an imperative to generalise what was alreadyimplicit within the status quo (thus Ricardo's advocacy of the inter-ests of industrial capitalism against the interests of the landlords),and, in the period of capitalism's maturity, a teleological justificationof the status quo (thus the concentration of neoclassical economicsupon equilibrium analysis) . Alternatively, critical theories withinthis framework can only reject the status quo on the basis of an arb-itrary ethical principle (thus petit-bourgeois or utopian socialism) .

8 !t might be claimed that a restriction is imposed, since the existenceof a general rate of profit ensures that it is capitalism which is beingtheorised. While that is true for Ricardo's theory of price, it saysnothing which restricts the use of embodied-labour to capitalism .It might further be claimed that a restriction to certain societies isimplied by the commensurability of labour times; that this com-mensurability applies only to those societies in which labour is mobile .This clearly does not restrict the use of the concept of embodied-labour to capitalism, for slave or communist societies would satis-this condition too .

9

For a clear exposition of such a parable see Baumol 1974 .10 The following footnotes classify the more prominent advocates of

each tendency we identify . But there is so much confusion in theliterature that many writers adopt vacillating and obscure positions ;

102

CAPITAL & CLASS

we have omitted these. It should also be clear that there are some fewauthors who recognise the problems that we have outlined. Amongsuch authors are Colletti 1972, Fine 1975, 1977, Fine and Harris1976 and 1977 Gerstein 1976, and Shaikh 1977 .

11 Examples are (implicitly) Glyn and Sutcliffe 1972, Harrison 1973a,1973b, Gough 1973, 1975, and Armstrong, Glyn and Harrison 1978 .

12 Examples are Samuelson 1971, Hodgson 1973, 1976, and Steedman1973, 1975, 1977 .

13 The term is used by Fine and Harris 1976 . Examples are Yaffe 1973,1975, Bullock and Yaffe 1975, Howell 1975, Williams 1975, andMurray 1977 .

14 The point of these provisos at the beginning of the charge is to look atthe simplest and most unproblematic cases. If there is only one avail-able method for the production of each commodity, each methodusing only circulating capital and producing only one product, the casesthat give rise to the "puzzles" which lead to the charge of inconsist-ency cannot arise .

15 Formally their equations can be shown to be equivalent to Seton'ssimultaneous equation system which we outline in the Appendix .Gerstein 1976 has an interesting discussion of the mathematics .Strictly speaking, it should be noted that Sraffa actually specified thecomposition of output, so that he worked with total amounts of in-puts necessary to produce the specified outputs rather than withinput-output coefficients . As a result he did not need to make any as-sumptions about returns to scale since he had precluded variation inoutputs .

16 The limited nature of Sraffa's results impels Sraffians sooner or laterto turn to the general linear production model of von Neumann . For avery clear exposition of what it is possible to do with such a model,see Bliss 1975 .

17 Cases of joint production proper are not at all uncommon, particularlyin the chemicals industries . Consider for example the cracking of oilin refining processes. The joint products thereby produced includetar, bitumen, high octane fuels such as kerosene, petroleum, lubri-cating oils, most precursors to plastics such as PVC, synthetic foodmaterials, and industrial alcohol . The latter is also produced in otherjoint production processes . Joint production is therefore a signifi-cant phenomenon, sufficient to problematise the quantification ofembodied labour, and, considering that most of the products justmentioned are themselves used as inputs in further production pro-cesses, to problematise the calculation of ratios based on quantitiesof embodied labour, even if one does not claim any significance forthem as "values" .

18 We are indebted to Ben Fine and Laurence Harris for letting us readthe draft manuscript of their book Rereading Capital (Macmillan,forthcoming) wherein this point is made .

19 The relevant parts of the often cited Frobenius-Perron theorems arethose which state that any non-negative matrix has a non-negativeeigenvalue with which a non-negative eigenvector can be associated .


103

And further, that if the matrix is also indecomposable, that eigenvalueis unique and positive, and can be associated with a positive eigen-vector. (Morishima 1964, p.195) . Elsewhere (Morishima 1974) Mori-shima points out that Frobenius was born in 1849 and Perron in1880, and that Marx therefore could not have known their theorems .

20 Morishima would agree, arguing that Marx"was very successful in using this social scientific approach to coverhis mathematical deficiency and . . . obtained practically the samesolution as we accept today by the rigorous application of theFrobenius-Perron theorem" . (Morishima 1974, p .614)

Shaikh 1977 adopts a similar approach .21 Murray 1977 advocates an historical transformation position and

makes reference to further literature on the subject .22 The development of paper money requires a corresponding further

development of the analysis .23 To obtain unique absolute "prices of production" requires a "normal-

isation condition", an equation which specifies some characteristicof the value equations which it is desired to remain invariant to thetransformation into "prices of production". The issue is a purelymathematical one ; there is no matching to a real world invariance andso the choice of normalisation is arbitrary . Gerstein 1976 discusses theissue clearly, but is quite wrong in feeling compelled to choose be-tween alternative conditions.

On the other hand, to specify the "price of production" of themoney commodity to be one also fixes absolute "prices of production" .But it does so, not by the arbitrary imposition of a meaningless in-variance postulate, but by a further transformation to a more con-crete form, that of money-prices .

BIBILOGRAPHY

Armstrong, P ., Glyn., and Harrison, J ., 1978, 'In defence of Value',Capital and Class 5, (summer) .

Baumol . W., 1974, 'The Transformation Problem : What Marx ReallyMeant', Journal of Economic Literature, (March) .

Bliss, C .J ., 1975, Capital Theory and the Distribution of Income, North-Holland, Amsterdam .

Bullock, P., and Yaffe, D ., 1975, 'Inflation, Crisis and the Post-war BoomRevolutionary Communist,3/4 (November) .

Colletti, L., 1972, From Rousseau to Lenin, New Left Books, London .Fine, B ., 1975, 'Marx's Economics : A Dual Theory of Value and Growth,

by M . Morishima. A Review' . Bulletin of the Conference of SocialistEconomists, IV . 12 (October) .

Fine, B ., 1977, 'The Political Economy of Value and the Value of PoliticalEconomy. An Essay in Honour of Maurice Dobb' . Mimeo .

Fine, B., and Harris, L ., 1976, 'Controversial Issues in Marxist EconomicTheory', in Millibrand, R ., and Saville, J ., (ed)Theory', in Milibrand, R ., and Saville, J ., (ed) Socialist Register 1976,

104

CAPITAL & CLASS

Fine, B ., and Harris, L ., 1977, 'Surveying the Foundations', in Milibrand,R .,and Saville, J ., (ed) Socialist Register 1977,(Merlin Press, London)

Fine, B., and Harris, L ., Forthcoming, Rereading Capital, (Macmillan,London

Gerstein, I ., 1976, 'Production, Circulation and Value : The Significance ofthe "Transformation Problem" in Marx's Critique of Political Economy',Economy and Society, Volume 3 .

Glyn, A., and Sutcliffe, B ., 1972, British Capitalism, Workers and theProfit Squeeze ~ Penguin, Harmondsworth .)

Gough, I ., 1972, 'Marx's Theory of Productive and Unproductive Labour,'New Left Review 76 .

Gough, I., 1973, 'On Productive and Unproductive Labour - A Reply,'Bulletin of the Conference ofSocialist Economists . 11 .7 . (Winter) .

Harrison, J ., 1973a, 'Productive and Unproductive Labour in Marx'sPolitical Economy,' Bulletin of the Conference of Socialist Econom-mists 11 .6. (Autumn) .

Harrison, J ., 1973b, 'The Political Economy of Housework,' Bulletin ofthe Conference ofSocialist Economists . 11 .7 (Winter) .

Hodgson, G ., 1973, 'Marxist Epistomology and the TransformationProblem', Bulletin of the Conference of Socialist Economists 11 .6(autumn) .

Hodgson, G ., 1976, 'Exploitation and Embodied Labour time,' Bulletin ofthe Conference ofSocialist Economists V.13 (February) .

Howell, P ., 1975, 'Once Again on Productive and Unproductive Labour,'Revolutionary Communist 314 (November) .

Marx, K ., 1938, Capital Volume },(Allen and Unwin, London), 1969, Theories of Surplus Value, Volume II .(Lawrence and Wishart,

London.1970, Capital Volume I (,(Lawrence and Wishart, London .)1972, Capital Volume I I (,(Lawrence and Wishart, London1973 . Grundrisse,(All en Lane, London .)1976a, Value Studies by Karl MarxLNew Park, London1976b, Capital Volume (,(Penguin, Harmondsworth)

Morishima, M ., 1964, Equilibrium, Stability and Growth,(OUP, Oxford)Morishima, M., 1973, Marx's Economics: A Dual Theory of Value and

Growth,(CUP, CambridgeMorishima, M ., 1974, "Marx in the Light of Modern Economic Theory",

Econometrica .Murray, R ., 1977, Value and Theory of Rent : Part One, Capital and

Class 3 (Autumn .)Ricardo, D., 1951, On the Principles of Political Economy and Taxation

Sraffa, P ., (ed),(CUP, Cambridge .Samuelson, P.A ., 1971, "Understanding the Marxian Notion of Exploit-

ation : A Summary of the So-called Transformation Problem BetweenMarxian Values and Competitive Prices,' Journal of Economic Liter-ature, Vol. 9, No .2 .

Seton, F ., 1957, 'The Transformation Problem,' Review of EconomicStudies, Vol . 24 .

THE ANOMALIES OF CA PI TAL

105

Shaikh, A., 1977, 'Marx's Theory of Value and the "TransformationProblem", in Schwartz, J ., (ed), The Subtle Anatomy of Capitalism,(Goodyear, Santa Monica .)

Smith, A., 1904 . An Enquiry into the Nature and Cause of The Wealthof Nations,(G rant Richards, London)

Sraffa, P., 1960, The Production of Commodities by Means of Commod-ities,(CUP, Cambridge .)

Steedman, I ., 1973, 'The Transformation Problem Again,' Bulletin of theConference of Socialist Economists 11 .6 (Autumn) .

Steedman, I ., 1975, 'Positive Profit and Negative Surplus Value,' EconomicJournal (March) .

Steedman, I ., 1977, Marx after Sraffa,(New Left Books, LondonWilliams, M ., 1975, 'An Analysis of South African Capitalism - Neo-Ricar-

dianism or Marxism?', Bulletin of the Conference of SocialistEconomists, IV.10 (February) .

Yaffe, D ., 1973, 'The Crisis of Profitability : A Critique of the Glyn-Sutcliffe Thesis,' New Left Review 80 .

Yaffe, D., 1975, 'Value and Price in Marx's Capital! RevolutionaryCommunist 1 (January) .

Pluto 41~ PressMarxism and the PartyJohn Molyneux

John Molyneux examines the work of Marx, Lenin, Luxemburg,Trotsky and Gramsci taking as his central theme their concern withthe relationship between the party and the working class . Heexplores such questions as the independence of the party, spontan-eity and consciousness, democracy and centralism ., and the impli-cations of these debates for the revolutionary party today .£2 .95 paper £6 .60 clothRights at WorkJeremy McMullen

Contracts/ equal pay/ lay-offs/ discrimination/ maternity/ dismissal/union recognition/ collective agreements/ strikes/ tribunals andother issues. Information essential to every working person .£2.25 paper £12 .00 cloth

Unit 10 Spencer Court, 7 Chalcot Road, London NW I 8LH

SURVEY

REGIONALISM: SOME CURRENT ISSUES

Doreen Massey

SOME DEFINITIONS AND THEMES

The term 'regionalism' is a very inadequate one . Recent discussions in theCSE Working Group, however, failed to come up with anything that wasboth more accurate and less than a paragraph long . For the purposes ofthis survey, 'regionalism' is taken to refer to the analysis of intra-nationalspatial differentiation. Its concern is to study the mechanisms by whichthe process of accumulation generates uneven spatial development, and theeffects of such unevenness on the development of a national social form-ation and particular areas within it . The scale is intra-national in the sensethat it is at this level at which the spatial unevenness which is the focus ofattention occurs . This does not mean, however, that such differentiation isproduced solely by mechanisms defined at the national or intra-nationallevel . Spatial unevenness in the process of accumulation, within a socialformation, may just as well be dominantly the product of mechanisms oper-ating at an international scale . The object of study, however, is spatial un-even development and its effects within a national economy . Such effectsmay occur at any spatial level within the social formation, from inequalitiesbetween major regions to patterns of growth and decline of particular cities (I) .

The process of accumulation within capitalism continually engendersthe desertion of some areas, and the creation there of new reserves of lab-our-power, the opening up of other areas to new branches of production,and the restructuring of the territorial division of labour and class rela-tions overall . The geographical distribution of population is typically farmore than a general tendency to agglomeration superimposed on a "terri-torial division of labour, which confines special branches of productionto special districts of a country" (Capital 1, p.353), as occasionally impliedby Marx . Even in those few areas where particular branches of productionhave entirely dominated the economy, it is not possible simply to assumethat such areas will be the same as others equally so dominated . It is morethan the branch of production which determines the characteristics of aregion. Thus Gervais, Servolin and Weil (1965) distinguish three types

REGIONALISM

107

of agricultural region in France, a distinction based primarily on the natureand stage of articulation of capitalism with peasant production, and on thedominant form of class relations (quoted in Lipietz, 1977, pp .48-52) . Suchdifferences in the economies and class structures of particular areas mayalso be associated with significant differences in political relations . Theresulting picture of 'regions' and of 'inter-regional relationships' is thusenormously complicated . The purpose of work within regionalism is tounderstand the formation, nature and effects of this spatial differentiation .

Why, however, should socialists be interested in the analysis of thisaspect of uneven development? Briefly, the fact and the form of spatialdifferentiation can affect both political and economic development . Thelevelling-out of employment rates between regions figured in the UK Nat-ional Plan as a means of increasing the available labour-force ; in Italy reg-ional disparities are argued to have been beneficial for accumulation(Secchi, 1977) . Analysis of spatial differentiation can therefore be an im-portant component simply in understanding the working of an economy .

But there are also much more immediate political reasons . Most ob-viously, analyses of uneven regional development can contribute to thedebate on regional separatist movements . More generally, however, thepresent crisis is affecting different parts of the country in different waysand to different degrees . Such spatial differentiation can frequently operatein a divisive way in the working class . When faced with massive declines inlocal industry and jobs, community groups and unions have frequentlyfought as though the problem was one of and for their area. This, of course,is the way in which the 'problem' is represented by capital, and it has twomain repercussions . First, it sets workers of one area against those of an-other in a chase after available jobs, for instance . The prime recent exampleof this has been the attempt to portray the inner cities as having lost outto the State-assisted peripheral regions . Here an important part of thework within regionalism can be to show the relation between the disparateproblems and struggles of different parts of the country . A second reper-cussion of 'regional problems' is that localised economic problems areoften understood as stemming from the supposed inadequacies of the part-ucular areas and its people . The Red Paper on Scotland attributes some ofthe problems of Scotland to a shortage of local entrepreneurs (see Firn,1975) ; the White Paper on Inner Cities of the present Labour Governmentlays much of the blame for the present decline of those areas on their in-habitants' lack of (appropriate) skills . A prime aim of studies in regionalismis to combat this spatial definition of phenomena and to analyse and pointto the real causes of such disparities .

The purpose of this survey is to present some of the issues currentlypre-occupying analysis within regionalism . However, because work in thisfield is as yet rather disparate (indeed coherent debates are only just nowbeginning to emerge), the paper has as an aim also to formulate somemajor lines of implicit contention and to argue for a particular approach toanalysis . Many of the debates hinge on methodological issues, but their im-plications go way beyond methodology . Such issues include : whether ornot one starts analysis from pre-given regions ; the potential or otherwise ofthe regional analysis of Stuart Holland ; the possibility of 'borrowing' form-ulations from underdevelopment theory . It is primarily around questionssuch as these that the structure of the paper is organised .

108

CAPITAL & CLASS

APPROACHES TO ANALYSIS

The concern, then, has been how to formulate approaches which enableanalysis of the complexity of spatial differentiation ; how to go beyondgeneral references to 'uneven development' . The present section brieflyindicates a number of approaches which have been attempted, each ofwhich has yielded insights and information, but each of which also has itsproblems .

Abstract formulation and general lawsThere have been a number of attempts to derive general propositions

concerning the spatial form and development of the capitalist mode ofproduction . First, the possibility has been investigated of elaborating a'law of value over space' (see, for example, Hein (1976), Lipietz (1977) ) .At different levels of analysis, both these authors reject such mechanismsand proportional ities. Indeed Lipietz interprets the absence of any regulatoryeconomic mechanisms over space as a fundamental reason for State inter-vention in the geographical organisation of capitalism (2) . A rather dif-ferent attempt at the formulation of general statements about the geo-graphy of capitalism has been to propose a characterisation of the system'scomponent parts . Thus, Castells (1977) defines the urban as the space ofcapitalist consumption, the region as the space of production . Such at-tempts have in general been roundly criticised (Harloe, (1978) and Sayer(1977) ), primarily as abstract and arbitrary .

Thirdly, there are a number of writings which propose a necessarytendency within capitalism towards spatial centralisation, not only of con-trol, over the process of production, but of production itself . Such conclu-sions have a degree of empirical backing and an apparent radicalism, andare clearly stimulated by a desire to counter the conclusions of equalitywhich emerge from neo-classical theory (see Holland, 1976 ; Purdy, 1977 ;Castells, 1977) . Marx, too, was inclined to see an inevitable tendency undercapitalism towards spatial concentration (see Capital, volume 1, p.352 andGrundrisse, p .587, both quoted in the discussion in Harvey, 1975) . But,apart from their dubious theoretical status, the vulnerability of such a-historical generalisations has become apparent in face of the recent ten-dency for the regional decentralisation of production (see later, and criti-cisms in Mellor (1975), Harloe (1978) and Massey (1976) ) . Empirically,neither the neo-classical nor the 'centralisation' school is correct . Thoughapparently opposed, they share the same problem of substituting for his-torical analysis predictions derived from an a-historical formal model .

Given its political importance, the work of Stuart Holland meritsa little more elaboration . Holland's (1976) argument is that a tendency toregional inequality is intrinsic to capitalism but that it has until recentlybeen offset, primarily by State action . The present dominance of multi-nationals has undermined this ability of the State since these firms are ableboth to play off States against each other and to locate in the Third World,thereby ignoring the peripheral regions of metropolitan countries . The ten-dency to regional inequality has therefore re-emerged . Empirical evidenceto the contrary is seen as an exception, merely 'disguising' the underlyingtrend (p .57) . This is not in any way a class analysis, and indeed, by con-centrating on regional rather than class relations, it has potentially divisive

REGIONALISM

109

implications . Equilibrium theory is simply replaced by an elaboration ofMyrdal and Perroux ; in order to account for the previous invisibility of theclaimed empirical tendency, State regional policy has to be interpreted asunambiguously directed towards regional equality ; the State is umpirebetween capital and the public interest - a role it would again play inHolland's proposed policy solution ; present regional problems are in factinterpreted as the result, not of capitalism, but of a 'meso-economic sector'which, with its super-profits, has broken free from economic imperatives -again an important proposition since it enables the proposed solution ofnationalised forms acting differently from their private competitors (3) .

Holland's work does not, then, provide a jumping-off ground foranalyses of spatial differentiation, nor even of the regional problem, thoughit has certainly raised some important empirical issues and highlighted thepolitical significance of certain aspects of spatial uneven development . Itsfrequent acceptance as 'Marxist' - or at least as the best we've got -should be a stimulus to further work .

Approaches borrowed' from underdevelopment theoryFew of the Marxist classics treat the subject-matter of regionalism to

more than a passing reference (cf, for instance, the comments of Harvey,1975, p.274). This lack of forebears has produced a sense of unease, animportant effect of which has been a tendency to adopt methods of anal-ysis developed at 'other spatial scales' . In particular this is true of work atthe international level, in imperialism and underdevelopment . However, thispaper will argue that, while much may be gleaned from such analysis forthe study of spatial differentiation within a social formation, it is notpossible simply to transplant them to 'a lower level of spatial disaggregation' .The relations between nation states within world imperialism are not to beequated with 'interregional relations' within a nation .

First, there are empirical differences between nation states and theirconstituent 'regions' . These include, for instance, monetary unificationand trade and customs policies (see, eg, Hechter, 1975) . More fundamentally,the State as a focus for class relations is usually less strong at regional levelthan national (Lipietz, 1977) . These are, of course, tremendous general-isations, and great variation exists in the degree to which such differenceshold, but, as we shall see, they are indicators of potential problems in anysimple transference of theories derived at the international level to prob-lems of intra-national spatial differentiation .

A second, and related, implication of such transference is that there isa general problematic of 'the spatial', of which the basic idea is that geo-graphical differentiation and 'inter-areal relations' at one scale are simplythose of another scale writ large, or small . As Anderson (1975) points out,in such a problematic "spatial form and scale are considered in the ab-stract, forgetting that we are dealing with social divisions of territory andsocially different types of territorial division" (p.15) . The object of anal-ysis is not arbitrary divisions of 'space' as such .

Thirdly, and most importantly, the theories discussed in the presentsection tend on the whole in their application at international level to takenation states as objects given to analysis . Whether or not this is correct at

110

CAPITAL & CLASS

an international level, this paper will argue that 'regions' are not necessarilypre-given to the study of intra-national spatial differentiation . Considerabledebate exists on this, and as will be seen there are a number of differentapproaches to the problem . This paper will argue that regions must be con-stituted as an effect of analysis ; they are thus defined in relation to spatialuneven development in the process of accumulation and its effects onsocial (including political) relations . Thus the analysis of the productionof uneven development does not imply a pre-given regionalisation .

This is not to say, however, that there can never be reasons for anal-ysing the place, within the overall process of spatially uneven development,of an already-specified region . The recent growth of 'regional nationalism'has inevitably brought such questions to the fore. For this to be a validprocedure, however, there has to be a clear reason for taking the regions asgiven. To take an example, what is the basis for analysing 'East Anglia' interms of its 'interregional relations? As far as I am aware, there is no signi-ficant and specifically East Anglian social or political force . And if indeedEast Anglia is a coherent entity in terms of economic criteria, this shouldbe the result of analysis and not assumed from criteria and boundaries con-structed in some other area of investigation . This point will be taken upagain later .

There is another theme which underlies much of the work discussed inthis section, though again it is by no means exclusive to it . Again, moreover,it is a subterranean theme rather than an explicit debate . This concerns thepresence, or not, within regions of metropolitan capitalist countries, ofpre-capitalist forms . The implications of most of the work reported here isthat no such forms exist . The study of the North East of England acceptsthis implicitly . Carter (1974) explicitly adopts a market definition of cap-italism and appear to equate being influenced by the dynamics of the CMPwith being capitalist . Lovering (1977) points out that the internal colon-ialism model frequently applied to Wales lacks any concept of modes ofproduction and their relations of conservation-dissolution . Moreover,having formulated a more coherent model, he finds no empirical evidenceof the reproduction of non-capitalist modes . Only Lipietz (1975 ; 1977),writing in France, presents argument and empirical evidence for the oppositepoint of view . The issue is of course far broader than the problems of regionalanalysis, and stems from underlying theoretical and political positions .

Finally, it should be stressed that the theories of underdevelopmentreferred to here are all subject to debate and criticisms in relation to theirapplication at international level . It is beyond the scope of this paper todeal with those debates . All that will be referred to here are those pointswhich concern their use at regional level .

In one of the more thoughtful attempts to use dependency theory,Carney, Hudson, Ive and Lewis (CHIL) (1975) draw "on this body oftheory . . . to suggest certain characteristic features of underdevelopedregions in the way Szentes (1971) has done for countries, and 'test' themagainst our British case study - the North East" (of England) (p.144) .They argue "that the temporary externalisations of economic contra-dictions that characterised an imperial phase of capitalist development are,in some capitalist societies, and especially France and the United Kingdom,

DEBA TE: REGIONAL ISM

111

being replaced, in part, by attempts to contain them internally" (p .157) .Their argument involves analysing the contradictory place of the NorthEast where "the basis of profitability . . . historically has involved depres-sion of wages as they enter into costs of production, and/or the reproduc-tion of a large reserve army of unemployed" (p .149), in an overall economy"the basis of (which) lies in high real wages and high demand for consumergoods within the domestic market, and on capitalist consumption and Stateexpenditure to prevent realisation crises re-emerging whilst allowing con-tinued capital accumulation" (p .149) .

In another example, Carter (1974), in a discussion of bourgeois analysesof the Scottish Highlands, uses Frank (1970) to challenge the typicalview of that region as the 'archaic' sector of a dual economy .

Significantly, at the empirical level the debates about this approach doreflect the problem of switching objects of analysis, from international tointerregional . This is particularly the case in relation to class structure .CHIL, in their paper on N .E. England, talk of an "indigenous bourgeoisie"and advocate the use of Frank's work to explain how the local bourgeoisiehas become increasingly controlled from outside the region (pp .153-4) .Considerable scepticism of this position is expressed in Anderson (1975)and in a discussion reported in Harloe (1975, p .166) . Lebas (1977, p .84)launches a general attack which clearly relates the identification of suchregional classes to the demands of this type of approach . She talks of "a'creeping parochialism', a characteristic often noted of research groupsdoing work 'on their region' . This incipient parochialism, compoundedwith the lack of concerted theoretical perspective, leads researchers toestablish the questionable existence of 'regional bourgeoisies' " . As alreadyargued in the present paper, however, such matters are clearly empiricalquestions . Mellor (1975) who severely criticises (on empirical grounds) theuse of dependency theory, herself gives evidence of regional class distinc-tiveness within both working class and bourgeoisie .

Closely related to dependency theory are the concepts of unequalexchange . Lipietz (1977) and Sayer (1977) both examine the usefulnessof this approach, and the concepts are referred to in a number of otherstudies . The positions follow those of Emmanuel (1972 ; 1975) and Amin(1973 ; 1976) with Lipietz's approach integrating concepts of 'externalarticulation' and 'unequal exchange in the broad sense' (spatial different-iation in the distribution of industries with high and low organic compos-itions), and 'integration' and 'unequal exchange in the narrow sense'(based on spatial differentiation of wage levels) with Rey's (1973) conceptsof stages in the process of articulation to capitalism of non-capitalist modes .

Starting with criticisms made at the empirical level, Sayer (1977)follows Emmanuel in arguing that "unequal exchange in the narrow senseis unlikely to take place within countries unless there is some institution-alised differentiation of wages within each sector (eg, apartheid)" . In fact,it is not clear that such a statement can be made a priori, but anyway theevidence for its empirical validity or otherwise is not unambiguous .

Empirical criticism is also made of unequal exchange in the broadsense. This is that, while the usual notion of this form of unequal exchange

112

CAPITAL & CLASS

would have low organic composition sectors in the peripheral regions, highorganic composition sectors in the 'central areas', and consequently a flowof value (with profit-equalisation) from periphery to centre, in fact one ofthe characteristics of recent industrial investment in intranational peri-pheral regions in Western Europe and the USA has been its high degree ofcapital-intensity relative to that of the centre . Thus Sayer refers to "someinteresting and possibly counterintuitive spatial and structural changes . . .and perhaps surprising inverse relationship between regional income andcapital investment per employee" (p .6) . (4)

The opposite point of view is put by Lipietz (1977) . For him, unequalexchange in the broad sense represents the articulation of different modes,or different stages of modes, of production (see, for instance, pp .58, 61),but its effect is one of the bases for regional inequality of wages typifiedin the phase of integration and implying unequal exchange in the narrowsense. Moreover, it is as an empirical question of unequal exchange in thenarrow sense that Lipietz raises current tendencies of manufacturing in-vestment (pp .58-59) . Here, however, the problem is not the specificationof the mechanism of unequal exchange but simply a worry as to why theinequality of wages has not provoked the equilibrating reaction to be anti-cipated from the simple equation form (i .e . why do capital intensiveplants form a significant proportion of the production processes presentlybeing established in peripheral areas?) . The 'answer', which Lipietz himselflater discusses, is that the response of capital to spatial differentiation maytake a number of forms, and cannot simply be predicted out of historicalcontext. The present attractiveness of peripheral regions as a location isdue to more than wage differentials .

There are, however, other questions to be answered about this approach .First, Lipietz follows Palloix in emphasising that even if unequal exchangein the narrow sense is occurring, an analysis which is simply confined tothat can only register the fact of its occurrence, as the result of an already-existing regional differentiation. He explains such inequality by unequalexchange in the broad sense .

The second question refers to unequal exchange in the broad sense .The equalisation of the rate of profit between sectors with different organiccompositions of capital is a tendency always in operation in a capitalisteconomy . The particular empirical phenomenon being referred to by thisapproach is the fact of systematic spatial pattern . What remains unclear areboth the implications of this in terms of the nature of regional 'inequality'(in what sense is this unequal exchange?) and the mechanisms of produc-tion of that inequality .

The third line of work which tries to formulate regional questions inan 'imperialism' framework is that which uses the internal colony model .We refer here only to attempts to apply the approach to regions withinmetropolitan capitalism . Hechter's (1975) discussion of the British Celticfringe is probably the best-known example . Hechter's own approach is notsquarely within the Marxist tradition, and although he uses terms such asmode of production, this tends to refer to rural/urban differences ratherthan to class relations and modes of appropriation of surplus labour . Hiswork has, however, been influential amongst Marxists and non-Marxists,

REGIONALISM

113

and particularly within the nationalist movements . Lovering (1977) pro-vides a detailed discussion and critique of the use of concepts of internalcolonialism within Plaid Cymru . His criticisms include the loose and in-correct use of the term 'exploitation', the conception of the State as adeliberate conspiracy, and the lack of empirical evidence for many of theclaims of the proponents of the model - in terms, for instance, of classstructure, and net flows of resources .

Finally, the use of all three of these approaches either implies or en-courages an analysis of the production of spatial differentiation whichstarts from pre-defined regions (a characteristic, as we have seen, related totheir original, international contexts) . All the authors are aware of thisproblem and its implications, but it is difficult, using such approaches, toescape them . Thus, Carney, Hudson, Ive and Lewis take as given, withoutany analytical justification (5), the North East of England as 'a region', asan adequate theoretical object . Indeed, they refer to it as a 'social formation'(pp .149, 151). Moreover, in spite of their correct insistence (pp .140, 155)that 'inter-areal relations' do not exist, the test of the dependence modelrefers to 'direct economic dependence' and to 'trade dependence' . Theiranalysis is referring to important real phenomena, yet just as the definitionof 'the North East' requires justification, so does the concept of a 'region' .Concepts of 'inter-regional relationships' imply the definition of spatialentities with some degree of internal coherence, whether economic orpolitical. Such definition must be the result of analysis ; it cannot be anintuitive or a priori starting point . The analyses of Wales referred to byLovering, on the other hand, do have a reason, at least at the politicallevel, for starting with a predefined region . Yet, as Lovering points out, thedivisions and dependencies within Wales are comparable to those betweenWales and England . Where there are such dislocations between 'political'and economic regionalisations, that itself may be an important phenom-enon to analyse .

The attempt to conduct analysis on the basis of given regions, especiallywhen combined with theories originating at the international level, canalso (though it does not necessarily) entail political implications whichhave been much subject to attack . Mingione (1977) in discussing southItaly, writes "it seems certain that the internal regional imbalances whichexist are not principally a result of imperialist exploitation . . . Rather theyresult from a process of centralisation and specialisation which is commonto all capitalist development"(p .95) . In afootnote,he adds"For this purposeI do not share the analysis of those authors who mechanically extendtheories of imperialism to apply them to under-developed regions . Thedivergence becomes yet wider when one considers the political conclusionswhich these authors draw, ending bygivingtheoretical support to separatism,local nationalism and the rebellion of all the classes in underdevelopedareas against a hypothetical colonial domination" (p .109). Clearly, theauthors of Plaid Cymru, for example, would take the opposite line. Again,it is not clear how, politically, one should understand "dependence", theconcept of "a structurally deformed economy" or "externally-orientedaccumulation" in a regional context . These are debates which link theanalyses of 'regionalism' to those of nationalism and regional separatism .

114

CAPITAL & CLASS

Lipietz (1977) is aware throughout of these problems (see, for example,pp.25-26), and his regions and inter-regional relations are therefore theproduct of his analysis. But even in Lipietz's work problems arise whenlinking regions defined in terms of their histories to regions defined interms of their relation to the presently-emerging spatial division of labour .Clearly there is no necessary one-to-one correspondence between the twoand it may not therefore be appropriate to start (as Lipietz tends to) froma specification of the first for an analysis of the second . Such change overtime in the 'regionalisation' of a social formation may involve a radical re-structuring both in the 'shape' of the spatial variation and in the nature ofthe use made by capital of any given form of differentiation . In fact, whatLipietz is doing here is to handle implicitly a change in regional structurewhich we would argue should be made explicit in the framework of ex-planation .

From accumulation to spatially uneven developmentThe approach which is suggested here begins from the process of accu-

mulation and analyses the production of spatially uneven developmentwithout any pre-specified regionalisation of that space . From analyses ofaccumulation, it produces concepts of geographical organisation in termsof the spatial division of labour .

We take as starting-point the historically-dominant processes of pro-duction, and define the uneven geographical distribution of the conditionsfor accumulation in relation to those processes . In general terms, thismeans beginning with those elements of accumulation which both have aneffect on the rate of profit and are unevenly spatially distributed (Hein,1976; Regional Social Theory Group, 1978) . It is the fact that regionalinequality is specified in relation to the evolving characteristics of pro-duction which makes this not an externally-provided regionalisation .

In any given period, new investment in economic activity will begeographically distributed in response to this pattern of spatial differen-tiation . But the nature of this response may vary . The term "spatial divi-sion of labour" is meant to refer to the way in which economic activityresponds to geographical inequality in the conditions of accumulation -the particular kind of use made by capital of such inequality . This willdiffer both between sectors and, for any given sector, with changingconditions of production . The term does not, therefore, refer to a divi-sion between regions .

The nature of capital's response to spatial unevenness is itself a productof the interaction between the existing characteristics of spatial differen-tiation and the requirements at any time of the dominant process ofproduction . This interaction is important - not only does productionshape geography, the historically-evolved geographical configuration (boththe fact of spatial differentiation and its particular nature) has its influenceon the course taken by accumulation . Thus, for instance, it may be preciselythe fact of spatial separation which enables the preservation for a longerperiod than otherwise of certain conditions of accumulation - low wagesand lack of militancy may be easier to ensure (for capital) in isolated areasdependent on one or two individual capitals . In turn, the preservation of

REGIONALISM

115

such conditions may influence the kind of technological changes pursuedby capital .

It should also be stressed that the forms of spatial differentiation rele-vant to the process of accumulation are by no means confined to `thepurely economic' . The degree of organisation and militancy of the labourforce are well-recognised 'location factors' even within neo-classical in-dustrial location theory . What such location theory does not recognise, ofcourse, is that it is the specific form taken by class relations which deter-mines these conditions . Such relations may be the basis for the lack oforganisation of the labour force (Mandel, 1963, gives a detailed example ofthis from Flanders) . Again, specific relations of land-ownership may pre-vent what would otherwise be the best location for a particular productionprocess (Lipietz, 1975) . State regional policies (which themselves may be aresponse to economic and local political conditions) may also be influential .

One schematic way of approaching this as a historical process is toconceive of it as a series of rounds of new investment, in each of which anew form of spatial division of labour is evolved . In fact, of course, theprocess of change is much more diversified and incremental, though cer-tainly there are periods of radical redirection . In general, however, anynew form of spatial division of labour will typify only the more advancedsectors of production, and may well vary between each sector . Betweenrounds, in other words, conditions will change . They will do so as a resultof the combination of 'more purely spatial' changes with those in therequirements of production . First, the process of accumulation may beaffected by changes in relative location through developments in trans-port and communication . The pressure towards improvements in thesederives from the requirement both to cut costs of production and toreduce the time of circulation (Grundrisse, pp .533-538 ; see also Harvey,1975) . The effect is that "the relative differences (in distances) may beshifted about by the development of the means of transportation and com-munication in a way that does not correspond to the geographical dis-tances". . . a fact "which explains the deterioration of old and the rise ofnew centres of production because of changes in communication andtransportation facilities" (Capital, 2, p.253), Such shifts in the spatial sur-face produce changes in the relative competitive positions of individualcapitals, in the relative prices of different commodities, in methods ofproduction, etc . At a more aggregate level, they will change the relativecompetitive position of branches of production in whole regions, and eventransform the conditions in a particular region to being favourable to abranch of industry not yet located there .

Second, changes in the characteristics of accumulation may occureither in the production requirements of specific branches - and there-fore in their locational requirements (see e .g. Dunford, 1977 ; Massey,1976) - or in the balance between branches of production with dif-ferent locational demands . In either case, a different regional distributionof production will result .

This new distribution of economic activity, produced by the evolu-tion of a new division of labour, will be overlaid on, and combined with,the pattern produced in previous periods by different forms of spatial

116

CAPITAL & CLASS

division . The combination of successive layers will produce effects whichthemselves vary over space, contributing to a new form and geographicaldistribution of inequality in the conditions of production, as a basis forthe next round of investment . A spatial division of labour is thereforenot equivalent to a 'regionalisation' . It is suggested, on the contrary, thatthe social and economic structure of any given local area will be a complexresult of the combination of that area's succession of roles within theseries of wider, national and international, spatial divisions of labour .

In general terms, there is probably an increasing degree of agreementthat analysis should start from accumulation rather than from regions .Within that context, new debates are now emerging . One of these concernsthe most appropriate unit for analysis. The preceding discussion is rele-vant both to individual capitals and to branches of production . Otherwork, however, has been done at a broader level . At a Departmental levela broad division has been postulated in the interwar period, between ex-panding Department I I production in the S .E. of England and a stagnatingDepartment I in the North (Carney, Lewis and Hudson, 1977) . In contrastit is the regional implications within Europe of "accumulation imperativesin relation to the component parts of capital" (variable capital, circulatingconstant capital and fixed constant capital) which are examined by theRegional Social Theory Group (1977) . In Italy, the distinction betweenleading, export-oriented and backward domestic sectors has been found tobe important . Thus, Garofoli (1975a) analyses the connection between un-balanced regional development and the development of leading sectors . Hedoes this by investigating the connection between the alternation of ex-tensive and intensive phases of accumulation and the forms of use oflabour-power . Secchi (1977) also distinguishes between extensive and in-tensive phases and analyses their different regional implications in Italyover the period from the late ninteenth century to the early 1970s, theextensive phases corresponding to a strong concentration of productionand employment in the developed areas, the intensive phases to a com-bination of de-industrialisation and the location of large leading-sectorplants in the backward areas .

So far, the discussion of the approach which starts from accumulationhas concentrated on the response of capital to spatial differentiation . Thesecond, and equally important, stage is the analysis of the effects of thatresponse . This will be taken up in a later section . For the moment, we usethis general approach to present some of the changes in regional economicpatterns at present going on in the U .K .

AN EMERGING FORM OF THE SPATIAL DIVISION OF LABOUR

For reasons of space, this section can only be indicative, but it seems im-portant to present at least the main features of what appears to be emergingas a new form of intra-national spatial division of labour, and one whichcharacterises certain expanding branches of productions, such as electronics .

Briefly, then, the characteristics of production which underlie thisnew use of space include : the increasing size of individual capitals and the

REGIONAL ISM

117

related features of a smaller number of larger plants in direct production(Dunford, 1977), complex units of production -e.g. chemical/petro-chem-ical complexes (Castells and Godard, 1974 ; Dunford 1977), the division ofproduction into autonomously-functioning stages which can be also sep-arately located (Lipietz, 1977 ; Massey, 1976), and the increasing separationwithin individual capitals of the function of overal control (Lipietz, 1977) .Within production too there have been major changes - in particular therecent apparent acceleration of deskilling of direct work alongside an in-crease in research and development . Finally, the role of the State is typicallyof growing importance both in financing major individual projects(Castells and Godard, 1974 ; Blietrach and Chenu, 1975) and in the pro-vision of 'regional infrastructure' .

Where such changes take place in an intranational context in whichthere is marked spatial differentiation in wage levels of direct workers, inlevels of skill, in degree of organisation of the labour movement, and inthe degree of presence of, for instance, banking and commercial capital,a new form of geographical organisation is arising . Such is the case in mostcountries of Western Europe and North America .

One use by capitals of such spatial differentiation is increasingly basedon the geographical separation of control and R & D functions from thosedirect processes of production still requiring skilled labourers and of thesein turn from mass-production and assembly work requiring only semi-skilled labour-power . (It should be noted that this is not some ideal-type-model, but simply a form frequently found amongst presently-leadingsectors.)

This third stage of production is increasingly located in areas wheresemi-skilled workers are not only available (since they are everywhere),but where wages are low, and where there is no tradition - at least amongthese workers - of militancy . Typically this will involve the incorporationof workers with no previous experience of capitalist relations of produc-tion - drawn either from the remnants of pre-capitalist modes, from thecollapse of a previously-dominant industrial branch (in which case it willbe the women, not the workers employed in the former specialisation,who will be employed) or from areas where workers (again mainly women)do not become totally dependent on (nor organised around) capitalistproduction relations (e .g . seaside resorts with seasonal self-employment intourism) . Although the introduction of these factories into such (frequentlydepressed) areas is hailed by the State as beneficial, (6) its positive effectsmay be minimal . Wages and skills remain low, and it is not even necessarilythe case that much new employment will result : one of the major charac-teristics of such factories is that they have few local links and stimulatelittle locally in terms of associated production (the Italians label them'cathedrals in the desert') . A good example in the UK is given by Carter(1974) in his analysis of the Highlands . Some of these plants may them-selves be relatively labour-intensive (such as electronics assembly), othersemploy very few workers (steel and chemical complexes are typicalexamples) .

The 'second-stage' of production is typically located in the old centresof skilled work - primarily nineteenth century industrial towns and cities :

118

CAPITAL & CLASS

the critical characteristic of this stage, however, is its decreasing (quanti-tative) importance . More and more, standardisation and automation areenabling capital to be locationally freed from its old ties to skilled labour-power. It is the link between such changes in the production process andthe possibilities open to capital as a result of the spatial differentiation oflabour-power (together, of course, with the collapse of other sectors,characterised by a different spatial division of labour and formerly basedin the cities) which is behind much of the present industrial decline ofthe inner cities (see Community Development Project, 1977 ; Massey andMeegan, 1978) . In such sectors as electronics, it appears that this rung inthe spatial division of labour is fast disappearing and that a simple dicho-tomy is emerging between the city-regions (rather than the inner cities) ofthe core, and the peripheral regions . Thus, changes in the labour-processas a result of competition, the use by capital of spatial differentation ; andthe reconstitution of the working class, here go hand in hand . The fact ofgeographical differentiation in the wages, skills, and organisational strengthof the working class both influences the form of, and enables, particulardevelopments in accumulation . The fact of the spatial basis of the organ-ised strength of skilled labour-power both encourages spatial decentralis-ation from those bases when capital's dependence upon skill decreases, andthereby enables a much more effective undermining of the strength of theworking class (7) .

Finally, the central metropolitan regions (such as London, Paris) aretypified by the presence of control functions, research, design and devel-opment, and by the significant presence of managerial and technical strata(it is this presence, rather than the absence of manual work, which isdistinctive) .

A number of points should be quickly made to round off this briefdescription . First, the pattern which has been described is an intra-nationaldivision of labour, but the precise form which it takes within any onenation will be determined also by the place of that nation itself within theinternational division of labour. (Thus Michon-Savarit (1975) analyses howthe future pattern of spatial organisation and interregional relations withinFrance might vary with different possible scenarios for the internationaldivision of labour .) Second, this is thus a very different form of spatialdivision of labour from, for instance, sectoral specialisation . Its economicrepercussions are also different - regions at the 'bottom' end of the hier-archy, for instance, are placed in direct competition with countries in theThird World. Such changes in the form of the spatial division of labour canbe misread as an end to spatial differentiation . Thus, to take one example,the development of locational hierarchies such as these has an importantbasis in the increasing level of 'technology' both in the production of givencommodities and as an element of competition between capitals . Com-menting on the relation between this change in production and regionaldifferentiation, Mandel (1975) writes that "regional or international dif-ferences in levels of productivity no longer provide the main source for therealisation (??) of surplus profits . This role is now assumed by such dif-ferences between sectors and enterprises . . . There thus develops a perm-anent pressure to accelerate technological innovation . For the dwindling of

REGIONALISM

119

other sources of surplus profits inevitably leads to a constant hunt for 'tech-nological renewal" (p.192) . While it is true that inter-regional productivitydifferentials (ie between firms) are no longer a dominant source of surplusprofit, the example just presented shows that this does not necessarilyimply either that regional differentials have disappeared, or that they haveceased to be relevant to capitalist investment strategies . On the contrary,such differentials would appear to be important in the appropriation ofprecisely those sources of surplus profit which Mandel now sees as domin-ant. An adequate use of regional differentials within the intro-firm divisionof labour is an important component of inter-firm competition, and part-icularly so in those branches of production which combine fast rates oftechnological change with the assembly-line-production of standardisedcommodities. Finally, it is clear that the latest form of the spatial divisionof labour is establishing not only a different form of use by capital ofspatial differentiation, but also a new shape of geographical 'regionalisation' .

THE EFFECTS OF SPATIAL DIFFERENTIATION IN THE PROCESSOF ACCUMULATION

The analysis of the evolution of a new spatial division of labour is,however, only the first stage in the study of spatial differentiation . It isnext necessary to analyse the way in which this new use of space is com-bined with the geographical pattern of previous uses . It is the effects ofthis combination which produce both the distinctive characteristics of localareas, and the overall pattern of regional variation in a social formation .

First, there are the effects on any particular geographical location orarea. Some examples of direct effects have already been referred to, butthere are broader implications . Thus, taking initially just the economiclevel, the presently-emerging spatial division of labour does not charac-terise every branch of production . It arises from the combination of cer-tain newly-dominant features of the process of production with a spatialconfiguration formed as a result of previously-dominant features . The 'newspatial division of labour' described here is therefore one (a) which is afeature primarily of new and advanced sectors of production, and (b) whichis articulated with an inherited, and different, form of spatial division .New branches of production may be introduced, affecting the conditionsof production of established local industry ; large inter-regional or multi-national capital may enter an area previously the preserve of local firms .This process of combination will therefore produce effects which go beyondthe direct implications of the locational strategies of capital, and whichwill possibily produce precisely that regional specificity which a number ofthe analyses referred to earlier (and which started from a regional base)have correctly been trying to grasp .

Moreover, these effects are not confined to production . They willinclude, for example, locally-differentiated effects on class structure(Lipietz, 1977, p.85, Lewis and Hudson, 1977, are good examples) .Gramsci's work also contains a number of comments on and analyses ofthis aspect of the impact of spatially-differentatied accumulation : Turin is

120

CAPITAL & CLASS

"the proletarian city, par excellence" . . ." precisely because of this power-fully united character of the city's industry" (in The Historical Role of theCities, Gramsci, 1977), and similar analyses are made of Milan, Piedmont,and the city-countryside relationship. Mandel (1963) analyses the form-ation of the 'two proletariats' of Belgium as a result of the distinct economicdevelopment of the regions of Wallonia and Flanders . As is clear from bothGramsci and Mandel, such processes may also imply a potentially politicallysignificant spatial differentiation in forms of class struggle. Castells (1977,ch.14) makes similar points in relation to 'urban social movements' .

It is the combination of effects such as these which produces the com-plex form of spatial variation which is the empirical phenomenon withwhich regional analysis is faced . This paper has argued so far that thecauses of such complex differentiation can not be explained adequately bystarting from any pre-given regionalisation . However, the examination ofthe resultant pattern of accumulation, and of its effects, may well requiresome method of spatial summary, and this may include the identificationof 'regions' . Considering that it is so central, there is relatively little debateon 'the concept of a region' (either its possibility or its nature) . One of theclearest positions is that of Lipietz (1977) who insists on the dominance,in the definition of any such entity, of distinctive social relations basedprimarily on the geographically-differentiated articulation of capitalismwith pre-capitalist modes (see, for instance, pp .33, 26; and Lipietz, 1975,p.419) . As already mentioned, such a position, also held by others inspecific analyses, is disputed at both empirical and political levels . In Italy,a related debate focuses on what is the correct class characterisation of theSouth: Mingione (1977) follows Gramsci (1949) in arguing that the bour-geois revolution did not involve the South ; Secchi (1977), discussing thetwentieth century, attacks the common thesis that "the Italian system ischaracterised by modern (capitalist) activities mainly concentrated in theNorthern regions, and by backward (pre-capitalist) activities concentratedin the southern regions" (p.36) .

To return, however, to the 'concept of region', a different questionwhich can be raised against Lipietz's definition concerns not whether theparticular criterion is appropriate but whether there is any point in attemp-ting to establish any criteria for universal application . It may be thatregional specificity and coherence may be established on a variety of dif-ferent bases - though 'class relations' in a general sense will evidently be adominant component . Mandel's (1963) work on Belgium again provides agood example. Having begun from an analysis of the process of accumu-lation in relation to Belgium as a whole, he analyses the spatially-different-iated form that this takes, and the impact of this in turn on class relations(see above) . His analysis is that Wallonia and Flanders are distinct in termsof date of industrialisation, the nature (branch, size, etc .) of industry, thedegree of urbanisation and the nature of internal spatial organisation, andin terms of language, culture and religion, and politics - in relation bothto nationalism and socialism . It emerges clearly not only that spatial separ-ation has been very important in the construction of these characteristicsof the Belgian national social-formation, but also that the integration be-tween form of accumulation and politics and ideology, and the effect of

REGIONALISM

121

that, clearly warrants the identification within the country of two dis-tinct regions .

It should not be assumed, however, that in every spatial analysis of anational capitalist economy such divisions will always emerge so clearly,nor that they necessarily cover the total geographical area of the state .Such a 'regionalisation' should not be forced on unwilling evidence . At theeconomic level, for instance, the combination of successive spatial divisionsof labour may not produce in any sense coherent economies . It has alreadybeen mentioned that Lipietz's regions switch from those constructedthrough historical analysis (and which are based on the articulation ofmodes of production) to those characterised on criteria (primarily type oflabour-power) relevant to an analysis of the present spatial division oflabour. 'Region' may mean many things - in this case both a coherentspatial entity in terms of social relations and a geographical disaggregationon the basis of a single economic variable . Lipietz's is a perfectly feasibleprocedure so long as the different status of these regional types is fullyrecognised . Moreover, it is possible to summarise and analyse the effects ofgeographical differentiation without the construction of coherent regions .In the UK over the last decade or so, for instance, a definite change hasbeen taking place in the form, composition and geographical distribution ofthe reserve army of labour . Some aspects of this have already been referredto (e .g . the decline of the 'inner cities') . This is an important phenomenonto recognise and to analyse but it is not necessary therefore to define, say,inner cities, as 'regions' the coherence of which extends beyond the dis-tribution of this aspect of accumulation . In such cases, different geograp-hical bases may well be appropriate for the analysis of different phenomena .

Finally, whether of not coherent regions may be defined from theanalysis, the rationale for any particular form of geographical summaryshould be related to its usefulness in analysing the effects of such differen-tiation. These effects will occur not only at the local level (as alreadydiscussed) but also as a result of the impact of the fact and form of spatialdifferentiation on the development of the social formation as a whole . Anumber of studies have been produced analysing this impact, in botheconomic and political terms . One point which emerges clearly from themis that no a priori assumptions should be made as to whether such effectsare problematic or positive for capital . There is some tendency to assumethat severe spatial inequality is necessarily a problem for capital, thatregional policy is designed to cope with these negative effects, but that it iscontinually subordinated to the more pressing demands of accumulation .At certain historical periods this is undoubtedly true (examples include theUK in both the 1930s and 1963), but spatial inequality may also be func-tional . Both aspects appear in Secchi's (1977) study which emphasises therole of territorical inequality in both periods of growth and the crises ofthe Italian economy . Secchi carries this analysis through to the politicalimplications of the spatial pattern, particularly in relation to systems ofintercapitalist alliances . Garofoli (1975b) takes up the same themes .Carney, Lewis and Hudson (1977) examine the contradictory effects inthe UK of the inter-war geographical specialisation in Departments I andII. They argue that : "The crucial restraint on the continued accumulation

122

CAPITAL & CLASS

of capital in the Department I I industries of the South was the depressedconditions of consumption in those areas dominated by Department Iproduction". Yet at the same time : "One of the conditions for the successof Department I I production in the South was that a large mass of skilledlabour was thrown out of work in the North and so acted as a reserve armysustaining reductions in production costs in the South by their presenceand sustaining production needs for labour-power by their migrationsouth" (p .58) . Again, of course, such effects have more than simplyeconomic implications . Mandel's article (Mandel, 1963) examines the veryimportant political repercussions of Belgium's patterns of regionalisation,this time in terms of the labour movement .

IN CONCLUSION

This review has been something of a mad dash through a disparate andsometimes confusing field of work . It is hoped, however, that a number ofpoints have been established . First, that there is such a field of study asregionalism, with a valid general object . Second, that within that field thereare a number of very different stages of analysis and distinct questions . Inparticular, attention has been focused in this paper on the difference be-tween the production of spatial unevenness, the effects of that unevenness,and the fortunes of particular regions . It is argued here that these must becarefully distinguished, both in terms of the questions being asked and interms of the direction of causality involved . Finally, spatial differentiationcan have important effects, both on the development of a national capitalisteconomy, and on the course of political struggle .

NOTES

Doreen Massey works at the Centre for Environmental Studies,62 Chandos Place, London W.C.2 . Much help was received in writingthis survey from discussion in the CSE Regionalism Group, and in theEditorial Committee of Capita/andC/ass, and in particular from detailedcomments by James Anderson, Mick Dunford, Mike Geddes, JohnHarrison, Jim Lewis, Richard Minns, Diane Perrons and Andrew Sayer .

1 For reasons of space the present review has a very restricted scope . Itis confined in its empirical basis, and to some extent in its propositions,to metropolitan capitalist countries, and it has had to omit consider-ation of a number of very closely related fields of work, in particularanalysis of nationalism, and of the burgeoning debate specifically on'urbanism'. Neither has there been room to consider the literature onstate intervention in this field, particularly regional policy .

2 Although there is clearly a point here, this does seem overformal, andmay have the effect of implying that the existence of theoretical equili-bria implies a problem-free capitalist economy . It is nonetheless worthnoting that the introduction of the spatial dimension plays havocwith neo-classical concepts of general equilibrium (see Massey, 1974) .

REGIONALISM

123

3

A number of these points are elaborated further in Anderson, 1977 .4 Earlier in the same article Sayer examines the difficulties of using

measures of capital intensity to indicate organic composition . It isclear, however, that he is referring to a real phenomenom in terms ofthe direction of differentials in organic composition . It is also the case,of course, that the workers in the different regions are applying labour-power of different skills, etc . (value) .

5 Which is not to say that such a justification could not be provided .Elements of such an argument appear both in this article and in Car-ney, Lewis and Hudson, 1977 .

6 Such developments are also frequently attributed to regional policy -which may well have encouraged them, but not in any sense againstthe trend of the changing requirements of accumulation . It is inter-esting to note that those who hold to the `inevitable spatial concen-tration' model of capitalism, are also forced to attribute such devel-opments solely to the effectiveness of state intervention .

7 Spatial separation and differentiation can also be important elementsin more immediate strategies, either of individual capitals (for whichgeographical mobility may enable total changes in production whichmight otherwise be fought by the unions on-site) or of State policy(the way in which inner city workers have been set against those ofperipheral regions is a good recent example) .

REFERENCES

Amin, S ., 1973, L'Echange inegal et la loi de la valeur, Anthropos - IDEPParis .

Amin, S ., 1976, Unequal development, Harvester .Anderson, J ., 1975, "The political economy of urbanism : an introduction

and bibliography", Department of Urban and Regional Planning,Architectural Association .

Anderson, J ., 1977, Stuart Holland's regionalism : reformism reheated,paper presented to CSE Regionalism Working Group, mimeo .

Bettelheim, C ., 1972, Appendix I Theoretical Comments, in Emmanuel(1972) .

Bleitrach, D ., and Chenu, A ., 1975, "Amenagement: regulation ou aggra-vation des contradictions sociales? Un exemple : Fos-sur-mer et fairemetropolitaine marseillaise" .

Brenner, R ., 1977, The origins of capitalist development: a critique of neo-Smithian Marxism . New Left Review, vol . 104, pp.25-92 .

Carney, J ., Hudson, R ., Ive, G. and Lewis, J ., 1975, "Regional underdevel-opment in late capitalism : a study of the North East of England" inMasser, I . (ed .) Theory and practice in regional science, Pion, London .

Carney, J ., Lewis, J . and Hudson, R ., 1977, "Coal combines and interreg-ional uneven development in the UK" in : Massey, D.B . and Batey P .W .J .(eds) London Paper in Regional Science, vol . 7, Alternative Frame-works for analysis, Pion, London .

124

CAPITAL & CLASS

Carter, I ., 1974, "The highlands of Scotland as an underdeveloped region"in : E de Kadt and G . Williams (eds), Sociology and Development,Tavistock Publications, pp .279-31 1 .

Castells, M ., 1977, The Urban Question, Edward Arnold .Castells, M . and Godard, F ., 1974, Monopolville : /'enterprise I'etat,

l'urbain, Mouton .Community Development Project, 1977, The Costs of Industrial Change.Dunford, M.F ., 1977, "Regional policy and the restructuring of capital",

Sussex University : Urban and Regional Studies, Working Paper 4 .Emmanuel, A., 1972, Unequal Exchange : a study of the imperialism of

trade, New Left Books .Emmanuel, A., et al, 1975, Un debat sur /'exchange inegal, Maspero, Paris .Firn, J ., 1975, External control and regional policy, in The Red Paper on

Scotland, edited by Gordon Brown .Frank, A.G ., 1970, Latin America : Underdevelopment or Revolution .

Monthly Review Press, New York .Garofoli, G ., 1975a, Produttivita del lavoro e safari : uni analisi dei differen-

ziali intersettoriali ed interregionali, Archivio di Studi Urbani eRegionali, no.3-4, pp .97-123 .

Garofoli, G ., 1975b, Un' analisi critica della politica di riequilibrio regionalein Italia : it caso del Mezzogiorno, Archivio di Studi Urbani e Regional,,no .3-4, pp.165-183 .

Gervais, M ., Servolin, C ., and Weil, J ., 1965, Une France sans Payson, LeSeuil, Paris .

Gramsci, A., 1949, llrisorgimento, Einaudi, Turin .Gramsci, A., 1977, "The historical role of the cities", in Selections from

political writings, 1910-1920. Lawrence and Wishart .Harloe, M . (ed), 1975, "Proceedings of the conference on urban change

and conflict", Centre for Environmental Studies, Conference Papers 14 .Harloe, M ., 1978, "Marxism, the state and the urban question : Critical

notes on two recent French theories", in : Crouch, C . (ed), : BritishPolitical Sociology Yearbook .

Harvey, D ., 1975, "The geography of capitalist accumulation : a recon-struction of the Marxian Theory", Antipode, vol . 7, No.2, pp.9-21(also in Peet, 1977) .

Hechter, M., 1975, Internal colonialism : The Celtic fringe in British nat-ional development, 1536-1966, International Library of Sociology,Routledge and Kegan Paul .

Hein, W., 1976, "The accumulation of capital on the world scale, the nationstate and uneven development ; Outline of a theoretical approach",Paper to CSE Working Group on the Neocolonial State, mimeo .

Holland, S ., 1976, Capital vs. the Regions, Macmillan Press .Lebas, E., 1977, "Regional policy research : some theoretical and meth-

odological problems" in M . Harloe (ed), Captive Cities, Wiley, London,pp .79-88 .

REGIONALISM

125

Lee, R., 1977, "Regional relations and economic structure in the EEC" in :Massey, D.B . and Batey, P.W .J . led), London Papers in RegionalScience, vol . 7, Alternative Frameworks for Analysis, pp.19-38 .

Lipietz, A ., 1975, "Structuration de I'espace, probleme foncier et amen-agementdu territoire ",Environment and Planning, A,vol .7, pp.415-425

Lipietz, A ., 1977, Le capital et son espace, Maspero : Economic et Social-isme, 34.

Lovering, J ., 1977, "The theory of the 'internal colony' and the politicaleconomy of Wales", mimeo .

Mandel, E ., 1963, "The dialectic of class and region in Belgium", NewLeft Review . vol .20, pp .5-31 .

Mandel, E., 1975, "Late Capitalism", New Left Books, London .Marx, K ., Capital, Lawrence and Wishart .Marx, K ., Grundrisse, Penguin .Massey, D.B ., 1974, "Towards a critique of industrial location theory",

Centre for Environmental Studies, Research Paper No 5 . (also inAntipode, Dec. 1973 and Peet, 1977) .

Massey, D.B ., 1976, "Restructuring and regionalism : some spatial effectsof the crisis". Paper presented to CSE Working Group on Regionalism .Centre for Environmental Studies, Working Note 449 .

Massey, D.B ., and Meegan, R.A ., 1978, "Restructuring vs . the Cities",Urban Studies, vol .1 5, No .3 .

Mellor, R ., 1975, "The British Experience : combined and uneven devel-opment" in : Harloe, M . led), 1975 .

Michon-Savarit, C ., 1975, "La place des regions francaises dans la divisioninternationale du travail : deux scenarios contrast6s", Environmentand Planning, A, vol . 7, No . 4, pp .449-454 .

Mingione, E ., 1977, "Theoretical elements for a Marxist analysis of urbandevelopment" in : Harloe, M . (ed), Captive Cities, Wiley, pp.89-110 .

Purdy, D ., 1977, "Review of Holland's 'Capital vs. the Regions' and 'TheSocialist Challenge' "Capital and Class, No.1 .

Regional Social Theory Group, 1978, "Accumulation, the regional problemand nationalism" in : P.W .J . Batey (ed) London Papers in RegionalScience, vol .8, Pion, London .

Rey, P-P ., 1973, Les alliances de classes, Maspero, Paris .Sayer, A., 1977, "The law of value and uneven development : some prob-

lems and possibilities for analysis" . Paper presented to the RegionalSocial Theory Group, Regional Science Association, Lanehead Work-shop, July. Mimeo .

Secchi, B ., 1977, "Central and peripheral regions in a process of economicdevelopment : The Italian case" in : Massey, D.B ., and Batey, P.W .J .(eds), London Papers in Regional Science, vol .7 : Alternative Frame-works for Analysis, Pion, London .

Szentes, T., 1971, The political economy of underdevelopment, AkademiaiKiado, Budapest .

REVIEW ARTICLE

INTELLECTUAL AND MANUAL LABOUR :AN INTRODUCTION TO ALFRED SOHN-RETHEL

INTELLECTUAL AND MANUAL LABOURBy Alfred Sohn-RethelMacMillan (London, 1978) . p .p.216. £3.95 paperback . (Available throughCSE Book Club)

Reviewed by Monika Reinfelder and Phil Slater

Alfred Sohn-Rethel's Intellectual and Manual Labour (1978b) isa major theoretical work, and could well prove a decisive turning point inthe Anglo-Saxon attempt at a Marxist theorisation of philosophy andscience . It takes the debate away from orthodox epistemology and movesit in the direction of a perspective that locates the formal determinants ofthought in the elementary form of the socio-economic realm, which for cap-italist societies is, of course, the commodity . However, while this under-taking comes as a welcome departure both from the naive materialism ofthe "reflection theory" and from the ultimate idealism of Althusserian"theoretical practice", one should not underestimate the tremendousbarriers to comprehension facing the reader, barriers rooted in the theor-etical and ideological constellation of the Anglo-Saxon tradition, in part-icular its lack of any deep-seated history of Marxism or dialectics generally .And yet, Sohn-Rethel's book is, for that very reason, ideally suited tomake a significant intervention in this constellation : aiming at a historicalmaterialist critique of the very logic of Western thought, Intellectual andManual Labour challenges that tradition at its roots, while simultaneouslystruggling to extricate the Marxist tradition from the dogmatism andsuperficiality that make it such easy prey for its political opponents inthis field (1) .

While a simple review would be inadequate to such an undertaking, asystematic critique, by contrast, would be somewhat premature, since itwould take as its object a theory devoted to questions which in the English-speaking world are not yet even recognised as questions . Thus, anybody

REVIEW ARTICLE

127

sensitive to the lack of any adequate historical materialist theorisation ofphilosophy and science should recognise that Intellectual and ManualLabour is a serious pioneer in this area, and that wide circulation and dis-cussion of it offer excellent propects for initiating a qualitatively higherlevel of debate all round . Given this strategic consideration, and given thatthe book is bound to strike many of even the most enthusiastic readers asobscure and "alien", the present contribution offers a brief introductionto Sohn-Rethel's life and work, an outline of the substance and structureof Intellectual and Manual Labour (which, with one important exception,contains the main fruits of his theoretical production) (2), a schematis-ation of the criticisms levelled at Sohn-Rethel (largely in Germany), and atabulation of his publications in German and English over the last four anda half decades .

LIFE AND WORKS

Alfred Sohn-Rethel was born in 1899, and thus belongs to the genera-tion of German-speaking individuals that boasts the names of Adorno,Benjamin, Horkheimer and Marcuse (Bloch, Korsch and Lukacs beingslightly older) . However, whereas these men began, at a relatively youngage, to publish works that were later to be of seminal importance for thestudent anti-authoritarian movement, Sohn-Rethel published his firstmajor work, Geistige and korperliche Arbeit (1970b), at a time when anti-authoritarianism had been superceded by what has been called the "Organ-isational Phase". This meant that if he was not to be dismissed as an ana-chronism, Sohn-Rethel had to address himself to the inflamed debate onpolitical economy, class struggle, Marxism and revisionism . As it turnedout, not only did Sohn-Rethel's theory contribute to the discussion oncommodity, value and money, as well as on monopoly capitalism, social-ism and technocracy, not to mention the very basis of historical mater-ialism itself, but his book became one of the central objects in the wholedebate on revisionism and anti-revisionism (3) . Despite the efforts of the`orthodox Marxist-Leninists', with their 'dialectical materialism' and'reflection theory', to preclude any substantive discussion of the book thefirst edition was a sell-out, and, faced with competition from a pirate-edition, Suhrkamp brought out the second edition in the popular 'editionsuhrkamp' series (1972b) .

Prior to this success, Sohn-Rethel had been almost totally unheard of,despite several attempts at partial exposition of his theory (see biblio-graphy), and despite an approving mention by Adorno in NegativeDialektik of 1966 to the effect that Sohn-Rethel has been the first topoint out the affinity between thought form and commodity form (4) .However, this relative obscurity soon changed after 1970 : a string of pub-lications followed (see bibliography), not only elaborating on specificthemes of the opus magnum, but also reconstructing the long and difficultroad that lay behind it. People began to take an interest in the personaland intellectual life of this mysterious figure who, at the age of seventy-one, had published a major work of Marxist theory .

128

CAPITAL & CLASS

Born in Paris in 1899, Sohn-Rethel was the son of an artist and thegodson of one of the big names in the German steel industry, Ernst Poens-gen (5) . Sohn-Rethel spent much of his boyhood with the latter in Dussel-dorf, being a frequent visitor to the factory, his interest always at the shop-floor level . Thus, contrary to what the structure of Intellectual and ManualLabour might suggest, the author's interest in the modern labour processwas riot the result of an originally theoretical concern, but a preoccupationdating back to his boyhood and early youth . Returning to his own familyin Berlin in 1912, Sohn-Rethel was thrown out of home in 1916 as a resultof his increasing political radicalisation . He went to Heidelberg Universityin 1917, where he became involved in anti-war activities, the latter provingso dangerous that he was forced to shift his studies to Munich . Conscriptedshortly before the German surrender, he then lived through the momen-tous years between 1918 and 1923, convinced of the necessity and immi-nence of revolution . The failure of the German revolution must be countedone of the most crucial phenomena driving Sohn-Rethel, like so many ofhis generation, to undertake a comradely but critical reassessment of somefundamental features of the entire Marxist tradition .

Sohn-Rethel had read Marx while still at school, but it was in the 1920sthat he began a sustained analysis of Capital, in particular the first threechapters, convinced that here alone could be found the basis for the polit-ically indispensable historical materialist critique of idealism. Pursuingwhat turned out to be a life-long, though often interrupted task, he cameinto early contact with Benjamin and Bloch, as well as Heidegger, Cassirerand Koyre, and last but not least, Adorno, with whom he corresponded inthe mid-1930s . It was around this time, after an intriguing spell as areasearch assistant in the Berlin office of the Mitteleuropaischer Wirt-schaftstag (Central European Economic Congress), (6) that Sohn-Rethelwas obliged to flee from the Gestapo as a result of his involvement withseveral resistance groups . He landed in England in 1937, where he was towork on his theory in almost total isolation for over thirty years .

At this stage, Sohn-Rethel's publications were limited to a short, butinsightful account of the collapse of reformism and the rise of fascism as anew strategy of capital (1932), and a shortened version of the doctoraldissertation he had submitted in 1928 (1936) . The latter was a fairly re-strained effort, presumably written with due consideration to its academicpurpose; however, the author's later preoccupation with the dialectic ofsocialisation/privatisation in thought/action already emerges in the argu-ment that the relation between the exchange nexus and the individualeconomic subject's Reason is one of strict "incongruence" (1936 : 43ff),as well as in the methodological emphasis on the exchange act.

By the time the dissertation appeared in print, however, Sohn-Rethel'stheorisation had made significant advances, and these find expression intwo expository pieces of the mid-1930s, both of which were finallyprinted in Warenform and Denkform (1971e) . Of the two, the first isperhaps the more important : in the form of a letter to Adorno (thenresiding in Oxford), it is remarkable for the consistency it reveals betweenSohn-Rethel's mature theory and the problematic he was trying to est-ablish decades earlier. For Sohn-Rethel, then as now, the Marxist critique

REVIEW ARTICLE

129

of political economy is "inadequate to its own purpose as long as itsconceptual tools (say the analysis of the commodity form and valuerelation) do not offer the prospect of contructing an exhaustive critiqueof the truth of bourgeois idealism" . And if the economic analysis ofcapitalism fails to provide an adequate theory of philosophy and science,"then at some point it will fall short on the tasks of social transformation"(1971e:1Of) . At this stage, Sohn-Rethel is more concerned with specifyingthe problem than with constructing bold answers, but even here he broachessome major arguments of his mature theory : for example, the abstractnature of the exchange act, and the synthetic function of money, as wellas the latter's mystified representation in Kant's "transcendental synthesisa priori" . As yet, however, there is no attempt at a historical materialistdeduction of the Kantian forms and categories, no theory of head andhand, and no notion of the "dual economics" of late capitalism . Thus, thethree major components of the opus magnum are, at this stage, absent .

Having settled in England in 1937, Sohn-Rethel worked, though withrepeated interruptions, on his project of a critique of philosophy andscience from the perspective of Marx's analysis of commodity, value andmoney . The only kindred spirit he found in England was George Thomson,the distinguished Professor of Greek and author of several major works ofMarxist theory . Henceforth, the two men drew extensively on each other'swork, always generous in their mutual acknowledgement ; indeed, Thomsonactually goes so far as to say that Sohn-Rethel "helped me to appreciatethe profound philosophical importance of the opening chapters ofCapital" (7) . However, despite Thomson's support, a 1950s manuscript ofSohn-Rethel's Intellectual and Manual Labour was turned down, not merelyby the 'respectable' publishing houses, but also by Lawrence and Wishart(which, given the intellectual tradition of the British CP, is not altogethersurprising) . In any case, Sohn-Rethel became increasingly tied up withschool teaching, only returning to intensive theoretical work in the 1960s,convinced that he would have to turn his attention to the German-speakingworld if his work was to be transmitted seriously .

The success of post-1970 proved him right . Not only did he becomean eminently publishable semi-cult figure (as well as a major bogeyman tothe 'orthodox Marxist-Leninist' camp), but in 1972 he was appointedGuest Professor for Theory of Knowledge and Society at the Universityof Bremen, holding the post (the very title of which is a tribute to hisstruggle against the 'autonomy' of epistomology) until 1976, the maxi-mum period tenable for a Guest Professorship . Upon his return to England,he concentrated on producing an Enlgish edition of his opus magnum,which, as it turns out, is no mere translation, but constitutes the mostadvanced version of his theory. Previous attempts at exposition in Eng-lish (see bibliography) will, of course, retain their topicality, both asbasic (though partial) introductions and as documents reconstructing thedifficult task of forcing a dialectical argument into a language dominatedfor centuries by an empiricism that masquerades as 'sound commonsense' . However, these monographs cannot adequately convey the sym-tematic frame of reference within which they were ultimately thought,and it is thus to Intellectual and Manual Labour (1978b) that seriousdiscussion of Sohn-Rethel's work must ultimately turn .

130INTELLECTUAL AND MANUAL LABOUR

CAPITAL & CLASS

Intellectual and Manual Labour opens with a short preface (xi-xiv) (8)by way of a brief intellectual autobiography, but also presenting thebasic argument in the bewildering, "crazy" form in which it first struckSohn-Rethel in the 1920s: "in the innermost core of the commoditystructure there was to be found the 'transcendental subject' " . This convic-tion of the "secret identity of commodity form and thought form" wasnever to be shaken : "I had grasped the beginning of a thread whose endwas not yet in sight" (xiii) . As it turns out, unravelling this thread throughhalf a century of hard work has led Sohn-Rethel over a range of problemsso vast and so (seemingly) diffuse as to be difficult to peruse, let alonereview. To take just one of the provocative questions posed in the Intro-duction (1-9) : "Is modern technology class-neutral?" (1) . Sohn-Rethel'srefreshing treatment of this perennial problem not only leads backwards tothe origin of Western thought in commodity production, but also returnsto present-day questions of workers' control, class consciousness, revolu-tionary strategy, technocracy and Meoism, thereby providing dramaticillustration of the political pertinence of what may seem, particularly inthe early sections of the book, an obscure and idiosyncratic undertaking .

The Introduction serves the useful purpose of presenting the basicarguments of the text in advance . Quoting Marx's '1859 Preface' to theeffect that it is men's social being that determines their consciousness,Sohn-Rethel argues that there is a lacuna not only in Marxism but in Marxhimself: there is no historical materialist analysis (indeed, in the '1859Preface' there is even no mention) of the "conceptual foundations of thecognitive faculty vis-a-vis nature which in one form or another is char-acteristic of the ages of commodity production from their beginnings inAncient Greece to the present day" (5) . Introducing the concept 'socialsynthesis' to designate the network of relations by which any one societyforms a coherent whole, Sohn-Rethel implicitly challenges the 'reflectiontheory' by stating a major methodological premise: "The conceptual basisof cognition is logically and historically conditioned by the basic formationof the social synthesis of its epoch" (7) . That the social synthesis of com-modity producing societies is effected via exchange conforms to thispremise, for the "constituent elements of the exchange abstraction un-mistakably resemble the conceptual elements of the cognitive facultyemerging with the growth of commodity production" (6) . In fact, this isno mere resemblance, but "true identity" (7) . To substantiate this thesisconstitutes the fundamental undertaking of the book .

Part One, 'Critique of Philosophical Epistemology' (11-79), which isto carry the weight of the entire book, begins by trying to justify theauthor's critical focus on a Kantian problematic, rather than on the moreorthodox Hegel-Marx constellation . Kant's question concerning the Dos-sibility of synthetic judgements a priori is a real question: namely, howdoes one explain the pre-given nature of the forms of intuition and cate-gories of the understanding? Kant's own reply ("Via the faculty of trans-cendental synthesis a priori") is a hypostatisation that merely restates theproblem in the form of a self-assertive 'answer', but this should not prompt

REVIEW ARTICLE

131

us to follow Hegel's 'sublation' (Aufhebung) of the Kantian problematicinto absolute idealism, for such a course blots out a paradox which, forSohn-Rethel, is objectively rooted in "the realities of capitalism" (15) .Rather than 'sublating' the problematic in the tradition of Hegel, andrather than discarding and vilifying it in the tradition of the Diamat,Sohn-Rethel asks (38) : what is the historical origin of our ability to con-struct mathematical hypotheses and the elements contributing to them?What is needed, according to the author, is the demonstration that ab-stract thought, while having the form of thought, does not originate outof thought, but out of a socio-historical act which, though abstract, con-stitutes a real abstraction (Realabstraktion) by virtue of being a spatio-temporal occurence (20) . For Sohn-Rethel, only one person has everaffirmed the possibility of such a real abstraction, and that is Marx (19),to whose analysis of the commodity the enquiry must turn .

Whereas Marx, in his critique of political economy, begins with thetwo-fold nature of the commodity (use-value and exchange-value), Sohn-Rethel, pursuing the critique of epistemology, concentrates on the corres-ponding polar activities : namely, use and exchange . If the social synthesisof the commodity producing societies is carried by the exchange nexus(29), then exchange should, to satisfy SohnRethel's methodological pre-mise, also constitute the real abstraction sought after . This is indeed thecase: the act of exchange, the abstraction from all use, provides a form ofequation (Gleichung) . that "abstracts quantity in a manner which con-stitutes the foundation of free mathematical reasoning" (47) . Paradoxi-cally, however, this abstractness of the action is not reflected in theactor's minds, which, on the contrary, remain occupied with the use-valueto be acquired : "the action is social, the minds are private" (29) . True, theabstraction does achieve "representation" in coined money, but (in linewith the fetishism of the value form as a whole) this "representation" is"disguised as a thing" and is thus not recognisable in its "true identity asabstract form" (33) .

However, Sohn-Rethel argues that the abstraction operative in ex-change does achieve an "identical" expression, namely in the so-called"pure understanding", the cognitive source of scientific knowledge (34) .To illustrate this "identity", he turns (putting tremendous intellectualdemands on his reader) to ancient Greece and lonia, which, as is oftenforgotten in philosophical discussion (George Thomson being a notableexception), achieved a 'Greek miracle' in the secular sense of creatingcoined money, thereby generating the violent class struggles that posedradically new problems for the human mind to ponder. The long line ofphilosophers from Thales to Aristotle applied their intellectual genius notleast to these very problems, but the most dazzling results of their labourtook a form no less fetishistic than the money form itself ; for example,Parmenides' 'the One' can be regarded as the first concept "fitting thedescription of the abstract material of money, but without any idea ofwhat his concept stood for and what had prompted him to conceive it"(65) . In fact, what came into being was "the capacity of conceptualreasoning in terms of abstract universals, a capacity which established fullintellectual independence from manual labour" (60) .

If Part One attempts to show how abstract thought is founded, log-

132

CAPITAL & CLASS

ically and historically, on exchange, the aim of Part Two, 'Social Synthesisand Production' (81-135), is to explain this in terms of economic exploit-ation: "intellectual in separation from manual labour arises as a means ofthe appropriation of products of labour by non-labourers" (90) . It is notpossible to summarise here the author's account of the changing relationof head and hand from ancient Egypt to the present, and, in any case, thispart of the book is far easier than the first and thus requires far less byway of introduction for the English reader. However, the coherence ofwhat Sohn-Rethel ultimately intends as a political text rests on his under-standing of the capitalist mode of production, and it is thus crucial tograsp the author's argument regarding the specific relation of head andhand requisite to that mode. Whereas in ancient Egypt exploitationmeant the non-labourer's appropriation of the products of labour of directproducers who were often their own 'masters' as regarded the precisestructure of the labour process, the rationale of capital valorisation isincompatiable with a labour process based on the labourer's 'know-how'and autonomous manual expertise, and must, on the contrary, establish an"unambiguous division of head and hand in the production processes"(113) . For Sohn-Rethel, this is the ultimate significance of the mathe-matico-experimental method of Galileo and Newton : while manual labouris necessary both to set up the experiment and to carry out in the produc-tion process the operations to which the results are applied, the actualexperiment itself is "safeguarded from any touch by human hand and madeto register specific measurements which are then read as indicated by theinstruments", which is only possible if, in direct opposition to the crafts-man's skill, the phenomenon under investigation can be "torn out of thecontext in which it occurs" (732) . The mathematico-experimental methodthus secures to capital "the possibiliby of a knowledge of nature fromsources other than manual labour" (122) .

Part two closes with a discussion (far less radical than one might haveexpected) of the c/ass nature of what we know as 'science', a problem that,like the analysis of capitalism generally leads directly (although the readerwill find the transition very demanding) into Part Three, 'The Dual Eco-nomics of Advanced Capitalism' (737-785) . Drawing on the work of Lenin,Baran and Sweezy, and Braverman, but also stressing the need for aMarxist critique of bourgeois theorists like Eugen Schmalenbach (thefounder of 'management sciences' in Weimar Germany), Sohn-Rethelargues that the economic background to the rise of what is known as'monopoly capitalism' lay in structural changes within the productionprocess itself. In line with Marx's concept of the increasing organic com-position of capital, modern management is faced with the increasing dom-inance of the so-called 'indirect' or 'fixed' element of cost, with the resultthat any attempt to reduce output, for whatever reason, will effect adrastic rise in unit costs . Thus, the rising organic composition of capitalmakes production "increasingly inadaptable to the market regulatives",and the capitalist class is forced "to try to obtain control of the move-ments of the market" (745) . This attempt, which is precisely what makesthem 'monopolists', is the point at which Baran and Sweezy begin theiranalysis of Monopoly. Capital, whereas for Sohn-Rethel the essence of the

REVIEW ARTICLE

133

matter lies at the level of capitalist production as a process of capital valor-isation : the major feature of monopoly capitalism is not some ill-defined`rising surplus' (Baran and Sweezy), nor even those phenomena Leninanalysed under the rubric of 'Imperialism', but a "substantial increase inthe rate of exploitation of the labour employed in the industries at home"(146) . For Sohn-Rethel, this strategy was pioneered by one man, Fred-erick Winslow Taylor .

The detailed discussion of Taylorism, which marks off Sohn-Rethel asone of the few Marxists to have attempted a serious analysis of the modernlabour process (9), focuses on the attempt to reduce the various operationsof the "collective labourer" to a uniform measure of time, the establish-ment and implementation of which presuppose, to quote Taylor, "takingthe control of the machine shop out of the hands of the many workmen,and placing it completely in the hands of the management" (152) . FromTaylor's "unit times" (154), via Frank Gilbreth's "synthetic timing"(155), to Henry Ford's "flow production" (159f) is a momentous, but(from the perspective of capital) quite logical development .

Important as this analysis is for an understanding of the capitalistlabour process, however, it gives rise in Sohn-Rethel's account to a farfrom unproblematic theory of the 'dialectic' of late capitalism : namely,the thesis of 'dual economics'. The author is apparently quite aware thatit is this aspect of his theory, rather than the formal analysis of the com-modity, which will come in for severest criticism in the English-speakingworld, for he emphasises in advance that this whole section is bound to be"a great deal more speculative" and is only intended to serve "as a basisfor further research by others" (139) . Nonetheless, the thesis of 'dualeconomics' is perfectly clear ; it begins by delimiting post-1896 capitalismfrom the "periods' presented in Capital : according to Sohn-Rethel, Marxanalysed the period of manufacture, where the transformation of themode of proudction takes labour-power as its starting-point, as well as theperiod of large-scale industry, where the instruments of labour are thestarting-point, but he did not live to see and analyse what Sohn-Rethelregards as the 'third period', namely, monopoly capitalist flow production,where, the author argues, "it is labour itself that forms the starting-point"(141). Despite the remarkably advanced 'extrapolations', particularly inthe Grundrisse, Marx did not, in Sohn-Rethel's eyes, show "the implica-tions carried by the external necessity of the continuity of the productionprocess (143), implications which constitute the heart of the concept of'dual economics' .

This theory runs as follows : whereas laissez-faire capitalism was amarket economy, uniformly resting on a commensuration of dead labour,monopoly capitalist flow production confounds this with a Taylorist-Fordist commensuration of living labour, of "labour in action" (171) .The logic of the latter, a logic of production rather than appropriation(carrying fantastic significance for the social nexus) is, of course, subord-inated to, and deformed by the primacy of capital valorisation ; nonethe-less, the irrepressible duality in the mode of commensuration of labour(hence, "dual economics") means that the modern labour process might"harbour potentialities which could assume a vital significance if society

134

CAPITAL & CLASS

were no longer subservient to capitalism" (165) . The critique of politicaleconomy thus leads into a critique of "scientific management" : whatmasquerades as an "objective, neutral science" is, in reality, the translationof the principles of the resocietisation of labour into the one-dimensional,fetishistic language of capital valorisation . Just as this fetishism is "one ofthe particular ideological concerns, not only of the capitalists themselves,but of the State" (1571), so any adequate socialist strategy must includethe transitional struggle of the resocietised labour force to itself become"the societising force" : only as such can it "bring about the unity of headand hand that will implement a classless society" (140) . Such a strategy,far from being an abstract demand "from without", is in fact objectibelyprepared by the development of capitalist exploitation : the fetishism ofcapital has, at least according to Sohn-Rethel, "worn thin in a type ofproduction where both labour and machinery assume compound structure"(163) . A rare example of a 'Sohn-Rethelian' struggle is offered in the formof the Pirelli strike of 1968, where the workers took over their assembly-lines, structuring 'counter-norms' and reducing the flow to one third of themanagement-rated speed (162f) . The same strategy of uniting head andhand constitutes, for Sohn-Rethel, the guiding principle of China's 'Cul-tural Revolution' (169, 182, 184), and of Mao's critique of USSR technoc-racy (178) . Finally, Sohn-Rethel emphasises that 'dual economics' does notin any way imply a mechanistic transition to socialism, and nor does itimply any inate necessity of a breakdown of capitalism "other than by itsrevolutionary overthrow" (165) . Part Three closes with a reminder that"the purpose of a study like the present must be seen against such abackground" (185) .

Part Four, "Historical Materialism as Methodological Postulate"(187-204), is by way of a methodological appendix (and appeared as suchin the German editions), underpinning the method, structure and logic ofthe entire analysis . As such, it returns to and elaborates on the methodol-ogical discussions featured in the Introduction and in the opening pages ofPart One . In particular, the author now makes explicit his evaluation ofthe Diamant's 'reflection' theory' : while conceding (over-generously) thelatter's political and ideological value, he not only declares its theoretical(that is, explanatory) value to be "nil", but goes on to castigate its "dam-aging effect" of militating against all "serious historical-material investi-gation of the phenomena of cognition" (189) (10) . In fact, Sohn-Rethelcannot really concede that the 'reflection theory' has even a healthy pol-itical value, since the theory must necessarily assume the "neutrality ofscience and technology towards social class", thereby distinguishing itself"as an ideology of technocracy, not of socialism" (191) . The book closeswith the conviction that it is only the "revolutionary commitment of ourexposition that yields the truth" (204) .

CRITICAL OUTLOOK

Given the absence of any adequate historical materialist theorisationof philosophy and science, constructive criticism of Sohn-Rethel should

REVIEW ARTICLE

135

not be such as to blur his significance as a pioneer in this field . Given thisproviso, however, it will not be amiss to map out the areas where his workhas been subject to serious criticism of a kind that aims to drive the argu-ment further, rather than demolishing it . Basically, this criticism covers therelated areas of value theory, capital theory, and socialist theory, and,common to all three, the question of Sohn-Rethel's reading of Marx . Thelatter, be it said in anticipation of charges of 'consulting holy scriptures',is an integral element in the assessment of a theory formulated on the basisof a specific reading of Capital .

Although the charge of 'revisionism' was all too often merely a pole-mical devise in the absence of any substantive discussion of Sohn-Rethel 'swork, J ost Halfmann and Tillman Rexroth make a serious attempt todemonstrate that his shifts of emphasis in the commodity analysis rest ona "misunderstanding" . According to this argument, Sohn-Rethel does notgrasp the relation between value and value form, and he fails to follow theMarxian progression from simple commodity circulation to capital theory ;this double flaw reveals itself in Sohn-Rethel's focus on the more 'tangible'phenomenon of the exchange act, implying that labour is abstractified onlyin the sphere of circulation (Halfmann and Rexroth, 1976, pp .80ff) . Theultimate expression of this misunderstanding comes, so his critics argue, inthe thesis of 'dual economics' : far from establishing any such 'duality' ofeconomic laws, the societisation_ of labour is fundamental to the consoli-dation of capital (Ibid, pp .87ff) . All in all, conclude Halfmann and Rexroth,the basic difference between Sohn-Rethel and Marx is that the latter "loc-ates the unity of the form in the contradictory nature of a value-determinedsocietisation principle, whereas Sohn-Rethel presents the same state ofaffairs in terms of two sets of structural laws" (Ibid, p .101) .

Equally fundamental is Hans-Dieter Bahr's criticism : he detects aresidual empiricism in Sohn-Rethel (11), but also proceeds to construct(on the basis of a serious study of the Grundrisse) a theory of machinerythat is implicitly critical of Sohn-Rethel for effectively replacing valuetheory with a theory of head and hand, and for reproducing, thanks to a'dialectical' analysis, precisely the sort of uncritical evaluation of machinetechnology that Sohn-Rethal had hoped, at least at the programmatic level,to loosen up . Bahr does not extend this implicit criticism to the politicallevel, but this .can readily be done by looking at other Marxists who, likeSohn-Rethel, have worked on the modern labour process, but who havedrawn radically different conclusions as regards class consciousness andstrategy . A particularly useful devise (one implied by the collection ofarticles entitled The Labour Process & Class Strategies) (12) is to confrontSohn-Rethel's 'dual economics' and socialist strategy with the Italian trad-ition associated with Mario Tronti . While agreeing to a great extent withSohn-Rethel's picture of the Taylorist-Fordist deskilling of the labourforce, the 'Trontians' would argue that this class composition is capitalistpure and simple, and that precisely because of its Taylorisation-Fordis-ation the working class can no longer meaningfully fight under the bannerof 'workers' control', but only in terms of a 'refusal of labour', that is, arefusal to reproduce itself as capital . Sohn-Rethel's perspective, by contrast,would then appear not only as reformist, technicist and even oppressive,but as anachronistic in terms of the class competition he himself depicts .

136

CAPITAL & CLASS

However, this schematic account of critical perspectives on Sohn-Rethel's work should not be seen as detracting in any way from his crucialachievements. Halfmann and Rexroth themselves subtitle their critique'Sohn-Rethel's Revision of Value Theory and the Productive Consequencesof a Misunderstanding' ; these "productive consequences" consist, aboveall, in the attempt to relate thought form and commodity form, andthereby to liquidate the idealist conceptualisation of philosophy andscience, particularly in its residual role inside Marxist theory . In the Eng-lish-speaking world, Sohn-Rethel's very departure has yet to establish itselfas a meaningful undertaking, and in this sense his work can only be des-cribed as seminal . Needless to say, every aspect of his theory will have tobe scrutinised critically, but the debate will only be constructive if con-ducted at, rather than be/ow the level at which Sohn-Rethel's analysisoperates. At the very least, Intellectual and Manual Labour is a reasonedcall to its readers to awaken from their dogmatic slumbers, and as such itcannot be recommended too highly .

NOTES

1

Dates and letters in parentheses refer to the bibliography below .2 The exception is Sohn-Rethel 1978c . For details, see note 6 .3 A brief outline of the German debate on Sohn-Rethel up to 1975 can

be found in Halfmann and Rexroth, 1976, pp .7ff.4 "Sohn-Rethel was the first to point out that hidden in the transcen-

dental principle, in the universal and necessary activity of the mind,lies labour of an inalienably social nature ." Adorno, 1973, p177 -modified . Adorno is referring here to an expose Sohn-Rethel sent himin 1936; for details, see below . See also Sohn-Rethel 1978a, p .137 .

5 The following account is based on the autobiographical material scat-terred throughout Sohn-Rethel 's publications, supplemented by aninterview with Sohn-Rethel in Birmingham, 31 January 1975 .

6 Here, Sohn-Rethel gained first-hand experience of the gradual fusionof private enterprise, finance capital and Nazi politics . His reflectionson this experience are collected in Economy and Class Structure ofGerman Fascism (1978c), some of which was written as early as 1932,and which includes personal recollections, attempts at a value analysisof Nazism, and indicators as to how these early sketches could bebrought under the later theoretical concept of 'dual economics' .

7 Thomson, 1961, p .7 . This book, which constitutes Volume Two ofThomson's Studies in Ancient Greek Society, is the crucial work inthe present context .

8

Throughout this section, all page references in italics are to Intell-ectual and Manual Labour (1978b) .

9 Apart from Marx himself, the list would include Gramsci, MarioTronti, Ferruccio Gambino, Sergio Bologna, Karl-Heinz Roth andHarry Braverman .

10 This part of the book is based on the author's 'Materialism and its

REVIEW ARTICLE

137

Advocacy' (1947/48), an attempt to loosen up the dogmatism on suchquestions in the British CP . In part a critique of the Party luminary,Maurice Cornforth, the article was printed with an appended replyfrom Cornforth, who, by quoting Lenin ("The fundamental premiseof materialism is the recognition of the external world, of the existenceof things outside and independent of the mind"), concludes thatSohn-Rethel "cannot claim the authority of Marxism for this positiv-istic critique of Marxist views" . Cornforth establishes the 'authority ofMarxism' without a single reference to Marx, and instead of reflectingon the Marxian proposition (quoted by Sohn-Rethel) that it is men'ssocial being that determines their consciousness, Cornforth discusses asignificantly impoverished proposition : "existence determines consciousness". Thereby, Cornforth reasserts without a moment's thought thevery proposition Sohn-Rethl had written the article against, and anydiscussion of the substance of Sohn-Rethel 's argument is precluded .

11 . In fact, Bahr's critique amounts to an accusation that Sohn-Rethel'stheory is marred by a helpless tangle of empiricism, idealism and dis-astrous silences : "On the one hand, Sohn-Rethel regards thoughtforms as 'arising' from exchange acts (whereby he merely identifies aproblem), and, on the other hand, he intercalates an - unnamed -act of reflection between commodity form and thought form . Yetreflection of one form in another medium presupposes the very under-standing that compares the real and reflected forms with one anotherin order to arrive at a judgement as to their formal identity ." Bahr .1973, pp .64f.

12 Apart from a reprint of Sohn-Rethel's "The dual economics of trans-ition", this collection includes reprints of seminal documents byTronti, Raniero Panzieri and Sergio Bologna .

Alfred Sohn-Rethel : a bibliography

1932 (anon.) "Die soziale Rekonsolidierung des Kapitalismus" in Deut-sche Fuhrerbriefe, 72 (16 Sept .) and 73 (20 Sept .) Rptd (with apostscript "Ein Kommentar nach 38 Jahren"), 1970a ; also rptd in1973b .

1936 Von der Analytik des Wirtschaftens zur Theorie der Volkswirt-schaft: Methodoloqische Untersuchunq mit besonderem Bezuq aufdie Theorie Schumpeters (Emdetten : Lechte). Rptd in 1978a .

1947/ "Materialism and its Advocacy" in Modern Quarterly, New Series,48

III, 1 (Winter) .1948 "Die politischen Buros der deutschen Grossindustrie" in Blick in

die Welt, 15 .1961 "Warenform and Denkform : Versuch einer Analyse des gesell-

schaftlichen Ursprungs des 'reinen Verstandes' " (includ . a resumein English) in Wissenschaftliche Zeitschrift der Humboldt-Univer-sitat zu Berlin, Gesellschafts-und Sprachwissenschaftliche Reihe,X, 2/3. Rptd in 1971e .

138

CAPITAL & CLASS

1965 "Historical Materialist Theory of Knowledge" in Marxism Today,IX, 4 (April) . Translated/expanded in 1971d .

1969 "Imperialism, the Era of Dual Economics: Suggestions for aMarxist Critique of 'Scientific Management' " in Praxis, V, 1/2 .

1970a "Die soziale Rekonsolidierung des Kapitalismus (September 1932)"in Kursbuch, 21 (Sept.) Rpt (with a postscript "Ein Kommentarnach 38 jahren") of 1932 .

1970b Geistige and korperliche Arbeit : Zur Theorie der gessellschaft-lichen Synthesis (Frankfurt: Suhrkamp) . 2nd edn (revised/ex-panded), 1972b ; English edn, 1978b .

1971 a "Die technische Intelligenz zwischen Kapitalismus and Sozialis-mus" in Neues Rotes Forum, 3 (June). Rptd (revised) in 1971d ;also reptd (revised), 1973c .

1971 b "Brief von Alfred Sohn-Rethel an die Redaktion des Neuen RotenForum" in Neues Rotes Forum, 4 (Oct.) Extract reptd in 1971 d .

1971 c "Aus Anlass van Joachim B i schoffs 'Materiel le and geisteige Produk-tion, Sohn-Rethels' "Siegeszug" durch die nichteveisionistischeLinke' " in Sozialistische Politik, III, 13 (Oct .) .

1971 d Materialistische Erkenntniskritik and Vergesellschaftung derArbeit: Zwei Aufsatze (Berlin : Merve) . Incl. translation/ expansionof 1965, rpt (revised) of 1971 a, and extract from 1971 b .

1971e Warenform and Denkform: Aufsatze (Frankfurt: Europaische Ver-lagsanstalt) . Incl. rpt of 1961 . Rptd (expanded), 1978a .

1972a "Mental and Manual Labour in Marxism" in Paul Walton a . StuartHall (eds), Situating Marx: evaluations and departures (London :Human Context Books) .

1972b Geistige and korperliche Arbeit : Zur Theorie dergesellschaftlichenSynthesis (Frankfurt : Suhrkamp) . 2nd . edn. (revised/expanded) of1970b. English edn, 1978b .

1972c "The dual economics of transition" in Bulletin of The Conferenceof Socialist Economists, II, 2 (Autumn). Rptd, 1976c .

1972d Die okonomische Doppelnatur des Spatkapitalismus (Neuwied :Luchterhand) .

1973a "Intellectual and manual labour : An attempt at a materialistictheory" in Radical Philosophy, 6 (Winter) .

1973b Okonomie and Klassenstruktur des deutschen Faschismus : Auf-zeichnungen and Analysen (Frankfurt: Suhrkamp) . Incl . rpt . of1932. English edn, 1978c .

1973c "Technische Intelligenz zwischen Kapitalismus and Sozialismus" inRichard Vahrenkamp (ed .), Technologie and Kapital (Frankfurt :Suhrkamp) . Rpt (revised) of 1971 a .

1974a "More on the Lin Piao-Confucius debate" in China Now, 42 (June) .1974b "Confucius and the changing family" in China Now, 43 (July/

August) .1974c "Die Formcharaktere der zweiten Natur" in Chris Bezzel a .o .

Das Unvermogen der Realitat : Beitrage zu enier anderen material-istischen Asthetik (Berlin : Wagenbach) .

1975a "Science as alienated consciousness" in Radical Science journal2/3 . This article is based on a paper given in 1974, which the RS]

REVIEW ARTICLE

139

collective, with the author's permission, completely rewrote ; theauthor's reservations vis-a-visthe result are contained in the"Author'sIntroduction" .

1975b "Eine Kritik der Kantschen Erkenntnistheorie" in Neues LotesFolum (sic), XXVII, 1 (May) .

1976a "Materialistische Erkenntnistheorie?" in Alternative, XIX, 106(Feb) .

1976b "Das Geld, die bare Munze des Apriori" in Paul Mattick a.o .,Beitrage zur Kritik des Geldes (Frankfurt: Suhrkamp) .

1976c "The dual economics of transition" in The Labour Process andClass Strategies, CSE Pamphlet no .] (London : stage 1) . Rpt of1972c .

1978a Warenform and Denkform : Mit zwei Anhangen (Frankfurt: Suhr-kamp) . Rpt (expanded) of 1971e ; incl . rpt of 1936 .

1978b Intellectual and Manual Labour: A Critique of Epistemology(London, Macmillan) . English edn of 1970b/1972b .

1978c Economy and Class Structure of German Fascism (London: CSEBooks) . English edn of 1973b .

Secondary ReferencesAdorno, Theodor W . (1973) . Negative Dialectics (London : RKP) .Bahr, Han-Dieter (1973) "Die Klassenstruktur der Maschinerie" in R .

Vahrenkamp (ed .) Technologie and Kapital (Frankfurt: Suhrkamp) .Halfmann, Jost and Rexroth, Tillman (1976) Marxismus GIs Erkennt-

niskritik (Munich: Hanser) .Thomson, George (1961) The First Philosophers (London, Lawrence &

Wishart) .

Capital and Land

Landownership by Capital in Great BritainDoreen Massey and Alejandrina CatalanoSocial Structure and Social Change Series

The pattern of landownership in Great Britain has undergone substantial changessince the war . The `land and property boom' of the early 1970s brought into sharperfocus questions concerning both this changing pattern and its economic and politicalimplications . This book collects together the available data on the present pattern oflandownership, and uses this as the basis of an analysis of landownership by capital inGreat Britain, and its role in relation to the economic structure as a whole . The bookexplores the changing nature of `the land problem' under capitalism, and whether ornot a distinct fraction of capital exists in Great Britain based on landownership. Theauthors argue that different types of private landowners pose different problems forcapitalism and go on to analyse the differential effect of recent legislation on thosedifferent types . The book contributes to the debate on the analysis of private land-ownership under capitalism . ;

® Edward Arnold41 Bedford Square, London WC 1B 3DQ

Cloth £9 .95 . Paper £3.50

REVIEWS

THE MAKING OF MARX'S 'CAPITAL'By Roman Rosdolsky, translated by Pete BurgessPluto (London, 1977), pp .581, £18

Reviewed by Simon Clarke and Ben Fine

1 The Making of 'Capital'. Simon ClarkeThe English publication of Rosdolsky's book is an important event .

The book has already had a strong influence, directly and indirectly, onthe development of Marxism in this country, most obviously in its in-spiration of the 'fundamentalist' stream of Marxism identified especiallywith David Yaffe, but also including thinkers like Ernest Mandel and PaulMattick . Fundamentalism involves an approach to Marx that regards theanalysis of Capital as something akin to an eternal truth, formulated at asufficient level of abstraction as to be untouched by history . While, as isargued below, this approach has serious weaknesses, it also has greatstrengths, for it constantly forces us back to examine and re-examineMarx's arguments before conceding revisionist claims, so many of whichare based on a failure to take Marx's own arguments seriously .

Rosdolsky's book is the product of twenty years work carried out in

EDITORIAL NOTE

Wal Suchting, one of the translators of the Marx article The Value-Form(the appendix to the first edition of volume 1 of Capital), which we pub-lished in Capital & Class 4, has asked us to point out that some of the ital-icisation, which is important to the article, is inaccurate . Also, the phrase"andre Ware" is translated on p .135 as "other commodities" where "an-other commodity" would be a better translation . These relatively minorinaccuracies were due to problems of communication between the trans-lators (who were in Germany and Australia at the time) and with theeditors of Capital & Class .

REVIEWS

141

the isolation of exile in the USA in a period in which even those fewMarxist theoreticians who survived had turned their backs on Marx's ownwork, and especially on Capital . As he notes in his Preface, he wrote thebook, ill-equipped as he was, because there was no better equipped totake on the task . The aim of the book is to reassert the centrality ofCapital in Marx's work and to affirm its continuing relevance to contem-porary capitalism. This is done firstly through avery extensive commentaryon the Grundrisse and secondly through a survey of debates that havearisen around Capital. The aim in the first case it to show that the Grund-risse anticipates, and sometimes complements, the argument of Capital,and in the second case is to reassert Marx's arguments against his critics .

Many of the arguments developed by Rosdolsky have already beenmade familiar to English readers, but Rosdolsky develops them with arigour and a scrupulous attention to Marx's texts that adds a new dimen-sion. Marx is not simply invoked as an authority, but rather Rosdolskyinsists that we read Marx for ourselves, to the extent that he often quotesat what some may feel is inordinate length, especially from the Grundrisse .The advantage of this procedure is that, because Rosdolsky lays his entirecase before us, we are able to interrogate his own interpretation, and touse Rosdolsky's book as a resource in developing our own understandingof Marx. While sophisticates may feel that much of Rosdolsky's account ispedestrian, he does change the emphasis of previous interpretations ofCapital in very significant ways, most notably in his focus on the questionof the relation between capital-in-general and many capitals and in hisinsistence on the centrality of the concept of use-value to Marxism . More-over his exposition of the Grundrisse is clear, though it would have bene-fitted from heavy editing, and includes some stimulating and even acuteobservations, while his cross-references to Capital, the Critique andTheories of Surplus Value are often very helpful .

The deficiences of Rosdolsky's book are the counterparts of itsmerits. Written in a period and a country in which to remain a Marxistrequired an unshakeable faith, this book is an expression of that faith .It was the faith of a few, such as Rosdolsky, that kept Marxism as acritique of capitalist society, distinct from the practice of those who pro-claimed themselves Marxist . It is works such as this, and those it hasinspired, that have provided the indispensable foundation on which thecontemporary renewal and revitalisation of Marxism can build . This is allthat Rosdolsky aspired to, as he indicates in his modest Preface . But whileRosdolsky's task was to sustain his faith in the unquestioned truth of thework he took it upon himself to preserve, it is our task now to questionand to develop that work in order to transform if from a fossil into a livingsocial force. Today it is once again possible and necessary to submitMarx's work to the test of history in order to give it new life . Todayfaith is not enough . Capital can no longer be seen as the sacred text, in-violate and inviolable, that contains within it all the secrets of the universe .While we must acknowledge our debt to Rosdolsky's fundamentalism, theway to do so is by moving beyond it and recognising its weaknesses .

Ben Fine takes up the weaknesses of fundamentalism in Rosdolsky's

142

CAPITAL & CLASS

treatment of specific theoretical issues . I would like to focus on its defic-iences in the interpretation of the Grundrisse . Since Rosdolsky's aim is todefend a Marxist 'orthodoxy' and to prevent the Grundrisse from beingcounterposed to Capital he seeks to show that the Grundrisse anticipatesCapital in every important respect . The differences between the two aretreated as being merely differences of presentation or of terminology andnot of substance . Thus Rosdolsky devotes the first fifty pages of his bookto a discussion of the structure of Marx's work that usefully surveys andconcludes past debates but that concentrates on presentation and not onsubstance . There is no discussion of the structural change between theGrundisse Marx starts with money and ends up with the commodity,Grundrisse Marx starts with money and ends up with the commodity,arguing in a manner that is certainly dialectical but that is also oftenidealist, in Capital Marx develops money out of the commodity in athoroughly materialist dialectical argument . This inversion of both argu-ment and method has enormous implications that fundamentalism misses .For fundamentalism remains idealist in believing that Marxism is based ona logical relation between essence and appearance, a relation that looks thesame from whichever side it is viewed . It fails to realise that in movingthrough the Grundrisse to Capital Marx developed a properly historicalmethod that made possible a properly materialist, and no longer purelyformal, analysis of capital .

The bulk of Rosdolsky's book consists of an exposition of theGrundrisse reorganised, more or less, according to the plan of Capital.This reflects Rosdolsky's belief that the difference between the two issimply one of presentation . While this approach is interesting for thecomparisons it throws up, it nevertheless suppresses the real differencesbetween the Grundrisse and Capital. In particular it destroys the co-herence of the Grundrisse so that the latter ceases to exist as a work in itsown right. Thus the main value of the Grundrisse to Marxists is lost, forsurely the exciting thing about the Grundrisse is that it is not simply a textthat anticipates Capital, it is the text in which we can actually see Marxat work, groping towards the solutions that would make Capital possible,developing the concepts that were to play a central role in Capital, asRosdolsky himself points out . Thus it is only in the course of the Grund-risse that Marx develops the distinction between labour and labour power,or the distinction between constant and variable as opposed tothat between fixed and circulating capital, that are central to Capital. Inthe Grundrisse Marx is only groping towards an understanding of thenature of capital . Without an analysis of the circulation of commodities andthe circulation of capital Marx does not clearly distinguish money fromcapital . Athough, as Rosdolsky emphasises, Marx takes the themes of theopposition of value and use-value as his guiding principle in both Capitaland the Grundrisse, the significance of the opposition changes . Thusinstead of seeing production as a twofold process, production of value andproduction of use-values, in the Grundrisse Marx tends to contrast pro-duction with circulation as use-value to value and so tends to see thefundamental contradiction of capitalist society as lying in the oppositionof production to circulation and not in production itself . This determines

REVIEWS

143

the theory of crisis, the understanding of socialism, and the importanceattached to Proudhonism in the Grundrisse . The absence of a series of keyconcepts from the Grundrisse, and the hesitancy with which others areapproached, takes us back to the starting point of the Grundrisse .

By starting with money instead of with the commodity Marx avoidsthe need to raise the question that is fundamental to Capital, the questionof the 'form of value' . It is only once this concept is developed that Marxcan develop the distinction between exchange value and value and so theconcept of abstract labour. It is correspondingly only when the historicalspecificity of value has been properly examined that Marx can rigorouslydistinguish use-value from value and so labour from labour power, fixedfrom constant capital etc . All this Rosdolsky misses because he is so intenton defending Marx's infallibility, because he does not treat Marxism as aliving theory, developing through Marx's work, and crying out to be fur-ther developed . While Rosdolsky recognises the absences from the Grund-risse, he treats the absence or incomplete development of concepts in theGrundrisse as mere terminological differences . Where whole arguments areabsent from the Grundrisse, Rosdolsky simply fills them in with referenceto later works. As an attempt to elucidate the making of Marx's CapitalRosdolsky's book must be judged a failure, for it never goes beyond con-sideration of the form of presentation of truths themselves consideredeternal and immutable . However to treat Rosdolsky as he treats Marx is todo him a disservice . As a testament to one man's faith the book is remark-able. Within its limitations the book contains many acute and novel insights .As a work of reference, as a source of erudite quotations, as a stimulus tofurther consideration of the development and originality of Marx'sthought, Rosdolsky's book is invaluable . Its availability in English, in avery clear and lucid translation, should go far to dispersing the aura thathas surrounded fundamentalism, if only it is treated as Rosdolsky wouldhave us treat it, as a basis on which to make Marx's theory "a living sourceboth of knowledge and the political practice which this knowledge directs" .

2 From Capital-in-General to Many Capitals . Ben FineIt is with some relief that I found myself able to review the parts of

Rosdolsky's book that were contributions in their own right to Marxisteconomics. For these are of considerable interest, whereas, with the oc-casional exception, his exegesis of the Grundrisse etc . i s so dominated bythe stringing together of quotations, that a reading of the originals is atleast as good as a reading of the book . Rosdolsky is the best representativeof the fundamentalist school . It is this which both determines the qualitiesand the limitations of his contributions . By fundamentalist, I mean thosethat draw the distinction between analysis at the level of capital-in-generaland that at the level of many capitals . Rosdolsky's achievement lies in thequality of his analysis of capital-in-general and his ability to draw the limitsof such an analysis . That is he poses the nature of the problems to bestudied at the level of analysis to be associated with many capitals (forexample, the transformation problem) . His limitations lie in the inabilityto undertake such studies, and to substitute a dogmatic assertion of the

144

CAPITAL & CLASS

validity of his interpretation of Marx's analysis for these (again the trans-formation problem but there are others) .

Let us begin with Rosdolsky's study of capital-in-general . His mostexcellent contribution concerns the status of use-value in Marx. For manyof us attempting to combat the dreaded spectre of individual utility theoryin economics teaching, it is easier to argue that the treatment of specificuse-values is absent from Marx's economics, since it would first require ananalysis of the economic and political relations which give rise to the ideo-logy that determines tastes . But Rosdolsky argues that specific utilities arepresent throughout capital, but as socially determined categories . The mostimportant is labour-power, but the others are to be associated with theforms of capital (money, productive, commodity, constant, variable,fixed, etc .) and the functions (i .e . use values) that they perform .

Rosdolsky also discusses the problem of the reduction of skilled tounskilled labour arguing correctly that in principle the reduction betweenthe two offers no more difficulties than the reduction of one form of un-skilled concrete labour to another. Beyond here, however, Rosdolsky doesnot recognise that analysis of capital-in-general has reached its limits andhe is reduced to some circular argument . Like others after him, he doesnot acknowledge that a more specific determination requires a specificationof the social relations in which the skilled labour is produced and used . Hewould have done well to have acknowledged the quotation from Marx herefers to elsewhere in the context of primitive accumulation 'the dialecticalmethod of presentation is only correct when it knows its limits' .

Before moving onto the limits posed by the fundamentalist method onRosdolsky's work, it is as well to examine critically the abstraction inmoving from capital-in-general to many capitals and, indeed the manylevels of abstraction that exist in Marx's economics . The point is thatmovement of analysis from the level of capital-in-general to many capitalsis not an undifferentiated concept . For the transformation problem itinvolves competition to equalise rates of profit. For the aggregate circul-ation of capital (and commodities) at the end of Volume II of Capital, itinvolves many capitals in the interaction of buying and selling but compet-ition in any form is not present . That the distinction between capital-in-general and many capitals is not simple is reflected in the other popularforms of locating levels of analysis - the abstraction that commodities ex-change at their value, abstracting from competition, abstraction from dis-tributional struggle, etc . Each of these in fact has different orders of sig-nificance and each cuts across the others in locating the level of analysis .Consequently, waving the magic wand of capital-in-general and manycapitals is a source of analytical confusion . For each object of analysis theproblem first has to be posed in terms of the many dimensions of abstrac-tion involved before it can possibly be solved .

Consider then Rosdolsky's treatment of the law of tendency of therate of profit to fall (TRPF) . He correctly identifies the tendency as ex-isting at the level of capital-in-general, the counteracting tendencies beingassociated with the movement of many capitals . But these counteractingtendencies exist therefore at different levels of analysis and must be anal-ysed as such (devaluation of capital, raising the rate of exploitation, foreign

REVIEWS

145

trade, etc .) . Marx borrowed his list of counteracting tendencies fromJ .S. Mill but in his chapter on the internal contradictions of the law of theTRPF, he only analysed those associated with the devaluation of capitalprecisely because these exist at a less complex level than the others . Wefind analysis of the other counteracting tendencies throughout Capital, butthey are not related to the Law as Such in a systematic way . This depend-ence by Marx on Mill for his "list" of counteracting tendencies was firstpointed out in English, as far as I know, in my article with Laurence Harrisin Socialist Register 1976 . Rosdolsky refers to Grossmann's note of thesimilarity in the latter's German work of 1929 (only reprinted in 1967) .The point of this is not to claim exegetical credit but to demonstrate thatMarx's list of undifferentiated counteracting tendencies is borrowed fromMill and consequently has not been integrated systematically into his owntheory and that this has escaped the majority of discussion of the law ofTRPF. However, this is strictly irrelevant for Rosdolsky (as for all funda-mentalists) for whom the level (this should be levels) of abstractiondividing capital-in-general from many capitals is identical with the order ofdetermination . The tendency must dominate the counteracting tendenciesbecause it exists at the level of capital-in-general . This is simply falsereasoning and false representation of Marx, whose concern is to analysethe Internal Contradictions of the Law not the quantitative effects on therate of profit (which in any case has not even been studied in the contextof the operation of all the counteracting tendencies) .

Rosdolsky's treatment of the reproduction schema is the best presentlyavailable to Marxism . For most of us, Rosa Luxemburg's criticisms ofMarx's reproduction schema are simply a tautologous dependence onunderconsumptionism . The production of surplus value necessarily createsthe revenue with which it can be purchased, even if disproportionality orthe separation of the acts of production and exchange do not render whatis logically possibly into what is actual . Thus, either consciously or other-wise, we accept some element of underconsumptionism, or we adopt Tugan-Bar-anovsky's scheme of equilibrium expansion as a logical rejoinder to theillogic of underconsumptionism. Rosdolsky adopts neither of these ap-proaches, rejecting Tugan's scheme because it cannot correspond to cap-italist reality . Why not is the question to be answered . Rosdolsky answersbecause capital accumulation does not produce sufficient demand torealise the surplus value it can produce, partly because the workers' con-sumption is limited (and this is clearly related to distributional struggle),and partly because capital does not produce means of production for thepurposes of producing more means of production (as is required by anincreasing organic composition) . But surely this is precisely the under-consumptionist thesis? The answer is no, because the effects of capital ac-cumulation are to be seen as the product at a complex level of the under-lying contradictions associated with the law of the TRPF, they are notaccidental effects of the anarchy of the market and distributional struggle .The point is that if underconsumption is theorised as in Luxemburg, thena solution is theorised as in Tugan . But both schemes of analysis misunder-stand the basis on which the problems of realisation arise . Capital can nomore solve the problems of realisation by increasing wages than it can by

146

CAPITAL & CLASS

decreasing them. The problem lies not in the spheres of distribution orexchange (effective demand) but in their interaction with and dependenceupon the accumulation process (fundamentally the expansion of capital inproduction) .

Perhaps I have read more into Rosdolsky than exists there . But as Isuggested at the beginning, he does pose the problems of analysing manycapitals (seen here as the relationship between production, distribution andexchange), he does not solve them . That we begin to do so is of increasingimportance, particularly in the context of the reproduction schema . Inthe modern world, with state economic intervention geared in part tomanipulation of effective demand and distributional struggle betweencapital and labour (incomes policy), Marxists are easily led to accept thesecategories of analysis uncritically . Consequently, the economy is deter-mined in their analysis by changes in distributional struggle and the levelof aggregate demand, but these are not constructed as categories fromanalysis of the accumulation of capital and its associated laws and struc-tures of abstraction and determination . Distributional struggle betweencapital and labour is based upon the reserve army, which is in turn deter-mined by the pace and nature (extent of expulsion of living labour) ofaccumulation, quite apart from the effect of trade union strength . Thesecannot be considered as simultaneous forces to be aggregated into a simpleunity pushing up the level of wages or not . More transparently, the level ofeffective demand, whilst often identified with class struggle over the levelof employment, has no such direct and simple correspondence . Indeed, itis a category which obscures the distinctive role played by capital, since itconfuses the expenditure of money as capital with its expenditure asrevenue, giving each an equal effectivity .

Rosdolsky discusses other topics critically - such as methodology,and the "Marxism" of Joan Robinson . Whilst informative and stimulatingthese do not go beyond nor surpass in quality other contributions in theseareas. For those looking for direct relevance of Rosdolsky's work to themodern world, there will be disappointment . On the rare occasions whenhe commits himself to statements that could not have been made by Marx,these are dogmatic and controversial (for example, money can only existultimately as a commodity embodying labour-time, the laws of value andplanning clash under socialism - following Preobrazhensky - but else-where the law of value does not exist under socialism). In my view, thisfailing of Rosdolsky is not imposed by the object of his book, but by itsmethod. Those constrained within the fundamentalist school prove againand again that they are unable to move to an understanding of the com-plex except by arbitrary theorising isolated from the laws of capital-in-general . It is Rosdolsky's achievement to have unconsciously demonstratedthis whilst elaborating the laws of capital-in-general and posing the problemof their concretisation .

REVIEWS

147

KARL KORSCH: REVOLUTIONARY THEORYEdited by Douglas KellnerUniversity of Texas Press (Austin & London, 1977), pp. 299, (11 .25

Reviewed by Phil Slater

The growing Anglo-Saxon interest in recent years in Karl Korsch(1886-1961) has been seriously hampered by the inaccessibility of manyof his most important writings; some have never been translated from theiroriginal German, while others, written in English, are buried in obscure,often defunct journals . Thus, while Marxism and Philosophy has, deserved-ly, received wide attention, it has usually been viewed in isolation from theKorschian opus as a whole . As a result, Korsch has either been assimilatedto the Lukacs of History and Class Consciousness, or else made the "hero/villain" in a retrospective legend according to which his entire theory andpractice were, from first to last, "ultra-left" . Buckmiller's serious attempt(1973) to rectify this picture has not (at least, to my knowledge) beenmade available to the English-speaking public .

In this situation, Douglas Kellner has decided against writing a bookon Korsch, and chosen instead to edit a selection of Korsch's writings thatadmirably reflects both the progression of political positions taken up byKorsch between the late 1910s and middle 1950s, as well as the vast rangeof burning political issues to which he addressed himself, and from which,in the light of historical experience, he drew the inspiration for any pro-spect of theoretical advance . These issues, further advance upon which willnow necessarily involve a serious consideration of Korsch's contributions,include: workers' control and the transition to socialism ; Social Demo-cratic reformism, anarcho-syndicalism, Leninism and Stalinism ; corporatestate, fascism and state-socialism ; Paris Commune, Weimar Republic,Spanish Civil War and New Deal USA . Kellner divides his selection into sixclear sections, each given a short, but lucid and informative introduction .

In a highly differentiated overall introduction which draws upon, andquotes from, Korsch's work as a whole, Kellner argues that the key to thisvast, complex and open-ended opus lies in what he calls Korsch's "revo-lutionary historicism", that is, the latter's concern "to derive his theoryfrom the requirements and possibilities of the historical situation" and tomake this theory "a material force in revolutionary struggle" (pp . 73-4) .Thus, it was not subjective inconsistency that turned Korsch into a Lenin-ist critic of the USSR, and finally into a critic of Leninism and Lenin him-self; rather, it was as a result of Korsch's truly remarkable sensitivity (de-veloped in and through political action) to the dangers of fetishising whathe called "the composite result of the class struggles of a previous historicalera" (p . 171) .

For Korsch, revolutionary theory must be dialectically related to thechanging "ground on which the workers' class struggle against capitalism"(p. 258) . For all his growing scepticism vis-a-vis a reified "Marxism", anddespite his attack on what he regards as Marx's own residual Jacobinism,

148

REVIEWS

Korsch remained faithful to Marx inasfar as the concern was to trace andexploit the dialectic between "the impasse that modern capitalism hasreached in the present phase of its historical development" (p . 254) and"what may be called, in a very comprehensive sense, the Marxist, that is,the independent revolutionary movement of the international workingclass" (p. 193) . It is this very dialectic that Korsch grappled with relent-lessly, and it is this same dialectic that, together with the de-dogmatisingprospect that it holds, was behind the recent renaissance of interest inKorsch, particularly (and indicatively) in Italy, as revealed in the short, butinformative review by Marramao (1975) .

But to pursue this dialectic today means, ultimately, to pass beyondKorsch's categorial apparatus, for in his incisive critique of the economismof the Second and Third Internationals, and in his liberating return to thenotion of class struggle as praxis, Korsch's own historical analyses reveal afundamental lack of economic depth in common with such diverse, butaffined writers as Sorel, Gramsci and the Frankfurt School . As early as1919, when Korsch was a pioneering proponent of the anti-mechanisticsignificance of the councils, there was no attempt on his part to relate thedeterminate historical role of this organisational form to the compositionof the German labour-force . Only thus could he write, after the destructionof the German councils movement, that "on the day of revolutionaryaction, the councils will again rise like the phoenix from its ashes" (p.21) .Thus, in the context of a consideration of the significance today ofKorsch's early writings, a reading of Bologna (1976) will suggest a funda-mentally critical appraisal, a suggestion already put into practice with spe-cific reference to Korsch by Paolo Perulli, who is quoted in Marramao(1975) . Now it is true that the principle of "revolutionary historicism"prevented Korsch from parroting the slogan of "all power to the councils"in subsequent years, and the criticism of him for basing his revolutionarystrategy on the "professional worker" is therefore of limited validity . Theunfortunate thing, however, is that despite his move away from any dog-matic model of the councils, and despite his serious consideration, again inthe light of historical experience, of such organisational forms as the tradesunions and the anarchist collectives, Korsch's writings on the advancedcapitalist and state-socialist economies failed to locate the changing formsof class struggle in the labour process . This failure to live up to his ownprinciple of relating theory dialectically to the changing terrain of classstruggle means that Korsch is unable to free himself entirely from thatvery fetishisation of living forces that he himself criticises so convincingly ;and he can actually reveal traces of that residual "Jacobinism" he claims todetect in Marx . For example, Korsch speaks in 1935 of "the destruction ofall remains of an independent proletarian class movement and even a classconsciousness through fascism" (p. 166) : implicitly hypostatising a histor-ically determinate form of proletarian movement and consciousness,Korsch views the political consolidation of the fascist state as the negationof all proletarian movement and consciousness, thereby remaining obliviousto the task of analysing the new composition of the labour-force and the

REVIEWS

149

new forms of its consciousness and struggle . In this regard, a critical re-ception of Korsch's work would benefit greatly from a comradely con-frontation with Karl Heinz Roth's as yet (at least, to my knowledge) un-translated Die 'andere"Arbeiterbewegung (1974) .

On the other hand, Roth's (as well as Bologna's and Perulli's) conceptsof recomposition, massification, autonomy, etc ., are the concepts of atheorisation based on active participation in a revolutionary struggle post-dating Korsch's work, which was effectively brought to an end in the late1950s . It would be a sterile and abstract undertaking simply to measureKorsch against the achievements of a subsequent practical-theoretical con-stellation, particularly in view of the fact that many organisations on therevolutionary left today cling to what Korsch called "the composite resultof the class struggles of a previous historical era" perhaps even more con-vulsively than was the case when Korsch wrote . In this situation, Korsch'swritings can play an important role in challenging dogmatism and thema-tising the need to relate theory dialectically to the ever changing "groundon which the workers' class struggle against capitalism" . All in all, this wel-come edition of Korsch's writings offers the chance of setting the spirit of"revolutionary historicism" to work on the theory and practice of thepost-Korschian left, and vice versa. To this end, Kellner's selection willserve as the indispensable basis for the Anglo-Saxon debate .

REFERENCES

Bologna, Sergio (1976) "Class composition and the theory of the party atthe origin of the workers councils movement" in The Labour Process& C/ass Strategies, CSE Pamphlet, 1, London .

Buckmiller, Michael (1973) "Marxismus als Realitat" in Uber Karl Korsch,Jahrbuch Arbeiterbewegung, 1, ed . C. Pozzoli, Frankfurt, Fischer .

Marramao, Giacomo (1975) "Korsch in Italy" in Te/os, 26 .Roth, Karl Heinz (1974) Die "andere"Arbeiterbewegung, Munich,

Trikont .

THE END OF PROSPERITY: THE AMERICAN ECONOMY IN THE1970sby Harry Magdoff and Paul M . SweezyMonthly Review Press (New York and London, 1977) pp. 136. £4 .75

Reviewed by Ben Fine

This book is a collection of ten essays written over the past five yearsand published in Monthly Review . It attempts to chart and explain the de-velopment of recession in the United States during the 1970s . For Sweezy,who has argued the underconsumptionist thesis with increasing vigour overa period now extending for more than thirty years, the recession should

150

REVIEWS

surely have proved to be the world's kindest response to his analysis . Para-doxically, just as Sweezy's thesis of tendency to stagnation attained itsheight of popularity during the "post-war boom", most notably withMonopoly Capital, so its decline has accompanied that of the world eco-nomy. Underconsumptionism has become the most frequently dismissedtheory of modern capitalism . On its errors we can all agree . Or can we?The elements of explanation in this book are all too familiar, and to befound both in Marxist and bourgeois analysis otherwise presumed removedfrom and critical of underconsumptionism .

Underlying all of the essays are the propositions of political economyassociated with the Monthly Review school. Monopoly and competitionare seen as anti-thetical, abolishing in one coup de plume the laws ofmotion of capitalism discovered by Marx and creating in their place anunder-investing and hence stagnating capitalism . Essentially the only socialconstraint on the operation of monopolies is the overall level of demandfor products. The failure of effective demand however becomes displacedinto an over-expansion of the credit system . This is the novel emphasis ofthe school in deference to the inflation of the 1970s . On the one hand,credit is expanded to boost effective demand and stave off recession. Onthe other hand, it is lavished on companies, unprofitable through demanddeficiency, because the consequences of bankruptcy in terms of loans fore-gone would far outweigh the cost of advancing an additional loan . It is acase of bad credit thrown after bad . Similarly credit is otherwise throwninto speculative adventures in the absence of real profitable outlets .

But what makes bad credit bad? If credit can be and is used to expandeffective demand, why should it fail to do so and lead to inflation whenthe economy is simultaneously characterised by chronic under-utilisationof capacity? These problems constitute the theoretical and practical crisisof Keynesianism, as is recognised by the authors . It is a crisis that cannot,however, be resolved theoretically, as the authors do, by adopting theKeynesian framework and simply asserting that bad credit is over-ex-panded whilst good production is not. (In fact recourse is made to that oldsubstitute for lack of argument-analogy, for example, punctured tyres ex-panding in size and pressure whilst simultaneously aggravating the deflatingpuncture . Better to have examined the engine?) . It does have the advantagethough of interpreting all events within a fool-proof framework . Thecharted levels of unemployment and over-capacity correspond to stag-nation, those of credit expansion to inflation (as "creeping stagnation"and "banks: skating on thin ice" bring "Keynesian chickens . . . home toroost") . To this can be added the cliches of modern political economy,that any state expenditure (e .g . the Vietnam war) is inflationary and anybalance of payment or exchange rate "crisis" demonstrates a weakening ofhegemony. If they do, neither in this book nor elsewhere have these pro-positions been given any satisfactory theoretical status, nor can they beuntil freed from a Keynesian problematic .

REVIEWS

151

MARXIST ECONOMICS FOR SOCIALISTS : A CRITIQUE OFREFORMISMBy John HarrisonPluto (London, 1978) pp .169, £2.40 pb ., £4 .80 hb .

Reviewed by Harry Newton

It is a truism, often overlooked, that every generation must be edu-cated anew. This applies with great force to the education of Socialists .Knowledge is not won on a once and forever basis . John Harrison's book istherefore an essential and perhaps long-overdue contribution to thetheoretical arming of the present generation of Socialist students .

The book not only explains in a clear and concise manner the groundplan and key concepts of Marxian economics but goes on to give a workingapplication of the theory to post-war capitalism . This application of thetheory demonstrates John Harrison's main strength which is that he under-stands, what so many alleged Marxists fail to understand, that all Marxisttheory and writing must be a positive guide to action . He keeps in mindthat "many men have philosophised about society but the point, however,is to change it . The difficult material in Part One of the book never be-comes boring or sterile . Part One makes clear that the true title of thesubject matter of the study is Political Economy, that is, a study of socialrelationships of power, class power, who gets what out of social produc-tion, and not a study of economics as defined by non-Marxist economistswhich is a futile exercise in supply and demand curves . The move fromMarx's concentration on production to the non-Marxist concentration onmarket distribution serving the purpose of concealing exploitation and theextraction from the working-class of surplus value . The Introduction andPart One of John's book strips the mask from the concealment .

Part Two : "Capitalism and the Productive Forces : the Critique ofSocial Democracy" traces the origin and development of capitalism . Theprogressive role of capitalism in organising the forces of production andthereby laying the base for the potential abundance of productionthereby laying the base for the potential abundance of production uponwhich communism can be achieved is well presented . The inner crisisover over-accumulation endemic in capitalism and its manifestation inworkers' movements and class struggle is explained as posing for theworking-class the choice of reform or revolution. Whilst 'reform' was, andstill is, a non-solution in real terms, the economic theories of John May-nard Keynes, especially his General Theory of Employment, Interest andMoney presented capitalist apologists outside and inside the labour move-ment with a theoretical fig-leaf to cover their intellectual exposure .J .M. Keynes promised the ruling-class that his theory would stabilise cap-italism. Leaders from the 'aristocracy of labour' promised the labourmovement that with the help of Keynes it would be possible to achievesocialism without the dreaded class-struggle . The more cautious of thecapitalist academics timidly asked Keynes if in the long-run his monetaryschemes would be inflationary, to which Keynes boldly replied "in thelong-run we will all be dead" . Keynes was writing in the capitalist crisis of

152 CAPITAL & CLASS

the 1930s, the outbreak of World War Two followed by the post-war re-construction boom offset the more acute manifestation of the chroniccrisis of the capitalist system .

The book investigates, in detail, the above mentioned period andconcludes with an annotated Guide to Further Reading . In conclusion mayI say that this book carries the hall-mark of the expert writer, i .e . it can beread at all levels of sophistication, the beginner will find a clear and rel-iable outline, the intermediate student will be guided via the reading listinto the deeper water and the experienced Marxist will find what we haveall been waiting for - the beginning of a serious Marxist treatment of post-war capitalism . If the last few sentences suggest a book which is all thingsto all men this should not be taken to suggest, also, political opportunism .The political integrity is beyond question .

kllamssnJournal of revolutionary socialists of the Middle East

khamsin S available nowThe first English-language editiontakes as its main theme : Oriental Jewryand includes articles on `Zionismand its Oriental subjects', 'EgyptianJewry -why it declined' and `Thedevelopment of class struggle inEgypt' .

Subscriptions :individuals £6.50, institutions £8for4issues . Alicheques/postalorders payable to Khamsin8 Honiton Road, London NW6

khamsin 6 available Aug 78with its main theme : Women in theMiddle East and includes articleson 'Women and politics in Lebanon','Islam and women', 'Problems andperspectives of women's emancipationin the Middle East' and 'Palestinianwomen in Israel' .

Single copies£2 from bookshops and direct fromPluto Press Unit 10 Spencer Court,7 Chalcot Road, London N W 18 LHTelephone 01-722 0141

ECONOMY AND CLASS STRUCTURE OF GERMAN FASCISMAlfred Sohn-Rethel

"The shattering defeat of the German working class at the hands ofAdolf Hitler is an event so momentous and awesome that historians,particularly on the Left, have often failed to consider the other side ofthe equation; and have not asked how it was that German Big Business,most of which held the Nazis in deep contempt, were forced to support(and indeed encourage) the seizure of power by a hysterical Austrianmystic at the head of a ragbag terrorist army .

Alfred Sohn-Rethel's book is important simply because it confronts thisknotty question . . ." David Edgar .£5 .95 cloth

TECHNOLOGY AND TOIL : Work and Capital in the Industrial Revo-lution edited by Maxine Berg

'Capital,' wrote Adam Smith, 'is command over labour .' This collectionof documents, provides acute insights into how this control was esta-blished - and fought against - in the 'industrial revolution' which tookplace in Britain from the late 18th to the late 19th centuries . The col-lection includes selections from early management theory on factoryorganisation, descriptions of typical labour processes, and accounts ofworkers' organisation against the capitalist division of labour . MaxineBerg of the CSE Labour Process Historians Group, has set each selec-tion in context and provided an historical and methodological intro-duction for the collection as a whole .£8 .95 cloth

CSE Books, 55, Mount Pleasant, London W .C .1 .

CSE bookclubJUST OUT

INTELLECTUAL AND MANUAL LABOUR : A CRITIQUE OFEPISTEMOLOGYAlfred Sohn-Rethel, was Professor of Epistemology and Social Theory atBremen University, West Germany until his recent retirement .

II socialism is to present an alternative to technocracy, if society is to gain control over technologyrather than the reverse, then one must establish, as a theoretical precondition, that srieni edependson social history for its very origin and its logic .

This hook demonstrates that the conceptual form of thinking of philosophy and science can betraced outside the immanency of the mind and, in fact, to what Marx calls the 'commodityabstraction' which is the key to the formation of societies in which production is carried on forexchange . I his is the novel element in this book : the view that exchange is the vehicle of a historicalprocess in time and space, by the agency of human action, not human thought .

I he theory argued in this hook sees development forced in the direction of production operated ona social scale substituting the capitalist system of private appropriation by the resources ofreunified intellectual and manual labour .

publisher's price £4 .95CSE bookclub price £2 .95

JUST OUT

INDUSTRY AND LABOURCLASS STRUGGLE OF WORK AND MONOPOLY CAPITALISMAndrew L. Friedman, Lecturer in Economics, University of Bristol

Throughout the history of capitalism radicals have been occupied with the struggle againstexploitation at work . Although Marx recognised the importance of worker resistance, he did notsystematically study the possibility that changes might occur under capitalism in response to thecontradictory or self-destructive forces created by capitalist accumulation .

In this hook Andrew Friedman analyses the capitalist mode of production in general, starting fromMarx's framework as set out in Capital. He examines the dav-to-day operation of workerresistance and managerial counterpressure . and also helps to explain why areas of deprivationcontinue to exist beside areas of prosperity in all advanced capitalist countries, despite concertedattempts by governments to remove the 'pockets' of deprivation .

publisher's price £4 .95

offer open in UK andCSE bookclub price £2 .95

Republic of Ireland only

Prices include postage and packing. Cheques/ postal orders to be made outto CSE and sent to:

CSE, 55 Mount Pleasant, London WCI

Please indicate amount of cash enclosed and address to which the order isto be despatched .

Capital & Class. - 1978. - Issue 6

Documents

cse localand

years conference

participatory conference

cse thatcapitaand classlags

years annual conference

cse workingand local

political tendenciesand

political activists