This article was downloaded by: ["Queen's University Libraries, Kingston"] On: 31 May 2013, At: 19:07 Publisher: Routledge Informa Ltd Registered in England and Wales Registered Number: 1072954 Registered office: Mortimer House, 37-41 Mortimer Street, London W1T 3JH, UK
The Journal of Peasant Studies Publication details, including instructions for authors and subscription information: http://www.tandfonline.com/loi/fjps20
Freedom, equality, property and Bentham: The debate over unfree labour J. Mohan Rao
a
a
Department of Economics, University of Massachusetts, Amherst, MA, 01003, USA Email: Published online: 05 Feb 2008.
To cite this article: J. Mohan Rao (1999): Freedom, equality, property and Bentham: The debate over unfree labour, The Journal of Peasant Studies, 27:1, 97-127 To link to this article: http://dx.doi.org/10.1080/03066159908438726
PLEASE SCROLL DOWN FOR ARTICLE Full terms and conditions of use: http://www.tandfonline.com/page/ terms-and-conditions This article may be used for research, teaching, and private study purposes. Any substantial or systematic reproduction, redistribution, reselling, loan, sub-licensing, systematic supply, or distribution in any form to anyone is expressly forbidden. The publisher does not give any warranty express or implied or make any representation that the contents will be complete or accurate
Downloaded by ["Queen's University Libraries, Kingston"] at 19:07 31 May 2013
or up to date. The accuracy of any instructions, formulae, and drug doses should be independently verified with primary sources. The publisher shall not be liable for any loss, actions, claims, proceedings, demand, or costs or damages whatsoever or howsoever caused arising directly or indirectly in connection with or arising out of the use of this material.
Downloaded by ["Queen's University Libraries, Kingston"] at 19:07 31 May 2013
Review Articles
Freedom, Equality, Property and Bentham: The Debate Over Unfree Labour J. MOHAN RAO
Definitional disputes may be disregarded as inconsequential aside from getting in the way of communicating substantive positions. In the present instance, such an attitude seems untenable if only because Free Labour and Capitalism are big words in wide currency. And debates involving these terms reflect deeper differences in theoretical and historical interpretation. Though, or perhaps because, a considerable part of the volume in review is energised by such disputes, it makes an eminently valuable contribution and provokes many substantive questions relating to labour and class relations, both contemporary and especially historical. While opinion may vary whether, on balance, the eighteen wide-ranging case studies in the volume shed useful light on the categories in contention and vice versa, their collective value transcends the debates themselves. Free and Unfree Labour: The Debate Continues, edited by Tom Brass and Marcel van der Linden, Bern: Peter Lang, 1997. Volume 5 in the series International and Comparative Social History. Pp.602. £52/US$78.95 (hardback). ISBN 3 906756 87 4; 0 8204 3424 8 (US) This sphere that we are deserting, within whose boundaries the sale and purchase of labour-power goes on, is in fact a very Eden of the innate rights of man. There alone rule Freedom, Equality, Property and Bentham [Marx, 1976a [1867]: 176]. J. Mohan Rao, Department of Economics, University of Massachusetts, Amherst, MA 01003, USA. e-mail:
[email protected] ass.edu. The Journal of Peasant Studies, Vol.27, No.1, October 1999, pp.97-127 PUBLISHED BY FRANK CASS, LONDON
98
THE JOURNAL OF PEASANT STUDIES
Downloaded by ["Queen's University Libraries, Kingston"] at 19:07 31 May 2013
I. FREEDOMS Work is not merely the characteristic form of human expression but also the critical imperative of human existence. The history of most 'civilised' societies reveals sharp disjuncture between the rewards of work and the compulsions to work. This disjuncture is importantly shaped by social institutions, not just by variations in nature. The compulsions to labour among dependent classes, best seen by contrast with the choices in leisure of the idle classes, are enforced through laws regarding status and legal obligations, non-legal institutional sanctions, religious or cultural norms and beliefs, hierarchies within the workplace, and, through material rewards from work and the material threat of not finding work. Class differentiation and coerced labour, rather than an egalitarian order and uncoerced labour, have been the rule. Class differentiation is itself at least partly based on the rules of property which must also be enforced. In this light, the notion of 'free labour' appears to be a contradiction in terms. Moreover, it seems obvious that no simple dichotomy between free and unfree labour could possibly do justice to the actual arrangements through which labour (and a surplus above subsistence) has been extracted from reluctant workers. Rather, there appear to be a variety of unfreedoms at work whose negations represent a corresponding variety of freedoms. Our understanding of these unfreedoms must also confront the distinction, often insidious, between the individual and the community, and the alleged 'western' provenance of worker rights, and also the omission, often blatant, of the linkages between patriarchy and property. Yet, Marx made the bold distinction between free and unfree labour a central part of his analytical framework for capitalism. Similarly, many adherents of free market economics have little difficulty in differentiating between, for example, serfdom and free wage labour. Although Marx used the dichotomy with ironic effect, and some neo-classical economists champion free markets as nirvana, both refer to free labour in the restricted 'bourgeois' sense of a negative freedom to dispose of one's own labour power without socio-politico-juridical restrictions. Marx's irony, targeted at bourgeois economics, derives from a wider, richer conception of positive freedom involving the participatory self-determination of the full range of conditions obtaining in and out of work.1 Further differentiation among various forms of unfree labour informs his analysis of pre-capitalist modes of production. At first glance, the very publication of the 600-page volume under review, devoted to theoretical and historical investigations of free and unfree labour, must appear to vindicate the value of the distinction. But dissenters from the classical view are very well represented within the
Downloaded by ["Queen's University Libraries, Kingston"] at 19:07 31 May 2013
THE DEBATE OVER UNFREE LABOUR
99
volume itself. For example, Kerr, who is otherwise sympathetic to Marxian analysis, sees both unfree and free labour to be subject to coercion and exploitation, and dismisses the significance of the dichotomy (p.425). Steinfeld and Engerman, adherents of neoclassical economics, find the distinction subjective and arbitrary. In their view, any reference to 'compulsion in labour relations' must imply 'situations in which the compelled party is only offered a choice between unpleasant alternatives and chooses the alternative which represents the lesser evil to Mm or her' (pp. 118-19, emphasis added). Generalised coercion or formal subjective choice-making are each seen as sufficient to render unfree labour logically arbitrary and historically irrelevant. While the former will have no truck with the negative freedom of 'free labour', the latter proclaim the omnipresence of freedom-as-subjective-choice. Is there, then, any use value to the bourgeois categories in question? Or are they just empty conceptual boxes designed to sustain mere polemical exchange? Besides free/unfree labour, the book is also centrally concerned with the other great classical tautology, that capitalism progresses in proportion to the growth of free wage labour. If capitalism is generalised commodity production, then the commodification of labour power is essential, not peripheral, to its development. Free wage labour, however, means not only free labour in the bourgeois sense but also labour dispossessed of the means of its subsistence. It is this double condition, Marx argued, that assures the commodity character of labour power. On both the theoretical and historical sides, a number of authors in the volume (obviously excluding those who see no purpose to the free/unfree distinction to begin with) deny any necessary link between capitalism and free labour. In their view, capitalist development has frequently involved the employment of dispossessed but unfree labour. On the premise that there has been a 'pervasive incidence of unfree wage labour throughout the history of capitalism', Brass proposes that 'in particular instances it might be more appropriate... to problematise the achievement of free wage labour rather than the existence of unfreedom' (p.60). On the one hand, this proposal affirms the value of the free/unfree distinction itself. On the other, by detaching capitalism from free labour, it raises the question whether, and where, a dividing line may be drawn between capitalism and other modes of production. Definitional disputes may be disregarded as inconsequential aside from getting in the way of communicating substantive positions. In the present instance, such an attitude seems untenable if only because Free Labour and Capitalism are big words in wide currency. And debates involving these terms reflect deeper differences in theoretical and historical interpretation. Though, or perhaps because, a considerable part of the volume in review is energised by such disputes, it makes an eminently valuable contribution and
Downloaded by ["Queen's University Libraries, Kingston"] at 19:07 31 May 2013
100
THE JOURNAL OF PEASANT STUDIES
provokes many substantive questions relating to labour and class relations, both contemporary and (especially) historical. As Brass points out in his Introduction, the collection is less a synthesis (least of all in definitional matters!) than an airing of several important disputes over theory, method and history. The debates over free and unfree labour, and over labour relations under capitalism, are cogently and even-handedly represented in conjunction with an excellent collection of historical studies. While there can be disagreement over whether, on balance, the historical studies shed useful light on the categories in contention and vice versa, their collective value transcends the debates themselves. The introductory piece by Brass sets up and clarifies a number of issues in the various papers while a concluding chapter by van der Linden gives a thoughtful account of the origins and spread of free wage labour. Part I of the volume, on theoretical issues, includes five papers (Kobben, Kossler, Steinfeld and Engerman, Roth, and Grossman) apart from two position papers by Lucassen and Brass written for a conference in 1995. The case studies on unfree labour, in Part II, are grouped under five regional headings: the United States and the Caribbean (Angelo, Krissman, Shlomowitz, Casanovas), Latin America (Johnson, Martins, McCreery, Pastore), Russia (van der Linden, Craveri), Asia (Olsen, Kerr, Baak), and Australasia (Munro, Markey). While there are seven chapters in Part I on theoretical issues, the logic of this arrangement is not entirely clear. Two of these chapters - on Nazi Germany and the Soviet Union - present both theoretical analyses and case studies but that holds good equally for many of the chapters placed in Part II. In fact, the article by Olsen, ostensibly on Indian agrarian relations, presents a view of neo-classical and Marxian alternatives which has broader theoretical significance. The extensive bibliography the book supplies should be of interest to both generalists and specialists. Whereas a wide range of themes is covered in the book, some rather specific to the regions studied (for example, slavery on the Amazonian frontier in contemporary Brazil), the issues of general interest may be ordered as follows: (1) objections to the value of the distinction between free labour and unfree labour; (2) the origins, incidence and demise of slavery; (3) capitalism and free labour (proletarianisation); (4) the relationship between capitalism and unfree labour; (5) rural indebtedness and unfree labour; and (5) the role of the state in coercing labour in ways more direct than through enforcing property rights. It is also the order in which the rest of this review essay is arranged. II. SELF-OWNERSHIP Some at least of the objections to the free/unfree labour dichotomy seem to arise from a misreading of the categories. Free labour obtains when the
Downloaded by ["Queen's University Libraries, Kingston"] at 19:07 31 May 2013
THE DEBATE OVER UNFREE LABOUR
101
worker is not, by means of social sanctions, coerced to supply her labour power or to remain in an employment. That is, the worker controls the disposal of her own labour power - self-ownership for short. Apart from peasants and other petty proprietors, self-owning wage labour is the canonical case. The worker is free to enter or quit the employment relationship at any time. The source of the social sanctions in the case of unfree labour may be the coercive authority of the state or of the community, and only those sanctions matter that are effectively enforced. Steinfeld and Engerman (S&E) deny the value of the free/unfree dichotomy on theoretical grounds. First, they object (as does Kerr) that since coercion of some kind has been ever present in labour relations, any such distinction is arbitrary. Second, S&E contend that all forms of coercion may ultimately be referred back to the law so that no useful distinction can be made between legal and economic compulsion. These arguments deserve closer scrutiny. Even if law alone were the source of all coercion (a proposition that is itself disputable), the free/unfree distinction has nothing to do with the polarity often drawn between economic coercion and legal or non-economic coercion. After all, the so-called economic coercion ('seek work or perish') of legally free (that is, self-owning) and dispossessed workers is itself sustained by legal backing to property in the means of production which makes such workers dispossessed in the first place. Rather, 'free/unfree labour' refers only to laws (or norms) regulating property rights in workers' labour power. Freedom and Property under capitalism stand, respectively, for legal backing for self-ownership by workers, and for ownership of the means of production by those that own them. While Freedom is 'nothing but' property in one's own labour power, distinguishing the two draws attention to the centrality of labour in production and to the varied coercions exerted by different configurations of property rights. For example, though dispossessed just as proletarians are, slaves also lack Freedom because they are the property of their owners. Whereas proletarians, moreover, are collectively coerced (with regard to their disposal of labour power), slaves are both jointly and severally coerced. S&E offer their own proposals for guiding historical research. First, since the free/unfree types are 'empty conceptual shells, our task becomes to offer historical accounts of labour relations in terms of specific coercive practices, all, in different ways, the product of law, rather than in terms of discontinuous "types" of labour' (p. 125). Second, the object of historical analysis should be 'to explain why a particular polity adopted the rules [of law] it did ... from among the rich variety of possibilities' (p. 125). In making the first of these proposals, they cite the example of wage labour and contract labour in the nineteenth century, purporting to show that there is no fixed content to what many researchers take to be clear-cut and
Downloaded by ["Queen's University Libraries, Kingston"] at 19:07 31 May 2013
102
THE JOURNAL OF PEASANT STUDIES
well-defined instances of free/unfree labour. S&E conclude that British wage labour in the nineteenth century was classically unfree whilst American contract labour in the same period was free. The former claim, especially, is surprising but this is best left for competent historians to judge. However, assuming the claim to be true, the categorisation involved is less than precise. Wage labour is not ipso facto self-owning labour nor contract labour unfree. As with the unfounded contrast between legal and economic coercion, these equations too have been made commonly. S&E's historical claim calls into question these unmindful equations rather than the free/unfree categories themselves. Besides, analysis entails judgments as to the fit between theoretical categories and the inevitably noisy raw data of history. For example, as noted above, the prevalence of unfree labour implies effectively enforced restraints on the labourer's self-ownership but determining whether a restraint is effective or not involves the observer's judgement. This methodological limitation applies equally to the categories in their second proposal - such as polity, rules, the state, enforceable laws and rights, etc. that S&E commend to us. Marx took pains to remind his readers of this methodological stricture: It is always the direct relationship of the owners of the conditions of production to the direct producers ... which reveals the innermost secret, the hidden basis of the entire social structure ... This does not prevent the same economic basis - the same from the standpoint of its main conditions - due to innumerable different empirical circumstances, natural environment, racial relations, external historical influences, etc. from showing infinite variations and gradations in appearance, which can be ascertained only by analysis of the empirically given circumstances [Marx, 1976b [1894]: 791-2]. There is also potential inconsistency between S&E's second proposal and their formal allegiance to an individualist choice-theoretic framework: formal because their substantive understanding of how social rules are determined seems much more nuanced. Thus, 'the rules governing social/economic life are established in ad hoc ways out of a variety of existing and new materials, and may be changed piecemeal, the outcomes depending on the relative political power of the participants, extant social views about permissible and impermissible forms of coercion, the normative persuasiveness of different groups, and a host of other factors' (p. 126). This betrays belief in a complex mixture of both negative and positive freedoms (see note 1) as the driving force of social choice 'from among the rich variety of possibilities'. One must doubt whether this could
Downloaded by ["Queen's University Libraries, Kingston"] at 19:07 31 May 2013
THE DEBATE OVER UNFREE LABOUR
103
be accommodated against historical evidence without stretching the rational egoism hypothesis to the point of a tautology. In effect, S&E's substantive rather than formal positions make a good deal of sense. Are they then justified in their active rejection of the classical free/unfree categories and, much more importantly, of the separability of the rules of property and the rules of the market? It is clear that Marx takes the exact opposite positions. The paradox is that no one was more concerned with the constraining effect of property-based power on market choices than was Marx; but as the theorist of capitalism, he found the free/unfree dichotomy a useful tool in dispelling commodity fetishism. The paradox also is that as neoclassical partisans, S&E must endorse the separability of markets and property that is the stuff of neoclassical practice; but as practising historians of 'unfree' labour, S&E must find that separation unhelpful in understanding, for example, the property-based imposition of slavery in the face of the market scarcity of self-owning labour. These paradoxical choices are methodological more than theoretical ones, informed less by deep differences of insight into the historical process than by the exigencies of descriptive convenience and analytical simplicity (perhaps also by normative concerns). Historical specificities matter but they must be combined with theoretical categories that provide the basis for an initial taxonomy and provisional hypotheses. True, the proof of the pudding is only in the eating and our categories must yield and adapt to advance our understanding. But from that standpoint, it seems premature to join S&E in jettisoning the categories of self-owning and unfree labour that have informed so much valuable historical research, especially the history of unfree labour, as well as the understandings of most peoples. III. SLAVERY Lucassen identifies three forms of labour in Europe, Africa and the Mediterranean over the two millennia since classical antiquity: independent workers (smallholders and artisans), unfree dependent workers and selfowning dependent workers. Their varied incidence and coexistence in time and space is one of the great facts of that broad sweep of history. Lucassen maintains that these three forms, where they coexisted, 'complemented' each other (p.53) but does not tell us what that might mean. While it might be argued that the existence of self-owning and unfree dependents, for example, was functional to the stability of the system as a whole, it might also be maintained that the division reduced the surplus appropriated. Other hypotheses that go beyond surplus appropriation and systemic stability may also be entertained. Thus, the differentiation in labour status may be related to the evolution and persistence of insider-outsider distinctions, distinctions
Downloaded by ["Queen's University Libraries, Kingston"] at 19:07 31 May 2013
104
THE JOURNAL OF PEASANT STUDIES
that transcend purely legal constructions and involve complex interaction between economy and culturally defined racial and ethnic categories, and among class, state and community. Attempts to link slavery directly with labour scarcity, originating in the work of Nieboer a century back, have found tantalising rather than definitive support. Bolland's question is central to any such purely 'materialist' explanation: 'How do we account for the fact that in some situations of labour shortage the workers are able to increase their benefits in terms of greater freedom and better remuneration, while in others the employers exercise even more powerful coercion, in the form of violence or bondage?' (cited in Lucassen, p.45). Kobben reexamines Nieboer's thesis that unfreedom is fundamentally due to the presence of open land, available for the taking. Nieboer's thesis was developed mainly for tribal societies. In extending it globally, however, Kobben argues that free land should be reinterpreted as part of a multicausal explanation in which its influence is enhanced or mitigated by other factors, particularly social stratification. Kobben cites work by Baks, Breman and Nooij who argue that only stratified societies, comprising two or more hereditary classes (excluding the slave class) that are unequal in status and power, can give rise to slavery whereas societies lacking such stratification would have only individual and non-hereditary differences. Based on a compilation of historical cases, they found that open resources and slavery were strongly correlated among stratified societies whereas no such correlation could be detected among egalitarian societies. This is indeed an interesting correlation in that the 'interaction' between material conditions and class structure (apart from slavery) is a critical conditioning factor so that the influence of labour scarcity is mediated by the macro social structure. But it also poses a problem in that Nieboer's original hypothesis was historically derived for 'tribal' societies which at least suggests the very opposite if we may take such societies to be unstratified virtually by definition. An alternative synthetic hypothesis suggests itself however. Slavery among tribes was probably driven by inter-tribal conquests and thus based on inter-tribal rather than intra-tribal differentiation. But where tribes or other egalitarian societies found themselves in the presence of stratified societies, enslaving conquests went in favour of the latter. This, in turn, may be due to stratification itself arising either from 'material' factors {land scarcity related to higher population density and therefore larger states) or from the accretion of conquests leading to progressive insider-outsider differentiation. But slavery by conquest is by no means the only historical 'type'. Drawing on Russian and east European history, Domar [1970] had
Downloaded by ["Queen's University Libraries, Kingston"] at 19:07 31 May 2013
THE DEBATE OVER UNFREE LABOUR
105
hypothesised that of the three elements, open land, 'free' (self-owning) labour and a surplus appropriating class, any two but not all three can coexist. In particular, free land can support surplus appropriation only if labour has been enserfed or enslaved. The main element of contingency he identified is the political factor, the willingness and ability of the superordinate class to impose unfreedom with political force. In Marxian parlance, the only safe generalisation would seem to be that slavery emerges from both 'external' and 'internal' contradictions with only an attenuated connection to labour scarcity. Apart from the availability of open resources, another often-claimed correlate of slavery or serfdom is the growth of long-distance trade. Yet, the same argument has been employed to show the demise of serfdom in western Europe and its second birth in eastern Europe. In relation to open land, one reason why Bolland's question is not easily answered is that 'labour scarcity' is not independent of the modes by which both production and the state are organised. This counter-argument seems to be even more potent against the 'commercialisation' hypothesis. The 'staples' or commercialisation hypothesis is the starting point of Pastore's analysis of slavery in early colonial Paraguay. Pastore does his best, moreover, to put a rational actor gloss on the troubled history of relations between the Spanish conquerors and indigenous peoples in the New World in general and Paraguay in particular. Both state and conquistadors are taken to be rational agents to explain the evolution of labour relations starting from primitive communism in the pre-conquest era, through slavery as a response to staples-led growth of demand, followed by a slavery-induced population collapse which was then met by a form of feudalism/serfdom, all under the tutelage of the Spanish crown. In short, colonised Paraguay eventually and tortuously converged on the politicaleconomic model of feudal absolutism that Spain itself had. As agents of the state which was to share in the spoils, private Spaniards took to enslaving the plains Indians with the rising demands for staples from the mining regions, itself a product of the conquests. Enslavement is seen as logically following from the abundance of lands to which the population had access. This ought to have benefited the crown as well. Yet, the toll of over-exploitation (not disease alone as Pastore makes abundantly clear) caused a vertical descent of the native population instead. The overexploitation is sought to be explained in terms of the greed and competition for slaves who came to be treated more as a common pool resource than as private capital. Not land alone but labour too was available for the taking. So here we have an instance of slavery, though a response to labour scarcity, actually exacerbating that scarcity. If the economy did not disappear into a black hole, this was because the crown eventually took matters into its own
Downloaded by ["Queen's University Libraries, Kingston"] at 19:07 31 May 2013
106
THE JOURNAL OF PEASANT STUDIES
hands and abolished slavery to stem depopulation. It then established a feudal labour-services model along Iberian lines (the encomiendas), and delegated police and military functions to the encomenderos. But the encomiendas of Paraguay took three different forms, each an experiment to address the contradiction thrown up by the others. Pastore does not ask why the unfortunate subjects had to be exposed to slavery and then decimation before the last-mentioned solution, rather solutions, were adopted. But his informative history suggests, at least in the mind of this reader, a plausible conjecture if not a manifest conclusion. The rationality of the conquerors was met by the rationality of the conquered leading not to any sense of an 'equilibrium' solution to the conflicting rationalities but to resistance in the small and in the large. That resistance coupled with the changing power balance between the crown and its conquering agents helps account for the conquerors' rational calculations going awry. But Pastore is disposed at the outset not to countenance any such explanation, preferring instead arguments founded in rational agents responding to factor endowments and market opportunities. This opposition is summed up, in relation to Robert Brenner, as follows: 'Brenner is sceptical of both population and trade-driven accounts of labour coercion, and prefers explanations based on class conflict. However, trade and population are both linked to relative scarcity and, therefore, to the development of property rights, without which Marxian classes or class conflict cannot be defined' (p.327, ff. 7). Even supposing this reduction of classes to relative scarcities to be true, it tells us nothing about the course of class conflict. What Pastore's own historical account yields effortlessly is the proposition that the conquest gave Paraguay what had been scarce prior to the conquest, that is, an exploiting class. Conquest itself preceded staplesdriven growth. For the rest, if a choice must be made between the primacy of individual rationality and of class conflict, Paraguayan history may be read more adequately as a history of class conflict. That seems more convincing than a tautological forcing into the Procrustean bed of rational egoism. In arguing the irrelevance of the free/unfree distinction, S&E invoke a sort of equivalence between the two: 'At various times, where labour was nominally free, attempts were made, by introducing taxation, limiting the land to be settled, increasing immigration, and other similar devices influencing labour incomes and land availability, to lower incomes and change the amount and/or nature of the work free workers were "willing" to do' (p. 110). In a footnote citing Engerman, they add: 'While individuals maintained legal property rights in themselves, controls over land and capital led to economic outcomes resembling those when property rights in persons belonged to others' (p.110, ff 5; emphasis added). In other words,
Downloaded by ["Queen's University Libraries, Kingston"] at 19:07 31 May 2013
THE DEBATE OVER UNFREE LABOUR
107
either the Mohammed of labour supply is forced up to meet the mountain of capital/land supply or the mountain is forced down to meet Mohammed to produce a perfectly 'equivalent' result. Leaving aside a formalism that evades the subjective experience of slave status, the argument still leaves two issues unaddressed. First, will all outcomes be equal under the two alternatives? Second, what explains which of the two alternatives is in fact socially chosen? Obviously, full equivalence need not hold. For example, while restrictions on land/capital supply can equalise the wage rate of self-owning labour and the rental rate of slave labour for any specified level of slave labour supply, surplus with slavery will be greater than with self-owning wage labour. On the other hand, if full equivalence does hold by other means, we are bereft of an explanation why slavery emerges at all rather than self-owning wage labour. While S&E would, consistent with their sensible substantive position described in the previous section, rely on a historical analysis of the political-economic-cultural determinants at play, why should variation in these determinants leave their consequences unaffected? IV. MANUMISSION The end of slavery in the west, Lucassen notes, has been explained on three distinct arguments: first, that it was abolished from above under the influence of Christian moral teachings; second, that it ceased when it became unprofitable; and third, that emancipation was won by revolutionary action of the unfree themselves. But are these arguments completely independent? Might not the profitability of unfree labour be influenced by resistance, not necessarily revolutionary action, from below? or ideological influences on those above also inform emancipatory struggles from below? Lucassen cites Engerman's argument that slavery was still expanding in the Americas at the time of abolition and was halted by military action (p.55). But it is unclear whether historians have generally ceased support for the unprofitability theory of the demise of slavery because they find the argument and evidence unpersuasive or because the evidence is simply not strong enough to conclude either way. One valuable reason for focussing on unfreedom is that it highlights the role of moral and political beliefs in structuring the economic system. For example, the abandonment of slavery in favour of serfdom in early medieval times and the abolitionist movement of recent times are both attributed by many to the Christianisation of the previously subject populations. When New World plantation owners resorted to slavery along with the growth of demand for colonial products (sugar initially), 'only African slaves were available at that time as all other traditional sources for
Downloaded by ["Queen's University Libraries, Kingston"] at 19:07 31 May 2013
108
THE JOURNAL OF PEASANT STUDIES
slavery in Europe were drying up. Europeans ... had by now all been converted to Christianity or Islam and for that reason could no longer be enslaved by coreligionists' (Lucassen, p.51). On the other hand, there is no necessary law or universal linkage here. The religious factor played little role in many countries including Italian cities of the early modern period, Denmark, Sweden, Holland, Portugal and Spain. Lucassen also cites Charles Verlinden who has argued that the prohibition of at least Christian slaves in Carolingian Europe was a consequence rather than a cause of changes in the streams of slaves available commercially. Casanovas's study of the end of urban slavery in Cuba is also instructive of the dynamics of resistance from below. Slave and self-owning labour coexisted in urban sectors in Cuba.2 This allowed Cuban urban elites to lower the living and working standards of Cuban self-owning labour as well as of slaves. Although rigid racial segregation was seen by the elite as a powerful tool in this effort, Casanovas argues that the building of ties between self-owning and slave labour, across race and ethnic lines, was a principal factor in the eventual elimination of slavery. This co-operation was crucial not only in the anti-slavery cause and anti-colonial struggles in Cuba but also helped eliminate the reliance on physical coercion inherited from the era of slavery to discipline all labour. This serves as a reminder for us that employers' resort to the physical coercion of free (self-owning) wage labour within the labour process does not contradict Freedom since the selfowning worker is free to quit work (and may be seen as choosing, in the words of Steinfeld and Engerman, 'the alternative which represents the lesser evil to him or her1). A central issue in the analysis of the transition from slavery has been the role of coercion within and outside 'market' exchanges. Shlomowitz uses the backdrop of this transition in the US South and the Caribbean to equate 'class analysis' with market collusion among employers plus non-economic coercion which have the effect of restricting labour mobility, tying down labour to particular employers and otherwise reducing wages. This is opposed to the market approach which emphasises the impossibility of market collusion, at least in the historical cases he refers to. This dichotomy ignores the fact that market analysis does not tell us how property rights get determined and also ignores coercion in the production process. Far from collusion and coercion being absent in the 'market model', these are rather segregated out of the Eden of commodity exchange into the sphere of the state and in the 'hidden abode of production'. Angelo takes a radically different viewpoint which emphasises coercion and collusion in and outside the 'markets' that got constructed during the transition in the post-bellum US South. Noting, correctly in our view, that free factor mobility of which self-owning labour is a critical element
Downloaded by ["Queen's University Libraries, Kingston"] at 19:07 31 May 2013
THE DEBATE OVER UNFREE LABOUR
109
'frames the analysis [of capitalism] for both Marxist and neo-classical economics' (p. 173), she asks why Southern wages persistently fell well below other regions of the US. With self-owning labour, why did labour in the South not move elsewhere in search of better options? Whereas the 'perfect markets school' of Southern history has insisted on such markets as the causes shaping the Southern economy, it has relied on given low productivity to explain Southern poverty while ignoring the fundamental question of arrested interregional labour mobility. The 'imperfect markets school' relates poverty to the imperfections of the market system. She finds that both approaches 'assume that market forces dominated the allocation of land, labour and credit without questioning whether the political and social institutions of the postbellum South were designed to create markets in these fundamental commodities' (p.175; emphasis added). Angelo uses historical evidence to argue that the lack of juridical equality between planters and labour, the use of physical coercion, and the severely unequal bargaining powers between the classes rendered labour unfree in a sense acceptable to both Marxian and neoclassical viewpoints. The arrangements that emerged did not even approximate a market in selfowning labour. Sharecropping not only made possible the annual contract which prevented the vicissitudes of seasonal labour scarcity from hitting the landowner. It also secured access to the labour power of the freedman's family by giving it the nominal cover of family farming. Changes in the law denied sharecroppers title to the crop they produced and thus denied independent access to credit on the crop lien system. This compelled borrowing from or through the landowner. Contract labour laws also rendered unpaid debts at the end of the season subject to criminal liability and established a presumption that they were a claim on the following year's crop cycle. The whole mechanism - laws, tenancy, and the crop lien requirements taken together - prevented the mobility of large segments of labourers and sharecroppers not only between the South and other regions but even from agriculture to industry within the South. These are important counterarguments against the perfect and imperfect markets schools. They would gain in strength if evidence or rationale can be adduced to make more precise how dependent labour was less than fully commodified by the arrangements. For example, does the fact that freedmen did move around among landowner-creditors make a difference? Similarly, in what way did sharecropping represent a collusive arrangement? V. PROLETARIANISATION The relationship between capitalism and unfree labour has been the source of much controversy and this is well represented in the volume under
Downloaded by ["Queen's University Libraries, Kingston"] at 19:07 31 May 2013
110
THE JOURNAL OF PEASANT STUDIES
review. Outside the sphere of commodity exchange, historically and actually existing capitalisms give ample proof of the use of politicaleconomic power to limit the substantive freedom of labour, rig the terms of the capital-labour bargain, and forestall serious challenges to the rules of the game. Power has been exercised through the state, in the marketplace and in the hidden abode of production. While these powers have also been amply resisted, to many observers it has seemed anything but obvious to suppose that capitalism is a system based on 'free' labour. What is not obvious may nonetheless be analytically useful. Capitalism may be characterised by three fundamental conditions: (a) labour free in the double sense of being self-owned and being dispossessed of its means of subsistence; (b) private ownership of the means of production; and (c) a substantial absence of non-economic constraints on commodity production and exchanges. These conditions are necessary and sufficient for generalised commodity production which Marx identified with capitalism. There are other definitions available, of course. Lucassen seems to reject self-ownership of labour as a criterion because he sees 'a strong relation between the occurrence of international economic development and unfree labour and that consequently unfree labour cannot be seen as a phenomenon restricted to some primitive or backward situations' (p.56). That is capitalism is international economic development. There are others (both the editors included) who see no antagonism, perhaps even symbiosis, between 'capitalism' and unfree labour. Though this is implicit, they tend to take capital accumulation as capitalism's defining feature. Capital accumulation and international economic development also fuse easily. Both reflect a focus on the 'logic' of accumulation counting on technology and trade as its motors. The choice among these definitions seems innocuous enough but, in practice, generates an analytical momentum predetermining questions and historical taxonomy. Perhaps, this is as it should be because the choice is probably always motivated by theoretical preconceptions. Ideas evidently are not quite like roses! At century's end, many parts of the world, particularly the more developed ones, would qualify as capitalist by all three definitions. But there are other parts and periods that do not, hence the controversy. Van der Linden asserts that both slavery and self-owning wage labour are fundamentally compatible with capitalism. Since the accumulation of capital (not including slaves) undoubtedly occurred under slavery, especially new world slavery, there really is no distinction to be made then between slavery and capitalism. Rather, bondage and freedom of labour are merely alternative instruments in the service of capital accumulation (capitalism), the 'logic' of which alone governs the choice. Apart from the functionalism embedded in this argument, it seems to leave no room for the
Downloaded by ["Queen's University Libraries, Kingston"] at 19:07 31 May 2013
THE DEBATE OVER UNFREE LABOUR
111
free/unfree distinction in shedding light on history. Having gone this far, van der Linden would have to include serfdom too in capitalism considering that from the twelfth century on at least, western Europe witnessed an efflorescence of commerce on the one hand and an active process of agricultural expansion through the opening up of new lands, a process that made no slight demands of capital. Van der Linden argues that the 'class analysis approach tends to become tautological rather quickly, i.e. it makes true by definition precisely that which needs to be investigated and explained' (p.503, note 6). But his summary shows that attempts to endogenise the mode of exploitation within capitalism also have an ex-post character. As we shall see below, van der Linden is surely incorrect in arguing that the endogeneity of capitalist class structure is preempted by Marx's definition of capitalism. While it should be granted that the emergence of doubly free labour and of the forms in which it is integrated into production (from putting-out and home work to piece-rated and wage-paid workers) must be investigated historically and analytically, van der Linden's position calls into question the usefulness of capitalism itself as a category. Besides, interchanging capitalism with capital accumulation is a different tautology than the one proffered by Marx but a tautology all the same. While not as contentious as van der Linden's position, Kerr's lukewarm attitude to the classical view is based on the observation that the achievement of self-owning wage labour has been reversible and uncertain. The observation is indeed accurate, but surely the classical definition does not guarantee the irreversibility of self-owning wage labour, at least at particular places and times? Nor does it imply that the commodification of labour power is not resisted by those controlling it in its uncommodified state. These may be slave-owners, feudal lords, or even corvee-exacting absolutist states. Or they may be petty proprietors and peasants. But we are unaware of instances where the resistance came from slaves, serfs, or corvee workers. Evidently, it would be useful to clarify the classical view further before we get to the historical case studies. Capitalism arose historically and primitive accumulation (the freeing of workers in the double sense) gets capitalism started. But its autonomous beginnings may be repeated; it is not started in one place and one time only. Moreover, the world is not conquered by capitalism in one fell swoop. Nor is this necessarily and everywhere a progressive and assured conquest. It is also a process that has potentially an 'intensive' (increased profits per worker) and an 'extensive' (inclusion of new sources of labour power) dimension. The extensive process has two important implications. First, the coexistence of rivals to the capitalist mode of production: workers under various forms of unfree labour and/or various degrees of dispossession of their means of subsistence. Second, when
Downloaded by ["Queen's University Libraries, Kingston"] at 19:07 31 May 2013
112
THE JOURNAL OF PEASANT STUDIES
extensive development proceeds successfully, capitalism overtakes and destroys its rivals, for example, feudal, communal, petty proprietor. This overtaking necessarily entails capture of new sources of labour to feed the accumulation process. Van der Linden asks the key question about this process: why has 'wage labour ... become the normal form of dependent labour in advanced capitalist societies' (p.519); though, adopting Marx's definition makes the qualifier 'capitalist' redundant. In short, how has capitalism proletarianised and then absorbed large parts of the world's labour force?3 We may answer without elaboration that the modalities of proletarianisation have included: (a) capital accumulation and productivity growth which enable capitalism to compete with rivals in product markets and thereby pauperise their selfowning labour, for example, self-owning peasants; (b) indirect influences from accumulation and growth which may hasten a process of dissolution of unfree labour, for example, women under patriarchy; (c) direct political assaults on precapitalist modes (primitive accumulation) from below or above, for example, various 'peasant' revolts and bourgeois revolutions, or from without, for example, imperialism. McCreery provides historical background to and a fascinating account of the transition to capitalist wage labour in Guatemala between the two wars. Accessibility to their own lands made the creation of a doubly free labour force among the Indians of Guatemala an uphill task for about four centuries. A mixture of coercive practices involving planters and the state therefore prevailed. The spread of coffee growing after 1871 weakened the incipient tendency toward self-owning labour, instead reviving and generalising labour based on direct coercion through state corvees on the one hand and coercion-backed debt peonage on the other. The 'attraction' of the latter to the Indian lay in its being the lesser evil as compared to the depredations of official corvees (being in debt to plantations exempted peasants from the corvee) and also because it gave Indians desirous of leaving communal lands an alternative source of subsistence. Moreover, there was enough competition of a sort among planters that exit from any particular plantation if not from debt and coercion altogether remained an option. The persistence of unfree labour relations (if they were indeed such) seems, at first glance, to be at odds with the persistence of communal property in Guatemala. The popular argument that peasant land holdings are functional to capitalism because they cheapen labour supplies and food supplies is logically incoherent (see Rao [1994]). The more apt reason which McCreery identifies is that 'indigenous peasant communities resisted commodification of their work and material resources' (p.323). From the 1920s onward, labour scarcity receded as population growth and ecological decline on the Indian lands compelled more and more
Downloaded by ["Queen's University Libraries, Kingston"] at 19:07 31 May 2013
THE DEBATE OVER UNFREE LABOUR
113
Indians to seek out work in the coffee plantations. This tendency was joined by the effects of reduced labour demand following the Great Depression. This account establishes, simultaneously and with great clarity, that (1) the vicissitudes of creating a self-owning wage labour force arise due to resistance from those controlling labour power in its uncommodified state (in this case, both communal peasants and semi-feudal classes); and (2) the state may play seemingly 'inconsistent' roles, imposing unfree labour relations on the one hand while being unable to privatise communal property on the other.4 VI. DEPROLETARIANISATION In view of the extensive aspect of capitalist advance, it is not a coincidence that much contemporary unfree labour, described in this volume, concentrates overwhelmingly on the capitalist periphery, especially in backward agrarian settings or where special circumstances such as illegal immigrant labour prevail. But several of the papers take the view that capitalism pursues profits employing unfree labour passively while some, notably Brass, also argue that capitalism actively converts self-owning labour into unfree labour. That such unfree labour relations 'occur within the process of the reproduction of capital on an extended scale, but not always immediately within the capitalist work process itself (Martins, p. 298) should give us pause to doubt whether the reproduction of capital has anything to do with them. Martins discusses Brazilian 'slavery' or rather temporary 'debt-peonage' in the Amazonian frontier. Data painstakingly put together from diverse survey and other sources allow Martins to estimate that up to a half million people found themselves in this labour form over the period 1970-85. This seems to be the cumulative total for the whole period; since terms of bondage were typically less than a year, one may suppose that the annual average of debt peons numbered 30,000. But this must surely count for a minuscule proportion of the total Brazilian labour force. Moreover, Martins describes the farms on which much of this bondage was to be found as enclaves 'governed by their own specific, although illegal, criteria of justice; places subject to the will of the landowners' (p.287). In short, the phenomenon being described seems both small-scale and a variant of serfdom in which political authority and economic power are fused together. In this light, his claim that 'today's slavery is a component of the very process of capital accumulation' (p.289) sounds implausible if it is taken to mean that such slavery is essential or even helpful to the reproduction of capitalism. Krissman's extensive study details the development of unfree labour relations in Californian agriculture. Non-white unfree labour, Krissman
Downloaded by ["Queen's University Libraries, Kingston"] at 19:07 31 May 2013
114
THE JOURNAL OF PEASANT STUDIES
argues, has been crucial to the development of many labour-intensive agricultural activities for two centuries. Seasonal labour shortages have provided the impetus to unfree labour. In an earlier phase, labour scarcity in the high season was due both to a large demand for labour and the availability of open resources to support independent agriculture, hunting and gathering by the non-white population. In this century up until the present, immigrants with no local options have been the main source of labour supplies. At first sight, the explanation for unfree labour seems circular: 'The unfree labour concept has proven useful in explaining the enigma of the use of unfree labour by dynamic and highly advanced capitalist crop industries' (p.235; emphasis added). As it happens, the effective denial of self-owning labour status to a series of non-white immigrant workers has been based on a denial of political freedom. The labour market has been, so to speak, driven underground both by immigration laws and by the use of race, ethnicity and citizenship to discriminate across the labour force. The benefits to 'capitalist' agriculture derive from the formal and informal repression that has kept down the current wage and other labour costs but also from the subsidy implicit in the fact that the labour force has been reproduced inter-generationally only in its home countries. Brass cites a number of his own theoretical and field studies in addition to those in the volume (including the ones just summarised) to present and support his main thesis concerning a process of deproletarianisation under capitalism. Before taking up his thesis, it will be useful to recount Brass's lucid if controversial dismissal of what he takes to be the three basic assumptions of much current understanding of third world rural labour: (1) that labour market imperfections symptomise peasant resistance to proletarianisation; (2) that capitalist penetration of agriculture necessarily transforms peasants into proletarians; and (3) that unfree relations are precapitalist forms of production. He targets three viewpoints within this literature. These are as follows. (1) Neo-classical economics accepts the presence of capitalism and redefines unfree labour as effective self-ownership. (2) The semi-feudal thesis inverts this by taking unfree labour, including 'debt-bondage' in particular (in India, Latin America and elsewhere), as forms of feudal or semi-feudal relations. (3) Postmodernist or culturalist arguments, such as survival strategies, moral economy and resistance theory, deny or dilute unfree labour. Brass's own view is that, in many instances including those noted above, 'instead of attempting to create a proletariat... capitalism was much rather trying to prevent this from taking place. In other words, it was not encouraging the formation of a proletariat but much rather using unfreedom
Downloaded by ["Queen's University Libraries, Kingston"] at 19:07 31 May 2013
THE DEBATE OVER UNFREE LABOUR
115
as a mechanism to prevent this from occurring ... by using forced labour' (p.36). Forced labour has been imposed on those already separated from the means of production and habituated to proletarian life. Such deproletarianisation serves not merely to raise capitalist profits immediately but also helps to curtail the development of a proletarian consciousness. Hence, it is functional to capital accumulation. In line with this thesis, Brass finds that modernisation theory, with its expectation that unfreedom will decline everywhere in proportion to the spread of capitalism, has not been vindicated. In particular, Brass sees the presence or creation of unfree labour with open resources as an example of deproletarianisation by capitalism designed to create a reliable labour supply where none is (voluntarily) forthcoming. This involves first the separation of workers from the means of production, and then their habituation to proletarian life by conditioning them through an unfree labour phase. Brass takes the case studies of Markey on nineteenth-century Australia, Martins for contemporary Brazil, and even McCreery on Guatemala as supporting his thesis. These arguments seem to imply 'capital accumulation ergo capitalism' by way of a definition of capitalism despite protestations to the contrary as when Brass explicitly allows that both aspects of wage labour - its separation from the means of production and its self-ownership - 'capture the difference between capitalism and pre-capitalist modes of production' (p.59). On the other hand, Brass is clear that there is a symbiosis between capitalism and unfree labour going so far as to state at one point that he regards unfree labour to be 'a capitalist and not a pre-capitalist relation' (p.19, note 2). That capital displays ability 'to engineer unfreedom in labour markets' (p.61) seems to be a fair statement, on balance, of his position. Is unfree labour necessary or even good for capitalist development? On the one hand, it all depends on what one means by 'capitalism' (and, as we shall see in the following section, also on the applicability of the selfowning labour category in the contemporary third world). By the classical criteria (see section V), this is a meaningless question. But by the criteria of capitalist deproletarianisation, capital accumulation, or the spread of global commodity production (or the modern world system), the question has meaning and all three posit unfree labour to be functional to 'capitalism' though the choice between self-owning and unfree labour forms has, it is argued, hinged on the 'logic' of capital accumulation. On the other hand, many take capitalism, by whatever criteria, to be the empirically dominant mode of production in many parts of the world (see also note 3). By implication, unfree labour is of second-order significance. Arguably, also, these definitional differences may matter more for historical taxonomy and for 'short-hand' rather than for 'long-hand'
Downloaded by ["Queen's University Libraries, Kingston"] at 19:07 31 May 2013
116
THE JOURNAL OF PEASANT STUDIES
historical analysis. But in practice, the definitional differences have tended to generate much heat and do seem to have an impact on the questions protagonists raise or consider relevant. All the same, it is necessary to point out that the assertion of consistency between capitalism and unfree labour poses problems for our understanding of capitalist competition. Recall the three modalities of proletarianisation discussed in section V which involve both political and economic factors. These modalities imply a relationship of competition or rivalry between capitalism and other modes, and this has been the long-term tendency. If profit-making based on unfree labour were a general modality if not necessity of capitalist development, then, two contradictions arise. First, the process of capitalist development itself becomes, as we have seen, historically ill-defined. Second, the process of capitalist competition in which a labour market with self-ownership may be the most important element will be attenuated to that extent and so, therefore, will be its entailments, including capital accumulation itself. However, competition for labour power does not preclude, in times and places, relationships of accommodation and/or parasitism from developing. Accommodation or compromise mediated politically, whether at national, local or even international levels, is usually effected by changing or reinforcing existing laws and state policies (rules of the game, for short) with the intention to change, if only temporarily, the terms on which capitalist competition (and therefore proletarianisation) takes place. Parasitism involves the transfer, within or across nations, whether through market exchange or via the state, of resources from non-capitalist modes to the capitalist mode, for example, gains from slavery and the slave trade feeding capitalist development or proletarianisation elsewhere, or certain types of unequal international or interregional exchange. But we should note that such transfers need hardly be unidirectional. Relationships of competition, accommodation and parasitism are by no means unique to the capitalist era. For example, the coexistence of slavery with both self-owning dependent and self-owning independent labour in the same socio-economic formation doubtless has involved similar relationships among those three modes.5 Angelo's analysis is a good example of what is at stake in these debates not only by way of definitions and taxonomy (she takes a classical view) but also by way of long-hand analysis. She shows how the political and social institutions of the post-emancipation American south failed to commodify labour power (in Brass's terms, labour was deproletarianised). But this was not an act of 'capital' so much as a consequence of state power and planter power seeking to shore up the class structure that emancipation by itself had rendered vulnerable to the freedmen's economic choices. She shows that even while making a surplus possible, these institutions, in the end, subverted capital accumulation and economic development.
THE DEBATE OVER UNFREE LABOUR
117
Downloaded by ["Queen's University Libraries, Kingston"] at 19:07 31 May 2013
VII. PEONAGE
A primary source of the debate surrounding the link between capitalism and unfree labour are the varied definitions of unfree labour itself. At its widest, the notion has ceased to be just an indicator of a particular type of negative freedom (self-ownership) and come to include elements of positive freedom as well, that is, social omissions and commissions that systematically weaken labour's bargaining power vis-a-vis capital. For example, Brass's view that capitalism thrives on and promotes unfree labour is due, in major part, to his uncommonly wide notion of the latter. He includes under that rubric such contemporary phenomena as the growth of extra-territorial production zones where legislation protecting workers does not apply and even the replacement, in advanced capitalist countries, of welfare provision with workfare (p.70). Equally relevant is the onerous problem of identifying, in practice, unfree labour even narrowly construed as self-ownership. One reason is that there remains, in large parts of the rural world, a substantial if uncertain disjuncture between the extent of unfree labour and that of labour that has not been dispossessed. Arguably, incomplete dispossession itself does not pose a problem to the classical viewpoint: it may be easily understood as the joint product of both resistance from workers and capitalism's inability (as yet) to establish its dominion completely rather than its 'will' or 'logic' to preserve semi-proletarians. While dispossession is objectively identified, unfree labour is another matter. The presence of such semi-dependent classes who have to trade in localised, fragmented and often monopolistic land, labour and credit 'markets', and are often deeply enmeshed in the local structures of both property and 'political' power, not to mention local ideologies of patronage and loyalty, makes the determination of unfree status empirically problematic. That is to say, the ambiguities that arise in identifying Freedom in these contexts are intimately tied to variations relating to Equality, Property and Bentham. Not surprisingly, researchers from the outside encounter difficulty in penetrating these thickets of doubtful Eden and come away with diverse categories and explanations. The tangle of both elastic definitions and elusive evidence is nowhere better seen than in the case of the persistent cycles of indebtedness, typically flowing from subsistence rather than production demands, in which large masses of both landless labourers and landed peasants, contemporary and historical, find or found themselves. The entry of a seemingly self-owning worker into a relationship based on debt that then compels him into a cycle of indebtedness serviced by the creditor's usufruct on his labour power constitutes the most important challenge to the practical application of the self-owning labour category.
Downloaded by ["Queen's University Libraries, Kingston"] at 19:07 31 May 2013
118
THE JOURNAL OF PEASANT STUDIES
Many take the view that it is indeed a form of unfree labour on the supposition that the worker once locked in finds it well-nigh impossible to escape it though there are also some, like Brass, who do not make even such lock-in necessary to qualify it as unfree labour. But the presumption that the worker is not coerced into debt by social sanctions and, the fact, to the contrary, that he is bound by contract to meet his obligation through mortgaging his future earnings seems only to be a working out of the 'capitalist' market imperatives of both Property and Freedom. Indeed, the relationship in backward rural contexts seems rather more 'free' than in advanced countries where laws preclude creditors from explicitly using future labour earnings as collateral (though they are often implicitly so used, for example, in housing loans) and where bankruptcy legislation gives the debtor-worker, in the event of his being 'insolvent' in terms of his current earnings and non-labour assets, a legal way out of his obligation. Even when present in third world settings, such protective laws are either unknown to the parties or effectively unenforceable since needy workers may be seen as voluntarily foregoing such 'protection'. The problem of evidence and interpretation, however, is scarcely reducible to the legal position. Apart from legal considerations, purely 'economic' models of indebtedness raise problems of their own. Capital in Ricardo's classical model of agrarian capitalism was, for all practical purposes, little more than the wages advanced to workers through the growing season and may, therefore, appear similar to credits extended to semi-proletarians and others in the rural third world. But the contracts are structured quite differently. The cost/return of/to Ricardian capital is borne/secured by the capitalist himself. As such, the Ricardian labour market is indistinguishable from a pure spot market in casual labour without capital. The difference is related to several observations: Ricardian workers, unlike many we are discussing here, are pure proletarians; the Ricardian labour market is unproblematically casual, and unlike real contexts, not powerfully localized in space or time; Ricardo did not have to worry about workers' own access to 'credit'; questions of enforcement in a casual market may be confined to the exogenous powers of the state which also Ricardo took for granted; and in view of all this, any question about workers' indebtednessbased bondage or lack of self-ownership does not arise. On the whole, neoclassical analysts too have assumed the selfownership of the contemporary rural debtor-worker as self-evident, whether or not the transaction ties labour and credit, or labour, land and credit, in a dyadic relationship with the same creditor. Neo-classical models improve upon the classical one by seeking to endogenise some elements while still leaving others within a black box of 'market imperfections'. Models may differ about the endogenous determination of contractual terms in the
Downloaded by ["Queen's University Libraries, Kingston"] at 19:07 31 May 2013
THE DEBATE OVER UNFREE LABOUR
119
absence of perfect enforceability of contracts due to the potential for wilful default by the borrower. That the future labour power of the debtor may be the basis of the loan is not, in these models, evidence of lack of selfownership not only because the debtor is free to enter any particular contract but also because, having entered it, he would be free to exit it provided he has met the terms of the contract, that is, paid the dues. The fact that many debtors cannot pay off the debt without borrowing again itself reflects calculated tradeoffs between present consumption and future. This 'voluntarism' may, however, be stretched to breaking point. But first consider a purely accounting point. Suppose the worker's income W received end-of-season is just sufficient to pay off the debt D (= his consumption) contracted at start-of-season, plus the interest at rate i. W
=
(l+i)D
(1)
Thus, the worker starts the next season with debt D again. At end-ofseason, the present value of the worker's lifetime labour power, PW for short, is (if we assume, for convenience, a deathless, though not debtless life) W/i whilst, assuming the debt cycle is perpetual, the present value of debts (also at end-of-season), PD for short, is (l+i)D/i. Hence, the net worth of the worker, given the perpetual debt condition (1), is PW-PD
=
W/i - (l+i)D/i =
0
(2)
The worker's present value of labour power is matched by the present value of his debts. Since the life-time credit-for-labour transaction implies PW as the life-time collateral, we may ask whether the labour power is self-owned or 'owned' by the creditor. The fact that the worker entered the contract voluntarily and is free to quit it, after paying the debts, to restart with another creditor is formally sufficient, from a free-entry-free-exit viewpoint, to suppose that the worker 'owns' his own labour power. But consider a slave economy with the same time structure of consumption-before-production and interest rate. Then, with a competitive market in slaves, (2) will hold for the slave as well with D standing for slave consumption (as for our worker) and W for 'his' income. However, PW now stands for the competitive market price of the slave rather than the life-time collateral of the perpetually indebted 'free' worker. Hence, from this viewpoint, the indebted worker is formally similar to a slave. As for freeentry-free-exit of the indebted, slaves too circulate among masters either on rent or by sale. The accounting is faithful to the long history, in Latin America, where debt peons were regarded as 'virtual' slaves. It may also have something to do with the fact that explicit collateralisation of labour power has been abolished by law in much of the capitalist world notwithstanding the abridgment of Property and Freedom that it represents.
Downloaded by ["Queen's University Libraries, Kingston"] at 19:07 31 May 2013
120
THE JOURNAL OF PEASANT STUDIES
Be that as it may, free entry and free exit do capture the classical definition of Freedom adequately. Some neo-classically-inclined historians have regarded European serfdom to have been a voluntarily entered 'contract', an exchange of protection in unsettled times for labour rents. Even if 'in certain areas at certain times, slavery and serfdom were formally initiated through agreements' (S&E, p. 119), subsequent generations of serfs were born into it so the proposition seems contrived. Similarly, even if indebtedness originates voluntarily due to 'exogenous' reasons, it also entails, where legal or social sanctions allow, the involuntary engagement of the services of kith and kin, even those unborn, in the service of the debt. That is, such inheritance of debt-based labour obligations is based no less on a coercive social rule (one's offspring are one's chattel) than is slavery. Voluntarism can be stretched here on the ground that the lender understood, at the time of the original debt contract, that the kin and offspring of the debtor were to be part of his collateral not to speak of the argument that the debt itself may have made possible such collateral! But it comes up against the individualism not simply of the classical definition but of neoclassical theory as well. Marxian theory, like neoclassical economics, analyses capitalist market exchanges starting from the assumption of Freedom, etc. However, Olsen points out that whereas neo-classical economics almost exclusively focuses on the implications of rational, choice-making (the exercise of negative Freedom) in markets, Marxian theory pursues the ways in which substantive freedom is or is not exercised thus changing the constraint sets that the rational choice makers face. We turn now to a consideration of case studies of debt-based labour transactions. It is noteworthy that such transactions are central in a significant majority of the cases studied. Olsen draws attention to the fact that in India, historically, forms of corporate bondage involving whole families of workers in one caste and owing beck-and-call labour services to upper caste Hindus prevailed. Similarly, she points out that even in the case of 'bondage' contracted on an individual basis, the obligations arising from the contract could be inherited from parents (pp.380-81). In particular, the debt obligations contracted voluntarily by a male could burden the female/s in his household or family. More generally we may note that the caste system, or at least segments of it, institutionalised an unequal division of labour and hierarchical dominance notwithstanding apparent 'reciprocity', enforced by local sanctions. To the extent that its members were tied into relationships involving obligations not voluntarily chosen, self-ownership was abridged. Kerr's study begins by noting that railway construction in nineteenth century India relied on both local village labour and migratory labour. Yet,
Downloaded by ["Queen's University Libraries, Kingston"] at 19:07 31 May 2013
THE DEBATE OVER UNFREE LABOUR
121
it was circulating (self-owning?) labour, lacking village ties, that came to increasingly dominate railway construction. Kerr argues that village-based labour, bound by ties of dependence on village notables, was essentially unfree. But the actual picture presented by the evidence is open to alternative interpretations. Much of the local labour was brought to the work sites under the tutelage of village 'powerholders' during the agricultural slack season. As such, this did not seem to conflict with the powerholders' own labour demands but, to the contrary, gave them a cut from the wages bill in the form of commissions. Similar sorts of paid intermediation were also common in the early modern Indian textile industry. Without necessarily denying unfree labour within the villages, might such intermediation not have been a service? Kerr's interpretation also slights his evidence that both circulatory labour and many among the village labourers had to be enticed to come to railroad work. While credit advances to the workers, whether as enticement or necessity, yielded interest, does this necessarily constitute unfree labour? Although Kerr states that debt was the key element to 'most forms of extraeconomic coercion applied to the construction workers' (p.417), he does not tell us wherein lay such coercion. On the contrary, he describes the informal contracting underlying the labour relationship thus built. While advances helped mobilise slack season labour, Kerr views the creditor as gaining 'considerable power over the disposition of the labour power of the recipient' (p.417) without confronting the fact that workers were not under duress to accept the credits. Baak's study of debt and unfree labour in the plantations of nineteenthcentury southwestern India is not unlike Angelo's study of the American south in that it details the transition from one form of unfreedom (slavery) to another. The abolition of slavery in 1855 was soon replaced by the Criminal Breach of Contract Act of 1865 which imposed criminal penalties for specific non-performance of debt-based labour commitments. But Baak also notes that the laws were not very effective in helping employercreditors to compel indebted workers to remain on the estates to clear the debts. What is viewed here as 'debt slavery' or the abridgment of Freedom may well be seen by others as the enforcement of Property arising from obligations contracted upon. As for contemporary India, two broadly different positions are discernible. Some scholars hold contemporary forms of attached labour to be 'free' and distinguishable from traditional forms of unfree tied labour arrangements or at least that contemporary labour-tying is based on the purely 'economic' compulsion of indebtedness.6 Others see contemporary forms of labour attachment as containing or based on significant elements of unfreedom.
Downloaded by ["Queen's University Libraries, Kingston"] at 19:07 31 May 2013
122
THE JOURNAL OF PEASANT STUDIES
Based on field studies in Haryana, India (see Brass [1990]), Brass believes not only that labour-tying by means of debt is a form of unfree labour but also that it is not an archaic pre-capitalist relation. Moreover, he argues that the shift from permanent attached labour (through debt) to seasonal and/or casual work is not the demise of unfreedom 'but only a change in its form' (pp.29, ff 18) on the consideration that the production relation in which workers find themselves after entering a contract does constrain their freedom to exit (for example, during the peak season). But this leaves unspecified what the constraining force in fact is. However, Brass found evidence that the heritability of debt was common. This establishes a presumption that some fraction - depending on the dynamics of debt and its inheritance - of the labour force are involuntarily caught in debt. While Brass found rules of heritability to be ubiquitous and concludes that both permanent and casual workers are unfree, heritability per se is not the theoretical basis of his conclusion as noted above. Debt-based unfreedom in the sense, at least, of collateralisation of kin and offspring has been enforced and policed by the community at large. This applies especially to women's labour, children's labour and the labour of lower social strata such as those based on caste or ethnicity which may all be considered fair game for unfreedom. Yet, Olsen found but one study (Karin Kapadia's work on Tamil Nadu) in her literature survey which dealt with the nature of women's involvement as direct or 'collateral' bonded labourers. Brass brings out the role of the 'community' clearly in a case from Purnea, Bihar, to conclude that a woman who receives support from her kin and caste in a confrontation with her employer over wages nonetheless is the target of hostility from the same groups when it comes to meeting debt obligations incurred by her menfolk to the same employer (P-40). With the important exception of collateralisation through inheritance, we believe that the phenomenon of indebted labour in rural India, both historical and contemporary, is not generally or easily classified by the classical yardstick. In many cases, a voluntaristic reading may be appropriate while in others self-ownership may indeed be abridged by direct coercion. In contemporary India, the term 'bonded labourer' has come to refer to a person whose debt to an employer obliges him, and often his family, to work for that employer on a first-call or low-wage basis (Olsen). This is in keeping with the law against having to mortgage one's future labour power that originated in the west. But the wide range of 'estimates' of this category illustrates the empirical difficulties we have already alluded to. In particular, the direct political-coercive powers that local notables may wield, in conflict or collusion with formal state power, may make for unfree
Downloaded by ["Queen's University Libraries, Kingston"] at 19:07 31 May 2013
THE DEBATE OVER UNFREE LABOUR
123
labour. Where, for example, the poor have inadequate means to access the powers of formal law, 'exogenous' enforcement by the state is voided and the poor must rely on informal local mechanisms in asserting their selfownership. Even if local notables arrogate coercive powers locally, they might be resisted through such mechanisms. In all such cases, the neoclassical separation of 'endogenous enforcement' through contract design from 'exogenous enforcement' by the state becomes blurred. This accounts for the frequent invocation of so-called extra-economic coercion in the rural Indian context and its relevance in determining self-ownership. What has just been stated may also help, together with local social sanctions and an ideology of the sanctity of family debts, in understanding why the relatively powerful may have to rely on debt in the first place to cement their authority over workers who are often neither proletarians nor sturdy independent peasants. Indebtedness is, indeed, voluntarily entered into and, in principle, continues only so long as the debt has not been repaid (over whatever time period). After all, the lock-in of debt may be broken by the very fluctuations (though not necessarily natural) in current income or expenditure that builds it in the first place. This makes for the 'vulnerability' of the local labour power supply, a supply that local property owners count on for its reliability, ease of recruitment, and other desired features arising from familiarity produced by the localised fragmentation of rural settlements which also makes for its monopolistic exploitability [Rao, 1988]. At the same time, local property owners' powers of 'exogenous enforcement' may themselves also be vulnerable to challenges from below or above, and to potential competition produced by caste or other types of local political factionalism among the powerful. This double vulnerability of power is met by the strategy of dual modes of control through which the line between self-ownership and unfree status becomes hopelessly blurred. It is one meaning that may be attached to the term 'semi-feudalism', a reflection of the somewhat fragile fusion of local political power and economic wealth in an economy in which the propertied-powerful are less than unified, the labour force is not wholly proletarianised-powerless, and the formal powers of the state unimplementable-contested. VIII. STATES The part played by state power in regulating and coercing labour is the focus of several case studies in the volume: Nazi Germany (Roth), Soviet-type systems (Kossler; van der Linden) and South Africa during and after the apartheid era (Grossman). In all these cases, state coercion of labour, the authors argue, has been strikingly more direct than just through enforcing property rights. This is not to say, of course, that states in the other cases
Downloaded by ["Queen's University Libraries, Kingston"] at 19:07 31 May 2013
124
THE JOURNAL OF PEASANT STUDIES
represented in the book confined themselves to the minimal functions of law and order. But states rarely come into focus in these other cases. Kossler, Roth and Grossman, on the other hand, seek to demonstrate that various forms of unfree labour were actively and intrusively created by the states' coordinating roles in rigging labour markets, in structuring labour processes or planning production, and in trying to control workers and their collective potential to challenge or change the systems. Kossler relates the despotism of the Soviet system not to the '"Oriental" features of the Russian past' (p.92) but to the Weberian hyper-rationality of market capitalism, on the one hand, and to the 'deep-rooted despotic feature of industrial capitalism' on the other. According to Weber, optimal work extraction from workers is achieved with the complete separation of all workers from the means of production (Kossler, p.91). The existence of a home base of independent production is apt to provide a shelter from the rigours and discipline of the market, and thus diminish the responsibility and compulsion that the market imposes upon a fully proletarianised work force. The achievement of this hyper-rationality, while allowing the Freedom of labour and the Freedom of consumption, also allows the full development of the despotic power of capitalist enterprise with which the 'free' worker must come to terms. While noting the convergence in factory organisation on Taylorist lines in both Soviet-type economies and the advanced west, he also draws several contrasts between the two. Factory despotism in the west has been abridged by two important developments i.e., the delinking of welfare from wagelabour, and the development of workers' rights under democracy. In the Soviet-style systems, however, the socialist project was conceived of as extending the rationality of the Taylorised factory to the economy and polity at large. This provided the basis not only for the development of Stalinist despotism but also for clipping the wings of plural institutions including workers' organisations. Kossler argues that the Soviet state, through party, police and other organs, came to exert great power over the entire workforce and enjoyed 'virtual monopsony [power] vis-a-vis labour power' (p. 102). One way of putting Kossler's thesis is that Soviet socialism, with its ambition to tame the (Marxian) irrationality and anarchy of market capitalism, ended up foregoing its (Weberian) hyper-rationality. The fusion of the political and the economic created a form of 'unfree' labour that 'transcends the limits of forced labour strictly speaking and slavery' (p.92). Van der Linden's essay focuses directly on forced labour proper under the Stalinist regime. He argues that forced labour in the Gulags - which he equates to unadulterated slavery - had not only a political importance (that is, terror as a substitute for legitimacy) but also considerable economic weight in that labour could be quickly and cheaply mobilised for the
Downloaded by ["Queen's University Libraries, Kingston"] at 19:07 31 May 2013
THE DEBATE OVER UNFREE LABOUR
125
massive early industrialisation: 'It was a type of labour mobilisation that fitted well into the stage of extensive industrialisation which lasted into the 1950s' (p.362). So when the transition to intensive industrialisation ensued, the economic weaknesses of forced labour became quickly apparent and caused its demise. This argument persuades easily. However, it is not in accord with his analysis, in the final essay of the volume, of the endogenous choice between slave labour and self-owning wage labour by accumulating elites. There, van der Linden rejects the classical argument that slavery proper was jettisoned because it became incompatible with the advance of technology. Roth presents the Nazi resort to forced labour in various forms during the Depression and the War as a clear instance of the state running the joint affairs of the bourgeoisie in deep crises. Although such unfreedom impinged on no more than 9.2 per cent of the employed (in 1934), campaigns to limit labour mobility generally led to a certain 'decommodification' of labour. With the extension of the Third Reich's geographical area during the War this decommodification greatly expanded in scope so that the proportion of unfree workers to German 'free' workers peaked at 43.67 per cent (in 1945, see Table 10). The main part of this 'unfree' class consisted of workers placed in a range from extreme contractlabour to open forced-labour relationships (p. 136, Table 2). He argues that this development was an organic part of the National Socialist response to the Depression which aimed to advance capital investment and rearmament. As the economy hit full employment in the wake of a programme of rearmament and deficit spending, the state used the decommodification of labour as a way to lower wages below what they would otherwise have been. Rather more implausible is Roth's claim that 'rapid advances in development and strategies for overcoming crises and bottlenecks are only possible on the basis of the widespread employment of unfree labour' (Roth, p. 129). That such strategies were employed is not in question for the politically implemented restrictions imposed on masses of workers under the Third Reich cannot be dismissed as a systemic 'aberration'. But that it should be seen, as Roth does, to be a necessity, indeed an ever present and expanding necessity, of capitalist modalities strains credibility. Grossman examines the trajectory of unfree labour and of the right to strike in South Africa during and after the apartheid regime. The systematic abrogation of the right of free entry of individual blacks into specified regions, occupations and industries was apartheid's defining characteristic. As such, it imposed major barriers to formal freedom in the labour market even while making full use of the labour so repressed. In addition, the system also denied blacks the collective right to organise and bargain with
Downloaded by ["Queen's University Libraries, Kingston"] at 19:07 31 May 2013
126
THE JOURNAL OF PEASANT STUDIES
capital to improve their wages and working conditions. Grossman regards bourgeois freedom as having little relation to the real experience of workers and particularly empty without the right to strike or otherwise collectively engage in improving the terms of work. As apartheid came down, momentous changes in the law were effected granting black workers the freedom of unrestricted mobility. Restrictions on individuals' choice of industry and occupation, and on the collective right to strike were both removed. Legal compulsion and penalties of contract labour have also been lifted. The right to strike, however, is now being sought to be suppressed again under the guise of an imperative of capital accumulation. Unlike business interests and their ideologues, Grossman does not assume that capitalism was opposed to apartheid. The defenders of a convergence of interests between capital and labour against apartheid emphasise the constriction in the supply of skilled labour that apartheid imposed. In other words, they tend to locate the pressures for change out of apartheid in the inexorable pressure of economic development and not in the role of mass defiance and struggle (p. 161). Indeed, we can agree on Grossman's fundamental point that apartheid is, in essence, 'a particular form of capitalist labour control' (p.161). In their respective analyses of market capitalism, both neoclassical economics and Marxian theory focus on the workings of markets under the assumption of the separability of state power and private power. For the most part, neoclassical economics takes the state as given and therefore leaves separability untheorised. On the other hand, while Marxians never lose sight of the question, most Marxian analyses either reduce the state to dominant ruling class interests or must rest content with a narrative view of state actions. The cases summarised in this section are, in a sense, 'easy' cases for Marxian class analysis because, in each of them, there was a close integration of political and economic power which abridged self-ownership one way or another and, more importantly, reduced the scope for workers' positive freedoms. But the question of separability remains very much open. The historical case studies throughout this volume show, at least implicitly, that the emergence and demise of various forms of unfree labour as also the emergence of self-ownership cannot be explained without addressing this question. It seems more than likely that an adequate formulation of an answer to this question will have to transcend the 'mere' negative dimension of freedom by which self-ownership is defined to comprehend positive freedom as well.
THE DEBATE OVER UNFREE LABOUR
127
NOTES
Downloaded by ["Queen's University Libraries, Kingston"] at 19:07 31 May 2013
1.
2.
3.
4.
5.
6.
'If, in general, freedom is the absence of restrictions upon options open to agents, one can say that the liberal tradition has tended to offer a very narrow construal of what these restrictions can be (often confining them to deliberate interferences), of what the relevant options are (often confining them to whatever agents in fact conceive or choose), and of agents (seen as separate individuals, pursuing their independently conceived ends, above all in the market-place). Marxism invokes wider notions of the relevant restrictions and options, and of human agency' [Lukes, 1983: 146]. Johnson's study of slavery in the Buenos Aires region in the early eighteenth century similarly notes that slavery was integrated into the fabric of urban production alongside selfowning labour. Less than twenty percent of slaves, however, were to be found in artisanal manufacturing and agriculture, the rest being used as household servants. The preponderance of small-scale slave-ownership is attributed to the comparatively high cost of imported and coerced African slaves which could only be supported in the Andean mining zone and the Caribbean. This point is related to capitalism being the world's dominant mode of production. Dominance may be defined ex-post and literally as the predominance of self-owning wage labour in the total labour force. Else, it may be, ex ante, said to obtain if the rate of growth of self-owning wage labour exceeds that of the total labour force. The latter is to be regarded as an hypothesis to be tested against historical experience. Recent attempts at giving so-called clear and individual titles to peasants on communal lands in some parts of the third world have been presented as the only way for these peasants to secure formal credit. But they may also have effects not unlike the enclosures of commons in Europe. Does the presence of a dominant mode of production necessarily preclude compromises among modes via the state? Brass is right to point out that we must know not merely that law forbids unfreedom, say, but also why it does so. But the state is not necessarily the instrument of a single class even if it be the dominant class. See Rao [1999] for the Indian debate over unfree labour. REFERENCES
Brass, Tom, 1990, 'Class Struggle and Deproletarianisation of Agricultural Labour in Haryana (India)', The Journal of Peasant Studies, Vol.18, No.l, pp.35-67. Domar, Evsey D., 1970, 'The Causes of Slavery or Serfdom: A Hypothesis', Journal of Economic History, Vol.XXX, pp.18-32, March. Reprinted in Evsey D. Domar, Capitalism, Socialism and Serfdom. Essays by Evsey D. Domar, Cambridge: Cambridge University Press, pp.225-38. Lukes, Steven, 1983, 'Emancipation', entry in Tom Bottomore, Laurence Harris, V.G. Kiernan and Ralph Miliband (eds.), A Dictionary of Marxist Thought, Cambridge, MA: Harvard University Press. Marx, Karl, 1967a [1867], Capital: A Critique of Political Economy, Vol.1, New York: International Publishers. Marx, Karl, 1967b [1894], Capital: A Critique of Political Economy, Vol.3, New York: International Publishers. Rao, J. Mohan, 1988, 'Fragmented Rural Labour Markets', The Journal of Peasant Studies, Vol.15, No.2, pp.238-57. Rao, J. Mohan, 1994, 'Agrarian Structure, Labour Supplies and the Terms of Trade', in A. Dutt (ed.), New Directions in Analytical Political Economy, Aldershot: Edward Elgar. Rao, J. Mohan, 1999, 'Agrarian Power and Unfree Labour', The Journal of Peasant Studies, Vol.26, Nos.2/3, pp.242-62.
