Weapons of the week, weakness of the weapons: Shifts and stasis in development theory

Abstract

Over the past half century the theory, practice and politics informing development studies have followed contrasting trajectories, a tangled epistemological pattern displayed inadvertently by some of the contributions to three of the four books reviewed here. This inconsistency has resulted in confusion, not least where current Marxist approaches to the agrarian question are concerned. Unsurprisingly, therefore, misinterpretations of unfree labour plus the jettisoning of class analysis have led to the abandonment of socialism, and its replacement with nationalism and bourgeois democracy as desirable political objectives. By locating rural class formation and agrarian struggle in a global capitalist context, however, one of the four books demonstrates the continuing importance of socialist politics to the study of development.

Notes

1 The impact on development studies of prevailing intellectual fashion is an issue that, for obvious reasons, still awaits a chronicler. When considered, the reason for a change of opinion is subject to a classic form of displacement: from the context of the viewer him/herself to that of the viewed. Its cause is invariably – and wrongly – attributed simply to exogenous factors, such as a new dynamic or a new set of circumstances operating in or affecting the area studied.

2 For an amusing depiction of a fictional university department in the US colonized by postmodern theorists, see the novel by Hynes [2001]. A description of a star academic theorist conveys humorously (but accurately) the stasis that is an inevitable outcome of postmodern aporia [Hynes, 2001: 204]: ‘In a discipline where scholarly heft was defined by being more postcolonial than thou [the theorist] was the heftiest of the lot. As a graduate student at an Ivy League school he had announced to his dissertation committee that doctoral theses at major Western universities were a primary locus of the objectifying colonialist gaze on native subjects, and he refused on principle to participate in the marginalization of indigenous voices or to become complicit with the hegemonic discourse of Western postcolonial cultural imperialism. In practice, this meant that for six years he refused to take classes, attend seminars, or write a dissertation. As a result of this ideologically engaged non-participation, he was offered tenured positions even before he had his Ph.D., but by refusing to write a book or any articles on his topic – publishing with major university presses being even more complicit with imperialism than writing dissertations – he provoked a fierce bidding war. Columbia won by offering him an endowed chair and a full professorship, and on Morningside Heights he courageously continued his principled refusal to teach any classes, hold any office hours, publish any books, serve on any committees, or supervise any dissertations. For this demanding and theoretically sophisticated subaltern intervention in the dominant discourse, [the theorist] made well into the six figures, more money than the president of the United States.’

3 It is perhaps significant that an espousal of systemically unspecific democracy prefigured the abandonment of socialism by those associated with ‘The God that Failed’[Koestler et al., 1950]. One of the contributors [Koestler, 1944: 176] had a short while earlier accepted that ‘conformism is often a form of betrayal which can be carried out with a perfectly clear conscience; and the temptation to exchange the miseries which intellectual honesty entails for the heart-warming satisfactions of [bourgeois] efficiency is great. The collapse of the revolutionary movement has put the intelligentsia into a defensive position; the alternative for the next few years is no more “capitalism or revolution” but to save some of the values of democracy … or lose them all; and to prevent this happening one has to cling more than ever to the ragged banner of “independent thinking”’ (original emphasis). This kind of justification (‘to save some of the values of democracy’) is repeated almost exactly by many ‘new’ postmodern populists in their discourse of disillusion.

4 Three of these Marxist dissidents were – and are – associated with this journal: Petras and Veltmeyer, on which see more below, and Brass [1991]. Excluded from the criticisms made here is Beverley [2004], who, unlike other converts to postmodernism, has not only acknowledged a change of mind but also attempted to confront the theoretical issues raised by this. Disagreement with his current espousal of postmodern theory [Brass, 2006] should not detract from this fact.

5 That followers of this third path were among the participants in a conference the objective of which, according to the editors of the Agrarian Studies volume (Ramachandran and Swaminathan, p.xi – emphasis added), ‘was to provide a forum for discussion and debate on new theoretical and empirical research from a left perspective on agrarian relations in less-developed countries’, is ironic indeed.

6 It is impossible to surmise from the confident assertion about how old and well-known are ‘elegiac accounts of the loss of rural ways of life, or of the despoliation and neglect of the countryside’, that one of the authors [Corbridge and Jones, 2005: 1] is the same person who a decade earlier missed precisely the longevity and pervasiveness of this very ideology when endorsing its latest manifestation, in the form of the ‘new’ postmodern populism. Then, the same author [Corbridge, 1994: 91] wrote just as confidently: ‘I acknowledge the power of the populist and post-modernist turns [and] the power of the post-modernist critique in regard to questions of difference and representation.’ Among those writing about peasants whose approach to the study of development underwent a similar kind of shift – from 1970s critic to a more benign view of populism in the early 1990s, and once again to critic in the late 1990s – was Lehmann. Whereas 1970s texts by Lehmann [1974b; 1978] contain many endorsing references to the existence and importance of class and class conflict, and are generally dismissive of populism (=‘airborne narodnism’), by the early 1990s he had made his peace with populism. Thus Lehmann [1990] is a paean to the virtues of non-specific grassroots agency (basismo) guided by NGOs, the epitome of a populist approach to development, whilst elsewhere he observes [Lehmann, 1993: 694] that a book under review was ‘a touch too populistic’, but dismisses this as ‘a minor reservation’. By the late 1990s, however, Lehmann [1997] had declared postmodernism wanting. Yet another example is Michael J. Watts, a geographer who writes about African development. At the start of the 1980s, in what is a staple of populist discourse, he [Watts, 1983] was extolling the virtues of what he took to be a pre-capitalist peasant subsistence ethic operating in northern Nigeria, categorized by him as an instance of ‘moral economy’. Nearly two decades later, by contrast, Watts [2000] – like Corbridge and Lehmann – is critical of the postmodern attack on development by ‘the millenarian populists, the romantics’.

7 Why two contributions to the volume edited by Moyo and Yeros (by Ampuero and Brittain, and by Veltmeyer) are regarded as the sole exceptions to this and the following criticisms is outlined below.

8 To say, as does the editor of A Radical History of Development Studies, that the volume ‘is a radical chronicle because it includes plural conceptions of development history and adopts a critical perspective towards, and engagement with, orthodoxies of development theory and practice’ (Kothari, p.1), is to say nothing, since stated thus the term remains politically meaningless. In short, it fails to differentiate between a ‘critical perspective’ of development that emanates from the political left and one that originates from the discourse of the political right. Both are covered by the spuriously progressive (and hence slippery) concept ‘radical’, and the volume simply repeats the same mistake as an earlier volume on the same subject [Horowitz, de Castro, and Gerassi, 1969], the difference being that now there is no longer any excuse for making this kind of error. That Kothari is wholly unaware of the different politics structuring this rubric is evident from her categorization as radical not just Marxist theory but also ‘post-colonial and feminist perspectives and analyses’, and the classification of them all as ‘thinking from the “left”’. Unlike Marxism, both post-colonial and feminist essentialisms have strong roots in conservative discourse (on which see Brass [2000; 2003; 2006]). An inability on the part of its editor to spot this difference augurs badly for the rest of A Radical History of Development Studies, a misgiving that is indeed borne out.

9 In the case of Moyo, this difficulty is prefigured in an earlier paper on the land question in Africa, where he incorrectly cites the editors of a special issue of this journal. In the bibliography of that paper [Moyo, 2003: 30], he mistakenly attributes the production of the special issue in question to one editor, Bernstein, whereas it actually had two [Bernstein and Brass, 1996/97].

10 This criticism applies also to other accounts of the development debate published recently. Among the more questionable assertions rebutted in this journal of late are claims to have participated in the critique of the Asian subaltern studies project, to have argued all along that postmodernism was politically reactionary, and to have insisted from the outset that capitalism was compatible with unfree labour.

11 Although unintended, the effect is to replicate post-1924 historiography in the USSR about the Russian Revolution, based on the absence of a narrative distinction between those who had been peripheral figures and those who had actually discharged heroic roles.

12 One remedy lies with publishers generally, the editors of which should seek to tighten up their refereeing procedures so as henceforth to ensure that contributors to edited volumes possess at least a basic level of competence where the relevant literature on development studies is concerned.

13 Although their remit is populist discourse with particular reference to Africa, Woodhouse and Chimhowu somehow manage to overlook not just earlier texts on this subject [Kofi, 1978], but also the classic article on African populism by Saul [1969]. Unsurprisingly, much of what is found in the latter analysis is encountered also in (but not transcended by) their own contribution. Similarly absent is another seminal text, published over a decade ago in this journal: that by Jackson [1993] dealing with environmentalism, gender and populism. That such lacunae occur in a volume the stated intention of which is ‘to stimulate new thinking on where the discipline may be moving’ is scarcely credible. It also invites a riposte that – on this evidence – development studies shows no sign of movement, never mind travelling with a particular destination in mind.

14 The claim to be ‘the first ever student of development’ accurately captures the flavour of this contribution. Its egocentric focus and narrow timeframe result in a corresponding failure to recognize that the study of socio-economic development, its causes and effects, has a very long lineage, stretching far back into history. To confine the study of development to the formally constituted academic rubric of development studies is rather like saying that the study of art and literature had no existence before the foundation of art colleges and university literature departments.

15 Much development theory of the kind Harriss currently favours assumes that the causes of underdevelopment and rural poverty are basically ones of knowledge. If only those in the state apparatus of capitalist nations really understood the true nature of the problem, this discourse proclaims, they would very quickly change policies for the better. Such a view is naïf, in that it overlooks or downplays the main reason why the capitalist state fails either to promulgate or – where promulgated – to implement plans/policies that actually are advantageous to poor peasants and agricultural labourers: namely, the political power in national or international contexts of those classes that own/control the means of production/distribution/exchange. For Harriss [2002b], therefore, the capitalist state remains a potentially disinterested – and thus benign – agent in the development process, capable/desirous of and indeed willing to engineer fundamental transformation. What adherents of this approach fail to realize is that without the expropriation of landowners and agribusiness enterprises on the one hand, and the capture of state power by workers and poor peasants on the other, fundamental transformation is either off the political agenda or stays only on paper. Much the same point has been made in an interesting article by Das [2005] on the state and social capital in India.

16 This very point was made in a critical review article published a decade ago [Brass, 1995: 517–18], and strangely is the only argument in that critique which Harriss chose not to heed. However, this same misplaced faith in the efficacy of piecemeal solutions has found a new champion. In his contribution to the volume edited by Kothari, it is Bernstein (p.111, original emphasis) who now advocates the importance to development studies of ‘applied knowledge of practical benefit in the formulation and implementation of development policies and interventions’. In this regard, still relevant are the comments made by Common [1992: 3, 5] of the pressures to conform brought to bear on certain groups/individuals belonging to the political left in the 1930s: ‘The socialist has the special problem of holding onto his vision of a world which is not yet, while maintaining himself in an environment which makes vision a handicap and tempts him to abandon it. What is behind the socialist parties and what assures their final victory is this vision of a new world; what ruins them all is that they must prove practical utility in the present capitalist day if they are to live at all. … Hence we have in this country a nominal Left which is incapable of supplying ideas of a revolutionizing character, Marxist and revolutionary though it is in its literature and in a literary way’.

17 When considering the same issue elsewhere in the collection edited by Kothari – namely, why was it that the study of development ‘has not only survived the current period of neo-liberal ascendancy … but has prospered in British universities’ – Bernstein (p.115) also overlooks this simple explanation. Development studies thrived in part because of a willingness on the part of its practitioners to apply market ‘solutions’ to problems of underdevelopment. The extent to which this connection eludes Bernstein is evident from a subsequent observation that, although the ‘wider intellectual, and political, understanding of development as a process of struggle and conflict’ has been ‘lost’, ‘such oppositional thinking thrives outside the institutional spheres and practices of development studies’ (pp.119, 134 note 17). The unintended irony structuring this observation – that political dissent vanished from the academy only to reappear on the streets – underlines as precisely as possible the point being made here (the rightwards turn taken by development studies), and thus requires no further comment.

18 See, for example, Harriss [1982a; 1982b]. At this period, therefore, one encounters the application by Harriss [1982: 204ff., 282ff.] to rural Tamil Nadu of a Marxist analytical framework. Elsewhere his view [Harriss, 1982b: 23] was that Marxism ‘seems to be most appropriate for studies relating to rural development (both as policy and process) because it is inherently inter-disciplinary, and because of all the approaches that we have reviewed it is the one which is most centrally concerned with issues of distribution and with poverty.’ In his contribution to the volume edited by Kothari under review here, Harriss (p.24) accepts that initially he subscribed to ‘a self-conscious project of restoring the unity of social science, around a broadly Marxist perspective’.

19 Details about this particular shift are set out in a critical review of the ‘impasse’ position [Brass, 1995]. Despite having contributed to a volume [Booth, 1994] espousing the notion of a development ‘impasse’, and arguing strongly [Harriss, 1994: 173] that the debate about agrarian change in India was indeed at an ‘impasse’, Harriss nevertheless felt able to write about the ‘impasse’ debate shortly thereafter [Harriss, 1998: 294–5] almost as though it had nothing to do with him.

20 This emerges most clearly in two contributions by Harriss, both published in the early 1990s. ‘The study of agrarian change’, Harriss [1994: 180–81] proclaimed at that conjuncture, ‘thus points to the limitations of conventional Marxian political economy because of the way in which, by its reduction of politics to economics, it consistently fails to account for politics.’ His own position, he accepted [Harriss, 1994: 192], now had ‘explicit continuities with the work of some members of the Subaltern School of Indian historians and with that of James Scott.’ Allegiance to the ‘new’ populist postmodernism is stated finally and unequivocally [Harriss, 1994: 193] when agreeing that his interpretation ‘owes something to the influence of post-modernist “deconstruction” of concepts such as that of class.’ In another edited volume [Harriss, Hunter, and Lewis, 1995], he declares a new-found enthusiasm for the ‘new institutional economics’ (NIE), a choice-theoretic, scarcity-assuming development paradigm associated with the work of the neo-classical economists such as Douglass North (see North [1990; 1997] and Wallis and North [1986]).

21 Those Marxists of the ‘I told you so’ variety who – against the grain – wrote critiques during the early 1990s of the ‘new’ populist postmodernism were Petras [1990] and Brass [1991; 1995]. For the epistemological link between on the one hand postmodernism, and on the other the subaltern studies project, the new social movements framework and the ‘everyday forms of resistance’ theory associated with James Scott, see Brass [2000].

22 For this particular identity, see Harriss [2002a] and – especially – Corbridge and Harriss [2000]. A critical analysis of all the claims made in the latter text, and especially those about having been a long-standing critic of the ‘new’ populist postmodernism, can be found in a review published in this journal (JPS, Vol.29, No.1, October 2001, pp.182–91). The extent of the shift can be gauged from a simple comparison of two statements. One, made in the mid-1990s and already cited (see above), was the acceptance by Harriss [1994: 192, 193] that his view exhibited ‘explicit continuities with the work of some members of the Subaltern School of Indian historians and with that of James Scott [and] owes something to the influence of post-modernist “deconstruction”’. The other, made eight years later, was the belated recognition by Harriss [2002a: 9] that ‘a post-modern refusal to accept “external” guidelines for political action or negotiation leaves the armoury of anti-secularism bare when it comes to the adjudication of disputes between religious and ethnic groups.’

23 There is abundant evidence of such confusion. In Harriss and Harriss [1989], for example, references to the existence of rich, middle and poor peasants (that is, Marxist concepts) jostle alongside endorsements of (neo-populist) Chayanovian theory.

24 Pace Moyo and Yeros, it is not the case that Bernstein argued – let alone initiated the argument – that agribusiness reproduced peasant economy for its own purposes. What he did claim was much rather the reverse: that petty commodity production was capable of reproducing itself, even where capitalism was already present. This, as all those who write about the peasantry ought by now to recognize, combines the theory about peasant family farming long associated with the work of Chayanov [1966] with the everyday-forms-of-peasant-resistance framework associated more recently with James Scott [1985]. Both are exemplars of agrarian populism, as was Bernstein until his volte face.

25 An input from the same source to their other contribution is also acknowledged by Moyo and Yeros (p.202, note 1).

26 Marxism, Bernstein (p.127) informs us in A Radical History of Development Studies, ‘has been largely displaced by the various currents of post-structuralism, postmodernism and the like … [i]n relation to development studies, the effect of the postmodern(ist) “turn” is to deny the validity of any conception of development other than as for an imperializing (Northern) discourse imposed on the South.’ These ‘various currents’ are described further as ‘combinations of nationalist, populist and deconstructionist elements’ (Bernstein, p.128), thereby reproducing precisely the case made against the ‘new’ populist postmodernism by this reviewer throughout the 1990s (see next endnote) and published in Brass [2000]. Instead of citing the latter, however, the single source for this view is identified by Bernstein (pp.127, 135 note 31) as being Cooper and Packard [1997], a very belated contribution to the debate about the impact on development studies of postmodern theory, and more accurately perceived as a synthesis of preceding critiques.

27 Articles criticizing the ‘new’ postmodern populism appeared in JPS 18/2 (1991), 21/3&4 (1994), 22/3 (1995), 24/1&2 (1996–97), and 24/4 (1997). Despite being significant critiques of the cultural turn – as suggested by, among many others, Chaturvedi [2000: vii–xix] – not one of them is cited by Bernstein in either of his contributions to the volumes edited by Kothari and by Moyo and Yeros.

28 The only alternative explanation – that the omission of any reference to these texts is deliberate – is unthinkable, and must be discounted automatically and resolutely, since not to do so would immediately raise awkward questions about the scholarly credentials of these volumes.

29 See the critique of Bernstein [1977; 1981; 1982] by Gibbon and Neocosmos [1985], the accuracy of which has been confirmed subsequently by Leys [1996: 12, 64] who identifies Bernstein as a fellow dependency theorist. Adherents of this view, as Leys [1996: 176] himself accepts, ‘tended to focus on the relative limitations of Africa's capitalist classes and tended to ignore their growth paths over time … They tended (like the Russian populists a hundred years earlier) to see capitalism as something occurring only in the visible, formal sector of the economy and to overlook the ongoing gradual but crucial transformation of rural relations of production into increasingly commercial and finally capitalist relations’. The Chayanovian theory informing Bernstein's arguments at that conjuncture has also been noted by Johnson [2004]: ‘What is definitive about the peasant form of production is that, regardless of ownership, the logic of production is subsistence. Building upon Alexander Chayanov's theory of a peasant mode of production, Henry Bernstein … argues [in 1979] that peasant production is distinguished from capitalism because there is no appropriation and realization of surplus value or accumulation of capital.’

30 See Gibbon and Neocosmos [1985: 162–3, 187, original emphasis]. Although at the beginning of the 1990s Bernstein was keen to distance himself from the State/peasant dichotomy that marked his earlier position, he nevertheless continued to adhere to an essentialist concept of peasant economy located inside the wider capitalist system. Indeed, he himself comes close to admitting that he has retained this central element of the populist claim [Bernstein, 1990: 72]: ‘The starting-point must be to view peasants today as agrarian petty commodity producers within capitalism. This is also the position of more sophisticated populists’ (original emphasis). For more of the same, see Bernstein [1992: 30].

31 See Bernstein [2001: 9 note 19, 17, 18]. The same is much in evidence in his contribution to the volume edited by Kothari, where one encounters many criticisms of populist approaches to development. For example, that ‘NGO activity … the jargon of “participation”, “empowerment”, “stakeholders” and the like is most pervasive … [a]long with tendencies to celebrate the “local” and “indigenous”’, etc., etc. (Bernstein, pp.120, 134 note 19). For yet more of the same see Bernstein [2006b: 401–2].

32 Nor, it should be added, are they the only ones. In his contribution to the volume edited by Kothari, Harriss (p.34) refers to the exchange between Bernstein, Gibbon and Neocosmos, but makes no mention of the fact that both the latter are extremely critical of the former. An earlier reference by him [Harriss, 1997: 7–8] to the same exchange does mention the element of criticism. Elsewhere another text cites both Bernstein and Gibbon and Neocosmos in a fashion not so different from that of Moya and Yeros, noting [Oya, 2004: 292 note 11] only that ‘emphasis is given on the prevalence of “petty commodity production” (PCP), which combines class locations of capital and labour and within which concrete processes of differentiation occur in Africa’. Since no mention is made of the fact that Gibbon and Neocosmos are highly critical of Bernstein, this kind of conflation is rather like saying that Charles I and Cromwell both had views about kingship and republicanism, which undoubtedly they did – but ones that were inimical.

33 See Bernstein [1994: 64; 2000: 41, 44], where he states unequivocally that of central importance to an understanding of class formation and class struggle in a capitalist context is the fight ‘against all forms of “tied” labour arrangements based in personal dependence, debt bondage, patronage, etc. – in short, “deproletarianization”… which denotes the denial or loss of the one positive “freedom” of the proletarian condition … namely workers' mobility within labour markets.’ In fact, his name appears as co-author of a text [Brass and Bernstein, 1992] where precisely this same case is made strongly. The obvious question to ask here is: if his view that ‘analytical class “purism”’, as he calls it, has no rôle either in the analysis of or in the struggle against capitalism, why then did he support this very position?

34 That he has abandoned class analysis is evident, for example, from his admission [Bernstein, 2006b: 406, note 14] that now ‘I use the term “classes of labour” in preference to inherited notions of “proletariat” (and proletarianisation) [which] are encumbered with too many problematic historical and ideological associations.’

35 Difficulties arising from a misunderstanding of the agrarian question is a problem Bernstein has in common with Byres (see below). Hence the observation by Bernstein (p.126) in the volume edited by Kothari – that ‘[i]t may well be that the questions … posed are not the right ones’ – is not just the only statement encountered in his contribution with which it is not possible to disagree but also the one that accurately sums up his own analytical approach. Moreover, it confirms the continuing relevance of the point made over a century ago by Karl Kautsky [1988: 2, emphasis added]: ‘My own opinion is that adding yet another specimen to the vast pile of monographs and inquiries is not what is required at present. Praiseworthy as these may be, there is hardly a current shortage of conclusions as to the state of agriculture. Year in, year out, governments, the sciences and ruling class journalism continue to batter the public with a crushing volume of such studies. No. What is needed is an exposé of the central thread running through this plethora of facts – an investigation into the fundamental tendencies which, although operating below, nevertheless determine how the observable facts appear to us. Instead of examining the various individual aspects of the agrarian question in isolation – the conventional current practice – we have to look at the relationship between these aspects: between the large and the small farm, indebtedness, the law of inheritance, the shortage of labour, foreign competition and so on – considered as individual manifestations of one overall process. The task is a difficult one, the subject immense. To my knowledge, the field has not yet been given a satisfactory treatment from a modern socialist viewpoint.’

36 Written at the end of the nineteenth century, and still relevant today, the two main analyses by Marxists of the agrarian question are those of Lenin [1964] and Kautsky [1988]. Unsuccessful attempts to improve on them include not just Bernstein [1996/97; 2006a] but also Byres (see below).

37 See especially Bernstein [2006a: 4, 6]. If all that Bernstein is saying is that those who purchase labour-power have different political interests from those who sell it, then this is self-evident and amounts to no more than a tautology. However, this political difference does not translate theoretically into a bifurcated agrarian question in the manner he proposes, not least because the political interests of those selling their labour-power are themselves not uniform. This is because the latter category is invariably composed of those whose class position is not the same, a point that Bernstein is unable to confront (let alone explain) now that he has jettisoned ‘analytical class “purism”’.

38 Where this leads is unsurprising. Hence the banal conclusion to a more recent presentation [Bernstein, 2006a: 13], where we are told that we have ‘to recognize and … to be able to analyze the contradictory sources and impulses – and typically multi-class character – of such struggles, in ways that can inform a realistic and politically responsible [sic] assessment of them.’ This is World Bank speak: ‘politically responsible’ – politically responsible for whom? Even non-Marxists start, not end, with this kind of question, one that Marxists do not themselves pose. Those rural struggles that are – and have been – ‘multi-class’ (=‘multitude’) do not resolve anything politically, nor can they. For an incisive critique of the pseudo-sociological notion of ‘multitude’ informing the currently fashionable postmodern nostrums of Hardt and Negri [2000; 2005], see Petras [2002].

39 Most notably the assertions made by Byres about the social relations of production (see below). Just how far Bernstein manages to distance himself from the analysis of Byres is evident from statements [Bernstein, 1996/97: 22, 51, 52] to the effect that his own analysis ‘builds on the project of T. J. Byres’ comparative political economy of agrarian transitions', that he himself has ‘been privileged by the fruits of [Byres'] intellectual labour, and the agenda it has enriched’, and that ‘[Byres] continues to be a beacon of light in our collective project’.

40 See Bernstein [1996/97: 23ff.], where the references are to Engels [1990], Lenin [1964], Kautsky [1988] and Preobrazhensky [1966].

41 On the centrality of ‘the methodology of class analysis’ to the agrarian question of labour, see Bernstein [1996/97: 25], where the specifically political component – that element which is retained as the agrarian question of ‘labour – is described as involving ‘struggles for democracy and socialism’ (emphasis added). The significance of this last objective will become clear below.

42 Hence the observation [Bernstein, 1996/97: 43, 50] that ‘a century of modern imperialism has extended the determinants of industrialization far beyond the prospects of agrarian transition in landscapes inhabited exclusively by classes of landed property and agrarian labour’, a situation in which ‘the end of the agrarian question without its resolution[amounts to] the elimination of any prospects of agrarian transition as a route to comprehensive industrialization in contemporary poor countries’.

43 On this point, see Bernstein [1996/97: 52].

44 Kautsky was concerned with the political opportunism of those such as Vollmar, who advocated building an electoral base among the peasantry, insisting that the German Social Democracy should seek a following among the crisis-ridden small and middle peasantry [Salvadori, 1979: 48ff.; Husain and Tribe, 1981; 1984]. Against this, Kautsky argued that as ultimately the small peasant farm was an historically doomed institution, no attempt should be made to revitalize it.

45 For this argument, see Preobrazhensky [1980] and also Bukharin and Preobrazhensky [1922: 316ff.], where the centrality of this same case emerges clearly (‘The rich peasant greeted the [Bolshevik] revolution inspired by the most rosy hopes and anticipations, but as a result of the revolution he found himself stripped of part of the land which he had owned before it occurred. As long as this class of rich peasants continues to exist, its members will inevitably prove to be irreconcilable enemies of the proletarian State and its agrarian policy … The petty-proprietor mentality of middle peasants … inclines them to form an alliance with the rich peasants’, etc.).

46 The text in question is Gerschenkron [1943], on the importance of which to his own analysis of the agrarian question see Bernstein [1996/97: 47ff., 52]. Although Gerschenkron is invoked by Bernstein mainly with regard to agrarian transition in Prussia, it is nevertheless clear that the influence of the former extends to much of re-interpretation presented by the latter.

47 For the centrality of economic backwardness to his ideas not just about Germany but also about Russia, see Gerschenkron [1955; 1962]. The latter is classified by Renton [2001: 146 note 3] as one of the exponents of the Sonderweg thesis, and according to Fishlow [2003] ‘Gerschenkron's analysis is conspicuously anti-Marxian’.

48 See Bernstein [1996/97: 49]. The inference that Junkers were in any way anti-capitalist is controversial, and some analyses [Tirrell, 1951: 12, 16] maintain that large landowners in Prussia took an active interest in agricultural production, and were already capitalist by 1800. Even if the latter conjuncture seems too early, it is clear from the United States Senate Commission [1913: 267ff.] appointed by President Wilson that on the eve of World War I the purpose of mortgage credit associations (landschaften) was to provide Prussian landowners with capital for agricultural improvements. Large proprietors were accordingly able to borrow for agricultural investment ‘at lower rates and under more favourable conditions than would otherwise be possible’, and in 1912 some two-thirds of estates larger than 100 hectares had availed themselves of such loans [United States Senate Commission, 1913: 381ff.]. In purely economic terms, therefore, the existence of financial institutions specifically in order to provide Junkers with capital for agriculture, plus a 60% uptake of these credit facilities, is difficult to reconcile with the categorization of the Prussian landlord class as ‘anti-capitalist’.

49 That socialism and not capitalism was the object of struggle conducted by Junkers is clear from many things, not least the fact that among the committees operated by the Chamber of Agriculture was one concerned with day labour on farms, the task of which was to give employers ‘advice in case of breach of contract’[United States Senate Commission, 1913: 371]. In other words, an institutional form of employer collusion in rural areas, so as to police and enforce unfree labour relations in agriculture, thereby attempting to pre-empt the emergence of conditions favouring the development of a working class consciousness.

50 For a critique of the Sonderweg thesis see Renton [2001: 130, original emphasis], who argues that: ‘Where it is accepted, Sonderweg theory changes our understanding of fascism. In blaming the Junker class for the success of Hitler in 1933, the suggestion appears to be that fascism was a pre-capitalist reaction against capitalism. The inference of the latter position is inescapable: such an argument breaks all links between capitalism as a system of production and fascism.’

51 See below. The target of the critique by Renton [2001] is Byres, whose analysis of Prussia is located within the Sonderweg tradition. Among the more surprising claims made by Byres that Bernstein fails to interrogate is that capitalism and unfree labour are incompatible, as the employment of latter prevents the installation of developed productive forces. Accordingly, Byres' argument that in the antebellum US slavery ‘effectively blocked the development of capitalism in the South’ because slave resistance placed ‘strong limits on the instruments of production that can be deployed’ (i.e. unfree labour was less efficient than free workers) is accepted by Bernstein [1996/97: 30–1] unchallenged. Although Bernstein notes that ‘Southern landowners did introduce technical change’, he nevertheless reiterates Byres' view that unfree labour was an obstacle to technical change. This is surprising, for two reasons. First, in another text [Brass and Bernstein, 1992 precisely this same claim – namely, the incompatibility between unfree labour and capitalist development – is strongly criticized. And second, the inference that unfree workers resist the imposition of discipline whereas free labour employed by agrarian capitalists does not, is clearly nonsense, as the many reported cases of worker sabotage in the capitalist labour process confirm. This old canard is found not just in Byres but also Kautsky 1988: 41], who notes: ‘In feudal agriculture, the large farm was not markedly superior to the small farm … The bulk of an estate's labour power, both human and animal, was supplied by peasants who had to use their own tools and implements, ploughs, carts and animals to perform their manual labour and plough on the lord's land. The difference between the large and small farm did not consist in the fact that the former had better equipment and a more extensive division of labour than the latter. The difference was that the same equipment which peasants used assiduously and enthusiastically when working for themselves, was used on the lord's land with that degree of indolence and carelessness only possible when undertaking forced labour for another.’ The problem with the claim that unfreedom is necessarily inefficient because slaves, serfs or bonded labourers do not fully apply themselves when working on behalf of an employer is clear. Not only could the same argument be applied to a capitalist factory, in that free workers similarly undertake acts of sabotage in order to slow down speed-ups or production schedules, but it fails to take into account the important role of coercion. The latter took the form of overseers who whipped slaves who did not maintain the desired level of work intensity, and the eviction of tenants who did not work efficiently on the feudal landlord demesne.

52 Nominally the political dimension of the agrarian question as interpreted by Bernstein [1996/97: 25, emphasis added] is about the politics of a transition ‘to democracy and socialism’. That Bernstein no longer perceives socialism as part of the way in which the agrarian question is to be ‘resolved’ politically is clear from observations [Bernstein, 2006a: 6] such as ‘[t]he most “virtuous” realization of the logic of the “classic” agrarian question, in transitions to both capitalism and (once) socialism.’ Now that he has abandoned socialism as a political objective, the politics of transition can only be about some form of bourgeois democracy.

53 Marxist theory has long recognized the crucial nature of the link between the agrarian question and a transition to socialism, so much so that without the former the latter cannot be achieved. Some seven decades ago, for example, the following comment was made in a Marxist Study Course [c. 1930: 38–9]: ‘But the desertion of Marxism in questions of agrarian theory has also another hidden aim. The revision of Marxian theory on the development of capitalism in agriculture must promptly lead to the renunciation of the theory as a whole and the abandonment of the revolutionary conclusions which are drawn from it. The small peasantry forms even in the most developed countries a notable part of the population and a majority of it in the whole world. The Marxian theory of the hopeless position of the small peasantry under capitalism provides the foundation for the alliance between the proletariat and the small peasantry as well as their joint struggle for the destruction of capitalist society. The theory of the [German] Social Democrats which implies the possibility of prosperity of small peasant economy under capitalism is nothing else than the abandonment of the struggle for Socialism.’

54 See Bernstein [1996/97: 41, emphasis added].

55 The case is put thus [Bernstein, 1996/97: 43]: ‘In short, a century of modern imperialism has extended the determinants of industrialization far beyond the prospects of agrarian transition in landscapes inhabited exclusively by classes of landed property and agrarian labour.’ If one were to state less categorically ‘extended but not closed off completely the determinants of industrialization’, his claims would be less problematic; but then the concept of an agrarian question of capital as having been accomplished – a core element in Bernstein's framework – would itself no longer hold.

56 That he has now joined the ranks of those for whom socialism is no longer part of the agenda of development studies is clear from what Bernstein (pp.127, 131) states in the volume edited by Kothari. Hence the observation that ‘in the light of historical experience to date it may prove impossible to rethink notions of any feasible socialism(s), and of socialist development, that can be projected into a forseeable future … a viable Marxist political project – the future of socialism – in current and forseeable conditions remains as problematic as ever, and even more unanswerable.’ As will be seen below, the abandonment of socialism as a political objective is yet another position Bernstein shares with Byres.

57 Hence the claim [Bernstein, 1996/97: 52] that the continuing efficacy of petty commodity production is itself supported by ‘the generalization of the “peasant path” in the discourses and programmes of anti-colonial and national democratic movements’. This view that the resolution of the agrarian question is in essence about the survival of peasant economy and the struggle for bourgeois democracy is once again a theoretical position that Bernstein shares with the exponents of the ‘semi-feudal’ thesis. For the latter, therefore, agrarian transition entails the replacement of landlord capitalism (= the Prussian path) with peasant capitalism (= the American path), a transformation in which one form of capitalism (unbenign, landlord, undemocratic) is followed not by socialism but by another ‘pure’ stage of capitalism (benign, peasant, democratic).

58 For the details and implications of Baüer's theory about nationalism, and the similarity between his views and the ones held by those associated with the subaltern studies project, see Brass [2006].

59 See Byres [1996] and also Byres [1995: 565ff.]. For yet earlier references to the feudal and/or semi-feudal character of the post-1947 Indian countryside, see Byres [1974: 223, 246, 247, 248, 251].

60 With some surprise one learns (Byres, p.59 note 4) that this already redundant thesis is to be expounded yet further, in no less that three additional volumes, to appear at some point in the future. There may, however, be another and simpler explanation for the frequent presence in his writings of references to forthcoming – but never appearing – volumes by him and others (see, for example, Byres [2000: 240]). It is noticeable that, whenever faced with a difficult or potentially contradictory point of theory, Byres states that the issue in question will either be the subject of a forthcoming book, or would need another book to address. Typical, perhaps, are the following two examples. First [Byres, 1996: 8]: ‘I stress that in this book no attempt is made at an exhaustive treatment of the criteria by which one might assess whether or not capitalist agriculture was developing. A separate, full-scale work would be necessary to do justice to such an undertaking.’ And second [Byres; 1998: 220]: ‘I can only promise to address some of these issues in a future treatment.’ The need to engage intellectually with an awkward piece of theory is thereby postponed sine die.

61 Japan is accordingly categorized as ‘a quite distinct capitalist agrarian transition’.

62 The prevalence in the article by Byres of references to texts about Latin America written in Spanish raises a difficult question – about sources cited in languages other than English – first encountered in his book [Byres, 1996]. The latter exhibited a mismatch – on which see Brass [2002: 472–3, note 45] – between on the one hand the citation by him of German texts, and on the other their contents. At least one of the texts contained arguments that were the opposite of those he attributed to it.

63 The irony of Byres [1987] taking Chakravarty [1987] to task for excluding class struggle from his analysis of the contribution by Marxist political economy to an understanding of development theory, when this is precisely what Byres himself does with regard to the acceptability to capitalist producers of unfree labour, requires no further comment.

64 For the concept ‘deproletarianization’, see Brass [1999].

65 Details about the unfree relations of production utilized by rich peasant capitalists in each of these case studies are presented in Brass [1999]. That agrarian capitalists still prefer unfree workers to labour-power that is free, not just in the eastern lowlands of Peru but also in both the Santa Cruz region and the north Amazon of Bolivia, is clear from current research into the operation of the enganche system in these areas [Assies, 2002; Bedoya Garland and Bedoya Silva-Santisteban, 2005a; 2005b]. For confirmation that in Haryana the incidence of unfree labour is as entrenched as ever, see Mukherji and Sahoo [1992], Chowdhry [1993] and Rawal [2004]. However, the errors permeating a subsequent and much altered version of the text by Rawal suggest that he is unfamiliar with the theory, the debate and the literature about unfree labour. Thus he mistakenly attributes to Ramachandran [1990] the argument made by this reviewer generally throughout the 1980s and about Haryana in 1990: namely, a critique of the ‘belief that the shorter the duration of a contract, the more free is the relationship’. Rawal also wrongly accuses me of not differentiating the workforce in terms of unfreedom. This overlooks the fact that my case is – and always has been – that attachment has shifted from permanent workers to increasingly landless agricultural labourers who are no longer unfree throughout the whole agricultural cycle or on an intergenerational basis. In other words, workers are indeed differentiated in terms of unfreedom, or the very argument Rawal claims is absent from my analyses of debt bondage.

66 Among them not just Byres but also Jan Breman and Utsa Patnaik.

67 This point – about the resurgent role of gangmasters in British agriculture – has been made in Brass [2004]. An indicator of the theoretical contradiction generated by this seemingly incompatible process – capitalist development on the basis of unfree production relations – is the argument made by neoclassical economists defending the employment of unfree labour by agribusiness enterprises. Gangmasters and unfree migrants, we are told, are necessary because they supply a need by filling jobs no indigenous worker wants to do. What is not said by these neoclassical economists is that such a defence in effect negates one of the central tenets advanced by laissez faire theory: namely, that where workers are scarce and/or not available in the desired quantities, wages should rise until this need is met, thereby establishing the equilibrium structuring the marginalist analytical framework. As Marxists have long argued, the operation of the market depends ultimately on the intervention of the state – the very institution and the very activity (= intervention) which neoliberalism wishes to ban from the economic sphere – both to regulate the supply of labour and to strip away legislative protections secured by workers in the course of the class struggle.

68 This is linked in turn to a very old fallacy circulating in the domain of bourgeois political economy (mainly in Germany towards the end of the nineteenth century), and one criticized almost a hundred years ago by Rosa Luxemburg [1954]. Even at the start of the twentieth century, therefore, Luxemburg [1954: 10–11] points out that ‘[a]ll of these advanced countries [with which Economics so exclusively concerns itself] produce for one another, partly also for the most distant parts of the world, and the products of every continent are utilized by them at each step, in consumption as well as production. In the face of such enormously developed mutual exchange, how can we separate the boundaries of the “economy” of this nation from those of another? How can we speak of so and so many “national economies,” as if they were economically closed areas which we could analyze each by itself?’. Her conclusion is salutary [Luxemburg, 1954: 24]: ‘We are discovering that these days [c. 1916] a “commodity” is being exported and imported which King Nebuchadnezzar did not know, nor did his contemporaries, nor did Antiquity, or the Middle Ages. This commodity is capital. This “commodity” is not used for the purpose of filling in “certain gaps” in foreign “national economies,” but, on the contrary, it produces gaps, creates cracks and fissures in the structures of antiquated types of economies. It penetrates them, and like a charge of dynamite, sooner or later, it reduces these economies to a heap of rubble. Together with this “commodity”, capital, even stranger “commodities” are being exported to the entire globe by a few older countries – and in constantly increasing quantities: modern means of transport and the annihilation of entire peoples, money economy and the economic ruin of the peasantry, wealth and poverty, the proletariat and exploitation, insecurity of existence and crises, chaos and revolutions’ (original emphasis). This view – that from the early twentieth century onwards capitalism was an increasingly non-national systemic form, and consequently opposition on the part of workers should be correspondingly international in scope – was undermined by the rise of Stalinist theory. Like bourgeois political economy, it recuperated the epistemology structuring nationalism, and restored the nation to its politically central role in struggles that were henceforth about a transition not to socialism but to capitalism.

69 Symptomatically, Byres is reduced to asking questions to which the answers have long been known. Is it the case, he inquires somewhat naïvely, that for capitalism to be present it is necessary for agriculture to be capitalist, or can accumulation occur in industry (and therefore industrialization take place) on the basis of a non-capitalist agriculture?

70 Rather touchingly, he informs the reader that it is his intention to address (in a forthcoming paper entitled ‘Structural Change, the Agrarian Question, and the Possible Impact of Globalization’ – emphasis added) whether or not the existence of international capital might have implications for agrarian transition. Until Byres pronounces on the issue, of course, all the rest of us will have to contain our excited anticipation, being as we are unaware of the fact that the existence of a global capitalism might just have implications for national agrarian transformation.

71 This emerges clearly from a comparison of early texts by him with more recent ones. Hence the following assertion [Byres, 1974: 235]: ‘Rich peasants were not the “masters of the countryside” at Independence – the landlord class was still very much in command – and they were not to be clearly seen in all parts of India. They were a class of capitalist farmers in embryo, in the womb of the old order.’ A quarter of a century on, nothing much has changed in this regard. ‘Rich peasants’, we are told by Byres [2000: 266] in a much later text, ‘constituted a class of proto-capitalists, that still bore the stigmata of a pre-capitalist order.’

72 Nowhere is this demonstrated more clearly than in the confused and unsuccessful attempt by Byres to dismiss the argument made by Bowman [1993], an earlier and infinitely better comparative study of the US planter class and the Prussian Junkers. Byres [1996: 125] cites in part a long section from Bowman [1993: 110–11], a fuller version of which goes as follows: ‘As growing numbers of rural East Elbians migrated to urban centres in the westerly provinces or emigrated to the United States, and as the expanding cultivation of sugar beets increased the demand for large numbers of seasonal labourers, Junker landowners turned rapidly in the 1880s to the seasonal employment of low-wage migratory workers from the Polish districts of the Russian and Austro-Hungarian Empires … Insofar as these migrant Polish labourers were subject to physical abuse, deemed racially and culturally inferior, enjoyed no political rights, and were not permitted to change employers during the agricultural season (April to November), the label “free labour” seems as inappropriate for them as for contemporaneous black sharecroppers. It was not until the German revolution of 1918–19 and the Weimar Republic that agricultural workers achieved a civil and political status equal to that of industrial workers’ (emphasis added). Against this view Byres makes the following objection: ‘That is simply to confuse concepts and misunderstand the meaning and nature of free wage labour. Such migrant labour was free of the means of production, and, despite restrictions during the agricultural season (a phenomenon not uncommon among agricultural labourers, and indeed other forms of labour, where capitalist relations prevail) free to move between seasons’ (emphasis added). Unfortunately, it is Byres, not Bowman, who misunderstands the conceptual distinction between free and unfree labour. What Byres forgets is that, for a worker to conform to the Marxist definition of free labour, s/he must have and retain permanently the capacity to commodify or recommodify his/her own labour-power. Where such workers are unable personally and unconditionally to sell their only commodity, labour-power, the social relations of production are not (and cannot be considered to be) free. That Polish workers employed on a seasonal basis in German agriculture were deproletarianized is clear from the trajectory of their employment history. In Poland, therefore, such workers were initially a proletariat, in that they offered their labour-power for sale. Once they had migrated to Germany, however, such workers were no longer permitted to offer their labour-power for sale to the highest bidder, and in fact were specifically prevented from doing so. In short, Polish workers had been deproletarianized: they were free but became unfree. The italicized portion of the quotation from Byres underlines neatly the dilemma he cannot avoid facing: those using these ‘restrictions’ (as he admits, ‘a phenomenon not uncommon among agricultural labourers’) are capitalists. Hence the need for Byres to have to define these production relations as ‘free’, any evidence to the contrary notwithstanding.

73 A contributory factor that cannot be overlooked is that, like other exponents of the semi-feudal thesis, Byres remains trapped in a political economy which, although its era came to an end formally in March 1953, endured well beyond that conjuncture.

74 For the centrality to his framework of the systemic conditions licensing increased agricultural productivity, see Byres [2000: 239]. It is, in his words (Byres, p.55), a question of ‘what is it that has allowed the unleashing of capital accumulation, in both town and country, that has lain at the heart of capitalist transformation’.

75 For the view that unfree relations of production (=‘semi-feudal features’) are incompatible with capital accumulation in agriculture, see Byres [1996: 29, 78]. For the relabelling by him of bonded labour employed by capitalist producers as ‘free wage labour’, see Byres [1996: 38]. For examples of the same views – the attempt to redefine as free those unfree workers employed in capitalist agriculture – but this time held by neo-classical economic historians, see Fogel and Engerman [1974] and Steinfeld and Engerman [1997]. Not the least of the many ironies about this epistemological overlap is that, over the years, Byres has always insisted that a vast theoretical gulf separated his own (semi-feudal) analytical framework from that of neo-classical economists such as Michael Lipton. This overlap has not escaped the notice of other contributors to the Agrarian Studies volume. In the subsequent discussion (pp.84ff.) of the papers by Byres and Keith Griffin – the latter dismissed by Byres as a neo-classical economist (p.91) – some present drew attention to the epistemological similarity between them (p.84).

76 Details about this are outlined in Brass [2002]. Socialism is not merely nowhere on the semi-feudal agenda, therefore, but it never has actually been a part of this (see, for example, Byres [1982: 136; 1991: 3; 2005: 85]). It is with some relief, one imagines, that Byres (p.54) is able finally to join all the other ex-socialists in proclaiming capitalism as the ‘saviour’ of the holy grail of agricultural productivity. About this shift there can be no question, since he makes it clear in the following observation: ‘Whether, with the demise of a socialist path seriously pursued and the disappearance of collective agriculture, one should continue to use the notion of a “third world” and suggest thereby a possible socialist solution to economic backwardness, is, perhaps, doubtful. We may proceed, rather, in terms of “rural societies” in contemporary economically backward, or poor countries, in which, unambiguously, it is capitalism that must be the focus of our attention. It is capitalism that drives such transformation.’

77 The palpable fact of unfree labour employed by agrarian capitalists has led to an equally palpable discomfiture on the part of exponents of the semi-feudal thesis. An example is the way in which Byres [1999] attempts to sidestep the persistence of bonded labour in the highly commercialized agriculture of the most advanced regions in India. In the states of Haryana, Punjab, Uttar Pradesh, and Andhra Pradesh, therefore, he is unable to deny that the ‘dominant classes’ – for really rather obvious reasons, he cannot bring himself to write ‘agrarian capitalists’ – do indeed utilize ‘various tied labour arrangements’. However, he then goes on to describe as ‘contradictions’ in the emancipatory process (unfree → free) the fact that in Andhra Pradesh the incidence of female attached labour has increased, and qualifies it yet further by casting doubt on this being a widespread a phenomenon [Byres, 1999: 18–19]. Symptomatically, therefore, he fails to question the extent to which such developments in the capitalist labour process (free → unfree) are typical, categorizing them as ‘anomalies’ (=‘contradictions’) instead of asking whether in fact they are anomalies.

78 The extent of the current overlap between the semi-feudal thesis and neoclassical economic theory regarding the necessity of yet more capitalist development in the so-called Third World is evident from a comparison of Byres' epistemology with that informing the marginalist approach of Lal [2004]. The latter, like Byres, argues both that economic growth per se is a desirable end and that it is capitalism that will deliver this throughout the Third World. Where such an argument leads is equally clear. Lal concludes that, as empires guarantee economic development, imperial conquests are ‘natural’, they result in mutual gains and have ‘promoted peace and prosperity’. That a neoliberal economist holds this view about the continuing and benign rôle of the market as a solution to the problems of underdevelopment is unremarkable; that much the same kind of argument underpins the view of a self-proclaimed Marxist is, to put it mildly, surprising.

79 On this subject generally, see also Brittain [2005].

80 The sub-title to this section – praise deemed appropriate – is a reference to the chorus ‘The many rend the skies with loud applause’ in Alexander's Feast by George Fredric Handel.

81 See Petras and Morley [1990] for contributions to and comprehensive examinations of debates about Latin American development. For an evaluation of the agrarian reform programmes in Latin America during the ‘development decade’, see especially Petras and LaPorte [1971].

82 These analyses – by Petras [1969], Petras and LaPorte [1971: 125ff., 233ff.], and Petras and Zemelman Merino [1972] – of rural Chile are based on an in-depth knowledge of the historical context combined with fieldwork research.

83 See Petras and Veltmeyer [1997; 2001b; 2003b; 2005], Petras [2003] and Veltmeyer [1997; 1999; 2000]; also Petras, Veltmeyer and Vasapollo [2005] and Saxe-Fernandez, Petras, Nuñez, and Veltmeyer [2001].

84 Unlike many others on the left, Petras[ 2006b] has not been afraid to tackle politically difficult issues.

85 For a critique of imperialism, and how this systemic form remains extant (claims by others to the contrary notwithstanding), see Petras [2002]. On the link between capitalism and overfishing in Newfoundland, see also the useful analysis by Overton [2000] about its impact on the rural economy and the resulting debate, an approach which complements that of Petras and Veltmeyer.

86 Hence the validity of the observation (Petras and Veltmeyer, p.180) that: ‘Much current non-Marxist – especially postmodern – analysis of the peasantry … either ignores the state altogether or recognizes its presence and impact but denies that its agency is based on class.’

87 This argument was made early on, in the now classic analysis by Petras 1990 and also in Petras and Morley [1990: 147ff.].

88 See for example Veltmeyer [1997].

89 ‘The major tactic of this approach’, he argued [Petras, 1970: 378, 379, 382], ‘is to ascribe the problems of Latin American development and motivation to “feudalism,”“traditionalism,” or “Hispanic culture”… Latin landowners who pay low wages and exploit their workers for profits [exponents of this approach maintain,] are not like their counterparts in California or Texas – they are “feudalists” acting out the Spanish heritage’ (see also Petras [1978: 137ff.]). Rightly, he aligned himself with the approach of Andre Gunder Frank, who ‘effectively destroys the myth of the feudal past so often cited and exploited by U.S. scholars as a means of defending capitalist development as a “democratic alternative” to Latin American stagnation.’ This critique remains as relevant today as it was then, with the difference that it no longer applies simply to US academics.

90 See also Petras and Veltmeyer [2003a], and Petras [2006a].

91 Brittain [2005] also notes that as capitalism has become a global relation, neoliberal economic policies advocated by US imperialism have fuelled the grassroots resistance that sustains the powerful guerrilla movements in rural Colombia.

92 In this they have been aided by those theorists (Adolfo Gilly, Guillermo de la Peña in Mexico; Gail Omvedt in India) – for the most part ‘new’ populist postmodernists – whose ideas reflect the prevailing views about such movements.

93 See also Petras [1998], and Petras and Veltmeyer [2001a]. That the source of individualistic ‘conservatism’ – insofar as it can be regarded as such – is not innate to peasant farmers and agricultural workers, but rather lies outside them, in terms of the wider economic context in which they operate, necessarily gives rise to optimism about the political direction of future change. This was recognized a long time ago by Mann [1929: 80, original emphasis], who wrote: ‘After long experience of Indian farmers in many parts of India, I think that this idea of innate conservatism among the rural classes is not correct, and possibly they are less averse to change than a very large proportion of the farmers of western countries. I have seen, again and again, within twenty years an old but less efficient implement replaced almost entirely, over large regions, by one more efficient, or an improved type of seed replace that in use for a hundred years, or the employment of artificial manure become general. And it would really seem to be true that readiness to adopt new methods is the characteristic of the Indian cultivators, provided they are proved, to their own satisfaction, to be of advantage, and provided they give a return which will warrant the borrowing of capital at high interest.’ As a socialist option has been in effect banished from the approach to the agrarian question of Byres and Bernstein, neither of them retains a progressive politics capable of transcending capitalism.